Development and characterization of a novel peroral peptide drug delivery system

Journal of Controlled Release 71 (2001) 307–318 www.elsevier.com / locate / jconrel

Development and characterization of a novel peroral peptide drug delivery system Farid A. Dorkoosh, J.C. Verhoef, G. Borchard, M. Rafiee-Tehrani, Hans E. Junginger* Department of Pharmaceutical Technology, Leiden /Amsterdam Center for Drug Research, Leiden University, P.O. Box 9502, 2300 RA Leiden, The Netherlands Received 17 November 2000; accepted 29 January 2001

Abstract Novel drug delivery systems were developed for peroral administration of peptide and protein drugs for site specific mechanical fixation at the gut wall and with specific release patterns. These so-called shuttle systems were designed by using superporous hydrogels (SPH) and SPH composite (SPHC) as the conveyor of a core which contained the model compound N-a-benzoyl-L-arginine ethylester (BAEE). Two different types of shuttle systems were evaluated: (a) core inside the shuttle system, and (b) core attached to the surface of shuttle system. Each of these systems was made of two parts: (1) the conveyor system made of SPHC which is used for keeping the dosage form at specific site(s) of the GI tract by mechanical interaction of the dosage form with the intestinal membranes, and (2) the core containing the active ingredient and incorporated in the conveyor system. The effect of formulation composition of the core on the release profile of BAEE was investigated by changing the type and amount of excipients in the formulations. In addition, the effect of various enteric-coat layers on the release profile and dissolving of the dosage form was investigated. The systems were also characterized for trypsin inactivation and Ca 21 binding. The release profile of BAEE from the core formulation consisting of PEG 6000 microparticles or small tablets showed the desired burst release. When these core formulations were incorporated into the conveyor system made of SPH and SPHC, a suitable time-controlled release profile was obtained. Changing the type, concentration and thickness of the enteric-coat layer influenced the starting time of BAEE release from the dosage form, which indicates the necessary lag time for dissolving of the dosage form at any desired specific site of drug absorption in the intestine. Both SPH and SPHC were found to partly inhibit the activity of trypsin, due to two mechanisms: Ca 21 binding and entrapment of the enzyme in these polymers. In conclusion, the presently developed delivery systems demonstrate suitable in vitro characteristics with an appropriate time-controlled release profile, making these systems very promising for effective peroral delivery of peptide and protein drugs.  2001 Elsevier Science B.V. All rights reserved. Keywords: Superporous hydrogel (SPH); SPH composite (SPHC); Intestinal drug delivery; Peptide and protein drugs; Double phase time-controlled release

*Corresponding author. Tel.: 131-71-527-4308; fax: 131-71527-4565. E-mail address: [email protected] (H.E. Junginger).

1. Introduction The design of novel delivery systems, specially for hydrophilic and macromolecular drugs such as pep-

0168-3659 / 01 / $ – see front matter  2001 Elsevier Science B.V. All rights reserved. PII: S0168-3659( 01 )00232-2

308

F. A. Dorkoosh et al. / Journal of Controlled Release 71 (2001) 307 – 318

tides and proteins, with the aim of increasing their peroral bioavailability is an ongoing process in pharmaceutical science. The increasing number of peptides and proteins by means of biotechnology as well as gentechnology will just serve to accent the importance of such development efforts. The majority of existing peptide and protein drugs is commonly administered via parenteral injection routes which are inconvenient, time- and money-consuming and occasionally dangerous. Since oral administration has a good patient compliance, this route attracts the greatest interest [1,2]. There are several issues, which are crucial for designing an effective delivery system for these categories of drugs. These include site-specific drug delivery to achieve a predictable and reproducible absorption in therapeutic doses, improving the poor peroral bioavailability of these drugs, and overcoming both transmucosal transport and metabolic barriers for achieving a desirable therapeutic effect [3,4]. The poor bioavailability of these drugs after oral administration is due to the interplay of low permeability characteristics, rapid degradation by proteolytic enzymes in gastrointestinal tract, and clearance mechanisms such as firstpass effect and excretion in the bile [2,5,6]. It has been reported that enzyme inhibition, opening of intercellular tight junctions and targeting of the drug to specific intestinal site(s) of absorption are important factors to be taken into account when developing a peroral peptide drug delivery system [7– 11]. Up to now, several approaches have been employed for site-specific peptide drug delivery such as using magnetic systems, unfoldable or expandable systems and mucoadhesive systems. Each one of these methods has a specific mechanism of action

which is different from our presently developed system. This system is able to keep the dosage form mechanically at the site of drug absorption. This is a new concept in drug delivery, and different from mucoadhesion. Since the mucus layer of the human intestine is regenerated with a turn-over time of 12–24 h [12], mucoadhesive dosage forms will stay — if at all — attached to the mucosal layer for a specific period at specific sites of the intestine, in contrast to mechanical attachment of the dosage form. In addition, by the use of this novel system it is possible to get an appropriate time-controlled release profile, necessary for intestinal absorption of peptide drugs (Fig. 1). For a normal release profile from the dosage form, drug release will be started from time zero (Fig. 1A), indicating that from the moment the dosage form is in the intestinal lumen the drug release is started. However, for delivery of peptide drugs a lag time of 20–30 min is necessary to inactivate proteolytic enzymes and to open the tight junctions. Thereafter a burst release is required in which the whole amount of peptide drug should be released from the dosage form in a short period of time. This type of drug release is depicted in Fig. 1B and called time-controlled release profile. For achieving mechanical interaction of this novel drug delivery system with the intestinal wall superporous hydrogel (SPH) and SPH composite (SPHC) polymers were used. Superporous hydrogels are a new generation of hydrogels, which are able to swell very quickly due to their highly porous structure [13,14]. The difference between SPH and SPHC is their swelling ratio and mechanical stability. SPH swells more quickly, but it is mechanically less stable, whereas SPHC is swelling less, but is mechanically more stable. In the present study the

Fig. 1. Kinetics of drug release from intestinal formulations.

F. A. Dorkoosh et al. / Journal of Controlled Release 71 (2001) 307 – 318

developmental procedure of this system is described, resulting in optimal formulations with appropriate time-controlled release profiles for the model compound N-a-benzoyl-L-arginine ethylester (BAEE). Moreover, the system is characterized with respect to drug release, capability for enzyme inactivation, and Ca 21 binding properties.

2. Experimental

2.1. Materials BAEE (N-a-benzoyl-L-arginine ethylester), BA (N-a-benzoyl-L-arginine), trypsin (from bovine pancreas, 8600 Units / mg) and MES (2-[N-morpholino]ethane-sulfonic acid]) were purchased from Sigma Chemical (Bornem, Belgium). Titriplex  III (ethylene dinitrilotetraacetic acid, dinatrium salt) was from Merck (Darmstadt, Germany). PEG 6000 (polyethylene glycol) was obtained from Fluka (Zwijndrecht, The Netherlands). Tetraglycerol pentastearate (TGPS, HLB 2.6) and tetraglycerol monostearate (TGMS, HLB 8.4) were applied as poly(glycerol esters of fatty acids) (Sakamoto Yakuhin Kogyo, Osaka, Japan). Carbomer (C934P) was supplied by BF Goodrich (Cleveland, OH, USA). Superporous hydrogel (SPH) and SPH composite (SPHC) were synthesized as described previously [13]. Polyvinylpyrrolidone (PVP) was from Brocacef (Maarssen, The Netherlands), and Eudragit S100 was ¨ supplied by Rohm (Darmstadt, Germany). Gelatine capsules were kindly donated by Capsugel  (Colmar, France). The water (MQ) used was filtered by a Milli-Q UF plus ultrapure water system from Millipore (Etten-Leur, The Netherlands). All other compounds were of analytical grade.

2.2. Preparation of delivery systems Basically, this novel delivery system (so-called shuttle system) has two different types: (a) core inside the shuttle system, (b) core attached to the surface of shuttle system (Fig. 2) [15]. Each one of these shuttle systems is composed of two components: a core and a conveyor system. Core is the part which consists of drug with appropriate excipients. BAEE (N-a-benzoyl-L-arginine ethylester) was

309

used as a model compound. Conveyor systems were made of superporous hydrogels (SPH) and SPHC.

2.2.1. Core inside the shuttle system Various formulations for the core were made, which are summarised in Table 1. All formulations were prepared in two different forms: microparticles and gross mass. For preparing the microparticles and gross mass of formulations 1 and 2, first PEG 6000 was melted and BAEE was dispersed completely in it, and then the mixture was cooled down in order to get the gross mass. For making microparticles, this mixture was crushed in a mortar and sieved through a sieve mesh size 400 mm. Microparticles smaller than 400 mm were used as a core formulation. For preparing the microparticles and gross mass of formulations 3 and 4, initially TGPS and TGMS were melted together; then BAEE was added and mixed completely. Thereafter, C934P was added and mixed throughly while the whole mixture was cooled down. Microparticles of these formulations were prepared as described for formulations 1 and 2. The second part of the present delivery system is the conveyor system (Fig. 2A), which is made of SPH and SPHC. SPH and SPHC were synthesized according to Chen et al. [16] with some modifications in order to make these polymers appropriate for intestinal drug delivery purposes [13]. SPHC, which is mechanically more stable than SPH, was used as the body of the conveyor system and SPH, which has a much higher swelling ratio than SPHC, was used as a cap for the conveyor system. Since the core has to be incorporated inside the SPHC, a hole was made inside SPHC. After synthesising SPHC and dialysing the polymer in order to get rid of monomers, the polymers are in the swollen state [13]. At this stage a hole was made inside of the polymer by use of a borer. Thereafter, the polymers were dried by one of the two following procedures: (1) at ambient temperature; the drying time was dependent on the average pore size of the system and approximately 1 week, and (2) under reduced pressure at 608C for 1 day. Both drying procedures allow the polymers to shrink, depending on their porosity in such a way that they can be fitted in gelatine capsules (size 000). This is the so-called body of the conveyor system consisting of SPHC. Then the core was incorporated

310

F. A. Dorkoosh et al. / Journal of Controlled Release 71 (2001) 307 – 318

Fig. 2. Schematic view of the mechanism of action of the drug delivery system. (A) Core inside the shuttle system; drug is being released from the core (dashed arrows) after swelling of the conveyor system (SPHC / SPH). (B) Core attached to the surface of shuttle system, drug is released (dashed arrows) after attachment of delivery system to intestinal wall.

inside the body of the conveyor system and this system was capped by a piece of SPH (Fig. 2A). One part of these systems was prepared in such a way in order to be used for dissolution studies. The other part of the systems was inserted in gelatine capsules (size 000) and enteric-coated in two different ways in order to study the influence of the gelatine capsule and enteric-coat layer on the behaviour of the systems and also to evaluate the drug release profiles. The gelatine capsules were enteric-

coated by two procedures: (1) using a 2% solution of polyvinylpyrrolidone (PVP) in isopropanol and subsequently by a 2% aqueous dispersion of Eudragit S100. The first layer is a protective layer and prevents the gelatin capsules from the deformation during the coating process. (2) Using a 6% solution of Eudragit S100 in isopropanol. For enteric-coating of gelatine capsules, the pan coating method was used with the ERWEKA apparatus (Heusenstamm, Germany) with 25 rpm and spraying 1 ml coating

F. A. Dorkoosh et al. / Journal of Controlled Release 71 (2001) 307 – 318 Table 1 Formulations of the core with either PEG 6000 or a mixture of 20:80 ratio of C934P/(TGPS,TGMS)a Formulation

BAEE (mg) PEG 6000 (mg) Carbomer (mg) TGPS (mg) TGMS (mg) Total

1

2

3

4

10 50 – – – 60

10 100 – – – 110

10 – 10 28 12 60

10 – 20 56 24 110

a The percent ratio of TGPS to TGMS was 70:30%, yielding the HLB value of 4.34. BAEE, N-a-benzoyl-L-arginine ethylester; PEG 6000, polyethylene glycol 6000; TGMS, tetraglycerol monostearate; TGPS, tetraglycerol pentastearate. All formulations were prepared in two forms: microparticles or gross mass.

solution every 30 s to 1 min (this time was related to the amount of heat which was applied for drying of the coating solution).

2.2.2. Core attached to the surface of shuttle system In this system (Fig. 2B) cores were in the form of small tablets which were formulated as follows: BAEE was dispersed in melted PEG 6000. After cooling, the mixture was crushed in a mortar and sieved through sieve mesh size 400 mm. Then, just before making the tablets, the microparticles were mixed with Mg stearate and the tablets with a hardness of 40 N were prepared in single punch tableting machine. The size of tablets was 3 mm in thickness and 4 mm in diameter. The second component of this delivery system is conveyor system which is made of only SPHC, as depicted in Fig. 2B. The procedure for making this conveyor system is the same as for the core inside the shuttle system, but in this case two holes were made on the counter-side of the conveyor system. An important issue in making this type of delivery system is the attachment of the small tablets to the surface of shuttle system. When the polymer swells, the size of holes in which the tablets are placed will be enlarged, and the tablets will come out of their place before they attach to the intestinal wall by the swelling of the polymer. Therefore, a bio-adhesive glue was needed to attach the tablets to the polymer in such a way that, when the polymer swells, the

311

tablets will be kept inside of the holes, up to the time the polymer swells and keeps the dosage form at the site of drug absorption. One drop of a cyanoacrylate glue is suitable to attach the tablets to the surface of the polymers. Even after complete swelling, the tablets remained attached to the polymer. After preparation of the system, it was placed in gelatine capsule size 000 and the capsules were entericcoated as described above.

2.3. Drug release studies The drug release studies were divided into three parts: (1) release of BAEE (model compound) from the core itself, (2) release of BAEE when the core was incorporated inside the conveyor system, and (3) release of BAEE when the system was inserted inside enteric-coated gelatine capsules. The dissolution for each step was investigated by using the standard USP XXIII paddle apparatus. The rotation speed was 100 rpm and the dissolution medium was 500 ml phosphate buffer solution (PBS) of pH 7.2 at 37618C. PBS was prepared according to USP XXIII. Three replicates for each sample were tested and their mean percent release was calculated. Volumes of 1 ml were withdrawn at predetermined interval times from the dissolution medium, diluted with 0.5 ml stop solution (phosphoric acid pH 1.8) and analysed for BAEE as described in Section 2.5.

2.4. Trypsin inhibition studies SPH and SPH composites were investigated for their ability to inactivate trypsin, which is the most predominant proteolytic enzyme in the intestinal tract. For this study, 10 ml of 24.5 U / ml trypsin solution were added to 200 mg of SPH or SPHC polymers. The polymers were allowed to suck up the trypsin solution, and after 20 min 10 ml of 1.5 mmol BAEE / l was added. As a positive control BAEE solution was added to trypsin solution without the presence of polymer, and as a negative control only the BAEE solution was kept under the same condition. All incubations were performed at 378C. After predetermined intervals, 100 ml samples were taken and diluted with H 3 PO 4 solution (pH 1.8) to stop the trypsin activity. The degradation of BAEE was

312

F. A. Dorkoosh et al. / Journal of Controlled Release 71 (2001) 307 – 318

studied by determining the formation of BA as a metabolite, using HPLC as described in Section 2.5.

2.5. HPLC analysis Analysis was performed by HPLC with a ThermoSeparations system, consisting of a P 200 gradient pump, AS 100 autosampler, ABI UV/ Vis detector and a datajet integrator. All measurements were evaluated with software package ‘Winner on Windows’ (Thermo-Separations, Breda, The Netherlands). The stationary phase was a Lichrosorb 7 RP 18 column 10033.0 mm (Chrompack, Middelburg, The Netherlands) accompanied with a RP 18 precolumn. For measuring BAEE, gradient elution was performed with the following two mobile phases: eluent A 90% 0.01 M ammonium acetate buffer pH 4.2 and 10% acetonitrile, and eluent B 50% 0.01 M ammonium acetate buffer pH 4.2 and 50% acetonitrile. The wave length for UV detection was 253 nm and the injection volume was 20 ml. Gradient elution was performed as follows: 0–3 min 92% A / 8% B, isocratic with flow rate of 0.75 ml / min; 3–8 min 50%A / 50%B, linear gradient with flow rate of 0.75 ml / min; 8–9 min 50%A / 50%B, isocratic with flow rate of 0.75 ml / min; 9–11 min 92% A / 8% B, linear gradient with flow rate of 0.75 ml / min; 11–13 min 92% A / 8% B, isocratic with flow rate of 1.05 ml / min; 13–14 min 92% A / 8% B, isocratic with flow rate of 0.75 ml / min. Retention time of BAEE was 7.0 min. For measuring BA (being the metabolite of BAEE), mobile phase A was used for isocratic elution at a flow rate of 0.75 ml / min; the retention time of BA was 3.1 min.

2.6. Ca 21 binding studies In order to study the mechanism of action for trypsin inhibition using SPH and SPHC, the capability of these polymers for Ca 21 binding was studied. First, 100 mg of SPH or SPHC polymers were placed in 4 ml of 40 mg / ml CaCl 2 solution in MES / KOH buffer (pH 7.2, containing 250 mM mannitol) for 20 h. Then the solutions were measured for free calcium by complexometric titration with Titriplex  III using calcein as an indicator.

Thereafter, the polymers were washed five times with MES / KOH buffer to take out the previously entrapped calcium (unbound Ca 21 ). These washedout solutions were also determined for calcium by complexometric titration. Bound calcium was calculated by subtracting the amount of free and entrapped calcium from the total amount of calcium in each solution.

3. Results

3.1. Preparation of dosage forms and in vitro release studies Initially, the release of BAEE from core alone was investigated. Fig. 3A depicts the release of BAEE for the formulations 1 (50 mg PEG) and 2 (100 mg PEG), either in gross mass or microparticles (core inside). BAEE was released from all formulations very quickly and the release patterns were identical between the various formulations. This means that there is a burst release after 5 min for all formulations and that the whole amount of incorporated BAEE is released after 15–20 min. It did not make any difference whether the core formulations were in gross mass or microparticles form. Because PEG 6000 is a water-soluble base, it dissolves quickly and BAEE is released from all formulations in the same manner. Therefore, all of them were used in the second step of dosage form development, in which BAEE core was incorporated inside of the SPHC. Fig. 3B shows the release from the core, when it is in the form of small tablets for the second system (core attached to the surface of shuttle system). Just similar as for the core inside of the shuttle system, the release of BAEE from small tablets was quite fast due to the aqueous solubility of PEG 6000. Therefore, these tablets were used for the preparation of the final dosage forms. Fig. 4 depicts the release of BAEE for the formulations 3 (10 mg C934P/ 40 mg PGEF) and 4 (20 mg C934P/ 80 mg PGEF), either in gross mass or microparticles. The release of BAEE from these formulations was observed to be delayed in comparison to the release from PEG 6000 containing formulations (Fig. 3). The reason for this is the

F. A. Dorkoosh et al. / Journal of Controlled Release 71 (2001) 307 – 318

313

Fig. 4. Release of BAEE from core formulations 3 and 4, either in gross mass or microparticles; Data are expressed as mean6S.D. (n53). (♦) Gross mass (core formulation 3); (j) gross mass (core formulation 4); (m) microparticles (core formulation 4); (3) microparticles (core formulation 3).

Fig. 3. Release of BAEE from (A) core inside the shuttle system, core formulations 1 and 2, either in gross mass or microparticles. (♦) Gross mass (core formulation 1); (j) microparticles (core formulation 1); (m) gross mass (core formulation 2); (3) microparticles (core formulation 2). (B) Core attached to the surface of shuttle system, small tablets as a core. Data are expressed as mean6S.D. (n53), (♦) small tablets of PEG 60001BAEE (core outside).

presence of carbomer (C934P) in the formulation which in contact with water forms a gel. Hence the release will be retarded. When the formulations were made in the form of gross mass, this retardation was more apparent (Fig. 4), because carbomer formed clamps and the release of BAEE was further delayed. However, when the formulations were in microparticulate form, the release of BAEE was quicker, because the microparticles do not form any clamp when absorbing water and therefore, the release is not delayed.

It is also obvious from Fig. 4 that, when there was a lower amount of carbomer in both the gross mass and the microparticles formulation (compare formulations 3 with formulations 4), the release of BAEE was faster, due to less gel formation of carbomer. Since it was the aim in the present study to develop dosage forms with a burst release of drug after a specific lag time (10–15 min), the release from the core must be rapid enough that, after swelling of the SPHC, the whole amount of the incorporated drug is released in a short period of time (10 min). Formulations 3 and 4 do not fulfill these requirements, and were therefore not used for further developmental experiments. Fig. 5 shows the release profiles of BAEE from core formulations 1 and 2 (in either gross mass or microparticles) incorporated in SPHC as the conveyor system and capped with a piece of SPH. For all formulations, BAEE release started after a lag time of 10–20 min. Thus, firstly the conveyor system absorbs water and swells and thereafter the core is wetted, and BAEE will be dissolved and released. As observed in Fig. 5, the release of BAEE from the gross mass was delayed in comparison to that from microparticles. The reason for this is that it takes more time for dissolving the whole compact core mass (low surface area) in comparison to the microparticles (high surface area).

314

F. A. Dorkoosh et al. / Journal of Controlled Release 71 (2001) 307 – 318

Fig. 5. Release profile of BAEE form core formulations 1 and 2 (either gross mass or microparticles) incorporated in SPHC as the conveyor system; mean6S.D. (n53). (♦) Microparticles (core formulation 1); (j) microparticles (core formulation 2); (m) gross mass (core formulation 1); (3) gross mass (core formulation 2).

On the basis of these results it was decided to use microparticles instead of the gross mass. In addition, because the release of BAEE from formulations 1 and 2 was approximately the same (Fig. 5), formulation 1 (containing lower amounts of excipients) was selected for further studies as a core inside the shuttle system. Both selected dosage forms, (a) core inside the shuttle system consisting of a core (containing 10 mg BAEE and 50 mg PEG 6000 in microparticles form) incorporated in SPHC as a conveyor system and finally capped with SPH, and (b) core attached to the surface of shuttle system consisting of two small tablets attached to the two holes on the surface of SPHC as a conveyor system, were inserted into gelatine capsules and enteric-coated in two different ways. The release profiles of BAEE from these final enteric-coated dosage forms are depicted in Fig. 6. When the gelatine capsules were enteric-coated with two layers, a lag time of approximately 40 and 60 min was observed for core outside and core inside the shuttle system, respectively, accompanied by a burst release of the whole amount of BAEE within the following 20 min (Fig. 6A). In the case of enteric-coating with one layer, the lag time was shortened to about 30 and 40 min for core outside and core inside the shuttle system, respectively (Fig. 6B).

Fig. 6. Release profiles of BAEE from the final dosage forms in enteric-coated gelatine capsules. (A) Enteric-coated with 2% solution of PVP in isopropanol and 2% aqueous solution of Eudragit S100. (d) Core outside; (j) core inside. (B) Entericcoated with 6% solution of Eudragit S100 in isopropanol. (d) Core outside; (j) core inside; mean6S.D. (n53).

3.2. Trypsin inhibition studies The results of trypsin inactivation are shown in Fig. 7. In case of negative control (only BAEE solution), there was almost no formation of BA during the time course of the experiment. On the contrary, for the positive control (BAEE with trypsin) BAEE was completely degraded into BA. In presence of the polymers SPH or SPHC, the formation of BA was less than observed for the positive control, which indicates these polymers are able to

F. A. Dorkoosh et al. / Journal of Controlled Release 71 (2001) 307 – 318

Fig. 7. Trypsin inhibition studies using SPH and SPH composite in comparison to negative control (BAEE solution) and positive control (BAEE with trypsin solution). (♦) Positive control; (j) SPH; (m) SPH composite; (3) negative control; mean6S.D. (n53).

partly inhibit the activity of trypsin. SPH appeared to be more potent than SPHC.

3.3. Ca 21 binding studies Table 2 depicts the Ca 21 binding studies using SPH and SPHC. The majority of calcium appeared to be entrapped in the interconnected porous structure of the polymers. SPH was able to bind more Ca 21 in comparison to SPHC, due to presence of more available carboxylic groups in the structure of SPH polymer.

4. Discussion The present study reports on the development of a

315

novel drug delivery system for improved intestinal absorption of peptide and protein drugs. The mechanism of action for this system is shown in Fig. 2. This drug delivery system is called a shuttle system, which is made of two distinct parts: (1) core consisting of the drug within an appropriate formulation, and (2) conveyor system consisting of superporous hydrogel (SPH) and SPHC. The entericcoated conveyor system takes the core to the desired intestinal site of drug absorption depending on the release time of the enteric coating. After dissolving of the enteric coated and gelatine layers, the SPHC conveyor systems will swell rapidly and be mechanically attached at the desired part of the gut wall. Two different delivery systems have been developed: (1) core inside the shuttle system (Fig. 2A), and (2) core attached to the surface of shuttle system (Fig. 2B) [15]. For the core inside the shuttle system, the core (including the drug of interest) is incorporated inside the conveyor system, which is made of SPHC. The system is closed with a plug made of SPH (Fig. 2A-A). SPH and SPHC differ substantially in mechanical stability and swelling ratio [13,16]. SPH swells quicker, but is mechanically less stable than SPHC. The use of SPHC as a conveyor system guarantees resistance to peristaltic movement of the intestine. Therefore, the system will be kept at the site of drug absorption by mechanical interactions (Fig. 2A-B). The SPH plug will swell more rapidly and separate apart, thus allowing for a burst release of the drug incorporated in the core (Fig. 2A-C). When the SPH polymers are swelling, the microparticles of the core are wetted and start to dissolve and come out from the conveyor system. For the core attached to the surface of shuttle system, SPHC is used as a conveyor system and the cores are attached to the surface of conveyor system using a biodegradable glue (Fig. 2B-A). In this case the core (which includes the drug of interest) is in

Table 2 Ca 21 binding studies using SPH and SPH composite polymers a

21

Entrapped within polymer (unbound Ca ) Bound Ca 21 to polymer Free Ca 21 a

SPH (%)

SPH composite (%)

70610 2364 761.5

6768 1062 2366

Data are presented as % of total amount of Ca 21 added, and expressed as mean6S.D. of four experiments.

316

F. A. Dorkoosh et al. / Journal of Controlled Release 71 (2001) 307 – 318

the nearest possible position of the gut wall for efficient drug absorption. Using a suitable bio-adhesive glue has the additional advantage of preventing drug diffusion into the SPHC; therefore, an unilateral diffusion of drug toward the intestinal epithelium is achieved. It should be mentioned that if a mechanically instable SPH is applied instead of SPHC, the conveyor system will break by the peristaltic pressure of the intestine into many parts and thus will not keep the dosage form long enough at the site of drug absorption. In the presently developed systems for intestinal drug delivery, a new concept of drug release has been taken into account, which is named ‘double phase time-controlled release profile’. It is wellknown that, for the absorption of hydrophilic compounds and macromolecules such as peptides and proteins, initially the luminal and brush border enzymes should be inactivated and the intercellular tight junctions should be opened [5,17]. Therefore, it is necessary to make use of the absorption enhancers and / or enzyme inhibitors at first and, after a specific period of time (ca. 20 min) in which the enzymes have been inactivated and the tight junctions have been opened, the burst release of the active compound should follow. This so-called ‘double phase time-controlled release profile’ is schematically shown in Fig. 8. It should be mentioned that the developed dosage forms are enteric-coated, leading to a lag time for dissolving of the enteric coat layer(s). According to the presented results, such desired

Fig. 8. Schematic view of the concept ‘double phase time-controlled release profile’.

consecutive release / dissolution phases (i.e. dissolving of the enteric-coated layers and gelatine, followed by the rapid swelling of SPHC and simultaneous mechanical fixation, enzyme inhibition and opening of tight junctions, and subsequently finished by the burst release of the (peptidic) drugs) are well possible. The release studies of the core formulations (Figs. 3 and 4) demonstrated that for the core inside the shuttle system, it is preferred to use microparticles instead of gross mass due to faster release of BAEE. On the other hand, for safety reasons it is advisable to use the formulation which consists of the lowest amounts of excipients. In this respect, core formulation 1 in the microparticles form (Fig. 5) is the most appropriate core composition. When using carbomer in the formulation, a delay in the release pattern of BAEE was observed, due to gel forming properties of these polymers. Carbomer forms a viscous jelly solution in an aqueous medium, making it unique for retardation of drug release. However, because for the presently developed dosage form a burst release is required, carbomer is not a suitable excipient. The high standard deviations observed in Fig. 5 are probably due to inhomogeneous dissolution of the core material in gross mass when it was incorporated in the conveyor system. It may also be possible that these cores did not absorb enough water and thus did not dissolve quickly enough in the same amounts of dissolution fluids, due to different porosity of the polymers and speed of water penetration into the SPHC. For the core attached to the surface of shuttle system, the prepared small tablets were appropriate for inserting into the system (Fig. 3B). As demonstrated in Fig. 6 and also reported previously [18,19], by changing the enteric-coat layer and the concentration or thickness of coating layer, it is possible to influence the length of lag time and to modulate the lag time period. Therefore, by adjusting these factors this novel drug delivery system can be targeted to any specific part of the intestine and even to the colon. It is also evident from Fig. 6 that the release of BAEE from the core attached to the surface of shuttle system is started earlier than from the core inside the shuttle system, due to the time period required for swelling of conveyor system and separation of the cap from the shuttle body. The present study has also shown that

F. A. Dorkoosh et al. / Journal of Controlled Release 71 (2001) 307 – 318

SPH and SPHC are able to inhibit to some extent the activity of the intestinal luminal protease trypsin. SPH was found to be more potent than SPHC, most likely due to two factors. Firstly, in the structure of SPH more carboxylic groups are present which can deprive Ca 21 ions from the structure of trypsin and inactivate this enzyme. Secondly, SPH polymers contain larger pores than SPH composites, allowing for a quicker entrapment and inactivation of luminal enzymes [13]. The results of the trypsin inhibition studies are in agreement with the present Ca 21 binding studies and previously published data on Ca 21 -dependent trypsin inhibition by poly(acrylates) [20]. Both SPH and SPHC are able to take up Ca 21 ions either by entrapment or by binding, but SPH is more powerful to bind Ca 21 due to presence of more carboxylic groups in the structure of SPH.

5. Conclusions The choice of an appropriate intestinal drug delivery system depends primarily on the desired site of application for the dosage form. Moreover, it is important that, before releasing of instable and hydrophilic active ingredients such as peptide and protein drugs, the luminal enzymes are inactivated and the tight junctions opened. Therefore, drug delivery systems with ‘double phase time-controlled release profile’ are essential for effective intestinal absorption of such drug molecules. In the present studies novel drug delivery systems based on superporous hydrogels (SPH) and SPHC were designed, which supplies a new mechanism for drug targeting in the intestine. These so-called shuttle systems keep the dosage form at the site of drug absorption by mechanical interaction of the dosage form with the intestinal membranes. The focus of these novel systems is on achieving a ‘double phase time-controlled release profile’ and targeting the dosage form to the desired site(s) of the intestinal tract. The in vitro characteristics of these polymers showed a promising time-controlled release profile for the model compound BAEE. Further studies are in progress for evaluating the absorption of peptide drugs across the intestinal wall ex vivo and in vivo in pigs. In addition, the enteric-coating of

317

the dosage form showed that by changing the type, concentration and thickness of the coating layer, it is well possible to target the dosage form to any specific site of the small intestine (particularly the jejunum and ileum) or to the colon.

References [1] J.A. Fix, Oral controlled release technology for peptides: status and future prospects, Pharm. Res. 13 (1996) 1760– 1764. [2] T. Jung, W. Kamm, A. Breitenbach, E. Kaiserling, J.X. Xiao, T. Kissel, Biodegradable nanoparticles for oral delivery of peptides: is there a role for polymers to affect mucosal uptake?, Eur. J. Pharm. Biopharm. 50 (2000) 147–160. [3] J.P. Bai, L.L. Chang, J.H. Guo, Targeting of peptide and protein drugs to specific sites in the oral route, Crit. Rev. Ther. Drug Carrier Syst. 12 (1995) 339–371. [4] J.A. Fix, Strategies for delivery of peptides utilizing absorption-enhancing agents, J. Pharm. Sci. 85 (1996) 1282– 1285. [5] S.S. Davis, Delivery systems for biopharmaceuticals, J. Pharm. Pharmacol. 44 (Suppl. 1) (1992) 186–190. [6] J.P. Bai, Distribution of brush-border membrane peptidases along the rabbit intestine: implication for oral delivery of peptide drugs, Life Sci. 52 (1993) 941–947. ¨ [7] M. Kratzel, R. Hiessbock, A. Bernkop-Schnurch, Auxiliary agents for the peroral administration of peptide and protein drugs: synthesis and evaluation of novel pepstatin analogues, J. Med. Chem. 41 (1998) 2339–2344. [8] C.-M. Lehr, J.A. Bouwstra, W. Kok, A.G. de Boer, J.J. Tukker, J.C. Verhoef, D.D. Breimer, H.E. Junginger, Effects of the mucoadhesive polymer polycarbophil on the intestinal absorption of a peptide drug in the rat, J. Pharm. Pharmacol. 44 (1992) 402–407. ¨ [9] A. Bernkop-Schnurch, The use of inhibitory agents to overcome the enzymatic barrier to perorally administered therapeutic peptides and proteins, J. Control. Release 52 (1998) 1–16. [10] A.L. Daugherty, R.J. Mrsny, Regulation of the intestinal epithelial paracellular barrier, Pharm. Sci. Technol. Today 2 (1999) 281–287. ¨ [11] T. Trenktrog, B.W. Muller, Preparation and characterization of a peptide containing w / o emulsion, Int. J. Pharm. 123 (1995) 199–207. [12] J.F. Forstner, Intestinal mucins in health and disease, Digestion 17 (1978) 234–263. [13] F.A. Dorkoosh, J. Brussee, J.C. Verhoef, G. Borchard, M. Rafiee-Tehrani, H.E. Junginger, Preparation and NMR characterisation of superporous hydrogels (SPH) and SPH composites, Polymer 41 (2000) 8213–8220. [14] J. Chen, K. Park, Synthesis and characterization of superporous hydrogel composites, J. Control. Release 65 (2000) 73–82.

318

F. A. Dorkoosh et al. / Journal of Controlled Release 71 (2001) 307 – 318

[15] H.E. Junginger, F.A. Dorkoosh, G. Borchard, M. RafieeTehrani, J.C. Verhoef, Double phase time-controlled release system, Patent Application 99203578.2, October 29, 1999. [16] J. Chen, H. Park, K. Park, Synthesis of superporous hydrogels: Hydrogels with fast swelling and superabsobent properties, J. Biomed. Mater. Res. 44 (1999) 53–62. [17] H.J. Lee, G.L. Amidon, Oral peptide delivery: improving the systemic availability of small peptides and enkephalin analogs, NIDA Res. Monogr. 154 (1995) 86–106. [18] T. Ishibashi, H. Hatano, M. Kobayashi, M. Mizobe, H. Yoshino, In vivo drug release behavior in dogs from a new

colon-targeted delivery system, J. Control. Release 57 (1999) 45–53. [19] R. Narayani, K.P. Rao, Polymer-coated gelatin capsules as oral delivery devices and their gastrointestinal tract behaviour in humans, J. Biomater. Sci. Polym. Ed. 7 (1995) 39–48. [20] H.L. Luessen, J.C. Verhoef, G. Borchard, C.-M. Lehr, A.G. de Boer, H.E. Junginger, Carbomer and polycarbophil are potent inhibitors of the intestinal proteolytic enzyme trypsin, Pharm. Res. 12 (1995) 1293–1298.