Doi:10.1016/j.ejpb.2004.02.008

European Journal of Pharmaceutics and Biopharmaceutics 57 (2004) 447–455 Investigation of the cytotoxicity and insulin transport of acrylic-based copolymer protein delivery systems in contact with caco-2 cultures Departments of Chemical Engineering, Biomedical Engineering, and Division of Pharmaceutics, The University of Texas at Austin, Austin, TX, USA Received 8 December 2003; accepted in revised form 2 February 2004 Microparticles or nanospheres of hydrogels of crosslinked poly(methacrylic acid) grafted with poly(ethylene glycol) as well as crosslinked poly(acrylic acid) grafted with poly(ethylene glycol) were prepared for use as oral insulin delivery carriers. The copolymer carriers weresynthesized by precipitation/dispersion polymerization that led to gel nanospheres or by bulk polymerization and subsequent size reductionof thin films to obtain gel microparticles. The cytotoxicity of these copolymers was investigated in contact with Caco-2 cell cultures using ametabolic assay to measure the effect of the presence of copolymers on the cell viability. The copolymers were found to exhibit no cytotoxiceffect on the cell cultures. Insulin-loaded formulations were also tested for cytotoxicity and insulin transport studies across cell monolayers.
The copolymers were shown to open the tight junctions between cells, increasing the available area for diffusion across the cell monolayer,and thus increasing the permeability of insulin across the monolayer.
q 2004 Elsevier B.V. All rights reserved.
Keywords: Acrylic-based copolymer; Copolymer nanospheres; Poly(ethylene glycol); Caco-2 cell culture; Insulin transport Numerous scientific obstacles must be overcome in order to create an oral protein delivery system. The greatest The harsh conditions of the gastrointestinal tract often barriers are the harsh conditions of the stomach and the denature an orally administered protein formulation before transport barrier, present in the intestine. The stomach has it reaches the blood stream. In the case of oral insulin an acidic environment full of proteolytic enzymes. Sensitive delivery devices, less than 0.1% of the orally dosed insulin proteins are often denatured or destroyed in the stomach.
reaches the blood stream intact . Clearly, to make use of The small amount of protein that makes it through the the ever-expanding library of proteins, proteins must be stomach intact must then be transported across the lining of the intestine to get into the bloodstream. This transport Unfortunately, injections are often painful, leading to process is greatly reduced by the mucus and no significant low patient compliance . Research has focused on insulin can be transported across the cell layers.
alternate ways of delivering proteins and other unstable Previously, we have shown that copolymers of therapeutic agents, including oral delivery. By minimizing methacrylic acid (MAA) containing pendent chains of injections, patient compliance could increase and the poly(ethylene glycol) have unique properties as carriers for treatment regime could be made less intrusive, improving transmucosal drug delivery . In this work, we have extended the previous studies by developing a group ofcopolymer hydrogels capable of protecting proteins whilethey are in transit through the stomach, and then aiding in * Corresponding author. Departments of Chemical Engineering, Biomedical Engineering, and Division of Pharmaceutics, The University increasing protein transport across the cellular barrier in the of Texas at Austin, 1 University Station C0400, Austin, TX 78712-0231, upper small intestine. Our present studies were conducted USA. Tel.: þ 1-512-471-6644; fax: 1-512-471-8227.
with insulin-loaded systems. The family of copolymers E-mail address: [email protected] (N.A. Peppas).
studied contained either methacrylic acid or acrylic acid Present address: Boston Scientific, 1 Scimed Place, Maple Grove, MN (selected, for their pH-sensitive nature and ability to bind 0939-6411/$ - see front matter q 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.ejpb.2004.02.008 A.C. Foss, N.A. Peppas / European Journal of Pharmaceutics and Biopharmaceutics 57 (2004) 447–455 calcium) and was grafted with poly(ethylene glycol) 50 wt% monomer in deionized water and ethanol was selected for its ability to stabilize and protect proteins.
purged with nitrogen to remove all oxygen present. Thesolution was then pipetted between two glass microscopeslides with Teflonw spacers of 0.9 mm separating the slides, and irradiated at 138 mW/cm2 for 20 min. The resulting filmwas removed from the slides and washed in deionized water 2.1. Synthesis of P(MAA-g-PEG) and P(AA-g-PEG) for 5 days to remove all unreacted monomer.
The copolymer films were then dried in a vacuum oven and crushed with mortar and pestle. The resulting dry Synthesis of copolymer hydrogel carriers was carried out copolymer gel particles were sieved to obtain micro- under nitrogen using UV-initiated free-radical polymeriz- particles in the size range of 150 – 220 mm, which were ation. Mixtures of methacrylic acid (MAA, Polysciences stored in a desiccator until use. The polymerization Inc., Warrington, PA) and poly(ethylene glycol) mono- kinetics, copolymer composition, and structural analysis methylether monomethacrylate (PEGMA with PEG of of these PEG-containing systems have been discussed molecular weight of 1000, Polysciences Inc., Warrington, PA), or of acrylic acid (AA, Polysciences Inc., Warrington,PA) and PEGMA were prepared.
The monomers were mixed in molar ratios of 4:1, 2:1, 1:1 or 1:2 of MAA:EG or AA:EG repeating units.
Tetraethylene glycol dimethacrylate (TEGDMA, Poly- All cell studies were conducted with Caco-2 cells.
sciences Inc., Warrington, PA) was added as the cross- Caco-2 cells spontaneously differentiate and possess tight linking agent in the amount of X ¼ 0:01 moles of TEGDMA junctions, which are characteristic of the cells lining the per mole of total monomers. Thus, a network prepared with intestine Thus, they serve as a good model for a nominal crosslinking ratio of X ¼ 0:01 is said to be 1% studying the effect of copolymer carriers on the tight crosslinked. 1-Hydroxylcyclohexyl phenyl ketone (Irga- junctions of the small intestine. The properties of the Caco-2 cure-184, Ciba-Geigy Corp., Hawthorne, NY) was used as cell line are passage number dependent thus, all studies the free-radical initiator and added in the amount of 0.5 wt% were carried out with passage numbers 60 – 80.
Caco-2 cells were cultured in 96-well plates (CoStar, Corning Incorp., Corning NY) with Dulbecco’s modified 2.2. Synthesis of copolymer gel nanospheres Eagle’s medium (DMEM, Sigma, St Louis, MO). After7 days, the cell culture reached 90% confluency. At that Dilute solutions of the monomer mixture were prepared point all DMEM were removed and replaced with Hank’s in deionized water. Depending on the monomers involved balanced salt solution (HBSS, Sigma, St Louis, MO).
and the monomer feed ratio used, most dilutions were Ranges of concentrations of copolymer gel particles were around 1 vol% monomer mixture in deionized water. The prepared in HBSS and placed in contact with the cell monomer solution was purged for 20 min with nitrogen to cultures. The pH of HBSS was 7.4, thus causing the gel remove all dissolved oxygen and polymerized under UV particles to be in their highly swollen state Cell light at an intensity of 138 mW/cm2 for 20 min. Asdiscussed before nanospheres were prepared by a viability studies of nanospheres and microparticles of both unique surfactant-free precipitation polymerization process.
P(MAA-g-PEG) and P(AA-g-PEG) were performed. While The resulting nanospheres were subsequently washed using these studies were carried out at pH 7.4, we have shown a dialysis tubing (molecular weight cut-off of 12,000 – before that they are relevant to the conditions of the upper 14,000, Spectrum, Rancho Dominguez, CA) against deio- nized water until all unreacted monomer and other small The 96-well plate was incubated for 2 h at 37 8C. Then, molecular weight oligomer and polymer chains were the HBSS with copolymer gel particles was removed and the cell culture was washed three times with 200 ml of fresh D(þ )-Trehalose was added in the amount of 1 g trehalose HBSS to remove all remaining copolymer gel particles. A per g of copolymer gel nanospheres and the mixture was sample of 100 ml of HBSS and 20 ml of CellTiter 96w frozen overnight and freeze-dried to remove all water aqueous solution cell proliferation assay (Promega, Madi- present. The resulting dry copolymer gel nanospheres were son, WI), an NADPH reactive assay, were placed in each well and incubated for 90 min. The NADPH produced bythe living cells reacted with the reagents of the assay, 2.3. Synthesis of copolymer gel microparticles resulting in a colorimetric change at 490 nm. The colori-metric change was measured via UV-absorption by a Copolymer microparticles were prepared by crushing microplate reader (Elx 800 NB Bio-Tek Instruments, thin films prepared by UV polymerization. A solution of A.C. Foss, N.A. Peppas / European Journal of Pharmaceutics and Biopharmaceutics 57 (2004) 447–455 ELISA (Bovine Insulin EIA, ALPCO Diagnostics, Wind-ham, NH).
Caco-2 cell cultures were grown in the apical chambers of 6-well Transwell plates (Costar, Corning Inc., CorningNY) with DMEM media for 21 days. The membranes in the plate separating the apical and basolateral chambers had apore size of 4.0 mm. The formation of the tight junctions Synthesis of the copolymer carriers for insulin delivery between adjacent cells was monitored by measuring the was conducted via two techniques. The first method was a resistance across the cell layer with an electrical resistance precipitation/dispersion polymerization to prepare gel nano- probe. After 21 days of growth, the resistance reached an spheres The second was an in situ free-radical equilibrium value, which indicated the monolayer’s tight polymerization leading to thin films, which were sub- sequently crushed to obtain gel microparticles In the present research we conducted studies with 2.6. Transepithelial electrical resistance studies hydrogels with a wide range of compositions as describedby the AA:EG and MAA:EG molar ratios in order toevaluate their suitability as carriers for protein delivery Transwell plates were used for transepithelial electrical This range included four monomer feed ratios of resistance studies (TEER). All DMEM growth media were 4:1, 2:1, 1:1 and 1:2 acid to EG repeating units. The removed and replaced with HBSS. The resistance across the monomer feed ratio played a dominant role in the physical cell monolayer was then measured to detect any difference in TEER value between different cell media. A sample of 10 mg As the monomer feed ratio of carboxylic acid groups was of copolymer gel particles was added to the apical chamber of increased, the equilibrium-swelling ratio increased and each well. Particles of four different copolymer gel approached that of a pure MAA network. Lowman and compositions were investigated. The first system studied Peppas concluded that the ability of the copolymer gel was nanospheres of crosslinked P(AA-g-PEG) with an initial network to hydrogen bond, and thus complex at lower pH, feed monomer ratio of 2:1 AA:EG. P(MAA-g-PEG) nano- went through a maximum at a monomer feed ratio of 1:1 spheres with a feed monomer ratio of 1:1 MAA:EG were also MAA:EG. Thus, the copolymers studied here were prepared studied in addition to copolymer gel microparticles of P(AA- with a wide range of comonomer feed ratios of MAA to EG g-PEG) with a monomer feed ratio of 2:1 AA:EG. The final to examine the different characteristics of the copolymer gel copolymer gel particles were P(MAA-g-PEG) microparti- networks with different compositions. It was also important cles with a monomer feed ratio of 1:1 MAA:EG. The control to maintain an adequate amount of PEG in the copolymer study used a blank well with cells and media subjected to gel network because PEG acts as a mucoadhesion promoter identical conditions as the experimental wells.
to the copolymer gel carriers as observed and discussed by The resistance across the cell monolayer was measured at Sahlin et al. P(AA-g-PEG) was also examined as a given intervals to determine the extent to which the cell tight possible protein carrier for oral delivery.
3.1. Preparation of copolymer nanospheres 2.7. Protein transport enhancement studies Transwell plates were also used for the protein The mechanism for insulin release from copolymer transport studies. The chamber on the apical side of nanospheres was based on the complexation/decomplexation the monolayer was filled with 2 ml of HBSS containing process due to hydrogen bonding and the associated increase 10 mg of copolymer and 0.4 mg of insulin (Insulin from of the network mesh size. The pH-sensitive nature of the bovine pancreas, Sigma, St Louis, MO). Four copolymer copolymer nanospheres was evident when the pH of the systems of: P(AA-g-PEG) microparticles with monomer environment was increased and the copolymer gel nano- feed ratio of 2:1 AA:EG; P(AA-g-PEG) nanospheres with sphere mixture became transparent. From previous studies monomer feed ratio of 2:1 AA:EG; P(MAA-g-PEG) on crosslinked P(MAA-g-PEG) and P(AA-g-PEG) gel microparticles with monomer feed ratio of 1:1 MAA:EG; networks, it is known that as the pH of the swelling medium and P(MAA-g-PEG) nanospheres with monomer feed increased, the swelling ratio of the network increased. All copolymer network carriers studied became transparent at The Transwell plate was kept at a constant 37 8C for the high pH values. The change in gel opacity as the pH was duration of the experiment. A sample of 10 ml from the increased was caused by the hydrogen bonding complexation basolateral chamber of the Transwell was taken at given between the etheric oxygen of PEG and the carboxylic acid time intervals. A sample of 10 ml was taken from the apical group of MAA or AA which was disrupted at high PH values chamber of the Transwell cell every hour. The insulin leading to incorporation of more solvent into the copolymer concentrations of the samples were analyzed via insulin network. This transition of the gel particles from A.C. Foss, N.A. Peppas / European Journal of Pharmaceutics and Biopharmaceutics 57 (2004) 447–455 a phase-separated structure to a one-phase system was aunique characteristic of these complexation hydrogels.
Residual monomers present in the post-reaction mixture along with small chain oligomers were removed from thegel nanospheres mixture by partitioning with deionizedwater. The newly formed copolymer gel nanospheres wereplaced in cellulose tubing with a molecular cut-off weight of12,000 – 14,000 and then placed in a deionized water bath.
Washings were analyzed via UV to ensure all unwantedmaterials were removed from the copolymers. D(þ )-Trehalose was added as a good cryoprotectant andparticle stabilizer.
The method of gel microparticle preparation required the use of solvent to create a favorable environment for the Fig. 1. Viability of Caco-2 cell culture after 2 h of contact time with various formation of hydrogen bonding complexation. A 50/50 wt% microparticles of P(MAA-g-PEG) with monomer MAA:EG feed ratios of: ethanol/water solvent was shown by Lowman and Peppas 4:1 (W), 2:1 (A), 1:1 ( # ), 1:2 ( # ). Cell viability was measured by NADPH as a good solvent for preparing these films.
production compared to control cell culture with no copolymer present.
Studies conducted with Caco-2 cell cultures ðn ¼ 6Þ: with a 4:1 MAA:EG monomer feed ratio exhibited a The main goal of the cytotoxicity studies was to reduction in NADPH production of nearly 35% at the determine the cell viability of Caco-2 cell cultures in the highest concentration of gel microparticles. As shown in presence of the copolymer micro- or nanospheres. The the cell viability, or ratio of NADPH production, results of the cell proliferation assay were expressed in approached unity as the concentration of copolymer gel terms of the amount of NADPH produced by the cells still functioning in the culture. These were compared to the data While the error bars of the data points overlapped for the from control cell cultures to calculate a ratio of NADPH P(MAA-g-PEG) microparticles with 1:2, 1:1 MAA:EG feed produced, by averaging the UV absorption of wells that ratio and on occasion for those prepared from the 2:1 contained a given copolymer gel particle concentration and MAA:EG monomer feed ratio, it was observed that the gel dividing it by the average UV adsorption of the control cell microparticles of P(MAA-g-PEG) with feed monomer ratio culture. This ratio was a measure of the gel particles’ effect of 1:2 MAA:EG consistently caused the lowest decrease in NADPH production. These results could be explained by the As the half-life of NADPH was short, on the order of fact that the carboxylic acid group in MAA could bind seconds, only NADPH produced by cells still viable in calcium, an important ion in cell function , culture was detected. Residual NADPH from dead cells was thus disrupting the calcium concentration necessary removed with the polymer, or decayed and did not react for cell function. The P(MAA-g-PEG) copolymer gel with the tetrazolium compound to produce a colorimetricchange. The results of these studies could be used tocompare the behavior of microparticles of P(MAA-g-PEG)and P(AA-g-PEG) with different monomer feed ratios.
shows the cell viability versus particle concen- tration as expressed by an NADPH production ratio for theset of P(MAA-g-PEG) crushed microparticles. Four differ-ent P(MAA-g-PEG) copolymers were synthesized withvarying monomer feed ratios, of 4:1, 2:1, 1:1, 1:2 MAA:EGrepeating units.
The P(MAA-g-PEG) copolymer microparticles caused no significant decrease in cell viability with the exception ofthe P(MAA-g-PEG) microparticles with 4:1 MAA:EG feedratios. The P(MAA-g-PEG) copolymer gel microparticles of1:2, 1:1 and 2:1 MAA:EG caused less than a 15% decreasein NADPH production compared to control cultures. Allthese copolymer microparticles would be suitable as oral Fig. 2. Viability of Caco-2 cell culture after 2 h of contact time with variousmicroparticles of P(AA-g-PEG) with monomer feed ratios of AA:EG ¼ 4:1 insulin delivery devices since even at the highest concen- (W), 2:1 (A), 1:1 ( # ), 1:2 ( # ). Cell viability was measured by NADPH tration studied, they did not exhibit any effect on the cell production compared to control cell culture with no copolymer present.
cultures. The P(MAA-g-PEG) copolymer microparticles Studies conducted with Caco-2 cell cultures ðn ¼ 6Þ: A.C. Foss, N.A. Peppas / European Journal of Pharmaceutics and Biopharmaceutics 57 (2004) 447–455 microparticles with 1:2 MAA:EG monomer feed ratio hadthe lowest calcium-binding ability.
shows the cell viability for the set of P(AA-g-PEG) microparticles. Four different P(AA-g-PEG) copolymerswere studied with monomer feed ratios of 4:1, 2:1, 1:1, 1:2AA:EG repeating units.
The results indicate that at the highest copolymer gel microparticle concentration the P(AA-g-PEG) gel micro-particles prepared with a monomer feed ratio of 1:1 and 1:2AA:EG repeating units exhibited no cytotoxicity. TheP(AA-g-PEG) gel microparticles prepared with monomerfeed ratio of 2:1 AA:EG repeating units caused only a minordecrease in NADPH production, suggesting a slight effecton the cell viability. The P(AA-g-PEG) gel microparticles Fig. 4. Viability of Caco-2 cell culture after 2 h of contact time with various with monomer feed ratio of 4:1 AA:EG exhibited a more nanospheres of P(AA-g-PEG) with monomer feed ratios of AA:EG ¼ 4:1 pronounced effect on the NADPH production, causing a (W), 2:1 (A), 1:1 ( # ), 1:2 ( # ). Cell viability was measured by NADPHproduction compared to control cell culture with no copolymer present.
decrease in NADPH production of nearly 30% at the Studies conducted with Caco-2 cell cultures ðn ¼ 6Þ: 20 mg/ml gel microparticle concentration. These resultsindicate that the presence of large amounts of acrylic acid dispersion polymerization technique, in contact with repeating units in complexation hydrogels with preponder- Caco-2 cell cultures. Cell viability was expressed as a ence of AA of EG units, leads to highly acidic local ratio of NADPH production with respect to the control.
microenvironments due to the ionization of the AA units.
Similar to the gel microparticles, the P(MAA-g-PEG) gel Therefore, hydrogels that do not exhibit a stoichiometric nanospheres with monomer feed ratios of 1:2 and 1:1 ratio of AA:EG may cause some cell cytotoxicity. The MAA:EG repeating units caused little decrease in the reader is drawn to a comparison of these results with those NADPH production of the cell culture. Both gel nano- of Torres-Lugo et al. who noted that in P(MAA-g-EG) spheres caused only a 10% loss in NADPH, even at the the ratio of MAA:EG did not affect cell viability. Clearly, highest concentration of nanospheres that was 20 mg/ml.
The P(MAA-g-PEG) gel nanospheres with a monomer effects on the physicochemical and cellular behavior of feed ratio of 2:1 MAA:EG caused only slightly more of these hydrogels. First, it stabilizes the hydrogen bonds a decrease in NADPH production, but the cell viability remained above 80%. The P(MAA-g-PEG) gel nano- it provides protection of the cells.
spheres with a monomer feed ratio of 4:1 MAA:EG The results of show the cell viability of caused a significant loss in NADPH production.
nanospheres of P(MAA-g-PEG) and P(AA-g-PEG) pre- The P(MAA-g-PEG) gel nanospheres with a monomer pared with different monomer feed ratios. In we feed ratio of 4:1 MAA:EG had a diameter of 3 mm in their plot the cell viability versus the concentration of P(MAA- fully swollen state as measured by photon correlation g-PEG) gel nanospheres, prepared by a precipitation/ spectroscopy (PCS). P(MAA-g-PEG) gel nanospheres withmonomer feed ratios of 2:1, 1:1 and 1:2 MAA:EG haddiameters of 1.5, 1.1 and 0.5 mm, respectively, in their fullyswollen states. At a pH of 2.0, all copolymer gelnanospheres had a diameter near 0.2 mm.
The larger decrease in cell viability observed at high concentrations of P(MAA-g-PEG) gel nanospheres with amonomer feed ratio of 4:1 MAA:EG was attributed to itshigher content of MAA in the network. The exact mechanismby which the increased MAA caused a decrease in NADPHproduction is unclear. It could be related to the largernanosphere size when compared to the other P(MAA-g-PEG)gel nanospheres at the conditions of the study. It could also bethat the higher amount of MAA, which bound ions in solution,disrupted the local ionic equilibrium and caused a loss in Fig. 3. Viability of Caco-2 cell culture after 2 h of contact time with various NADPH production. While the mechanism was unclear, the nanospheres of P(MAA-g-PEG) with monomer feed ratios of results indicated the P(MAA-g-PEG) gel nanospheres with a MAA:EG ¼ 4:1 (W), 2:1 (A), 1:1 ( # ), 1:2 ( # ). Cell viability was monomer feed ratio of 4:1 cause a loss in NADPH production measured by NADPH production compared to control cell culture with nocopolymer present. Studies conducted with Caco-2 cell cultures ðn ¼ 6Þ: at concentrations of 10 and 20 mg/ml. All other monomer A.C. Foss, N.A. Peppas / European Journal of Pharmaceutics and Biopharmaceutics 57 (2004) 447–455 feed ratio P(MAA-g-PEG) nanospheres caused no significantNADPH production loss. Thus, as long as the monomer feedratio was kept at or below 2:1 MAA:EG, the copolymer gelcarriers could be expected to cause no decrease in cellviability, as measured by NADPH production.
shows the cell viability versus the concentration of P(AA-g-PEG) gel nanospheres, prepared by a dispersionpolymerization technique, in contact with Caco-2 cellcultures. The P(AA-g-PEG) gel nanospheres with monomerfeed ratios of 1:2 MAA:EG repeating units caused littledecrease in the NADPH production of the cell culture. Itcaused only a 10% loss in NADPH even at the highestconcentration of nanospheres that was 20 mg/ml. Unlike theP(MAA-g-PEG) gel nanospheres, the copolymer with a Fig. 5. TEER values of Caco-2 cell culture monolayers in contact with monomer feed ratio of 1:1 AA:EG caused a significant P(MAA-g-PEG) with a monomer feed ratio of 1:1 MAA:EG nanospheres decrease in the production of NADPH at concentrations of (A), microparticles ( # ) and control (W) as a function of time. TEER values 10 and 20 mg/ml. The P(AA-g-PEG) gel nanospheres with a expressed as a ratio of measured resistance value to pre-experimentalresistance value ðn ¼ 3Þ: monomer feed ratio of 2:1 MAA:EG caused no decrease inNADPH production. The P(AA-g-PEG) gel nanospheres as a function of contact time to record the different effects with a monomer feed ratio of 4:1 MAA:EG caused a caused by the copolymer gel carriers. presents the significant loss in NADPH production.
results of the TEER experiments conducted with P(MAA-g- The P(AA-g-PEG) gel nanospheres with a monomer feed PEG) gels with monomer feed ratios of 1:1 MAA:EG ratio of 2:1 AA:EG caused no significant loss in cell repeating units. A control study was also plotted as a viability. The lack of effect on the cell viability meant the reference on . As each experiment used different P(AA-g-PEG) gel nanospheres with a monomer feed ratio cellular monolayers, the initial TEER value for each of 2:1 AA:EG could be good candidates for oral insulin Transwell was measured and compared to the experimentally delivery devices. This conclusion is based on the fact that measured TEER values. The TEER value of the cellular copolymers with increased amounts of AA show an monolayer did not decrease below 80% of the initial TEER increased response to pH changes, since AA can disas- during the 6 h studied in any of the cases shown in The P(MAA-g-PEG) gel nanospheres with a monomer feed ratioof 1:1 MAA:EG caused a decrease in the TEER values more 3.3. Transepithelial electrical resistance quickly than the P(MAA-g-PEG) gel microparticles. Mostlikely, nanospheres had a shorter calcium ion binding time It is well known that the transepithelial electrical resistance which affected the opening of the tight junctions and led to (TEER) measures the resistance to electric current across a layer of cells. In these studies, a monolayer of Caco-2 cells was presents the results of the TEER experiments grown on a porous membrane and the TEER value of the cell conducted with P(AA-g-PEG) gel carriers with monomer monolayer was measured while the culture was in contact withcopolymer gel carriers in cell media. The integrity of the tightjunctions between the cells could influence transport acrossthe cell monolayer via the paracellular route . TheP(MAA-g-PEG) gel microparticles were shown by Torres-Lugo et al. to decrease the TEER values depending on theconcentration of copolymer gel microparticles. The decreasein TEER value was attributed to the opening of the tightjunctions via the calcium ion binding done by the P(MAA-g-PEG) gel microparticles. Calcium ions in solution have beenshown to be very important to the integrity of the tightjunctions. Any significant loss of calcium concentrationopens the tight junctions . Madsen and Peppas showed the P(MAA-g-PEG) gel copolymers to have asignificant calcium-binding ability.
The TEER studies in this work investigated the four best Fig. 6. TEER values of Caco-2 cell culture monolayers in contact withP(AA-g-PEG) with a monomer feed ratio of 2:1 AA:EG nanospheres ( candidates that were selected after the cytotoxicity studies.
microparticles ( # ) and control (W) as a function of time. TEER values Each copolymer gel carrier was placed in the medium in expressed as a ratio of measured resistance value to pre-experimental contact with the cell culture. The TEER value was measured A.C. Foss, N.A. Peppas / European Journal of Pharmaceutics and Biopharmaceutics 57 (2004) 447–455 feed ratios of 2:1 AA:EG repeating units. It was observed a transport study performed without copolymer added to the that the P(AA-g-PEG) gel carriers with monomer feed ratios apical chamber. It must be noted that these results and the of 2:1 AA:EG both caused a reduction in the TEER value by associated analysis of Eq. (1) are only approximate as they nearly 30 – 35% after 6 h. This result was much larger than refer to steady state conditions in the apical and basolateral the 10% observed with the P(MAA-g-PEG) gel carrier in sides. Considering the very low permeability values for As in the copolymer gel nanospheres caused a these large proteins, the variation in drug concentration in quicker reduction in the TEER values, but after a few hours the microparticles produced a similar effect on the TEER The presence of these copolymer gel carriers increased the permeability of insulin across the monolayer. The The ability of the copolymer gel carriers to increase the P(AA-g-PEG) gel microparticles with monomer feed ratios transport of insulin across the Caco-2 cell monolayer was of 2:1 AA:EG repeating units caused the greatest enhance- investigated by conducting insulin transport studies. Torres- ment of insulin transport, nearly doubling that of the control Lugo et al. identified the mechanism of insulin value. The P(MAA-g-PEG) gel nanospheres had a smaller transport across a Caco-2 cell monolayer as paracellular impact on the transport of insulin across the monolayer than transport. Borchardt et al. had previously shown that all other copolymer gel carriers. The P(MAA-g-PEG) gel chelating calcium in solution caused a significant increase in microparticles exhibited greater enhancement of insulin paracellular permeability via the opening of the tight transport compared to P(MAA-g-PEG) nanospheres as Insulin concentration in the apical chamber was measured The copolymer gel particles of P(AA-g-PEG) had a every hour to monitor the insulin there. Small samples were greater effect on the insulin permeability than the P(MAA- taken from the basolateral chamber for insulin analysis. The g-PEG) particles. This increased effect on the permeability 10 ml sample of the total 2.5 ml was deemed small enough to could be explained by their quicker swelling dynamics and not affect the overall insulin concentration in the basolateral higher equilibrium-swelling ratio. The hydrogen bounding chamber. The studies were conducted with 3 wells for each complexation in the P(AA-g-PEG) system was not as strong copolymer gel carrier system. The insulin permeated through as the P(MAA-g-PEG) due to the lack of the methyl group the cell monolayer to the basolateral chamber versus time to stabilize the hydrogen bonding that occurred Thus, was plotted and the slope, dQ=dt; was determined. Then, the the P(AA-g-PEG) copolymer probably bound calcium faster insulin permeability could be calculated from Eq. (1).
and caused a quicker opening of the tight junctions than the m was the permeability of the insulin across the Caco-2 cell monolayer. A was the surface areaof the membrane the cells were grown on and C We studied the behavior of P(MAA-g-PEG) and the concentration of insulin in the apical chamber. The P(AA-g-PEG) nanospheres and microparticles prepared surface area of the membrane was 4.71 cm2. The concen- by free radical polymerization of comonomer feed ratios of 4:1, 2:1, 1:1 and 1:2 MAA:EG or AA:EG units. The cell viability studies indicated that all copolymer gel The permeability of insulin across the Caco-2 cell carriers studied had a negligible effect on the NADPH monolayer in the presence of the copolymer gel carriers production of Caco-2 cell culture. This suggests that studied is presented in The control experiment was these materials would be good candidates for oral insulindelivery devices. Cell viability studies indicated that gel carriers with high MAA:EG or AA:EG monomer feed Permeability of insulin across Caco-2 cell monolayers in contact with ratios, those of 4:1 MAA:EG or AA:EG, caused a significant decrease in NADPH production and would not be a suitable candidate for an oral insulin deliverydevice. The best candidates for an oral insulin delivery system were the P(MAA-g-PEG) gel carriers prepared with monomer feed ratios of 1:1 MAA:EG and the P(AA-g-PEG) gel carriers prepared with monomer feed TEER studies determined the copolymer gel carriers’ ability to open the tight junctions between cells in a Caco-2 cell monolayer. P(AA-g-PEG) gel carriers had a greater ability to disrupt the tight junctions of the Caco-2 cell monolayer than the P(MAA-g-PEG) gel carriers.
A.C. Foss, N.A. Peppas / European Journal of Pharmaceutics and Biopharmaceutics 57 (2004) 447–455 Insulin transport across Caco-2 cell monolayers showed [16] G. Borchard, H.L. Luessen, A.G. de Boer, J.C. Verhoef, C.M. Lehr, that all copolymer gel carriers contributed to an increase in H.E. Junginger, The potential of mucoadhesive polymers in enchan-cing intestinal peptide drug absorption. III: effects of chitosan- insulin permeability. The copolymer gel carrier with the glutamate and carbomer on epithelial tight junctions in vitro, greatest effect was the P(AA-g-PEG) gel microparticles J. Control. Release 39 (1996) 131 – 138.
with a monomer feed ratio of 2:1 AA:EG prepared by in situ [17] R.T. Borchardt, G.M. Pauletti, F.W. Okumu, Effect of size and charge polymerization of thin films that were crushed to obtain on the passive diffusion of peptides across caco-2 cell monolayers via the paracellular pathway, Pharm. Res. 14 (1997) 164 – 168.
[18] R.T. Borchardt, T.J. Raub, I.J. Hidalgo, S.L. Kuentzel, Characteriz- ation of the caco2 cell line as an intestinal epithelial model system,Fed. Am. Soc. Exp. Biol. J. 2 (1988) A734.
[19] R.T. Borchardt, I.J. Hidalgo, T.J. Raub, Permeability properties of an intestinal epithelial model system (caco-2 cell), Fed. Am. Soc. Exp.
Biol. J. 2 (1988) A733.
This work was supported in part by NIH grant EB [20] R.T. Borchardt, K.M. Hillgren, A. Kato, In-vitro systems for studying intestinal drug absorption, Med. Res. Rev. 15 (1995) 83 – 109.
[21] H. Yu, T.J. Cook, P.J. Sinko, Evidence for diminished functional expression of intestinal transporters in caco-2 cell monolayers at highpassages, Pharm. Res. 14 (1997) 757 – 762.
[22] A.M. Lowman, M. Morishita, M. Kajita, T. Nagai, N.A. Peppas, Oral delivery of insulin using pH-responsive complexation gels, J. Pharm.
[1] F. Banting, H. Best, Pancreatic extracts, J. Lab. Clin. Med. 7 (1922) [23] J. Klier, N.A. Peppas, Complex-forming hydrogels sensitive to [2] J.R. Campbell, New Opportunities in Diabetes Management, Miles physiological conditions, Proc. Adv. Biomed. Polym. 1 (1989) [3] A. Lowman, N.A. Peppas, A complexation/decomplexation mechan- [24] C.L. Bell, N.A. Peppas, Complexation in P(EG-g-MAA) hydrogels, ism in ph-responsive copolymer networks, Polym. Prepr. 38 (1) Polym. Prepr. 34 (1) (1993) 831 – 832.
[25] N.A. Peppas, K.B. Keys, M. Torres-Lugo, A.M. Lowman, [4] A. Lowman, N.A. Peppas, Complexation graft copolymers as oral Poly(ethylene glycol) containing hydrogels in drug delivery, drug delivery systems, Polym. Prepr. 38 (2) (1997) 56 – 57.
J. Control. Release 62 (1999) 81 – 87.
[5] A. Lowman, N.A. Peppas, Pulsatile drug delivery based on a [26] M. Torres-Lugo, N.A. Peppas, Molecular design and in vitro studies complexation/decomplexation mechanism, in: S.M. Dinh, J.D.
of novel pH-sensitive hydrogels for the oral delivery of calcitonin, DeNuzzio, A.R. Comfort (Eds.), Intelligent Materials for Controlled Macromolecules 32 (1999) 6646 – 6651.
Release, ACS Symposium Series, Washington, DC, vol. 728, 1999, [27] N.A. Peppas, P. Bures, W. Leobandung, H. Ichikawa, Hydrogels in pharmaceutical formulations, Eur. J. Pharm. Biopharm. 50 (2000) [6] M. Torres-Lugo, N.A. Peppas, Preparation and characterization of [28] A.M. Lowman, N.A. Peppas, Hydrogels, in: E. Mathiowitz (Ed.), J. Nanoparticle Res. 4 (2002) 73 – 81.
Encyclopedia of Controlled Drug Delivery, Wiley, New York, NY, [7] D.N. Robinson, N.A. Peppas, Preparation and characterization of ph- responsive poly(methacrylic acid-g-poly(ethylene glycol) nano- [29] A.M. Lowman, N.A. Peppas, Molecular analysis of interpolymer spheres, Macromolecules 35 (2002) 3668 – 3674.
complexation in graft copolymer networks, Polymer 41 (2000) [8] C. Donini, D.N. Robinson, P. Colombo, F. Giordano, N.A. Peppas, Preparation of nanospheres for methacrylic monomers for pharma- [30] J.J. Sahlin, N.A. Peppas, Hydrogels as muco- and bioadhesive ceutical applications, Intern. J. Pharm. 245 (2002) 83 – 91.
materials: a review, Biomaterials 17 (1996) 1553 – 1561.
[9] M. Torres-Lugo, M. Garcia, M. Record, N.A. Peppas, pH-Sensitive [31] M. Joshi, A. Misra, Dry powder inhalation of liposomal Ketotifen hydrogels as gastrointestinal tract absorptions enhancers: transport fumarate: formulations and charaterization, Intern. J. Pharm. 223 mechanisms of salmon calcitonin and other model molecules using the caco-2 cell model, Biotechnol. Progr. 18 (2002) 612 – 616.
[32] A.M. Lowman, N.A. Peppas, Solute transport analysis in pH- [10] M. Torres-Lugo, M. Garcia, M. Record, N.A. Peppas, Physicochem- responsive, complexing hydrogels of poly(methacrylic-g-ethylene ical behavior and cytotoxic effects of P(MAA-g-EG) nanospheres for glycol), J. Biomater. Sci. Polym. Ed. 10 (1999) 999 – 1009.
oral delivery of proteins, J. Control. Release 80 (2002) 197 – 205.
[33] S. Yee, In vitro permeability across caco-2 cells(colonic) can it predict [11] R.A. Scott, N.A. Peppas, Highly crosslinked, PEG-containing in vivo (small intestinal) absorption in man-fact or myth, Pharm. Res.
copolymers for sustained solute delivery, Biomaterials 20 (1999) [34] B. Gumbiner, Biochemistry and assembly of epithelial tight junctions, [12] J.H. Ward, A. Shahar, N.A. Peppas, Kinetics of living radical Am. J. Physiol. 253 (1987) C749 – C758.
polymerizations of multifunctional monomers, Polymer 43 (2002) [35] R.T. Borchardt, P.S. Burton, R.A. Conradi, N.F.H. Ho, A.R. Hilgers, How structural features influence biomembranes permeability of [13] R.A. Scott, N.A. Peppas, Kinetics of copolymerization of PEG- peptides, J. Pharm. Sci. 85 (1996) 1336 – 1340.
containing multiacrylates with acrylic acid, Macromolecules 32 [36] M. Torres-Lugo, H. Ichikawa, N.A. Peppas, Cytotoxicity and transport enhancement of proteins through cell monolayers using [14] R.A. Scott, N.A. Peppas, Compositional effects on network structure novel pH-sensitive hydrogels, Proc., World Meet. APV/APGI 3 of highly crosslinked copolymers of PEG-containing multiacrylates with acrylic acid, Macromolecules 32 (1999) 6139 – 6148.
[37] A.W. Sedar, J.G. Forte, Effect of calcium depletion on the junctional [15] N.A. Peppas, Y. Huang, M. Torres-Lugo, J.H. Ward, J. Zhang, complex between oxyntic cells of the gastric glands, J. Cell Biol. 22 Physicochemical foundations and structural design of hydrogels in medicine and biology, Ann Rev. Biomed. Engng 2 (2000) [38] F. Lacaz-Vieira, Calcium site specificity, early Ca þ 2-related tight junction events, J. Gen. Physiol. 110 (1997) 727 – 740.
A.C. Foss, N.A. Peppas / European Journal of Pharmaceutics and Biopharmaceutics 57 (2004) 447–455 [39] F. Madsen, N.A. Peppas, Complexation graft copolymer networks: [47] H. Ichikawa, N.A. Peppas, Novel complexation hydrogels for oral swelling properties, calcium binding, and proteolytic enzyme peptide delivery: in vitro evaluation of their cytocompatibility and inhibition, Biomaterials 20 (1999) 1701 – 1708.
insulin-transport enhancing effects using caco-2 cell monolayers, [40] R.T. Borchardt, G. Knipp, N. Ho, C. Barsuhn, Paracellular diffusion in J. Biomed. Mater. Res. 67 (2003) 609 – 617.
Caco-2 cell monolayers: effect of perturbation on the transport of [48] J.E. Lo´pez, N.A. Peppas, Cellular evaluation of insulin transmucosal hydrophilic compounds that vary in charge and size, J. Pharm. Sci. 86 delivery, J. Biomater. Sci., Polym. Ed. (2004) in press.
[49] J.E. Lo´pez, N.A. Peppas, Effect of PEG molecular weight and [41] B. Kim, N.A. Peppas, Synthesis and characterization of pH-sensitive microparticle size on oral insulin delivery from P(MAA-g-EG) glycopolymers for oral drug delivery systems, J. Biomater. Sci. 13, microparticles, Drug Dev. Ind. Pharm. (2004) press in press.
[50] N.A. Peppas, Devices based on intelligent biopolymers for oral [42] O. Sipahigil, M. Torres-Lugo, N.A. Peppas, FTIR spectroscopic protein delivery, Intern. J. Pharm. (2004) in press.
analysis of protein/carrier interactions in novel protein delivery [51] R. Langer, N.A. Peppas, Advances in biomaterials, drug delivery, systems, STP Pharma 12 (2002) 345 – 350.
and bionanotechnology, Am. Inst. Chem. Eng. J. 49 (2003) [43] B. Kim, K. La Flamme, N.A. Peppas, Dynamic swelling behavior of pH-sensitive anionic hydrogels used for protein delivery, J. Appl.
[52] K. Nakamura, R.J. Murray, J.I. Joseph, N.A. Peppas, M. Morishita, Polym. Sci. 89 (2003) 1606 – 1613.
A.M. Lowman, Oral insulin delivery using P(MAA-g-EG) hydrogels: [44] B. Kim, N.A. Peppas, Analysis of molecular interactions in P(MAA- effects of network morphology on insulin delivery characteristics, g-EG) hydrogels, Polymer 44 (2003) 3701 – 3707.
J. Control. Release 95 (2004) 589 – 599.
[45] B. Kim, N.A. Peppas, PEG-containing hydrogel microparticles for [53] M. Morishita, T. Goto, N.A. Peppas, J.I. Joseph, M.C. Torjman, C.
oral protein delivery applications, Biomed. Microdevices 5 (2003)333 – 341.
Munsick, K. Nakamura, K. Takayama, A.M. Lowman, Mucosal [46] B. Kim, N.A. Peppas, In vitro release behavior and stability of insulin insulin delivery systems based on complexation polymer hydrogels: in complexation hydrogels as oral drug delivery carriers, Intern.
effect of particle size on insulin enteral absorption, J. Control. Release

Source: http://web67.cc.utexas.edu/research/biomat/PDFReprints/EurJPharmBio_57_447-455(2004).pdf