Langmuir 2000, 16, 1485-1488

1485

Enzyme Encapsulation in Layer-by-Layer Engineered Polymer Multilayer Capsules

Frank Caruso,*, Dieter Trau, Helmuth Mohwald, and Reinhard Renneberg
Max Planck Institute of Colloids and Interfaces, D-14424 Potsdam, Germany, and Department of Chemistry, The Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, SAR Hong Kong, China Received August 27, 1999. In Final Form: November 23, 1999
We report on the encapsulation of enzyme (catalase) by the controlled polymer multilayer coating of biocrystals, achieved by the sequential adsorption of oppositely charged polyelectrolytes onto enzyme crystal templates. An extremely high enzyme loading in each polymer capsule is obtained, and the activity of the encapsulated enzyme is preserved. The polymer-encapsulated enzyme is stable against protease degradation: The polymer-coated enzyme retains 100% of its activity after incubation for 100 min with protease, whereas uncoated, solubilized catalase loses more than 90% of its initial activity within 100 min under the same conditions. This simple, general, and versatile approach can potentially be applied for the encapsulation of various crystallized substances for catalysis and drug delivery applications.

Encapsulation technologies are utilized in medicine, pharmaceutics, agriculture, and the cosmetic industries for the development of controlled-release delivery systems.1-3 Carrier systems of colloidal dimensions, such as liposomes, polymeric particles, and microemulsion droplets, are used for the sustained release of drugs, pesticides, fragrances, and other substances.3,4 Although such delivery systems are widely employed, problems associated with their stability and permeability are often encountered, thereby limiting their general application. Here we present a novel and general approach to using biocrystals as templates for the controlled encapsulation of biomolecules (enzyme) by polyelectrolyte multilayers, as illustrated in Figure 1. The method takes advantage of the fact that the enzyme is a crystalline suspension in water at pH 5-65 and therefore can be treated as a colloidal particle. We earlier demonstrated that charged polymer particles can be successfully employed as colloidal cores for the assembly of alternating charged polyelectrolyte multilayers.6-9 Removal of the colloidal core via its decomposition and penetration of its constituent oligomers through the polyelectrolyte walls resulted in hollow polyelectrolyte capsules.7,8 A similar strategy, where nanoparticles were deposited in alternation with polyelectrolyte, and the core subsequently removed, was recently also employed to produce inorganic and inorganic* To whom correspondence should be addressed. Fax: +49 331 567 9202. E-mail: frank.caruso@mpikg-golm.mpg.de. Max Planck Institute of Colloids and Interfaces. The Hong Kong University of Science and Technology.
(1) Langer, R. Nature 1998, 392, 5-10. (2) Park, K. Controlled Drug Delivery: Challenges and Strategies; American Chemical Society: Washington, DC, 1997. (3) Kreuter, J. Colloidal Drug Delivery Systems; Marcel Dekker: New York, 1994. (4) Yokoyama, M.; Okano, T. Adv. Drug Delivery Rev. 1996, 21, 7780. (5) Biochemica Information; Boehringer: Mannheim Germany, 1987; pp 15-16. (6) Caruso, F.; Donath, E.; Mohwald, H. J. Phys. Chem. B 1998, 102, 2011-2016. (7) Sukhorukov, G. B.; Donath, E.; Davis, S. A.; Lichtenfeld, H.; Caruso, F.; Popov, V. I.; Mohwald, H. Polym. Adv. Technol. 1998, 9, 759-767. (8) Donath, E.; Sukhorukov, G. B.; Caruso, F.; Davis, S. A.; Mohwald, H. Angew. Chem. 1998, 110, 2324-2327; Angew. Chem., Int. Ed. 1998, 37, 2201-2205. (9) Caruso, F.; Lichtenfeld, H.; Donath, E.; Mohwald, H. Macromol ecules 1999, 32, 2317-2328.

hybrid hollow spheres.10 Enzyme crystal templating, however, presents several challenges that do not apply when templating, for example, latex particles. First, the crystals are formed under delicate solution conditions; hence suitable conditions that facilitate polymer multilayer deposition on the crystal surface and do not destroy the enzyme crystal morphology (i.e. to avoid its solubilization) need to be found. Second, the permeability of the polymer capsule walls must be such that it permits encapsulation of the enzyme. In addition, since the primary usefulness of enzymes is their biological function, their activity should be preserved when encapsulated. Here we describe a method for enzyme encapsulation that satisfies all of these criteria. In our enzyme crystal templating procedure,11 the catalase crystals were separated from already solublilized protein by washing and centrifuging several times with 1 M potassium acetate buffer of pH 5 at 4 C. Chilled solutions were used to avoid significant solubilization of the enzyme crystals. A tapping mode atomic force microscopy image12 of a catalase crystal, approximately 8 12 m in size, is shown in Figure 2A. The catalase crystals exhibit a positive surface charge in water at pH 5 (zeta ()-potential ) +20 mV, catalase isoelectric point ) 5.8), as determined by electrophoretic mobility measurements.13 This positive charge on the surface of the crystals
(10) Caruso, F.; Caruso, R. A.; Mohwald, H. Science 1998, 282, 1111 1114. (11) The enzyme crystals were encapsulated as follows: The catalase crystals (Sigma, C-100) were washed twice with a chilled (4 C) solution of 1 M potassium acetate at pH 5 (buffer), with intermittent centrifugation steps (500 g, 4 min, 4 C) to remove the supernatant. The polymer layers were then assembled onto the enzyme crystals by the sequential deposition of poly(sodium 4-styrenesulfonate) (PSS), Mw 70 000, and poly(allylamine hydrochloride) (PAH), Mw 8000-11000, both obtained from Aldrich. PSS was dialyzed against Milli-Q water (Mw cutoff 14 000) and lyophilized before use. The first layer was deposited by adding a 0.5 mL aliquot of a 5 mg mL-1 aqueous PSS solution (containing 1 M potassium acetate, pH 5, 4 C) to 0.2 mL of the crystal suspension, occasionally shaking the suspension, and allowing 25 min for adsorption. The excess polyelectrolyte was removed by three repeated centrifugation (500g, 4 min, 4 oC)/chilled buffer wash/redispersion cycles. The next layer, PAH, was deposited from a 5 mg mL-1 solution containing 1 M potassium acetate (pH 5 at 4 C) using the same procedure and conditions. Subsequent alternating PSS and PAH layers were deposited in identical fashion until the desired number of polymer multilayers was achieved. (12) AFM images were obtained with a Digital Instruments Nanoscope IIIa AFM in tapping mode (TM) on samples deposited onto cleaned glass slides and air-dried.

10.1021/la991161n CCC: $19.00 2000 American Chemical Society Published on Web 01/29/2000

1486

Langmuir, Vol. 16, No. 4, 2000

Letters

Figure 1. Scheme showing the process used to encapsulate enzymes by using biocrystals as templates for the deposition of polymer multilayers, subsequent enzyme solubilization and release, and the formation of hollow polymer capsules. (1, 2) Polyelectrolyte layers are deposited stepwise onto the crystals by making use of the surface charge reversal that occurs upon adsorption of each layer. Each polyelectrolyte layer deposited bears an opposite charge to that already adsorbed. Excess, unadsorbed polyelectrolyte is removed by repeated centrifugation/wash redispersion cycles before the next layer is deposited. (3) Solubilization of the enzyme inside the polymer capsule by exposure to solutions of pH > 6 or acidic solution (pH < 4) results in a morphology change of the polymer capsule. (4) Release of the enzyme by rupturing the polymer capsule, achieved by exposure to solutions of pH > 11. (5) Exposure of the encapsulated enzyme to an oxidizing solution results in decomposition of the enzyme which then is expelled from the interior through the polymer walls, leaving behind hollow polymer capsules that originally encapsulate the enzyme.

in principle makes them suitably charged templates for the deposition of polyelectrolyte layers. Next, the first polyelectrolyte layer, poly(styrenesulfonate) (PSS), which is negatively charged, was added to the crystal suspension and allowed to self-assemble onto the crystal (Figure 1). Removal of the excess, unadsorbed polyelectrolyte was achieved by centrifugation of the crystal suspension, removing the supernatant, and repeated washing with a chilled potassium acetate solution. At this stage the enzyme crystal surface charge was reversed (-potential of -30 mV), indicating the successful adsorption of PSS. Repeated sequential deposition of poly(allylamine hydrochloride) (PAH) and PSS produced crystals with alternating positive (+20 mV) and negative (-30 mV) -potentials, respectively. These data clearly demonstrate a reversal of the surface charge, which is characteristic of polyelectrolyte multilayer growth on colloidal templates.6-10 To verify that the enzyme crystals could be successfully coated with polyelectrolyte multilayers, a fluorescently labeled polyelectrolyte (fluorescein isothiocyanate (FITC)PAH) was substituted for each alternate polyelectrolyte (polycation) in the buildup process. The fluorescence, as seen under a fluorescence microscope, originates from the surface of the polymer multilayer-coated enzyme crystals (Figure 2B). The fluorescence signal was found to increase systematically with an increasing number of fluorescent layers (data not shown). No noticeable changes in size or shape of the crystals were observed with the polyelectrolyte multilayer shell coating. Further, the polymer-coated enzymes could be stored for at least 30 days at 4 C without any noticeable change in morphology. These results show that enzyme crystals can be encapsulated within a polymer multilayer capsule via the stepwise, regular assembly of oppositely charged polyelectrolytes and that the crystal size and shape are retained upon their coating.
(13) Electrophoretic mobilities of the uncoated and polymer multilayer-coated enzyme crystals were measured with a Malvern Zetasizer 4 as described elsewhere.6-9 The mobility u was converted into a -potential using the Smoluchowski relation ) u/ , where and are the viscosity and permittivity of the solution, respectively.

Exposure of the polymer multilayer-encapsulated enzyme crystals to a solution of pH 2 resulted in the polymer capsules assuming a more spherical shape (Figure 3A). The morphology changes of the polymer capsules at pH 2 are attributed to solubilization of the enzyme crystal, which occurs because the polyelectrolyte multilayers are permeable to small mobile molecules (e.g., ions and water) in the solution phase. (Previous fluorescence spectroscopy studies have shown that similar polymer multilayers assembled on solid core colloidal particles are permeable to small polar molecules of diameter 1-2 nm.6,9) The more spherical shape is most likely due to the osmotic pressure buildup in the polymer capsules, caused by the high concentration of dissolved enzyme. The capsules did not rupture solely as a result of enzyme solubilization, indicating their high stability. Similar morphology changes were observed for other coated crystals. Catalase solubilization readily occurred at pH < 4 and pH > 6. Evidence that the enzyme was encapsulated within the polymer capsules was obtained by deliberately rupturing the polymer capsules. The polymer multilayer capsules were found to partially rupture upon exposure to alkaline solutions of pH > 11.14 An optical micrograph of capsule rupture and enzyme release is displayed in Figure 3B. Soon after capsule wall rupture, solubilized enzyme was expelled from the interior of the capsule. Video microscopy verified that the solubilized catalase was expelled upon exposure to solutions of pH > 11 as a result of capsule breakage. Capsule rupture and the associated expulsion of the enzyme are caused by the combination of osmotic pressure increase in the interior of the polymer capsules
(14) Several drops of the polymer multilayer-coated crystal solution were placed on a cleaned glass slide and most of the water was allowed to evaporate. An alkaline solution of pH > 11 was then pipetted onto an area on the slide near that which contained the encapsulated enzyme crystals. The alkaline solution was then allowed to make contact with the coated enzymes, and the polymer capsule rupture and enzyme release were monitored using optical and video microscopy. Rupture occurred immediately upon contact with the alkaline solution front. It should be noted that this release process may not be suitable for certain applications. To this end, we are currently employing a range of other polymers in order to effect enzyme release.

Letters

Langmuir, Vol. 16, No. 4, 2000 1487

Figure 2. Images of uncoated and polymer multilayer-coated catalase crystals. (A) AFM image of an uncoated crystal showing its shape and micrometer dimensions. (B) Fluorescence optical micrographs of catalase crystals coated with eight [(PSS/FITCPAH)4] polyelectrolyte multilayers, showing the applicability of the coating process to enzyme crystal templates. The fluorescently labeled (fluorescein isothiocyanate, FITC) polycation, FITC-PAH, is deposited in alternation with the polyanion, PSS.

and most probably a weakening of the electrostatic interactions between the charged moieties of the polymers due to the basic solution (pKa (PAH) is about 10). Figure 4 shows a representative TEM image of an airdried polymer capsule,15 which was obtained after exposing the polymer-coated enzyme crystals to an oxidizing solution (deproteinizer).16 The enzyme was decomposed by the deproteinizing treatment, allowing the expulsion of its fragment constituents from the interior by permeating the polymer capsule walls. The drying process (evaporation of the aqueous content by air-drying) induces a number of folds and creases in the polymer capsules. The capsules are also flattened and some spreading occurs; the diameters of the dried polymer capsules are ap(15) Samples for transmission electron microscopy (TEM, Philips CM12 microscope operating at 120 kV) were prepared by deposition of aqueous solutions of the hollow polymer capsules upon a carbon-coated copper grid, allowing them to air-dry for 1 min, and then blotting off the extra solution. (16) Hollow polymer capsules were obtained by exposing the polymer multilayer-coated capsules to a deproteinizer solution (Medical instruments; active constituent is sodium hypochlorite) for 15 min and washing three times with 100 mM sodium chloride solution. Although the pH of the deproteinizer solution is approximately 12, no capsule rupture was observed. This is most likely due to the rapid decomposition of the enzyme, hence avoiding any significant increase in osmotic pressure buildup in the capsules.

Figure 3. Representative optical micrographs of the polymer capsules encapsulating solubilized enzyme. (A) A polymer multilayer capsule containing solubilized enzyme in its interior. The close to spherical shape is assumed upon solubilization of the enzyme as a result of exposure to a solution of pH 2. Solubilization also occurs upon exposure to solutions of pH < 4 or pH > 6. (B) Rupturing of a polymer multilayer capsule causes the release of the entrapped, solubilized enzyme. Rupturing is achieved by subjecting the polymer capsules to alkaline solutions of pH > 11.

proximately 20 m. The texture of the capsules is characteristic of the polyelectrolyte film,8 although some residual (undecomposed) catalase can be seen. From AFM examination of air-dried polymer capsules (data not shown), it can be deduced from the lowest height dimension (35 nm), which is equivalent to twice the polymer capsule wall thickness, that the average thickness per polyelectrolyte layer is approximately 2 nm (the enzyme crystal templates were coated with eight layers). This value is consistent with data obtained for multilayers on solid core particles.7-9 The activity of the encapsulated catalase and the stability imparted by the polymer multilayer coating with respect to proteolysis were examined. The catalase activity was measured after its solubilization and release from the polymer capsules. This was achieved by following its reaction with hydrogen peroxide (substrate) to produce water and oxygen.5,17 A recovered specific activity of 97% was obtained, compared with 100% for the uncoated catalase. This shows that the polymer multilayer coating of the catalase crystals proceeds without causing any significant loss of enzyme activity. The results of the

1488

Langmuir, Vol. 16, No. 4, 2000

Letters

Figure 4. TEM image of an air-dried hollow polymer capsule comprising eight [(PSS/PAH)4] polyelectrolyte layers, obtained after decomposition of the encapsulated enzyme. The polymer capsule spreads out on the carbon surface on which it is dried, and folds and creases can be seen. Some undecomposed enzyme can still be seen in the interior of the capsule.

Figure 5. Stability of (a, d, e) solution-solubilized catalase and (b, c) polymer-multilayer encapsulated (solubilized) catalase with respect to proteolysis: (a) solution-solubilized catalase crystals, no protease incubation (control); (b) [(PSS/PAH)2]coated (four layers) catalase, protease incubation; (c) [(PSS/ PAH)4]-coated (eight layers) catalase, protease incubation; (d) and (e) repeat experiments for solubilized catalase, protease incubation. Proteolysis of the catalase was determined by measuring the decrease in the catalase enzyme activity (see ref 18 for details).

proteolysis experiments18 are shown in Figure 5. Solubilized, uncoated catalase (curves d and e) was inactivated by protease to more than 90% during an incubation time of 100 min. In contrast, no measurable loss in enzyme activity was observed for the polymer-encapsulated (solubilized) catalase within 100 min under the same conditions (curves b and c). (The catalase was solubilized and retained within the polymer multilayer capsule at the experimental pH of 7.) Further experiments showed that the activity of the encapsulated enzyme was fully retained after 24 h of exposure to protease. These results clearly demonstrate that a thin polymer coating of four layers (thickness of ca. 8 nm) is sufficient to prevent proteolysis of the polymerencapsulated catalase. This finding is consistent with the observation that proteins of diameter greater than ap(17) The enzyme activity was determined by first releasing the enzyme from the polymer capsules by exposure to ultrasound for 5 min at 20 C in 1 M phosphate buffer at pH 7 (this caused rupturing of the polymer walls) and using a standard enzyme activity assay;5 that is, by measuring the decomposition of the substrate hydrogen peroxide (H2O2) at 240 nm. The concentration of the catalase was determined spectrophotometrically at 280 and 405 nm (soret band). The specific activity was obtained by dividing the measured activity by the enzyme concentration. The specific activity of the uncoated catalase was measured in the same way and used for comparison. See also, for example: Tijsen, P. Practice and Theory of Enzyme Immunoassays, 8th Impression, Elsevier: Amsterdam, 1993; pp 202-203.15. (18) 10 mg of protease (Streptomyces griseus; Sigma P6911; activity ) 5.2 U/mg solid) was dissolved in 500 L of phosphate buffered saline (PBS, pH 7.0) (100 U/ml) and used as the stock solution. Samples (200 L) of the polyelectrolyte-coated catalase and solublilized catalase were separately incubated in a water bath at 37 C with either 30 L of the stock protease solution or 30 L of PBS (control experiment). The starting activity of the catalase for all samples was approximately 12 000 U/mL catalase. The starting activities were normalized to 100%. The final protease activity in the sample was 13 U/mL. Proteolysis of the catalase was determined by measuring the decrease in the catalase enzyme activity. The catalase activity measurements where carried out according to the previously described method (i.e., by spectrophotometric detection (240 nm) of the decomposition of hydrogen peroxide, see ref 17). Samples (5 L) were diluted and used for the activity measurements.

proximately 5 nm do not penetrate polyelectrolyte multilayer films.19 Important advantages of the enzyme crystal templating procedure presented are its versatility and generality: The thickness and composition of the polymer walls can be controlled by varying the number of polymer deposition cycles, thus potentially providing a straightforward and simple means to vary the permeability of the polymer capsules. It is expected that this process can be extended to encapsulate other biological materials in crystalline (or amorphous) form, e.g. peptides, antibodies, or enzyme catalysts in order to perform chemical reactions. Furthermore, the volume in the polymer capsule consists of an active catalyst (catalase). The layer-by-layer engineering of polymer multilayer capsules on biocrystal templates also represents a novel way to prepare immobilized catalysts with enhanced stability and activity, with the polymer multilayer coating providing a protective barrier for the enzyme in environments where enzyme-degrading substances (e.g., macromolecules, microorganisms, etc.) may be present. Overall, the strategy employed in this work provides a new route to produce tailored and optimized systems for numerous applications in biotechnology. Our current work is focusing on using various crystal templates, pH swellable, thermally responsive, and biocompatible polymers as the capsule constituents, on constructing microreactor systems for sequential enzymatic biocatalytic reactions, and on functionalizing the outer polyelectrolyte layers for targeting. Acknowledgment. This work was supported by The German Federal Ministry of Education, Science, Research and Technology (BMBF) and the Max Planck Society. We thank C. Durr for technical assistance and M. Giersig for help with electron microscopy.
LA991161N
(19) Caruso, F.; Niikura, K.; Furlong, D. N.; Okahata, Y. Langmuir 1997, 13, 3427-3433.