WO1992001060A1

WO1992001060A1 - A fermentation process for riboflavin-producing organisms

Info

Publication number: WO1992001060A1
Application number: PCT/US1991/004941
Authority: WO
Inventors: Richard B. Bailey; George William Lauderdale; Donald L. Heefner; Craig A. Weaver; Michael J. Yarus; Linda A. Burdzinski; Annette Boyts
Original assignee: Zeagen, Inc.
Priority date: 1990-07-13
Filing date: 1991-07-12
Publication date: 1992-01-23
Also published as: EP0539507A4; AU8309391A; EP0539507A1

Abstract

A high productivity method for efficiently fermenting yeast for the production of riboflavin. The method includes restricting the carbon source uptake rate while restricting growth in a linear manner by restriction of a micronutrient. The method is especially useful in the fermentation of Candida yeast. In preferred embodiments, the nitrogen source, copper and iron concentrations and pH are controlled. Using this method, volumetric productivity has been increased to at least about 0.17 g total riboflavin per liter per hour.

Description

A FERMENTATION PROCESS FOR RIBOFLAVIN-PRODUCING ORGANISMS

Field of the Invention

The present invention is directed to a method for the efficient fermentation of yeast which produce riboflavin and, in particular, a continuous fermentation process for

Candida fa ata yeast in a manner designed to improve the production of riboflavin by the yeast.

Background of the Invention Riboflavin, also known as Vitamin B₂, Vitamin G, and lactoflavin, is typically produced by chemical synthesis or biosynthesis. Riboflavin is biosynthesized by a wide variety of microorganisms in amounts which greatly exceed the metabolic requirements of the organisms. Riboflavin produced by these organisms can be used as a food or feed additive. Ascomycetes, such as Ashbya qossypii and Eremothecium ashybii are known for production of riboflavin by fermentation. Riboflavin production by other microorganisms is also known. For example, the bacteria belonging to the genera Clostridiu and Bacillus, as well as various genera of yeast, including Candida, Saccharomyces. Hansenula, and Pichia are known for riboflavin production. More specifically, for example, U.S. Patent No. 3,433,707 (1969) to Matsubayashi, et al. describes the production of riboflavin by three species of Pichia yeast. Yields of riboflavin of between 10.5 mg/1 and 51 mg/1 in 12 days were reported. Riboflavin overproduction by Ashbya qossypii of 6.42 g/1 has been -2- reported by Szczesnika, et al. (1973) as discussed in Perlman, Primary Products of Metabolism, 2 Econ. ^•Microbiology at 312 (1978).

Commercial production of riboflavin by fermentation processes has been hampered by a number of problems in large scale production. For example, in batch fermentation processes, there is the problem of the microorganism population increasing to levels at which the growth rate significantly decreases or ceases altogether due to the build-up of toxic by-products generated by the fermentation or for other reasons. As such, batch fermentation runs require substantial amounts of non- operative time which results in decreased production.

Another concern in large scale fermentations is the cost of nutrients. It is desirable to achieve improved yields of end product to fermentation medium nutrients and thereby decrease production costs.

A further concern in the commercial production of riboflavin is the prevention of contamination of fermentation medium. It is known to sterilize medium components by filtration of contaminating microorganisms. However, such sterilization increases equipment costs and complexity, as weld as carrying some risk of contamina ion. Another concern in the commercial production of riboflavin is the production of substances which are toxic to the fermentation. For example, such substances can be nutrients which are chemically converted into toxic substances or metabolic by-products.

Successful large-scale fermentation also requires simplified process controls. For example, minimization of the number of feed inputs for various nutrients, as well as for control of process parameters, such as pH, is desirable.

Large scale fermentations also require that carbon dioxide be maintained below a level which is toxic to the fermentation. Carbon dioxide levels can be controlled by aeration, however, aeration can be difficult in large fermentation vessels. Consequently, fermentation conditions which minimize the production of carbon dioxide are also beneficial. Commerciallyviable production processes also require high riboflavin production rates. It is known that ribo¬ flavin production rates are affected by the growth rate of microorganisms in the fermentation medium and by the amount of riboflavin produced per microorganism ("specific formation") . While it is known that some compounds, such as zinc and cobalt, improve specific formation of ribo¬ flavin, identification of other such compounds is useful in improving riboflavin production.

In view of the above-discussed limitations in the commercial production of riboflavin, there is a need to address and overcome these problems. Summary of the Invention

In accordance with the present ^'invention, a fermen¬ tation process has been developed which results in a high level of growth and productivity that is sustainable over an extended period of time. The present invention includes a fermentation method for riboflavin-producing microorganisms which includes restricting a nutrient feed rate such that the nutrient uptake rate is limited. In this manner, several unexpected advantages are obtained. An unexpectedly high increase in the yield coefficient for carbon is obtained which is greater than that expected from the observed decrease in production of observable by¬ products. Moreover, production of arabitol as a by¬ product was unexpectedly observed. In addition ethanol production was also stopped. In a further embodiment of this method, the limiting nutrient is glucose.

Another embodiment of the foregoing fermentation process includes conducting the fermentation with ammonium salts, ammonia gas or ammonia hydroxide as the nitrogen source in the fermentation medium.

The present invention also includes a method for improving the specific formation of riboflavin in a fermentation by increasing the copper concentration in a fermentation medium up to a non-toxic level. In a further embodiment of this method, a riboflavin-producing strain of microorganism is selected which has a decreased sensitivity to copper toxicity. A further ^' specific embodiment of the present invention includes a fermentation method for riboflavin producing microorganisms which includes heat sterilizing a nitrogen source for the fermentation medium and controlling the pH of the fermentation medium only with the addition of base. This embodiment of the invention further includes the use of ammonium sulfate as the nitrogen source in the fermentation.

Yet another embodiment of the present fermentation process is a continuous fermentation method for the production of riboflavin which involves diluting a fermentation broth by adding a fresh fermentation medium which includes a nutrient. The rate of addition of the nutrient is such that the nutrient uptake rate is limited. This method further includes providing an ammonium salt, ammonia gas, or ammonia hydroxide as a nitrogen source. In a preferred embodiment of this process, the nutrient in the fresh fermentation medium is glucose. Another embodiment of the method includes use of Candida famata having the identifying characteristic of ATCC 20849. Still further embodiments of the process include the regulation of pH within the range of about 3.0-5.5 and regulation of the iron concentration.

Detailed Description of tne Invention The present invention involves a method for fermentation of microorganisms in a fermentation broth, and, in particular, for fermenting microorganisms which produce riboflavin. By practicing the present invention, high product to nutrient yields can be achieved, and the productive phase of a fermentation can be effectively extended. As a result, depending upon the nature of the fermentation, high cell densities can be achieved and/or the production of materials, such as vitamins, can be increased. The method further includes controlling various other fermentation parameters, including nitrogen source, copper and iron concentration, and pH, to improve cell growth and riboflavin formation in yeast.

The present method is particularly advantageous for microorganisms such as yeast of the genera Candida, Saccharomyces, Hansenula. and Pichia and, in particular, Candida famata. More particularly, this method is advantageous for strains of Candida famata identified by ATCC Nos. 20755, 20756, 20849, and 20850, these strains having been deposited with the ATCC under the terms of the Budapest Treaty. Most particularly., this method is most advantageous for the strain of Candida famata identified by ATCC 20849 and mutant strains thereof.

An embodiment of the present invention includes con¬ ducting a fermentation at a restricted growth rate while restricting the carbon uptake rate in the fermentation. In this manner, it has been found that the yield coefficient (^_p/s-^{1 can be} increased. As used herein, the term "yield coefficient" (Y_p/S) refers to the amount of riboflavin (p) formed for every mass unit of substrate (S) consumed. In the present instance, the substrate is the carbon source. In the development of the present process, production and accumulation of ethanol was observed. To address this problem, the process was modified to restrict the carbon source uptake. In so doing, Applicants have developed a process which unexpectedly prevents the production of arabitol and which leads to an unexpected increase in the Y _/s for carbon source which is non-proportional for the observed decrease in production of for example ethanol and arabitol. Restriction of the microorganism growth rate may be accomplished by any means known in the art. Specifically, one means by which to accomplish growth rate restriction is by nutrient limitation. For example, rate restriction can be accomplished by restriction of micronutrients, such as iron. Alternatively, growth rate can be restricted by restriction of macronutrients, as well.

As used herein, the term "growth rate" (which will be indicated by the symbol "dx/dt"). is the mass of cells formed per mass of cells in the fermentation broth per unit time. Maximum growth rate is indicated by the symbol "dx/dt^ax". Unless indicated otherwise, the units for growth rate will be hr^"1, and the symbol will be "dx/dt^hr" ^1,1. Specifically, it has been determined that it is useful to restrict the growth rate of microorganisms to a range from about 2% of dx/dt^max to about 20% of dx/dt^max, and more preferably to about 5% of dx/dt^max.

The foregoing aspect of the present invention further involves restricting the rate of addition of the carbon source to a rate which restricts the carbon source uptake rate, measured in grams per liter per hour (g/l/h-r) . In other words, if the rate of addition of the carbon source were to be increased, the carbon source uptake rate would increase. Without intending to be bound by theory, it is believed that by restricting the availability of the carbon source, the production of toxins, including ethanol and arabitol, is prevented or substantially reduced because microorganisms will be forced to preferentially utilize available carbon for growth. It is further believed that by restricting the carbon source but not having it at a growth rate limiting concentration, the microorganisms do not detect carbon source limitation and thus continue normal metabolic processes without inter- rupting non-essential metabolic processes, such as for example, non-essential vitamin production. It has also been found that by conducting a fermentation in a carbon restricted manner, that the respiration quotient (volume of CO., evolved/volume of oxygen uptake, "RQ") is typically decreased from greater than 1.0 to less than 1.0. This result indicates more efficient use of carbon.

A further embodiment of the present method of the invention may include restricting the addition rate of the carbon source to a rate which is growth rate limiting. It has been unexpectedly found that the yield coefficient of a carbon source limited fermentation is greater than that of an unrestricted carbon source fermentation although less than that of a carbon source restricted fermentation in which carbon is not growth rate limiting.

The rate of addition of the carbon source can be monitored and the carbon source can be fed at a rate to achieve the desired restricted carbon uptake rate. The carbon source is fed on a continuous basis to maintain the desired restricted level. Specifically, it has been determined that it is useful to maintain the feed of the carbon source at a rate designed to restrict the uptake rate to a range from about 20% of the maximum uptake rate to about 80% of the maximum uptake rate, more preferably from about 40% of the maximum uptake rate to about 60% of the maximum uptake rate and most preferably to about 50% of the maximum uptake rate. The embodiment of the present invention relating to carbon source restriction is effective for use with any carbon source used in a fermentation. Specifically, the invention is useful for any type of, carbohydrate and, in particular, any monosaccharide, disaccharide or trisaccharide. In particular, for the fermentation of Candida, the carbon source is preferably glucose.

In the practice of the present method of restricting the feed rate of a carbon source to obtain the benefits of extended fermentation and improved product yield, it is recognized that by restricting the carbon source addition rate too greatly, the ability to grow and/or produce materials may be negatively affected to an undesirable extent. Therefore, it is desirable to maintain the carbon source feed rate at the desired restricted level but to also maintain a sufficient carbon source feed rate and concentration to avoid having a negative impact on growth and/or production of extracellular materials. In a preferred embodiment of the present invention the fermentation broth is diluted with fresh fermentation medium and the carbon source uptake rate is restricted. Preferably, this method includes three phases: (1) an inoculation phase; (2) a batch growth phase; and (3) a carbon source restricted growth rate restricted, fed-batch or continuous culture growth phase. The inoculation phase includes providing an inoculum of actively growing microorganisms to a fermentation vessel. The batch growth phase includes propagating microorganisms in the fermen- tation vessel with initial medium additions having an initial carbon source concentration at from about 20 grams per liter to about 50 grams per liter. Typically, the carbon source concentration is then depleted to between about 1 g/1 and about 5 g/1 during the batch period. During this time, the carbon source uptake rate is typically above about 5 g/l/tir. During the transition from the batch phase to the continuous phase, medium additions, including a carbon source are initiated. Typically, the carbon source feed rate is between about 0.5 g/l/hr and about 5.0 g/l/hr, more preferably between about 1.5 g/l/hr and about 3.5 g/l/hr and most preferably at about 2.5 g/l/hr. The carbon source uptake at this phase is essentially the same as the corresponding feed rate and is, therefore, limited by the feed rate. During this phase, the concentration of glucose in the fermen¬ tation medium is about 250 parts per million. The feed stream includes most commonly glucose and is otherwise nutritionally balanced for growth (i.e., including nutrients such as potassium, phosphate, iron, sulfate, citric acid, magnesium and other trace metals) .

The carbon source uptake rate during this phase is less than the maximum rate, such as that which occurs during the batch phase. The fermentation growth rate is also not carbon limited during this phase. Rather, the fermentation growth rate is restricted, preferably by a microπutrient, such as iron. During this phase, unexpectedly high carbon source product yields are achieved. For example, Y _/s values of greater than about .060 can be attained and more preferably of greater than about .067 can be attained.

In conjunction with the process parameters of the present invention, as described above, the present invention includes the control of additional process parameters. In the fermentation of yeast, nitrogen source, copper and iron concentrations, and pH are controllable process parameters. The present invention further includes control of these variables to obtain optimal growth rates and high yields of riboflavin production in an efficient riboflavin production process. Preferably, the volumetric productivity of riboflavin is at least about 0.12 g of riboflavin per liter of fermentation broth per hour, more preferably at least about 0.15 g/l/hr and most preferably at least about 0.17 g/l/hr.

With respect to control of nitrogen source in a fermentation, an ammonium salt, ammonia gas or ammonium hydroxide is used as the nitrogen source in the fermentation medium. Preferably an ammonium salt is used. Suitable ammonium salts are ammonium sulfate, ammonium phosphate, ammonium carbonate and ammonium chloride. Most preferably, ammonium sulfate is used. In particular, ammonium sulfate is provided in amounts sufficient to maintain the nitrogen concentration at levels necessary to meet nitrogen requirements. Typically, ammonium sulfate is provided in a feed stream in amounts between 40 g/1 and 60 g/1, more preferably 45 g/1 and 55 g/1, and most preferably 48 g/1 and 52 g/1, to meet nitrogen requirements.

The use of ammonium sulfate alone and in combination with other aspects of the invention has several unique advantages over use of other nitrogen sources, such as urea. A first advantage is that necessary fermentation apparatus and its operation is simplified, because pH of the fermentation medium can be controlled throughout the fermentation with only a base titrant, rather than an acid and a base titrant. As discussed above with respect to the glucose-limited fermentation process, the fermentation medium initially has a high concentration of glucose. As the glucose is metabolized, acid is generated requiring pH control of the fermentation broth by a base. When an ammonium salt is used as a nitrogen source, as the ammonium salt is metabolized, ammonium groups are deprotonated, thereby producing acid. Accordingly, as the fermentation switches from uncontrolled to controlled carbon uptake, hydroxide ion production decreases, and the effect of nitrogen source metabolism becomes greater, pH can still be controlled with the same base titrant as in the earlier part of the fermentation. In contrast, for example, if urea is used as a nitrogen source, as it is metabolized, ammonia is produced which becomes protonated, thereby creating hydroxide ions which make t e bioth basic. Therefore, an acid titrant is requ_- xor pH control in addition to the initial need for a base titrant. Another advantage of using an ammonium sε-lt in the present fermentation method is that the risk of contamination can be substantially reduced because it can be heat sterilized prior to its introduction into the fermentation. By heat sterilization, any bacterial or other contamination present in the ammonium salt solution is killed prior to addition of the ammonium salt solution to the fermentation medium.

In contrast, a nitrogen source such as urea cannot be heat sterilized because urea breaks down when heated. Thus, urea must be sterilized by filtration which adds additional process steps and equipment, thereby further increasing the risk of contamination. Such equipment raises overall costs and adds complexity to the process. Ammoriium salts, and in particular ammonium sulfate, have the further advantage of not 'being converted to substances toxic to the fermentation. Urea, however, has the disadvantage of breaking down to biuret, which can be toxic to a fermentation. Biuret can significantly lower riboflavin production and microorganism growth.

A further embodiment of the subject invention includes regulation of the popper concentration to increase the specific formation of riboflavin. It is known that copper is an essential trace metal for biological systems. It has been unexpectedly found that specific formation of riboflavin is directly related to increased concentration of copper up to about a copper concentration which is toxic to the fermentation. More specifically, the present invention includes a copper concentration, as referenced by CuS0₄-5H₂0, from about 0.66X10^"5 g/1 to about 3.0X10^"5 g/1, more preferably from about 1.0X10^"5 g/1 to about 2.5X10^"5 g/1, and most preferably from about 1.8X10^"5 g/1 to about 2.2X10^'5 g/1. By regulating the concentration of copper present in the fermentation medium, values for the specific formation of riboflavin of .1 gram of riboflavin per gram of cell can be achieved, more preferably .2 gram of riboflavin per gram of cell can be achieved, and most preferably .3 gram of riboflavin per gram of cell can be achieved.

Copper can be provided by the addition of copper sulfate or any other salt of copper. Most preferably, the copper source in the present invention is copper sulfate. Another embodiment by which to increase the specific formation of riboflavin is to practice the method of the subject invention with microorganisms that have been selected which have a high tolerance for copper toxicity. In this manner, higher specific formation of riboflavin can be achieved because higher copper concentrations can be used with such microorganisms than with microorganisms that have not been selected for copper tolerance. Such selection for tolerance to copper toxicity can be, for example, from screening naturally occurring organises to identifying natural occurring variations in tolerance for copper toxicity. Alternatively, genetic manipulation, such as mutagenesis and selection or recombinant techniques, can be used to increase the specific formation of ribo- flavin beyond what is possible with microorganisms having a normal tolerance for copper toxicity. Specifically, the above method of selecting can be used with strains of Candida famata and more specifically with Candida famata identified by ATCC Nos. 20755, 20756, 20849 and 20850. For example, microorganisms with increased tolerance for copper toxicity can be selected by standard mutagenesis and plating techniques. First, the minimum inhibitory concentration of copper on which riboflavin- producing microorganisms can survive is determined, and a starting population of riboflavin-producing microorganisms is subjected to mutagenesis, either by chemical or physical means. The method of mutation employed in the selection methods of the present invention can be any of various chemical or physical mutation methods known in the art.

For example, subjecting a microorganism to various concentrations of N-methyl-N'-nitro-N-nitrosoguanidine

(NTG) , ethylmethane sulfonate, hydrazine, or nitrous acid induces mutagenesis in microorganisms. A culture of microorganisms can also be mutated by subjecting the culture to physical mutagens, such as ultraviolet or gamma radiation.

Once the starting population has been mutated, it is then cultured on mediums with varying, incremental copper concentrations. The culture surviving on the highest copper concentration^{^} medium, or at least cultures surviving on the minimum inhibitory concentration, can be selected and used to produce riboflavin, such culture strain having an increased tolerance for copper toxicity above what a parent strain can tolerate, thereby allowing for increased specific formation of riboflavin. The above selection process may be repeated to develop and select a strain for which the specific formation of riboflavin may be yet further increased.

The preεent invention further involves conducting fermentation while diluting the fermentation broth by the addition of fresh fermentation medium. The fresh fermentation medium is generally an aqueous solution added to a fermentation vessel which includes a carbon source. In particular, the fresh fermentation medium can be water, or water with nutrients, such as glucose, and other materials added to it.

The fermentation broth can be diluted either by increasing the overall volume with the addition of fresh fermentation medium, i.e., a fed-batch system, or, prefer¬ ably, by maintaining a substantially constant volume of fermentation broth by the addition of fresh fermentation medium and the removal of spent fermentation broth. In the latter case, the dilution rate (D) is equal to the volumetric flow rate of fresh fermentation medium into the fermentation vessel and spent fermentation broth out of the vessel divided by volume of the fermentation broth. The units for D are reciprocal time, and for present purposes will be hr^"1 unless indicated otherwise. Dilution is equal to the number of vessel liquid volumes which pass through the vessel per hour, and is the reciprocal of *nean residence time.

The addition of fresh fermentation medium and the removal of spent fermentation broth can be either constant or periodic, although it has been found that high growth rates and formation of riboflavin can be obtained with periodic addition and/or removal. The addition of ..resh fermentation medium and the removal of spent fermentation broth can either be simultaneous or not, but is preferably simultaneous.

The present method requires a dilution rate which iε less than the maximum growth rate (dx/dt^ma ) . In the case of maintaining a relatively constant volume of fer en- tation broth for the-growth of a typical culture, dilution is typically effective at between about 0.003 -hr^"1 to about 0.013 hr^"1, more preferably between about 0.006 hr^"1 to about 0.012 hr^"1 and most preferably between about 0.009 hr^"1 and about 0.011 hr^"1.

At steady state, D = dx/dt in the fermentation method of the present invention. Dilution rate (D) is dictated by the maximum growth rate (dx/dt^max) of a particular organiεm. Thiε can be determined by meaεuring dx/dt^max in a batch culture using a nutrient-rich medium (e.g., one having an excess of required nutrients) under optimal conditions, One may set the dilution rate in the culture at any level less than dx/dt^max. For optimal productivity, D is preferably from about 2% to about 20% of dx/dt^max, and more preferably is about 5% of dx/dt^max. The system will tend to be more unstable as dx/dt^max is approached. Elimination of all microorganisms from and cesεation of the fermentation occurs if D is gre_.ater than dx/dt.

It should be recognized that with the introduction of fresh fermentation medium, the nutrient levels in the fermentation broth must be maintained at concentrations sufficient to support growth and/or production of extra¬ cellular material. Such nutrients can be added either in the fresh fermentation medium or independently of it. Specifically, such nutrients include sourceε of carbon; nitrogen; phosphates; sulfates; and magnesium, iron, copper and other trace metals. It is known that iron is required in a fermentation medium in an amount sufficient to sustain microorganism growth. It is also known that higher concentrations of iron can repress riboflavin production. Therefore, iron must be maintained in the fermentation at an amount suf¬ ficient to enhance cell growth, but greater than an amount at which riboflavin production is repressed. The method of the present invention further includes maintaining the concentration of iron in the fermentation medium sufficient to enhance cell growth and to ensure continued riboflavin production. More particularly, the present invention includes providing an initial iron concentration in the fermentation medium and subsequently feeding iron during the fermentation to maintain sufficient iron con- centrations. The initial iron concentration is preferably between about .25 ppm and about 6 ppm, more preferably between about .75 ppm and about 4 ppm, and most preferably between about 1.5 ppm and about 2.5 ppm. The fermentation medium is subsequently fed iron at a rate sufficient to maintain cell growth and an iron concentration of at least about 1 parts per billion and more preferably, at least about 10 parts per billion.

The foregoing iron concentrations are particularly useful for microorganisms which are tolerant for iron. Such microorganisms include, for example, Candida famata identified by ATCC Nos. 20755, 20756, 20849 and 20850.

In another embodiment of the present invention, the pH of the fermentation medium is controlled. The pH is preferably maintained at a range from about 3.0 to about 5.5, more preferably from about 3.5 to about 4.5, and moεt preferably at about 4.0.

In the preεent invention, control of the pH of the fermentation medium has several advantages. First, the specific formation of riboflavin has been found to be increased. Second, cell growth is improved over that realized when pH is maintained at conventional pH ranges of about 5.0 to about 7.5. Third, bacterial contamination is prevented because the medium is too acidic to support bacterial growth.

The following experimental results are provided for purposes of illustration and are not intended to limit the scope of the invention.

Example 1 A serieε of three riboflavin fermentation production runs were conducted. The initial nutrient media components are identified below in Table 1. The glucose feed formulation is identified below in Table 2. The salt feed formulation is identified below in Table 3. The media was inoculated with an exponentially growing culture of the strain of Candida famata identified by ATCC No. 20849. The fermentation was conducted without additions until initial glucose concentrations were substantially depleted. The resultε of these fermentation runs, identified as R8822, R8901, R8908, are shown below in Table 4.

Table l. Media For t₀ Riboflavin Fermentation.

Table 2. Glucose Feed Formulation For Riboflavin Fermentation.

*Representative fermentation performed without glycine in the t- media. Measured NH₄ in the fermenter was 50% less than other fermentations reported. Initial higher O.D. (Optical Density.)

Example 2 Comparative fermentation runs were conducted to illustrate the elimination of arabitol and ethanol by restriction of glucose concentration. The fermentation media for both runs were identical and were the same as the fermentation media in Example 1, except for glucose concentrations. The fermentations were conducted with the strain of Candida famata identified by ATCC No. 20849. The results of the glucose restricted fermentation are shown in Figure 1 and the results of the control are shown in Figure 2. For both fermentations, the media contained 40 percent solids and 14 percent of the carbon source was derived from fructose and 86 percent of the carbon source was derived from glucose. In Figure 1, the concentration of glucose in the fermentation media was allowed to be depleted. In Figure 1, curve (2) plots the depletion of the glucose source over time; curve (4) plots the increase in the cell mass optical density at 620 nanometers over time; and curves (6) and (8) plot the reduction in the production of the toxins ethanol and arabitol, respectively, over time. In Figure 2, the concentration of glucose in the fermentation media was allowed to be substantially depleted and thereafter maintained at a concentration of 20 g/L. In Figure 2, curve (2) plots the increase in cell mass optical density at 620 nanometers; curves (4) and (6) plot the reduction in the production of the toxins arabitol and ethanol, respectively, over time; and curve (8) plots the utilization of the glucose source over time.

Example 3 A fermentation was conducted to identify cell growth and riboflavin production in a low glucose fermentation with no glycine.- The fermentation was conducted with a strain of Candida famata identifed by ATCC No. 20849. The fermentation conditions and media for this run are the same as those identified in Example 1, except that no glycine was present in any media. The results of this fermentation are identified in Figure 3. In Figure 3, curve (2) plots the increase in cell optical density over time and curve (4) plots the increase in the production of riboflavin over time. Example 4

A fermentation run was conducted to examine the effect of trace metals on riboflavin and cell optical density. The fermentation conditions and media were the same as those identified in Example 1, except for the control of trace metals. The results of this fermentation are identified in Figure 4 and the arrows (6) , (8) , (10) in Figure 4 indicate increases in the trace metal feed. In Figure 4, curve (2) plots the increase in cell optical density over time and curve (4) plots the increase in the production of riboflavin over time.

While various embodiments of the present invention have been described in detail, it is apparent that modifi¬ cations and adaptations of those embodiments will occur to those skilled in the art. However, it iε to be expreεsly underεtood that εuch modifications and adaptations are within the scope of the present invention, set forth in the following claims.

Claims

What Is Claimed Is;

1. A fermentation method for riboflavin-producing microorganisms comprising: a. maintaining a nutrient feed rate, wherein the nutrient uptake rate is restricted and the fermentation growth rate is not limited by such nutrient; and b. providing a nitrogen source selected from the group consisting of ammonium salts, ammonium hydroxide and ammonia gas.

2. A method, as claimed in Claim 1, wherein said nutrient is a carbon source.

3. A method, as claimed in Claim 2, wherein said carbon source is glucose.

4. A method, as claimed in Claim 3, wherein the feed rate of glucose in the fermentation medium is from about 0.5 grams per liter per hour to about 5.0 grams per liter per hour.

5. A method, as claimed in Claim 3, wherein Y _/s of the glucose is at least about 0.60.

6. A method, as claimed in Claim 1, wherein said ammonium salt is selected from the group consisting of ammonium sulfate, ammonium phosphate, ammonium carbonate and ammonium chloride.

7. A method, as claimed in Claim 1, wherein said ammonium salt is ammonium sulfate.

8. A method, as claimed in Claim 1, wherein said microorganiεmε are Candida famata.

-26-

9. A method, as claimed in Claim 8, wherein said Candida famata are selected from the group conεisting of ATCC 20755 ATCC 20756, ATCC 20849, ATCC 20850, mutants thereof, and mixtures thereof.

10. A method, as claimed in Claim 8, wherein said Candida famata have the identifying characteristic of ATCC 20849, or mutants thereof.

11. A method, as claimed in Claim 1, wherein said microorganismε produce riboflavin at a rate of at leaεt about 0.12 g/l/hr.

12. A method, as claimed in Claim 1, wherein said microorganismε produce riboflavin at a rate of at leaεt about 0.17 g/l/hr.

13. A method -^for the production of riboflavin, wherein the copper concentration in the fermentation medium aε referenced by CuS0.5H₂0, iε from about 0.66X10^"5 g/1 to about 3X10^"5 g/1.

14. A method, as claimed in Claim 13, wherein εaid concentration of copper is up to about a maximum non-toxic concentration.

15. A method, as claimed in Claim 13, wherein the specific formation of riboflavin for copper is at least about .1 gram of riboflavin per gram of cell.

16. A method, as claimed in Claim 13, wherein said copper is selected from the group consisting of copper salts and mixtures thereof.

17. A method, as claimed in Claim 13, wherein said copper is copper sulfate.

18. A method, as claimed in Claim 13, wherein said fermentation is conducted by Candida famata.

19. A method, as claimed in Claim 18, wherein said Candida famata are selected from the group consisting of ATCC 20755, ATCC 20756, ATCC 20849, ATCC 20850, mutants thereof and mixtures thereof.

20. A method, as claimed in Claim 18, wherein εaid Candida famata have the identifying characteristic of ATCC 20849, or mutants thereof.

21. A fermentation method for riboflavin-producing microorganisms comprising: a. heat sterilizing a nitrogen εource for a fermentation medium; and b. controlling the pH of the fermentation medium only with the addition of base.

22. A method, as claimed in Claim 21, wherein said nitrogen source is an ammonium salt.

23. A method, as claimed in Claim 22, wherein said ammonium εalt iε selected from the group consisting of ammonium sulfate, ammonium phosphate, ammonium carbonate and ammonium chloride.

24. A method, as claimed in Claim 22, wherein said ammonium salt is ammonium sulfate.

25. A method, as claimed in Claim 21, wherein said microorganisms are Candida famata.

26. A method, as claimed in Claim 25, wherein said Candida famata are selected from the group consiεting of ATCC 20755, ATCC 20756, ATCC 20849, ATCC 20850, mutants thereof, and mixtures thereof.

27. A method, as claimed in Claim 25, wherein said Candida famata have the identifying characteristic of ATCC 20849 or mutants thereof.

28. A continuous fermentationmethod for riboflavin- producing microorganisms comprising: a. diluting a fermentation broth comprising said microorganismε by adding fresh fermentation medium comprising a carbon source to said fermentation broth, wherein the carbon source uptake rate is restricted and the fermentation growth rate is not limited by the carbon source; b. providing an ammonium salt as a nitrogen source; c. providing a copper concentration in the fermentation medium, as referenced by CuS0₄'5H₂0, from about 0.66X10^"5 g/1 to about 3X10^"5 g/1; d. maintaining a concentration of iron in the medium between about .25 ppm and about 6 ppm; and e. maintaining a pH of the medium between about 3.0 and about 5.5.

29. A riboflavin-producing strain of microorganism, wherein said strain has been selected for decreased sensitivity to copper toxicity.

30. A strain of microorganism, as claimed in Claim 29, wherein said strain iε of the genuε and εpecieε Candida famata.

31. A strain of microorganism, as claimed in Claim 29, wherein said εtrain is selected from Candida famata having the identifying characteristic of ATCC 20849 and mutants thereof.

32. A fermentation method for riboflavin-producing microorganisms comprising: a. diluting a fermentation broth comprising said microorganisms by adding fresh fermentation medium comprising a nutrient to said fermentation broth, wherein the nutrient uptake rate is restricted and the fermen¬ tation growth rate is not limited by the nutrient; and b. providing an ammonium salt as a nitrogen source.

33. A method, as claimed in Claim 32, wherein the rate of dilution iε from about 0.003 hr^"1 to about 0.013 hr^"1.

34. A method, aε claimed in Claim 32, wherein the rate of dilution is from about 0.009 hr^"1 to about 0.011 hr^"1.

35. A method, as claimed in Claim 32, further compriεing removin- spent fermentation broth, wherein the volume of εaid ermentation broth iε maintained εubεtantially constant by maintaining substantially equal in-flow of fresh fermentation medium and out-flow of spent fermentation broth.

36. A method, as claimed in Claim 32, wherein the growth rate of said microorganisms is from about 2% to about 20% of the maximum growth rate.

37. A method, as claimed in Claim 32, wherein the growth rate of said microorganisms is about 5% of the maximum growth rate.

38. A method, as claimed in Claim 32, wherein said ammonium salt is εelected from the group conεiεting of ammonium sulfate, ammonium phosphate, ammonium carbonate and ammonium chloride.

39. A method, as claimed in Claim 32, wherein said ammonium salt is ammonium sulfate.

40. A method, as claimed in Claim 32, wherein said microorganisms are Candida famata.

41. A method, as claimed in Claim 40, wherein said Candida famata are εelected from the group conεisting of ATCC 20755, ATCC 20756, ATCC 20849, ATCC 20850, mutants thereof, and mixtures thereof.

42. A method, as claimed in Claim 40, wherein said Candida famata have the identifying characteriεtic of ATCC 20849, or mutants thereof.

43. A method, as claimed in Claim 32, wherein said nutrient is a carbon source.

44. A method, as claimed in Claim 43, wherein said carbon source is glucoεe.

45. A method, as claimed in Claim 44, wherein Y of the glucose is at leaεt about 0.060.

46. A method, as claimed in Claim 44, wherein the feed rate of glucose in the fermentation medium is from about 0.5 grams per liter per hour to about 5.0 grams per liter per hour.

47. A method as claimed in Claim 43, wherein the rate of addition of said carbon source is a minimum rate necessary to suεtain growth.

48. A method, as claimed in Claim 32, wherein said microorganismε produce riboflavin at a rate of at leaεt about 0.12 g/l/hr.

49. A method, aε claimed in Claim 32, wherein said microorganisms produce riboflavin at a rate of at least about 0.17 g/l/hr.

50. A method, as claimed in Claim 32, further comprising removing spent fermentation broth.

51. A method, as claimed in Claim 32, wherein said addition of freεh fermentation medium iε conεtant.

52. A method, as claimed in Claim 32, wherein said addition of fresh fermentation medium is periodic.

53. A method, as claimed in Claim 50, wherein εaid removal iε conεtant.

54. A method, aε claimed in Claim 50, wherein εaid removal iε periodic.

55. A method, as claimed in Claim 50, wherein said removal is simultaneous with said addition.