Journal of Applied Bacteriology 1989,66,331-337 2892/08/88

The effect of cooling rate, freeze-drying suspending fluid and culture age on the preservation of Campylobacter pylori

R . J . OWEN*, S.L.W. ON & M . COSTAS National Collection of Type Cultures, Central Public Health Laboratory, London N W 9 5HT, UK

Received 19 August 1988, revised 21 October 1988 and accepted 26 October 1988

OWEN, R.J., ON, S.L.W. & COSTAS, M. 1989. The effect of cooling rate, freeze- drying suspending fluid and culture age on the preservation of Campylobactm pylori. Journal of Applied Bacteriology 66,33 1-337.

The effects of freezing rate, suspending fluid and age of culture on the ability of four strains of Campylobacter pylori to survive and recover from freeze-drying were examined. Freeze-drying by standard procedures generally resulted in an overall loss in viability of between 3 and 7 log units. The exact cause of poor recovery by C. pylori was not established but strain differences were detected, with NCTC 11637 (type strain) surviving better than NCTC 11638 and NCTC 11639. Recovery of the poorest growing strain (NE 26695) was notably more erratic. The largest loss in viability occurred at the primary drying stage. Losses resulting from freezing and secondary drying were less marked and the rate of freezing had only a marginal effect on recovery. Nineteen different freeze-drying suspending fluids were investi- gated. Overall the best recovery results were obtained with 5% inositol-broth (or horse serum) plus 25% glucose, at pH 7.0, in which loss of viability was typically about 4 log units. Other factors, such as age of culture and number of viable bac- teria in the before-dry suspension, did not have a significant effect on survival. We conclude from these results that C. pylori can survive freeze-drying, albeit in small numbers, but the degree of recovery is apparently largely strain dependent.

Freeze-drying or lyophilization of micro- organisms is a three-stage process in which the cell suspension is frozen, 90-95% of the water content is removed by sublimation (primary drying stage) and the remaining water, except for about 1%, removed slowly under vacuum (secondary drying stage). The process is used by culture collections to preserve a wide variety of bacteria, yeasts, fungi and some viruses (Kirsop & Snell 1984). Freeze-drying has become a popular method of preservation because it is convenient for batch production, long-term via- bility is excellent in most cases, and storage and distribution requirements are simple.

Although the majority of bacteria of medical interest survive freeze-drying without significant loss of viability even after 30 years (Rudge 1984) there are notable exceptions, one of which is

* Corresponding author.

Campylobacter pylori. This organism was first isolated from human gastric mucosa (Marshall 1983) and has emerged as a new gastrointestinal pathogen associated world-wide with active chronic gastritis (Dooley & Cohen 1988). The intense medical interest in this novel organism has stimulated a considerable demand for refer- ence cultures over the past five years. Efforts to supply C. pylori have been frustrated because strains are highly susceptible to freeze-drying (unpublished observations), although stocks can be maintained frozen at -50°C (Morgan et a!. 1988) or in liquid nitrogen (McNulty & Dent 1987). However, neither of these methods are suitable for culture collection dispatch purposes.

The aim of the present study was to deter- mine at which stage cell damage occurs during the freeze-drying process and to investigate factors that might improve the survival of C. pylori. This paper describes the viability results

332 R. J . Owen et al. obtained with different freezing-rates and pres- ervation fluids, and pre-dry cultures of different ages.

Materials and Methods

BACTERIA

The following strains of C. pylori were used: NCTC 11637 (type strain), NCTC 11638, NCTC 11639 and NE 26695. They were obtained from the National Collection of Type Cultures (NCTC), and the collection of Dr D.R. Morgan, Norwich Eaton Pharmaceuticals Inc. (NE), Norwich, New York, USA.

MEDIUM A N D C U L T U R E C O N D I T I O N S

All strains were grown on nutrient agar plates containing 5% v/v defibrinated horse blood. Cultures were incubated for 2-6 d at 37°C under microaerobic conditions (about 5% oxygen) attained by incubating an anaerobic jar without catalyst from which 75% of the air had been withdrawn (580 mm Hg below atmo- spheric pressure) and replaced with an anaero- bic gas mixture containing 10% hydrogen, 10% carbon dioxide and 80% nitrogen. The anaero- bic jars were regassed daily using the same mixture. Strains were maintained by regular subculture on blood agar for periods up to 4-6 weeks. Stocks of all strains were also preserved in 10% v/v glycerol in Nutrient Broth No. 2 (Oxoid: CM67) over liquid nitrogen ( - 176°C).

FREEZE-DRYING

Dense suspensions from 3-d-old cultures on blood agar were made in 0.5 ml of inositol- serum consisting of 5% (w/v) m-inositol in horse serum (Wellcome No. 3), and filter sterilized (Redway & Lapage 1973). This was the stan- dard suspending medium used unless stated otherwise. It was dispensed in 0.2-ml volumes (in triplicate) in sterile glass freeze-drying ampoules which were placed in the centrifuge carrier plate of an EF4 Modulyo freeze-dryer (Edwards High Vacuum, Crawley, West Sussex). The bacterial suspensions were spin-frozen at a condenser temperature of -59°C for 15 min and dried under vacuum (pressure < 10- mbar) for 3 h. Secondary drying was on an Edwards

Manifold dryer under vacuum (< lo-' mbar) over phosphorus pentoxide for approximately 18 h. Full details of the freeze-drying methods have been described by Rudge (1984). Freeze- dried samples from the primary and secondary stages were reconstituted with 0.2 ml of sterile distilled water. Spin-frozen suspensions were equilibrated to room temperature. Viable counts, expressed as colony forming units (cfu/ ml), were obtained for each suspension by a modified Miles & Misra (1938) technique with Nutrient Broth No. 2 (Oxoid: CM67) as the diluent. The counts were recorded on pre-dried 5% horse blood agar plates at 37°C under microaerobic conditions for 3 d (NCTC 11637, NCTC 11638 and NCTC 11639) and 6 d (NE 26695).

FREEZING RATE EXPERIMENTS

Ampoules containing 0.2 ml of inositol-serum in freeze-drying ampoules were placed at four dif- ferent temperatures (4, - 30, - 40 and - 70°C). The decrease in temperature in each was record- ed at 30-s or I-min (4°C) intervals on a digital thermometer (Comark Electronics, Sussex) inserted directly into the serum. Suspensions (0.2 ml) of C. pylori in 5% inositol-serum were distributed in ampoules and the initial tem- perature of each was recorded. The suspensions were then cooled at the respective temperature (see above) until completely frozen. The frozen suspensions were equilibrated to room tem- perature and viabilities were determined as above.

S U S P E N D I N G F L U I D S

The 18 different suspending fluids listed in Table 1, in addition to the standard medium (No. 1) were prepared and sterilized by filtration. Sus- pensions of C. pylori were prepared in each of the fluids, 0.2-ml amounts were freeze-dried, and viable counts were determined as described above.

C U L T U R E AGE

Campylobacter pylori cultures 2, 3, 4, 5 and 6 d old were freeze-dried in 5% inositol-serum plus 25% (w/v) glucose at pH 7.0. Before-dry (BD)

Preservation of Campylobacter pylori 333

Table 1. Preservation media used and viability counts of different Campylobacter pylori strains

Preservation medium used Viability counts (cfuiml)

Strain Before-dry After-dry Number Media base and additives* number (BD) (AD) Log drop

5% m-inositol-horse serum base 1 - (standard medium)

2 10% glycerol

3 5% dimethylsulphoxide (DMSO)

4 10% dimethylsulphoxide

5 25% glucose

6 5% polyvinylpyrrolidone + 3% L-glutamic acid

5% m-inositol-broth base 7

8 25% glucose

9 25% glucose pH 6.0

10 25% glucose pH 7.0

11 25% glucose pH 8.0

12 30% glucose

13 40% glucose

14 50% glucose

15 5% polyvinylpyrrolidone + 3% L-glutamic acid

5% m-inositol-fetal calf serum base 16 17 5% L-glutamic acid

Fetal calf serum base 18 19 5 % L-glutamic acid

NCTC 11637 NCTC 11639 NCTC 11637 NCTC 11639 NCTC 11637 NCTC 11639 NCTC 11637 NCTC 11639 NCTC 11637 NCTC 11639 NCTC 11637 NCTC 11639

NCTC 11637 NCTC 11638 NCTC 11637 NCTC 11637 NCTC 11637 NCTC 11639 NCTC 11639 NCTC 11639 NCTC 11637 NCTC 11639 NE 26695 NCTC 11637 NCTC 11638 NCTC 11639 NE 26695 NCTC 11637 NCTC 11638 NCTC 11639 NE 26695 NCTC 11637 NCTC 11639 NCTC 11637 NCTC 11639 NCTC 11637 NCTC 11639 NCTC 11637 NCTC 11639

NCTC 11637 NCTC 11637

NCTC 11637 NCTC 11637

7.5 x 107 6.0 x 107 4.2 x 107 1.3 x 107 4.8 x 107 1.5 x 107 6.2 107 0.3 x 107 1.5 x 10' 0.7 x lo6 4.2 x 10' 0.7 x lo6

0.2 x 10' 1.7 x 10' 0.3 x 10'

5.6 x 10' 0.5 x lo6 2.5 x 10'

3.8 x 10'

7.0 x 10' 4.0 x 10' 2.8 x 10' 8.5 x 10' 1.4 x 10' 9.2 x 10' 3.5 x 10' 3.5 x 10' 3.6 x lo6 0.8 x 10' 1.3 x 10' 0.4 x lo6

5.5 x 107

1.1 x 109

1.0 x 109

3.3 x 107 0.1 x 107 3.9 x 107 3.5 x 107 1.9 x 107

1.4 x 109 1.2 x 109

1.7 x 109 1.2 x 109 3.6 x lo6

0 0

0 0 0 0 0

0

0

7.5 x 103

1.1 x 105

3.5 x 104

2.2 x 103

1.8 x 105 6.0 x 105

2.4 x 107 1.2 104 2.0 x 103

5.1 x 103 1.3 x 105

1.3 x 104 1.8 x 104

0 2.9 x lo2

0 0

0 1.2 x 105

0 5.0 x lo2

0 6.2 x 10' 1.0 x 10' 1.8 x 10' 0.5 x 10' 3.6 x 10'

0 2.9 x 104 2.2 x 103

1.6 x 103 3.6 x 103

1.7 x 103 1.0 x lo2

0

7.0 7.0 3.8 7.0 7.0 7.0 7.0 6.0 2.1 5.0 3.1 6.0

3.0 8.0 4.0 2.5 3.0 6.0 7.0 1.7 4.5 5.7 7.0 3.5 4.7 3.8 7.0 5.3 3.4 4.3 6.0 5.1 6.1 4.3 6.8 4.5 7.0 3.1 4.0

5.9 5.5

7.2 5.8 6.0 NE 26695

* Unless stated otherwise, all media were at pH 6.3.

334 R. J . Owen et al. Table 2. Viable counts (cfuiml) of Campylobacter pylori at different stages of freeze-

drying in 5% inositol-serum

Before-dry Secondary-dry Strain (BD) Spin-freeze Primary-dry (AD)*

NCTC 11637 7.75 x lo7 4.3 x 10' 3.5 x 103 5 x lo2

NCTC 11639 1.9 x lo8 4.1 x lo7 4.4 x 1 0 3 3.25 x 103 (0.25)t (4.34) (5.19)

(0.66) (4.63) (4.76) NE 26695 3.0 108 2.1 10' 2.75 x 103 2.5 x lo2

(1.15) (5.03) (6.1) * AD, resultant after dry count. t ( ) cumulative logarithmic losses

and after-dry (AD) viable counts were deter- mined as above.

Results

EFFECT OF F R E E Z E - D R Y I N G

The viability counts on three C. pylori strains (NCTC 11637, NCTC 11639 and NE 26695), which were suspended in 5% inositol-serum (medium No. l), were recorded before freeze- drying (BD count), after spin-freezing, and after primary and secondary drying. Table 2 shows that the major loss in viability (average 3.98 0.50 log units) occurred at the primary drying stage. Smaller losses of less than 1 log unit were observed after spin-freezing and of about 1 log unit after secondary drying. Although the three strains all exhibited overall low recovery rates, there were slight strain-to- strain differences: NE 26695 was more suscep-

tible to freeze-drying than NCTC 11637 and NCTC 11639 despite the fact that it had the highest BD count.

EFFECT OF F R E E Z I N G R A T E

Table 3 shows the viabilities of C . pylori before and after freezing in 5% inositol-serum at six different temperatures with freezing rates between 1 and ZWC/min. Only small losses in viability (on average 1 log unit) were detected and these were independent of the cooling rate. The largest viability losses shown by NCTC 11639 and NCTC 11637 were 1.59 and 0.54 log units respectively, and these occurred during spin-freezing. For strain NE 26695, the largest loss (1.70 log units) occurred at -70°C. Although the before-freezing (BF) counts dif- fered between strains, the relative overall losses in viability due to freezing were very similar for each strain.

Table 3. Rates of cooling and viable counts* of Campylobacter pylori after freezing at different temwratures in 5% inositol-serum

Rate of NCTC NCTC NE Temperature coolingt 11637 11639 26695

("C) ("C/min) ( x 108) ( x 108) ( x 106)

Room temperature (before freezing) 0 6.0 15.5 1 .o

4 1.0 4.0 12.5 0 .5

-40 - 30 - 70 6.8 5.0 10.0 0.02

4.0 4.5 5.0 3.51 '!'y'9.8 :: ] 1.4

Spin-freezing 10.0 1.7 Liquid nitrogen 200.0 2.55 5.5 0.2

( - 176°C)

* cfu/ml. t Values were determined experimentally except for those for spin-freezing and

1 Mean value of viable count ( x lo8). liquid nitrogen, which were estimated (R. Rudge, personal communication).

Preservation of Campylobacter pylori EFFECT OF SUSPENDING FLUID

Table 1 shows the before-dry (BD) and after-dry (AD) counts and the corresponding log drop in viability of C. pylori in 19 different suspending fluids. NCTC 11637 was recoverable from 16/19 of the suspending media although freeze-drying generally resulted in a drop in counts of 2-7 log units. The best recovery rates (2-3 log drops) were obtained with: 5% inositol-serum plus 25% glucose (medium 5), 5% inositol-serum plus 5% polyvinylpyrrolidone (PVP) and 3% L- glutamic acid (medium 6), 5% inositol-broth (medium 7), 5% inositol-broth plus 25% glucose (medium 8), 5% inositol-broth plus 25% glucose at pH 7.0 (medium lo), and 5% inositol-broth plus 5% PVP and 3% L-glutamic acid (medium 15).

Strain NCTC 11639 was more susceptible to freeze-drying than NCTC 11637 and was not recovered from 8/14 media tested. NCTC 11639 was recovered with viability decreases of 4-7 log units from inositol-serum based media but was not tested in the fetal calf serum based media. The other two strains (NCTC 11638 and NE 26695) were tested only in 5% inositol-horse serum plus 25% glucose at different pH values (medium numbers 9-11). Strain NE 26695 was not recovered from any of these media but NCTC 11638 was recovered from the pH 7.0 and 8.0 media (numbers 10 and 11). The inclu- sion of cryoprotectants, such as glycerol and dimethylsulphoxide (DMSO), did not signifi- cantly improve survival rates, although some improvement was observed when PVP and glu- tamic acid were both present in the suspending medium. Fetal calf serum instead of horse serum did not improve recovery rates and viability drops of 5-7 log units were recorded irrespective of the cryoprotectant added.

EFFECT OF CULTURE AGE

NCTC 11639 had the fastest generation time of the four strains examined with colonies clearly visible after just 2 d. Strains NCTC 11637 and NCTC 11638 grew well in 3 d, but NE 26695 was the slowest growing strain usually taking 5-6 d for distinct colonies to appear. To investi- gate the effect of culture age on the ability of C. pylori to survive freeze-drying, suspensions were prepared from cultures grown for different periods up to 6 d. The resultant losses in via-

335 Table 4. Effect of initial culture age on viability loss (log units) after freeze-drying in 5% inositol-serum

containing 25% glucose, pH 7.0

Age of culture used for suspension (d)*

Strain 2 3 4 5 6

NCTC 11637 3.4 4.5 t 3.51

NCTC 11638 3.7 NCTC 11638 3.5 NE 26695 3.7

6.2

3.4 5.5 3.8 4.2 3.6 6.2 NT 3.21 4.01 NT

2.9 3.9 3.8 5.0 4.7 3.5 3.6 NT 5.9 3.5 NR 2.3

* Before-dry counts were typically 6 8 log units. t Results obtained in duplicate from the same bac-

NR, not recovered. NT, not tested. terial suspension.

bility of the four strains after freeze-drying in 5% inositol-serum plus 25% glucose, pH 7.0 are shown in Table 4. The average drop in viability for each of these strains was about 4 log units. In most cases this viability drop was indepen- dent of the age of the initial culture although the results on NCTC 11637 and NCTC 11638 indicated that the 6-d-old cultures survived freeze-drying less well. The results on NE 26695, the slowest growing strain, tended to be less reproducible even when viability counts were determined in duplicate on the same suspension.

Discussion

The factors affecting the survival of micro- organisms after freezing and drying stresses are complex and not well understood (Mackey 1984) but the composition of the preservation medium is one of the most important. Loss of viability may first occur during freezing as a result of cold shock or damage to the cell mem- brane and DNA. Our results showed that C. pylori was damaged to some extent by freezing stress in the presence of sugars and serum, which act as cryoprotectants, although there were no obvious differences in recovery when the cooling rates were either slow or rapid. It was not possible to establish if the damage to C. pylori at this stage was due to solution effects or to internal ice formation. The longevity and sta- bility of C. pylori strains using liquid nitrogen preservation has not been reported but we have found that cultures can be recovered after 1 year

R. J . Owen et al. at -190°C with 10% glycerol as the cryo- protectant (results not presented). Westblom el al. (1987) reported that C. pylori was extremely sensitive to freezing and did not survive when frozen in regular freezing media. They recom- mended freezing at - 70°C in defibrinated horse blood or skimmed milk, however, both of which yielded viable bacteria after 6 months, and attributed loss of viability to length of storage rather than to freezing and thawing stresses. A similar procedure (-50°C in whole rabbit blood) was used by Morgan et al. (1987) to store C. pylori strains.

After the initial freezing step, freeze-dried micro-organisms are subjected to drying and rehydration stresses, which in C. pylori caused significant loss of viability. Our results demon- strated that the principal damage, which was probably to membranes and nucleic acids, occurred at the primary drying stage with further but smaller losses after secondary drying. In this series of experiments the freeze- dried material was not subjected to long-term storage so damaging effects due to amino- carbonyl reactions, oxidation or free-radical reactions were not thought to be contributing factors. Overall, the highest and most consistent level of after-dry recovery was achieved for NCTC 11637 with either 5% inositol-serum plus 25% glucose or 5% inositol-broth plus 25% glucose. The inclusion of proven cryo- protectants, such as glycerol or DMSO in the suspension fluid, did not provide further improved protection during freeze-drying, but the use of 5 % PVP and 3% L-glutamic acid pro- duced results that were comparable to 25% glucose. It is not known if PVP or glutamic acid were equally effective alone in inositol-serum (or broth) based media but the results using either alone in fetal calf serum (medium 17 and 19) suggested that individually they did not provide additional protection.

There was considerable variation between the four strains of C. pylori examined in their sensi- tivity to freeze-drying. NCTC 11639 resembled NCTC 11637 in showing best recoveries ( < 4 log unit drop) after preservation in 5 % inositol broth plus 25% glucose pH 7.0, and NCTC 1 1 638 also gave comparable recovery rates from this fluid. However, NE 26695 was the most sensitive strain to freeze-drying and was recov- ered only in low numbers from 5% inositol serum plus 5% glucose pH 7.0 (see Table 4).

The reason for these strain differences is not known as they were from similar clinical material and had identical conventional test characteristics. The results of other factors, such as age of cultures, pH of suspending fluid, and number of viable bacteria in initial suspension before freeze-drying, showed that they did not have a significant effect on the resultant via- bilities of C. pylori preparations.

We conclude that the best suspending medium overall was 5% inositol broth and 25% glucose pH 7.0. The main cell damage was pre- sumed to have occurred during primary drying but it appeared that the combination of pep- tones in the nutrient broth of the suspending fluid and the high total sugar concentration of 30% provided some protection to C. pylori. The high sugar concentration may also have provid- ed protection during the rehydration stage when further stress to cytoplasmic membranes occurs. Annear & Goodwin (personal communication) also found that high concentrations of glucose improved C. pylori survival when dried from the liquid state in a suspending medium comprising nutrient broth plus 20% glucose. Precise pH of the suspending fluid is possibly another impor- tant factor because membranes, already damaged by freezing and drying, become more sensitive to acid and alkali (Mackey 1984).

It is of interest to note that there is growing evidence from ultra-structure studies and ribo- somal RNA sequences that C. pylori is not a true campylobacter (Paster & Dewhirst 1988). Its sensitivity to freeze-drying suggests some fundamental difference at the cellular level from other campylobacters and wolinellas, which survive the process, and provides further evi- dence to support the view that the species should be reclassified.

We wish to thank Dr L.R. Hill for helpful dis- cussions. S.L.W.O. was supported by Norwich Eaton Pharmaceuticals Inc., Norwich, USA.

References

DOOLEY, C.P. & COHEN, H. 1988 The clinical signifi- cance of Campylobacter pylori. Annals of Internal Medicine 108, 7G79.

KIRSOP, B.E. & SNELL, J.J.S. 1984 Maintenance oj Micro-organisms. A Manual of Laboratory Methods. London: Academic Press.

MACKEY, B.M. 1984 Lethal and sublethal effects of refrigeration, freezing and freeze-drying on micro-

Preservation of Campylobacter pylori 337 organisms. In The Reoiual oflnjured Microbes ed. PASTER, B.J. & DEWHIRST, F.E. 1988. Phylogeny of Andrew, M.H.E. & Russell, A.D. pp. 45-75. Campylobacters, Wolinellas, Bacteroides gracilis, London: Academic Press. and Bacteriodes ureolyticus by 16 S ribosomal ribo-

MARSHALL, B. 1983 Unidentified curved bacillus on nucleic acid sequencing. International Journal of gastric epithelium in active chronic gastritis. Lancet Systematic Bacteriology 38, 5662. i, 1273-1275. REDWAY, K.F. & LAPAGE, S.P. 1974 Effect of carbo-

MCNULTY, C.A.M. & DENT, J.C. 1987 Rapid identifi- hydrates and related compounds on the long term cation of Campylobacter pylori (C. pyloridis) by pre- preservation of freeze-dried bacteria. Cryobiology

11,73-79. 25, 1683-1686. RUDGE, R.H. 1984 Maintenance of bacteria by freeze-

MILES, A.A. & MISRA, S.S. 1938 The estimation of the drying. In Maintenance of Micro-organisms ed. bacteriocidal power of the blood. Journal of Kirsop, B.E. & Snell, J.J.S. pp. 23-33. London: Aca- Hygiene, Cambridge 38,732-749. demic Press.

WESTBLOM, T.U., BARTHEL, J.S., HAW, A.D., GONZA- KRAFT, W.G. 1987 Susceptibility patterns of LEZ, F.J., TARKA, E.F. & EVERETT, E.D. 1987 Long Campylobacter pyloridis. FEMS Microbiology term freeze storage of Campylobacter pyloridis. Letters 42, 245248.

formed enzymes. Journal of Clinical Microbiology

MORGAN, D.R., FITZPATRICK, P.M., DAVID, K.L. &

Journal of Clinical Pathology 40,353-356.