THE EFFECTS OF SPRAY-DRYING ON THE VIABILITY OF FUNGOUS SPORES'

PETER MAZUR2 AND WILLIAM H. WESTONThe Biological Laboratories, Harvard University, Cambridge, Massachusetts

Received for publication July 11, 1955

Much effort has been devoted recently to meansof conserving viability by reducing the activityof cells through the withdrawal of water. Chiefof these methods of desiccation is the lyophilor freeze-drying technique. The rationale behindfreeze-drying is the supposition that water canbe removed from a cell without causing death ifthe water is removed directly from the frozenstate-that is, by sublimation. In support of thisconcept, drying from the frozen state has beensuccessful in many cases (Flosdorf, 1949) in whichdehydration from the liquid state has not.On the other hand, a considerable number of

instances have been reported in which freeze-drying killed most or all of the organisms. Forexample, Atkin et al. (1949) reported that 99.98per cent of brewer's yeast cells fail to survivedesiccation; Elser et al. (1935) were, except inrare cases, unable to obtain viable protozoansfollowing freeze-drying; Glover (1946) reportedthat the technique is exceedingly harmful toMycobacterium tuberculosis; Raper and Alexan-der (1945) have found that spores of the Ento-mophthorales do not survive lyophilization.

Until recently freeze-drying has been anempirical technique, and attempts to improve andextend its applications to living cells have beenempirical as well. The authors felt, therefore, thatstudies on the fundamental factors involved insurviving dehydration would be of both theoret-ical and practical interest. Since intracellularwater is apparently a requisite for cell life, anunderstanding of the factors which determine

I A portion of this study was presented beforethe Microbiological Section of the Botanical So-ciety of America and the Mycological Society ofAmerica, New York, December, 1949 (Mazur andWeston, 1949).The study is a portion of a thesis submitted on

June, 1953, by Peter Mazur to the Faculty of Artsand Sciences of Harvard University in partial ful-fillment of the requirements for the degree of Doc-tor of Philosophy.

2Present address: Air Research and Develop-ment Command, Baltimore, Maryland.

whether cells will or will not survive the removalof this water should be of interest to both the cellphysiologist and the ecologist.With this in view a quantitative study was

made of the effects on the viability of fungousspores of the separate factors of the lyophil tech-nique (Weston, Buell, and Mazur, unpubli8heddata). One finding was that the standard freeze-drying procedure for fungous spores (Raper andAlexander, 1945) and for yeasts (Wickerham andAndreasen, 1942) killed more than 90 per cent ofthe spores of three species of fungi tested: Asper-gillus flavus, Pestalotia palmarum, and Curvularialunata. A second finding of interest was that al-though a higher percentage of P. palmarumspores survived immediately following treatmentwith certain modifications in procedure, subse-quent storage of the dehydrated spore suspensionsresulted in marked loss in viability in most in-stances.

This standard freeze-drying method involvedsuspending the spores in horse serum, freezing thesuspension rapidly by immersion in a dry icebath, and slowly dehydrating it while in thefrozen state. Low viability resulted. If either thesub-zero temperatures or the slow dehydrationwere responsible for death, then superior re-coveries should be obtained by using some methodproducing rapid dehydration from the liquidstate. Such a method is spray-drying.The rapidity of dehydration obtained is a re-

sult of the fact that in this technique an aqueoussuspension or solution is reduced to a mist inwhich the droplets have an exceedingly largesurface area. Because of the large surface area,heated air of low humidity can rapidly withdrawthe water present. It seemed possible to use theprinciple of spray-drying on a laboratory scale todehydrate suspensions of fungous spores in someproteinaceous liquid. It was decided to use horseserum as the vehicle in order to facilitate a com-parison of the results previously obtained whenspores in serum were freeze-dried with those tobe obtained when they are spray-dried.

257

on August 7, 2019 by guest

http://jb.asm.org/

Dow

nloaded from

MAZUR AND WESTON

EXPERIMENTAL METHODS

The spray-drying apparatus. To spray-dry suc-

cessfully, it was necessary to devise some way

of dispersing the spore suspension into a fine mist.Rather than forcing it through a pressure nozzle,as in industrial driers, it was decided to spray

using a partial vacuum on the principle of an

atomizer. However, the atomizer had to be con-

structed in such a way that the spray could bedrawn into a heated chamber in which dehydra-tion would take place. The apparatus utilized is

diagrammed in figure 1.

Figure 1. Diagram of the spray-drying appa-ratus showing the tube-within-a-tube arrange-ment of the nozzle. For details, see text.

The spray nozzle consisted of two lengths ofglass tubing, one (A) suspended coaxially withinthe other (B) by means of two star-shaped copperstrips (C). The outer tubing (B) was made frompyrex tubing, 15 mm in outside diameter and8 in long. The lower or nozzle end of this tube wasdrawn out to form a tapered tip with an openingof about 3 mm in diameter. The inner tube (A)was constructed from pyrex tubing 7 mm inoutside diameter, and its lower end was drawn outto form an aperture of 1 mm or slightly less. The

two tubes were positioned so that the opening ofthe inner extended just beyond that of the outer.

In the manner to be described shortly, thespore suspension was allowed to pass into theinner tube (A) and, simultaneously, heated airwas forced rapidly through the space between theinner and outer tubes. As the suspension movedthrough the small opening of the inner tube, itwas immediately broken into a very fine spray bythe jet of air passing out the tip of the outer tube(B).These two glass tubes forming the nozzle

proper were placed in a large rubber stopper wh.ichfitted into the neck of either a 5-gal pyrex carboy,or a large vacuum bell jar. These vessels servedas the drying chamber, and were heated by twoelectric reflector heaters (D) placed on oppositesides. The temperature inside the chamber wascontrolled by the distance between the carboy andthe heaters.To the untapered ends of the two glass tubes

were affixed lengths of rubber tubing (E and F) asshown in figure 1. A strong spring clamp (H) wasapplied just above the untapered end of the innerglass tubing; the spore suspension was pipettedinto the inner rubber tubing (E) above the clampand was, for the moment, prevented from movinginto the nozzle proper by that clamp. Finally,another clamp was applied at point (G) to closethe upper, or basal, end of the outer rubber tub-ing (F). This clamp functioned in preventing thepassage of any air through the nozzle and into thedrying chamber.

After the clamps had been applied, the dryingchamberwas evacuated bya "CencoHyvac" pumpattached through tube I. When most of the airhad been withdrawn, as evidenced by the collapseof the outer rubber tubing (F), the spore suspen-sion was permitted to descend into the inner glasstubing by releasing the lower clamp. Thereupon,the upper clamp was released, and the resultingblast of air reduced the suspension to fine dropletsas it emerged from the tip of the inner tube.Drying took place only if the carboy was heated

by the electric heaters mentioned above, andwas more effective if the air drawn through thenozzle was preheated. Two bunsen burners heatedthe air passing through a large (2 in by 2 ft) pyrextube (J). This air was drawn into the dryingchamber upon the removal of the clamps, partlyby the vacuum in the drying chamber, and partlyby the impelling action of a small turbo fan at-

258 [VOL. 71

on August 7, 2019 by guest

http://jb.asm.org/

Dow

nloaded from

FUNGOUS SPORE VIABILITY AND SPRAY-DRYING

tached to the distal end of the heating tube. Theair temperature as measured by a thermometerwas 100 C or higher as it entered the nozzle.

Tests showed that a portion of the dried sus-pension was swept toward the vacuum pumpthrough the tube (I). Accordingly, the dried ma-terial was collected by placing S-shaped trap tubes(K and L) in the vacuum line. The dried suspen-sion collected in the bendc of the "S." These traptubes were made from 7-mm pyrex tubing. Whenthe dehydrated product was to be stored, thelower tube (L) was fused off in vacuo with a gas-oxygen hand torch, sealing off first the end farth-est from the vacuum pump, then the other end.

Obtaining the spores and preparing suspensions.Spores of the imperfect fungi A. flavus and P.palmarum were used in the investigation. TheAspergillus cultures were descendants of a cultureisolated by Dr. C. M. Williams of HarvardUniversity and Dr. Alfred Sussman of the Uni-versity of Michigan. Dr. Kenneth Raper identi-fied the fungus as A. aus Link. The source cul-ture of P. palmarum was culture no. 78, kindlyfurnished by the Australian Mycological Panel.Cultures were grown on 75 ml of 1 per centpotato-dextrose agar in 12-cm diameter Kolleflasks which were incubated at room tempera-ture (20 to 25 C).To prepare a suspension, a loopful of mature

spores was initially dispersed in 1 ml of a 0.1 percent solution of the wetting agent, "Naccanol" NR,keryl benzene sulfonate (Schwartz and Perry,1949) manufactured by the National AnilineDivision of the Allied Chemical and Dye Corpora-tion. This surface-active agent was completelynon-toxic (control, figure 2). The resulting sus-pension was then thoroughly mixed with 5 ml ofnormal horse serum.Spray-dring the suspension. One or 2 days

prior to a run, the nozzle and all rubber and glasstubing were autoclaved or oven sterilized. Thecarboy, too large to sterilize in either an oven oran autoclave, was washed thoroughly with soapand water, rinsed with alcohol, and allowed todry for 2 days with a cotton plug in the mouth.

In a drying run, the spensions were sprayedin 5 batches of 0.5 ml each. A 0.5-ml portion waspipetted into the inner rubber tubing (E, figure1) and held there with the lower clamp (H). Theupper clamp (G) was then applied, and the ap-paratus evacuated. Next, the tube leading fromthe large pyrex tube (J), in which air was being

heated, was attached to the upper end of thenozzle, and the suspension sprayed into the heateddrying chamber by removing first the lower andthen the upper clamp. The procedure was re-peated for the other 4 portions of the suspension,the entire run requiring 15 to 20 min.The material collecting in the bends of the trap

tubes was similar in color and texture to powderedskim milk. While the amount of suspension de-posited was too scant to determine its water con-tent quantitatively, its appearance indicated alow moisture content, probably under 10 percent.

Storing the dried spension. When the effectof storage was under investigation, the traptubes containing the dried suspension were sealedin vacuo in the manner already described, andstored in a refrigerator at 3 C for up to 2 years, 8months. Prior to determining the viability ofthese spores, the tubes were tested for the pres-ence of a vacuum with a Cenco high frequencycoil of the Tesla type.

Rehydration and the detemination of viability.When studying the immediate effects of spray-drying, the suspensions were rehydrated within2 hr of being dried. The same procedure was in-volved with both the above suspensions and thosestored. The outside of the trap tube was surfacesterilized with a small amount of 70 per centethanol; a scratch was made in the center with atriangular file, and the tube was carefully brokenopen. With the aid of a transfer needle, the endof which was flattened to form a tiny spatula, thedried suspension was then transferred to a small(10- by 75-mm) test tube. To rehydrate, 2 or 3drops of room temperature 0.1 per cent Naccanolwere added with a capillary pipette, and stirredwith the transfer needle until the highly solubledried horse serum had dissolved and the sporeswere once again evenly distributed.

Spore viability was evaluated in terms ofpercentage germination. To promote germination,two drops of the rehydrated suspension wereplaced on disks of potato-dextrose agar with asterile capillary pipette and spread over thesurface. These agar disks were supported bystandard microscope slides placed in moist cham-bers. Details of preparing the disks and the moistchambers are given elsewhere (Weston, Buell,and Mazur, unpublished data).The moist chambers, in turn, were held at room

temperature to induce germination of the spores.

25919561

on August 7, 2019 by guest

http://jb.asm.org/

Dow

nloaded from

MAZUR AND WESTON

Incubation was continued until the germ tubeswere, on the average, 5 to 10 times as long as the3- to 5-,u diameter of the spores. At the end ofthis interval, which ranged from 12 to 20 hr,germination was maximal. The percentage ger-mination was determined by direct microscopicobservation. At least 200 spores were counted ina randomized manner as described by Weston,Buell, and Mazur (unpublished data). A sporewas considered to have germinated when a defi-nite germ tube could be observed. In each experi-ment, duplicate agar disks were inoculated withthe suspensions of each control and each spray-drying run.

EXPERIMENTAL RESULTS

The immediat effect of spray-drying on sporeviability. Although preliminary trials had shownthat spray-dried horse serum was deposited in thebends of the trap tubes as a dry white powder, theamount deposited was too small to permit a quan-titative measurement of the amount of water re-moved. Because of this lack of quantitative data,a finding of a high survival following spray-dry-ing could be interpreted in two ways: Eithersurvival might be high because dehydration isrelatively incomplete, or it might be high in spiteof relatively extensive dehydration-that is,because desiccation by spray-drying is inherentlyinnocuous.To distinguish between these two possibilities,

two experiments were performed: one with sporesof A. flavu1 and one with those of P. palmarum.In these experiments the apparently dry suspen-sion in the trap tubes remained in contact with adraft of heated air for various lengths of timeafter spraying had been completed. The tempera-ture of the air was above 100 C when it enteredthe drying chamber, and presumably was warmenough to exert further desiccating action on thesuspension deposited in the trap tube. Hence, afinding that viability was high immediately fol-lowing spraying, and remained high after furthercontact with the drying effect of heated air, wouldsupport the belief that the high immediate via-bility was not merely due to incomplete dehydra-tion.The suspensions were sprayed in the manner

already described. When 2.5 ml of the suspensionof spores in serum had been processed and somehad been dried and deposited in the trap tube, thevacuum pump was allowed to run for an addi-

tional 0, 10, 20 or 30 min, during which time itdrew heated air from the large pyrex heating tube(J, figure 1) through the trap tube containing thedeposited suspension. The temperature of theheated air was 102 and 108 C in the experimentsinvolving spores of A. flavus and P. palmarum,respectively. In both cases the drying chamber,prior to spraying, was heated to 65 C by the ac-tion of the two electric heaters. Upon rehydration,the contents of each trap tube were distributedevenly between two agar disks, thus giving twosamples on which to base the percentage germina-tion. For a control, spores were initially suspendedin 0.1 per cent Naccanol to which was added anequal volume of horse serum. The resulting mix-ture was not dried, but was maintained at roomtemperature during the spraying run.The spore recoveries following spray-drying and

additional heating are given in figure 2. The mostsignificant aspect of these results is the high per-centage of germination of the spray-dried spores,irrespective of the length of time they were ex-posed to the heated air. These recoveries, 64 to83 per cent for A. flavus and 77 to 90 per cent forP. palmarum, contrast markedly with those ob-

100

901

801

70

60-

50L% OF

GERMINATION40r30

20

lo

* Aflavus []R palmarum

Control 0 Min 10Min 20M;n

TIME IN CONTACT WITHHEATED AIR AFTER SPRAYING

30 Min

Figure 2. The percentage germination of spray-dried spores of Aspergillus flaws and Pestalotiapalmarum after contact with heated air for vari-ous lengths of time following the completion ofspraying.

260 [VOL. 71

on August 7, 2019 by guest

http://jb.asm.org/

Dow

nloaded from

FUNGOUS SPORE VIABILITY AND SPRAY-DRYING

tained when these spores were freeze-dried inhorse serum, 2 to 4 per cent for A. flavus andless than 10 per cent for P. palmarum (Weston,Buell, and Mazur, unpublished data).With the spray-dried spores, however, there

was an interesting relationship between the lengthof time the spores were exposed to heated air andthe percentage germination. The spores of bothspecies exhibited this relationship (figure 2);namely, the percentage of germination decreasedas the length of exposure to heated air followingspraying was increased from 0 to 10 min; itthen increased again as the time was increased to30 min. The pattern appears significant, for itoccurred in both experiments; however, its basisremains obscure.These differences in recovery are, we believe,

of less significance than the fact that the per-

centage germination of all the spray-dried sporeswas much higher than the percentage germination

following freeze-drying, and also continued to bemuch higher with exposure to the desiccatingaction of additional heated air. Even if t4he spore

suspensions were further dehydrated by theheated air, any increased dehydration loweredthe viability only slightly.As a further check on the results of the above

experiments, another was performed involving a

somewhat different approach to the question ofthe extent of dehydration as a factor in viability.The temperature of the drying chamber intowhich the spore suspensions are sprayed would beexpected to influence both the rate and the ex-tent of dehydration; therefore, an experimentwas performed to see whether various tempera-tures would produce corresponding changes inviability.

It was observed in preliminary trials that a

sprayed suspension underwent little or no de-hydration when the drying chamber was below35 C prior to spraying. The lack of desiccationwas manifested either by the absence of anydeposit in the trap tubes, or by the deposition ofviscous liquid droplets rather than white powder.Dehydrated serum began to be deposited, how-ever, when the temperature of the drying chamberwas raised above 40 C, and was deposited in in-creasing amounts as the temperature was raisedhigher.

Using spores of A. flau, three runs were

made in which the temperatures in the center ofthe drying chamber, just below the outlet of the

nozzle, were 43, 59, and 81 C. These temperatureswere obtained by placing the two electric heatersvarious distances from the outside wall of thedrying chamber.The spores were suspended in horse serum and

sprayed in the usual way. The air which passedthrough the nozzle was preheated to 97 C. Thetrap tubes, in which the spore suspension was-deposited, were removed from the apparatus im-mediately after the completion of spraying, andthe suspension then rehydrated to permit ger-mination as previously described. Once again, forcontrol, spores were suspended in a 50:50 mix-ture of Naccanol and serum, but were not spray-dried.The results, given in table 1, may be sum-

marized briefly by stating that nearly 100 percent of the A. flas spores survived, irrespectiveof the temperature of the drying chamber. Yetit is likely that the suspension of spores in serumwas more extensively dehydrated with a dryingchamber temperature of 81 C than with a tem-perature of 43 C.

TABLE 1Relationship between the percentage germination of

spray-dried spores of Aspergillusflavus andthe temperature of the drying chamber

Temperature of % Germination of Mean %Drying Chamber Each Sample Germination

C

43 96.0, 96.2 96.159 98.1, 96.7 97.481 96.1, t 96.1

Control* 97.6, 95.8 96.7

Percentages of germination were based oncounts of from 210 to 410 spores for each sample.

* The control consisted of spores suspended ina 50:50 mixture of Naccanol and horse serum, heldat room temperature and not dried.

t One sample lost through accident.

The three experiments reported above showthat from 60 to 100 per cent of the spores of A.flavus and P. palmarum survive the rapid de-hydration occurring in a laboratory-scale spray-drier; moreover, the data suggest that the highsurvival is not a result of incomplete dehydra-tion. Rather, it seems to be a result of the methodof effecting desiccation; namely, rapid dryingfrom the liquid state.Weston et al. (1955) have found that P. pal-

19561 261

on August 7, 2019 by guest

http://jb.asm.org/

Dow

nloaded from

MAZUR AND WESTON

marum spores can under certain conditions sur-vive the immediate effects of freeze-drying, andyet fail to survive subsequent long-term storage.There existed the possibility, therefore, that thehigh survival of the spores immediately followingspray-drying would also decrease upon storage.

The effect of prolonged storage on viaiity.Spray-dried spore8. In order that the viabilityof spores, or any cells, be conserved by laboratorydesiccation for periods of time longer than foundin nature, it appears essential that sufficientwater be removed to render the cells less suscep-tible to both their internal and their externalenvironment. It seemed important, therefore,to determine the storage stability of spray-driedspores, for such a study would provide evidenceas to whether the resulting dehydration is suf-ficient to promote stability.To obtain the material for a storage experi-

ment, suspensions of A. flaus spores in horseserum were spray-dried in the manner alreadydescribed. The temperature in the drying cham-ber was 70 C. Ten spraying runs were made. Ineach case, immediately after spraying had beencompleted, the trap tube was fused off at bothends while evacuated. After a check with a. highfrequency coil had shown that the tubes had re-tained a vacuum, 8 of them were placed in arefrigerator at 3 C for storage. The. remaining 2tubes were then broken open to determine theimmediate viability of the spores.

Six months later, one of the sealed tubes wasremoved from the refrigerator and, after a checkwith the coil had shown that vacuum was stillpresent, the spores were rehydrated and theirviability determined in the usual fashion.Similar determinations were made after storageperiods of 14, 25, and 32 mo. The percentages ofrecovery after 0, 6, 14, 25 and 32 mo were 73,67, 50, 61 and 40, respectively.

Comparison of storage survival of spray-dried,freez-dried, and naturaUy air-dried spores of A.flavu. It was felt that the data obtained on thestorage survival of spray-dried A. flavus sporeswould be more meaningful if they could be com-pared with data obtained on the storage survivalof freeze-dried spores and of spores left on nutrientagar.

Suspensions of these spores in horse serum hadbeen freeze-dried and their immediate viabilitydetermined by Buell (1948). Evacuated lyophiltubes containing these spores were also main-tained at 3 C for 54 mo and the viability of thespores then determined by the methods described

earlier in the present report. Before storage 1.8per cent were viable; after storage, only 0.14per cent.

Cultures of A. flavus grown on the usual po-tato-dextrose agar (1% or 2 per cent agar) wereused to determine the longevity of naturally air-dried spores. The cultures were maintained atroom temperature, 20 to 27 C, for various lengthsof time up to 32 mo. One week after inoculation,sporulation was heavy. As time progressed theagar dried to a shrunken film, and the sporescollected as a fine powder on the surface of thedried agar. The viability was ascertained bytransferring a small amount of the sporulatingmaterial from each culture to a sterile serumtube (10 by 75 mm), containing 1 ml of sterile0.1 per cent Naccanol NR to aid in suspendingthe spores. Using graduated pipettes, 0.03-mlaliquots of the resulting suspensions were placedon each of three agar disks. Germination was in-duced and the percentage germination deter-mined in the usual way. After storage for X,1%, 4, 10 and 32 months, the recoveries were99, 95, 53, 72 and 1.8 per cent.The effects of storage on spray-dried, freeze-

dried, and naturally air-dried spores are sum-marized in figure 3.

z0

4z

LUcc

Li0D4I--zLiO

LIta.

* FREEZE- DRIEDo SPRAY-DRIED0 AIR-DRIED

% LENGTH OF STORAGE (MONTHS)

Figure S. The percentage germination of Asper-gillus flavus spores after storage for variouslengths of time following various treatments.

262 [VOL. 71

on August 7, 2019 by guest

http://jb.asm.org/

Dow

nloaded from

FUNGOUS SPORE VIABILITY AND SPRAY-DRYING

Examining first the storage stability of thespray-dried A. flam spores, it is evident that,in general, these spores survived admirably evenafter 32 mo in a desiccated state. Not only didthe percentage germination remain high, but alsoa culture derived from 32-mo material grewluxuriantly and sporulated in a normal manner.The data do seem to show, however, a slow de-cline in viability from 73 to 41 per cent as thelength of storage increased from 0 to 32 mo.Whether this decline will continue will be deter-mined by experiments in the future.These results with spray-dried spores contrast

markedly with the storage stability of freeze-dried and naturally air-dried spores. Not onlywas survival very low immediately followingfreeze-drying in horse serum (1.8 per cent), butit decreased 10-fold to 0.14 per cent after storagefor 54 mo.

Spores maintained on agar behaved differently.Initially, nearly 100 per cent were capable ofgerinating, but viability dropped comparativelyrapidly with the passage of time. Finally, after 32mo, only 1.8 per cent of the spores in the culturegerminated while 40 per cent of spray-dried ma-terial was viable; yet, the spores of the latter hadactually been obtained from that very agar cul-ture some 32 mo earlier.These experiments on storage stability show

clearly that the viability of A.flaus spores spray-dried in serum is markedly higher than that offreeze-dried spores both immediately after treat-ment- and after prolonged storage. They alsoshow that the viability of spray-dried sporesremains higher than that of naturally air-driedspores maintained under environmental condi-tions simulating those found in nature.

DISCUSSION

Spray-drying differs from freeze-drying in twoessential respects. In the first place, dehydrationis effected rapidly in the former, but slowly inthe latter. Secondly, the temperatures involvedin spray-drying result in dehydration from theliquid state; whereas, in freeze-drying, the tem-peratures are such that the water is completelyor partially frozen and dehydration is by sub-limation.We have seen that these two basically different

processe exert markedly different effects on theviability of spores of A. jlavwu and P. palmarum.While only from 1 to 10 per cent survive freeze-drying in serum, 60 to 98 per cent survive theimmediate effects of spray-drying in serum.

Since the techniques of growth and asay and alsothe suspending vehicle were the same in bothcases, the marked difference seems most likely todepend on one or more of the following possi-bilities: (1) the difference in the rate of dehydra-tion; (2) the presence or absence of freezingtemperatures and the consequent effect on thephysical state of the spore suspension; and (3) adifference in the extent of dehydration achievedby the two methods.The available evidence makes it unlikely that

the superiority of spray-drying over freeze-drying is a result of the third possibility; namely,a difference in- the extent of dehydration. In thefirst place, the spraying process dehydrated thesuspensions of spores in serum to the extent offorming a white powder appearing as dry as thepellet produced by freeze-drying. In the secondplace, the viability of spray-dried spores wasfound to remain high irrespective of the timethey were exposed to the desiccating action ofheated air. Similarly, viability was unaffectedby the temperature of the drying chamber. Ifthe spores had survived spray-drying only be-cause dehydration was incomplete, one wouldhave expected viability to have decreased whenthe spores were exposed to heated air for longertimes or after they had been sprayed into awarmer drying chamber. Finally, the superiorstorage stability of spray-dried A. flavus sporeswould have been unlikely had dehydration beenless thorough than in the case of freeze-drying.In support of this last statement, Weston et al.(1955) found that the storage stability of P.palmarum spores was better when dehydrationwas more extensive.

If the extent of dehydration is not a factor,then the harmful effects of freeze-drying wouldappear to be a consequence of either the slow rateof dehydration, (1) above, or else of the presenceof freezing, (2) above. On the basis of the datareported, we cannot decide which of these twoaccounts for the low recoveries of freeze-dried A.flavus spores. There is evidence, however, indi-cating that the harmful effects of freeze-dryingon P. palmarum spores are a result of slow de-hydration. Thus, an analysis of the effects of theseparate factors in the freeze-drying technique(Weston, Buell, and Mazur, unpublished data)showed that slow dehydration was primarily re-sponsible for death of spores of the latter. It wasalso found, on the other hand, that merely freez-ing at -78 C did little or no damage. Moreover,the data given in figure 2 show that the spores

26319561

on August 7, 2019 by guest

http://jb.asm.org/

Dow

nloaded from

MAZUR AND WESTON

suffered little reduction in viability when rapidlydehydrated by spray-drying. We are led to be-lieve, therefore, that P. palmarum spores are notkilled by dehydration; rather, they are killed byslow dehydration.

It seems justifiable at this point to speculate,briefly, as to why the slow dehydration in freeze-drying should be harmful. The simplest explana-tion is that slow desiccation exposes the sporesto a relatively high concentration of solutes formany minutes, whereas the exposure time inspray-drying is on the order of a second. It maybe argued, with justice, that freeze-drying oughtnot to expose cells to any concentrated liquidsolution, for dehydration is supposed to takeplace from the solid state. In the freeze-dryingtechnique referred to in the introduction, how-ever, the spore suspension is held in a bath main-tained at -5 to -10 C during desiccation. As a

result, it will probably be only partially frozen,and will consist of two phases; ice and a concen-

trated liquid solution. The basis for this state-ment is that the chief ionic components of serumare sodium and chloride ions, and they will bethe chief determiners of the freezing behavior ofthe suspension. Pure sodium chloride solutionsbecome completely olidified only below -21 C(the eutectic point). A temperature of -21 C,then, should represent the highest temperatureat which horse serum could be completely frozen.Future experimentation should confirm or

contradict the above speculation as to why slowdehydration is harmful to P. palmarum spores.The basis of the injurious effects of freeze-dryingon A. flavu8 spores, on the other hand, has beenstudied (Mazur, 1953) and will be the subjectof future papers.The primary purpose of studying spray-drying

was to investigate the basic effects of dehydrationon viability. However, the initial and continuinghigh percentage recovery of spray-dried sporesraises the question of whether this techniqueoffers a practical means of conserving the viabil-ity of microorganisms.The relatively crude apparatus used in these

preliminary studies is not practical. First of all,only a very small percentage of the 2.5 ml of sus-

pension sprayed into the carboy was recovered inthe trap tubes as a dry product. The major

fraction fell to the bottom of the chamber as a

liquid. In the second place, in contrast to a

laboratory lyophil apparatus, only one sealedevacuated tube could be obtained in a given run.

However, if spray-drying proves capable of con-serving the viability of organisms which succumbto freeze-drying, it might be worth while to at-tempt to design an apparatus eliminating boththese drawbacks. Such an apparatus might be auseful adjunct to a freeze-drier in the preservationof cultures. "Laboratory" spray-driers have beendesigned by Wilkinson et al. (1942), and Wood-cock and Tessier (1943), and these driers haveyielded efficient recoveries of dried sera, plasma,milk, and other biologicals. They might wellprove valuable to those interested in the produc-tion of comparatively large quantities of driedliving organisms; however, they are rather elab-orate and costly, and do not appear adaptable tothe simultaneous drying of small quantities ofmany samples of microorganisms.

It is difficult to determine the degree to whichthe findings reported in the present study can beextended to cells other than the spores of A. Jlavusand P. palmarum. True, there have been ex-tensive bacteriological studies on importantspray-dried foods such as milk (Crossley andJohnson, 1942; Macy, 1928) and egg (Gibbonsand Moore, 1944; Watson, 1948; McFarlane et al.,1945; Hartsell, 1944). But for the most part,these investigators have been concerned withthe effects of spray-drying on the mixed popula-tions of bacteria naturally occurring in eggs andmilk rather than its effect on single species as inthe present study.Although the manner in which spray-drying

affects fungi has not been studied heretofore, itseffect on several bacterial species has beendescribed. Gibbons and Moore (1944) found that99.99 per cent of Salmonella organisms in wholeegg were killed by spray-drying using air heatedto 121 C. Apparently, Salmonella is much moresusceptible to the treatment than the other or-ganisms found in whole egg, for only 65 to 82 percent of this mixed population was killed. Cellsof Bacterium lactie aerogene8 also succumbed tospray-drying. Even with air temperatures aslow as 60 C, 99.6 per cent were killed, accordingto work of Bullock and Lightbown (1947). Inmarked contrast to the fate of these vegetativebacteria, bacterial spores appear much more re-sistant. Bullock and Lightbown (1947) reportthat only 8 to 10 per cent of the spores of BaciUlussubtilis were killed by spray-drying in spite of airtemperatures as high as 170 C (the air tempera-ture was 100 C in the present study).We have seen that the spores of the fungi A.

264 [VOL. 71

on August 7, 2019 by guest

http://jb.asm.org/

Dow

nloaded from

FUNGOUS SPORE VIABILITY AND SPRAY-DRYING

lavus and P. palmarum are also highly resistantto spray-drying. But although surviving thisrapid dehydration from the liquid state, theysuccumb to the slow dehydration from the frozenstate involved in freeze-drying. Perhaps, there-fore, the technique of spray-drying is worthy offurther studies-studies both to investigate thebasic effects of dehydration on cell viability andalso to investigate the possibility that spray-drying may be of practical value as a method ofconserving the viability of microorganisms.

A simple inexpensive laboratory experimentalspray-drier, spraying on the principle of anatomizer, was developed. The spray nozzle con-sisted of two lengths of pyrex tubing with onesuspended coaxially within the other. The nozzlewas placed in the top of a 5-gal carboy whichserved as a drying chamber. To spray, thedrying chamber was evacuated; then, by re-leasing the vacuum, the spore suspension wasforced through the inner tube and broken into afine mist by a jet of heated air drawn through theouter tube. The dehydrated material passed outof the carboy toward the vacuum pump and col-lected in the bends of "S"-shaped trap tubesplaced between the drying chamber and thepump.The apparatus described was used to dehy-

drate horse serum suspensions of spores of Asper-gillus flavus and Pestalotia palmarum. More than80 per cent of the spores were found to survivethe initial spraying and more than 65 per centwere viable even when the dried powder in thetrap tubes was exposed to heated air for lengthsof time up to 30 min. Other data show that fewerthan 10 per cent of the spores of A. lavus werekilled by spray-drying irrespective of whether thetemperature in the drying chamber was 43, 59or 81 C.

Spores of A. flavu8 were spray-dried in horseserum and stored in vacuo at 3 C for as long as32 mo. In this period of time, the percentagegermination decreased from 73 to 40 per cent.The death rate of naturally air-dried sporesstored on agar for the same length of time wasmuch greater. The viability of freeze-dried A. Jla-vus spores was not only very low (1.8 per cent)immediately after treatment, but also decreasedto 0.14 per cent after 54 mo of storage in vacuo.The results are interpreted as indicating that

the high survival following spray-drying is not a

result of incompleteness of dehydration, butrather is an indication that the rapid dehydrationfrom the liquid state involved in spray-drying isconsiderably less harmful to the spores than theslow dehydration from the frozen state occurringin freeze-drying.

REFERENCES

ATKIN, L., MosEs, W., AND GRAY, P. P. 1949The preservation of yeast cultures by lyo-philization. J. Bacteriol., 57, 575-578.

BUELL, CAROLINE B. 1948 Studies of the effectof the lyophil process on fungous spores.M.A. Thesis, Wellesley College, Wellesley,Massachusetts.

BULLOCK, K. AND LIGHTBOWN, J. W. 1947 Theinactivation of enzymes and micro-organismsin oils and powders. II. The effects of dry-ing on the viability of bacteria and theirthermolability in powders. Quart. J. Pharm.and Pharmacol., 20, 313-328.

CROSSLEY, E. L. AND JOHNSON, W. A. 1942Bacteriological aspects of the manufacture ofspray-dried milk and whey powders, includingsome observations concerning moisture con-tent and solubility. J. Dairy Research, 13,5-44.

ELSER, W. J., THOMAS, RUTH A., AND STEFFEN,G. I. 1935 The desiccation of sera and otherbiological products (including microorgan-isms) in the frozen state with the preservationof the original qualities of products so treated.J. Immunol., 28, 433-473.

FLOSDOR?, E. W. 1949 Freeze-drying (drying bysublimation). Reinhold Publ. Co., New York.

GIBBONS, N. E. AND MOORE, R. L. 1944 Driedwhole egg powder. XI. Occurrence and dis-tribution of Salmonella organisms in Canadianpowder. XII. The effect of drying, storage,and cooking on the Salmonella content. Can.J. Research, F, 22, 48-63.

GLOVER, R. E. 1946 The effects of (a) freeze-drying and (b) low temperature on the via-bility of Mycobacterium tuberculosis. J. Pa-thol. Bacteriol., 58, 111-114.

HARTSELL, S. E. 1944 Studies on the bacteriol-ogy of stored, dried egg powder. Food Re-search, 9, 505-511.

MACY, H. 1928 Some observations on the bac-terial content of dried milk. J. Dairy Sci.,11, 516-526.

MAZUR, P. 1953 Studies on the effects of lowtemperatures and dehydration on the viabil-ity of fungous spores. Ph.D. Thesis, Har-vard University, Cambridge, Mass.

MAZUR, P. AND WESTON, W. H., JR. 1949 Ef-

19561 265

on August 7, 2019 by guest

http://jb.asm.org/

Dow

nloaded from

MAZUR AND WESTON

fects of spray-drying on the viability of cer-tain fungous spores (abstract). Am. J.Botany, 36, 811-812.

MCFARLANE, V. H., WATSON, ALICE J., AND GORE-BLINE, H. E. 1945 Microbiological controlin the production of spray dried whole eggpowder. U. S. Egg and Poultry Mag., 51,250-257, 270-273, 275-277, 279-286.

RAPER, K. B. AND ALEXANDER, DOROTHY F.1945 Preservation of molds by the lyophilprocess. Mycologia, 37, 499-525.

SCHWARTZ, A. M. AND PERRY, J. W. 1949 Sur-face active agents. Their chemistry and tech-nology. Interscience Publ., New York.

WATSON, ALICE J. 1948 Microbiology of spray-dried, whole-egg powder. V. Aerobic meso-philic sporeforming bacilli isolated at 56 C.J. Bacteriol., 56, 696.

WICKERHAM, L. J. AND ANDREASEN, A. A. 1942The lyophil process: Its use in the preserva-

tion of yeasts. Wallerstein Lab. Commun.,5, 165-169. VA

WILKINSON, J. F., BULLOCK, K., AND COWEN, W.1942 Continuous method of drying plasmaand serum. Lancet, 242, 281-284.

WOODCOCK, A. H. AND TESSIER, H. 1943 Alaboratory spray drier. Can. J. Research, A,21, 75-78.

266 [vOL. 71

on August 7, 2019 by guest

http://jb.asm.org/

Dow

nloaded from