Changes in the Paper Electrophoretic Protein Patterns ofRefrigerated Skim Milk Accompanying Growth of Three

Pseudomonas Species

J. D. SKEAN AND W. W. OVERCAST

Departnment of Dairying, T'ennessee A gricultural Experim,ent Station, Knoxville, T'ennessee

Received for publication January 19, 1960

Changes in milk proteins effected by bacterial actionare important in various aspects of the ripening ofcheese and the proteolytic spoilage of dairy productsin refrigerated storage. Paper electrophoresis hasafforded a new method for the study of protein systemssuch as those in milk (MacRae and Baker, 1958; andWeigt, 1959), including the detection of certainchanges occurring in such systems. This method hasbeen used to study changes in casein during the ripen-inlg of various cheeses (Lindquist et al., 1953; andStorgards and Lind(quist, 1953) and changes in thewhey protein of milk from cows with acute mastitis(Lecce and Legates, 1959). A discussion of the generalapplication of the method to dairy science is available(Holdsworth, 1956a and 1956b). The present studywas made to determine what changes, if any, occur inthe paper electrophoretic patterns of casein and wheyproteins as a result of the growth of some commondairy psychrophiles in refrigerated skim milk.

MATERIALS AND METHODS

Bacterial cultures and sample inocula. The speciesused in these studies were Pseudomonas fragi strainiNo. ATCC 4973,1 Pseudomonas fluorescens strain No.ATCC 11251,1 and Pseutdomonas putrefaciens.2 Cul-ture maintenance and sample inocula preparationswere carried out as previously described by Overcastand Skean (1959).

M1ilk samples. All milk samples were mixed herdmilk obtained from the university creamery. Themilk was commercially separated and pasteurized bythe holding method. Within 5 min after the end of the30-min hold period, duplicate samples of approximately1500 ml were aseptically transferred to closed, sterile2-L, Erlenmeyer flasks held in an ice bath. When cooledto near ice bath temperature, one sample receiveda 0.5 to 1 per cent inoculation of a pure culture andthe other was held as the control sample. The sampleswere subsequently stored at 3 to 5 C for a period of42 days.

1 Obtained fromn the American Type Cultuire Collection,Washington, D. C.

2 Maintained in the cultture collectioii of the University ofTennessee D)airy Mlicrobiology Laboratory, Knoxville, Ten-nessee.

Sample examinations. The trial samples of skimmilk were examined on the day of inoculation and at1.4-day intervals thereafter throughout the storageperiod. The examinations included enumeration ofbacterial populations, organoleptic evaluation, andelectrophoretic separation of the casein and wheyprotein fractions. All examinations were made on sub-samples aseptically transferred from the trial samplesto suitable, sterile glass containers.

Bacterial counts. Both total and psychrophilic countswere made according to recommended procedures(American Public Health Association, 1953) except thata gelatin-phosphate dilution medium of Naylor andSmith (1946) was substituted for the phosphate buffernormally used. All plating was done with milk-proteinhydrolyzate (M-PH) agar. Total count plates wereincubated 48 ± 3 hr at 32 C, and psychrophilic countplates were incubated 7 days at 5 C.

Organoleptic evaluations. The organoleptic evalu-ations were made by an experienced judge of dairyproducts. The flavor of the sample was criticized withno attempt being made to assign a numerical score.

Electrophoretic separation. The proteins were ob-tained for electrophoretic separation by precipitatingthe casein at pH 4.6 with the dropwise addition of42.5 per cent lactic acid to 200 ml of milk under vigorousagitation. The precipitate and whey were separated by a30-min centrifugation followed by decantation of thewhey. The centrifugations were made at 2500 rpm using250-ml bottles on a head 7 in. in diameter. The re-covered precipitate was washed twice by resuspensionin 50 ml of distilled water at 5 to 10 C, followed byvacuum filtration through Whatman No. 1 paper.The washed casein was resuspended in a minimum(uantity of distilled water, then freeze-dried andpulverized by mortar and pestle. This treatmentyielded a fine, white powder containing approximately14 per cent nitrogen. The decanted whey was passedthrough a Seitz filtei. A 150-ml quantity was thenplaced in a cellophane dialyzer bag and dialyzed 44 hrat 4 to 6 C against 6 L of distilled water which wasrefreshed three times at 12-hr intervals. A 100-mlquantity of the dialyzed whev was freeze-dried to yielda material containing approximately 12 per cent

335

on May 22, 2018 by guest

http://aem.asm

.org/D

ownloaded from

J. D. SKEAN AND W. W. OVERCAST

nitrogen. All nitrogen contents were determined by a

micro-Kjeldahl method (Association of Official Agri-cultural Chemists, 1950).The proteins were prepared for electrophoresis by

resuspending the dried material in a suitable buffer at a

concentration of 1 per cent protein nitrogen. A boratebuffer was used with casein, and a barbital buffer was

used with the whey proteins. The borate buffer was ofpH 11.5 and contained 4.0 g of sodium hydroxide and4.96 g of boric acid per L. The barbital buffer was ofpH 8.6 and contained 3.7 g of barbital and 20.5 g ofsodium barbital per L. The buffers were made up indistilled water at room temperature. Each buffer hada calculated ionic strength of 0.1. The compositionof the buffer used in the electrophoretic cell was thesame as that used to resuspend the protein beingseparated.The electrophoretic separations and characteristic

curves of the proteins were obtained with a SpincoModel3 RB Paper Electrophoresis System, with theexception of the strips which were scanned with a

Model R Analytrol.3 The cell was operated in a coldroom controlled at 4 to 6 C. A 0.01-ml quantity of a

protein suspension from the control and the inoculatedsample was applied to each of four strips. The electro-phoretic run was continued for 16 hr. Constant voltagewas applied to the strips. Casein was separated at 140 vwith a current of 5.5 to 7 ma. Whey proteins were

separated at 200 v and 6.5 to 9 ma. The strips were

prepared for scanning by a modified Spinco procedurefor blood serum analysis (Spinco Division, InstructionManual4). The sequence of preparation included: (a)30-min drying at 120 to 130 C, (b) two 15-min prerinsesin methanol, (c) immersion for 1 hr in a 0.1 per centsolution of bromphenol blue in methanol, (d) three

3 Spinco Division, Beckman Instruments, Inc., BelmontCalifornia.

4Model R Paper Electrophoresis System Instruction ManualRIM-5. Spinco Division, Beckman Instruments, Inc., PaloAlto, California.

successive 5-min rinses in 5 per cent aqueous aceticacid, (e) gentle blotting between clean sheets of blottingpaper, (f) 10-min drying at 120 to 130 C, and (g)15-min exposure to an ammonia saturated atmosphere.The separated protein fractions were not identified.However, as the characteristic patterns obtained were

similar to those previously reported (Larson andRolleri, 1955; Lecce and Legates, 1959; MacRae andBaker, 1958; and Weigt, 1959), the nomenclature ofJenness et al. (1956) appears applicable.

RESULTS AND DISCUSSION

The pure cultures studied increased in numbersthroughout the storage period. A comparison of bothcounts (table 1) on the inoculated samples shows thatthe psychrophilic count always exceeded the totalcount with P. fragi and P. putrefaciens, but with P.fluorescens the reverse was true.A cooked flavor prevailed in the control samples

throughout the storage period. All inoculated samplesdeveloped a bitter flavor during the storage period.The bitterness varied in intensity and in the timerequired for its development. With P. fragi and P.fluorescens bitterness was easily detected on the 28thday of storage. Extreme bitterness was produced byP. putrefaciens by the 14th day of storage. All inoculatedsamples were extremely bitter when examined on thelast day of storage. Bitterness was the only off-flavordetected with P. putrefaciens. However, with P. fragia rancid and/or fruity flavor preceded the developmentof bitterness. Only a fruity flavor preceded the de-velopment of bitterness with P. fluorescens. Along withthe development of bitterness, protein flocculation hadoccurred by the end of the storage period in samplesinoculated with P. putrefaciens.

Paper electrophoretic separation of the casein andwhey proteins into their constituent fractions showedthat during the storage period both protein patternsfrom the inoculated samples were altered from those of

TABLE 1Total plate counts (TPC) and psychrophilic plate counts (PPC) on control (C) and inoculated (I) samples of skim milk*

0 Dayst 14 Days 28 Days 42 DaysCulture Sample

TPC PPC TPC PPC TPC PPC TPC PPC

plate counts/miPseudomonas fragi C 280.0 <10.0 220.0 <10.0 5.2 T <10.0 11.0 T <10.0

I 150.0 Tt 420.0 T 6.3 M§ 190.0 M 14.0 M 350.0 M 71.0 M 850.0 MPseudomonas fluorescens C 200.0 <10.0 360.0 <10.0 1.1 T <10.0 110.0 T 20.0 T

I 43.0 T 29.0 T 15.0 M 9.4 M 320.0 M 79.0 M 140.0 M 95.0 MPseudomonas putrefaciens C 75.0 <10.0 70.0 <10.0 1.8 T <10.0 2.3 T <10.0

I 4.5 T 20.0 T 20.0 T 2.8M 2.5M 9.8M 10.0 M 44.OM

* Milk was stored at 3 to 5 C and examined at 14-day intervals during a 42-day period. Each count stated is an average of two

trials per culture.t Within 1 hr after incubation.$ T = thousands.§ = millions.

336 [VOL. 8

on May 22, 2018 by guest

http://aem.asm

.org/D

ownloaded from

ELECTROl'HORETIC PROTEIN PATTERNS OF PSEUDOMIONAS

the control samples. Typical changes are shown infigures 1 and 2. The prominent change in casein was adecrease in fraction 2 and in the whey proteins adecrease in fraction 4. These two changes occurredwith each of the three organisms studied. However,other changes occurred that appeared peculiar to aparticular organism.

Atypical fractions appeared in the patterns of wheyproteins (figure 2) from milk inoculated with P. fragiand P. putrefaciens. The atypical fraction, whichmigrated toward the cathode (designated as fraction Ain figure 2), was evident with P. fragi on the 42ndday of storage. The atypical fraction with P. putrefaciensmigrated toward the anode (designated as fraction Bin figure 2), and was present on the 14th, 28th and42nd days of storage. The appearance of these atypicalfractions does not justify the assumption that theyresulted solely from bacterial action. The appearanceof a fraction located in the relative position of atypicalfraction B observed in this study has been reportedpreviously (Larson and Rolleri, 1955; and Lecce andLegates, 1959). The opposing electrophoretic migrationtendencies of the two atypical fractions observedindicate that they are of widely different composition.Proteins or peptides composed primarily of basicamino acids would be expected to migrate toward thecathode while those composed primarily of acidicamino acids would be expected to migrate toward theanode (Storgards and Lindquist, 1953).The sharp peak representing fraction 1 in the casein

patterns (figure 1) from the inoculated samples issimilar to that obtained by other workers in a studyof the ripening process of various cheeses (Lindquistet al., 1953). This peak appears only at the point ofsample application and its presence seems to be directlyrelated to bacterial action on the casein. Apparentlythe bacterial action causes a greater amount of theprotein in the sample to be adsorbed or bound to thepaper strip at the point of application. This is evidenton the strips shown in figure 1.The most rapid changes in the protein patterns were

effected by P. putrefaciens. With this culture thecasein pattern obtained after 14 days of storage ap-parently was no different from that obtained after 42days, indicating relatively rapid degradation of caseinby that organism. The pattern for the wtiey proteinswas altered less rapidly but continued to show changeoni the 28th anld 42nd days of storage.The degradationi of specific fractions in a protein

system implies a reduction in the total protein of thesystem. A reduction in the amount of total protein wasnot always indicated by the method used in this studybecause all samples were standardized at 1 per centprotein nitrogein content before the constituent frac-tions were electrophoretically separated. However,with P. putrefaciens a reduction in total amount ofwhey proteins was indicated despite sample standardi-

STRIP

Normrvl

' ,.

P fr

P fluorescens

P. put refociens

2

CA SE INPATTERN

FRACTION

-(-1------ > +(-j- - -- "I +MIGRATION DIRECTION

Figure 1. Paper electrophoretic strips and correspondingcharacteristic patterns of casein from high quality (normal)pasteurized skim milk and similar skim milk inoculated withPseudomnonas fragi, Pseudomonas fluorescens, and Pseudomnonasputtrefaciens after 42 days of storage at 3 to 5 C.

STRIP

Normol

WHEY PROTEiNSPATTERN

Pw

p, pUtVOfacienis

"A' 1 2 3 4 5 6 aB "MAl 2 3 4 5 6 -FRACTION

MIGRATION DIRECTION

Figure 2. I'aper electrophoretic stIrips and correspondingcharacteristic patterns of the whey proteins from high quality(normal) pasteurized skim milk and similar skim milk iniocu-ated with Pseudonmonas fragi, Pseudomonas flioorescens, an(dPseitdomnonas putrefaciens after 42 days of storage at 3 to 5 C.

1960] 337

on May 22, 2018 by guest

http://aem.asm

.org/D

ownloaded from

J. D. SKEAN AND W. W. OVERCAST

zation. This is shown in figure 2 by the difference inthe total area enclosed within the characteristic curveof the whey proteins from normal (control) milk andthat inoculated with P. putrefaciens. One explanationfor this may be that not all the nitrogen recovered infreeze-dried whey from milk inoculated with P. putre-faciens represented protein with the dye-binding ca-pacity necessary for paper electrophoretic analysis.This would be particularly true if the protein degrada-tion had proceeded to the amino acid stage.

SUMMARYSamples of high quality pasteurized skim milk were

inoculated with pure cultures of Pseudomonas fragi,Pseudomonas fluorescens, and Pseudomonas putrefaciensand stored 42 days at 3 to 5 C. On the day inoculatedand at 14-day intervals thereafter the samples wereexamined for total bacterial count, psychrophilic bac-terial count, organoleptic quality, and changes in thepaper electrophoretic pattern of casein and wheyproteins. The bacterial counts increased throughoutthe storage period. With P. fragi and P. putrefaciensthe psychrophilic count always exceeded the totalcount, whereas with P. fluorescens the reverse heldtrue. Each of the cultures produced a bitter flavorby the end of the storage period. Changes occurringin the paper electrophoretic protein patterns indicated:(a) changes in the relative amounts of the containedprotein fractions, (b) decreases in total protein con-tent, and (c) the appearance of electrophoreticallydifferent fractions.

REFERENCES

American Public Health Association 1953 Standard methodsfor the examination of dairy products, Ed. 10. AmericanPublic Health Association, Inc., New York, New York.

Association of Official Agricultural Chemists 1950 Officialmethods of analysis, p. 745. Ed. 7. Washington, D. C.

HOLDSWORTH, E. S. 1956a Paper chromatography and paperelectrophoresis as applied to dairy science. Part I. DairySci. Abstr., 18, 98-110.

HOLDSWORTH, E. S. 1956b Paper chromatography and paperelectrophoresis as applied to dairy science. Part II. DairySci. Abstr., 18, 186-198.

JENNESS, R., LARSON, B. L., MCMEEKIN, T. L., SWANSON,A. M., WHITNAH, C. H., AND WHITNEY, R. 1956 No-menclature of the proteins of bovine milk. J. Dairy Sci.,39, 536-541.

LARSON, B. L. AND ROLLERI, G. D. 1955 Heat denaturationof the specific serum proteins in milk. J. Dairy Sci., 38,351-360.

LECCE, J. G. AND LEGATES, J. E. 1959 Changes in the paperelectrophoretic whey-protein pattern of cows with acutemastitis. J. Dairy Sci., 42, 698-704.

LINDQUIST, B., STORGARDS, T., AND GORANSSON, MAJ-BRITT.1953 Electrophoresis in paper as a means of studying theripening process in cheese. 13th Intern. Dairy Congr.,.3, 1261-1268.

MACRAE, H. F. AND BAKER, B. E. 1958 Application of elec-trophoresis on paper to the estimation of alpha-, beta-and gamma-casein. J. Dairy Sci., 41, 233-240.

NAYLOR, H. B. AND SMITH, P. A. 1946 Factors affecting theviability of Serratia marcescens during dehydration andstorage. J. Bacteriol., 52, 565-573.

OVERCAST, W. W. AND SKEAN, J. D. 1959 Growth of certainlipolytic microorganisms at 4 C and their influence on freefat acidity and flavor of pasteurized milk. J. Dairy Sci.,42, 1479-1485.

Spinco Division, Beckman Instruments, Inc. Model R paperelectrophoresis system instruction manual RIM-5. PaloAlto, California.

STORGARDS, T. AND LINDQUIST, B. 1953 A comparison of theripening process of different cheese types, based on somenew methods of investigation. 13th Intern. Dairy Congr.,2, 625-628.

WEIGT, U. 1959 Papierelektrophoretishe Untersuchungenan normalen Rindermilchseren. Milchwissenschaft, 14,61-64. (English summary).

338 [VOL. 8

on May 22, 2018 by guest

http://aem.asm

.org/D

ownloaded from