C&ids and Surfaces A: Physicochemical and Engineering Aspects, 83 (1994) 187-191 Q921-7151/94/$Q7OO 0 1994 - Elsevier Science B.V. All rights reserved.
Ice nucleation by bacteria: production and activity” N. Cachet*,
C. Blanc, M.P. Luquet, D. Bouabdillah,
Ddpartement Gtkie Chimique, Universitd de Technologie de Compidgne, BP 649-60206 CompiPgne Cddex, France (Received
26 June 1992; accepted
Abstract The ability of Pseudomonas syringae to catalyse ice formation at relatively high temperatures from -2 received considerable attention. A fermentation method was used in order to obtain a high nucleation activity expressed in ice nuclei was based on the optimization of the medium composition and control of the fermentation parameters. medium appeared to provide the best conditions for biomass production and ice nucleation activity. observed a positive influence on the nucleation activity due to Mn*+ ion addition. The pH control of the culture seemed to influence negatively the bacterial nucleation activity. Key words: Fermentation;
The ability of certain bacteria to catalyse ice formation at relatively high temperatures (from -2 to - 5°C in 1 ml) has received considerable attention. For several strains such as Pseudomonas syringae, Pseudomonasjuorescens and Erwinia herbicola, ice nucleation activity (INA) has been described [ 1,2]; the strains are termed INA bacteria. Genetic studies carried out with P. syringae resulted in the cloning and expression of an ice nucleation gene in Escherichia coli . The gene product was a protein which appeared to be located in the outer membrane of the bacteria . Few works have dealt with the production of these nucleating agents [ 51. However, the possibility of employing INA bacteria as cloud seeding *Corresponding author. “The preliminary form of this paper was presented at the 7th International Conference on Colloid and Surface Science held in Compiigne, France, 7-13 July 1991, and was coordinated for publication by Professor M. Clausse.
per cell This Meat peptone Moreover, we
agents, in artificial snow or for the freezeconcentration of foodstuffs, has already been reported [ 61. The objective of our work was the optimization of the biomass production and the INA of the bacteria.
Material and methods Strain Pseudomonas syringae was obtained Institut Pasteur Collection (no. 74-20).
Media and culture conditions Glucose was added at a concentration of 40 g 1-l and the medium had the following composition: OSg 1-r; KH,PO,, 0.5 g 1-r; K,HPG,, Na,C,H,07.2H,0, 0.2 g I-‘; solution 1, 5 ml I-‘; solution 2, 5 ml 1-l; solution 3, 0.1 ml 1-r. The
N. Cachet et al.JColloids
I-‘; H,BO,, Solution
2.0 g l- ‘; ZnS0,6H20,
5.4 g 1-l; MnC1,.4H20, 0.17 g 1-r; CaC1,.6H,O,
0.99 g 0.238 g
expressed as FNU cell-’ (where FNU denotes a freezing nucleus unit). This method is more conve-
0.062 g 1-l; HCl, 0.155 M. 2: M,O,
10 g 1-i; HCI, 0.524 M.
Solution 3: Na,Mo0,.2H,O, 24.1 g 1-l. The nitrogen sources were added to the medium, maintaining a constant C/N value of ten. After inoculation, the 500 ml flasks containing 100 ml of medium were incubated at 30°C on an oscillatory shaker (150 rev min-‘). The bioreactor (Setric; model SET 2M; 2 1) was equipped with temperature, agitation, oxygen and pH measurement and control systems. Determination
of ice nucleation activity
The use of the calorimeter for testing the activity of the bacteria was not convenient as several samples had to be tested and each experiment was expected to last for 2 h. Furthermore, when this activity has to be tested during biomass production, time is very important and it is necessary to obtain a rapid result from the test. Therefore, for this study, we chose a faster method which allowed us to study several samples at the same time . samples
as 100 droplets
gation (5000 rev min- ‘; 20 min) and were resuspended in water in order to reach 7.5 x lo9 cells ml _ ‘. One hundred drops equivalent to 10 mm3 were deposited upon an aluminium sheet and were subjected to a cooling programme. The temperature was slowly reduced from the ambient (19°C) to -20°C at a rate of 1 “C min-‘, and was maintained for 5 min at each 1 ‘C step. The number of frozen drops was evaluated at each temperature step. Freezing temperature spectra were established for each set of culture conditions and the percen-
at the same
time. Also, as the sample volume to be cooled was important, it was difficult to perform perfect continuous occur
each step. However,
as the aim of
these tests was only to compare the activity of bacteria obtained in different ways, Vali’s method was used. We should have preferred to use the emulsion technique, which is surely the best method, but there would still be a problem to solve as ice nucleating agents were found to be less active when they are introduced in microsize droplets instead
In previous work [7,8] we have tested the INA of bacteria in microsize droplets dispersed within an emulsion or in bulk samples (l-2 mm3).
Eng. Aspects 83 (1994 i 187-191
tage of frozen drops was expressed as a function of temperature. The nucleation activity values were then calculated using Vali’s method  and were
compositions. Solution 1: CaCO,, 1-i; FeC13.6H,0, 1-r; CuCl,.2H,O,
Surfaces A: Physicochem.
of in bulk samples
Biomass evaluation Biomass evaluation was performed by plate counts on nutrient agar. Biomass dry weight was determined by gravimetry after centrifugation (10000 rev min -‘. , 20 min) and drying (24 h at 105 _+2°C). Results The preliminary assays were carried out in flasks in order to study the influence of the nature of the nitrogen source on biomass production and nucleation activity. Among the nitrogen sources tested, the four different peptones supplied all the necessary nutrients for growth, as the biomass production values were similar. However, the nucleation activity was significantly enhanced by meat peptone, but the components involved in this enhancement were not identified (Table 1). The meat peptone was then selected as the complex nitrogen source for further assays. The medium composition previously defined during the flask experiments was used to carry out the batch cultures in the bioreactor.
N. Cachet et al./Colloids
Surfaces A: Physicochem
Table 1 Influence
of the nitrogen
on the growth
Eng. Aspects 83 (1994)
of P. syringae
Meat (liver) peptone
Final biomass (g 1-r) Nucleation activity (%) Nucleation activity (FNU
1.83 1 1.32 x lO_”
4.13 0 0
0.76 0 0
5.69 90 3.03 X 10-s
5.64 0 0
5.59 12 1.67 x 10-a
6.24 0 0
The investigation was performed in 100 ml flasks after a 48 h culture “Percentage of frozen drops at -3°C.
at 150 rev mini’,
In Table 2 are reported the values of biomass production (g l-i), growth rate (h-l) and maximal nucleation activity (expressed as FNU cell-‘), obtained for cultures performed under different conditions. From these results, the meat peptone medium (culture 2) appeared to provide the best choice for biomass production and INA as deduced from the flask experiments. Furthermore, during experiments performed in the bioreactor, we tested the influence of pH control on two parameters: growth and nucleation activity. Without pH regulation, we observed an improvement of growth and INA for cultures 3 and 4 with monosodium glutamate (MSG) and meat peptone respectively. When the pH was maintained at 7.0 the nucleation activity values were lower (culture 1 with MSG and culture 2 with meat peptone). The culture referred to as culture 2 in Table 2 was compared with those conducted with a medium
(culture 3). The effect of pH control was also tested with meat peptone medium (culture 4). Finally, on MSG medium only, we tested the influence of Mn2+ ion addition. The results obtained with culture 3 (MSG, without pH regulation) and culture 5 (with 1 uM Mn2+ ions) were compared. Since the growth was less important with Mn2+ ion addition, we confirmed the finding of other workers [lo] concerning the positive action of Mn2+ ion addition on INA (culture 5). The growth of P. syringae for different nitrogen sources was evaluated during the batch cultures and is shown in Figs. 1 and 2 with pH regulation, and in Figs. 3 and 4 without pH regulation. As we previously reported, the highest nucleation activities were obtained at the beginning of the stationary growth phase [ 111. The activity did not appear until the cell concentration reached 2 x 10’ cell ml-’ (time t=24 h). The nucleation activity obtained at the beginning of the stationary growth phase expressed as FNU cell- ’ at each temperature step, is reported in Fig. 5 for the five cultures previously described,
including MSG as nitrogen source, as described by Lawless and Laduca [S]. Also, two sets of experiments were performed: one without pH regulation (culture 1) and the other with pH regulation at 7.0
Table 2 Influence
Growth rate (h-r) Biomass production (g 1-r) Nucleation activity” Nucleation activity (FNU cell-‘)
0.69 5.7 6 8.15 x 10-r’
of frozen drops
at - 3°C.
as a summary
and nucleation Culture
at 30°C with a C/N ratio of 10.
0.55 15.3 26 3.96 x 1O-9
of our results.
with P. syringae Culture
0.69 13.2 12 1.68 x lo9
0.46 15.0 90 3.03 x lo-*
0.27 1.9 30 4.70 X 1om9
N. Cachet et al.lColloids Surfaces A: Physicochem. Eng. Aspects 83 ( 1994 ) 187-191
lb Time (II)
Fig. 4. Culture 4. Evolution of growth ( W) and pH (0) during a batch culture of P. syringae (glucose, 90 g I-‘: meat peptone, 28 g 1-l; without pH regulation).
Fig. 1. Culture 1. Evolution of growth during a batch culture of P. syringae (glucose, 90 g 1-l; MSG, 45 g I-‘; pH maintained at 7.0).
Fig. 5. Evolution of nucleation activity during batch cultures of I? syringae carried out under different conditions: W, culture 1 (glucose, 90 g 1-l; MSG, 45 g I-‘; pH 7); A, culture 2 (glucose, 90 g 1-l; meat peptone, 28 g 1-l; pH 7); 0, culture 3 (glucose, 90 g 1-l; MSG, 45 g I-‘; without pH regulation); A, culture 4 (glucose, 90g 1~‘; meat peptone, 28g I-‘; without pH regulation).
1 zl z
Fig. 2. Culture 2. Evolution of growth during a batch culture of p. syringae (glucose, 90 g 1-l; meat peptone, 28 g 1-l; pH maintained at 7.0).
I 6 0
Fig. 3. Culture 3. Evolution of growth (m) and pH (0) during a batch culture of P. syringae (glucose, 90 g 1-l; MSG, 45 g 1-l; without pH regulation).
From flask experiments, meat peptone was chosen as the nitrogen source for Pseudomonas syringae cultivation. Using a bioreactor, we confirmed that this complex nitrogen source was better than MSG for biomass production and nucleation activity. Moreover, we recommended that the cells be cultivated without pH regulation. At the end of the culture, the maximal biomass production reached around 15 g 1-l and the harvested cells
N. Cachet et al./Colloids Surfaces A: Physicochem. Eng. Aspects 83 (1994)
presented a high nucleation activity with 3.03 x 10e8 FNU cell-‘. However, the addition of Mnzf ions to the culture medium led to an improvement in the INA, since the growth was not enhanced. We may assume that the specific activity obtained under these conditions (culture 5) was very attractive. We are currently testing other medium compounds in order to obtain a culture giving a maximal specific INA.
S.E. Lindow, D.C. Arny and CD. Upper, 68 (1978) 831.
3 4 5 6 7
8 9 10
J. Luisetti and J.L. Gaignard, CR. Seances Acad. Agric. Fr., 75(6) (1989) 93. C. Orser, B.J. Staskawicz, N.J. Panopoulos, D. Dahlbeck and SE. Lindow, J. Bacterial., 164(l) (1985) 359. L.H. Sprang and SE. Lindow, Phytopathology, 71 (1981) 256. R.J. Lawless and R.J. Laduca, Eur. Patent 0 281 145, 1988. M. Watanabe, 3. Watanabe, K. Kumeno, N. Nakahama and S. Arai, Agric. Biol. Chem., 53 (1989) 2731. D. Clausse, N. Cachet, M. Bouzoubaa and M. Aguerd, in P.E. Wagner and G. Vali (Eds.), Lecture Notes in Physics, Vol. 309, Springer Verlag, 1988, pp. 721-724. D. Clausse, D. Bouabdillah, N. Cachet, M.P. Luquet and S. Pulvin, Pure Appl. Chem., 63( 10) (1991) 1491. G. Vali, J. Atmos. Sci., 28 (1971) 402. L.M. Kozloff, M.A. Turner, F. Arellano and M. Lute, J. Bacterial., 173(6) (1991) 2053. M.P. Luquet, N. Cachet, D. Bouabdillah, S. Pulvin and D. Clausse, Cryoletters, 12 (1991) 191.