Preservation of Microorganisms by Freeze-Drying

I. Cell Supernatant, Naylor-Smith Solution, and Salts of Various Acids as Stabilizersfor Serratia marcescens'

R. G. BENEDICT, J. CORMAN, E. S. SHARPE, C. E. KEMP, H. H. HALL, AND R. W. JACKSON

Fermentation Laboratory, Northern Utilization Research and Developnment Division,2 Peoria, Illinois

Received for publication April 7, 1958

Inasmuch as all living things usually require a highcontent of water either internally or externally, or both,it is remarkable that microorganisms, especially sporeforms, can sometimes withstand an extreme desiccatingenvironment. Any success achieved by chemical orphysical means in modifying viability, that is in hasten-ing or preventing death of the desiccated cell, has impli-cations in many different fields. The various techniquesexploited during the last 45 years in the preservation ofbacteria by drying have been reviewed in an interestingand very useful manner by Fry (1954). Although hecommends Naylor and Smith (1946) on the logicaldevelopment of their stabilizing medium for the freeze-drying of Serratia marcescens, he points out that itsgeneral applicability would have to be tested with moresensitive organisms. The present study of preservationby drying is also limited to S. marcescens but a con-siderable variety of new substances has been triedbesides the Naylor-Smith components. Some of ourresults should find application in the preservation ofother organisms freeze-dried under similar conditions.Experiments with sensitive strains of yeast and bacteriawill be reported in a later paper.

MATERIALS AND METHODS

The culture of S. marcescens was obtained from theChemical Corps Biological Laboratories as their strain8 UK and is hereinafter designated as NRRL B-1481.Cells were grown in a medium containing 5 per centskim milk solids, 3 per cent Protopeptone,3 and 2 percent glucose designated SPG medium and described byBenedict et al. (1957). Viable cell concentrates contain-ing approximately 400 X 109 organisms per ml wereprepared by centrifugation, washed, and combined withan equal volume of stabilizer solution. The final concen-tration of cells therefore was approximately 200 X 109cells per ml. Three-ml amounts of the mixtures of

1 This work was supported by a contract with the ChemicalCorps, Fort Detrick, Frederick, Maryland.

2 Agricultural Research Service, United States Departmentof Agriculture.

3 No. 343 of Wilson and Company, Chicago, Illinois. Themention of trade name products does not imply endorsementof these over others of equal quality.

401

stabilizer, neutralized to pH 7 as required, and of cellswere dispensed into 15-ml serum bottles for the freezingand drying operations. The preparations were frozen byimmersing the spinning bottles in a bath of Dry Ice andmethyl cellosolve. The temperature of the bath was-40 to -60 C. The frozen cell preparations were heldfor a short time in a box refrigerated with Dry Iceuntil ready for placement in the drier. The machine usedfor drying in this study was the Experimental Dehydra-tion Unit no. 3501.4 See figure 1. The drying chambercontained a platen with 36 depressions, 1 in. in depthand 1I4 in. in diameter, to hold the serum bottles. Aspecial stoppering device with a vacuum seal permittedthe operator to stopper all bottles simultaneously undervacuum at the end of each drying run. In effect, theapparatus is a modification of that described by Huttonet al. (1951).

After a number of preliminary trials, a standard

National Research Corporation, Newton, Massachusetts.

Figure 1. The drying chamber of the experimental dehydra-tion unit showing enclosed platen and vacuum-seal stopperingdevice.

on June 22, 2020 by guesthttp://aem

.asm.org/

Dow

nloaded from

BENEDICT, CORMIAN, SHARPE, KEMP, HALL, AND JACKSON

drying schedule was adopted and used throughout theinvestigation. Prior to placing the frozen cell suspen-sions in the apparatus, the platen temperature waslowered to -30 C by means of a cooling system,5 andthe greased stoppers were arranged in position on theplate above the platen. After pressure in the chamberhad been reduced to 100 ,u of Hg by a gas ballast pump,an auxiliary diffusion pump was turned on, and thebottles were allowed to remain in the drier for a total of18 hr. The final pressure usually obtained prior tostoppering ranged from 15 to 25 A, and the final chambertemperature ranged from 30 to 32 C. The attached coldtrap was cooled with a mixture of Dry Ice and cello-solve, and was recharged twice with Dry Ice duringeach experiment. The total entrapping capacity,amounting to 7 pounds of ice, was only partly ex-hausted in a drying cycle. Although it was possible todry cell preparations in a shorter cycle, that is, 5 to 7hr, by elevating the platen temperature to 40 C duringthe last 2 hr of the run, the longer cycle neverthelesspermitted processing of dried samples on the followingwork day.The 36 positions of the platen permitted simultaneous

testing of 5 variables, with 6 bottles for each variableand 6 for controls. Occasionally the replicates werereduced to 5 for each of 6 variables, again with 6 bottlesas controls. Each dried sample was rehydrated undervacuum with 5 ml of a sterile diluent composed of 0.5per cent NaCl and 0.1 per cent tryptose. The 5-mlaliquot was removed from 100 ml of diluent in a 150-mldilution bottle and injected into the bottle of dried sam-ple with a hypodermic syringe. The rehydrated sample,thoroughly shaken, plus another portion of diluent usedto rinse the inner part of the rubber stopper, was returnedas quantitatively as possible to the original dilutionbottle. Dilutions were continued until a 0.3 ml aliquotfrom the final bottle was predicted to contain between30 and 300 viable cells. An aliquot of this amount waspipetted onto the dry surface of each of 9 agar plates andrapidly spread with a Z-shaped glass rod. The plateswere incubated at 30 C for 20 hr and estimates of viablecounts were made according to the procedure previouslydescribed.

RESULTSProtective effect of culture supernatant. Early results

indicated that cells of S. marcescens resuspended inculture supernatant were protected during subsequentdrying. The quantitative relationship between survivalof cells and amount of accompanying supernatantliquor is shown in figure 2.We also noted that "cell juices" would serve as

drying protectants. In establishing this point, the first6 We are indebted to Mr. V. F. Pfeifer of the Engineering

and Development Laboratory, Northern Division, for theconstruction of the cooling system.

operation was to centrifuge the cells, wash them inphysiological saline and recentrifuge, discard the saline,and resuspend the cells in distilled water. Four alternateslow freezings and rapid thawings, followed by centrif-ugation and filtration of cell juices through UFPyrex6 sintered glass filters, gave a solution whichprotected fresh cells during drying to the extent of 61per cent. Hutton and Shirey (1951) had reported thatthe cell-free juices from heavy suspensions of Brutcellaabortus strain 19, containing 600 X 109 cells per ml,gave drying protection to dilute suspensions of the sameorganism and that dilution of such cell-free fluids withwater decreased the drying protection in proportion tothe amount of dilution. Shortly after we noted theprotective effect of culture supernatant liquor, Recordaind Taylor (1953) reported that, when water suspen-sions of Escherichia coli were dried, reconstituted inwater, and freeze-dried a second time, most of the cellswere killed, but that soluble protective material wasrecoverable. This cell-free material was freeze-dried anddissolved at various levels in 5 per cent glucose solution.The mixtures were then used as stabilizers in the freeze-drying of E. coli cells. Cells dried in this manner andthen subjected to drying over P205 at a pressure of0.01 mm Hg or less evidenced 51 per cent viability after1 day and 38 per cent after 330 days. The stabilizingmaterial derived from E. coli cells was reported to beeffective also in protecting S. marcescens during drying.

6 Corning Glass Works, Corning, New York.

100Ir

90

801-

z 70u

CL

w-_ 60

D: 50

0z

40

z 30

201-

10

I I I I0 25 50so 75 100

SUPERNATANT, PERCENT

Figure 2. Effect of increasing amounts of culture liquorsupernatant as stabilizer on survival after drying of Serratiamarcescens.

402 [VOL. 6

on June 22, 2020 by guesthttp://aem

.asm.org/

Dow

nloaded from

PRESERVATION- BY FREEZE-DRYING

Effort was made to learn the chemical nature of theprotective factor in our supernatant liquors. A separa-tion of the material which protected cells during dryingcould be made either by (a) exhaustive dialysis againstdistilled water at 34 F or (b) ethanol precipitation.When the low and high molecular weight fractionsobtainied by dialysis, hereinafter referred to as LMWor H1IW, respectively, wvere adjusted to their originalvolumes by appropriate means, the LMW fraction wasfound to contain practically all of the protective factor.It gave about 65 per cent drying protection as opposedto about 3 per cent for the HMW fraction. The LMWfractioni adjusted to pH 6 to 7 lost a third of its protec-tive capacity upon being heated at 100 C, but none at80 C. When LMW fractions were distilled in vacuo atpH 4.0, 7.0, and 10.0, the results, as shown in table 1,indicated that volatile substance was not responsiblefor the protection afforded. When the fermentationacids present were adsorbed by passing the LMWfraction through a column containing Dowex-1 resinand the neutralized effluents concentrated to 4X, 2X,1X, and 0.5>X concentration in terms of the controlsupernatant, they permitted cell survivals of 100, 100,85, anid 65 per cent, respectively. A LMW control with-out chromatographic treatment as above gave 54 percent drving recovery.A LMW diffusate giving 56 per cent drying protection

at 1)X concentration was freeze-dried and method ofBorkenhagen (1953) was applied,that is, treatment of thelyophilized solids with anhydrous MeOH containing NH3

TABLE 1

Fractions of LMW' supernatant naterial as viabilityprotectants during drying.for washed cells of

Serratia mnarcescens*

Mlethod of Preparing SupernatantFractiont

pH 4.0Residue ....................

Distillate ..................

pH 7.0Residue ....................

D)istillate ...........pH 10.0Residue........Distillate ..................

Untreated LMW superna-tant .. ................ ..

Viable CellCount and 95

Per CentConfidence

Limits beforeDrying

X 109/ml

237 16213 i 23

184 4 13198 i 3

RangeSurvivalafter

Drying

43-606-8

60-713

210 + 12 59-69224 + 8 0

200 i 11 53-65

* NRC drier used throughout work reported in these tables.

t Three 500-ml aliquots of the material having low molecu-lar weight (LMW) were concentrated under reduced pressureat 30 C at the pH indicated until 250 ml of distillate were col-lected and 250 ml remained as residue. Aliquots of the fractions

were mixed with equal amounts of cell concentrate and then

subjected to freeze-drying.

to extract amino acids and other solubles. The MeOH-NH3 extract, readjusted to 1 X concentration, still gave46 per cent cell protection, whereas the insoluble residue,similarly adjusted, gave only 8 per cent. These resultssuggested that amino acids may have been partiallyresponsible for the drying protection afforded, althoughexcessive tailing did not permit the identification ofindividual amino acids by paper strip chromatography.Mild acid-hydrolysis of the MeOH-NH3 extract did notchange its activity as a drying stabilizer. A few experi-ments were run with neutralized solutions of the follow-ing single amino acids as stabilizers: alanine, glycine,arginine, aspartic acid, and glutamic acid. Survivalsof cells at the 0.15 M level of each amino acid were 17,13, 56, 23, and 30 per cent, respectively. Decreasing thelevel of stabilizer to 0.075 M increased the survivals witharginine and alanine, whereas the reverse was true withglutamic acid. When a peptide, glutathione, was testedat the 0.05 M level, 57 per cent of the cells surviveddrying. Various salt-free, acid- or enzyme-hydrolyzedcaseins, tested at levels from 0.5 to 2 per cent, wereexceedingly poor stabilizers for S. marcescens.Drying survival of cells with respect to medium, growth

rate, OAR,7 and temperature during fermentation. Theextent to which resistance of cells is developed dependssomewhat on the medium in which the cells are grown.When strain B-1481 was grow-n in the synthetic mediumof Smith and Johnson (1954) and dried with the neutral-ized culture li(quor as stabilizer, the average survival inseven vials was only 18 per cent. In general agreementwith Naylor and Smith (1946) and Fry and Greaves(1951) we found that aerated cells harvested during anearly stage of the log phase, or statically grown cul-tures, do not withstand freeze-drying as well as oldercells from well aerated cultures. Cells growni in SPGmedium at an OAR of 1, harvested at 7, 12, and 24 hr,and resuspended in supernatant liquor gave recoveriesof 11, 40, and 54 per cent, respectively. With super-natant liquor as the suspending agent, cells propagatedin 20-L fermentors at an OAR of 1 and temperatures of33 and 29 C both averaged 67 per cent drying survival,whereas those cultured at 25 C and 21 C averaged only28 and 18 per cent, respectively. In the latter two cases,cell production was slower but the total viable countsare equivalent to those obtained at the higher tempera-tures.We also foun-d that greater aeration of cells seems to

exert a sparing action on the amount of NaCl stabilizerrequired subsequently to achieve a given amount ofdrying protection. Cells produced in SPG medium atOAR of 0.2, 0.7, and 3.0 were washed with water and themaster cell pools equilibrated to contain approximatelythe same concentration of viable cells. The cells grownat the lowest OAR gave a maximum viable cell re-

7Oxygen absorption rate (OAR) was measured with sulfitesolution and is expressed in millimols of 02 per L per min.

190-8] 403

on June 22, 2020 by guesthttp://aem

.asm.org/

Dow

nloaded from

BENEDICT, CORAIAN, SHARPE, KEMP, HALL, AND JACKSON

covery of only 16.4 per cent at 0.108 M NaCl. Cellsgrown at the intermediate OAR showed about twice thissurvival rate in 0.054 M NaCl. Cells grown at the highestOAR level showed a significantly higher viable re-

covery in 0.054 M NaCl than did the cells propagated atthe intermediate OAR.

Fermentation conditions were therefore standardizedso that cells for drying on a laboratory scale were grownfor 20 to 22 hr at 29 C and at an OAR of 0.7. Viable cellcounts usually ranged between 105 and 125 X 10'cells per ml. Furthermore, tests of stabilizers were more

reproducible when the following technique was adopted:(a) centrifuge cells; (b) discard supernatant; (c) resus-

pend cells in a volume of distilled water equal to that ofthe original supernatant and recentrifuge, discardingthe wash water; and (d) resuspend cells in neutralizedstabilizing solutions such that the viable cell concen-

trate count prior to drying would be around 200 15X 109 per ml. This technique is more sensitive becauseit eliminates the protective power of the supernatant.The drying survivals of cells in supernatant liquoradjusted to pH 5.0, 6.0, 7.0, and 8.0 averaged 35, 66,78, and 62 per cent, respectively. Consequently, pH 7was selected as optimal for all mixtures to be dried.

Protective effects of sodium salts of various organic acidsincluding those of the tricarboxylic acid cycle. Early in our

work, it was suggested that salts of some acids of theterminal respiratory cycle might account for some of thedrying protection afforded. Before the washed celltechnique was adopted, numerous trials were made withsodium salts of compounds of the tricarboxylic acidcycle alone and in various combinations. The highestinitial drying survival of 81 per cent was obtained witha mixture of 0.02 M each of Na acetate, lactate, pyru-

vate, a-ketoglutarate, succinate, fumarate, malate,citrate, and aconitate. Although the situations are notparallel, we noted with interest that Heinmets et al.(1954) found that the highest reactivation of E. colistrain B/r, treated with heat, chlorine, "zephiran"chloride, and ethyl alcohol, also occurred when the cellswere incubated in a mixture of metabolites of the tri-carboxylic acid cycle. We found that, wvhen water-washed cells were stabilized with the sodium salts ofvarious acids (table 2), the maximum drying survivalusually resulted at 0.05 to 0.1 M concentration of thesalt. The difference between survivals with succinateand malate and the other salts was not accounted for.

Inorganic salts as stabilizers. In further search forinexpensive protective compounds, a few experimentswere made with three salts of HCl. The results are

shown in table 3. The poor survivals obtained withNH4C1 alone suggested that it could be omitted as a

component of the Naylor-Smith stabilizer under our

conditions.Survival of cells as affected by components of Naylor-

Smith stabilizer (NS). The original NS contained 0.5

per cent each of thiourea, ascorbic acid, and NH4Cl,plus 2 per cent dextrin. Dextrin was omitted in ourexperiments since drying survivals of 1 per cent or lesswere obtained with 2 per cent dextrin alone as stabilizer.Investigation of the remaining components of NSproved to be most interesting. Table 4 shows the re-sults of an experiment to evaluate each of the three

TABLE 2Effect of sodium salts of organic acids at variousconcentrations on viability after drying of washed

cells of Serratia marcescens

Conc of Salt (at)

Stabilizer 0.4 0.2 0.1 0.05 0.025

Mean survival* (%)

Na acetate .8 43 68 55 36Na lactate .40 56 69 44 22Na malate . - 15 17 29 22Na pyrivate .- 26 58 59 36Na succinate . - 3 10 30 19

* Mean count before drying was 213 i 10 X 109/ml. Fivebottles were employed per sample. Mean suirvival for controlcells in Naylor-Smith stabilizer was 72 per cent and in NS/10,4.5 per cent.

TABLE 3Effect of inorganic salts at various concentrations on

viability after drying of washed cells of Serratiamarcescens

Conc of Salt (mt)

Stabilizer 0.216 0.108 0.054 0.027 0.013

Mean survival* (%)

NH4Cl .1 3 13 1 -

NaCl .9 20 50 29 13KCI .7 21 35 44 24

* The mean cell counts before drying with NH4CI, NaCl,and KCI were, 227 i 17, 210 i 15 and 206 i 15 X 109 per ml,respectively.

TABLE 4Naylor-Smith components as stabilizer.s in drying of washed

cells of Serratia marcescens at concentration of265 i 22 X 109/ml

Stabilizer Employed Range Survival Meanafter Drying Survival*

Ascorbic acid, 0.5 .......... ........ 11-14 12Thiourea, 0.5 ...................... 47-63 56NH4Cl, 0.5 ....................... 3-8 5Ascorbic acid, 0.5, and Thiourea, 0.5 61-74 68Ascorbic acid, 0.5, and NH4Cl, 0.5.. 67-78 72Thiourea, 0.5, and NH4Cl, 0.5...... 13-22 17NS/lot ........................... 3-5 4

* Values are means of 5 replicates.t 0.05 Per cent each of ascorbic acid, thiourea, and NH4CI.

[VOL. 6404

on June 22, 2020 by guesthttp://aem

.asm.org/

Dow

nloaded from

PRESERVATION BY FREEZE-DRYING

components singly, and in combination of two. Al-though both ascorbic acid and NH4Cl used singly werepoor stabilizers, combination of the two permitted 72per cent of the cells to survive. Combination of ascorbicacid and thiourea gave a cell survival of 68 per centwhich is the exact sum of the per cent survivals ob-tained with the two components used separately. Oiithe other hand, the combination of thiourea and NH4Clgave a cell survival of only 17 per cent, far less than the56 per cent survival obtained with thiourea by itself.The high recoveries of cell viability with thiourea aloneled to further studies on related compounds as sta-bilizers.

Relationship of cell and stabilizer concentrations. In hisreview, Fry emphasized that the results of studies con-ducted by Otten (1930), Stamp (1947), and Fry andGreaves (1951) on the relation of cell concentration and

TABLE 5

Effect of cell concentration on viability after drying ofwashed cells of Serratia marcescens with NS or

NS/1O as stabilizer*

Viable Cell Count and 95 Range Survival MeanStabilizer Conc Per Cent Confidence Ranger Survival MeviantLimits before Drying after Drying Survivalt

X 109/ml % %NS/10 289 4 29 5-6 5

204 15 6-9 8105 9 28-29 28

52.6 ± 3.0 52-57 55

NS, full conc 289 i 29 66-77 73105 ± 9 53-68 61

52.6 i 3.0 40-46 42

* NS: 0.5 per cent each of ascorbic acid, thiotirea, aiidNH4Cl.

NS/10: 0.05 per cent each of the three ingredients.t Five bottles representing 5 replicates were employed per

sample.

-3007x

200Lu 150zZ 100oo0

370

uJ 5040

301I Il I I I 1 I I I21'- - - -

D'VING cIJPVIVAL. PERCENT

Figure 3. Survival after drying of washed cells of 4 pools of

Serratia marcescens with predrying counts of 289, 204, 105,and 53 X 101 cells per ml with NS or NS/10 as stabilizers.

Plotted on log probability paper. (See table 5 for details.)

survival were not the same. However, the methods ofdrying and the types of organisms employed weredifferent and comparisons could not logically be made.Miller and Goodner (1953) tested 62 different stabilizersin the freeze-drying of 10 strains of BCG (BacillusCalmette-Guerin). Glucose and glutamic acid werefound to be best in their respective carbohydrate andamino acid groups and were therefore employed inexperiments to show that the concentration of astabilizer is important. We also found that it was neces-sary to take into account the ratio of cells to amount ofstabilizer and confirmed this in two series of experi-ments. First, we employed NS at two levels with four

TABLE 6Comparison of viabilities obtained after drying with those

calculated when concentration of stabilizer was varied fromtNS/10 to 2 NS for washed cells of Serratia marcescens atconcentration of 203 4 16 X 109 per ml

Stabilizer Conc Survival Obtained Survival Calcu-

% NoNS/JO (0. 10) 5 4NS/8 (0.13) 8 ] 1NS/6 (0.17) 16 21NS/4 (0.25) 43 39NS/3 (0.33) 60 55NS/2 (0.50) 80 783 NS/4 (0.75) - 901 NS (1. 00) 79 782 NS (2.00) 66

* See equat ion in text.

9998

99,_. 90zu 80u 70_' 60> 50> 40D 30O 20z

ad 105

21

x

X/

II I.100.167.250.333 .500 .750 1.00

CONCENTRATION OF NS STABILIZER

Figure 4. Effect of changing cell-stabilizer ratio on viablecell survival after drying. Cell count of Serratia marcescensbefore drying was 203 X 109 per ml. Plotted on semi-log paper.(See table 6 for details.)

1958] 405

-

I1

on June 22, 2020 by guesthttp://aem

.asm.org/

Dow

nloaded from

BENEDICT, CORMAN, SHARPE, KEMP, HALL, AND JACKSON

different concentrations of cells, each originating fromthe master cell pool. Table 5 shows the results obtainedwhen NS and NS/10 were employed as drying sta-bilizers for washed cell concentrates with initial pre-drying counts of 289, 204, 105 and 53 X 109 cells perml. It will be noted that with NS/10 the average percent survival of drying increases with decreasing cellconcentration, whereas with NS the reverse is true. Theaverage per cent survivals were plotted on log probabil-ity paper and are shown in figure 3.

Next, when cell population was held constant andstabilizer concentrations were varied from NS/10 upto 2 NS, the initial drying recoveries were the lowestat NS/10, but increased as the NS concentrationsincreased up to NS/2, and thereafter began to decreaseas the NS levels became excessive (figure 4). Since thegraph closely resembles one-half of a parabola, thegeneral equation (X2 = -2 py) or its modified form[(X - c)2 = 2p (y - d)] may be utilized. The constantsof this equation (c, p, and d) were evaluated throughexperimental data by the method of simultaneousequations. Substituting NS concentration for X and per

cent survival for y, the final form becomes (NS concen-

tration 0.75)2 = - 0.0049 (per cent survival -90)which may be used to calculate survivals. Calculatedfigures are in fair agreement with the experimentalvalues shown in table 6.The results of these experiments show that the ratio

of cells to stabilizer warrants attention by future inves-tigators. Indeed, the use of stabilizers without regardfor cell-stabilizer ratio could certainly be a factor in pastfailures to preserve cells which have a history of beingdifficult to freeze-dry. This cell-stabilizer relationshipmight also explain the low drying survival of a strain ofS. marcescens reported by Christian and Stockton(1956). Counts were not given, but their cells were

grown at an extremely low OAR (250 ml mediumshaken in 500 ml flasks) which does not permit rapidgrowth and high terminal counts. Although straindifference would doubtless affect the survival, the lowdrying recovery of only 4 per cent obtained by themmight well be due in part to the very low ratio of cellsto Naylor-Smith stabilizer. Extrapolation of our NSdrying curve, figure 3, indicates that 7 per cent survivalcorresponds to 1 X 109 cells per ml.

DISCUSSIONThe techniques employed in most of the experiments

described in this paper with respect to use of wet cellconcentrates and quantities of cells dried are quitedifferent from those utilized in many laboratories topreserve cultures by freeze-drying. For this purpose a

majority of workers in the interest of economy or

safety, or both, operate on a smaller scale and harvestthe cells from surface growth on agar plates. With thisprocedure, probably not more than 1 X 108 or 1 X 109

juvant prior to drying and the per cent recovery ofviable cells remaining after drying is not determined.Success in the operation is measured by what Fry (1954)has aptly termed the "death or survival" method.

In tests of the sodium salts of various acids as dryingstabilizers, either singly or in admixture, the highestdrying survivals usually occurred between the 0.05and 0.1 M levels for a viable cell population of 200 i 15X 109 per ml. As tested with NaCl as stabilizer, highlyaerated cells grown in flasks were better fitted to survivefreeze-drying and required less stabilizer than thosepropagated at a low OAR.

Since cell liquor is so effective in protecting cells, itwas hoped that a specific and unique compound mightbe isolated and identified. This hope was not realized.It may well be that there is no one specific protectivefactor or even just a few. The fact that inorganic salt,salts of organic acids, sugars, amino acids, and othercommon organic chemicals have been found underproper conditions to afford high or relatively highdegrees of protection to dried cells lends considerableweight to this possibility. If the effect is unspecific, oreven largely so, it would seem that it is related tochanging osmotic pressure and perhaps to membranepermeability during drying or rehydration, or duringboth. This relationship could, in turn, be complicated bynumerous factors such as metabolic use and reservesupply of oxygen and carbon source.

Results obtained with various combinations of the NSingredients indicate that it is impossible to predictwhether the effects of stabilizers in combination will beadditive, antagonistic, or synergistic. Miller and Good-ner found that combinations of glucose and Na gluta-mate together gave less drying protection for BCG thaneither component used singly. Relative to the freeze-drying of cells grown in flasks, our findings stronglysuggest that the full potential of a drying stabilizercannot be realized unless the following are considered:(a) the organism should be propagated in a nutritionallycomplete medium at optimal temperature; (b) propaga-tions should be made at various oxygen absorption ratesto determine the optimal OAR for maximal cell produc-tion; (c) cells should be harvested at the experimen-tally determined physiological age when they are wellequipped to withstand freeze-drying; and (d) theoptimal ratio of drying stabilizer to viable cell numbersshould be established and employed.

SUMMARYA carefully standardized procedure for preparing and

drying cells of Serratia marcescens w,a,s developed in thisstudy.

Cells harvested from skim milk protopeptone glucosemedium after 22 hr growth showed an unusually highinitial survival when dried in their own supernatantculture liquor.

Tests of the sodium salts of various acids in a range of

[VOL. 6406

viable cells per tube are initially combined with ad-

on June 22, 2020 by guesthttp://aem

.asm.org/

Dow

nloaded from

ACTIVITY OF S. AGALACTIAE ON CASEIN

concentration showed that the maximum drying protec-tion per 200 X 109 cells per ml was generally affordedat stabilizer levels between 0.05 and 0.1 M.

Proof that cell-stabilizer ratios markedly affect sur-vival after drying was obtained by varying the cellpopulation at each of two levels of modified Naylor-Smith stabilizer, and by employing eight levels of thelatter while holding the cell population constant.

REFERENCESBENEDICT, R. G., KOEPSELL, H. J., TSuCHIYA, H. MW., SHARPE,

E. S., CORMAN, J., KEMP, C. E., MEYERS, G. B., ANDJACKSON, R. W. 1957 Studies on the aerobic propaga-tion of Se7ratia mlarcescens. Appl. Microbiol., 5, 308-313.

BORKENHAG(EN, L. K. 1953 Process for preparing aminoacids. U. S. I'atent, no. 2,657,232 (Chem. Abst., 48, 2994).

CHRISTIAN, R. T. AND STOCKTON, J. J. 1956 Influience ofsealing pressure on survival of Serratia marcescens andMicrococciis pyogenes var. au'reus desiccated from thefrozen state. Appl. M\icrobiol., 4, 88-90.

FRY, R. NI. 1954 The preservation of bacteria. In Biologi-cal applications of freezing and dIrying, Ch. 10, pp. 215-252, Edited by R. J. C. Harris. Academic Press, Inc.,New York, New York.

FRY, R. M. AND GREAVES, R. I. N. 1951 The survival ofbacteria during an(d after drying. J. Hgy., 49, 220-246.

HEINMETS, F., TAYLOR, W. W., AND LEHMAN, J. J. 1954 Thie

use of metabolites in the restoration of the viability ofheat and chemically inactivated Escherichia coli. J.Bacteiriol., 67, 5-12.

HUTTON, It. S., HILMOE, R. J., AND ROBERTS, J. L. 1951Some physical factors that influence the survival ofBrulcella abortus during freeze-drying. J. Bacteriol., 61,309-319.

H1JTTON, R. S. AND SHIREY, W. N. 1951 Evidence that sus-pensions of Brucella abortus contain a substance which in-fluencies the degree of killing caused by freeze-drying.Bacteriol. Proc., pp. 34-35.

NIILLER, R., JR., ANI) GOODNER, K. 1953 Stu(dies on thestability of lyophilized BCG vaccine. Yale J. Biol. andMed., 25 (4), 262-283.

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

OTTEN, L. 1930 D)ie Trockenkoinservierung von pathogenenBakterien. Zentr. Bakteriol. Parasitenk I. Abt. Orig.,116, 199-210.

RECORD, B. R. AND TAYLOR, R. 1953 Some factors influenc-ing the suirvival of Bacteriumi coli on freeze-drying. J.Gen. Microbiol., 9, 475-484.

SMITH, C. G. AND JOHNSON, M1. J. 1954 Aeration require-ments for the growth of aerobic microorganiSmS. J.Bacteriol., 68, 346-350.

STAMP, L. 1947 The preservation of bacteria by drying. JGen. Microbiol., 1, 251-265.

Activity of Streptococcus agalactiae on Casein of Heat-SterilizedCow's Milk

G. W. MALANEY, H. H. WEISER, Q. VAN WINKLE, AND J. M. PEEK

Departmtients of Bacteriology and Chemistry, The Ohio State University, Columulbus, Ohio

Received for publication April 7, 1958

Inverso et al. (1953) presented data which indicatedthat Streptococcus agalactiae is capable of inducing a

change in the proportions of the components of casein,yielding an increase in the relative concentration of/3-casein with an accompanyinig decrease in a-casein.Their findings have prompted further study of thephysicochemical changes produced in the casein ofcow's milk by S. agalactiae.

MIATERIALS A.ND METHODS

Organisms. Streptococcus agalactiae 6683 from thestock culture collection of the Department of Bac-

teriology, The Ohio State University, was used in thisstudy.

Preparations of samples. Skim milk or "reconstituted"milk was distributed into two 500-ml flasks and steri-

lized in the autoclave at 120 C for 7 to 10 min. This wasthe heat exposure used by Inverso et al. One of theflasks was inoculated with the test organism. Samples

takeni fromn this flask are referred to as the "inoculatedsamples." The second flask served as the uninoculatedcontrol. The flasks were incubated at 37 C. At selectedtime intervals, samples were removed aseptically. Acontrol sample was run in parallel with each inoculatedsample. The bacterial population was followed by platecounts on tryptone glucose extract agar.The casein of the sample was separated by a modified

acetic acid precipitation method, essentially that recom-mended by the Associationi of Official AgriculturalChemists (1950). Thirty-five ml of milk in a 100-mlcylinder was diluted to 80 ml with distilled water at40 to 42 C. To this was added sufficient 10 per cenitacetic acid solution to reduce the pH to 4.5. The mixturewas inverted several times and let stand for 10 min. Avolume of M/1 sodium acetate solution equal to thevolume of acetic acid solution used was then added andthe total volume made up to 100 ml with distilled water.The precipitated casein was separated from the super-

1958] 407

on June 22, 2020 by guesthttp://aem

.asm.org/

Dow

nloaded from