Biotechnol. Bog. 1093, 9, 291-297
28 1
Effect of Chelatants on Gellan Gel Rheological Properties and Setting Temperature for Immobilization of Living Bifidobacteria I. Camelin,t C. Lacroix,*$$ C. Paquin,*H. Prbvost,t R. Cachon,t and C. Diviest Centre de recherche Stela, Faculte des Sciences de YAgriculture et de YAlimentation, Universite Laval, Sainte-Foy, Qubbec, Canada G1K 7P4,and Laboratoire de Microbiologie, ENS BANA, Universite de Bourgogne, 21000 Dijon, France
The effect of various concentrations of sequestrants (sodium citrate, sodium metaphosphate, and EDTA) was studied on gellan gel (1.5-2.5% (w/v)) setting temperature and rheological properties. Addition of EDTA between 0 and 0.8% (w/v) led to a progressive decrease of setting temperature. Citrate and metaphosphate decreased this parameter when added up to 0.4 or 0.6%,depending on gellan gum concentration, eventually resulting in the absence of gel forma$ion a t room temperature for the 1.5% gellan solution containing 0.4% citrate. This effect was accompanied by a significant decrease of gel strength and stiffness and might be attributed to the binding of the divalent cations required for chain association during gelation by chelatants. With the aim of lowering the gel setting temperature during the cell entrapment process while maintaining high mechanical properties, a gel made of 2.5% gellan gum and 0.2% sodium citrate was used to entrap Bifidobacterium longum ATCC 15707. Ions and pH of the inoculum during the immobilization step influenced the long-term mechanical stability of the gel beads during continuous fermentation in a stirred tank reactor. High stability as well as high biocatalyst activity was obtained when a washed cell suspension was used as the inoculum. Gellan gel produced by dissolving gellan gum in a sodium citrate solution may be a promising entrapment matrix for temperature-sensitive cells such as mesophilic lactic acid bacteria and eukaryotic cells.
Introduction Gellan gum is a gel-forming polysaccharide secreted by Pseudomonas elodea (Kang and Veeder, 1982). When the polysaccharide concentration is sufficient, gellan gum produces a thermoreversible gel by increasing ionic strength or decreasing temperature (Rinaudo,1988). The gel rheological properties depend upon the gum concentration, the ionic strength, the type of the stabilizing ion, and the acylation of the molecule (Moorhouse et al., 1981; Kang et al., 1982;Norton and Lacroix, 1990). The acylated form of the polysaccharide produces soft, elastic, and cohesive gels, whereas hard, firm, and brittle gels are obtained from the deacylated form (Rinaudo, 1988). The rheological properties of deacylated gellan gel are superior to those of other common polysaccharide gels such as agar, K-carrageenan, and alginate at similar concentrations (Sanderson et al., 1989). According to these authors, the hardness, brittleness, and elasticity of gellan gel at 0.5% is comparable to, and its stiffness is much higher than, 1%K-carrageenan, 1.5% agar, or 4.0% gelatin gels. Gellan gum has recently been approved for food use by the FDA (Dziezak,1990). It is already used as a substitute for agar in the preparation of microbiological media with improved clarity, reduced toxicity to sensitive microorganisms, and high melting temperature (Baird et al., 1983). Since most applications in food products are at a low concentration (0.05-0.2 % 1, only limited information on the rheological properties of the more concentrated gels is available (Nussinovitch et al., 1990). Gellan gel, which is stabilized by a large variety of cations (Grasdalen and
* Corresponding author. t
Universit4 de Bourgogne,
* Universi~Laval.
8756-7938/93/3009-0291$04.00/0
Smidsrod, 1987), has recently been proposed as an entrapment matrix to be used for fermentation of whey or whey permeate with immobilized bacteria at temperatures up to 60 "C (Norton and Lacroix, 1990). However, the high setting temperature of the commercial gellan gum at high concentration (over 50 "C) makes ita use for entrapment of temperature-sensitive mesophilic bacteria difficult. Sequestrant addition has been reported to promote molecule hydration and to lower the setting temperature of low-concentration gels (Sanderson et al., 1989). This work examined the feasibility of using gellan gel as an entrapment matrix for a mesophilic lactic acid bacterium, Bifidobacterium longum. Preliminary work on immobilization of this bacteria in a K-carrageenan/locust bean gum mixed gel developed and currently used for lactic acid bacteria entrapment (Arnaud et al., 1989a,b;Audet et al., 1992) revealed the high sensitivity of B. longum to K+ ions, which are reqiiired for the maintenance of gel bead integrity during continuous fermentation of a wheybased medium (Paquin et al., 1990). In this research, the effect of sequestrant addition on setting temperature and rheological properties of gellan gum gel at different concentrations was examined. A preparation method for gellan gum gel was proposed for B. longum or other mesophilic bacteria entrapment, and immobilized cells were used in continuous fermentation experiments to assess the mechanical stability of the gel. Materials a n d Methods Preparation of Gellan Gum Solutions. Deacetylated gellan gum (Gelrite, lot no. C.0051, Scott Laboratories Inc., West Warwick, RI) was dispersed in hot deionized water (70 "C) containing sequestrant. The solution was
0 1993 American Chemical Society and American Institute of Chemical Engineers
292
stirred for 5 min, heated to 90 "C to complete dissolution, and then sterilized for 10 min at 121 "C to simulate the sterilization step during an immobilization process (Audet et al., 1990). Three gellan gum concentrations (1.5, 2.0, and 2.5% (w/v))and three chelating agents (EDTA, sodium citrate, and sodium metaphosphate; Fisher Scientific, Nepean, Ontario, Canada) each at five concentrations (from 0 to 0.8% w/v) were investigated. Setting Temperature Measurements. Setting temperature of the gellan gum solutions containing various amounts of chelatant was measured in duplicate, using a mercury thermometer plunged into the setting solution. The setting temperature was recorded when a 8-10-mm print was observed. Rheological Tests. Sample Preparation. Medical syringes were filled with the hot liquid solution, allowed to cool at room temperature, and then stored for 24 h at 4 "C. Gels were cut into three cylinders with a height and a diameter of 27 mm. These samples were then soaked for 24 h in 5 vol of a 0.15 5% CaCl2 solution (pH 6.4) changed three times to limit the dilution effect brought about by the gel (Norton and Lacroix, 1990). Rheological Measurements. Two compression tests between two parallel plates were carried out at room temperature with an Instron 1101 Universal testing machine (Model TMS, Instron Canada Ltd., Burlington, Ontario, Canada) interfaced with a microcomputer. For the first test, samples were compressed until failure using a compression rate of 1cm/min and a 500-N Instron load cell. The force at failure was recorded, and the gel strength (N/cm2)was calculated by dividing the force at failure by the original cross-sectional area of the sample. During a second compression test, samples were compressed by 6 % of their height at a rate of 0.5 cm/min using a 20-N Instron load cell, as indicated by Norton and Lacroix (1990).This deformation was selected in order to remain in the linear portion of the stress-strain curve where the gel exhibits elastic behavior (Mitchell, 1980). Apparent modulus (N/ m2or Pa), also called gel stiffness, was calculated from the force-deformation plot, using the following equation (Gouarraze and Grossiord, 1983): apparent modulus = u/cL = (F/Ao)/(AH/H0) where u and e~ are, respectively, the compressive stress and strain, F (N) is the strength measured, HO(m) is the initial height of the sample, A0 (m2)is the cross-sectional area of the sample, and AH (m) is the deformation. Reported data are the averages of five replications. Experiments. Influence of Type and Concentration of Chelatants. Rheological measurementswere performed on gel samples made from gellan gum solutions at three concentrations (1.5, 2.0, and 2.5% (w/v)) containing chelatants between 0 and 0.8% (w/v). The pH values of the gels studied, measured by plunging a pH electrode into the sample, were 6.9 for gels without chelating agent (control) and 4.2, 7.0, and 5.5 for gels containing EDTA, sodium citrate, and sodium metaphosphate, respectively. Influence of p H . Two types of tests were carried out: (1)The gellangum powder was dissolvedin the sequestrant solution adjusted to a pH in the range 4-7 using 1N HC1 or 1 N NaOH. Gel samples were soaked for 24 h in a 0.15% CaCl2 solution (pH 6.4) before measurement, as indicated above. (2) Gel samples produced without pH adjustment were soaked in a CaC12 solution adjusted to pH 4.0 with 1 N HC1 to determine the possible influence of acidification of the medium. Influence of Added Cells. A fresh bacterial culture of Bifidobacterium longum ATCC 15707 was prepared as
Biotechnol. Prog., 1993, Vol. 9, No. 3
detailed below and was mixed at a 2 % (v/v) level with the 2.5 % gellan gum solution supplemented with 0.2 % citrate or metaphosphate and cooled down to 45 "C. Before addition, the pH of the inoculum was previously adjusted, in some cases, to 7.0 with 6 N NaOH. Controls were noninoculated gels made by the same procedure. Cell Immobilization Tests. Sterilized 2.5 % gellan gum solutions supplemented with 0.2 % citrate or metaphosphate were used to prepare gel beads with immobilized living cells. A fresh bacterial culture of Bifidobacterium longum ATCC 15707 was prepared by incubating a 2% (v/v) inoculum in a broth medium composed of Lactobacillus MRS (Rose11 Institute Inc., Montreal, Quebec, Canada) and a whey permeate, supplemented with agar, Na2C03, and L-cysteine (C. Paquin, M. Le Roy, and C. Lacroix, manuscript in preparation), at 37 "C for 16 h. The immobilization procedure was based on a dispersion process in a two-phase system described previously (Audet and Lacroix, 1989;Arnaud and Lacroix, 1991). The sterile polymer solution was inoculated at 5% (v/v) and 47 "C and dispersed in a stirred hydrophobic phase at 47 "C (commercial soya oil) to form a macroemulsion. Different inocula were tested including a fresh culture of B. longum, with or without pH adjustment at 7.0 using 11N NaOH, and a washed cell suspension obtained by centrifuging (10 OOOg at 4 "C) a fresh culture and washing three times with the same volume of sterile saline solution. Gel beads were formed by cooling down the macroemulsion and hardened by soaking for 24 h in a 0.15 % CaCl2 solution. Beads with diameters in the range 1.0-2.0 mm were separated by wet sieving and used in fermentation. This entire procedure was carried out aseptically. Survival rate after immobilization was determined from viable cell counts in the inoculum and in the freshly prepared gel beads. Colony-forming-unit counts (CFU/ mL) were obtained by duplicate pour plating on the culture medium with 1.5% agar (DifcoLabs, Detroit, MI), followed by incubation at 37 "C for 48 h in anaerobic jars (BBL Microbiology System, Becton Dickinson, Mississauga, Ontario, Canada). Viable cell count determination in the gel beads was performed after gel dissolution in a 1% EDTA solution at pH 7.0 and 45 "C, using a high-speed vortex and glass beads. Continuous fermentation was carried out at 37 "C in a 0.75-L (Bioflo Model C-30, New Brunswick, Edison, NJ) or a 1.25-L(Bioflo 3, New Brunswick) stirred tank reactor (CSTR) equipped with an automatic pH control, similar to Audet et al (1992). The reactor was fed with the autoclaved MRS-whey permeate medium suplemented with 0.15 % (w/v) CaC12 at a dilution rate between 0.5 and 2.0 h-l, and the pH was maintained at 5.5 with 4 N NaOH. Agitation set at 50-60 rpm was provided by two marine impellers (5.0 cm in diameter) in the 0.75-L reactor (Audet et al., 1992) and a single flat blade impeller (6.6 cm in diameter, 6.2 cm high, with four 1.6 cm wide rounded edge blades mounted at 60" to horizontal) in the 1.25-L reactor (Arnaud et al. 1992). Bead appearance and integrity were evaluated by visual observation and by comparing bead volume, as measured by displacement in a graduated cylinder, at the beginning and at the end of the continuous fermentations (Audet et al., 1992). Statistical Analysis. A one-factor analysis of variance with the Scheffe multiple comparison test was carried out (two and five replications per treatment for temperature measurements and rheological tests, respectively). The average coefficients of variation were 1% for setting temperature, 5% for gel strength, and 10% for apparent modulus.
Biotechnol. hog., 1993, Vol. 9, No. 3
5 h
d
45
I
293
-
I
1.a
16
N
40
E P
35
12
E
10
v1
.'
a
4
3 0,O
0,6 0,8 Chelatant concentration (%w/v)
0,2
I
I
1:
30
'.
14
E
2
".
1,0
0,4
0,o
0,4
02
1,o
08
05
Chelatant concentration (%w/v)
-#-
Metaphosphate
Metaphosphate m
254 0,O
I
I
.
I
I
+ Citrate I
0,2 0,4 0,6 0,8 Chelatant concentration (w/v)
e
