Artificial Organs 12(3):248-251, Raven Press, Ltd., New York 0 1988 International Society for Artificial Organs

Drug Delivery Systems in Biotechnology

T. M. S. Chang, *R. Langer, TR. E. Sparks, and $G. Reach

Artificial Cells and Organs Research Centre, Faculty of Medicine, McGill University, Montreal, Quebec: Canada, *Department of Nutrition and Food Science, Massachusetts Institute of Technology, Cambridge, Massachusetts,

U.S.A.; fSchool of Engineering and Applied Science, Washington University, St. Louis, Missouri, U.S.A.; and $Unitt de Recherches sur le diabkte et la nutrition chet l'enfant, INSERM, Facultt Villemin, Paris, France

A number of reviews have already discussed in some detail different types of drug delivery systems (1-6). Recent rapid progress in biotechnology has resulted in the availability of an increasing number of biotechnologically derived materials. Some of these require special drug delivery systems. The present discussion concentrates on these systems, with emphasis on injectable or orally administered systems.

DESIGN OF CONTROLLED RELEASE SYSTEMS FOR MACROMOLECULES

Theoretical and experimental analyses demon- strate that a hemispheric polymer-drug matrix lam- inated with an impermeable coating, except for an exposed cavity in the center face, can be used to achieve zero-order release kinetics (7). Hemispher- ic systems for high molecular weight drugs were prepared by casting ethylene-vinyl acetate copoly- mer and protein in a hemispheric mold at -80°C. This was followed by a two-step drying procedure (-20 and 20°C). In both systems, cavities were made in the center face of the hemispheres and the remainder of the matrices was coated with an im- permeable material. Zero-order release for 60 days at a rate of 0.5 mg/day was achieved from polymer matrices containing bovine serum albumin.

A lattice random-walk model is used to simulate diffusion in a porous polymer (8). This model may be useful for the practical design of drug release systems. Both interacting and noninteracting parti- cles were allowed to diffuse through a pore with a

single exit hole. It was found that the specific inter- actions among the diffusing particles have little in- fluence on the overall release rate. Diffusion through more complicated structures was investi- gated by simulating the diffusion of particles through two pores connected by a constricted chan- nel whose length and width were varied. The over- all rate of release was found to be proportional to the width of the constricted channel. When the length of the channel was greater than or equal to the length of the pore, the rate of release was also inversely proportional to the channel length. From a practical standpoint, release rates can be de- creased (and times for release increased) by one or two orders of magnitude by decreasing the width and expanding the length of the intercxmnecting channels in the polymer matrix.

Small magnetic spheres or cylindrical magnets were embedded within the polymer matrnx, which was then subjected to an oscillating magnetic field (9). In this fashion baseline release rates could be increased 5- to 10-fold with 5-10% standard error. Parameters critical to the regulation of this release included the position, orientation, and magnetic strength of the embedded objects and the amplitude and frequency of the applied magnetic field.

The kinetics of drug release from polymer-drug matrices containing an embedded magnet was con- tinuously monitored in vitro and in vivo (10). The application of an oscillating magnetic: field in- creased the rate of drug release from the polymer matrices. Within the limits of detection, the in- crease in release occurred immediately. remained stable for as long as the field was applied, and re- turned exactly to baseline upon withdrawal of the

Chang at Faculty of Medicine, McIntyre Medical Science Bldg., The increase in was propor- 3655 Drurnmond Street, Montreal, PQ, Canada H3AlY6. tional to field amplitude. The same pattern of re-

This work was a panel summary presented at the VIth World suits was observed in viva as in vitr-, though Congress of the International Society for Artificial Organs and the XIVth Congress of the European Society for Artificial or- higher-strength fields were required in vivo to gans. achieve the same effect observed in vitro.

Address correspondence and reprint requests to Dr. T. M. S.

248

DRUG DELIVERY SYSTEMS IN BIOTECHNOLOGY 249

BIODEGRADABLE CONTROLLED RELEASE SYSTEMS

Since the first report on the use of biodegradable polylactic acid for the slow release of insulin and other biologically active macromolecules (1 I), there has been much research on this approach (1-6). This includes the recent report of the use of poly- lactic acid for the slow release of prostaglandin (12). Prostaglandin E, has a very short life in the body. Polylactic acid for its slow release resulted in much more effective protection of rats with fulminant he- patic failure from cerebral edema (1 2).

Antischistosomal drugs are toxic and rapidly me- tabolized. Astiban acid is an antimonial drug that is effective against Schistosomiasis mansoni. Micro- capsules of astiban acid in poly(d, [-lactic acid) were formed by coacervation (13). Release studies show that an appreciable fraction of the drug is available at the surface for rapid solution. After the initial burst, the release of drug follows Higuchi’s equa- tion up to -80% release, with exponentially de- creasing release rates thereafter.

Another drug delivery system is based on bio- erodible polyanhydrides (14). With the water-labile anhydride linkage, a wide range of matrix degrada- tion and drug release rates can be obtained from these drug carriers. The possibility of enhancing the release externally by an ultrasonic source has also been explored. The polymers tested showed good tissue biocompatibility and their breakdown prod- ucts showed no adverse toxicological effects. Pre- liminary in vivo results confirmed the efficacy of this approach.

The feasibility of polymerizing naturally occur- ring amino acids via their side chains by bonds other than their amide bond was investigated (15). This avoids problems associated with many con- ventional polyamino acids produced by amide bonds.

CARRIER SYSTEM FOR ENZYMES AND MULTIENZYMES

A number of carriers have been used to irnmobi- lize enzymes for medical applications (5,6,16-22).

Artificial cells containing catalase, asparaginase, or urease are effective for replacing hereditary en- zyme deficiency in acatalasemic mice, suppressing lymphosarcoma in mice, and decreasing systemic urea levels (5). An extension of this approach is the use of liposomes for the microencapsulation of en- zymes (19). Liposomes have the advantage of being biodegradable; however, they appear to enhance immune response to the entrapped protein. Red

blood cells have also been used to microencapsulate enzymes by hemolysis and resealing (20). Rh anti- body-coated human red blood cells containing P-glucosidase have been used for targeting in Gau- cher’s disease. Other immobilization techniques in- vestigated include the cross-linkage of enzymes with proteins (21) or with polymers (22).

Progress in biotechnology has resulted in the availability of a bacterial enzyme, phenylalanine ammonia lyase. By using artificial cells containing this simpler enzyme system, one does not need to use the complex multienzyme system (23). Using oral administration one has also solved the problem of in vivo accumulation (23). Randomized control studies in phenylketonuric rats showed that oral ad- ministration for 7 days lowered systemic phenylal- anine levels from 331.4 & 26.4 mgldl in the control group to 82.7 ? 7.0 mg/dl in the treated group (p < 0.001). The level in the treated group is not signif- cantly different from that of normal rats (33.6 ’-+ 29.3 mg/dl). This is more effective than using phe- nylalanine-free diets. The simplicity and the ease of oral administration should allow clinical applica- tion.

Artificial cells were prepared to contain a multi- enzyme system of urease, leucine dehydrogenase, glucose dehydrogenase, and dextran-NAD (24). In this way each artificial cell can convert urea into ammonia, which is then converted into essential amino acids like leucine, isoleucine, and valine. In liver failure this can be used for the conversion of ammonia into leucine, isoleucine, and valine, thus replacing the need to remove ammonia and infuse these amino acids. Tyrosinase in artificial cells has been used in extracorporeal systems for hemoper- fusion to remove tyrosine in rats with galactos- amine-induced fulminant hepatic failure (25).

CARRIER SYSTEM FOR CELL CULTURES Some time ago human cells enclosed in artificial

cells were proposed for use as islet cells, hepato- cytes, and other cells to avoid immunological rejec- tion (5) . With the progress in biotechnology of cell cultures, this approach is now being extended by a number of groups using microencapsulation of islet cells.

Prolonged survival of islet allografts in streptozo- cin-induced diabetic rats was achieved by encapsu- lating individual islets in protective, biocompatible alginate-polylysine-alginate membranes. A single intraperitoneal transplant of encapsulated islets re- versed the diabetic state for up to I year. In con- trast, a single injection of unencapsulated islets was effective for <2 weeks (26).

A r t f Organs, Vol. 12, No . 3, 1988

250 T. M . S . CHANG ET AL.

Rat islets encapsulated in alginate-polylysine membranes were implanted intraperitoneally into nonimmunosuppressed streptozocin-induced dia- betic mice (26). Diabetes was reversed within 3 days, and the animals remained normoglycemic for up to 144 days, with a mean xenograft survival of 80 days. This was significantly greater than nonencap- sulated islets, which functioned for €14 days. The graft survival rate at SO days was 3 0 % . Xenografts of rat islets encapsulated in alginate-polyornithine membranes also had a prolonged survival rate. This study demonstrates that encapsulation of pancre- atic islets in semipermeable membranes can prolong xenograft survival in the absence of immunosup- pression.

The selective permeability of alginate microcap- sules, containing isolated rat islets of Langerhans or insulin-secreting RINmSF cells, was investigated in vitro (27). An increase in insulin release was ob- served when microencapsulated islets were stimu- lated by glucose plus theophylline, and when mi- croencapsulated RINmSF cells were stimulated by arginine plus theophylline. These findings demon- strate the permeability of the microcapsule mem- brane to these B-cell secretagogues and to insulin. Immunoisolation of RINmSF cells by microencap- sulation was assessed using a 'lCr cytotoxicity test. Significant "Cr release was observed when nonen- capsulated cells were incubated with complement and either the serum of a rabbit immunized with RIN cells or the sera of two patients with recently diagnosed Type 1 (insulin-dependent) diabetes. This effect was not observed with encapsulated cells. Both free and encapsulated cells released 80% of their initial radioactivity when incubated in the presence of HCl. These results clearly demonstrate pancreatic cell immunoisolation by microencapsu- lation. They also provide a method for the in vitro evaluation of the functional characteristics of mi- crocapsules, in terms of both insulin permeability and immunoprotection. Hepatocytes encapsulated within artificial cells significantly increased the sur- vival time of rats with galactosamine-induced fulmi- nant hepatic failure (28). Extensive research in this approach is being carried out by a number of cen- ters around the world.

REFERENCES

1. Chandry IA, Thies C, eds. Proceedings 13th International Symposium on control release of bioactive materials. Lin- colnshire, IL: Controlled Release Society, 1986:245.

2. Chien YW, ed. Novel drug delivery systems: fundamentals, developmental concepts and biomedical assessments. New York: Marcel Dekker, 1982.

3. Kydonieus AF, ed. Controlled release technologies: method, theory and applications. Boca Raton: CRC Press, 1980520.

4. Langer R, Blackshear PJ, Chang TMS, Klein MD, Schultz JS. Enzyme and drug delivery systems. Trans Am SOC Arfif Intern Organs 1986;32:639-55.

5 . Chang TMS. Artificial cells. Springfield, 11,: Charles C Thomas, 1972.

6. Chang TMS, ed. Microencapsulation including artificial cells. New York: Humana, 1984.

7. Hsieh DS, Rhine WD, Langer R. Zero-order controlled- release polymer matrices for micro- and macrclmolecules. J Pharm Sci 1983;72:17-22.

8. Balazas AC, Calef DF, Deutch JM, Siege1 RA, Langer R. The role of polymer matrix structure and interparticle inter- actions in diffusion-limited drug release. Biophys J 1985;

9. Edelman ER, Kost J, Bobeck H, Langer R. Regulation of drug release from polymer matrices by oscillating magentic fields. J Biomed Muter Res 1985;19:67-83.

10. Edelman ER, Brown L, Taylor J, Langer R. In vitro and in vivo kinetics of regulated drug release from polymer matri- ces by oscillating magentic fields. J Biomed Muter Res 1987;21:339-53.

1 1 . Chang TMS. Biodegradable semipermeable microcapsules containing enzymes, hormones, vaccines, and other biolog- icals. J Bioeng 1976;1:25-32.

12. Zhou MX, Chang TMS. Effects of polylactic acid microcap- sules containing prostaglandin E, on the suivi.va1 rates of grade I1 coma galactosamine-induced fulminant hepatic fail- ure rats. J Biomat Artif Cells Artif Organs 1987;15:549-58.

13. Gopalratnam PC, Mason NS, Sparks RE. Mic:roencapsula- tion of astiban acid for the treatment of Schistosomiasis mansoni. Appl Biochem Biotechnol 1984;10:2 13-20.

14. Leong KW, Kost J, Mathiowitz E, Langer R . Polyanhy- drides for controlled release of bioactive agents. Biomateri- als 1986;7:36471.

15. Kohn J , Langer R. Polymerisation reactions involving the side chains of alpha-L-amino acids. J Am C'hem SOC (in press).

16. Chang TMS, ed. Biomedical applications of immobilized en- zymes andproteins, vols 1 and 2 . New York: Plenum, 1977.

17. Holcenberg JS, Roberts J , eds. Enzymes as drugs. New York: Wiley Interscience, 1981.

18. Chang TMS. Artificial cells in medicine and biotechnology. Appl Biochem Biotechnol 1984;lO:S.

19. Gregoriadis G. Drug carrieris in bioiogy and medicine. New York: Academic Press, 1979.

20. Ihler G, Glew R. Enzyme loaded erythrocytes,. In: Chang TMS, ed. Biomedical applications of immobi!i;:ed enzynes and proteins. New York: Plenum, 1977:219.

21. Poznansky MJ. Enzyme-albumih polymer: new approaches to the use of enzymes in medicine. J Appl Bimhern Biotrc./r- no1 1984;2:41.

22. Abuchowski A, Davis FF. Polymer bound enzymes. In: Hol- cenberg JS, Roberts J, eds. Enzymes as drugs. New York: Wiley Interscience, 198'1 : 367.

23. Bourget L, Chang TMS. Phenylalanine ammonia-lyase im- mobilized in semipermeable microcapsules for enzyme re- placement in phenylketonuria. A c f a Bioch,im Biophys

24. Gu KF, Chang TMS. Conversion of ammonia or urea to

47:97-104.

1986 ;883: 432-38.

Artif Organs, Vol. 12. No. 3, 1988

DRUG DELIVERY S YSTEMS IN BIOTECHNOLOGY

L-leucine. L-valine and L-isoleucine by artificial cells, immo- bilized multienzyme system. Int J Biomater Artif Cells Artif Organs 1987; 15:297-3O4.

25. Shu CD, Chang TMS. Tyrosinase immobilized within artifi- cial cells for detoxification in liver failure: 11. In vivo studies in fulminant hepatic failure rats. Znt J Artif Organs 1981 ;4:82.

26. O’Shea GM, Sun AM. Encapsulation of rat islets of Lang-

251

erhans prolongs xenograft survival in diabetic mice. Diabe- tes 1986;35:943-6.

27. Darquy S, Reach G. Immunoisolation of pancreatic B cells by microencapsulation. An in vitro study. Diabetologia 1985 ;28: 776-80.

28. Wong H, Chang TMS. Bioartxcial liver: implanted artificial cells microencapsulated living hepatocytes increases sur- vival of liver failure rats. Int J Artif Organs 1986;9:335-6.

Artif Organs, Voi. 12, No. 3, 1988