- Email: [email protected]
Mucoadhesive and cohesive properties of poly(acrylic acid)-cysteine conjugates with regard to their molecular mass
European Journal of Pharmaceutical Sciences 18 (2003) 89–96 www.elsevier.com / locate / ejps
Mucoadhesive and cohesive properties of poly(acrylic acid)-cysteine conjugates with regard to their molecular mass ¨ A. Bernkop-Schnurch ¨ * V.M. Leitner, M.K. Marschutz, Institute of Pharmaceutical Technology and Biopharmaceutics, Centre of Pharmacy, University of Vienna, Althanstr. 14, A-1090 Vienna, Austria Received 4 September 2002; received in revised form 1 November 2002; accepted 7 November 2002
Abstract The objective of this study was to evaluate the influence of the molecular mass and accordingly the polymer chain length on mucoadhesion and cohesion of thiolated polymers. Linear poly(acrylic acid)-cysteine (PAA-Cys) conjugates of 2-, 45-, 250- and 450 kDa (PAA 2 -Cys, PAA 45 -Cys, PAA 250 -Cys and PAA 450 -Cys) and polycarbophil-cysteine (PCP-Cys, 750–3000 kDa), all displaying on average 404.1665.5 mMol thiol groups per gram polymer were compressed into tablets to perform disintegration tests, mucoadhesion studies and viscosity measurements. Moreover, the influence of free unbound cysteine on mucoadhesion was evaluated. Disintegration tests showed a stability of the tablets as following: PAA 2 -Cys,PAA 45 -Cys,PAA 250 -Cys,PAA 450 -Cys5PCP-Cys. According to tensile studies and tests on the rotating cylinder the following rank order in mucoadhesive properties could be established: PAA 2 -Cys,PAA 45 -Cys,PCP-Cys, PAA 250 -Cys,PAA 450 -Cys. Evidence for the formation of disulphide bonds between thiolated polymers and mucin could be provided by the addition of free cysteine resulting in strongly decreased mucoadhesion and by viscosity studies showing comparatively higher viscosity of conjugate / mucin mixtures than of unthiolated polymer / mucin mixtures. The results of the present study contribute to the development of new polymers displaying further improved mucoadhesive properties. 2002 Elsevier Science B.V. All rights reserved. Keywords: Mucoadhesion; Poly(acrylic acid); Molecular mass; Thiomers
1. Introduction In recent years, considerable interest has been shown in the use of mucoadhesive polymers in controlled drug delivery. Mucoadhesive dosage forms should provide an optimum contact with the mucosal surface and should be able to prolong the residence time of the dosage form at the site of drug absorption, such as the oral and nasal cavity or the gastrointestinal tract. These features should help to reduce the need for multiple dosing, resulting in better patient compliance (Mortazavi and Smart, 1994a). The interactions between mucoadhesive polymers and the mucus are so far believed to be based on the formation of non-covalent bonds such as hydrogen bonds, ionic interactions and van der Waals forces or physical interpenetration effects of polymer chains and mucus (Peppas and Buri, 1985; Chickering III and Mathiowitz, 1999). *Corresponding author. Tel.: 143-142-775-5413; fax: 143-142-779554. E-mail address: [email protected] (A. Bernkop¨ Schnurch).
While these interactions are comparatively weak, covalent bonds are much stronger and not any more influenced by factors such as ionic strength and pH. Functional groups that are able to form covalent bonds are overall thiol groups. Recently, our research group has generated a promising new class of multifunctional polymers with thiol-bearing side chains, the thiolated polymers or thiom¨ ers (Bernkop-Schnurch et al., 1999). In contrast to traditionally used mucoadhesive polymers, their thiol groups are believed to form strong covalent bonds with cysteinerich subdomains of mucus glycoproteins (Gum et al., 1992) resulting in enhanced mucoadhesive properties. Thiomers are synthesised by immobilising thiol moieties on well established hydrophilic polymers such as poly¨ (acrylates) (Bernkop-Schnurch and Steininger, 2000), ¨ chitosan (Kast and Bernkop-Schnurch, 2001) or alginate ¨ (Bernkop-Schnurch et al., 2001a). Apart from their strongly improved mucoadhesive properties, thiomers have also ¨ permeation enhancing (Clausen and Bernkop-Schnurch, 2000) and enzyme inhibitory properties (Bernkop¨ Schnurch et al., 2001b), which render them useful excipients particularly for the noninvasive application of peptide
0928-0987 / 02 / $ – see front matter 2002 Elsevier Science B.V. All rights reserved. PII: S0928-0987( 02 )00245-2
V.M. Leitner et al. / European Journal of Pharmaceutical Sciences 18 (2003) 89–96
90
drugs. Calcitonin, for example, was incorporated in an oral carrier system based on thiolated chitosan resulting in about a 10% decrease of the plasma calcium level of rats for at least 10 h, whereas no effect was observed by using the same delivery system with unmodified chitosan (Guggi et al., 2002). To improve the already promising properties of so far developed thiomers, a better understanding of the mechanisms involved in mucoadhesion is required. The main physical mechanism of mucoadhesion is believed to be based on chain flexibility (Peppas and Buri, 1985). Flexible polymer chains favour interpenetration between polymer chains and mucus to a sufficient depth to create a strong adhesive bond. Crosslinking or the covalent attachment of large sized ligands leads to a reduction in chain flexibility and results in a strong decrease in mucoadhesion ¨ (Bernkop-Schnurch, 2000). Therefore, non-crosslinked polymers of low molecular mass displaying a higher chain flexibility may exhibit improved mucoadhesive properties. According to this theory, it was the objective of this study to evaluate the influence of the molecular mass of the polymer and accordingly the polymer chain length on mucoadhesion of thiomers. Polymers used for the study were linear poly(acrylic acid)-cysteine (PAA-Cys) conjugates of 2-, 45-, 250- and 450 kDa (PAA 2 -Cys, PAA 45 Cys, PAA 250 -Cys and PAA 450 -Cys) and a polycarbophilcysteine (PCP-Cys) conjugate, which is a poly(acrylate) of 750–3000 kDa being crosslinked via divinylglycol. All polymers exhibited comparable amounts of immobilised thiol groups. The disintegration and swelling behaviour of polymer tablets was determined and mucoadhesion studies with the rotating cylinder method as well as tensile studies were performed. Finally, the influence of free unbound cysteine on mucoadhesion was investigated and polymer / mucin mixtures were evaluated rheologically as the viscosity of polymer / mucin mixtures is believed to be an indicator for mucoadhesive strength (Hassan and Gallo, 1990).
water at 4 8C overnight and the pH of the polymer solutions were adjusted to 5 with 5 M HCl. They were dialysed according to a method previously described for ¨ PCP-Cys (Bernkop-Schnurch and Steininger, 2000). Control polymers without immobilised cysteine were prepared and purified in the same way. After dialysis, the pH of all samples were adjusted to 5 with 1 M NaOH and the frozen polymer solutions were dried by lyophilisation at 230 8C and 0.01 mbar (Christ Beta 1–8 K; Osterode am Harz, Germany). All polymers were stored at 4 8C until further use.
2.3. Determination of the thiol group content The amount of thiol groups immobilised on the polymer-cysteine conjugates was determined spectrophotometrically using Ellman’s reagent (Ellman, 1959) as described ¨ previously (Bernkop-Schnurch et al., 1999).
2.4. Evaluation of unbound cysteine in the polymer The amount of unconjugated cysteine in the dialysed and lyophilised polymer was quantified using the TNBS reagent (2,4,6-Trinitrobenzenesulfonic acid, Sigma, St. Louis, MO, USA). TNBS reacts with the primary amino groups of cysteine in a nucleophilic aromatic substitution developing an orange dye. Polymers (5 mg) were swollen in 0.5 ml of demineralised water containing 1% (m / v) NaCl to reduce the viscosity. A 0.1-ml aliquot of this solution was transferred to a 96-well microtitration plate and incubated with an equal volume of 0.1% (w / v) TNBS in 8% (w / v) NaHCO 3 at 37 8C for 2 h. The absorbance was measured at 450 nm. The amount of free amino groups was calculated using a standard curve of cysteine over a concentration range of 1–30 mg / ml prepared in 1% (m / v) aqueous NaCl solution.
2.5. Preparation of tablets 2. Materials and methods
2.1. Materials Poly(acrylic acid) (PAA) of 2-, 45-, 250-, 450 kDa and hydrochloride hydrate were purchased from Sigma-Aldrich, Steinheim, Germany; polycarbophil (PCP, mol. mass.700 kDa; Noveon AA1) from BF Goodrich, Brecksville, OH, USA. Poly(acrylic acid)-cysteine conjugates of 2-, 45-, 250-, 450 kDa (PAA 2 -Cys, PAA 45 -Cys, PAA 250 -Cys and PAA 450 -Cys) and the polycarbophil-cysteine conjugate (PCP-Cys) were supplied by MucoBiomer (Leobendorf, Austria).
Lyophilised PAA-Cys and PCP-Cys conjugates and controls were compressed (Hanseaten Type El, Hamburg, Germany) into 5.0 mm diameter flat-faced tablets of 30 mg. The compaction pressure was kept constant during the preparation of all tablets.
L-cysteine
2.2. Preparation of polymer-cysteine conjugates and controls All polymers were completely hydrated in demineralised
2.6. Disintegration studies The disintegration behaviour of the tablets in 0.1 M phosphate buffer pH 6.8 at 37 8C was analysed with a disintegration test apparatus according to the European Pharmacopeia. The oscillating frequency was adjusted to 0.5 s 21 .
2.7. Evaluation of the swelling behaviour The water absorbing capacity was determined by a gravimetric method. Tablets were fixed to the needle of a
V.M. Leitner et al. / European Journal of Pharmaceutical Sciences 18 (2003) 89–96
syringe and placed in a beaker with 0.1 M phosphate buffer pH 6.8 at 37 8C. At various time intervals the hydrated tablets were taken out of the incubation medium, excess water was removed and the amount of water uptake was determined gravimetrically (Kast and Bernkop¨ Schnurch, 2001).
2.8. In vitro evaluation of the adhesive properties 2.8.1. Tensile studies Tensile studies were carried out on native porcine intestinal mucosa. A tablet was glued to a stainless steel flat disc (5 mm in diameter), which was attached to a laboratory stand with a nylon thread (15 cm). The porcine mucosa was fixed to a glass support using a cyanoacrylate adhesive, placed in a beaker and completely immersed with 0.1 M phosphate-buffered saline pH 6.8. The beaker was placed on a balance and carefully raised by a mobile platform until the mucosa came in contact with the tablet. After an incubation time of 30 min at 25 8C, the mucosa was pulled down from the tablet at a rate of 0.1 mm / s. Data points were collected every second by a personal computer (WINWEDGE software; TAL Technologies, Philadelphia, PA, USA) connected to the balance. The total work of adhesion (TWA) representing the area under the force / distance curve and the maximum detachment force (MDF) were calculated with EXCEL 97 (Microsoft, USA) ¨ (Kast and Bernkop-Schnurch, 2001). 2.8.2. In vitro mucoadhesion studies with the rotating cylinder method Polymer tablets were attached to freshly excised intestinal porcine mucosa, which has been attached to a stainless steel cylinder (diameter: 4.4 cm, height 5.1 cm; apparatus 4-cylinder, USP XXII) using a cyanoacrylate adhesive (Loctite, Henkel, Austria). The cylinder was placed in the dissolution apparatus according to the USP, fully immersed with 0.1 M phosphate buffer pH 6.8 at 37 8C and agitated with 125 rpm. The detachment, disintegration and / or erosion of test tablets was monitored over a 24-h time ¨ period (Bernkop-Schnurch and Steininger, 2000). Additionally, the influence of free unbound cysteine in the polymer tablet on mucoadhesion was evaluated. Cysteine (2.5%, w / w) was mixed with hydrated PAA 450 -Cys or unthiolated control polymer and the pH was adjusted to 5 with 1 M NaOH. Polymers were lyophilised, compressed into tablets and the mucoadhesion time was determined.
91
PAA-Cys conjugates and controls were hydrated in demineralised water to give a concentration of 6% (m / v). The polymer solutions were added to an equal volume of mucin stock solution, mixed with a spatula and the pH of the mixture was adjusted to 6.8 with 2 M NaOH. After incubation for 20 min at room temperature, 0.8 ml of polymer / mucin incubates were transferred to a cone-plate viscometer (RotoVisco RT20, Haake GmbH, Karlsruhe, Germany) and allowed to equilibrate on the plate for 3 min at 20 8C60.5 8C. Then, dynamic oscillatory tests within the linear viscoelasticity region were performed at 1 Hz ¨ and Bernkop-Schnurch, ¨ frequency (Marschutz 2002). The parameters obtained thereby were the phase angle (d ) and the shear deformation (g ). The storage modulus (G9), the loss modulus (G0) and the complex viscosity (uh *u) were calculated by the following equations: G9 5 (t /g ) ? cos d G0 5 (t /g ) ? sin d ]]]] uh *u 5œ(h 9)2 1 (h 0)2 (whereas h 9 5 G0 /v, and h 0 5 G9 /v ) where t is the shear stress, v is the angular frequency which is related to the oscillatory frequency y by the relationship v 5 2 p y.
2.10. Statistical data analysis Statistical data analyses were performed using the Mann–Whitney U-test with P,0.05 as the minimal level of significance. Calculations were done using the software Xlstat version 5.0 (b8.3).
3. Results
3.1. Characterisation of the PAA-Cys conjugates The chemical substructure of the PAA-Cys conjugates used in this study is illustrated in Fig. 1. Determination of the thiol groups attached to the polymer by the Ellman’s
2.9. Rheological studies of polymer /mucin mixtures Porcine gastric mucin (2 g, Type II: crude, Sigma, St. Louis, MO, USA) was hydrated in 12.5 ml of demineralised water under continuous stirring at 4 8C overnight. The mucin solution was adjusted to pH 6.8 with 1 M NaOH and diluted to a final volume of 25 ml with 0.1 M phosphate buffer pH 6.8. The mucin stock solution (8% m / v) was stored at 4 8C no longer than 24 h.
Fig. 1. Chemical substructure of poly(acrylic acid)-cysteine conjugates.
92
V.M. Leitner et al. / European Journal of Pharmaceutical Sciences 18 (2003) 89–96
test demonstrated that on average 404.1665.5 mMol thiol groups were immobilised per gram polymer. Remaining traces of unbound cysteine after dialysis were determined with TNBS-reagent to be less than 0.5% (m / m) of the total mass of the polymer. The lyophilised polymer-cysteine conjugates and control polymers appeared as white, odourless powder of fibrous structure. They were easily hydratable in aqueous solutions forming thereby solutions of initially low viscosity.
3.2. Disintegration studies Disintegration studies demonstrated for all polymer tablets comprising a conjugate or control polymer that the disintegration time increased with increasing molecular mass of the polymers (Fig. 2). Apart from the PAA 2 -Cys matrix tablets, thiolated polymer tablets displayed a significantly (P , 0.05) higher stability than control tablets. Moreover, with increasing molecular mass of the polymers, the difference in the disintegration time between unthiolated and thiolated polymer matrix tablets became more pronounced. In particular matrix tablets of PAA 450 Cys and PCP-Cys were stable for more than 2 days and no erosion could be observed over this time period.
3.3. Evaluation of the swelling behaviour The swelling behaviour of mucoadhesive polymers exerts a great influence on their adhesive properties, drug release and stability (Mortazavi and Smart, 1993). The extent and rate of swelling are affected by the degree of crosslinking and chain length of the polymers (Smart, 1999). Mucoadhesive polymers are believed to take water from the underlying mucosal tissue by absorbing, swelling and capillary effects leading to considerably strong adhesion (Duchene and Ponchel, 1992). However, such ‘adhesion by hydration’ usually ceases when swelling continues and an excessive water uptake transforms the mucoadhesive dosage form into an over-hydrated slippery mucilage (Lehr, 1996). In order to evaluate the correlation between the swelling behaviour and mucoadhesion, water uptake studies were carried out with polymer and control tablets in 0.1 M phosphate buffer pH 6.8 at 37 8C. The experiments were only performed at this physiological pH as the buffer capacity within the polymer matrix is sufficiently high that the pH of the surrounding medium exerts only little influence on the swelling behaviour (data not shown). The swelling behaviour of tablets containing the PAA 2 -Cys and the PAA 45 -Cys conjugate as well as their corresponding controls could not be determined as they dissolved in the buffer within 10 min. Results of this study are shown in Fig. 3A. The initial weight of the PAA 250 Cys, PAA 450 -Cys and PCP-Cys matrix tablets (30 mg) increased 22.9-, 15.2- and 18.2-fold, respectively, over a time period of 90 min. The tablets appeared as highly cohesive and fully transparent. Fig. 3B shows the swelling behaviour of PAA 450 and PCP control tablets. For the PAA 250 control polymer, the swelling behaviour could not be determined because of rapid dissolution. After 30 min, the weight of the unthiolated PAA 450 tablets decreased due to erosion of the tablets.
3.4. Mucoadhesive properties of the PAA-Cys and PCPCys conjugates
Fig. 2. Disintegration behaviour of matrix tablets (30 mg, 5 mm I.D.) based on PAA-Cys conjugates (black bars) or corresponding unmodified polymers (grey bars). Studies were carried out with a disintegration test apparatus (Pharm. Eur.) in 0.1 M phosphate buffer pH 6.8 at 37 8C. PAA 450 -Cys and PCP-Cys matrix tablets did not disintegrate within an observation period of 48 h. Indicated values are means (6S.D.) of three experiments.
Mucoadhesion studies were performed in order to evaluate the influence of the molecular mass on mucoadhesion using two different test methods. On the one hand, tensile studies were carried out on porcine intestinal mucosa. Results in Fig. 4 show a strong influence of the molecular mass and crosslinking of the polymer tablets on their adhesive properties. The maximum detachment force (MDF) thereby did not increase with increasing total work of adhesion (TWA) but was in almost the same range for all polymers (data not shown). On the other hand, in order to confirm the results by another mucoadhesion test system, mucoadhesion tests were carried out on the rotating cylinder. Thereby obtained results, as shown in Fig. 5, were in good accordance with the results of tensile studies. Due to the immobilisation of thiol groups on poly(acrylic acid) the mucoadhesive prop-
V.M. Leitner et al. / European Journal of Pharmaceutical Sciences 18 (2003) 89–96
93
Fig. 4. Influence of the molecular mass on mucoadhesive properties of PAA-Cys conjugates (black bars) and the corresponding unthiolated polymers (grey bars). Represented values are means (6S.D., n 5 3–6) of the total work of adhesion (TWA) determined via tensile studies at pH 6.8 with tablets of indicated test material. *, differs from according unthiolated polymer, P , 0.01.
Additionally, the influence of free unbound cysteine on mucoadhesion of PAA 450 -Cys matrix tablets was evaluated. Results are shown in Fig. 6. The addition of free cysteine to the polymers had only an effect on the mucoadhesion behaviour of thiolated polymers, but not on the unthiolated polymer control.
3.5. Rheological evaluation of polymer /mucin mixtures
Fig. 3. A. Swelling behaviour of PAA 250 -Cys (O), PAA 450 -Cys (3) and PCP-Cys (m) tablets in 100 mM phosphate buffer pH 6.8 at 37 8C. B. Swelling behaviour of controls tablets based on PAA 450 (3) and PCP (m). All indicated values are means (6S.D.) of at least three experiments.
erties of the non-crosslinked polymers PAA 45 -Cys, PAA 250 -Cys and PAA 450 -Cys conjugates were 1000-, 93and 16-fold improved, respectively, while the crosslinked polymer PCP-Cys exceeded its control only 1.7-fold.
The viscosity of polymer / mucin mixtures is the net result of the resistance to flow exerted by individual chain segments, physical chain entanglements and (non)covalent intermolecular interactions, which are the same as the interactions involved in the process of mucoadhesion. Therefore, it has been suggested that the forces of interactions involved in a mucoadhesive system could be evaluated by viscosity measurements (Hassan and Gallo, 1990). According to this, the viscoelastic properties of the PAACys and PCP-Cys conjugates and controls were determined after incubation with mucin. Viscosity studies were performed with commercially available mucin instead of native mucus resulting in more reproducible and comparable results. In addition, orientating studies demonstrated that there is no significant difference in the results obtained with native mucus and hydrated mucin (data not shown).
94
V.M. Leitner et al. / European Journal of Pharmaceutical Sciences 18 (2003) 89–96
Fig. 6. Mucoadhesive behaviour of matrix tablets with and without free unbound cysteine determined via the rotating cylinder method. The black bars represent the time of mucoadhesion of PAA 450 -Cys tablets, the grey bars the mucoadhesion time of PAA 450 control tablets with and without the addition of 2.5% free cysteine (means6S.D., n 5 3). *, differs from according polymer without L-Cys, P , 0.05. Fig. 5. Influence of the molecular mass on mucoadhesive properties of PAA-Cys conjugates (black bars) and corresponding controls (grey bars). Test tablets of indicated polymers were attached to excised porcine mucosa, which has been spanned on a cylinder and agitated at 125 rpm in 0.1 M phosphate buffer pH 6.8 at 37 8C. The indicated time of adhesion represents the mean (6S.D.) of at least three experiments. *, differs from control, P , 0.05.
With increasing molecular mass of the polymers, storage modulus G9 and loss modulus G0 increased from 0.3260.11 / 2.9560.44 (G9 /G0) for PAA 250 -Cys to 727.776174.70 / 188.20667.94 (G9 /G0) for PCP-Cys. As depicted in Fig. 7, PAA 250 -Cys / mucin, PAA 450 -Cys / mucin and PCP-Cys / mucin mixtures showed significantly (P , 0.05) increased viscosity compared to unthiolated polymer / mucin mixtures; for PAA 250 -Cys / mucin the improvement was even more than 6-fold. The rheological properties of PAA 2 -Cys / mucin and PAA 45 -Cys / mucin mixtures and their controls could not be detected as they were too low to be determined with the available equipment. The mucin dispersion itself did not exhibit any detectable viscoelastic properties at the concentration employed (4% w / v), as well.
4. Discussion The molecular mass—correlating with the molecular chain length—determines the strength of mucoadhesive bonds, as molecular mobility and flexibility affect the ability of polymer and mucus to form entanglements. It is believed that an optimally high molecular mass corre-
Fig. 7. Influence of the molecular mass on the viscosity of PAA-Cys / mucin mixtures (black bars) and corresponding control / mucin mixtures (grey bars). Oscillatory measurements at 1 Hz frequency were carried out immediately after an incubation period of 20 min at room temperature. Indicated values are means (6S.D.) of at least three experiments.
V.M. Leitner et al. / European Journal of Pharmaceutical Sciences 18 (2003) 89–96
sponds to maximum adhesiveness (Caramella et al., 1999). However, if the polymer chains are too long, their ability to diffuse into the mucus gel is limited (Huang et al., 2000). According to this theory, it was the objective of this study to evaluate the influence of the molecular mass and chain length of thiomers on their mucoadhesive and cohesive properties. An important factor for mucoadhesion are the cohesive properties in conjunction with the swelling behaviour of polymers (Mortazavi and Smart, 1993). In contrast to the theory that rapidly swelling polymers will also quickly interact with the mucin (Mortazavi and Smart, 1993), a correlation between the swelling behaviour of investigated polymers and their mucoadhesive properties could not be found. The water uptake of PAA 250 -Cys tablets was the comparatively highest, whereas its mucoadhesive properties were lower than those of PAA 450 Cys tablets. These results are in good agreement with ¨ previous published studies (Bernkop-Schnurch and Steininger, 2000). Furthermore, a direct correlation between the molecular mass of the polymers and their swelling behaviour was not observed. Tensile studies as well as mucoadhesion studies with the rotating cylinder method showed that among the different PAA-Cys conjugates tested, the PAA 450 -Cys conjugate displayed the strongest mucoadhesive properties. The thiolated PAA 450 polymer used for this study exhibited even the highest TWA so far determined in comparison to other thiolated polymers tested under the same conditions. On the rotating cylinder, PAA 450 -Cys matrix tablets remained attached to porcine mucosa for about 16-times longer than corresponding control tablets, while PCP-Cys matrix tablets exceeded their controls only 1.7 times. These results strongly support the theory that non-crosslinked polymers display higher chain flexibility (Duchene and Ponchel, 1992) and therefore are able to interpenetrate the mucus layer more deeply resulting in enhanced mucoadhesion. Within this study, thiomers of comparatively low molecular mass were investigated for their mucoadhesive properties according to the assumption that their smaller and therefore more flexible chains would favour even stronger interpenetration and chain entanglements with the mucus resulting in improved mucoadhesion. However, mucoadhesion, disintegration and swelling tests of PAA 2 -Cys, PAA 45 -Cys and PAA 250 -Cys tablets resulted in decreased adhesive and cohesive properties of the tablets with decreasing molecular mass of the conjugates. This could be explained by too low intermolecular cohesive forces. The small molecules bearing only a few thiol groups were not able to form a highly cohesive three-dimensional network via the formation of intermolecular disulphide bonds. An important factor for mucoadhesion and cohesion of thiolated polymers is their ability to form covalent bonds within the polymer itself as well as between the polymer ¨ and the mucus layer (Bernkop-Schnurch and Steininger,
95
2000). The formation of disulphide bonds within the polymer resulted in much higher stability of thiomer tablets compared to control tablets, as shown in disintegration tests. This high cohesiveness of thiolated polymer matrix tablets resulted in an almost zero order release of calcitonin for at least 8 h as described recently by our research group (Guggi et al., 2002). The formation of covalent bonds between thiomer and mucus was also investigated by incubating commercially available mucin with a thiolated polymer and the corresponding unmodified polymer. While the mucin could be completely removed from the unmodified polymer / mucin mixture by centrifugation and removing the supernatant, it remained bound to the thiomer and could only be washed out by the addition of the disulphide bond breaker dithiothreitol (Bernkop¨ Schnurch et al., 1999). Within this work, additional evidence for covalent bonds between thiomer and mucus could be provided by the addition of free cysteine to PAA 450 -Cys. As cysteine is also known to be a disulphide bond breaking agent (Lightowler and Lightowler, 1971) and its addition to the polymer led to a massive reduction in mucoadhesion, it may be concluded that disulphide bonds between polymer and mucus are responsible for the enhanced mucoadhesive properties of thiomers. Similar results were obtained by Mortazavi, who reported a significant reduction in mucoadhesive strength of Carbopol 934P in the presence of highly concentrated hydrogen bond breakers such as KCNS and urea. It was concluded that the formation of hydrogen bonds between polymer and mucus was an important component for mucoadhesion (Mortazavi, 1995). The results obtained from viscosity studies showed that the viscosity of the polymer / mucin mixtures and controls increased with increasing molecular mass. As expected and previously reported by Mortazavi and Smart (1994b), the shorter chain length polymers with inherently poor gel forming properties were found to be less effective in promoting gel strengthening. The much higher viscosity of conjugate / mucin mixtures compared to controls can be explained by the formation of disulphide bonds between the thiolated polymer and mucin leading to a strengthened gel network, as mentioned above.
5. Conclusion This study compares the mucoadhesive and cohesive properties of five different poly(acrylic acid)-cysteine conjugates with a molecular mass in the range of 2 kDa– 1–3 million kDa. Among all conjugates tested, the PAA 450 -Cys conjugate displayed the most favourable properties concerning mucoadhesion and cohesion of matrix tablets. Results even demonstrate that in comparison to all other tested mucoadhesive polymers, this conjugate exhibits the highest total work of adhesion and accordingly mucoadhesion ever reached. Additionally, the
96
V.M. Leitner et al. / European Journal of Pharmaceutical Sciences 18 (2003) 89–96
polymer tablets showed an excellent stability comparable to PCP-Cys tablets being the most stable ones described so far. The findings of this study also contribute valuable basic information about the mechanisms involved in mucoadhesion of thiomers. Low molecular mass polymers with flexible chains favouring strong interpenetration are not cohesive enough for optimal mucoadhesion, whereas high molecular weight crosslinked polymers showing high cohesiveness do not display enough chain flexibility. The optimal mucoadhesive polymer represents therefore a compromise of these two extremes. In addition, this study provides further evidence for the formation of covalent bonds between thiolated polymers and mucin glycoproteins as mucoadhesion is strongly decreased when a disulphide bond breaking agent is added to the polymer. The present study gives an extensive characterisation of a broad range of thiolated poly(acrylic acids) and should therefore facilitate the development of new drug delivery systems providing controlled drug release and a greatly prolonged residence time on various mucosal tissues.
Acknowledgements This work was supported by Grant No. P15373-MOB ¨ from the Fonds zur Forderung der wissenschaftlichen ¨ Forschung (FWF) to A. Bernkop-Schnurch. ¨ and co-workers The authors wish to thank Mr. Strobl from the slaughterhouse Totzenbach for the supply of porcine intestinal mucosa.
References ¨ Bernkop-Schnurch, A., Schwarz, V., Steininger, S., 1999. Polymers with thiol groups: a new generation of mucoadhesive polymers? Pharm. Res. 16, 876–881. ¨ Bernkop-Schnurch, A., 2000. Mucoadhesive polymers. In: Dumitriu, S. (Ed.), Polymeric Biomaterials. Marcel Dekker, New York, pp. 147– 165. ¨ Bernkop-Schnurch, A., Steininger, S., 2000. Synthesis and characterisation of mucoadhesive thiolated polymers. Int. J. Pharm. 194, 239–247. ¨ Bernkop-Schnurch, A., Kast, C.E., Richter, M.F., 2001a. Improvement in the mucoadhesive properties of alginate by the covalent attachment of cysteine. J. Control. Rel. 71, 277–285. ¨ Bernkop-Schnurch, A., Zarti, H., Walker, G.F., 2001b. Thiolation of polycarbophil enhances its inhibition of soluble and intestinal brush border membrane bound aminopeptidase N. J. Pharm. Sci. 90, 1907– 1914.
Caramella, C.M., Rossi, S., Bonferoni, M.C., 1999. A rheological approach to explain the mucoadhesive behavior of polymer hydrogels. Drugs Pharm. Sci. 98, 25–65. Chickering III, D.E., Mathiowitz, E., 1999. Definitions, mechanisms, and theories of bioadhesion. Drugs Pharm. Sci. 98, 1–10. ¨ Clausen, A.E., Bernkop-Schnurch, A., 2000. In vitro evaluation of the permeation-enhancing effect of thiolated polycarbophil. J. Pharm. Sci. 89, 1253–1261. Duchene, D., Ponchel, G., 1992. Principle and investigation of the bioadhesion mechanism of solid dosage forms. Biomaterials 13, 709– 714. Ellman, G.L., 1959. A colorimetric method for determining low concentrations of mercaptans. Arch. Biochem. Biophys. 74, 443–450. ¨ Guggi, D., Kast, C.E., Bernkop-Schnurch, A., 2002. In vivo evaluation of an oral calcitonin delivery system for rats based on a thiolated chitosan matrix, Proceed. 11th Inter. Pharm. Technol. Symp., pp. 41–42. Gum, J.R.J., Hicks, J.W., Toribara, N.W., Rothe, E.-M., Lagace, R.E., Kim, Y.S., 1992. The human MUC2 intestinal mucin has cysteine-rich subdomains located both upstream and downstream of its central repetitive region. J. Biol. Chem. 267, 21375–21383. Hassan, E.E., Gallo, J.M., 1990. A simple rheological method for the in vitro assessment of mucin-polymer bioadhesive bond strength. Pharm. Res. 7, 491–495. Huang, Y., Leobandung, W., Foss, A., Peppas, N.A., 2000. Molecular aspects of muco- and bioadhesion: tethered structures and site-specific surfaces. J. Control. Rel. 65, 63–71. ¨ Kast, C.E., Bernkop-Schnurch, A., 2001. Thiolated polymers–thiomers: development and in vitro evaluation of chitosan-thioglycolic acid conjugates. Biomaterials 22, 2345–2352. Lehr, C.M., 1996. From sticky stuff to sweet receptors — achievements, limits and novel approaches to bioadhesion. Eur. J. Drug Metab. Pharmacokinet. 21, 139–148. Lightowler, J.E., Lightowler, N.M., 1971. Comparative mucolytic studies on dithiothreitol, N-acetylcysteine and L-cysteine on human respiratory mucus in vitro and their effects on the rate of flow of mucus in the exposed trachea of the rat on topical application. Arch. Int. Pharmacodyn.Ther. 189, 53–58. ¨ M.K., Bernkop-Schnurch, ¨ Marschutz, A., 2002. Thiolated polymers: selfcrosslinking properties of thiolated 450 kDa poly(acrylic acid) and their influence on mucoadhesion. Eur. J. Pharm. Sci. 15, 387–394. Mortazavi, S.A., Smart, J.D., 1993. An investigation into the role of water movement and mucus gel dehydration in mucoadhesion. J. Control. Rel. 25, 197–203. Mortazavi, S.A., Smart, J.D., 1994a. An in vitro method for assessing the duration of mucoadhesion. J. Control. Rel. 31, 207–212. Mortazavi, S.A., Smart, J.D., 1994b. Factors influencing gel-strengthening at the mucoadhesive-mucus interface. J. Pharm. Pharmacol. 46, 86–90. Mortazavi, S.A., 1995. An in vitro assessment of mucus / mucoadhesive interactions. Int. J. Pharm. 124, 173–182. Peppas, N.A., Buri, P.A., 1985. Surface, interfacial and molecular aspects of polymer bioadhesion on soft tissues. J. Control. Rel. 2, 257–275. Smart, J.D., 1999. The role of water movement and polymer hydration in mucoadhesion. Drugs Pharm. Sci. 98, 11–23.
Mucoadhesive and cohesive properties of poly(acrylic acid)-cysteine conjugates with regard to their molecular mass ¨ A. Bernkop-Schnurch ¨ * V.M. Leitner, M.K. Marschutz, Institute of Pharmaceutical Technology and Biopharmaceutics, Centre of Pharmacy, University of Vienna, Althanstr. 14, A-1090 Vienna, Austria Received 4 September 2002; received in revised form 1 November 2002; accepted 7 November 2002
Abstract The objective of this study was to evaluate the influence of the molecular mass and accordingly the polymer chain length on mucoadhesion and cohesion of thiolated polymers. Linear poly(acrylic acid)-cysteine (PAA-Cys) conjugates of 2-, 45-, 250- and 450 kDa (PAA 2 -Cys, PAA 45 -Cys, PAA 250 -Cys and PAA 450 -Cys) and polycarbophil-cysteine (PCP-Cys, 750–3000 kDa), all displaying on average 404.1665.5 mMol thiol groups per gram polymer were compressed into tablets to perform disintegration tests, mucoadhesion studies and viscosity measurements. Moreover, the influence of free unbound cysteine on mucoadhesion was evaluated. Disintegration tests showed a stability of the tablets as following: PAA 2 -Cys,PAA 45 -Cys,PAA 250 -Cys,PAA 450 -Cys5PCP-Cys. According to tensile studies and tests on the rotating cylinder the following rank order in mucoadhesive properties could be established: PAA 2 -Cys,PAA 45 -Cys,PCP-Cys, PAA 250 -Cys,PAA 450 -Cys. Evidence for the formation of disulphide bonds between thiolated polymers and mucin could be provided by the addition of free cysteine resulting in strongly decreased mucoadhesion and by viscosity studies showing comparatively higher viscosity of conjugate / mucin mixtures than of unthiolated polymer / mucin mixtures. The results of the present study contribute to the development of new polymers displaying further improved mucoadhesive properties. 2002 Elsevier Science B.V. All rights reserved. Keywords: Mucoadhesion; Poly(acrylic acid); Molecular mass; Thiomers
1. Introduction In recent years, considerable interest has been shown in the use of mucoadhesive polymers in controlled drug delivery. Mucoadhesive dosage forms should provide an optimum contact with the mucosal surface and should be able to prolong the residence time of the dosage form at the site of drug absorption, such as the oral and nasal cavity or the gastrointestinal tract. These features should help to reduce the need for multiple dosing, resulting in better patient compliance (Mortazavi and Smart, 1994a). The interactions between mucoadhesive polymers and the mucus are so far believed to be based on the formation of non-covalent bonds such as hydrogen bonds, ionic interactions and van der Waals forces or physical interpenetration effects of polymer chains and mucus (Peppas and Buri, 1985; Chickering III and Mathiowitz, 1999). *Corresponding author. Tel.: 143-142-775-5413; fax: 143-142-779554. E-mail address: [email protected] (A. Bernkop¨ Schnurch).
While these interactions are comparatively weak, covalent bonds are much stronger and not any more influenced by factors such as ionic strength and pH. Functional groups that are able to form covalent bonds are overall thiol groups. Recently, our research group has generated a promising new class of multifunctional polymers with thiol-bearing side chains, the thiolated polymers or thiom¨ ers (Bernkop-Schnurch et al., 1999). In contrast to traditionally used mucoadhesive polymers, their thiol groups are believed to form strong covalent bonds with cysteinerich subdomains of mucus glycoproteins (Gum et al., 1992) resulting in enhanced mucoadhesive properties. Thiomers are synthesised by immobilising thiol moieties on well established hydrophilic polymers such as poly¨ (acrylates) (Bernkop-Schnurch and Steininger, 2000), ¨ chitosan (Kast and Bernkop-Schnurch, 2001) or alginate ¨ (Bernkop-Schnurch et al., 2001a). Apart from their strongly improved mucoadhesive properties, thiomers have also ¨ permeation enhancing (Clausen and Bernkop-Schnurch, 2000) and enzyme inhibitory properties (Bernkop¨ Schnurch et al., 2001b), which render them useful excipients particularly for the noninvasive application of peptide
0928-0987 / 02 / $ – see front matter 2002 Elsevier Science B.V. All rights reserved. PII: S0928-0987( 02 )00245-2
V.M. Leitner et al. / European Journal of Pharmaceutical Sciences 18 (2003) 89–96
90
drugs. Calcitonin, for example, was incorporated in an oral carrier system based on thiolated chitosan resulting in about a 10% decrease of the plasma calcium level of rats for at least 10 h, whereas no effect was observed by using the same delivery system with unmodified chitosan (Guggi et al., 2002). To improve the already promising properties of so far developed thiomers, a better understanding of the mechanisms involved in mucoadhesion is required. The main physical mechanism of mucoadhesion is believed to be based on chain flexibility (Peppas and Buri, 1985). Flexible polymer chains favour interpenetration between polymer chains and mucus to a sufficient depth to create a strong adhesive bond. Crosslinking or the covalent attachment of large sized ligands leads to a reduction in chain flexibility and results in a strong decrease in mucoadhesion ¨ (Bernkop-Schnurch, 2000). Therefore, non-crosslinked polymers of low molecular mass displaying a higher chain flexibility may exhibit improved mucoadhesive properties. According to this theory, it was the objective of this study to evaluate the influence of the molecular mass of the polymer and accordingly the polymer chain length on mucoadhesion of thiomers. Polymers used for the study were linear poly(acrylic acid)-cysteine (PAA-Cys) conjugates of 2-, 45-, 250- and 450 kDa (PAA 2 -Cys, PAA 45 Cys, PAA 250 -Cys and PAA 450 -Cys) and a polycarbophilcysteine (PCP-Cys) conjugate, which is a poly(acrylate) of 750–3000 kDa being crosslinked via divinylglycol. All polymers exhibited comparable amounts of immobilised thiol groups. The disintegration and swelling behaviour of polymer tablets was determined and mucoadhesion studies with the rotating cylinder method as well as tensile studies were performed. Finally, the influence of free unbound cysteine on mucoadhesion was investigated and polymer / mucin mixtures were evaluated rheologically as the viscosity of polymer / mucin mixtures is believed to be an indicator for mucoadhesive strength (Hassan and Gallo, 1990).
water at 4 8C overnight and the pH of the polymer solutions were adjusted to 5 with 5 M HCl. They were dialysed according to a method previously described for ¨ PCP-Cys (Bernkop-Schnurch and Steininger, 2000). Control polymers without immobilised cysteine were prepared and purified in the same way. After dialysis, the pH of all samples were adjusted to 5 with 1 M NaOH and the frozen polymer solutions were dried by lyophilisation at 230 8C and 0.01 mbar (Christ Beta 1–8 K; Osterode am Harz, Germany). All polymers were stored at 4 8C until further use.
2.3. Determination of the thiol group content The amount of thiol groups immobilised on the polymer-cysteine conjugates was determined spectrophotometrically using Ellman’s reagent (Ellman, 1959) as described ¨ previously (Bernkop-Schnurch et al., 1999).
2.4. Evaluation of unbound cysteine in the polymer The amount of unconjugated cysteine in the dialysed and lyophilised polymer was quantified using the TNBS reagent (2,4,6-Trinitrobenzenesulfonic acid, Sigma, St. Louis, MO, USA). TNBS reacts with the primary amino groups of cysteine in a nucleophilic aromatic substitution developing an orange dye. Polymers (5 mg) were swollen in 0.5 ml of demineralised water containing 1% (m / v) NaCl to reduce the viscosity. A 0.1-ml aliquot of this solution was transferred to a 96-well microtitration plate and incubated with an equal volume of 0.1% (w / v) TNBS in 8% (w / v) NaHCO 3 at 37 8C for 2 h. The absorbance was measured at 450 nm. The amount of free amino groups was calculated using a standard curve of cysteine over a concentration range of 1–30 mg / ml prepared in 1% (m / v) aqueous NaCl solution.
2.5. Preparation of tablets 2. Materials and methods
2.1. Materials Poly(acrylic acid) (PAA) of 2-, 45-, 250-, 450 kDa and hydrochloride hydrate were purchased from Sigma-Aldrich, Steinheim, Germany; polycarbophil (PCP, mol. mass.700 kDa; Noveon AA1) from BF Goodrich, Brecksville, OH, USA. Poly(acrylic acid)-cysteine conjugates of 2-, 45-, 250-, 450 kDa (PAA 2 -Cys, PAA 45 -Cys, PAA 250 -Cys and PAA 450 -Cys) and the polycarbophil-cysteine conjugate (PCP-Cys) were supplied by MucoBiomer (Leobendorf, Austria).
Lyophilised PAA-Cys and PCP-Cys conjugates and controls were compressed (Hanseaten Type El, Hamburg, Germany) into 5.0 mm diameter flat-faced tablets of 30 mg. The compaction pressure was kept constant during the preparation of all tablets.
L-cysteine
2.2. Preparation of polymer-cysteine conjugates and controls All polymers were completely hydrated in demineralised
2.6. Disintegration studies The disintegration behaviour of the tablets in 0.1 M phosphate buffer pH 6.8 at 37 8C was analysed with a disintegration test apparatus according to the European Pharmacopeia. The oscillating frequency was adjusted to 0.5 s 21 .
2.7. Evaluation of the swelling behaviour The water absorbing capacity was determined by a gravimetric method. Tablets were fixed to the needle of a
V.M. Leitner et al. / European Journal of Pharmaceutical Sciences 18 (2003) 89–96
syringe and placed in a beaker with 0.1 M phosphate buffer pH 6.8 at 37 8C. At various time intervals the hydrated tablets were taken out of the incubation medium, excess water was removed and the amount of water uptake was determined gravimetrically (Kast and Bernkop¨ Schnurch, 2001).
2.8. In vitro evaluation of the adhesive properties 2.8.1. Tensile studies Tensile studies were carried out on native porcine intestinal mucosa. A tablet was glued to a stainless steel flat disc (5 mm in diameter), which was attached to a laboratory stand with a nylon thread (15 cm). The porcine mucosa was fixed to a glass support using a cyanoacrylate adhesive, placed in a beaker and completely immersed with 0.1 M phosphate-buffered saline pH 6.8. The beaker was placed on a balance and carefully raised by a mobile platform until the mucosa came in contact with the tablet. After an incubation time of 30 min at 25 8C, the mucosa was pulled down from the tablet at a rate of 0.1 mm / s. Data points were collected every second by a personal computer (WINWEDGE software; TAL Technologies, Philadelphia, PA, USA) connected to the balance. The total work of adhesion (TWA) representing the area under the force / distance curve and the maximum detachment force (MDF) were calculated with EXCEL 97 (Microsoft, USA) ¨ (Kast and Bernkop-Schnurch, 2001). 2.8.2. In vitro mucoadhesion studies with the rotating cylinder method Polymer tablets were attached to freshly excised intestinal porcine mucosa, which has been attached to a stainless steel cylinder (diameter: 4.4 cm, height 5.1 cm; apparatus 4-cylinder, USP XXII) using a cyanoacrylate adhesive (Loctite, Henkel, Austria). The cylinder was placed in the dissolution apparatus according to the USP, fully immersed with 0.1 M phosphate buffer pH 6.8 at 37 8C and agitated with 125 rpm. The detachment, disintegration and / or erosion of test tablets was monitored over a 24-h time ¨ period (Bernkop-Schnurch and Steininger, 2000). Additionally, the influence of free unbound cysteine in the polymer tablet on mucoadhesion was evaluated. Cysteine (2.5%, w / w) was mixed with hydrated PAA 450 -Cys or unthiolated control polymer and the pH was adjusted to 5 with 1 M NaOH. Polymers were lyophilised, compressed into tablets and the mucoadhesion time was determined.
91
PAA-Cys conjugates and controls were hydrated in demineralised water to give a concentration of 6% (m / v). The polymer solutions were added to an equal volume of mucin stock solution, mixed with a spatula and the pH of the mixture was adjusted to 6.8 with 2 M NaOH. After incubation for 20 min at room temperature, 0.8 ml of polymer / mucin incubates were transferred to a cone-plate viscometer (RotoVisco RT20, Haake GmbH, Karlsruhe, Germany) and allowed to equilibrate on the plate for 3 min at 20 8C60.5 8C. Then, dynamic oscillatory tests within the linear viscoelasticity region were performed at 1 Hz ¨ and Bernkop-Schnurch, ¨ frequency (Marschutz 2002). The parameters obtained thereby were the phase angle (d ) and the shear deformation (g ). The storage modulus (G9), the loss modulus (G0) and the complex viscosity (uh *u) were calculated by the following equations: G9 5 (t /g ) ? cos d G0 5 (t /g ) ? sin d ]]]] uh *u 5œ(h 9)2 1 (h 0)2 (whereas h 9 5 G0 /v, and h 0 5 G9 /v ) where t is the shear stress, v is the angular frequency which is related to the oscillatory frequency y by the relationship v 5 2 p y.
2.10. Statistical data analysis Statistical data analyses were performed using the Mann–Whitney U-test with P,0.05 as the minimal level of significance. Calculations were done using the software Xlstat version 5.0 (b8.3).
3. Results
3.1. Characterisation of the PAA-Cys conjugates The chemical substructure of the PAA-Cys conjugates used in this study is illustrated in Fig. 1. Determination of the thiol groups attached to the polymer by the Ellman’s
2.9. Rheological studies of polymer /mucin mixtures Porcine gastric mucin (2 g, Type II: crude, Sigma, St. Louis, MO, USA) was hydrated in 12.5 ml of demineralised water under continuous stirring at 4 8C overnight. The mucin solution was adjusted to pH 6.8 with 1 M NaOH and diluted to a final volume of 25 ml with 0.1 M phosphate buffer pH 6.8. The mucin stock solution (8% m / v) was stored at 4 8C no longer than 24 h.
Fig. 1. Chemical substructure of poly(acrylic acid)-cysteine conjugates.
92
V.M. Leitner et al. / European Journal of Pharmaceutical Sciences 18 (2003) 89–96
test demonstrated that on average 404.1665.5 mMol thiol groups were immobilised per gram polymer. Remaining traces of unbound cysteine after dialysis were determined with TNBS-reagent to be less than 0.5% (m / m) of the total mass of the polymer. The lyophilised polymer-cysteine conjugates and control polymers appeared as white, odourless powder of fibrous structure. They were easily hydratable in aqueous solutions forming thereby solutions of initially low viscosity.
3.2. Disintegration studies Disintegration studies demonstrated for all polymer tablets comprising a conjugate or control polymer that the disintegration time increased with increasing molecular mass of the polymers (Fig. 2). Apart from the PAA 2 -Cys matrix tablets, thiolated polymer tablets displayed a significantly (P , 0.05) higher stability than control tablets. Moreover, with increasing molecular mass of the polymers, the difference in the disintegration time between unthiolated and thiolated polymer matrix tablets became more pronounced. In particular matrix tablets of PAA 450 Cys and PCP-Cys were stable for more than 2 days and no erosion could be observed over this time period.
3.3. Evaluation of the swelling behaviour The swelling behaviour of mucoadhesive polymers exerts a great influence on their adhesive properties, drug release and stability (Mortazavi and Smart, 1993). The extent and rate of swelling are affected by the degree of crosslinking and chain length of the polymers (Smart, 1999). Mucoadhesive polymers are believed to take water from the underlying mucosal tissue by absorbing, swelling and capillary effects leading to considerably strong adhesion (Duchene and Ponchel, 1992). However, such ‘adhesion by hydration’ usually ceases when swelling continues and an excessive water uptake transforms the mucoadhesive dosage form into an over-hydrated slippery mucilage (Lehr, 1996). In order to evaluate the correlation between the swelling behaviour and mucoadhesion, water uptake studies were carried out with polymer and control tablets in 0.1 M phosphate buffer pH 6.8 at 37 8C. The experiments were only performed at this physiological pH as the buffer capacity within the polymer matrix is sufficiently high that the pH of the surrounding medium exerts only little influence on the swelling behaviour (data not shown). The swelling behaviour of tablets containing the PAA 2 -Cys and the PAA 45 -Cys conjugate as well as their corresponding controls could not be determined as they dissolved in the buffer within 10 min. Results of this study are shown in Fig. 3A. The initial weight of the PAA 250 Cys, PAA 450 -Cys and PCP-Cys matrix tablets (30 mg) increased 22.9-, 15.2- and 18.2-fold, respectively, over a time period of 90 min. The tablets appeared as highly cohesive and fully transparent. Fig. 3B shows the swelling behaviour of PAA 450 and PCP control tablets. For the PAA 250 control polymer, the swelling behaviour could not be determined because of rapid dissolution. After 30 min, the weight of the unthiolated PAA 450 tablets decreased due to erosion of the tablets.
3.4. Mucoadhesive properties of the PAA-Cys and PCPCys conjugates
Fig. 2. Disintegration behaviour of matrix tablets (30 mg, 5 mm I.D.) based on PAA-Cys conjugates (black bars) or corresponding unmodified polymers (grey bars). Studies were carried out with a disintegration test apparatus (Pharm. Eur.) in 0.1 M phosphate buffer pH 6.8 at 37 8C. PAA 450 -Cys and PCP-Cys matrix tablets did not disintegrate within an observation period of 48 h. Indicated values are means (6S.D.) of three experiments.
Mucoadhesion studies were performed in order to evaluate the influence of the molecular mass on mucoadhesion using two different test methods. On the one hand, tensile studies were carried out on porcine intestinal mucosa. Results in Fig. 4 show a strong influence of the molecular mass and crosslinking of the polymer tablets on their adhesive properties. The maximum detachment force (MDF) thereby did not increase with increasing total work of adhesion (TWA) but was in almost the same range for all polymers (data not shown). On the other hand, in order to confirm the results by another mucoadhesion test system, mucoadhesion tests were carried out on the rotating cylinder. Thereby obtained results, as shown in Fig. 5, were in good accordance with the results of tensile studies. Due to the immobilisation of thiol groups on poly(acrylic acid) the mucoadhesive prop-
V.M. Leitner et al. / European Journal of Pharmaceutical Sciences 18 (2003) 89–96
93
Fig. 4. Influence of the molecular mass on mucoadhesive properties of PAA-Cys conjugates (black bars) and the corresponding unthiolated polymers (grey bars). Represented values are means (6S.D., n 5 3–6) of the total work of adhesion (TWA) determined via tensile studies at pH 6.8 with tablets of indicated test material. *, differs from according unthiolated polymer, P , 0.01.
Additionally, the influence of free unbound cysteine on mucoadhesion of PAA 450 -Cys matrix tablets was evaluated. Results are shown in Fig. 6. The addition of free cysteine to the polymers had only an effect on the mucoadhesion behaviour of thiolated polymers, but not on the unthiolated polymer control.
3.5. Rheological evaluation of polymer /mucin mixtures
Fig. 3. A. Swelling behaviour of PAA 250 -Cys (O), PAA 450 -Cys (3) and PCP-Cys (m) tablets in 100 mM phosphate buffer pH 6.8 at 37 8C. B. Swelling behaviour of controls tablets based on PAA 450 (3) and PCP (m). All indicated values are means (6S.D.) of at least three experiments.
erties of the non-crosslinked polymers PAA 45 -Cys, PAA 250 -Cys and PAA 450 -Cys conjugates were 1000-, 93and 16-fold improved, respectively, while the crosslinked polymer PCP-Cys exceeded its control only 1.7-fold.
The viscosity of polymer / mucin mixtures is the net result of the resistance to flow exerted by individual chain segments, physical chain entanglements and (non)covalent intermolecular interactions, which are the same as the interactions involved in the process of mucoadhesion. Therefore, it has been suggested that the forces of interactions involved in a mucoadhesive system could be evaluated by viscosity measurements (Hassan and Gallo, 1990). According to this, the viscoelastic properties of the PAACys and PCP-Cys conjugates and controls were determined after incubation with mucin. Viscosity studies were performed with commercially available mucin instead of native mucus resulting in more reproducible and comparable results. In addition, orientating studies demonstrated that there is no significant difference in the results obtained with native mucus and hydrated mucin (data not shown).
94
V.M. Leitner et al. / European Journal of Pharmaceutical Sciences 18 (2003) 89–96
Fig. 6. Mucoadhesive behaviour of matrix tablets with and without free unbound cysteine determined via the rotating cylinder method. The black bars represent the time of mucoadhesion of PAA 450 -Cys tablets, the grey bars the mucoadhesion time of PAA 450 control tablets with and without the addition of 2.5% free cysteine (means6S.D., n 5 3). *, differs from according polymer without L-Cys, P , 0.05. Fig. 5. Influence of the molecular mass on mucoadhesive properties of PAA-Cys conjugates (black bars) and corresponding controls (grey bars). Test tablets of indicated polymers were attached to excised porcine mucosa, which has been spanned on a cylinder and agitated at 125 rpm in 0.1 M phosphate buffer pH 6.8 at 37 8C. The indicated time of adhesion represents the mean (6S.D.) of at least three experiments. *, differs from control, P , 0.05.
With increasing molecular mass of the polymers, storage modulus G9 and loss modulus G0 increased from 0.3260.11 / 2.9560.44 (G9 /G0) for PAA 250 -Cys to 727.776174.70 / 188.20667.94 (G9 /G0) for PCP-Cys. As depicted in Fig. 7, PAA 250 -Cys / mucin, PAA 450 -Cys / mucin and PCP-Cys / mucin mixtures showed significantly (P , 0.05) increased viscosity compared to unthiolated polymer / mucin mixtures; for PAA 250 -Cys / mucin the improvement was even more than 6-fold. The rheological properties of PAA 2 -Cys / mucin and PAA 45 -Cys / mucin mixtures and their controls could not be detected as they were too low to be determined with the available equipment. The mucin dispersion itself did not exhibit any detectable viscoelastic properties at the concentration employed (4% w / v), as well.
4. Discussion The molecular mass—correlating with the molecular chain length—determines the strength of mucoadhesive bonds, as molecular mobility and flexibility affect the ability of polymer and mucus to form entanglements. It is believed that an optimally high molecular mass corre-
Fig. 7. Influence of the molecular mass on the viscosity of PAA-Cys / mucin mixtures (black bars) and corresponding control / mucin mixtures (grey bars). Oscillatory measurements at 1 Hz frequency were carried out immediately after an incubation period of 20 min at room temperature. Indicated values are means (6S.D.) of at least three experiments.
V.M. Leitner et al. / European Journal of Pharmaceutical Sciences 18 (2003) 89–96
sponds to maximum adhesiveness (Caramella et al., 1999). However, if the polymer chains are too long, their ability to diffuse into the mucus gel is limited (Huang et al., 2000). According to this theory, it was the objective of this study to evaluate the influence of the molecular mass and chain length of thiomers on their mucoadhesive and cohesive properties. An important factor for mucoadhesion are the cohesive properties in conjunction with the swelling behaviour of polymers (Mortazavi and Smart, 1993). In contrast to the theory that rapidly swelling polymers will also quickly interact with the mucin (Mortazavi and Smart, 1993), a correlation between the swelling behaviour of investigated polymers and their mucoadhesive properties could not be found. The water uptake of PAA 250 -Cys tablets was the comparatively highest, whereas its mucoadhesive properties were lower than those of PAA 450 Cys tablets. These results are in good agreement with ¨ previous published studies (Bernkop-Schnurch and Steininger, 2000). Furthermore, a direct correlation between the molecular mass of the polymers and their swelling behaviour was not observed. Tensile studies as well as mucoadhesion studies with the rotating cylinder method showed that among the different PAA-Cys conjugates tested, the PAA 450 -Cys conjugate displayed the strongest mucoadhesive properties. The thiolated PAA 450 polymer used for this study exhibited even the highest TWA so far determined in comparison to other thiolated polymers tested under the same conditions. On the rotating cylinder, PAA 450 -Cys matrix tablets remained attached to porcine mucosa for about 16-times longer than corresponding control tablets, while PCP-Cys matrix tablets exceeded their controls only 1.7 times. These results strongly support the theory that non-crosslinked polymers display higher chain flexibility (Duchene and Ponchel, 1992) and therefore are able to interpenetrate the mucus layer more deeply resulting in enhanced mucoadhesion. Within this study, thiomers of comparatively low molecular mass were investigated for their mucoadhesive properties according to the assumption that their smaller and therefore more flexible chains would favour even stronger interpenetration and chain entanglements with the mucus resulting in improved mucoadhesion. However, mucoadhesion, disintegration and swelling tests of PAA 2 -Cys, PAA 45 -Cys and PAA 250 -Cys tablets resulted in decreased adhesive and cohesive properties of the tablets with decreasing molecular mass of the conjugates. This could be explained by too low intermolecular cohesive forces. The small molecules bearing only a few thiol groups were not able to form a highly cohesive three-dimensional network via the formation of intermolecular disulphide bonds. An important factor for mucoadhesion and cohesion of thiolated polymers is their ability to form covalent bonds within the polymer itself as well as between the polymer ¨ and the mucus layer (Bernkop-Schnurch and Steininger,
95
2000). The formation of disulphide bonds within the polymer resulted in much higher stability of thiomer tablets compared to control tablets, as shown in disintegration tests. This high cohesiveness of thiolated polymer matrix tablets resulted in an almost zero order release of calcitonin for at least 8 h as described recently by our research group (Guggi et al., 2002). The formation of covalent bonds between thiomer and mucus was also investigated by incubating commercially available mucin with a thiolated polymer and the corresponding unmodified polymer. While the mucin could be completely removed from the unmodified polymer / mucin mixture by centrifugation and removing the supernatant, it remained bound to the thiomer and could only be washed out by the addition of the disulphide bond breaker dithiothreitol (Bernkop¨ Schnurch et al., 1999). Within this work, additional evidence for covalent bonds between thiomer and mucus could be provided by the addition of free cysteine to PAA 450 -Cys. As cysteine is also known to be a disulphide bond breaking agent (Lightowler and Lightowler, 1971) and its addition to the polymer led to a massive reduction in mucoadhesion, it may be concluded that disulphide bonds between polymer and mucus are responsible for the enhanced mucoadhesive properties of thiomers. Similar results were obtained by Mortazavi, who reported a significant reduction in mucoadhesive strength of Carbopol 934P in the presence of highly concentrated hydrogen bond breakers such as KCNS and urea. It was concluded that the formation of hydrogen bonds between polymer and mucus was an important component for mucoadhesion (Mortazavi, 1995). The results obtained from viscosity studies showed that the viscosity of the polymer / mucin mixtures and controls increased with increasing molecular mass. As expected and previously reported by Mortazavi and Smart (1994b), the shorter chain length polymers with inherently poor gel forming properties were found to be less effective in promoting gel strengthening. The much higher viscosity of conjugate / mucin mixtures compared to controls can be explained by the formation of disulphide bonds between the thiolated polymer and mucin leading to a strengthened gel network, as mentioned above.
5. Conclusion This study compares the mucoadhesive and cohesive properties of five different poly(acrylic acid)-cysteine conjugates with a molecular mass in the range of 2 kDa– 1–3 million kDa. Among all conjugates tested, the PAA 450 -Cys conjugate displayed the most favourable properties concerning mucoadhesion and cohesion of matrix tablets. Results even demonstrate that in comparison to all other tested mucoadhesive polymers, this conjugate exhibits the highest total work of adhesion and accordingly mucoadhesion ever reached. Additionally, the
96
V.M. Leitner et al. / European Journal of Pharmaceutical Sciences 18 (2003) 89–96
polymer tablets showed an excellent stability comparable to PCP-Cys tablets being the most stable ones described so far. The findings of this study also contribute valuable basic information about the mechanisms involved in mucoadhesion of thiomers. Low molecular mass polymers with flexible chains favouring strong interpenetration are not cohesive enough for optimal mucoadhesion, whereas high molecular weight crosslinked polymers showing high cohesiveness do not display enough chain flexibility. The optimal mucoadhesive polymer represents therefore a compromise of these two extremes. In addition, this study provides further evidence for the formation of covalent bonds between thiolated polymers and mucin glycoproteins as mucoadhesion is strongly decreased when a disulphide bond breaking agent is added to the polymer. The present study gives an extensive characterisation of a broad range of thiolated poly(acrylic acids) and should therefore facilitate the development of new drug delivery systems providing controlled drug release and a greatly prolonged residence time on various mucosal tissues.
Acknowledgements This work was supported by Grant No. P15373-MOB ¨ from the Fonds zur Forderung der wissenschaftlichen ¨ Forschung (FWF) to A. Bernkop-Schnurch. ¨ and co-workers The authors wish to thank Mr. Strobl from the slaughterhouse Totzenbach for the supply of porcine intestinal mucosa.
References ¨ Bernkop-Schnurch, A., Schwarz, V., Steininger, S., 1999. Polymers with thiol groups: a new generation of mucoadhesive polymers? Pharm. Res. 16, 876–881. ¨ Bernkop-Schnurch, A., 2000. Mucoadhesive polymers. In: Dumitriu, S. (Ed.), Polymeric Biomaterials. Marcel Dekker, New York, pp. 147– 165. ¨ Bernkop-Schnurch, A., Steininger, S., 2000. Synthesis and characterisation of mucoadhesive thiolated polymers. Int. J. Pharm. 194, 239–247. ¨ Bernkop-Schnurch, A., Kast, C.E., Richter, M.F., 2001a. Improvement in the mucoadhesive properties of alginate by the covalent attachment of cysteine. J. Control. Rel. 71, 277–285. ¨ Bernkop-Schnurch, A., Zarti, H., Walker, G.F., 2001b. Thiolation of polycarbophil enhances its inhibition of soluble and intestinal brush border membrane bound aminopeptidase N. J. Pharm. Sci. 90, 1907– 1914.
Caramella, C.M., Rossi, S., Bonferoni, M.C., 1999. A rheological approach to explain the mucoadhesive behavior of polymer hydrogels. Drugs Pharm. Sci. 98, 25–65. Chickering III, D.E., Mathiowitz, E., 1999. Definitions, mechanisms, and theories of bioadhesion. Drugs Pharm. Sci. 98, 1–10. ¨ Clausen, A.E., Bernkop-Schnurch, A., 2000. In vitro evaluation of the permeation-enhancing effect of thiolated polycarbophil. J. Pharm. Sci. 89, 1253–1261. Duchene, D., Ponchel, G., 1992. Principle and investigation of the bioadhesion mechanism of solid dosage forms. Biomaterials 13, 709– 714. Ellman, G.L., 1959. A colorimetric method for determining low concentrations of mercaptans. Arch. Biochem. Biophys. 74, 443–450. ¨ Guggi, D., Kast, C.E., Bernkop-Schnurch, A., 2002. In vivo evaluation of an oral calcitonin delivery system for rats based on a thiolated chitosan matrix, Proceed. 11th Inter. Pharm. Technol. Symp., pp. 41–42. Gum, J.R.J., Hicks, J.W., Toribara, N.W., Rothe, E.-M., Lagace, R.E., Kim, Y.S., 1992. The human MUC2 intestinal mucin has cysteine-rich subdomains located both upstream and downstream of its central repetitive region. J. Biol. Chem. 267, 21375–21383. Hassan, E.E., Gallo, J.M., 1990. A simple rheological method for the in vitro assessment of mucin-polymer bioadhesive bond strength. Pharm. Res. 7, 491–495. Huang, Y., Leobandung, W., Foss, A., Peppas, N.A., 2000. Molecular aspects of muco- and bioadhesion: tethered structures and site-specific surfaces. J. Control. Rel. 65, 63–71. ¨ Kast, C.E., Bernkop-Schnurch, A., 2001. Thiolated polymers–thiomers: development and in vitro evaluation of chitosan-thioglycolic acid conjugates. Biomaterials 22, 2345–2352. Lehr, C.M., 1996. From sticky stuff to sweet receptors — achievements, limits and novel approaches to bioadhesion. Eur. J. Drug Metab. Pharmacokinet. 21, 139–148. Lightowler, J.E., Lightowler, N.M., 1971. Comparative mucolytic studies on dithiothreitol, N-acetylcysteine and L-cysteine on human respiratory mucus in vitro and their effects on the rate of flow of mucus in the exposed trachea of the rat on topical application. Arch. Int. Pharmacodyn.Ther. 189, 53–58. ¨ M.K., Bernkop-Schnurch, ¨ Marschutz, A., 2002. Thiolated polymers: selfcrosslinking properties of thiolated 450 kDa poly(acrylic acid) and their influence on mucoadhesion. Eur. J. Pharm. Sci. 15, 387–394. Mortazavi, S.A., Smart, J.D., 1993. An investigation into the role of water movement and mucus gel dehydration in mucoadhesion. J. Control. Rel. 25, 197–203. Mortazavi, S.A., Smart, J.D., 1994a. An in vitro method for assessing the duration of mucoadhesion. J. Control. Rel. 31, 207–212. Mortazavi, S.A., Smart, J.D., 1994b. Factors influencing gel-strengthening at the mucoadhesive-mucus interface. J. Pharm. Pharmacol. 46, 86–90. Mortazavi, S.A., 1995. An in vitro assessment of mucus / mucoadhesive interactions. Int. J. Pharm. 124, 173–182. Peppas, N.A., Buri, P.A., 1985. Surface, interfacial and molecular aspects of polymer bioadhesion on soft tissues. J. Control. Rel. 2, 257–275. Smart, J.D., 1999. The role of water movement and polymer hydration in mucoadhesion. Drugs Pharm. Sci. 98, 11–23.