IMPLANTS, MICROBEADS, MICROCAPSULES, AND METHOD FOR THEIR PREPARATION

The invention relates to novel implants, microbeads and microcapsules and t a method or their preparation.

A wide variety of substances have previously been formed into microcapsule and microbeads in order to improve their ease of handling and for other reasons For example, it is known to microencapsulate small drops of dyes and to coat th microencapsulated dye onto one face of a backing sheet. The coated sheet the serves as a substitute for carbon paper, since the impact of a typing element of typewriter upon the coated sheet ruptures the microcapsules and allows the dye t flow from the ruptures and mark the desired character upon an adjacent sheet o plain paper. The coated sheet has the advantage over normal carbon paper tha mere finger pressure will not rupture the microcapsules and thus the sheet can b freely handled without risk of staining the fingers.

During microencapsulation, a small drop or crystal of an active substance, typically about 0.1mm in diameter, is surrounded by a thin film of a capsule-formin carrier material. In contra^st, a microbead comprises a solid, substantially spherica particle of the bead-forming carrier material with the active substance disperse therethrough. The carrier πiaterial not only prevents the active material from running (if it is a liquid) but also isolates the active material from the external environment. For example, an easily-oxidizable substance can be stored in microencapsulated form without undergoing oxidation by atmospheric oxygen. Many biologically-active materials are susceptible to chemical changes during storage and microencapsulation is one possible way of protecting such substances from chemical change during storage. Moreover, if a microencapsulated biologically-active ma¬ terial could be produced which was suitable for injection into an animal body the active material could be released slowly into the blood stream as the capsule- forming material dissolves, thereby achieving the same type of "controlled release" action with injectable preparations as is achieved with orally administered large capsules which gradually dissolve within the alimentary canal.

Unfortunately, most previously known microcapsules or microbeads cannot safely be injected into an animal body (a term which is used herein to include the human body) without producing anaphylactic shock. The coating carrier material of an injectable microcapsule ex microbead is desirably one which will dissolve in the

bl-ood stream in - order to allow the active material within the microcapsules or microbeads to be released and to prevent undissolved microcapsules or microbeads blocking blood and other vessels. This obviously rules out synthetic resins as carrier materials. It is known to produce microbeads and microcapsules from proteins such as albumin (either natural albumin or albumin derivatives such as albumin active esters and albumin active intermediates), and it might be thought that such microbeads and microcapsules would be suitable for injection, since of course, •albumin in its native state can readily be metabolized in an animal's bloodstream. However, albumins are globular proteins having a complicated stereochemistry which depends upon subtle chemical interactions between various parts of the protein molecule. The natural stereochemical configuration of a globular albumin molecule is easily destroyed by heat, changes in pH or chemical reagents and once this natural stereochemical configuration has been changed, the albumin is no longer readily metabolized in an animal's bloodstream. In order to employ albumin as a carrier material in microbeads and microcapsules, it is necessary to effect some form of cross-linking between the albumin molecules and the previously known techniques for effecting such cross-linking of albumin (temperatures of 50 °C or higher, sudden changes in pH, or chemical cross-linking agents such as formaldehyde and glyoxal) cause physical denaturation of the albumin and changes in its stereochemical configuration, usually by formation of lysino-alanine bridges. The resultant physically denatured protein microcapsule or microbead is not immuno- logieally-aeeeptable. Although microcapsules or microbeads made from polylactic acid can be injected, the preparation of polylactic acid microcapsules and micro- beads presents great technical difficulties. It has now been discovered that, by cross-linking albumin under appropriate, mild conditions, microbeads and microcapsules can be produced which can safely be injected into an animal's blood stream. The preparation of such albumin microbeads and microcapsules is not technically difficult. Moreover, although the mild cross- linking technique of the invention is primarily useful for the formation of micro- beads and microcapsules, it may also be employed to produce macroscopic masses of cross-linked albumin having non-albumin materials dispersed therethrough and these larger masses of cross-linked albumin may be useful as implants.

Accordingly, the invention provides a composition comprising a cross-linked native albumin, a cross-linked albumin active ester or a cross-linked albumin active intermediate, and a non-albumin substance held within said albumin CD? the albumin is in a cross-linked but physically-native state.

OMPI

The method used to form the compositions of the invention differs slightl depending upon whether the albumin used is a native albumin (in which case it i necessary to add a cross-linking agent) or a self-polymerizable albumin native ester or intermediate (in which case no cross-linking agent is required). 5 Accordingly, the invention provides a method of incorporating a non-albumin substance in a matrix of cross-linked albumin, which method comprises: f orming a solution of a native albumin; and dispersing the non-albumin substance and a cross-linking agent in the albumin solution, thereby forming at least one body of cross-linked albumin containing the 0 non-albumin substance within the solution, characterized in that: the cross-linking agent is a bi-functional cross-linking agent having the property of cross-linking but not physically denaturing albumin; after addition of the cross-linking agent to the albumin solution a single phase containing both the cross-linking agent and the albumin is formed; and the albumin solution is maintained at a temperature not in exce.ss of 37°C and at a pH in the range of 4 to 10 while the non-albumin substance and the cross-linking agent are added thereto and thereafter until the body forms, thereby ensuring that the albumin in the body is in a crossed-linked but physically-native orm. The invention also provides a method of incorporating a non-albumin substance in a matrix of cross-linked albumin, which method comprises: forming a solution of a s-elf-polymerizable albumin derivative, this derivative being an albumin active ester or an albumin active intermediate; dispersing the non-albumin substance in the solution of the albumin derivative; and __„_ thereafter maintaining said solution under conditions which cause at least one body of cross-linked albumin containing the non-albumin substance to form within the solution, characterized in that the solution of the albumin derivative is maintained at a temperature not in excess of 37°C and at a pH in the range of 4 to 10 while the non- albumin substance is added thereto and thereafter until the body forms, thereby ensuring that the albumin in the body is in a cross-linked but physically-native form. There are three principal variations of the method of the invention, depending upon whether implants, microcapsules or microbeads are desired. If a quantity of the albumin solution is allowed to cross-link so as to form a single body of physically-native albumin containing the non-albumin substance incorporated there-

O PI

in, the resultant albumin body is usable as an implant in an animal's body. If it is desired to produce microcapsules, the non-albumin substance is added to the albumin solution dissolved in a solvent immiscible with the albumin solution and is main¬ tained in the form of droplets dispersed in the albumin solution, usually by vigorous stirring. Under these conditions, there are formed microcapsules comprising a core of the solution of the non-albumin substance surrounded by a coat of cross-linked but physically-native albumin. If desired, after these microcapsules have been formed, they may be washed to remove at least part of the immiscible solvent from their cores. Finally, if is desired to produce microbeads, after the non-albumin substance has been added to the albumin solution, the albumin solution is dispersed as droplets in a liquid immiscible therewith while maintaining the albumin solution at a temperature not in excess of about 37 °C and a pH in the range of about 4 to about 10. Again, the necessary dispersion of the albumin solution in the immiscible liquid can usually be effected by vigorous stirring. Under these conditions, the dispersed droplets of albumin solution form microbeads of cross-linked but physical¬ ly-native albumin having the non-albumin substance dispersed therein.

It should be noted that, when carrying out the method of the invention using a combination of a native albumin and a cross-linking agent, it is -esential that the cross-linking agent and the albumin be present in the same phase, since otherwise the desired physically-native but cross-linked albumin is not obtained. For example, U. S. Patent No. 4,147,767 describes a process for producing microbeads in which the albumin is cross-linked with glutaraldehyde at room temperature. However, in the method described in this patent the substance to be encapsulated and a native- albumin are dissolved in a single aqueous solution which is then injected via a hypodermic needle into a body of water-immiscible oil containing the cross-linking agent so that droplets of the aqueous solution are suspended in the oiL Thus, in this prior art process the albumin is present in the aqueous phase and the cross-linking agent is present in the oil phase. Accordingly, contact between the albumin and the cross-linking agent occurs at the surface of the aqueous droplets where a layer of heavily cross-linked albumin, which is not physically native, is formed. It will be appreciated that, as soon as any significant degree of cross-linking occurs in the albumin molecules adjacent the water/oil interface, the partially cross-linked albumin molecules are immobilized near the interface and are thereafter con¬ tinuously exposed to the cross-linking agent in the oil phase until complete cross- linking and denaturation of these albumin molecules occurs. On the other hand, albumin molecules which are originally present in the interior of the aqueous

droplets are protected from the cross-linking agent by the layer of cross-linked albumin formed on the surface of the droplet and are never exposed to the action of the cross-linking agent. The resulting microbead, containing heavily cross-linked albumin, will be strongly immunogenic, in contrast to the relatively non-immuno- genie microbeads of the present invention.

As previously mentioned, the albumin used in the implants, microcapsules and microbeads of the invention may be a native albumin, an albumin active ester or an albumin active intermediate. For example, the native albumin may be bovine serum albumin or rabbit serum albumin. Advantageously, the albumin is one native to the animal into which the microcapsules are to be injected or implanted.

A wide variety of substances may be incorporated in the implants, micro¬ capsules and microbeads of the invention. However, because the implants, micro¬ capsules and microbeads can be formed under very mild conditions, they are particularly suitable for biologically-active substances, especially those which are sensitive to heat ex to large changes in pH and which thus cannot be micro- encapsulated or microbeaded by previously known methods. For example, the implants, microcapsules and microbeads may contain steroids such as progesterone or norgestrel or enzymes such as asparaginase and proteases. Moreover because the temperatures used in the preparation of the implants, microcapsules and microbeads are low enough not to harm living cells, such cells may be incorporated into the implants, microcapsules and microbeads and thus immobilized. As is well-known, such immobilized cells may be employed to carry out chemical reactions, for example by allowing a solution of a substrate to flow through a column of microbeads containing cells having an enzyme capable of acting upon the substrate. Yeast cells are especially suitable fear incorporation into the implants, microcapsules and microbeads of the invention.

In the process of the invention, the temperature at which the albumin solution is maintained during and after the dispersion of the non-albumin substance therein desirably does not exceed 20 °C and is preferably in the range of 0 to 10 °C, most preferably 0 to 5 C; the optimum temperature in most cases appears to be about 4°C. The pH of the albumin solution is preferably maintained in the range of about 5.5 to about 8.5 and most preferably about 7. To ensure reasonably rapid formation of the implants, microcapsules or microbeads, the albumin solution should be relatively concentrated, being at least 10% w/v. However, it is preferred that the concentration of the albumin solution not exceed about 50% w/v and preferably not exceed 30% w/v. Although the albumin solution may be a purely aqueous solution, in some cases it is found advantageous to use an aqueous alcoholic solution, of albumin. ..

O PI

When native albumin is being employed, appropriate bi-functibnal cross-linking agents include drthiobissuceinimidyl propionate, glutaraldehyde and ethylene glycolyl bis(suceininidyl succinate), hereinafter referred to as EGS. A description of the preparation and properties of EGS is given in Abdella, P.M., Smith, P.K. and Royer, G.P., "A new cleavable reagent for cross-linking and reversible immobiliza¬ tion of proteins", Biochemical and Biophysical Research Communications, 87(3), 734-742 (1979).

As already mentioned, to produce microcapsules the non-albumin substance should be added to the albumin solution in a solvent immiscible with the albumin solution. For example, if the albumin is dissolved in aqueous methanol or ethanol, the non-albumin substance may be added in the form of a chloroform solution. On the other hand, to produce implants or microbeads, the non-albumin substance is preferably added to the albumin solution in the form of crystals. In some of the implants and microbeads of the invention, crystals of the non-albumin substance may be seen within the implant or microbeads under a microscope, whereas in other instant implants and microbeads no such crystals are visible. However, the presence or absence of microscopically visible crystals appears to have little or no effect on the release of the non-albumin substance from the implants or microbeads.

Obviously, the implants will be directly implanted surgically in an animal body, whereas the microcapsules and microbeads will normally be admimstered by injection. However, the microcapsules and microbeads may be administered by other methods such as orally ex in the orm of implants.

After injection or implantation into an animal body, the implants, micro¬ capsules and microbeads of the invention release the non--albumin substance contained therein relatively slowly as the cross-linked albumin is slowly dissolved, mainly by protease present within the animal body. Especially in the case of the microcapsules and microbeads, this sustained release action is desirable in that It avoids the initial high levels of active material produced in the blood stream by injection of -simple solutions of the active material and thus provides a much more consistent level of active material within the bloodstream. The sustained release action may also render it possible to increase the interval between injections and to reduce the total number of injections necessary. Furthermore, it is an important advantage of the implants, microcapsules and microbeads of the invention that the rate of release of active material therefrom may be controlled by Incorporating thereinto an inactive form of a protease capable of dissolving albumin. This inactive form of a protease is preferable a zymogen of a protease, preferably an

acyl derivative of serine protease such as trypsin or chymotrypsin. The inactiv form of the protease is incorporated into the implants, microcapsules or microbead by simply dispersing it in the albumin solution with the active substance. The inactive form of the protease has no effect upon the microcapsules and microbeads when they are stored dry (and thus does not effect their shelf li e) but when the microbeads or microcapsules are injected into an animal's bloodstream, the inactive form of the protease is gradually converted into the active orm (usually by simple hydrolysis) and the active form of the protease attacks the albumin present in the microcapsules or microbeads, thereby speeding up the release of the active substance therefrom. If a zymogen of a protease is used, it will be appreciated that the first portion of the zymogen to be hydrolyzed to its active form will thereafter rapidly hydrolyze the remaining zymogen, thereby securing rapid release of en¬ capsulated material from the microcapsules or microbeads.

Preferred compositions and methods of the invention will now be described, though by way of illustration only, with reference to the accompanying drawings, in which:

Figs. 1 and 2 are views of two different types of microcapsules produced in the following examples; and

Figs 3 and 4 show progesterone levels in rabbits after, respectively, sub- cutaneous and intramuscular administration to the animal of a steroid encapsulated in microbeads of the invention.

Example 1

This example illustrates the preparation of microcapsules containing a steroid.

Bovine serum albumin (50 mg.) was dissolved in 5 m of 40% methanol. Norgestrel (15 mg.) and 1.13mg. of dithiobissuccinimidyl propionate (a bi-funetional cro-ss-linking agent which does not physically denature -albumin) were dissolved in 1 ml. of chloroform. Both solutions were cooled to 4°C. and were near neutral pH. A chloroform solution was then slowly added to the albumin solution under vigorous stirring, thereby forming an emulaon of tiny chloroform droplets in the albumin solution. The size of these droplets is dependent on the speed of stirring the higher the stirring speed, the smaller the droplets. The emulsion was maintained at 4°C. for three hours; at the end of this time, the droplets had formed into microcapsules.

After one hour's further standing at room temperature the surface of the -droplets

"wrinkled" indicating the formation of an albumin membrane. The chloroform core of the microcapsules was then removed by washing with methanol, after which the

O PI ^~

microeapsules could be stained with coomassie blue, thus confirming that the skin of the microcapsules consisted of physically native albumin.

Example 2 This example illustrates the preparation of microbeads containing steroids and their injection into rabbits.

Rabbit or bovine serum albumin (600 mg.) was dissolved in 2.4 ml. of 0.1% sodium dodecyl sulfate in 1 mM. sodium phosphate buffer, pH 7.5. The solution was cooled to 4°C and 39 mg. of a finely divided steroid (progesterone or norgestrol) were dispersed in the solution, followed by 0.6 ml. of a 5% v/v aqueous solution of glutaraldehyde. The resultant dispersion was quickly pipetted into 150 mL of a vigorously stirred 14 mixture of corn oil and petroleum ether in a 250 ml. beaker. The stirring was continued for 15 minutes, at the end of which time solid microbeads had settled to the bottom of the beaker. The corn oil/petroleum ether mixture was decanted and the microbeads washed three times with petroleum ether and dried in a vacuum desiccator. The dried beads were then incubated twice with 25 mL of 0.1% w/v s-erum albumin in 0.05 M Tris-Cl (tris(hydroxymethyl)aminomethane chloride) buffer, pH 8.6, for fifteen minutes. The microbeads were then washed with about 500 mL of 1 mM. hydrochloric acid in a sintered glass filter funnel, the acid was washed off with distilled water and the beads were again dried in a vacuum desiccator.

The dried rabbit serum -albumin/progesterone microbeads, which are illustrated in Fig. 1, were dark brown in color and 100-200 microns in diameter. Under 100 x magnification, crystals of progesterone could be seen entrapped within the albumin. 70-250 mg. portions of the microbeads were suspended in corn oil and injected into rabbits weighing about 4 kg. Three rabbits received the injection subcutaneously, while three other rabbits received it intramuscularly. Progesterone in the serum was measured quantitatively by radioimmunoassay using the method of Powell, J.E. and Stevens, V. C, Clinical Chemistry 19(2), 210-215 (1973). The progesterone levels in the rabbits which had received the microbeads subcutaneously are shown in Fig. 3, while the progesterone levels in the rabbits which received the microbeads intra¬ muscularly are shown in Fig. 4. It will be seen that in both cases significant progesterone levels were present in the rabbit's bloodstream for a period of 20 days after injection. Control animals received similar amounts of progesterone dispersed in corn oil but not encapsulated. In the control animals, progesterone levels in the bloodstream diminished to zero in less than two days.

OMPI

The injections were repeated at intervals of one month. Six days after the third injection, one of the rabbits which had received the progesterone intra¬ muscularly was sacrificed. The injection sites were examined and found to be devoid of inflammation or any other abnormal appearance. The microbeads had completely disappeared from the site of the third injection. The injections into the other rabbits were continued at monthly intervals for seven months. Throughout the experimental period no adverse allergic responses were observed in either rabbit and the body weight, temperature and general condition of the rabbits appeared normal. No inflammation or necrosis was observed at the injection sites even in the rabbit in which the injections were continued over seven months. In the rabbits receiving the injections subcutaneously, the microbeads disappeared completely although their disappearance was slower than in the case of the rabbits receiving the intramuscular injections.

A further test using similar capsules which had been kept at 140°C for four hours (and in which the albumin was therefcsre denatured) showed a very much sslloowweerr rraattee ooff rreelleeaassee ooff pprrooggesterone. Moreover, these capsules heated to 140 °C are not degraded by proteases.

Example 3 This example illustrates the formation and assay of bovine serum albumin/- .asparaginase microbeads.

Bovine serum albumin (400 mg.) was dissolved in L6 ml. of 0.1M sodium phosphate buffer, pH 7.4, at 4°C. Asparaginase (15 mg.) was dispersed in this albumin solution and then 0.4 ml. of a 5% v/v aqueous solution of glutaraldehyde was mixed quickly with the albumin solution. In the same manner as in Example 2, the mixed solution was then suspended in 150 ml. of a 1:4 corn oil/petroleum ether mixture with vigorous stirring. After fifteen minutes, microbeads had settled to the bottom of the beaker containing the oil phase, which was then decanted. The microbeads were washed three times with petroleum ether and dried in a vacuum desiccator. The resultant microbeads, which are illustrated in Fig. 2, displayed physical characteristics similar to the microbeads produced in Example 2, except that no crystals of the enzyme were visible within the albumin polymer.

The activity of the microencapsulated asparaginase was assayed. 7.6 Ml. of

0.01M. asparagine in 0.05M. Tris-Cl buffer, pH 8.6, and 0.4 ml. of the same buffer were mixed and brought to a temperature of 37 C. 20 Mg. of the asparaginase

o?.?ι

microbeads were added to the asparagine solution with stirring. After ten minutes, 0.1 mL of the supernatant liquor was added to a mixture of 2 ml. of water and 0.2 ml. of Sigma Ammonia reagent (described in Winston, J.C., Jr., Methods in Enzymology (ed. Tabor, H. and Tabor, C.W.) XVII, 732, Academic Press, New York, 05 1970) to carry out Nessler's reaction. The mixture was then incubated at room temperature for 10 minutes and its absorbance at 480 nm was recorded, using an appropriate blank. The specific activity of the microbeaded asparaginase was 12.34 micromoles of asparagine hydrolyzed per minute per milligram of asparaginase, which is about 15.4% of the specific activity of native asparaginase.

10 Example 4

This example illustrates the formation and assay of bovine serum albumin/yeast alcohol dehydrogenase microbeads.

Bovine serum albumin (600 mg) was dissolved in 2.4 ml. of 0.1M. sodium phosphate buffer, pH, 7.4 at 4.°C. The albumin solution was stirred vigorously and

15 9.6 mg. of yeast alcohol dehydrogenase, 54.21 mg. of nicotinamide adenine di- nucleotide (NAD ) and 0.6 mL of a 5% v/v aqueous solution of glutaraldehyde were quickly dispersed in the albumin solution. The mixed dispersion was then suspended in 150 mL of the same oil mixture as in Examples 2 ^* and 3. After 15-20 minutes, microbeads separated at the bottom of the oil mixture, which was then decanted. 0 _The microbeads were washed three times with petroleum ether and dried in a vacuum desiccator for several hours.

The dried microbeads were then suspended in 50 mL of the same sodium phosphate buffer as before with stirring overnight at 4°C. At the end of this stirring, the microbeads were hydrated and swollen, while the supernatant buffer 5 was cloudy, probably due to an oil/water emulsion formed by residual oil on the microbeads. The microbeads were then washed with distilled water in a sintered glass filter funnel and some were dried by lyophilization while others were dried in a vacuum desiccator.

The physical characteristics of the microbeads dried in the vacuum desiccator 0 were similar to those produced in Examples 2 and 3, while the lyophilized microbeads had a fluffy appearance, but the alcohol dehydrogenase activity of the two preparations was substantially identicaL

The microbeaded alcohol dehydrogenase was assayed in a 3 ml. cuvette in a Cary l5 spectrophotometer equipped with a micro-submersible magnetic stirrer. The 5 production of NADH (the reduced form of NAD ) was followed by measuring the

absorbance at 340 nm. The specific activity was determined as about 60 units per gram of microbeads.