Comparative analysis of immobilization carriers for a endopolygalacturonase producing yeast strain

Text-only Preview

30th Conference SSCHE, Proceedings on CD ROM, Tatranské Matliare (SK), 26 – 30 May, 2003
30th Conference SSCHE, Proceedings on CD ROM, Tatranské Matliare (SK), 26 – 30 May, 2003
Comparative analysis of immobilization carriers for a
endopolygalacturonase producing yeast strain

Catarina Almeida 1,2, Tomáš Brányik 1, Pedro Moradas-Ferreira3, José Teixeira 1*

1 – CEB - Instituto de Biologia e Química Fina Universidade do Minho, Campus de
Gualtar 4710-057 BRAGA – PORTUGAL, Email: [email protected]
2 - Instituto Superior de Ciências da Saúde-Sul Quinta da Granja, 2829-511 MONTE
3 – Instituto de Biologia Molecular e Celular Rua do Campo Alegre, 823, 4150-180
PORTO - PORTUGAL Instituto de Ciências Biomédicas Abel Salazar Universidade do
Porto, Largo Prof. Abel Salazar, 2 - P-4099-003 PORTO - PORTUGAL
(*Corresponding author)

Key words: Continuous reactors, immobilized yeast cells, pectinase
Pectinases are a group of enzymes acting on pectin and other pectic substances
found in vegetable tissues. Pectin consists of a ?-1,4 polymer of D-galacturonic acid; the
main chain is 60 – 90 % methylated and includes rhamnose units and side chains of
arabinan, galactan and arabinogalactan. This complex polymer has a structural, scaffolding
function in the primary cell wall and in the middle lamella of plant tissues (Naidu and
Panda, 1998). In natural habitats, several microorganisms excrete different pectolytic
enzymes to invade the cell walls and thus grow on the rich substrates found in plants. In the
industrial world, pectinases have found use in any process that deals with extracting juices
from fruits and vegetables and in the processing of plant tissues. Some examples are: fruit
juice (apple, banana, mango, orange, guava…) clarification and viscosity reduction, as a
preliminary grape treatment in wine industries, in tomato pulp extraction, in chocolate and
tea fermentation, vegetable waste treatment, fiber degguming in textile industries and paper
industries (Kashyap, et al. 2001). Aspergillus niger is currently the only microorganism
used for pectinase industrial production. It excretes a mixture of pectolytic enzymes
(polygalacturonases, polymethylgalacturonases, pectin lyases and pectin esterases) together
with other degrading enzymes such as arabinofuranosidases and amyloglucosidases
(Blanco et al., 1999). In fact, commercial pectinase is a blend of enzymes. This can be
useful due to the complexity of plant tissues and all the different chemical bonds to
hydrolyse, but in some industrial cases a specific type of pectinase, or a specific blend is
needed (Manachini et al., 1988). The yeast strain Kluyveromyces marxianus CCT 3172 was
selected from a cocoa fermentation as a good endopolygalacturonase producer (Schwan and
Rose, 1994, Schwan et al., 1997).
Continuous production of an endopolygalacturonase from yeast would be an
interesting alternative to the current fungal batch production. To increase continuous
reactors’ productivity, high cell density systems can be used. Some carriers have been


successfully tested for yeasts and bacteria immobilization, namely porous materials such as
porous glass and ceramic, synthetic polymers, cellulosic fibers and cellulose derivatives,
activated charcoal, artificial polymers and gel matrixes as k-carrageenan, Ca and Ba
alginate, and pectate (Hartmeier, 1988, Mensour et al. 1996, Pilkington et al., 1999, Tata et
1999, Barranco-Florido et al. 2001, Navrátil et al. 2002).
In this work, a packed bed reactor (PBR) was chosen for pectinase production. Two
carriers were tested for cell immobilization: a commercial porous silicate glass (Siran) and
a recently tested cellulosic support, prepared from spent-grains, a by-product of the brewing
industry. Although Siran has been used with satisfying results for cell and enzyme
immobilization (Virkajärvi and Kronlöf, 1998, Castein, et al., 1999, Srivastava and
Onodera, 1998, Pérez et al., 1997, Racher and Griffiths, 1993), its high cost can be a
drawback for industrial productions. The cellulosic carrier has already been tested with a
brewing strain (Saccharomyces uvarum) and was found to be very efficient, having a high
yeast loading capacity, together with an easy preparation, reusability, availability and an
inert, non-toxic nature (Brányik et al., 2001, 2002).
Materials and methods
Media and strain
The wild type Kluyveromyces marxianus CCT 3172 used to inoculate both
continuous reactors was pre-grown in 200 ml (in 500 ml Erlenmeyer flasks) semi-synthetic
media at 30 ºC, 120 r. p. m. for 24 h.
The semi-synthetic medium (SS) for yeast growth included (g l-1): 5 K2HPO4,
2 (NH4)2SO4, 0,4 MgSO4.7H2O, 1 yeast extract, and different glucose concentrations (20,
40, 80).
The carriers used were porous silicate glass beads (Siran, SIKU012/05/120A, QVF
Engineering, Mainz, Germany with bead diameters 2 to 3 mm and porosity 50 to 65 %) and
a cellulosic support prepared from spent grains, a by-product of the brewing industry. Siran
beads were washed in distilled water and autoclaved twice before the first use. The steps
followed to obtain the carrier from spent grains are described in Brányik et al., (2001).
Enzyme assays
The endopolygalacturonase activity in the reactor effluent was assessed using the
method described by Honda and co-workers (1982). On unit (U) is defined as 1 µmol of
galacturonic acid released after 1 min of hydrolysis of polygalacturonic acid in the presence
of the enzyme at 40 oC, pH 4.1.
Analytical methods
Glucose concentration was determined by the DNS method for reducing sugars
quantification (Miller, 1954). For pectinase activity determination, the samples from
reactors were centrifuged, filtered and then dialyzed with a 14 000 MWCO membrane
against cold distilled water for 16 h. Lactose was used as a tracer for hydrodynamic studies.


Its concentration was determined using the specific enzymatic kit for detection of lactose
and D-galactose from Boehringer Manheim / Roche.
Cells contact angle measurements
A solution of 20 g l-1 of agar and 10 % glycerol was cast on a microscope slide. Cell
samples were taken from a continuous reactor outflow and washed with a solution with
increasing ethanol concentration (10, 20 and 50 % (w/v)). 1 ml of a cell suspension in 50 %
ethanol with a Abs 600 nm = 2.0 was spread on the solidified agar and glycerol and allowed
to dry. This step was repeated 4 times (Henriques et al. 2002). Contact angles were
measured at room temperature using water, formamide and ?-bromonaphtalene in a contact
angle apparatus (Kruss-GmgH, Germany) by the sessile drop technique. The total surface
tension (?tot) and its components (?LW, ?+, ?-, ?AB), the values of the free energy of
interaction between cells and water
and the components (
, G
) were
calculated according to van Oss and co-workers (van Oss et al., 1988).
Scanning electron microscopy (SEM)
A sample of biocatalyst was taken from the CSTR reactor, washed with water and
with a solution with increasing ethanol concentration (10, 25, 50, 75, 90, 100 %). It was
allowed to dry for 5 days in an exicator and covered with a thin gold layer to allow for
SEM observation.
Biomass quantification
The free biomass concentration at the reactors’ outlet was measured by reading the
absorbance of samples at 600 nm and then converting this value to dry weight per volume
using an appropriate calibration curve.
At the end of the reactor’s operation, samples of the biocatalyst were withdrawn
from different heights of the fixed bed. For the spent grains, the carrier with adsorbed yeast
cells was gently washed with 200 ml of distilled water. The resulting suspension was
filtered and washed with water; the filter paper with carrier and immobilized cells was dried
at 105°C for 16 h. A washing step (during 24 h at 120 r.p.m.) with a 3 % (w/v) NaOH
solution released the attached biomass. After washing with distilled water and filtering, the
biomass free carrier was dried at 105°C for 5 h. The biomass weight was calculated as
being the weight difference of the dry carrier before and after the NaOH washing. (Brányik
et al., 2002).
For the Siran carrier, the biocatalyst samples were dried for 48 h at 105oC and
weighted. The biomass was then combusted at 550oC for 2 hours and the residual material
was weighted. The biomass load (Xi) was calculated as the weight difference between the
dry biocatalyst and the clean carrier. Biomass loads were expressed in g biomass g-1 carrier.
Bioreactor start-up and operation
Before operation, the reactor was sterilized with a hypochlorite solution (3 days).
After this period of time, 10 reactors’ volumes of sterile distilled water were used to wash


the column. The packed bed reactor (PBR) was a “Perspex” column, with height to internal
diameter rate (H/Di) of 12 and an operation volume of 310 ml.
When using the spent-grains as cells support, 25 g of sterile dry carrier were
aseptically inserted in the column and then inoculated with a pre-grown yeast culture. After
24 h of batch growth in the reactor, continuous operation started by feeding SS medium
with 40 g l-1 glucose at the bottom of the column. A recycle rate of 40 ml h-1 was used
during the entire operation time, by re-introducing a part of the outflow to the bottom of the
When Siran was used as carrier, approx. 150 g of sterilized dried beads where
introduced into an Erlenmeyer flask with SS medium (20 g l-1 glucose) than inoculated with
a pre-grown 100 ml culture. After 24 h of incubation at 30 oC, 120 r.p.m., the Siran beads
were transferred into the reactor which was then filled with fresh SS medium (40 g l-1).
Continuous operation started 24 hours after the transfer. For this experiment the recycle rate
was 200 ml h-1.
The dilution rate was considered to be D (h-1) = volumetric feed rate / total working
volume of the reactor. All the assays were performed at 25 oC.
Results and discussion
Cell immobilization
The cellulosic carrier from spent grains is both irregular in shape and
non-homogeneous in chemical composition, originating “active sites” preferably colonized
by yeasts (Brányik et al., 2001). Siran beads are composed of silicate glass with an open
pore structure, relatively uniform in size but with an irregular orientation and shape (Fig 1).
Contact angles measured by the sessile drop technique (Henriques et al., 2002) were
used to calculate the yeast surface properties according to van Oss (1988, 1995). The values
of total surface tension and free energy of interaction for K. marxianus CCT 3172 cells, for
base-treated spent grains carrier (Brányik T., unpublished results) and for Siran (Nakari-
Setälä et al., 2002) are presented in Table 1. The high positive ? tot
G value found for the
Siran carrier is associated with its surface hydrophilic character, and the negative value
determined for the cellulosic carrier shows the presence of highly hydrophobic areas. From
these values, the free energy of interaction between K. marxianus cells and the two
different carriers was calculated. The values of
=1.86±8.70 mJ m-2 (the wide error
bar is associated with the non-uniform surface composition of the cellulosic carrier) for the
interaction cells-water-spent grains and
=55.5±5.0 mJ m-2 for cells-water-Siran
showed an energetically less favourable adhesion between yeast cells and the surface of
Siran beads. These results are in agreement with the observed behaviour of cells in the two
different biocatalyst beds (Fig 1). The Siran carrier seems to immobilize cells only by
spatial retention on its porous structure. It has an open pore matrix with an important
presence of void spaces, allowing liquid motion and cell percolation through the fixed bed.
This originates a difference in the biomass load at different heights in the column (for
distances from the top of 0, 8, 18 and 31 cm biomass loads (Xi) of 0.072, 0.054, 0.065, and
0.143 g g-1carrier were found at the end of reactors operation). As it was not possible to
withdraw biocatalyst samples during reactors operation, four separated assays were
performed using a smaller column and 1/10 of the Siran weight (approx. 15 g). A fixed


dilution rate of D = 0.33 h-1 was used and the four assays were stopped at 120, 240, 408 and
672 hours for immobilized biomass quantification. After 240 hours of continuous
operation, biomass load reached a stationary value (Figure 2). From these experiments, it
can be assumed that the immobilized biomass inside the PBR was nearly constant after the
first 240 hours of reactor’s operation.
When the cellulosic carrier was used, cells were attached to the irregular surface not
only by retention inside fibres, threads and crevices, but also by cell - surface adhesion due
to different interaction forces. In addition to that, the spent grains packing worked as a
“filter layer” giving rise to zones of local accumulation of yeasts. The CO2 bubbles formed
during the experiment were to account for a sponge-like bed structure. The gas was
periodically liberated through the top of the column thus mixing the packed bed and
releasing parts of the biomass deposits. At the end of the operation time, the PBR reactor
had a biomass load of 0.247 g g-1 carrier at the bottom and 0.204 g g-1 at the top showing a
higher homogeneity in the bed colonisation than for the PBR with Siran.
Hydrodynamic studies (Residence time distribution)
Experiments were carried out to study the hydrodynamic behaviour inside the PBR
for both tested carriers. Lactose was used as a tracer since this particular strain of K.
is unable to metabolise it efficiently. Besides, glucose is not totally consumed
during the experiments with the tracer, which acts as a catabolic repressor to lactose
A lactose concentration step was imposed at steady state conditions for both
reactors. The residence time distribution is presented in Figures 3a and 3b. When the Siran
carrier was used, the best fit for experimental tracer response seems to be an ideal CSTR.
For the axial dispersion model, Peclet number is the fitting parameter, defined as uL/D
(u being the linear velocity, L the height of the biocatalyst bed and D the axial dispersion
coefficient); Pe = ? corresponds to ideal plug flow, and Pe = 0 to ideal mixed flow. The
low Peclet (uL/D) value and the poor correlation obtained (Pe=1.07± 1.01 r2=0.87) for the
fitting with the axial dispersion model also suggest a nearly perfect mixing inside this
biocatalyst bed. In fact, as already discussed above, this packed bed has an open pore
matrix with a high void volume, which is likely to have a low resistance to mass transfer
and fluid motion (Fig. 1).
Using spent grains as cell carrier, a good agreement was found both for the axial
dispersion model with a Pe number of 5.51 ± 1.01 (r2 = 0,98) and for a series of 3 CSTRs.
From this, it can be implied that the reactor mixing is not negligible, however it has a lower
extent than in the packed bed with Siran. The fibrous structure of the spent grains bed,
working like a thick filter media, justifies the lower mixing found for the experiment with
this cellulosic carrier.
A recycle was used in both situations but in the case of the spent grains bed it’s
volumetric flow was five times lower than the one used with Siran to avoid fluidisation of
the lighter spent grains carrier. This fact can also account for the differences in the mixing


Pectinase production
During the operation of both packed-bed reactors, the free biomass concentration,
glucose concentration and pectinase activity were measured at the reactors outlet (Fig. 4 for
the PBR with spent grains; the data for the PBR with Siran is not shown). Similar values
were found for pectinase activities: in the PBR with spent grains the values oscillated
between 2.45 U ml-1 and 7.82 U ml-1, while in the PBR with Siran the pectinase activities
ranged from 3.08 U ml-1 to 7.72 U ml-1.
Using the PBR with spent grains, the volumetric productivity (PV) values range
from 0.61 to 0.98 U ml-1 h-1 and increase with the dilution rate (Figure 5). For the Siran
packed reactor, productivity also increased with the dilution rate and ranged from 0.39 to
1.68 U ml-1 h-1(Figure 5). The highest value of PV in the PBR (1.68 U ml-1 h-1) was found
for the Siran bed working with a D = 0.298 h-1 and an inlet sugar concentration of 40 g l-1.
When the specific pectinase production rate (qP) and the specific glucose
consumption rate (qS) were plotted against D (Sousa et al., 1994), the same tendency of
linear increase was found (Figure 6). As no samples were taken from the packed bed
throughout the operation time, the values of the immobilized biomass were estimated using
the biomass accumulation trend from Figure 2. In the case of the Siran bed after the first
240 h of continuous operation the average biomass concentration was considered constant
and equal to 34 g l-1 reactor volume. A similar behavior of the bed colonization was
assumed for the spent grains, and therefore the biomass concentration was considered also
constant (18.5 g l-1) after 240 h of the reactor’s operation.
For the spent grains PBR, the qS and qP values found using Sin = 40 g l-1 and
Sin = 80 g l-1 are similar, indicating that there is a limitation for glucose conversion to
pectinase in this column. In fact, increasing D with high glucose concentrations in the inlet
is useless, since it results in a higher sugar concentration at the outlet. This was not
noticeable for Sin = 20 g l-1 with the tested dilution rates.
In the Siran PBR glucose total consumption was achieved only at a D = 0.105 h-1
and Sin = 40 g l-1. Using the same dilution rates and the same inlet sugar concentration (Sin)
of 40 g l-1, the qS of the cells in the spent grains bed is slightly higher than for yeast
immobilized in Siran which consequently results in a higher pectinase production rate (qp)
in the immobilized system with spent grains. Although the packed bed with Siran has a
higher biomass concentration, its non-uniform distribution throughout the column is
possibly the reason for this difference. In fact, the bottom of the reactor (about 15 to 20 %
in terms of reactor volume) had a large and, in some places, compact yeast accumulation
(estimated as 25 to 30 % of the total biomass in the column) that probably imposed a high
mass transfer resistance for glucose and pectinase. The values found for qS at Sin = 80 g l-1
are higher than for Sin = 40 g l-1, but this difference was not, as should be expected,
reflected in the qP values, which are lower for Sin = 80 g l-1. This contradictory results can
also be a consequence of the uneven biomass distribution in the Siran column. As the
assays with Sin = 80 g l-1 were performed after the ones with Sin = 40 g l-1, the biomass
accumulation at the bottom of the column was increased and these clustered cells are
probably less efficient in converting sugar to pectinase due to nutrient restrictions imposed
by the high mass transfer resistance.
Both carriers can be successfully used for cell immobilization although, to increase
the biomass loads and improve its distribution throughout the column, the surface of the
Siran carrier should be activated. In fact, a pre-treatment with trimethylchlorosilane


favorably changes its surface properties for yeast attachment, as a six-fold decrease is found
for the free energy of interaction (
) (Nakari-Setälä et al., 2002).
From the performed assays, it can be concluded that the best results for pectinase
production can be achieved using a high Xi in the column and a high D, together with total
glucose consumption.
The better biomass distribution throughout the column obtained in the spent grains
PBR and the advantages related to this available by-product from brewing industries make
it a suitable option as a cell carrier.
Barranco-Florido, E., Garcia-Garibay M., Gomez-Ruiz, L., Azaola, A., Immobilization
system of Kluyveromyces marxianus cells in barium alginate for inulin hydrolysis. Process
Biochem., 37 513-519 (2001).

Blanco, P., Sieiro, C., Villa, T. G., Production of pectic enzymes in yeasts. FEMS
Microbiol. Lett., 175, 1-9 (1999).

Brányik, T., Vicente, A. A., Machado Cruz, J. M., Teixeira, J. A., Continuous primary
beer fermentation with brewing yeast immobilized on spent grains. J Inst Brew 108 (4)
410-415 (2002).

Brányik, T., Vicente, A.A., Machado Cruz, J.M., Teixeira, J. A. Spent grains - a new
support for brewing yeast immobilisation. Biotechnol. Lett. 23:1073-1078 (2001).

Canstein, H., Li, Y., Timmis, K.N., Deckwer, W.-D., Wagner-Dobler, I., Removal of
mercury from chloralkali electrolysis wastewater bu mercury-resistant Pseudomonas
, Appl. Environ. Microbiol. 65 (12) 5279-5284 (1999)

Hartmeier, W., Immobilized Biocatalysts – An Introduction, Springer Verlag ed. (1988).

Honda, S., Nishimura, Y.: Takahashi, M., Chiba, H., Kakehi, K., A manual method for
the spectrophotometric determination of reducing carbohydrates with 2-cyanoacetamide,
Anal. Biochem. 119, 1, 194-199 (1982).
Henriques, M., Gasparetto, K., Azevedo, J., Oliveira, R., Experimental methodology to
quantify Candida albicans cell surface hydrophobicity. Biotechnol. Lett. 24, 1111-1115

Kashyap, D.R., Vohra, P.K., Chopra, S. and Tewari, R., Applications of pectinases in
the commercial sector: a review. Bioresource Technol.77, 215-227 (2001).

Manachini, P. L., Parani, C., Fortina, M. G., Pectic enzymes from Aspergillus pullulans
LV10. Enzyme Microbiol. and Technol., 10 (11), 682-685 (1988).

Mensour, N. A., Margaritis, A., Briens, C. L., Pilkington, H., Russel, I., Application of
immobilized yeasts in the brewing industry. In: Wijffels R. H., Buitelaar, R. M., Bucke,


C., Tramper, J., eds, “Immmobilised Cells: Basics and Applications” Amsterdam,
Elsevier Science B. V., 661-671 (1996).

Miller, G. L., Use of dinitrosalicylic acid reagent for determination of reducing sugars.
Anal. Chem. 31, 426-428 (1959).

Naidu, G. S. N. , Panda, T., Production of pectolytic enzymes –a review. Bioprocess Eng.
19, 355-361 (1998).

Nakari-Setälä, T., Azeredo, J., Henriques, M., Oliveira, R., Teixeira, J., Linder, M.,
Expression of a fungal hydrophobin in Saccharomyces cerevisiae cell wall: effect
on cell surface properties and immobilization, Appl. Environ. Microbiol. 68 (7) 3385-3391

Navrátil, M., Gemeiner, P., Klein, J., Sturdik, E., Malovikova, A., Nahalka, J.,
Vikartovska, A., Domeny, Z., Smogrovicova, D.,
Properties of hydrogel materials used
for entrapment of microbial cells in production of fermented beverages. Art. Cells, Blood
Subs., and Immob. Biotech. 30, 3, 199-218 (2002).

Pérez, M., Romero, L.I., Nebot, E., Sales, D., Colonisation of porous sintered-glass
support in anaerobic thermophilic bioreactors. Bioresource Technol, 59, 177-183 (1997).

Pilkington, H., Margaritis, A., Mensour, N., Sobczak, Hancock, I., Russel I., Kappa-
carrageenan gel immobilisation of lager brewing yeast. J. I. Brewing 105, 6, 398 404

Racher, A.J., Griffiths, J.B., Investigation of parameters affecting a fixed bed bioreactor
process for recombinant cell lines Cytotechnology;13 (2):125-31 (1993).

Schwan, R.F., Cooper R.M., Wheals, A.E., Endopolygalacturonase secretion by
Kluyveromyces marxianus and other cocoa pulp-degrading yeasts. Enzyme Microb.
Technol. 21, 234-244 (1997).

Schwan, R.F., Rose, A.H., Polygalacturonase production by Kluyveromyces marxianus:
Effect of medium composition. J. Appl. Bacteriol. 76, 62-67 (1994).
Sousa, M. L., Mota, M., Teixeira, J.A., Influence of operational parameters on the start up
of a flocculation airlift bioreactor . Colloids and Surf. B: Biointerfaces 2, 181-188 (1994).

Srivastava, P., Onodera, R. J., A comparative evaluation of Cephalosporin C production
using various immobilization modes. Gen. Appl. Microbiol 44(2), 113-117. (1998).

Tata, M., Bower, P., Bromberg, S., Duncombe, D., Fehring, J., Lau, V. V., Ryder, D.,
Stassi, P.,
Immobilized yeast bioreactor systems for continuous beer fermentation
Biotechnol. Prog. 15,1, 105-113 (1999).

van Oss, C. J., Chaudhury, M. K. Good R. J., Interfacial Lifshitz-van der Walls and
polar interaction in macroscopic systems. Chem. Rev. 88, 927-941, (1988).


van Oss, C. J., Hydrophobicity of biosurfaces – origin, quantitative determination and
interaction energies. Colloids and Surfaces B: Biointerfaces 5, 91-110 (1995).

Virkajärvi, I., Kronlöf, J., Long-term stability of immobilized yeast columns in primary
fermentation. J. Am. Soc. Brew. Chem., 56 (2) 70-75 (1998).


Figures and Table

Figure 1 - SEM photos. Top images - cells on spent grains. Bottom images – cells on Siran
beads. Left images- bar corresponds to 20 µm Right images– bar corresponds to 100 µm.
carrier 0,06
-1 g 0,04
biomass 0,02
X i
t (h)

Figure 2 – Immobilized biomass load (Xi) in the Siran packed bed (assays with the smaller
column reactors).


Document Outline
  • bookmarks
    • Main Page
    • About Conference
    • List of Papers
    • Author Index
    • Exit