BIOTECHNOLOGY AND BIOENGINEERING, VOL. XIX, PAGES 301-309 (1977)
Bacteriolysis by Immobilized Enzymes ISAO KARUBE, TOSHIRO SUGANUMA, and SHUICHI SUZUKI, Research Laboratory of Resources Utilization, Tokyo Institute of Technology, Ookayama, Meguro-ku, Tokyo, J a p a n
Summary Bacteriolytic enzymes produced by Achromobacter lunatus were immobilized in collagen membrane. Intact bacteria such as Pseudommas solanacearum, Xanthomonas oryzae, Staphylococcus aureus, and Pseudomonas aeruginosa were lyzed with the bacteriolytic enzyme-collagen membrane. Relative activity of the bacteriolytic enzyme-collagen membrane against Pseu. solanacearum was about 2% of that of native bacteriolytic enzymes. No difference in the optimum pH was observed between immobilized enzymes and native enzymes. The bacteriolytic enzymes in the collagen membrane were stable in the p H range from 4 to 11. The enzyme-collagen membrane was stable against sodium chloride which was an inhibitor of the native bacteriolytic enzymes. Xanthomonas oryzae and Pseu. aeruginosa were continuously lyzed by a reactor containing the rolled bacteriolytic enzyme-collagen membrane.
INTRODUCTION A large number of enzymes capable of lyaing bacteria have been found.' The peptidoglycan which contributes to the rigidity of the bacterial cell wall has been lyaed by the action of multienzymes such as protease and glucanase. These bacteriolytic enzymes have mainly been used for studies of the structure and the composition of the bacterial cell walls.2 It has been found that intact bacteria are lyzed by bacteriolytic enzymes produced by Achromobacter l u n a t ~ s . These ~ bacteriolytic enzymes can therefore be used to sterilize bacteria. Recently, many immobilized enzymes have been prepared and used for various practical as well as experimental purpose^.^^^ It is difficult to couple or adsorb multienzymes on the same carrier at the same time. On the other hand, the entrapping method proves to be suitable for immobilization of multienzymes. An effective method for preparation of enzyme-collagen membranes without loss of enzymatic activity has been developed in this laboratory.6 301
@ 1977 by John Wiley & Sons, Inc.
302
KARUBE, SUGANUMA, AND SUZUKI
This method for immobilizing enzymes is advantageous in that it permits stable and continuous use of enzymes. I n the present studies, bacteriolytic enzymes were immobilized in the collagen membrane. Properties of the bacteriolytic enzymecollagen membrane and its use in continuous lysis of bacteria were examined.
MATERIALS AND METHODS Materials Peptone and meat extract were purchased from Kyokuto Pharmaceutical Co. Collagen was purified by the method previously reported.' Bacteriolytic enzymes were obtained from Amano Pharmaceutical Co. (Ach. Zunatas Type YL, 16,000 units/g). Other solvents and reagents were commercially available analytical reagents or laboratory grade materials. Deionized water was used in all procedures.
Culture of Microorganisms Pseudomonas solanacearum 6501 (Op type) Xanthomonas oryzae 44, Staphylococcus aureus 209P1 and Pseudomonus aeruginosa ATCC 7933 were used in these studies. Pseudomonas solanacearum and X a n . oryzae were cultured under aerobic condition at 30°C for 48 hr in 50 ml of t a p water (pH 7.0) containing 1% glucose and potato extract. Pseudomonas aeruginosa and Staph. aureus were cultured in glucose-nutrient broth (pH 7.0) a t 30°C for 48 hr. Preparation of Enzyme-Collagen Membrane Purified collagen fibril suspension was prepared by the method previously reported. Thirty-three g of 0.6% collagen fibril suspension (pH 3.0) and 5.3 ml of bacteriolytic enzyme solution (30 mg enzyme/ml) were mixed and casted on a Teflon plate. The casted enzyme-collagen membrane was dried at room temperature for 24 hr. Then, the enzyme-collagen membrane was suspended in 0.1yo glutaraldehyde solution (pH 8.0) for 1 min and washed sufficiently with 0.1M phosphate buffer solution (pH 7.0). The enzyme content of the collagen membrane was 14%.
Enzyme Assay Approximately 20 mg of enzyme-collagen membrane (2 X 2 cmz) or 20 pg of enzymes were added to 10 ml of bacterial suspension (0.1M phosphate buffer, 0.2 mg cells/ml) and incubated for 1 hr at 37°C.
BACTERIOLYSIS BY IMMOBILIZED ENZYMES
303
Fig. 1. Block diagram of system for continuous lysis of bacteria. 1, Bacteria reservoir; 2, magnetic stirrer; 3, peristaltic pump; 4, reactor; 5, water bath.
Activity was determined from the turbidity decrease (at 660 nm) of the reaction mixture or from the ribonucleic acid (RNA) increase. RNA was determined by the method of Mejbaum."
Experimental Apparatus for Continuous Lysis of Bacteria The enzyme-collagen membrane (4 X 30 cmz) was rolled with plastic net and inserted into a biocatalytic reactors (4 1.8 X 5 cmz, acrylic plastic). The system used for continuous lysis of bacteria is diagrammatically illustrated in Figure 1. The bacterial suspension (20 ml containing 4 mg bacterial cells and 0.5-1.5mM sodium lauryl sulfate) was circulated with a peristaltic pump a t the flow rate of 1 ml per min. The reactor was held in a water bath a t 37°C.
RESULTS The bacteriolytic enzymes (produced by Ach. lunatas) were protease and fi-1,3-gl~canase.~ These enzymes were immobilized in the collagen membrane. The enzyme content of the collagen membrane was 14%. Table I TABLE I Lysis of Bacteria with the Ensyme-Collagen Membrane Bacteria
Pseu. solanacearum 6501 (Op type) Xan. oryzae 44 Staph. aureus 209 P Pseu. aeruginosa ATCC 7933
Lysis
++ +
+++ ++ +
5 ml of bacteria solution (0.1M phosphate buffer, pH 7.0) containing 1 mg of cells.
304
KARUBE, SUGANUMA, AND SUZUKI 100 0
-
0 I7
-0.2z
3
. 3
v
- 01 E,
shows the lysis of bacteria by the enzyme-collagen membrane. As shown, Pseu. solanacearum, Xan. oryzae, and Staph. aureus were lyzed by the enzyme-collagen membrane. Pseudomonas aeruginosa was the one least lyzed by the enzyme-collagen membrane. However, it was lyzed to a high degree when a detergent such as sodium lauryl sulfate, poly(oxyethy1ene) lauryl ether, lauryl trimethyl ammonium chloride, or poly(oxyethy1ene) sorbitan mono-oleate was added to the reaction mixture. The relative activity of the enzyme-collagen membrane determined in a test tube against Pseu. solanacearum was 2%. Figure 2 shows the time courses of Pseu. solanacearum lysis by the enzyme-collagen membrane. Lysis of bacteria was determined from the turbidity decrease and the ribonucleic acid (RNA) increase of the reaction mixture. As the lysis curve determined from turbidity was almost the same shape as that determined from the increase of RNA concentration of the reaction mixture, the turbidity determination a t 660 nm was used in ensuing determinations of bacteriolysis. As shown in Figure 2, Pseu. solanacearum was almost completely lyzed within 30 niin. Figure 3 shows pH-activity profiles obtained for the enzymecollagen membrane and native bacteriolytic enzymes when Pseu. solanacearum was used for the experiments. No difference in the pH-activity profiles was observed between the enzyme-collagen membrane and native enzymes. Figure 4 shows the pH stability obtained for the enzyme-collagen membrane when Pseu. solanacearum suspension was used for the substrate. As shown, the enzyme-collagen membrane was stable over a pH range from 4 t o 11.
BACTERIOLYSIS BY IMMOBILIZED ENZYMES
305
Fig. 3. pH-Activity profiles of the enzyme-collagen membrane and native enzymes. Reaction was carried out under the standard conditions except for buffer solution employed. Glycine buffer (pH 2.0), acetate buffer (pH 4-5), phosphate buffer (pH 6-7), Tris buffer (pH 8-9), and glycine buffer (pH 10-11) were employed, respectively, I = 0.1. (0) Enzyme-collagen membrane; ( 0 ) native enzyme.
Thermostability of the enzyme-collagen membrane was examined with Pseu. solanacearum suspension. The enzyme-collagen membrane was stable below 50°C. Figure 5 shows the effect of sodium chloride on the activity of the enzyme-collagen membrane and native enzymes. It is known that sodium chloride is an inhibitor of these bacteriolytic enzymes3 The activity of native enzymes in 0.2M sodium chloride solution decreased to 50% of that in sodium chloride free solution. On the contrary, the enzyme-collagen membrane was stable against sodium
1.4.
0 2
,
4
,
,
,
6 PH
,
0
,
,
,
10
Fig. 4. pH Stability of the enzyme-collagen membrane. Enzyme-collagen membrane was suspended in the buffers mentioned in Fig. 3, for 30 min a t 37°C. Enzyme assays were carried out under the standard conditions.
306
KARUBE, SUGANUMA, AND SUZUKI
0
01
02
NaCl ( M )
Fig. 5. Effect of sodium chloride on the activity of the enzyme-collagen membrane and native enzymes. Reaction was carried out under standard conditions at various sodium chloride concentrations. (0) Enzyme-collagen membrane; ( 0 )native enzymes.
chloride. Ninety percent of the bacteriolytic activity was retained in 0.2M sodium chloride solution. Furthermore, the activity of the enzyme-collagen membrane was increased with sodium chloride below the concentration of 0.1M. Figure 6 shows the continuous lysis of X a n . oryzae by the enzymecollagen membrane. As shown, approximately 50% of X a n . oryzae was lyzed in 15 min. However, the lysis rate of X a n . oryzae decreased after 15 min. I n this experiment, lysis of X a n . oryzae was performed in the presence of 1.5mM sodium lauryl sulfate. On the other hand, the lysis of Pseu. aeruginosa was also examined
loo
r--3 Time( min)
Fig. 6. Continuous lysis of Xan. oryzae. 20 ml of bacterial suspension containing 4 mg of bacterial cells and 1.5mM sodium lauryl sulfate. Flow rate: I ml/min, 37OC.
BACTERIOLYSIS BY IMMOBILIZED ENZYMES
307
using the same biocatalytic reactor. Approximately 80% of the Pseu. aeruginosa was lyzed by the enzyme-collagen membrane in the presence of 0.5mM sodium lauryl sulfate.
DISCUSSION A large number of enzymes capable of lyzing certain gram-positive bacteria have been reported.l0 These enzymes hydrolyze the peptidoglycan which contributes to the rigidity of the bacterial cell wall. On the other hand, cell walls of gram-negative bacteria are complex both in morphology and in chemical composition. The outer layer contains a complex mixture of lipid, protein, and lipopolysaccharide which envelops the inner layer of peptidoglycan. Therefore, gram-negative bacteria are difficult t o lyze by use of bacteriolytic enzymes.'O Pseudomonas solanacearum, X a n . oryzae, and Pseu. aeruginosa, which were used in this study, are gramnegative bacteria. The bacteriolytic enzymes (produced by Ach. lunatas) are capable of lyzing intact bacteria. I n the present study, these bacteriolytic enzymes were immobilized in collagen membrane. As the enzyme-collagen membrane was treated with glutaraldehyde solution, a part of the enzymes near the surface of the collagen membrane were crosslinked to the collagen fibril. Bacteria such as Pseu. solanacearum, Xan. oryzae, and Pseu. aeruginosa were lyzed by the enzyme-collagen membrane. The activity of the enzymecollagen membrane was 2% that of the native enzymes. These bacteriolytic enzymes contain protease. Therefore, the activity of the enzyme-collagen membrane was determined with 1% milk casein solution. The activity of the enzyme-collagen membrane against casein was 10% that of the native enzymes. This activity of the bacteriolytic enzyme-collagen membrane was almost the same as that of other enzyme-collagen membranes such as the alcohol dehydrogenase-collagen membrane and the lipase-collagen membrane." Therefore, this low activity may be caused by the difficulty of contact between the immobilized enzymes and the bacterial cell walls. The size of the bacteria used in this study was about 0.5-2.0 pm. From diffusion experiments using proteinslothe size of the meshes of collagen fibril matrix is about 300 X 300 A.2 It is difficult for bacteria to penetrate the collagen membrane. Therefore, only bacteriolytic enzymes on the surface of the collagen membrane can react with the bacterial cell walls. Pseudomonas aeruginosa was lyzed in the presence of a detergent such as sodium lauryl sulfate. As mentioned above, the outer layer of gram-negative bacteria is
308
KARUBE, SUGANUMA, AND SUZUKI
composed of lipid, protein, and lipopolysaccharide. This outer layer can be removed with detergent treatment.12 Therefore, Pseu. aeruginosa could be lyzed by the enzyme-collagen membrane after the complex outer layer of cell walls was removed with sodium lauryl sulfate treatment. The immobilized bacteriolytic enzymes showed stability against acidic and alkaline conditions. As previously reported,11 immobilized enzymes in collagen fibril network are stabilized against acid and alkali. The mechanism of the stabilization may be the same as that in othrr enzyme-collagen membranes. As mentioned above, inhibition of the native bacteriolytic enzyme with sodium chloride was observed. On the other hand, the enzyme-collagen membrane was stable against sodium chloride. Furthermore, it was activated with sodium chloride (-0.lM). It is well known that collagen fibrils swell in a sodium chloride s01ution.l~ Therefore, the swelling of collagen fibrils may increase the surface area of the membrane. The activity of the enzyme-collagen membrane was increased with increase in sodium chloride concentration (-0.1M). Then, the equilibrium of swelling was attained, and the excess sodium chloride inhibited the immobilized enzymes in the membrane. Therefore, the activity of the enzyme-collagen membrane decreased with increase in sodium chloride concentration beyond a certain level. Bacteria such as Pseu. aeruginosa and Pseu. solanacearum were continuously lyzed by a reactor containing rolled enzyme-collagen membrane. As the bacterial content of the suspension was high, the bacterial suspension was circulated through the reactor. I n conclusion, this study suggests the possibility of a new type of germ sterilization. Further developmental studies in this laboratory are being directed toward using the enzyme-collagen membrane reactor for the sterilization of air. The authors would like to thank Dr. S. Tsuru for his many valuable suggestions and for supplying bacteria.
References 1. J. L. Strominger and J. M. Chuysen, Science, 156, 213 (1967). 2. J. W. Costerton, J. M. Ingram, and K. J. Cheng, Bacteriol. Rev., 38, 87 (1974). 3. S. Tsumura, Abstract of dlst Annual Meeting of Society of Food Industry 18, (1974). 4. K. E. Pye and L. B. Wingard, Jr., Eds., Enzyme Engineering 2, Plenum Press, New York, New York, 1974. 5 . M. Aizawa, I. Karube, and S. Suxuki, Anal. Chim. Acta, 69, 431 (1974).
BACTERIOLYSIS BY IMMOBILIZED ENZYMES
309
6. S. Suzuki, M. Aizawa, and I. Karube, Immobilized Enzyme Technology, H. H. Weetall and S. Suzuki, Eds., Plenum Press, New York, 1975,p. 253. 7. I. Karube, S. Suzuki, S. Kinoshita, and J. Mizuguchi, Ind. Eng. Chem. Prod. Res. Develop., 10, 160 (1971). 8. W. Mejbaum, 2.Physiol. Chem., 258, 117 (1939). 9. K. Venkatasubramanian and W. R. Vieth, Biotechnol. Bioeng., 15, 583 (1973). 10. S. Murao and Y. Takahara, Agr. Biol. Chem., 38, 2305 (1974). 11. S.Suzuki, I. Karube, and I. Satoh, Biomedical Application of Immobilized Enaymes and Proteins, T. M. S. Chang, Ed., Plenum Press, New York, 1976. 12. C. A. Schnaitman, J . Bacteriol., 108, 553 (1971). 13. A. Veis, The Macromolecular Chemistry of Gelatin, Academic Press, New York, 1964, p. 29.
Accepted for Publication September 22, 1976
Bacteriolysis by Immobilized Enzymes ISAO KARUBE, TOSHIRO SUGANUMA, and SHUICHI SUZUKI, Research Laboratory of Resources Utilization, Tokyo Institute of Technology, Ookayama, Meguro-ku, Tokyo, J a p a n
Summary Bacteriolytic enzymes produced by Achromobacter lunatus were immobilized in collagen membrane. Intact bacteria such as Pseudommas solanacearum, Xanthomonas oryzae, Staphylococcus aureus, and Pseudomonas aeruginosa were lyzed with the bacteriolytic enzyme-collagen membrane. Relative activity of the bacteriolytic enzyme-collagen membrane against Pseu. solanacearum was about 2% of that of native bacteriolytic enzymes. No difference in the optimum pH was observed between immobilized enzymes and native enzymes. The bacteriolytic enzymes in the collagen membrane were stable in the p H range from 4 to 11. The enzyme-collagen membrane was stable against sodium chloride which was an inhibitor of the native bacteriolytic enzymes. Xanthomonas oryzae and Pseu. aeruginosa were continuously lyzed by a reactor containing the rolled bacteriolytic enzyme-collagen membrane.
INTRODUCTION A large number of enzymes capable of lyaing bacteria have been found.' The peptidoglycan which contributes to the rigidity of the bacterial cell wall has been lyaed by the action of multienzymes such as protease and glucanase. These bacteriolytic enzymes have mainly been used for studies of the structure and the composition of the bacterial cell walls.2 It has been found that intact bacteria are lyzed by bacteriolytic enzymes produced by Achromobacter l u n a t ~ s . These ~ bacteriolytic enzymes can therefore be used to sterilize bacteria. Recently, many immobilized enzymes have been prepared and used for various practical as well as experimental purpose^.^^^ It is difficult to couple or adsorb multienzymes on the same carrier at the same time. On the other hand, the entrapping method proves to be suitable for immobilization of multienzymes. An effective method for preparation of enzyme-collagen membranes without loss of enzymatic activity has been developed in this laboratory.6 301
@ 1977 by John Wiley & Sons, Inc.
302
KARUBE, SUGANUMA, AND SUZUKI
This method for immobilizing enzymes is advantageous in that it permits stable and continuous use of enzymes. I n the present studies, bacteriolytic enzymes were immobilized in the collagen membrane. Properties of the bacteriolytic enzymecollagen membrane and its use in continuous lysis of bacteria were examined.
MATERIALS AND METHODS Materials Peptone and meat extract were purchased from Kyokuto Pharmaceutical Co. Collagen was purified by the method previously reported.' Bacteriolytic enzymes were obtained from Amano Pharmaceutical Co. (Ach. Zunatas Type YL, 16,000 units/g). Other solvents and reagents were commercially available analytical reagents or laboratory grade materials. Deionized water was used in all procedures.
Culture of Microorganisms Pseudomonas solanacearum 6501 (Op type) Xanthomonas oryzae 44, Staphylococcus aureus 209P1 and Pseudomonus aeruginosa ATCC 7933 were used in these studies. Pseudomonas solanacearum and X a n . oryzae were cultured under aerobic condition at 30°C for 48 hr in 50 ml of t a p water (pH 7.0) containing 1% glucose and potato extract. Pseudomonas aeruginosa and Staph. aureus were cultured in glucose-nutrient broth (pH 7.0) a t 30°C for 48 hr. Preparation of Enzyme-Collagen Membrane Purified collagen fibril suspension was prepared by the method previously reported. Thirty-three g of 0.6% collagen fibril suspension (pH 3.0) and 5.3 ml of bacteriolytic enzyme solution (30 mg enzyme/ml) were mixed and casted on a Teflon plate. The casted enzyme-collagen membrane was dried at room temperature for 24 hr. Then, the enzyme-collagen membrane was suspended in 0.1yo glutaraldehyde solution (pH 8.0) for 1 min and washed sufficiently with 0.1M phosphate buffer solution (pH 7.0). The enzyme content of the collagen membrane was 14%.
Enzyme Assay Approximately 20 mg of enzyme-collagen membrane (2 X 2 cmz) or 20 pg of enzymes were added to 10 ml of bacterial suspension (0.1M phosphate buffer, 0.2 mg cells/ml) and incubated for 1 hr at 37°C.
BACTERIOLYSIS BY IMMOBILIZED ENZYMES
303
Fig. 1. Block diagram of system for continuous lysis of bacteria. 1, Bacteria reservoir; 2, magnetic stirrer; 3, peristaltic pump; 4, reactor; 5, water bath.
Activity was determined from the turbidity decrease (at 660 nm) of the reaction mixture or from the ribonucleic acid (RNA) increase. RNA was determined by the method of Mejbaum."
Experimental Apparatus for Continuous Lysis of Bacteria The enzyme-collagen membrane (4 X 30 cmz) was rolled with plastic net and inserted into a biocatalytic reactors (4 1.8 X 5 cmz, acrylic plastic). The system used for continuous lysis of bacteria is diagrammatically illustrated in Figure 1. The bacterial suspension (20 ml containing 4 mg bacterial cells and 0.5-1.5mM sodium lauryl sulfate) was circulated with a peristaltic pump a t the flow rate of 1 ml per min. The reactor was held in a water bath a t 37°C.
RESULTS The bacteriolytic enzymes (produced by Ach. lunatas) were protease and fi-1,3-gl~canase.~ These enzymes were immobilized in the collagen membrane. The enzyme content of the collagen membrane was 14%. Table I TABLE I Lysis of Bacteria with the Ensyme-Collagen Membrane Bacteria
Pseu. solanacearum 6501 (Op type) Xan. oryzae 44 Staph. aureus 209 P Pseu. aeruginosa ATCC 7933
Lysis
++ +
+++ ++ +
5 ml of bacteria solution (0.1M phosphate buffer, pH 7.0) containing 1 mg of cells.
304
KARUBE, SUGANUMA, AND SUZUKI 100 0
-
0 I7
-0.2z
3
. 3
v
- 01 E,
shows the lysis of bacteria by the enzyme-collagen membrane. As shown, Pseu. solanacearum, Xan. oryzae, and Staph. aureus were lyzed by the enzyme-collagen membrane. Pseudomonas aeruginosa was the one least lyzed by the enzyme-collagen membrane. However, it was lyzed to a high degree when a detergent such as sodium lauryl sulfate, poly(oxyethy1ene) lauryl ether, lauryl trimethyl ammonium chloride, or poly(oxyethy1ene) sorbitan mono-oleate was added to the reaction mixture. The relative activity of the enzyme-collagen membrane determined in a test tube against Pseu. solanacearum was 2%. Figure 2 shows the time courses of Pseu. solanacearum lysis by the enzyme-collagen membrane. Lysis of bacteria was determined from the turbidity decrease and the ribonucleic acid (RNA) increase of the reaction mixture. As the lysis curve determined from turbidity was almost the same shape as that determined from the increase of RNA concentration of the reaction mixture, the turbidity determination a t 660 nm was used in ensuing determinations of bacteriolysis. As shown in Figure 2, Pseu. solanacearum was almost completely lyzed within 30 niin. Figure 3 shows pH-activity profiles obtained for the enzymecollagen membrane and native bacteriolytic enzymes when Pseu. solanacearum was used for the experiments. No difference in the pH-activity profiles was observed between the enzyme-collagen membrane and native enzymes. Figure 4 shows the pH stability obtained for the enzyme-collagen membrane when Pseu. solanacearum suspension was used for the substrate. As shown, the enzyme-collagen membrane was stable over a pH range from 4 t o 11.
BACTERIOLYSIS BY IMMOBILIZED ENZYMES
305
Fig. 3. pH-Activity profiles of the enzyme-collagen membrane and native enzymes. Reaction was carried out under the standard conditions except for buffer solution employed. Glycine buffer (pH 2.0), acetate buffer (pH 4-5), phosphate buffer (pH 6-7), Tris buffer (pH 8-9), and glycine buffer (pH 10-11) were employed, respectively, I = 0.1. (0) Enzyme-collagen membrane; ( 0 ) native enzyme.
Thermostability of the enzyme-collagen membrane was examined with Pseu. solanacearum suspension. The enzyme-collagen membrane was stable below 50°C. Figure 5 shows the effect of sodium chloride on the activity of the enzyme-collagen membrane and native enzymes. It is known that sodium chloride is an inhibitor of these bacteriolytic enzymes3 The activity of native enzymes in 0.2M sodium chloride solution decreased to 50% of that in sodium chloride free solution. On the contrary, the enzyme-collagen membrane was stable against sodium
1.4.
0 2
,
4
,
,
,
6 PH
,
0
,
,
,
10
Fig. 4. pH Stability of the enzyme-collagen membrane. Enzyme-collagen membrane was suspended in the buffers mentioned in Fig. 3, for 30 min a t 37°C. Enzyme assays were carried out under the standard conditions.
306
KARUBE, SUGANUMA, AND SUZUKI
0
01
02
NaCl ( M )
Fig. 5. Effect of sodium chloride on the activity of the enzyme-collagen membrane and native enzymes. Reaction was carried out under standard conditions at various sodium chloride concentrations. (0) Enzyme-collagen membrane; ( 0 )native enzymes.
chloride. Ninety percent of the bacteriolytic activity was retained in 0.2M sodium chloride solution. Furthermore, the activity of the enzyme-collagen membrane was increased with sodium chloride below the concentration of 0.1M. Figure 6 shows the continuous lysis of X a n . oryzae by the enzymecollagen membrane. As shown, approximately 50% of X a n . oryzae was lyzed in 15 min. However, the lysis rate of X a n . oryzae decreased after 15 min. I n this experiment, lysis of X a n . oryzae was performed in the presence of 1.5mM sodium lauryl sulfate. On the other hand, the lysis of Pseu. aeruginosa was also examined
loo
r--3 Time( min)
Fig. 6. Continuous lysis of Xan. oryzae. 20 ml of bacterial suspension containing 4 mg of bacterial cells and 1.5mM sodium lauryl sulfate. Flow rate: I ml/min, 37OC.
BACTERIOLYSIS BY IMMOBILIZED ENZYMES
307
using the same biocatalytic reactor. Approximately 80% of the Pseu. aeruginosa was lyzed by the enzyme-collagen membrane in the presence of 0.5mM sodium lauryl sulfate.
DISCUSSION A large number of enzymes capable of lyzing certain gram-positive bacteria have been reported.l0 These enzymes hydrolyze the peptidoglycan which contributes to the rigidity of the bacterial cell wall. On the other hand, cell walls of gram-negative bacteria are complex both in morphology and in chemical composition. The outer layer contains a complex mixture of lipid, protein, and lipopolysaccharide which envelops the inner layer of peptidoglycan. Therefore, gram-negative bacteria are difficult t o lyze by use of bacteriolytic enzymes.'O Pseudomonas solanacearum, X a n . oryzae, and Pseu. aeruginosa, which were used in this study, are gramnegative bacteria. The bacteriolytic enzymes (produced by Ach. lunatas) are capable of lyzing intact bacteria. I n the present study, these bacteriolytic enzymes were immobilized in collagen membrane. As the enzyme-collagen membrane was treated with glutaraldehyde solution, a part of the enzymes near the surface of the collagen membrane were crosslinked to the collagen fibril. Bacteria such as Pseu. solanacearum, Xan. oryzae, and Pseu. aeruginosa were lyzed by the enzyme-collagen membrane. The activity of the enzymecollagen membrane was 2% that of the native enzymes. These bacteriolytic enzymes contain protease. Therefore, the activity of the enzyme-collagen membrane was determined with 1% milk casein solution. The activity of the enzyme-collagen membrane against casein was 10% that of the native enzymes. This activity of the bacteriolytic enzyme-collagen membrane was almost the same as that of other enzyme-collagen membranes such as the alcohol dehydrogenase-collagen membrane and the lipase-collagen membrane." Therefore, this low activity may be caused by the difficulty of contact between the immobilized enzymes and the bacterial cell walls. The size of the bacteria used in this study was about 0.5-2.0 pm. From diffusion experiments using proteinslothe size of the meshes of collagen fibril matrix is about 300 X 300 A.2 It is difficult for bacteria to penetrate the collagen membrane. Therefore, only bacteriolytic enzymes on the surface of the collagen membrane can react with the bacterial cell walls. Pseudomonas aeruginosa was lyzed in the presence of a detergent such as sodium lauryl sulfate. As mentioned above, the outer layer of gram-negative bacteria is
308
KARUBE, SUGANUMA, AND SUZUKI
composed of lipid, protein, and lipopolysaccharide. This outer layer can be removed with detergent treatment.12 Therefore, Pseu. aeruginosa could be lyzed by the enzyme-collagen membrane after the complex outer layer of cell walls was removed with sodium lauryl sulfate treatment. The immobilized bacteriolytic enzymes showed stability against acidic and alkaline conditions. As previously reported,11 immobilized enzymes in collagen fibril network are stabilized against acid and alkali. The mechanism of the stabilization may be the same as that in othrr enzyme-collagen membranes. As mentioned above, inhibition of the native bacteriolytic enzyme with sodium chloride was observed. On the other hand, the enzyme-collagen membrane was stable against sodium chloride. Furthermore, it was activated with sodium chloride (-0.lM). It is well known that collagen fibrils swell in a sodium chloride s01ution.l~ Therefore, the swelling of collagen fibrils may increase the surface area of the membrane. The activity of the enzyme-collagen membrane was increased with increase in sodium chloride concentration (-0.1M). Then, the equilibrium of swelling was attained, and the excess sodium chloride inhibited the immobilized enzymes in the membrane. Therefore, the activity of the enzyme-collagen membrane decreased with increase in sodium chloride concentration beyond a certain level. Bacteria such as Pseu. aeruginosa and Pseu. solanacearum were continuously lyzed by a reactor containing rolled enzyme-collagen membrane. As the bacterial content of the suspension was high, the bacterial suspension was circulated through the reactor. I n conclusion, this study suggests the possibility of a new type of germ sterilization. Further developmental studies in this laboratory are being directed toward using the enzyme-collagen membrane reactor for the sterilization of air. The authors would like to thank Dr. S. Tsuru for his many valuable suggestions and for supplying bacteria.
References 1. J. L. Strominger and J. M. Chuysen, Science, 156, 213 (1967). 2. J. W. Costerton, J. M. Ingram, and K. J. Cheng, Bacteriol. Rev., 38, 87 (1974). 3. S. Tsumura, Abstract of dlst Annual Meeting of Society of Food Industry 18, (1974). 4. K. E. Pye and L. B. Wingard, Jr., Eds., Enzyme Engineering 2, Plenum Press, New York, New York, 1974. 5 . M. Aizawa, I. Karube, and S. Suxuki, Anal. Chim. Acta, 69, 431 (1974).
BACTERIOLYSIS BY IMMOBILIZED ENZYMES
309
6. S. Suzuki, M. Aizawa, and I. Karube, Immobilized Enzyme Technology, H. H. Weetall and S. Suzuki, Eds., Plenum Press, New York, 1975,p. 253. 7. I. Karube, S. Suzuki, S. Kinoshita, and J. Mizuguchi, Ind. Eng. Chem. Prod. Res. Develop., 10, 160 (1971). 8. W. Mejbaum, 2.Physiol. Chem., 258, 117 (1939). 9. K. Venkatasubramanian and W. R. Vieth, Biotechnol. Bioeng., 15, 583 (1973). 10. S. Murao and Y. Takahara, Agr. Biol. Chem., 38, 2305 (1974). 11. S.Suzuki, I. Karube, and I. Satoh, Biomedical Application of Immobilized Enaymes and Proteins, T. M. S. Chang, Ed., Plenum Press, New York, 1976. 12. C. A. Schnaitman, J . Bacteriol., 108, 553 (1971). 13. A. Veis, The Macromolecular Chemistry of Gelatin, Academic Press, New York, 1964, p. 29.
Accepted for Publication September 22, 1976
