APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Dec. 1976, P. 769-776Copyright © 1976 American Society for Microbiology

Vol. 32, No. 6Printed in U.S.A.

Bacterial Fermentation of Cheese Whey for Production of aRuminant Feed Supplement Rich in Crude Protein1

C. A. REDDY,* H. E. HENDERSON, AND M. D. ERDMAN

Department ofMicrobiology and Public Health, Michigan State University, East Lansing, Michigan 48824 *-Cooperative Extension Service, Colorado State University, Fort Collins, Colorado 80523; and Department of

Animal Husbandry, Michigan State University, East Lansing, Michigan 48824

Received for publication 17 June 1976

A simple and efficient process for the production of a ruminant feed supple-ment, rich in crude protein (defined as total N x 6.25), by bacterial fermentationof cheese whey has been developed. The lactose in unpasteurized whey isfermented to lactic acid by Lactobacillus bulgaricus at a temperature of 43°Cand pH 5.5. The lactic acid produced is continually neutralized with ammonia toform ammonium lactate. The fermented product is concentrated by evaporationto a solids content of about 70% and adjusted to pH 6.8 with additional ammonia.The concentrated product contains about 55% crude protein. Approximately 6 to8% of the crude protein is derived from bacterial cells, 17% from whey proteins,and 75 to 77% from ammonium lactate. The efficiency of conversion of lactose tolactic acid usually exceeds 95%. The fermentation time is greatly reduced uponthe addition of 0.2% yeast extract or 0.1% corn steep liquor as a source of growthfactors. Whey containing lactose at concentrations up to 7% can be fermentedefficiently, but at higher concentrations lactose is fermented incompletely. Theprocess has been scaled up to a pilot plant level, and 40 tons of concentratedproduct were produced for animal feeding trials, without ever encounteringputrefactive spoilage.

Whey is a lactose-rich, watery by-product ofthe cheese manufacturing process (42). Nearly,1.36 x 10'° kg of whey is produced annually inthe United States (20). Out of this, only 56% ofthe whey solids are currently processed intoanimal and/or human food products (20).

Several problems are encountered in dispos-ing or reutilizing whey. It is uneconomical totransport whey because of its high water con-tent. It is readily subject to bacterial and fungalspoilage and cannot be stored for any length oftime unless special precautions are taken.Drying whey requires a large capital invest-ment, is energy intensive, and is not economi-cally profitable (6, 28). Disposal of whey withinfederal environmental standards is expensivebecause of its high biological oxygen demand.Besides, whey contains large amounts of poten-tially recyclable nutrients. It is estimated thatdiscarded whey contains about 3.6 x 108 kg oflactose and 6.9 x 107 kg of protein. Thus, thereis an obvious need for finding a solution to thewhey disposal problem and for recycling thenutrients in whey as feed and/or food material.A number of processes have previously been

described for the batch bacterial fermentation

' Article no. 7729, Michigan Agricultural ExperimentalStation.

of lactose in cheese whey and production oflactic acid (36, 43; H. C. Jansen, Dutch Patent57848, 1945; A. H. Johnson, S. M. Weisberg, J.J. Johnson, and M. E. Parker, U.S. Patent2071346, 1937; S. M. Weisberg, F. L. Chappell,E. Stringer, S. Stevens, and H. A. Trebler,U.S. Patent 2071368, 1937) or microbial protein(4, 6). A continuous fermentation process forlactic acid production from cheese whey wasalso described (44). Further, the production ofammonium lactate, as a source ofcrude protein(defined as total N x 6.25) for cattle, by fermen-tation of cheese whey or other materials rich incarbohydrates has been described (2; E. J. Czar-netzky, U.S. Patent 2,094,437, 1959; Jansen,Dutch Patent 57848, 1945; L. H. Perquin, DutchPatent 58545, 1946). Several investigators haveshown that ammonium salts of short-chain or-ganic acids, including ammonium lactate, pro-duced either synthetically or by fermentationwere superior to urea and equivalent to soy-bean meal as nitrogen supplements to rumi-nants (1, 5, 17, 22-24, 39).The previous methods for the production of

lactic acid, salts of lactic acid, or SCP (singlecell protein), by fermentation of cheese whey,suffered from one or more of the following dis-advantages: (i) there was a long lag period priorto active lactic acid fermentation and/or unus-

769

on May 23, 2021 by guest

http://aem.asm

.org/D

ownloaded from

770 REDDY, HENDERSON, AND ERDMAN

ually long fermentation times that requiredgreater fermentor capacity and increased op-erational costs (4, 36; Czarnetzky, U.S. Patent2,094,437, 1959); (ii) manual procedures usedfor the addition of ammonia or calcium ion toneutralize the lactic acid produced during thefermentation were tedious and contributed toincreased product variability (18, 36; Czarnet-zky, U.S. Patent 2,094,437, 1959; Jansen, DutchPatent 57848, 1948; Perquin, Dutch Patent58545, 1946); (iii) large levels of inocula (10 to20% by volume) were required to obtain vigor-ous fermentation and, hence, were less eco-nomical (2, 4, 6, 36, 43; Czarnetzky, U.S. Patent2,094,437, 1959; Jansen, Dutch Patent 57848,1945); (iv) putrefactive spoilage of the fermen-tations was encountered (2); (v) the crude pro-tein content of the product was relatively low(4, 6; Czarnetzky, U.S. Patent 2,094,437, 1959;Perquin, Dutch Patent 58545, 1946); (vi) rela-tively high costs were involved in the pastueri-zation or sterilization of whey prior to fermen-tations (2, 6, 36; Czarnetzky, U.S. Patent2,094,437, 1959); (vii) there were high aerationcosts during fermentation (4, 6, 32); and (viii)there were palatability problems for the ani-mals with the product that resulted in de-creased dry matter intake (21).The objectives of this study were twofold: (i)

to develop a simple, efficient, and nonasepticfermentation process for the production of anitrogenous feed supplement, for ruminant ani-mals, from cheese whey and (ii) to investigatethe feasibility of producing such a product on apilot plant scale.

(This paper was presented in part at the 75thAnnual Meeting of the American Society forMicrobiology, 27 April-2 May 1975, New York.)

MATERIALS AND METHODSBacterial culture. After a careful examination of

a large number ofLactobacillus strains representingseveral different species, L. bulgaricus strain 2217(supplied by Robert Sellars, Chris Hansen Labora-tories, Milwaukee, Wis.) was used in this studybecause of its high rate of acid production in the pHrange 5.0 to 6.0, greater acid tolerance, and shortgeneration time. This organism was routinely main-tained in sterile skim milk medium (SSMM), 10%(wt/vol) skim milk powder in distilled water,and autoclaved at 121°C for 15 min. About 0.1 ml ofthe culture was transferred once every 4 weeks to 10ml of SSMM in a screw-cap tube. Inoculated tubeswere incubated at 440C for 24 h and stored at 40Cuntil further use.

Bench-top fermentor. The inoculum for a 14-literbench-top fermentor (model Ma 140 Fl, Fermenta-tion Design, Allentown, Pa.) was prepared by trans-ferring 0.1 ml of a 24-h culture of SSMM-grown L.bulgaricus to 10 ml of the same medium, in a screw-cap tube, and incubating for 12 h at 440C. Coagula-

APPL. ENVIRON. MICROBIOL.

tion of SSMM occurred in 8 to 10 h as a result ofbacterial growth and lactic acid production. About0.1 ml of this culture was added to another tube ofSSMM and incubated as above. The contents of thesecond tube were transferred to 500 ml ofSSMM in a1-liter foam-plugged Erlenmeyer flask. After incu-bation, the contents of the flask were used to inocu-late 9.5 liters of whey in the fermentor.

All fermentations were conducted as follows.Nonsterile, reconstituted whey (spray-dried, cottagecheese whey powder [Michigan Milk Producers,Ovid, Mich.], reconstituted with distilled water togive a lactose concentration of approximately 5.0%[wt/vol]) was added to the fermentor. Unless speci-fied otherwise, sterile corn steep liquor (CSL; CornProducts Corp. International, Argo, Ill.) was addedto the whey (0.1% final concentration) as a source ofsupplemental growth factors for L. bulgaricus. Themedium was continuously agitated at 150 rpm. Thetemperature was adjusted and maintained at 43 +0.5°C. The pH was adjusted to and maintained at 5.5+ 0.1 by an automatic pH recording controller(model PHRT, Fermentation Design, or model pH22, New Brunswick Scientific Co., New Brunswick,N.J.). Either anhydrous (99.9%; Matheson GasProducts, Chicago, Ill.) or aqueous (27% in water;Carrier Stephens Co., Holt, Mich.) ammonia wasused as the neutralizing agent. Inoculum was addedat the 5% level unless otherwise specified. No asep-tic precautions were taken for adding the inoculum,collection of samples, addition of growth factors, orother manipulations. Fermentation was terminatedat the end of 16 to 24 h. Samples (-40 ml) werecollected in screw-cap glass sampling bottles justbefore adding the inoculum, at selected intervalsduring the fermentation, and at the end of fermen-tation and stored at -18°C until analyzed. All sam-ples were analyzed for lactose, lactic acid, total ni-trogen, and ammonia nitrogen.

Pilot plant fermentor. All pilot plant scale fer-mentations were conducted in a 2,270-liter stainless-steel fermentor (model FS-3, Bernard and Leas, Ce-dar Rapids, Iowa) operated at a 1,940-liter capacity.The temperature in the fermentor was controlledwith an ACCO-Bristol. temperature controller(model 624-IL, ACCO-Bristol Div., Waterbury,Conn.), and the pH was continually measured andcontrolled at a set point by an electrode (model 764-31, Ingold Electrodes Inc., Lexington, Mass.) fittedinto the side of the fermentor and connected to anautomatic pH recording controller described above.

All fermentations using a 2,270-liter fermentorwere conducted as follows. About 1,890 liters offreshcottage cheese whey or sweet whey was added to thefementor, agitation was provided, and 20 liters ofCSL was added. The temperature and pH were ad-justed to and maintained at 43 + 0.5°C and 5.5,respectively, as described above. The inoculum wasprepared by transferring 10.0 ml of a 12-h-old cul-ture ofL. bulgaricus, grown in SSMM, to 2 liters ofthe same medium in a 3-liter foam-plugged Florenceflask. After incubation for 12 h, the contents of theflask were transferred to 45 liters of commerciallyavailable pasteurized skim milk, and after 12 h ofincubation the latter served as the inoculum. Imme-diately after adding the inoculum, a sample was

on May 23, 2021 by guest

http://aem.asm

.org/D

ownloaded from

PROTEIN FEED SUPPLEMENT FROM WHEY 771

collected. Fermentation was terminated after 24 h,another sample was collected, and the fermentedammoniated whey (FAW) was concentrated 10-foldby using a 378-liter, single-effect, rising film evapo-

rator (Rogers Co., Detroit, Mich.). The pH of theconcentrated product was adjusted to 6.8 with addi-tional ammonia to further increase the crude pro-

tein level in the product. This final product, referredto as fermented, ammoniated, condensed whey(FACW), was stored in 3,785-liter steel tanks untilfurther use.

Analytical procedures. Total nitrogen was deter-mined by the micro-Kjeldahl procedure (27). Lactosewas determined by the procedure of Dubois et al.(16) as modified by Montgomery (33). Ammonia ni-trogen was determined by Nesslerization (34) and byConway microdiffusion analysis (11). Lactic acidwas quantitatively determined by the procedure ofBarker and Summerson (3) as modified by Davidson(15). Metabolic end products produced during thefermentation were qualitatively analyzed on a gaschromatograph (Dohrmann Anaerobic BacteriologyAnalyzer, Clinical Analysis Products Co., Sunny-vale, Calif.), using the procedures described byHoldeman and Moore (25). Optical rotation of thelactic acid produced during the fermentation was

determined by the procedure of Cato and Moore (10).The percentage of solids in FACW was determinedby the procedure of Hood et al. (26) or by air dryingin a forced-air oven at 60°C for 48 h.

RESULTS

Rate study. Decreases in lactose concentra-tion and increases in concentrations of lacticacid and ammonia during the fermentation are

shown in Fig. 1. Lactose was rapidly metabo-lized and in many cases fermentation was es-

sentially complete within 14 to 16 h. About 95 to98% of the lactose was metabolized during thefermentation. As the lactose concentration de-creased, there was a proportional rise in theconcentration of lactic acid. Lactic acid was theonly.major acid produced during the fermenta-tion, suggesting that the lactose was metabo-lized by a homolactic acid type of fermentation.

6

2.5.0040

uO)3

i 2.0

For each gram of lactose fermented, about 0.189g of NH3 was required to neutralize the lacticacid produced, and the ammonium ion concen-

tration curve essentially paralleled the lacticacid concentration curve. Therefore, the prog-

ress of the fermentation could be monitoredequally efficiently by measuring either the de-crease in concentration of lactose or the in-crease in concentration of lactic acid or ammo-

nium ion.Effect of pH. As shown in Fig. 2, the rate of

utilization of lactose appeared to be better atpH 5.5 than at 5.0, 5.7, 5.9, and 6.1, especiallyin the first 16 h of fermentation. However, notethat at the end of 24 h the residual lactoseconcentration is the same regardless ofwhetherthe fermentation was conducted at pH 5.5 or

5.0. At pH 7.0 there was little utilization oflactose, even at the end of 24 h.Inoculum size. As shown in Fig. 3, rates of

lactose utilization and ammonia uptake were

similar with 10 and 20% inocula. With a 1%inoculum the rate of fermentation was consid-erably lower, especially in the first 12 h of

TIME (hours)

FIG. 2. Effect of pH on the rate of utilization oflactose. Symbols: 0, pH 5.0; *, pH 5.5; A, pH 5.7;0, pH 5.9; *, pH 6.1; A, pH 7.0.

1.0 ::.

z

Q26

Q4a

QL2

0 2 4 6 8 10 12 14 16 4

TIME(hOUrs)

FIG. 1. Lactose utilization (O), lactic acid produc-tion (-), and ammonia uptake (A) during the fer-mentation of whey by L. bulgaricus at 430C and pH5.5.

TIME (hours)

FIG. 3. Effect of inoculum level on lactose utiliza-tion and ammonia uptake. Inoculum levels: *, 1 %;0, 10%; 0, 20%.

Q

-ao~00

1.0=K:~~~~~~~~IE

VOL. 32, 1977

1.C

D

D

on May 23, 2021 by guest

http://aem.asm

.org/D

ownloaded from

772 REDDY, HENDERSON, AND ERDMAN

fermentation. However, note that at the end of24 h the residual lactose concentration and theammonia concentration were comparable withall three levels of inocula. We have obtainedadditional data which showed that 2.5 and 5%inocula gave as fast a rate of fermentation asthose obtained with 10 and 20% inocula. Noputrefactive spoilage was encountered in any ofthe fermentations.

Effect of addition of growth factors. Lacto-bacilli are nutritionally fastidious organismsand are known to require a number of growthfactors (7, 32, 38). The effect of adding differentsources of growth factors on whey fermenta-tion, as measured by the rate of lactose utiliza-tion, was investigated (Fig. 4). Higher rates offermentation were obtained when 0.1% CSL or0.2% yeast extract (Difco) was added. At about10 h after the initiation of the fermentation,control whey medium (without any growth fac-tors added) contained 3.4% lactose, whereas thesame medium supplemented with 0.1% CSL or0.2% yeast extract contained only 1.24 and1.7% lactose, respectively. Supplementationwith 0.1% yeast extract also resulted in an ap-preciably higher rate of fermentation. Trypti-case (BBL), Casamino Acids (Difco), or maltextract (Difco), when added to whey at a 0.1%level (data not shown), had no stimulatory ef-fect on fermentation, especially in the first 16 hof fermentation. At the end of 24 h, however,the differences in residual lactose concentra-tions were less marked.

Considering the stimulatory effect of CSL onfermentation and the fact that it is less expen-sive than yeast extract, the effect of differentlevels of CSL on fermentation was determined.As shown in Fig. 5, supplementation with 0.1%CSL appears to give a higher initial rate offermentation, whereas with 0.4, 0.6, and 1%

TIME (hours)

FIG. 4. Effect of the addition of different growthfactor sources on fermentation. Symbols: A, 0.1%CSL; 0, 0.1% yeast extract; 0, 0.2% yeast extract; *,control with no growth factor.

0

-0J

TIME (hours)

FIG. 5. Effect of different levels ofCSL on lactoseutilization. Levels of CSL: A, no CSL; 0, 0.1%lo; *,0.4%o; A, 0.6%o; and M, 1.0%.

levels of CSL, the rate of utilization of lactosewas somewhat lower. This suggested thathigher levels of CSL contained a compound(s)at concentrations high enough to inhibit fer-mentation. The effect of CSL at levels lowerthan 0.1% has not been tested. Interestingly,the residual lactose levels at the end of 16 hwere comparable in all fermentations, irrespec-tive of the level of CSL added.

In other experiments a combination of 0.2%yeast extract and 0.1% CSL with or withoutmineral solution (composition and percentageas in reference 8) added to the whey mediumdid not increase the rate of fermentation anyhigher than that obtained with either yeastextract or CSL added individually.

Effect of initial lactose concentration. Al-though whey normally contains about 5% lac-tose, higher lactose whey can be obtained byconcentrating the raw whey as it comes fromthe cheese vat. If such concentrated whey couldbe fermented efficiently, transportation .andhandling costs, the size ofthe fermentors, evap-oration costs, and the cost per ton of FACWcould be significantly reduced than when wheycontaining 5% lactose was used. Therefore, theeffect of initial lactose concentration on fermen-tation was investigated. Figure 6 shows that atinitial lactose concentrations up to 7%, the fer-mentation proceeded well, as determined bythe rate of ammonia input with time. When11.5% lactose whey was fermented, ammoniaconsumption during the fermentation was nogreater than that observed with 7% lactosewhey. An analysis of the lactose levels at theend of fermentation showed that only traceamounts of residual lactose (0.05 to 0.10%) werepresent when 3, 5.4, or 7% lactose whey wasfermented, whereas the residual lactose levelwas close to 4% when whey with an initial

APPL. ENVIRON. MICROBIOL.

on May 23, 2021 by guest

http://aem.asm

.org/D

ownloaded from

PROTEIN FEED SUPPLEMENT FROM WHEY 773

6 8 10 1TIME(hours)

FIG. 6. Effect of initial lactose concentration on

fermentation. Initial lactose concentrations: 0, 3.0%;*, 5.4%; *, 7.0%; and A, 11.5%.

lactose concentration of 11.5% was fermented.These results suggested that lactose levels up to7% in whey could be fermented efficiently andthat lactose levels of 11.5% were fermented in-completely.The incomplete fermentation observed with

11.5% lactose whey may be due, either whollyor in part, to the buildup of undissociated lacticacid and/or inhibition by the ammonium ionused for neutralizing the lactic acid. To find outmore about the nature of this inhibition, we didan experiment in which ammonium lactate,sodium lactate, or calcium lactate was added tonormal whey (containing 5% lactose) to give a

final concentration of5% (wt/vol) added lactateion. The medium was inoculated as before, andthe rate of utilization of lactose during the fer-mentation was measured. There was completeinhibition of fermentation as evidenced by thelack of utilization of lactose. In the control fer-mentor, without any added lactate, the fermen-tation proceeded normally. These results sug-gested that, irrespective of the cation involved,exogenous lactate, added at a 5% level at thebeginning of the fermentation, completely in-hibited whey fermentation. However, therewas a distinct possibility that different cations(Ca2+, Na+, or NH4+) added in the above experi-ment may also have contributed to the inhibi-tions. Therefore, another experiment was con-

ducted in which the amount of cation requiredto neutralize the 5% level of lactate ion was

calculated and was added to the fermentor inthe form of ammonium sulfate, sodium sulfate,or calcium sulfate. We found no significant in-hibition of fermentation with any of the threecations. This again suggested that it was lacticacid and not the concentration of cation thatplayed a primary role in inhibiting the fermen-tation when whey with a high lactose content

was fermented. However, we do realize thatthese results are not conclusive, and it is possi-ble that other interpretations may be equallyvalid.

Large-scale fermentations. The basic infor-mation obtained from the above series of exper-iments was utilized in scaling up fermentationsto a pilot plant level. FACW, described above,was a thick, tan-colored, syrupy liquid and hada mild dairy flavor and salty taste. The compo-

sitions of unfermented whey, FAW, and FACWare given in Table 1. A comparison of normalwhey and FACW shows that there is a 50- to 60-fold increase in crude protein and about a 100-fold increase in lactic acid content in FACW as

compared with unfermented whey. Approxi-mately 75 to 77% of the crude protein in FACWwas in the form of ammonium lactate, 17% was

in the form of whey proteins (lactalbumin andlactoglobulins), and the remaining 6 to 8% was

contributed by the bacterial cells. Lactic acidwas generally the only detectable acid producedduring the fermentation. However, traces ofacetic acid were occasionally observed. About54% of the lactic acid in FACW is D(-)-lacticacid and 46% is L(+)-lactic acid. FACW is sta-ble and resistant to microbial spoilage at tem-peratures of 4, 25, 37, and 60°C for more than 12months.

DISCUSSIONThe results show that the batch fermentation

of lactose in whey could be carried to near

completion in 24 h or less by using L. bulgari-cus. This is a considerable improvement over

previous whey fermentation methodology,which in most cases required 42 to 72 h for thecomplete fermentation of lactose to lactic acid(4, 36; Czarnetzky, U.S. Patent 2,094,437, 1959).Also, the fermentation process described here israther unique in that it is a means to an endrather than an end in itself; that is, lactose isconverted to lactic acid primarily as a means totrap ammonium ion and thereby enrich thecrude protein level in the product.

TABTz 1. Composition of whey, FAW, and FACW

Component Whey (gl100 g)

FAW (g/100 g)

FACW (gl100 g)

Lactose 5.0 0.10 1.0Lactic acid 0.5 5.10 49.0Ammonia nitro- 0.05 0.69 6.6gen

Total nitrogen 0.17 0.90 8.8Crude protein 1.1 5.63 55.0

equivalentaa Crude protein equivalent is defined as total N x

6.25.

VOL. 32, 1977

on May 23, 2021 by guest

http://aem.asm

.org/D

ownloaded from

774 REDDY, HENDERSON, AND ERDMAN

It was previously reported that by using re-strictive fermentation temperatures of 43 to440C, pH 5.5, and acid-tolerant organisms suchas L. bulgaricus, unsterilized and unpasteurizedwhey could be fermented to produce lactic acidwithout any putrefactive spoilage (36, 44; Weis-berg et al., U.S. Patent 2071368, 1937). This wasthe case in this study where several hundredbatches ofwhey were fermented both in labora-tory and pilot scale fermentors without encoun-tering putrefactive spoilage. The lack of therequirement for sterilization or pasteurizationof whey would reduce the operational costs in-volved in any industrial scale fermentation.Our results indicate that the fermentation of

whey proceeds at a faster rate at pH 5.5 but lessefficiently at pH 5.7 to 6.1 or at 5.0. Our resultsare, thus, practically identical to those ob-tained by Finn et al. (18) for the fermentation ofglucose by lactobacilli. These workers observedmaximum bacterial densities at a pH of 5.2 to5.6 and reduced bacterial densities at a higheror lower pH level. Luedeking and Piret (30)also obtained essentially similar results withbatch fermentations of glucose conducted be-tween pH 5.2 to 6.0. He reported a significantinhibition offermentation at pH 4.8. In contrastwith the above results, Kempe et al. (29) ob-served a pH optimum of 4.6 for the productionof lactic acid from wheat grit mashes supple-mented with nutrient salts and yeast extract.They found that the conversion of sugar to lac-tic acid varied inversely with the pH of the fer-mentation. More recently, Keller and Gerhardt(28) observed lower residual lactose concentra-tions when the continuous fermentation ofwhey was conducted at pH 6.0 instead ofpH 5.5.It is somewhat difficult to reconcile the dif-ferences in results obtained by different investi-gators, and we have no clear explanation forthese discrepancies at this time.To have successful fermentation with unster-

ilized carbohydrate-rich substrates, the inocu-lum size should be high enough to producerapid fermentation and thus reduce the chancesof contamination by undesirable bacteria. Atthe same time, from an economic standpoint,the inoculum should be small enough so as notto significantly increase the operational costs.In this study we found that a 2.5% inoculumproduced just as rapid and contamination-freefermentation as did that obtained with a 10%inoculum. This is a considerably smaller-sizedinoculum than that used by previous investiga-tors (2, 4, 6, 36, 43; Czarnetzky, U. S. Patent2,094,437, 1959; Jansen, Dutch Patent 57848,1945) and is one of the strengths of this process.The results indicate that high-lactose whey is

fermented incompletely under the particular

APPL. ENVIRON. MICROBIOL.

set of experimental conditions used in thisstudy. Our results suggest that the high con-centration of lactic acid produced during thefermentation may be a primary factor responsi-ble for the incomplete fermentation observedwith 11% lactose whey. Whittier and Rogers(44) and others (19, 28, 37) have previouslyindicated that in a high-lactose whey medium,the concentration of undissociated lactic acid isthe primary factor limiting the fermentation.In fact, it has been shown recently (Gerhardt,personal communication) that whey containinglactose concentrations up to 15 to 20% could befermented effilciently if the lactate producedduring the fermentation could be removed bydialysis. It must be emphasized, however, thatfurther work needs to be done to understandmore clearly the factor(s) limiting the fermen-tation of high-lactose whey.To assess the value of FACW as a protein

feed supplement to ruminant animals, exten-sive toxicity and feeding trials have been con-ducted with feedlot steers (13, 22-24, 37). Re-sults of feeding trials showed that there are nopalatability problems with FACW and therewas no decrease in daily dry matter consump-tion, even when FACW was fed to the animalsat two times the daily requirement for supple-mental protein. Growth response and feed effi-ciency were superior to urea and comparable tosoybean meal. Feedlot steers can consume atleast twice as much FACW nitrogen as ureanitrogen before reaching toxic levels. More re-cently FACW has been shown to be equivalentto soybean meal as a supplemental nitrogensource for dairy cows (J. T. Huber, unpublisheddata). A preliminary cost analysis indicatedthat FACW could favorably compete with soy-bean meal as a cattle feed supplement, and theprocess is being commercially adopted now.The process described has several obvious ad-

vantages. It is a simple and economical processfor the fermentative recycling of waste wheyinto a nitrogenous feed supplement. There areno wastes generated in this process. The prod-uct is stable and nontoxic, thus eliminating theneed for complicated storage equipment andprefeeding conditioning periods that are re-quired for other nonprotein nitrogen productssuch as urea. The substrate used does not com-pete with man's foodchain. The product con-tains bacterial protein and milk proteins that,when separated from ammonium lactate, havea potential as a protein source for humans.

ACKNOWLEDGMENTSWe wish to thank D. Bergman, C. P. Cornell, S. Cuppett,

and P. Whetter for technical assistance, G. Thompson, Uni-versity of Minnesota, for determining the water content in

on May 23, 2021 by guest

http://aem.asm

.org/D

ownloaded from

PROTEIN FEED SUPPLEMENT FROM WHEY 775

the concentrated product, Robert Sellars for supplying theL. bulgaricus cultures, T. Hedrick, Department of FoodScience and Human Nutrition, for allowing the use of theevaporator, Michigan Milk Producers Association, Ovid,Mich., for providing whey, and Ruminant Nitrogen Prod-ucts, Adrian, Mich., for donating the CSL.

LITERATURE CITED1. Allen, C. K., and H. E. Henderson. 1972. Ammonium

salts as a source of crude protein for feedlot cattle, p.5-17. In Report of beef cattle research. MichiganState University Agricultural Experimental Stationreport 174, East Lansing, Mich.

2. Arnott, D. R., S. Patton, and E. M. Kesler. 1958. Amethod for manufacturing a high-nitrogen low-lac-tose product from whey. J. Dairy Sci. 41:931-941.

3. Barker, S. B., and W. H. Summerson. 1941. The colori-metric determination of lactic acid in biological mate-rial. J. Biol. Chem. 138:535-554.

4. Bechtle, R. M., and T. J. Claydon. 1971. Acceleratedfermentation of cheese whey. Developing the system.J. Dairy Sci. 54:1959-1604.

5. Belasco, I. J. 1954. New nitrogen feed compounds forruminants. A laboratory evaluation. J. Anim. Sci.13:601-610.

6. Bernstein, D., and T. C. Everson. 1973. Protein produc-tion from acid whey via fermentation, p. 103-113. InFood processing waste management. Cornell Agricul-tural Waste Management Conference, Cornell, N.Y.

7. Buchanan, R. E., and N. E. Gibbons (ed.). 1974. Ber-gey's manual of determinative bacteriology, 8th ed.,p. 576-593. Williams and Wilkins Co., Baltimore,Md.

8. Caldwell, D. R., and M. P. Bryant. 1966. Medium with-out rumen fluid for nonselective enumeration andisolation of rumen bacteria. Appl. Microbiol. 14:794-801.

9. Casida, L. E., Jr. 1968. Anaerobic fermentations, p.304-314. In L. E. Casida, Jr. (ed.), Industrial microbi-ology. John Wiley and Sons, Inc., New York.

10. Cato, E. P., and W. E. C. Moore. 1965. A routinedetermination of optically active isomers of lactic acidfor bacterial classification. Can. J. Microbiol. 11:319-324.

11. Conway, E. J. 1963. Microdiffusion analysis and volu-metric error. Chemical Publishing Co., Inc., NewYork.

12. Cordes, W. A. (ed.). 1967. Manual for milk plant opera-tors, 3rd ed., p. 194-221. Milk Industry Foundation.

13. Crickenberger, R. G., H. E. Henderson, and C. A.Reddy. 1975. Effect of feeding elevated levels of fer-mented, ammoniated, condensed whey (FACW) ondaily dry matter consumption of steer calves, p. 70-75. In Report of beef cattle forage research. MichiganState University Agricultural Experimental Stationreport 288, East Lansing, Mich.

14. Crickenberger, R. G., H. E. Henderson, and C. A.Reddy. 1975. Toxic effects of FACW, NH4 acetate,NH4 lactate and urea when infused intraruminally,p. 76-91. In Report of beef cattle forage research.Michigan State University Agricultural Experimen-tal Station report 288, East Lansing, Mich.

15. Davidson, J. 1959. The colorimetric determination oflactic acid in milk and milk products. J. Dairy Res.16:209-216.

16. Dubois, M., K. A. Gilles, J. K. Hamilton, P. A. Rebers,and R. Smith. 1956. Colorimetric method for determi-nation of sugars and related substances. Anal. Chem.28:350-356.

17. Dutrow, N. A., J. T. Huber, and H. E. Henderson. 1974.Comparison of ammonium salts and urea in rationsfor lactating dairy cows. J. Anim. Sci. 38:1304-1308.

18. Finn, R. K., H. 0. Halvorson, and E. L. Piret. 1950.Lactic acid fermentation rate. Effect of continuously

controlled pH. Ind. Eng-Chem. 42:1857-1861.19. Friedman, M. R., and E. L. Gaden, Jr. 1970. Growth

and acid production by Lactobacillus delbrueckii in adialysis culture system. Biotechnol. Bioeng. 12:961-974.

20. Gridstaff, D. A. 1974. Processing modified whey prod-ucts, p. 60-66. In Proc. Whey Products ConferenceERRC no. 3996. U.S. Department of Agriculture,Philadelphia, Pa.

21. Hazzard, D. G., E. M. Kesler, D. R. Arnott, and S.Patton. 1958. Use of high-nitrogen feedstuff madefrom whey in the rations of dairy animals. J. DairySci. 41:1439-1445.

22. Henderson, H. E., R. G. Crickenberger, and C. A.Reddy. 1975. Fermented ammoniated, condensedwhey as a source ofcrude protein for feed-lot cattle, p.161-170. In Report of beef cattle forage research.Michigan State University Agricultural Experimen-tal Station research report 288, East Lansing, Mich.

23. Henderson, H. E., R. G. Crickenberger, C. A. Reddy,and E. Rossman. 1974. Fermented, ammoniated con-densed whey and high wax shelled corn for feed-lotcattle, p. 1-13. In Report of beef cattle research. Mich-igan State University Agricultural ExperimentalStation research report 245, East Lansing, Mich.

24. Henderson, H. E., R. G. Crickenberger, C. A. Reddy,and E. Rossman. 1974. Sources of crude protein andsulfur for feed-lot cattle, p. 14-23. In Report of beefcattle research. Michigan State University Agricul-tural Experimental Station research report 245, EastLansing, Mich.

25. Holdeman, L. V., and W. E. C. Moore. 1973. Anaerobelaboratory manual, 2nd ed., p. 113-115. VirginiaPolytechnic Institute and State University, Blacks-burg, Va.

26. Hood, R. L., C. E. Allen, R. D. Goodrich, and S. C.Meiske. 1971. A rapid method for the direct chemicaldetermination of water in fermented feeds. J. Anim.Sci. 33:1310-1314.

27. Horwitz, W. (ed.). 1975. Official methods of analysis ofthe Association of Official Agricultural Chemists,12th ed., p. 927-928. Association of Official Agricul-tural Chemists, Washington, D.C.

28. Keller, A. K., and P. Gerhardt. 1975. Continuous lacticacid fermentation ofwhey to produce a ruminant feedsupplement high in crude protein. Biotechnol.Bioeng. 17:997-1018.

29. Kempe, L. L., H. 0. Halvorson, and E. L. Piret. 1950.Effect of continuously controlled pH on lactic acidfermentation. Yield and conversion. Ind. Eng. Chem.42:1852-1857.

30. Luedeking, R. L., and E. L. Piret. 1959. A kinetic studyof the lactic acid fermentation. Batch process at con-trolled pH. J. Biochem. Microbiol. Technol. Eng.1:393-412.

31. McCullough, M. E., W. E. Neville, Jr., and W. J. Mon-son. 1972. Ammoniated whey as an ingredient incomplete livestock rations. Feedstuffs 27:233-235.

32. Marth, E. H. 1970. Fermented products from whey, p.43-81. In B. H. Webb and E. 0. Whittier (ed.), By-products from milk. Avi Publishing Co., Westport,Conn.

33. Montgomery, R. 1961. Further studies of the phenol-sulfuric acid reagent for carbohydrates. Biochim. Bio-phys. Acta 488:591-593.

34. Oser, B. L. (ed.). 1965. Urine: quantitative analysis,p. 1218-1221. In Hawk's physiological chemistry,14th ed. The Blakiston Division, McGraw-Hill BookCo., New York.

35. Peppler, H. J. 1967. Ethyl alcohol, lactic acid, acetone-butyl alcohol and other microbial products, p. 403-416. In H. J. Peppler (ed.), Microbial technology.Reinhold Publishing Corp., New York.

36. Pmrscott, S. C., and C. G. Dunn. 1959. The production of

VOL. 32, 1977

on May 23, 2021 by guest

http://aem.asm

.org/D

ownloaded from

776 REDDY, HENDERSON, AND ERDMAN

lactic acid by fermentation, p. 304-331. In Industrialmicrobiology, 3rd ed. McGraw-Hill Book Co., Inc.,New York.

37. Rogers, L. A., and W. 0. Whittier. 1928. Limitingfactors in the lactic fermentation. J. Bacteriol.16:211-219.

38. Tittsler, R. P., C. S. Pederson, E. E. Snell, D. Hendlin,and C. F. Niven, J. R. 1952. Symposium on the lacticacid bacteria. Bacteriol. Rev. 16:227-260.

39. Varner, L. W., and W. Woods. 1971. Influence ofammo-nium salts of volatile fatty acids upon ration digesti-bility, rumen fermentation and nitrogen retention bysteers. J. Anim. Sci. 33:110-117.

40. Ward, K. W. 1970. Industrial processing of whey today,

APPL. ENVIRON. MICROBIOL.

p. 30-35. In Proc. Whey Utilization Conference, ARS-73-69. U.S. Department of Agriculture, Philadelphia,Pa.

41. Webb, B. H. 1970. Utilization ofwhey in food and feeds,p. 102-111. In Proc. Whey Utilization Conference,ARS-73-69. U.S. Department of Agriculture, Phila-delphia, Pa.

42. Webb, B. H., and E. 0. Whittier. 1948. The utilizationof whey: a review. J. Dairy Sci. 32:139-164.

43. Whittier, E. 0. 1944. Lactose and its utilization: a re-view. J. Dairy Sci. 27:505-537.

44. Whittier, E. O., and L. A. Rogers. 1931. Continuousfermentation in the production of lactic acid. Ind.Eng. Chem. 23:532-534.

on May 23, 2021 by guest

http://aem.asm

.org/D

ownloaded from