APPLIED MICROBIOLOGY, Nov., 1965Copyright © 1965 American Society for Microbiology

Vol. 13, No. 6Printed in U.S.A.

Storage Stability of Clostridium botulinum Toxin andSpores in Processed Cheese

NICHOLAS GRECZ, R. 0. WAGENAAR,' AND G. M. DACK2

Biophysics Laboratory, Biology Department, Illinois Institute of Technology, Chicago, Illinois

Received for publication 23 July 1965

ABSTRACTGRECZ, NICHOLAS (Biophysics Laboratory, Illinois Institute of Technology, Chicago,

Ill.), R. 0. WAGENAAR, AND G. M. DACK. Storage stability of Clostridium botulinumtoxin and spores in processed cheese. Appl. Microbiol. 13:1014-1022. 1965.-Growth initi-ated from detoxified spores of Clostridium botulinum 62A resulted in toxin productionof 50 to 10,000 mouse lethal doses (MLD) per gram of processed soft surface-ripenedcheese. Regular assays during subsequent storage of toxic samples at 2 to 4 C revealeda characteristic two- to fivefold increase in toxin titer during the initial 1 week to 12months of storage. Thereafter, the toxin titer remained constant for 2 to 4 years, afterwhich the toxicity declined rapidly. At the end of 6 years of storage at 2 to 4 C, thesamples still contained 20 to 5,000 MLD of toxin per gram, with the usual toxin levelat 200 to 500 MLD. Toxic culture filtrates of C. botulinum incorporated into cheese andstored at 30 C for 60 days showed no decline in toxin in processed type I cheese, buttoxin decreased slightly in processed type II and type III cheese. The surface flora ofthese cheeses did not attack but, on the contrary, protected C. botulinum toxin duringstorage at 30 C. Initial difficulties in recovering C. botulinum organisms from type Icheese were traced to growth inhibitory activity which could be removed by washingwith distilled water in a centrifuge. Viable spores or vegetative cells could be recoveredfrom all samples after 4 to 5 years of storage at 2 to 4 C. After 6 years, organisms were

recovered from all except three samples of type I cheese. Two other samples showed a

large decrease in viable organisms. In type III cheese, spores remained remarkablystable for 6 years at the level of the initial inoculum, i.e., approximately 105 sporesper gram.

This paper presents some additional findingsobserved during a long-term investigationprompted by the occurrence of Clostridium bot-ulinum toxin in vacuum-packed cheese spread(Meyer and Eddie, 1951). This singulax occur-rence stimulated interest in a research programdesigned to study the many factors influencinggrowth and toxin production in cheese spreadsexperimentally inoculated with spores of C. bot-ulinum types A and B (Wagenaar and Dack,1958).During the course of experiments in which C.

botulinum produced toxin in experimental cheesepreparations, there was doubt regarding thestability of the toxin during subsequent incuba-tion at 30 C, or during refrigerated storage at 2 to4 C. The present study was designed to test thestability of toxin under these two storage condi-tions.

1 Present address: General Mills, Inc., ResearchCenter, Minneapolis, Minn.

2 Present address: Food Research Institute,University of Chicago, Chicago, Ill.

MATERIALS AND METHODS

Cultural methods. Two cultures of type A C. botu-linum were employed: strain 62 A from the culturecollection of K. F. Meyer, and strain 33 A fromW. E. Perkins, National Canners Association,Berkeley, Calif. The methods of preparing sporesand maintaining stock cultures were describedelsewhere (Grecz, Wagenaar, and Dack, 1959a).

Cheese preparations. Cheese preparations weremade as described by Wagenaar and Dack (1958)from surface-ripened cheeses designated as typesI, II, and III. The surface flora of types I and IIIwas composed of yeasts and bacteria; type II wasmold-ripened. The experimental cheese prepara-tions differed from commercial cheese spread inthat no emulsifier was added, the moisture andsalt content were varied to suit the experimentaldesign, and the cheese was inoculated with heat-shocked spores of C. botulinum tubed in 30-g por-tions and stored under anaerobic conditions at30C.

In general, experimental cheese was prepared asfollows. Sufficient water was added to the basicproduct to yield a favorable moisture content forgrowth of C. botulinum: 10% to type I, 15% to

1014

on May 26, 2020 by guest

http://aem.asm

.org/D

ownloaded from

STABILITY OF C. BOTULINUM TOXIN AND SPORES

type II, and 20% to type III. The cheese was

heated at 70 C and was thoroughly stirred to ob-tain a homogeneous mixture. Two 100-g portionswere weighed into 400-ml beakers, and the cheesewas either pasteurized by heating at 90 C for 10min, sterilized at 121 C for 10 min, or left un-

heated. The cheese was cooled to approximately50 C, restored to the original weight with steriledistilled water, and inoculated by thorough mixingwith heat-shocked spores or with sterile crudetoxin of C. botulinum 62 A. The inoculated cheesewas weighed in 30-g portions into sterile culturetubes (25 by 150 mm) and incubated anaerobicallyin desiccator jars at 30 C. Anaerobiosis was estab-lished by use of 25 g of pyrogallic acid mixed with25 g of sodium carbonate and 100 ml of water, andevacuation of the atmosphere was done withDack's anaerobic apparatus (Jordan and Burrows,1945). Before sealing, 10% CO2 was added to thejar. Samples were removed for toxin and pH anal-yses after 3, 7, 14, 30, and 60 days of incubation.

Spore inoculum. Spores of C. botulinum 62 Awere heated at 85 C for 10 min prior to incorpora-tion into the cheese. Heating was necessary todestroy any heat-labile toxin and viable vegeta-tive cells in the spore suspension, as well as toactivate the spores for germination. The inoculumconsisted of 100 spores per gram of type I cheese,and 100,000 spores per gram of type III. Generally,type III was the poorer medium for growth ofC. botulinum and, hence, usually required a largerinitial spore inoculum for growth and toxin pro-duction.

After appropriate periods of incubation at30 C, the tubes were transferred to a refrigerator,and the samples were analyzed for toxin after 0,0.5, 1, 2, 3, 4, 5, and 6 years of storage at 2 to 4 C.During the latter part of the storage period an

attempt was made to recover viable organismsfrom the stored cheese samples.

Sterile crude toxin. Supernatant fluid from cul-tures of C. botulinum 62 A grown in 5% Trypticase(BBL) broth medium was filtered through quali-tative grade filter paper to remove suspendedsolids. The clear broth was then sterilized by filter-ing through a Seitz filter. A representative portionof the sterile filtrate was titrated for its approxi-mate level of C. botulinum toxin by intraperitonealinjections of appropriate dilutions into whitemice. The quantity of filtrate needed to give thecheese the desired initial toxicity was calculatedfrom the results. The cheese preparations were

cooled to below 50 C before the toxic filtrate was

added. The cheese was incubated at 30 C, and thetoxin titer was determined after 0, 3, 7, 14, 30, and60 days.

Storage conditions. C. botulinum toxin may bedestroyed by sunlight under certain conditions,e.g., in the presence of methylene blue (Weil et al.,1957). All storage in the present experiment was

carried out in the dark, however, and no specialprecautions were taken during normal handlingto exclude light.

Toxin assay. Samples were thoroughly trit-

urated in gelatin-phosphate buffer solution com-posed of 0.2% gelatin, 0.73% NaH2PO4- H20, and0.37% Na2HPO4, pH 6.8, after autoclaving for20 min at 121 C (Naylor and Smith, 1946). Thegelatin protein appeared to minimize inactivationof toxin and to reduce nonspecific reactions inmice (Wentzel, Sterne, and Polson, 1950; Boor,Tresselt, and Shantz, 1955).The quantity of toxin was determined by intra-

peritoneal injection into duplicate mice of 0.4, 0.5,and 1.0 ml of appropriate dilutions of the sample.The death of any one of the mice indicated a posi-tive reaction when a similarly injected animal wasprotected with 0.1 ml of homologous antitoxin(Jensen-Salsbery Laboratories, Inc., Kansas City,Mo.). Tests for toxin titers were standardized atfixed levels of 10, 20, 50, and 100 mouse lethaldoses (MLD) plus decimal multiples of theselevels.

RESULTSShort-term study. The study of the stability of

C. botulinum toxin in soft surface-ripened cheeseduring storage at 2 to 4 C was prompted by theoccasional observation that the toxin titer of somesamples stored in a refrigerator appeared to in-crease in repeat assays. A series of systematicrechecks after storage for 1 to 2 weeks at 2 to 4 Cconfirmed this observation. The results indicatedthat the initial toxin titer either remained con-stant (in the majority of the samples tested) oroccasionally increased by a factor of 2.5 to 5(Table 1). In no case was there a decrease in theinitial toxin titer. These data suggested that per-haps more samples would show an increase intoxin titer if the storage period were prolonged.

Long-term study. The initial small-scale inves-tigation was followed by long-term storage oi

TABLE 1. Increase in the titer of Clostridiumbotulinum toxin in soft surface-ripened cheeseduring storage at 2 to 4 C (short-term study)

Toxin titer (MLD/g)Sample no.* Storage at2 to 4 Ct

Initial Recheck

days

11,266 2,000 5,000 1311,342 2,000 5,000 911,391 2,000 10,000 711,392 <5 20 7

* Sample numbers pertain to samples from aproject concerned with factors affecting growthand toxin production of C. botulinum in soft sur-face-ripened cheese (Wagenaar and Dack, 1958).All above samples contained type A toxin pro-duced by C. botulinum 62 A.

t Days elapsed between initial toxin assay andthe recheck after storage at 2 to 4 C.

1015VOL. 13, 1965

on May 26, 2020 by guest

http://aem.asm

.org/D

ownloaded from

GRECZ, WAGENAAR, AND DACK APPL. MICROBIOL.

TABLE 2. Stability of Clostridium botulinum toxin in soft surface-ripened cheese preparationsstored at 2 to 4 C (long-term study)*

Toxicity of cheese (MLD/g)

Type of Expt Sample Initial toxin assay 5 Dec em- 9 June 12 May 15 May 31 July 18 July 29 Julycheese no. no. her 1958 9Jn 2My1 a 1Jl 8Jl 9Jl, (ca 0 5 1959 1960 1961 1962 1963 1964

Date MLD yr) (1 yr) (2 yr) (3 yr) (4 yr) (5 yr) (6 yr)

Type I 130 12,733 20 May 1958 2,000 5,000 5,000 5,000 5,000 5,000 2,000 500129 12,749 6 June 1958 10,000 20,000 20,000 20,000 10,000 5,000 2,000 200129 12,750 6 June 1958 2,000 5,000 5,000 5,000 2,000 2,000 2,000 200129 12,753 6 June 1958 5,000 5,000 10,000 10,000 10,000 10,000 SEt SE129 12,754 6 June 1958 5,000 10,000 10,000 10,000 10,000 5,000 1,000 500131 12,870 26 August 1958 1,000 1,000 2,000 2,000 2,000 1,000 500 200131 12,878 15 September 10,000 20,000 20,000 20,000 20,000 20,000 10,000 5,000

1958

Type 35 12,785 25 June 1958 200 500 500 500 500 500 SE SEIII 35 12,789 25 June 1958 200 500 500 500 500 500 200 20

35 12,790 16 June 1958 50 200 200 200 200 200 SE SE

* Clostridium botulinum 62 A was used to produce toxin in the cheese. Samples of type I cheese wereinoculated with 100 spores per gram, whereas type III cheese received 100,000 spores per gram. Toxintiter was assayed by intraperitoneal injection of appropriate dilutions of toxic cheese into duplicatewhite mice. The specificity of toxin was established by neutralization with type-specific antitoxin.

t SE = Sample exhausted.

10'

10°

zx

2 10''

1-2

0 2 3 4 5 6

YEARS AT 2 - 4 C

FIG. 1. Changes in toxin titer in two types ofsurface-ripened cheese during storage at 2 to 4 C.Toxin titers were expressed in fractional changesN/No, where N was the toxin titer in MLD at a

given time, and No was the toxin titer at the start ofthe storage period.

toxic cheese samples for 6 years. Periodic assaysduring storage at 2 to 4 C revealed a two- to five-fold increase in toxin titer within the first 12months in every one of the cheese samples inves-tigated (Table 2). Thereafter, the toxin titer re-

mained constant at its peak level for 1.5 to 3.5years. All samples showed a reduction in toxintiter after 5 years and, to a progressively greaterextent, after 6 years of storage at 2 to 4 C. Figure1 presents the fractional changes in toxin titerduring 6 years of storage at 2 to 4 C in type I and

10'

100

z

z

a

z

0

102

0 2 3 4

YEARS AT 2 - 4C

5 6

FIG. 2. Decline in toxin titer in three experi-mental preparations of type I cheese.

III cheeses. Most of the original samples havebeen used up; the few remaining samples will betested after 10 years of storage.The rate of decline of toxin titer in type- I

cheese from experiment 129 appeared to be morerapid than in samples from parallel experiments130 and 131 (Fig. 2). Examination of the changesin toxin titer of the original experiment 129 dur-ing incubation at 30 C suggested that in thischeese the toxin tended to decline consistently

1016

o 0 0 0 0

* *

00

BOTULINUM TOXIN TITER IN0

TYPE CHEESEO----O0 TYPE III CHEESE

*~ ~ ~ ~~~ a **f* t

TYPE CHEESE

t-->bs~~~~~~~--

AVERAGES FOR EXPERIMENTS

130 & 131 (3 SAMPLES)

O------ 129 (4 SAMPLES)

I on May 26, 2020 by guest

http://aem.asm

.org/D

ownloaded from

STABILITY OF C. BOTULINYUM TOXIN AND SPORES

TABLE 3. Developmnent and subsequent stability ofClostrid iuni botulinuin toxin in type I

cheese (luring incubation at 30 C

Toxin titer in thousands of AILDExpt No. of sampleno.

3 7 14 30 60days* days days days days

129 12,) 53 (60)t 0 0.5 10 10 5129 12,754 (60) 0 0 5 10 5130 12,733 (14) 0 0 2 2 2131 12,870 (14) 0 0 1 2 5131 12,878 (30) 0 5 10 10 10

* Days of incubation at 30 C prior to toxini assay.t Numbers in parentheses indicate the sample

used for the study of toxin stability at 2 to 4 C(cf. Table 2).

3 7 14 30 6010

0 02878

10

I10

:E _ 89'

3 7 14 30 60

DAYS AT 30

FIG. 3. Relation7 of the stage of toxin synth,esi.s byClostridilium, botulinum 62 A in cheese (luring incu-batil'on at 30 C to the initial increase in toxin titerdulring subsequent storage of the cheese samiples at2 to 4 C. The curves in(licate changes in toxin titerduring incuibation at 30 C; the arrows are placedl atthe respective staie at w,hich a sample uas trans-ferred to storage at 2 to 4 C. The lengths of the arrowsindicate the mnagnitude of initial increase of toxicityin the respective samnples wvithin 6 to 12 mnonths ofstorage at 2 to 4 C.

after reaching the peak titer for this batch ofcheese, i.e., 10,000 MLD (Table 3, samples12,753 and 12,754). In the parallel experiments130 and, especially, 131, toxin was more stable at2 to 4 C as well as at 30 C (Table 3, samples12,733, 12,870, and 12,878).The initial increase in toxin titer during storage

at 2 to 4 C appeared to be independent of theduration of incubation of the inoculated cheese at30 C prior to transfer to a refrigerator. Incubationat 30 C of different samlples N-as terminated after

3 7 14 30 60

A

10e530 33 c 10

E-10010 ___________J~~~~12 3 4 5 6

YEARS AT 2-4C

FIG. 4. Changes in toxin titer in type I cheeseinoculated with Clostridiumi botulinum 62 A duringincntbation at 30 C (graph, A), as relatedl to stabilityof the pr eformeii ed toxin in the 14-day sample (no.12,870) during subsequent storage at 2 to 4 C for6 years (insert B).

2 3 A 5 6

YEARS AT 2 -AC

FIG. 5. Changes in toxin titer in type I cheeseinoculated with Clostridium botulinumt 62 4 (luringincubation at 30 C (graph A), as related to stabilityof the preformtied toxin in the 30-day samlple (no.12,878) during subsequent storage at 2 to 4 C for 6years (insert B).

7,14, 30, and 60 days (Fig. 3). The initial rise intoxin at 2 to 4 C in these samples was by a factorof 2.5, 2.0, 2.0, and 2.5, respectively.The characteristic behavior of toxin at 2 to 4 C

also appeared to be independent from the changesin toxin titer in the original experiment at 30 Cfrom which the sample was derived, i.e., it didnot seem to matter whether the original titer wasincreasing, constant, or declining. This point isillustrated by Fig. 4, 5, and 6. In Fig. 4, sample12,870 was transferred to 2 to 4 C after 14 daysat 30 C. At this time the sample contained only1,000 MILD) of toxin, and the experiment showedthat on continued incubation at 30 C the toxintiter would have increased fivefold, i.e., to 5,000

N OL. 13, 1965' 1017

on May 26, 2020 by guest

http://aem.asm

.org/D

ownloaded from

GRECZ, WAGENAAR, AND DACK

.E

E

1 2 3 4 5 6

YEARS AT 2-4C

FIG. 6. Changes in toxin titer in type I cheeseinoculated with Clostridium botulinum 62 A durinyincubation at 30 C (graph A), as related to stabilitgof the preformed toxin in the 60-day sample (no.12,754) during subsequent storage at 2 to 4 C for 6years (insert B).

Z'l'°rnP1287~~~~~~~~~~1878p

a 0.1 12750

1 %S~~~~~~~~~~~~~12 754

P Posturized by heating at 90c for lo miniutes P

S: Sterilized by heoting at 21c ftor 10 minutes '12749

0.01 * * * *

1 2 3 4 5 6

FIG. 7. Correlation between stability of preformedClostridium botulinum toxin in type I cheese andthe sterilization or pasteurization of the cheese priorto inoculation with C. botulinum 62 A.

MLD after 60 days. However, after transfer to a

refrigerator the toxin in sample 12,870 increasedonly twofold, i.e., from 1,000 to 2,000 MLD.

In Fig. 5, sample 12,878 was transferred from30 C to storage at 2 to 4 C after toxin had reachedits peak level. Additional incubation at 30 Cwould not have resulted in additional toxin pro-duction. Yet, on transfer to a refrigerator the ef-fective toxin titer doubled in potency.Most surprising, perhaps, was sample 12,754

(Fig. 6) in which toxin at 30 C was on the decline,but, after transfer to a refrigerator, the toxintiter doubled again. On the other hand, sample12,749 was derived from an experiment duringwhich this particular sample showed an unusually

TABLE 4. Stability of crude type A Clostridiumbotulinum toxin added to cheese*

MLD of toxin (avg of six samples) afterType of Initialcheese pH 0 3 7 14 30 60

days days days days days days

I 6.0-6.5 200 200 200 200 200 200II 5.7-6.4 150 200 150 117 100 83

III 7.1-7.3 200 200 150 133 133 67

* Toxin was produced in 5% Trypticase brothby C. botulinum 62 A, filtered, and added to cheese.The cheese samples were placed into a desiccatorfrom which air was evacuated by Dack's anaerobicapparatus (Jordan and Burrows, 1945), and 10%CO2 was added. Storage was in a dark incubatorat 30 C.

high toxin titer, and, again, on transfer to 2-4 Cthere was a doubling of toxin titer.The indigenous microbial flora or enzymes

present in cheese appeared not to play a role indegradation of toxin during storage at 2 to 4 C asmay be expected. Toxin showed an average stabil-ity in type I cheese in which microorganisms andenzymes were destroyed by autoclaving at 121 Cfor 10 min prior to experimental inoculation withC. botulinum spores (sample 12,754) comparedwith cheese which was pasteurized at 90 C for10 min (Fig. 7). In type III cheese, there appearedto be no difference between pasteurized (sample12,785) and sterilized (samples 12,789 and 12,790)samples.

Stability at 30 C of sterile toxin added to cheese.Culture filtrates of C. botulinum 62 A in 5%Trypticase broth, sterilized by filtration througha Seitz filter, were mixed into cheese at a level ofapproximately 200 MLD of toxin per gram. Of atotal of six samples for each type of cheese, threewere pasteurized by heating at 90 C for 10 min,prior to addition of toxin. The other three wereleft unheated to preserve the natural cheese flora.

Assays after 0, 3, 7, 14, 30, and 60 days at 30 Cshowed that the toxin titer was stable in type Icheese (Table 4). However, there was a definitedecline in types II and III; the decline com-menced after 7 to 14 days, and only one-half toone-third of the original toxin remained after 60days.There was no easily detectable relationship of

pH to toxin stability, although in the case oftype III cheese pH 7.1 to 7.3 may have been highenough to contribute to the instability of toxin,since it is known that basic pH leads to deteriora-tion of toxin activity (Spero, 1958).Comparison of pasteurized and nonpasteurized

cheese showed no consistent or significant differ-ences. It was expected that the natural flora of

1018 APPL. MICROBIOL.

on May 26, 2020 by guest

http://aem.asm

.org/D

ownloaded from

STABILITY OF C. BOTULINUM TOXIN AND SPORES

cheese with its proteolytic enzymes would attackC. botulinum toxin.

Effect of additives. The following substances hadno noticeable effect on the storage stability of

103

10 2 300 O'

plusorbganisrm

z

0

10 aln0 20 30 40 50 60

DAYS AT 30C

FIG. 8. Effect of natural flora of type III cheese

on stability of Clostridium botulinum toxin in sterile

6% Trypticase broth (averages of six tubes each).

Symbols: 0, broth plus sterile toxic broth;@0, broth

plus sterile toxic broth plus organisms from type

III cheese.

C. botulinum toxin in experimental cheese prep-arations: sodium chloride, 0.5, 1.0, and 1.5%;locust bean gum (stabilizer), 0.2%; sodium citrateor sodium phosphate emulsifiers, both at 2.5%.

Effect of cheese flora. To study the effect of themicrobial flora of cheese on C. botulinum toxin,12 tubes of 5% Trypticase broth were supple-mented with sterile toxic broth at a level of 1,000MLD per ml. Six tubes were inoculated with a

small block cut directly from the surface of typeIII cheese. These tubes developed heavy turbid-ity overnight. The other six tubes were not inoc-ulated. Both groups showed considerable declinein toxin titer during incubation at 30 C for 60days (Fig. 8). The toxin appeared not to be at-tacked by the actively growing microorganisms;on the contrary, the organisms appeared to offersome protection against toxin degradation.

Recovery of viable organisms from cheese storedat 2 to 4 C. Initial attempts at recovery of viableorganisms from type I cheese after 2 and 3 yearsof storage at 2 to 4 C failed because of surpris-ingly strong growth-inhibitory activity of agedsurface-ripened cheese to C. botulinum. All type Icheese samples received initially a low spore inoc-ulum (100 spores per gram).

In general, there appeared to be no increase in

TABLE 5. Type A Clostridium botulinum toxin in most probable number (MPN) tubes inoculated withdecimal dilutions of cheese samples stored at 2 to 4 C for 3 years

Type A toxin present (+) or absent (-) in MPN tubes inoculated with indicateddilutions of cheese samplesa No. of sporesTyeof Sample recovered___ _per__

cheese no. gram1:10 100spors wse0 1:100 1:1000 1:10,000 1:100,000 1:1,000,000

I 12,733 - - + + + + + 3 X10512,749 - - + -

d-

12,750 - - + + + - -_ 3X10312,754 - - + - - -

12,870 - - ± A

12,878 - - + + _ _ 2 X 102III 12,785 + + + + + + + 1.7 X 105

12,789 + + + + + + + - 2.0 X 105

a The amount of cheese added to recovery tubes at the 1:10 dilution level was approximately 5 mg/ml,i.e., 0.1 g of cheese suspended in approximately 20 ml of medium. The original experimental inoculumwas 100 spores of C. botulinum 62 A per gram of type I cheese and 100,000 spores per gram of type IIIcheese. The spores were heated at 85 C for 10 min prior to inoculation. Recovery medium was Wynne'sbroth (Wynne et al., 1955). Spores were estimated by use of the MPN method with the statistical tablesof Fisher and Yates (1953); samples were heated at 85 C for 10 min to destroy vegetative cells and anypreformed toxin, and to activate spores for germination. Toxin was estimated by intraperitoneal inocu-lations of 0.5 ml of the broth culture into white mice; specificity of toxin was checked with type Aantitoxin.

b These tubes each received 100 newly harvested, heat-shocked spores of C. botulinum 62 A.Sample was washed three times with distilled water in refrigerated centrifuge to remove inhibitory

substance(s) from the cheese.d At the 1:10 dilution inoculated with washed cheese, specific toxin and typical C. botulinum smell

was detected, but no definite growth could be recognized because of the strong turbidity from theadded cheese.

VOL. 13, 1965 1019

on May 26, 2020 by guest

http://aem.asm

.org/D

ownloaded from

1020 ~~~GRECZ, WAGENAAR, AND DACK AP.McoLL

TABLE, 6. Recovery of viable microorganisms from soft surface-ripened cheese preparations stored at 2 to 4 C

Initial expt* No. of organisms recovered after storage at 2 to 4 Ct

TYe Expt Sample - 4 yr 5 yr 6 yrno. no. Incuba-

tion at30C Spores Total Spores Total Spores TotalIdays

I 130 12,733 101 7 2.73 X 10' 3.31 X 105 2.7 X 105 3.1 X 105 3.0 X 101 1.5 X 101129 12,749 102 60 1.0 X 102 1.0 X 102 0.7 X 102 <102 0 0129 12,750 102 60 0.58 X 102 1.0 X 101 1.3 X 102 <102 0 0129 12,754 102 60 0.58 X 102 3.0 X 101 <102 <102 0 0131 12,870 102 14 0.58 X 102 3.0 X 101 <102 <102 0 2 X 101131 12,878 102 30 0.58 X 102 1.0 X 101 <102 <102 0 1 x 102

III 35 12,789 10' 7 2.1 X 10' 1.8 X 10' 2.0 X 10'5 1.7 X 10'5 4.5 X 10' 1.0 X 10'

*All samples were initially inoculated with spores of Clostridium botulinum 62 A heated at 85 C for10 min to activate spores, and to destroy vegetative cells and any preformed toxin. Therefater, thesamples were incubated at 30 C for the indicated number of days, and subsequently stored at 2 to 4 Cfor 6 years.

t The number of surviving organisms was determined by direct count and MPN methods by use ofcheese samples which were triple-washed in distilled water to remove from cheese the inhibitor activeagainst C. botulinum. For spore counts, the samples were heated at 85 C for 10 min; for total counts,no heat was applied.

the initial spore inoculum; there was, however, apronounced decrease and sometimes completedisappearance of the initial spore inoculum intype I cheese. An exception was found in sample12,733 in which the number of spores increa-sedfrom 102 to 3.0 X 10', per gram, i.e., a 3,000-foldincrease. In spite of the extensive spore multipli-cation, the amount of toxin synthesized was rela-tively low. Sample 12,733 developed 2,000 MLDof toxin as contrasted to 10,000 MLD in someother samples of toxic type I cheese. This obser-vation suggests that nutrient requirements forsporulation and for toxin synthesis seem to bediametrically opposite; i.e., media promotingsporulation seem to depress toxin synthesis andvice versa (Schneider, Grecz, and Anellis, 1963).

Inhibition of growth and toxin production ofC. botulinum in Wynne's broth (Wynne, Schmied-ing, and Daye, 1955) was noted in all tubes inoc-ulated with 1:10 (and occasionally with 1:100)dilutions of type I cheese. This effect was espe-cially evident with sample 12,733 (Table 5),which showed no growth of C. botulinum and notoxin in tubes inoculated with 1:10 dilutions ofcheese, but developed normal turbidity and toxinat the 10-2 through 10'1 levels where the growth-inhibitory activity of cheese was eliminated bydilution of the cheese suspension. It seems thatthe amount of inhibitory substance(s) necessaryto arrest growth must be very small, since thetotal amount of cheese at the 1 :10 dilution levelwas 0.1 g suspended in 20 ml of Wynne's broth,or approximately 0.005 g/ml. This amount wassufficient to inhibit the growth of the original

inoculum as well as that of 102 to 10' newlyadded spores of C. botulinum 62 A and 33 A. Nonoticeable inhibitory activity was found in typeIII cheese samples 12,785 to 12,789. The inhibi-tor(s) could be removed from the cheese inoculumby triple washing in distilled water in a refriger-ated centrifuge. This provided a method for therecovery of viable organisms from aged cheese(Table 6).Except for sample 12,733 which showed spore

multiplication as discussed above, the number ofspores in all other samples of type I cheese de-creased consistently but rather slowly. The or-ganisms seemed to survive predominantly in thespore state. The spores, apparently, were largelydormant since they yielded higher counts afterheat activation. The results suggested that sporesgerminated prior to final disappearance, as wasseen in samples 12,870 and 12,878; these samplescontained no recoverable spores after 6 years at2 to 4 C but still had a considerable number ofheat-sensitive (vegetative) cells.The large spore inoculum in type III cheese,

especially in sample 12,789 inoculated with 10'spores per gram, appeared to remain stable duringstorage at 2 to 4 C.Noteworthy was the observation that the dis-

appearance of spores was similar to decline intoxin titer, which was rapid in experiment 129,slower in experiments 130 and 131, and relativelystable in samples of type III cheese.

There was also the possibility that decline at2 to 4 C in both toxin titer and number of sporesmay have been related to the length of the initial

1020 APPL. MICROB10L.

on May 26, 2020 by guest

http://aem.asm

.org/D

ownloaded from

STABILITY OF C. BOTULINUM TOXIN AND SPORES

incubation period of the samples at 30 C. Thus,samples incubated for 7 days (experiments 130and 35) were most stable, whereas those incu-bated for 60 days (experiment 129) were leaststable. It is conceivable that, during 60 days at30 C, there was progressive deterioration whichmay have subsequently become additive with thedeterioration of toxin titer and spores at 2 to 4 Cin samples 12,749, 12,750, and 12,754.

DISCUSSION

In spite of its scientific and practical impor-tance, there is a remarkable paucity of knowledgeconcerning the longevity of C. botulinum sporesor the stability of C. botulinum toxin in food prod-ucts. C. botulinum spores survived for 5 and 6years in cheese, and, in two cases, the number ofspores did not decrease after 6 years of storage at2 to 4 C. This observation is not surprising, sinceit has been reported that spores survived incanned and hermetically sealed veal for as long as115 years (Curran, 1952). Our experiments sug-gested that the number of spores in cheese mayincrease, remain relatively constant, or decreaseto complete sterility (to a point where they couldnot be detected by availablemethods).An important consideration in the recovery of

C. botulinum spores from cheese is the develop-ment of inhibitory activity in cheese preparationspreventing outgrowth of spores at low dilutions.In the present study, washing of spores was usedsuccessfully to remove the inhibitor. However,other methods, such as dialysis or subculturinginto very large volumes of media, are of potentialpromise. Growth-inhibitory activity was prob-ably related to the antimicrobial agent(s) of agedcheese which was studied in some detail in ourlaboratory (Grecz et al., 1959a, b; Grecz, Dack,and Hendrick, 1961, 1962).

C. botulinum toxin was remarkably stable at2 to 4 C. Toxin stability at 30 C was not as great,although in the present experiments it was notexhaustively tested. In some lots of cheese thetoxin titer remained constant for as long as 2months. Hence, complete deterioration of toxinactivity in cheese would have taken a consider-ably longer time.

Especially interesting was the increase in toxintiter by a factor of two- to fivefold in sampleschanged from storage at 30 C to storage at 2 to4 C. It appears that this increase is Lest explainedin terms of activation of protoxin in nonprolifer-ating degenerating C. botulinum cells. This con-clusion is based on the reports of Bonventre andKempe (1960a, b). These authors presented in-direct evidence that early cell multiplication inC. botulinum types A and B is characterized by

the synthesis of a nontoxic intracellular precursorof the botulinal toxin molecule, termed protoxin.Later, during cell degeneration, the protoxin isconverted into active C. botulinum toxin. A simi-lar activation can be achieved with trypsin.Our results suggest that a certain rather con-

stant proportion of activatable toxin was presentin cheese cultures of C. botulinum. Activationtook place in all cheese samples, including thosewhich were heated for 10 min at 90 or 121 C priorto the start of the experiment. This severe heatingwould be expected to destroy all enzymes presentin the cheese prior to experimental inoculationwith spores. Therefore, the proteolytic enzymesfor activation of protoxin in cheese cultures musthave developed in the C. botulinum culture itself.

Bonventre (1957) reported that proteolytic ac-tivity, if continued too long, led to eventual de-struction of the toxin. This probably accountedfor the gradual deterioration of toxin in cheeseduring prolonged storage.The properties of individual batches of cheese

were the most important factors in determiningthe storage stability of spores and toxin of C.botulinum in cheese. This observation could beexplained in at least two ways. (i) The propertiesof individual batches of cheese affected the syn-thesis of toxin and enzymes by C. botulinum dur-ing active growth by providing nutrients, properpH, salt concentration, etc. (ii) The cheese af-fected the enzyme action responsible for activa-tion as well as deterioration of toxin after comple-tion of growth by its pH, ionic strength, colloidalabsorption of toxin (Greez and Dack, 1963) andenzymes, and other factors.

ACKNOWLEDGMENTSThis investigation was supported by a grant

from the National Cheese Institute and by agrant-in-aid from the Public Health Service(RG 5837). The investigation was initiated at theFood Research Institute, University of Chicago,Chicago, Ill., and was completed at the Quarter-master Food & Container Institute, and the Bio-physics Laboratory, Illinois Institute of Tech-nology, Chicago, Ill.We wish to thank A. A. Walker and Daniel and

Olga Berkowitz for technical assistance, and T. C.Tang for the drawing of all figures in this paper, inaddition to technical assistance.

LITERATURE CITEDBONVENTRE, P. F. 1957. The physiological basis

of toxigenicity of Clostridium botulinum types Aand B, p. 77-85. Ph.D. Thesis, The Univ. ofMichigan, Ann Arbor.

BONVENTRE, P. F., AND L. L. KEMPE. 1960a.Physiology of toxin production by Clostridiumbotulinum types A and B. I. Growth, autolysis,and toxin production. J. Bacteriol. 79:18-23.

VOL. 13, 1965 1021

on May 26, 2020 by guest

http://aem.asm

.org/D

ownloaded from

GRECZ, WAGENAAR, AND DACK

BONVENTRE, P. F., AND L. L. KEMPE. 1960b.Physiology of toxin production by Clostridiumbotulinum types A and B. IV. Activation of thetoxin. J. Bacteriol. 79:24-32.

BOOR, A. K., H. B. TRESSELT, AND E. J. SHANTZ.1955. Effects of salts and colloids on potency ofbotulinum toxin. Proc. Soc. Exptl. Biol. Med.89:270-272.

CURRAN, H. R. 1952. Symposium on the biology ofbacterial spores. V. Resistance in bacterialspores. Bacteriol. Rev. 16:111-117.

FISHER, R. A., AND F. YATES. 1953. Statisticaltables for biological, agricultural and medicalresearch, 4th ed. Hafner Publishing Co., NewYork.

GRECZ, N., AND G. M. DACK. 1963. Toxin produc-tion in naturally separated liquid and solid com-ponents in preparations of heated surfaceripened cheese inoculated with Clostridiumbotulinum. J. Food Sci. 28:276-282.

GRECZ, N., G. M. DACK, AND L. R. HEDRICK.1961. Antimicrobial agent of aged surfaceripened cheese. I. Isolation and assay. J. FoodSci. 26:72-78.

GRECZ, N., G. M. DACK, AND L. R. HEDRICK. 1962.Antimicrobial agent of aged surface ripenedcheese. II. Source and properties of the activeprinciple(s). J. Food Sci. 27:335-342.

GRECZ, N., R. 0. WAGENAAR, AND G. M. DACK.1959a. Inhibition of Clostridium botulinum andmolds in aged, surface ripened cheese. Appl.Microbiol. 7:33-38.

GRECZ, N., R. 0. WAGENAAR, AND G. M. DACK.1959b. Relation of fatty acids to the inhibitionof Clostridium botulinum in aged surface ripenedcheese. Appl. Microbiol. 7:228-234.

JORDAN, E. 0., AND W. BURROWS. 1945. Textbookof bacteriology, 14th ed., p. 26. W. B. SaundersCo., Philadelphia.

MEYER, K. F., AND B. EDDIE. 1951. Perspectivesconcerning botulism. Z. Hyg. Infektionskrankh.133:255-263.

NAYLOR, H. B., AND P. A. SMITH. 1946. Factorsaffecting the viability of Serratia marcescensduring dehydration and storage. J. Bacteriol.52:565-573.

SCHNEIDER, M. D., N. GREcZ, AND A. ANELLIS.1963. Sporulation of Clostridium botulinumtypes A, B, and E, Clostridium perfringens, andputrefactive anaerobe 3679 in dialysis sacs. J.Bacteriol. 85:126-133.

SPERO, L. 1958. The alkaline inactivation of botu-linum toxin. Arch. Biochem. Biophys. 73:484-491.

WAGENAAR, R. O., AND G. M. DACK. 1958. Factorsinfluencing growth and toxin production incheese inoculated with spores of Clostridiumbotulinum types A and B. I. Studies with sur-face-ripened cheese, type I. J. Dairy Sci. 41:1182-1190.

WEIL, L., T. S. SEIBLES, L. SPERO, AND E. J.SCHANTZ. 1957. Photooxidation of crystallineClostridium botulinum type A toxin in the pres-ence of methylene blue. Arch. Biochem. Bio-phys. 68:308-313.

WENTZEL, L. M., M. STERNE, AND A. POLSON.1950. High toxicity of pure botulinum type Dtoxin. Nature 166:739-740.

WYNNE, E. S., W. R. SCHMIEDING, AND G. T. DAYE,JR. 1955. A simplified medium for countingClostridium spores. Food Res. 20:9-12.

1022 APPL. MICROBIOL.

on May 26, 2020 by guest

http://aem.asm

.org/D

ownloaded from