Chapter 15

Membranes of Cellulose Derivatives as Supports for Immobilization of Enzymes

R. Lagoa1, D. Murtinho2, and M. H. Gil1

1Department of Chemical Engineering, University of Coimbra, 3000 Coimbra, Portugal

2Department of Chemistry, University of Coimbra, 3000 Coimbra, Portugal

Membranes were prepared from cellulosic derivatives (cellulose acetate, cellulose propionate and cellulose acetate butyrate). Their chemical and physical properties were determined by DSC, water vapor sorption and contact angle evaluation. Catalase, alcohol oxidase and glucose oxidase were covalently linked to these membranes and catalytic activity and stability were examined. The activity results of the immobilization and the stability of the coupled enzymes were found to correlate well with the studied properties of the supports. Cellulose acetate membranes yielded the most active conjugate of support and enzyme. Highly hydrophobic membranes from propionate and butyrate esters of cellulose yielded lower activities, but better storage stability.

Immobilized enzymes, proteins and cells offer advantages because they can be easily handled and recovered; i.e., easily removed from reaction mixtures and repeatedly used in continuous processes. In this way, expensive enzymes can be economically recycled for industrial and clinical applications. To be of value, enzymes attached to a support must retain some of their initial activity. The activity retained by the immobilized enzyme is dependent on a large number of parameters, including the coupling method, the enzyme used and the nature of the support.

Due to their good chemical and physical properties, cellulosic materials have been widely used as supports for enzyme immobilization (7,2). Examples include membranes for hemodialysis (5), for reverse osmosis (4), for microfiltration (5), in biosensors (/) and as chromatographic supports (<5).

In our group we have been interested in suitable supports for the fixation of various biological compounds. We have used polyethylene (7), cellulose and pectin (#), and agar (9) based graft copolymers, as well as cellulosic materials for the immobilization of enzymes, proteins and cells. Simultaneously, we have studied the possibility of applying these systems to industry and in medicine.

228 © 1999 American Chemical Society

Dow

nloa

ded

by U

NIV

OF

RO

CH

EST

ER

on

Aug

ust 2

8, 2

013

| http

://pu

bs.a

cs.o

rg

Pub

licat

ion

Dat

e: M

arch

25,

199

9 | d

oi: 1

0.10

21/b

k-19

99-0

723.

ch01

5

In Biopolymers; Imam, S., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1999.

229

In this paper, we report the immobilization of enzymes of importance in biosensor technology onto well characterized membranes derived from cellulose derivatives.

Experimental

Materials. Cellulose propionate (CP) and acetate-butyrate (CAB) were purchased from Aldrich (Dorset, UK) and cellulose acetate (CA) from Sigma (Dorset, UK). Glucose oxidase (p-D-glucose:oxygen 1-oxireductase; EC 1.1.3.4) from Aspergillus niger, catalase (H 20 2:H 20 2 oxireductase; EC 1.11.1.6) from bovine liver and alcohol oxidase (alcohol:oxygen oxireductase; EC 1.1.3.13) from Candida boidinii were supplied by Sigma. Sodium periodate, hexamethylene diamine and glutaraldehyde were obtained from Merck (Darmstadt, Germany). All other reagents were of analytical grade.

Preparation of the Membranes. Solutions of cellulose derivatives in tetrahydrofuran (10% w/v) were cast on a glass plate (20x20 cm), with the aid of a casting knife (0.33 mm). The solvent was evaporated at room temperature.

Determination of Water Vapor Sorption. Prepared membranes were conditioned over a saturated solution of copper sulfate (98% R.H.) at 25°C., until constant weight was achieved. Afterwards, the polymer was, in each instance, weighed every minute for ten minutes. The initial sorption capacity was obtained graphically after extrapolation to zero time. Each sample was dried to constant mass, under reduced pressure, at 100°C. The percentage of water uptake is given by:

% sorption = (Mt - M f)/ M f x 100

where M { is the initial mass at zero time and M f is the final dry mass.

Determination of Static Contact Angles. Static contact angles were determined with a Contact-6-Meter (Livereel, Durham, England). A sessile drop of water was placed on the surface of the membranes and the contact angle was determined at 20°C.

Characterization of the Membranes by DSC. Differential scanning calorimetric analyses were carried out on the membranes using a Polymer Laboratories, PL-DSC analyzer. Samples of membranes, weighing 5-6 mg, were sealed in aluminum pans. The samples were heated at a rate of 10°C min"1 in a N 2 gas purged atmosphere, using a flow rate of 10 cm3 min1.

Enzyme Immobilization. Membrane sections were allowed to react with 10 cm3

sodium periodate (0.5 M) for 2 hr in the dark. They were then treated with 10 cm3 of a 1% hexamethylene diamine (w/w) solution for 18 hr. This step was followed by immersion for 2 hr at 4 °C in 10 cm3 of 5% glutaraldehyde (v/v) in 0.05 M phosphate

Dow

nloa

ded

by U

NIV

OF

RO

CH

EST

ER

on

Aug

ust 2

8, 2

013

| http

://pu

bs.a

cs.o

rg

Pub

licat

ion

Dat

e: M

arch

25,

199

9 | d

oi: 1

0.10

21/b

k-19

99-0

723.

ch01

5

In Biopolymers; Imam, S., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1999.

230

buffer at pH 7.5. Membranes were thoroughly washed with distilled water between each step (10). Enzyme solutions (glucose oxidase, 9 cm3, 2 mg/cm3; alcohol oxidase, 15 cm3, 4 mg/cm3; and catalase 15 cm3, 2 mg/cm3, all in 0.05 M phosphate buffer, pH 7.5) were allowed to couple to activated membranes over a period of 20 hr at 4°C. The enzyme-coupled membranes were thoroughly washed with 0.05 M phosphate buffer, pH 7.5, and stored in the same buffer at 4°C until activity assays were performed.

Enzyme Activity Assays. The activity of immobilized glucose oxidase (GOX) and free glucose oxidase was determined by the 0-dianisidine procedure (11) (1 IU = 1 mmol D-glucose oxidized per minute). Free enzyme solution (0.05 cm3) was added to a buffered reaction mixture (50 mM acetate, pH 5.1) that consisted of 2.4 cm3 of 0.21 mM o-dianisidine, 0.5 cm3 of 10% D-glucose and 0.1 cm3 of 60 U/cm 3peroxidase. In the case of the immobilized enzyme, because the chromogen adsorbs onto the membranes, the process was separated into two steps. A weighed membrane portion containing immobilized glucose oxidase was immersed in 5 cm3 D-glucose (10% solution in 50 mM acetate buffer, pH 5.1). Then, 0.1 cm3 aliquots were withdrawn from the reaction vessel and reacted with 2.4 cm3 of 0.21 mM 0-dianisidine and 0.1 cm 3

60 U/cm3 peroxidase. The activity of free alcohol oxidase (AOD) and of immobilized alcohol oxidase

was determined by the 4-aminoantipirine procedure (72), using 50 mM phosphate buffer and methanol as the substrate (1 IU = 1 mmol methanol oxidized per minute). Free alcohol oxidase (0.1 cm3) was added to 1 cm3 of 6 U/cm3 peroxidase and 1.9 cm3

of an indicator solution that consisted of 0.4 mM 4-aminoantipirine, 11 mM phenol and 1.5 M methanol. In the case of immobilized enzyme, weighed portions of membranes were immersed in 3 cm3 of peroxidase solution and 6 cm3 of indicator solution.

The activity of immobilized catalase was determined by following the decomposition of H 2 0 2 at 240 nm (1 IU = 1 mmol of H 2 0 2 degraded per minute). A weighed portion of membrane was immersed in 15 cm3 of 100 mM H 2 0 2 in 50 mM phosphate buffer, pH 7.0.

Storage Stability of Free and Immobilized Glucose Oxidase. The storage stability of glucose oxidase was assessed. A 0.2 mg/cm3 of free enzyme solution, in 50 mM acetate buffer, pH 5.1, as well as membranes containing immobilized enzyme in the same buffer, were stored at 4°C for 3 days. Residual activity was then determined.

Results and Discussion

Membranes from cellulose acetate, cellulose propionate and cellulose butyrate were characterized by determination of water vapor sorption, thermal properties and static contact angle. The obtained results are presented in Table I. Among the tested materials, cellulose acetate has the highest water sorption capacity and the lowest contact angle. This can be explained by an increase in hydrophobic character due to the alkyl groups of propionate. Cellulose acetate butyrate presents two considerations:

Dow

nloa

ded

by U

NIV

OF

RO

CH

EST

ER

on

Aug

ust 2

8, 2

013

| http

://pu

bs.a

cs.o

rg

Pub

licat

ion

Dat

e: M

arch

25,

199

9 | d

oi: 1

0.10

21/b

k-19

99-0

723.

ch01

5

In Biopolymers; Imam, S., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1999.

231

a) the increased alkyl chain length of butyrate groups and b) the presence of two substituent groups.

Some of the thermal characteristics of the various cellulosic ester membranes are also given in Table I. All the membranes have a Tg. Cellulose acetate has the highest Tg, probably because this polymer is more ordered than cellulose propionate or cellulose acetate butyrate. Cellulose propionate and cellulose acetate butyrate are amorphous membranes.

Table I - Physical Characteristics of the Cellulosic Membranes Water Sorption Contact

Capacity Angle AH Tm Membrane (%) n (°C) (cal/g) (°C) Cellulose Acetate 13.6±0.5 73±3 191.2 0.49 210.6

Cellulose Propionate 7.7±0.1 78±3 152.2 - -Cellulose Acetate 4.6±0.4 79±3 173.8 6.23 -butyrate

These tliree membranes were used to covalently link catalase, glucose oxidase and alcohol oxidase. The enzymes were covalently linked to the supports by use of sodium periodate as reported by Gil et al (10). Relative activities are shown in Figure 1. The data in the graph suggest that the behavior of the enzyme-linked-membrane is not dependent on the enzyme, but rather on the support and that cellulose acetate membranes better couple the tested enzymes than do cellulose propionate or cellulose acetate butyrate.

These results can be related to the physical characteristics indicated in Table I. More hydrophilic membranes allow a better yield of coupled activity due to a better interaction between the enzyme solution and the support. On the other hand, the access of the enzyme to cellulose propionate and cellulose acetate butyrate membranes may be more difficult due to steric problems, as well as due to the higher substitution of the hydroxyl groups. Apparently, enzyme activities are strongly dependent upon the characteristics of the support.

The stability of GOX was investigated by monitoring the activity after 3 days of storage at 4°C (Table II). In comparison with the initial activities, supports which yield greater activity are those that offer less stabilization to the enzyme. Thus, enzymes immobilized onto CA membranes are less stable than those linked to CP and CAB membranes. This implies that a hydrophobic microenvironment reduces enzyme hydration and also contact with the aqueous solution, thus diminishing conformational mobility which leads to denaturation.

Dow

nloa

ded

by U

NIV

OF

RO

CH

EST

ER

on

Aug

ust 2

8, 2

013

| http

://pu

bs.a

cs.o

rg

Pub

licat

ion

Dat

e: M

arch

25,

199

9 | d

oi: 1

0.10

21/b

k-19

99-0

723.

ch01

5

In Biopolymers; Imam, S., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1999.

232

Figure 1. Relative activity of various enzymes covalently linked to various cellulosic supports.

Dow

nloa

ded

by U

NIV

OF

RO

CH

EST

ER

on

Aug

ust 2

8, 2

013

| http

://pu

bs.a

cs.o

rg

Pub

licat

ion

Dat

e: M

arch

25,

199

9 | d

oi: 1

0.10

21/b

k-19

99-0

723.

ch01

5

In Biopolymers; Imam, S., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1999.

233

Table II. Storage Stability of Free (IU / g Protein) and Immobilized (IU / g Membrane) Glucose Oxidase

DayO Day 3

Membrane Activity au/g)

Activity au/g)

Activity Retention (%)

Free enzyme 0.144 21.4 xlO"3 14

Cellulose acetate 1.9 266 xlO"3 14

Cellulose propionate 0.3 236x10 3 79

Cellulose acetate butyrate 0.7 239x10° 34

The stability of CP-immobilized GOX is distinctly greater as a result of specific interactions between the enzyme and support. Alternatively, processes which could lead to denaturation may have become diffusion limited. Immobilized biocatalytic systems, in which activity is limited by diffusion of the reactants, can have an apparent stability which is greater than the stability of the free enzyme molecules.

Conclusions

Characteristics, such as crystalinity, hydrophilicity and permeability, can be correlated with the activity and stability of enzymes immobilized onto membranes. Membranes of cellulose derivatives offer a good model for the study of the factors which influence the behavior of immobilized enzyme systems. Data from this model indicate that fusion enthalpies and water sorption values are useful in establishing practical parameters for the creation of effective supports for enzyme immobilization. Contact angle phenomena are of limited importance.

Acknowledgments

The authors gratefully acknowledge the financial support provided by JNICT (Portugal) for R. Lagoa.

Literature Cited

1. Gil, M. H.; Alegret, S.; Alves da Silva, M.; Alegria, A. C.; Piedade, A. P. In Cellulosics: Materials for Selective Separations and Other Technologies; Kennedy, J. F.; Philips, G. O.; William, P. A., Eds.; Ellis Horwood Ltd.: New York, NY, 1993; pp. 163-171.

2. Enzymes in Industry and Medicine; Bickerstaff, G. F. Ed.; Whitstable Litho: Whistable, Kent, 1987; 1st Ed.

Dow

nloa

ded

by U

NIV

OF

RO

CH

EST

ER

on

Aug

ust 2

8, 2

013

| http

://pu

bs.a

cs.o

rg

Pub

licat

ion

Dat

e: M

arch

25,

199

9 | d

oi: 1

0.10

21/b

k-19

99-0

723.

ch01

5

In Biopolymers; Imam, S., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1999.

234

3. Ohno, M.; Suzuki, M; Miyagi, M.; Yagi, T.; Sakurai, H.; Ukai, T. In Cellulosics: Chemical, Biochemical and Material Aspects; Kennedy, J. F.; Philips, G. O.; William, P. A., Eds.; Ellis Horwood Ltd.: New York, NY, 1993; pp. 415-420.

4. Koros, W. J.; Fleming, G. K.; Jordon, S. M.; Kim T. H.; Hoehn, H. H. Prog. Polym. Sci. 1988, 13, 339-340.

5. Manabe, S.; Kamata, Y.; Iijima, H.; Kamide, K., Polym. J. 1987, 19, 391-404. 6. Enzyme Technology, Chaplin, M. F.; Bucke, C., Eds.; Cambridge University

Press: 1990; 1st Ed., Chapter 2. 7. Ramos, M. C.; Gil, M. H.; Garcia, A. P.; Cabral, J. M. S.; Guthrie, J. T.

Biocatalysis 1992, 6, 223-234. 8. Beddows, C. G.; Gil, M. H.; Guthrie, J. T. Polymer Bulletin 1984, 11, 1-6. 9. Alves da Silva, M.; Beddows, C. G.; Gil, M. H.; Guthrie, J. T.; Guiomar, A. J.;

Rotov, S.; Piedade, A. P. Radiat. Phys. Chem. 1990, 35, 98-101. 10. Gil, M. H.; Piedade, A. P.; Alegret, S.; Orellana, A. Bios. and Biol. 1992, 7, 645-652.

11. Huggett, A. G.; Nixon, D. A. Lancet 1957, 2, 368. 12. Biosensors - A Practical Approach; Cass, A. E. G., Ed.; IRL Press-Oxford

University Press: Oxford, 1990; pp. 117.

Dow

nloa

ded

by U

NIV

OF

RO

CH

EST

ER

on

Aug

ust 2

8, 2

013

| http

://pu

bs.a

cs.o

rg

Pub

licat

ion

Dat

e: M

arch

25,

199

9 | d

oi: 1

0.10

21/b

k-19

99-0

723.

ch01

5

In Biopolymers; Imam, S., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1999.