In order to immobilize proteins covalently under gentle conditions
to a surface of any polysaccharide-based matrix, an activation of the
matrix is required. Numerous activation chemistries are now available to
couple proteins through the amino, thiol, aldehyde, or carboxyl groups
[1-3]. The decision about the activation method of carriers is based on
the functional groups on the surface of the matrix and of the protein.
The activated complex then has generated a covalent linkage between the
protein and the carrier, resulting in protein immobilization .
The coverage with proteins on the solid-phase surface will affect
working capacities up to a point where steric hindrance may diminish the
performance. Therefore, a spacer arm is interposed between the matrix
and the protein to facilitate the efficiency. Spacer arms must be
designed to maximize binding without non-specific binding effects .
Matrices based on cellulose are frequently used as a carrier for
protein immobilization because of their biocompatibility and
hydrophylicity. Moreover, cellulose has good potential for varied
derivatization due to the presence of three hydroxyl groups in every
monomeric unit . There are well-known methods for the activation of
cellulose-based matrices with cyanogen bromide , bisepoxirane ,
organic sulphonyl chlorides , or carbodiimidazole .
In this work a method of the activation of a cellulose-based matrix
with pentaethylenehexamine (PEHA) was developed. The method gives active
primary amino groups via a long spacer arm. PEHA-activated cellulose was
evaluated for the preparation of lectin-affinity adsorbents, as well as
for the immobilization of enzymes.
Cellulose-based matrix Granocel was prepared by saponification of
diacetylcellulose (Roshal, Russia, 55.0% bond acetic acid) by the
procedure described previously . Pentaethylenehexamine and
glutaraldehyde were purchased from Fluka (Steinheim, Germany), and
1-chloro-2,3-epoxypropane (epichorhydrin) was from Aldrich (Poznan,
Poland). Liquid glucoamylase (Amidase, Batch 8187/SPE 0551) was kindly
donated by Gist-Brocades (The Netherlands). The lectins Concanavalin A
(type V) and Wheat Germ Agglutinin (WGA), the glycoproteins glucose
oxidase (GOD) and fetuin, as well as sodium borohydride and Bradford
reagent were purchased from Sigma (Munich, Germany). All other chemicals
and solvents were of analytical-reagent grade.
Two-step preparation of PEHA-cellulose
Epoxidation of the cellulose-based matrix
A 25 g amount of sucked cellulose-based matrix Granocel was
suspended in 30 mL of 5% NaOH solution, the required amount of
epichlorhydrin (EC1H) and 30 mg of sodium borohydride were added, and
the mixture was stirred at 40[degrees]C for 2 h. Then the activated
matrix was thoroughly washed with water.
The concentration of epoxy groups was determined as follows: 1 g of
epoxydated cellulose was immersed in 10 mL of 1.3 mol [L.sup.-1] sodium
thiosulphate and the obtained sodium hydroxide was titrated with 0.1 mol
[L.sup.-1] HCl keeping the pH near 7.0. The 1 mL of HCl used for
titration corresponds to 100 [micro]mol of epoxy groups.
Coupling of the PEHA spacer
One gram of sucked epoxidated cellulose matrix was mixed with 2 mL
of PEHA, and the mixture was stirred at 40[degrees]C for 2 h. The
product was washed with 0.5 mol [L.sup.-1] NaOH and water.
One-step preparation of PEHA-cellulose
Fifty grams of sucked cellulose Granocel was suspended in the
solution containing the required amounts of 4% NaOH solution and PEHA.
The reaction mixture was heated up to 50[degrees]C and then the required
amount of epichlorhydrin and also sodium borohydride NaB[H.sub.4] were
added. The reaction mixture was stirred for 3.5 h at 50[degrees]C. The
product was washed with 0.5 mol [L.sup.-1] NaOH and water. The amounts
of reagents and characteristics of products are presented in Results and
Evaluation of nitrogen content and primary amino groups
The primary amino groups were determined by desamination with
sodium nitrite. Therefore PEHA-cellulose was heated at 70[degrees]C for
7 h in a solution containing 0.1 mol [L.sup.-1] of NaN[O.sub.2] and 0.2
mol [L.sup.-1] of acetic acid (30 mL per g of cellulose). After
desamination the cellulose was thoroughly washed with water and the
nitrogen content was determined by the Kjeldahl method. The nitrogen
content of the primary amino groups was calculated as a difference
between the nitrogen content in the cellulose before and after the
Activation of PEHA-cellulose with glutaraldehyde
Five grams of sucked PEHA-cellulose was mixed with 9.4 mL of 0.05
mol [L.sup.-1] phosphate buffer, pH 8.5, and 2.7 mL of 25% (v/v)
glutaraldehyde. The pH was regulated with NaOH up to 9.3. The reaction
mixture was stirred at room temperature for 2.5 h. After reaction, the
activated cellulose was filtrated and washed with 0.05 mol [L.sup.-1]
phosphate buffer, pH 7.5.
Blocking of the unreacted aldehyde groups
NaBH4 was used for the reduction of the residual aldehyde groups.
One gram of the support was immersed into 2 mL of 0.1 mol [L.sup.-1]
phosphate buffer (pH 7.4), adding 2 mg NaB[H.sub.4]. The suspension was
kept in a refrigerator for 1 h. Afterwards the adsorbent was washed with
the same buffer.
Immobilization of the lectins
One gram of the sucked activated support was immersed into 3 mL
solution of either lectin ConA or lectin WGA (10 mg [mL.sup.-1]) in 0.05
mol [L.sup.-1] phosphate buffer containing 5 x [10.sup.-3] mol
[L.sup.-1] of [Mg.sup.2+] (pH 7.5). Afterwards the sorbent was washed
with the coupling buffer containing 0.5 mol [L.sup.-1] of NaCl to
eliminate protein-protein interactions. The washing supernatants were
collected and the protein concentration was measured by the Bradford
Immobilization of glucoamylase
The swollen carrier (5 mL) was rinsed 5 times with distilled water
and the 0.1 mol [L.sup.-1] phosphate buffer (pH 7.0) as was described
previously . Activation of the N[H.sub.2]-groups by glutaraldehyde
(pH 7.0) was stopped by sucking out all the liquid and rinsing with
distilled water and buffer. After filtering, the activated carrier was
suspended in 10 mL of protein in the same buffer. The excess protein was
washed off with 0.1 mol [L.sup.-1] phosphate buffer, the same buffer
with 0.5 mol [L.sup.-1] NaCl, and 0.1 mol [L.sup.-1] acetate buffer (pH
5.0). All the eluates were collected and analysed for the presence of
protein and activity. The enzyme activity was assayed in the presence of
1.25% gelatinized soluble starch (pH 4.5, 50[degrees]C) and the released
glucose was measured by a glucose oxidase-peroxidase enzymatic assay kit
(Glucoza EO, POCh Gliwice, Poland). The enzyme activity unit (U) was
defined as the amount of enzyme liberating 1 mmol [L.sup.-1] glucose per
minute. Protein concentration was determined by Lowry's method
(Sigma procedure P-5656).
RESULTS AND DISCUSSION
Activation of the cellulose-based matrix with PEHA
In this work macroporous cellulose Granocel  was used as a
matrix (particle size of 200-315 [micro]m). As it was determined by
means of inverse gel-permeation chromatography, the pores of the matrix
Granocel-4000 are accessible to molecules of molecular mass up to 2 x
Carriers based on cellulose Granocel can be produced easily in a
wide variety of particle diameter, surface area, and pore size and they
have a good potential for varied derivatization. In order to indroduce
primary amino groups via a spacer arm, the activation of the Granocel
matrix with PEHA was studied. The activation procedure comprises two
steps: (i) epoxy activation of cellulose and (ii) coupling of PEHA to
epoxy groups. The synthesis pathway is shown in Fig. 1, a-c.
The epoxy groups were introduced by means of a well-known method
based on the reaction of cellulose with epichlorohydrin in the presence
of sodium hydroxide [14, 15]. The results in Table 1 show the dependence
of the density of epoxy groups on the ratio EClH/Cel in the reaction
[FIGURE 1 OMITTED]
Afterwards the reaction of epoxydated cellulose with PEHA was
performed (Fig. 1, c) according to the procedures described in Methods.
The dependence of the content of primary amino groups on the epoxy
groups density was investigated keeping the ratio PEHA/Cel constant at 2
mL [g.sup.-1]. Results presented in Table 2 show that the content of
primary amino groups may be regulated by adjusting the density of epoxy
The density of the primary amino group required for protein
immobilization depends on the protein. Usually a low density of active
groups is preferable in order to avoid multipoint attachment of the
protein. This activation method allows getting a carrier with a desired
content of the primary amino group.
While epoxy groups are not very stable, the PEHA coupling should
preferably be performed directly after the epoxydation step. In order to
simplify the procedures, it is possible to join the epoxydation and the
PEHA coupling steps together (see Methods). In this case the reaction
mixture contains both EClH and PEHA.
An immobilized PEHA spacer adds a 19-atom arm containing hydroxyl
and amino groups onto which a further immobilization of the proteins may
be performed. The presence of spacer molecules ensures that steric
limitations during the protein immobilization are kept to a minimum,
which is especially important considering the size of the protein
molecules. A second benefit arising from the use of a spacer is that the
protein is kept away from the particle surface, thus protein-surface
interactions could be minimized.
The primary and secondary amino groups of the PEHA spacer arm are
of weak basicity. This means that they are not dissociated at pH above
6.5-7. Therefore, the possibility of non-specific ionic interaction
between amino groups and proteins is negligible. A long hydrocarbon arm
chain could be involved into hydrophobic interactions. However, it was
found  that sometimes some additional forces provided by a spacer arm
are needed for the stronger ligand binding.
The method developed is not restricted just to cellulose-based
matrices, but may be used for other matrices containing free hydroxyl
groups, such as agarose and dextran.
PEHA-activated cellulose was evaluated for the preparation of
lectin-affinity adsorbents, as well as for the immobilization of
Immobilization of lectins on PEHA-cellulose
PEHA-activated cellulose was used as a carrier for the preparation
of lectin-affinity adsorbents. With the aim to immobilize any lectin,
the primary amino groups of the spacer arm may be activated with
glutaraldehyde. The reaction of primary amino groups of the PEHA spacer
arm with glutaraldehyde results in aldehyde end-groups onto which a
further immobilization of specific ligands may be performed (Fig. 1, d).
The covalent attachment of the lectin onto the aldehyde-activated
supports takes place at the primary amino groups of the lectins. This
reaction results in obtaining the so-called Schiff's base:
[FORMULA NOT REPRODUCIBLE IN ASCII]
The subsequent reduction with sodium borohydride led to the
stabilization of the bonds between the protein and the polysaccharide,
and to the reduction of the residual aldehyde groups .
Two supports, PEHA(0.37)-Cel and PEHA(1.1)-Cel with 0.37% and 1.1%
nitrogen, respectively, were prepared for lectin immobilization. The
content of primary amino groups was 0.10 and 0.16 mmol [g.sup.-1],
respectively. Two lectins of different molecular weight, such as
Concanavalin A (ConA, MM = 108 kDa) and Wheat Germ Agglutinin (WGA, MM =
36 kDa) were employed as ligands for the immobilization onto the
The same matrix Granocel activated with sodium periodate (OXY-Cel)
was used for comparison. OXY-Cel contains aldehyde groups introduced
directly into the cellulose chain without any spacer.
As Fig. 2 demonstrates, the immobilization behaviour of ConA on the
two supports is different. The faster kinetics and the higher amount of
the coupled ligand were achieved on the PEHA(1.1)-Cel support with the
spacer arm. The coupling of WGA is identical on both supports. The
immobilization kinetics mainly depends on the accessibility of the
active groups of the support to ligand molecules. Thus, the spacer arm
evidently increases the accessibility of voluminous molecules of ConA to
the active groups of the support.
ConA and WGA adsorbents were evaluated for the sorption of glucose
oxidase and fetuin, respectively (Table 3). The effect of the spacer arm
on the chromatographic behaviour of the adsorbents in the column was
followed as well. Although the lectin density on PEHA-Cel and OXY-Cel is
very similar, the glycoprotein sorption capacity is significantly higher
for PEHA-activated supports. The best recovery (93%) was reached on
PEHA(0.37)-Cel(ConA) with a low ligand density (7.5 mg [mL.sup.-1]).
[FIGURE 2 OMITTED]
Immobilization of glucoamylase on PEHA-Cel
The characteristics of PEHA-Cel used for the enzyme immobilization
are presented in Table 4.
According to the procedures described in Methods, glucoamylase was
immobilized on activated matrices using glutaraldehyde as an activator.
The procedures of immobilization were selected to apply protein amino
groups during the enzyme-carrier coupling.
The usefulness of PEHA carriers for enzyme immobilization is
characterized by four main parameters: the amount of bound protein,
enzyme activity, immobilization yield, and storage stability. The
results in Table 5 show that the immobilized glucoamylase exhibits high
specific activity. The higher the amount of the anchor groups was, the
better the activity observed. PEHA-Cel activated with glutaraldehyde
seemed to be a good carrier for glucoamylase covalent attachment.
Immobilized glucoamylase has satisfactory stability in the buffer and at
The activation of the polysaccharide-based matrix with
pentaethylenehexamine is a straightforward and convenient method, which
results in active primary amino groups coupled to the matrix via a long
spacer arm. PEHA-activated carriers may be used for the preparation of
biospecific sorbents as well as for enzyme immobilization.
Received 24 October 2005, in revised form 12 December 2005
[1.] Gupta, M. N. Methods for Affinity-Based Separations of Enzymes
and Proteins. Birkhauser, Basel, 2002.
[2.] Affinity Chromatography. Principles and Methods. Amersham
Pharmacia Biotech, 2004.
[3.] West, I. & Goldring, O. Lectin affinity chromatography.
Methods Mol. Biol., 2004, 244, 159-166.
[4.] Turkova, J. Bioaffinity Chromatography. Elsevier, Amsterdam,
[5.] Taylor, R. F. (ed.). Fundamentals and Applications. Marcel
Decker, Inc., New York, 1991.
[6.] Trevan, M. D. (ed.). An Introduction and Applications in
Biotechnology. J. Wiley & Sons, Chichester, 1980.
[7.] Bartling, G. J., Brown, H. D., Forrester, L. J., Koes, M. T.,
Mather, A. N. & Stasiw, R. O. A study of the mechanism of cyanogens
bromide activation of cellulose. Biotechnol. Bioeng., 1972, 14,
[8.] Matejtschuk, P. Affinity Separations: A Practical Approach.
IRL Press, Oxford, 2002.
[9.] Gustavsson, P. E., Mosbach, K. & Nilsson, K. Superporous
agarose as an affinity chromatography support. J. Chromatogr. A., 1997,
[10.] Comfort, A. R., Mutton, C. J. & Langer, R. The influence
of bond chemistry on immobilized enzyme systems for ex vivo use.
Biotechnol. Bioeng., 2004, 32, 554-563.
[11.] Maruska, A., Liesiene, J. & Serys, A. Lithuanian patent.
[12.] Bryjak, J. Glucoamylase, a-amylase and [beta]-amylase
immobilisation on acrylic carriers. J. Biochem. Eng., 2003, 16, 347-355.
[13.] Liesiene, J., Racaityte, K., Morkeviciene, M., Valancius, P.
& Bumelis, V. Immobilized metal affinity chromatography of human
growth hormone: effect of ligand density. J. Chromatogr. A, 1997, 764,
[14.] Safaric, I. & Safarikova, M. Black substrate for
spectrophotometric determination of cellulose activity in coloured
solutions. J. Biochem. Biophys. Methods, 1991, 23, 301-306.
[15.] Cartilier, L. & Chebli, Ch. Cross-linked cellulose as a
tablet excipient. United States Patent. 5989589, 1997.
Jolita Aniulyte (a), Jolanta Bryjak (b), and Jolanta Liesiene (a,c)
(a) Department of Chemical Technology, Kaunas University of
Technology, Radvilenu pl. 19, LT-50254 Kaunas, Lithuania
(b) Department of Chemistry, Wroclaw University of Technology,
Wybrzeze Wyspianskiego 27, 50-373 Wroclaw, Poland
(c) Department of Chemistry, Radom Technical University, Chrobrego
27, 26-600 Radom, Poland
* Corresponding author, firstname.lastname@example.org
Table 1. Effect of the ratio EClH/Cel in the reaction mixture on the
epoxy groups density in cellulose
EClH/Cel, mol [g.sup.-1] Density of epoxy groups, [micro]mol
Table 2. Effect of the epoxy groups density on the content of primary
Nitrogen of primary
Density of epoxy groups, Total nitrogen, amino groups,
[micro]mol [g.sup.-1] % %
144 0.43 0.09
375 1.19 0.46
405 2.59 1.81
490 4.63 2.11
PEHA/Cel = 2 mL [g.sup.-1].
Table 3. Chromatographic performance of the lectin affinity adsorbents
Adsorbent Ligand density, Glycoprotein sorption
mg [mL.sup.-1] capacity
mg [mL.sup.-1] %, of max *
PEHA(1.1)-Cel(WGA) 18.5 2.6 10.5
OXY-Cel(WGA) 17.5 1.3 8.7
PEHA(0.37)-Cel(ConA) 7.5 7.4 62.9
OXY-Cel(ConA) 9.0 3.1 22.0
* Theoretical maximum sorption capacity was calculated from the ligand
density considering one binding site per ligand molecule.
Table 4. Preparation and characterization of PEHA-Cel
Carrier Reaction mixture
Cellulose, PEHA, EClH, 4% NaOH,
g mL mL mL
PEHA(1.1)-Cel 50 3.4 5.1 62
PEHA(0.5)-Cel 50 1.7 2.6 31
Carrier Nitrogen, %
PEHA(1.1)-Cel 1.1 0.3
PEHA(0.5)-Cel 0.5 0.2
Table 5. Immobilization of glucoamylase on cellulose carriers
Carrier Bound Activity, Immobilization
protein, U [mL.sup.-1] yield
mg [mL.sup.-1] (protein),
PEHA(1.1)-Cel 4.8 14.2 11.3
PEHA(0.5)-Cel 0.6 5.3 1.4
Carrier Immobilization Activity after
yield 1 month storage,
(activity), U [mL.sup.-1]
PEHA(1.1)-Cel 0.402 12.1
PEHA(0.5)-Cel 0.150 5.6