Chitosan Ascorbate: A Chitosan Salt with Improved

Pharmaceutical Development and Technology, 13:513–521, 2008 Copyright © Informa UK, Ltd. ISSN: 1083-7450 print / 1097-9867 online DOI: 10.1080/10837450802288865

LPDT

Chitosan Ascorbate: A Chitosan Salt with Improved Penetration Enhancement Properties Silvia Rossi, Marzia Marciello, Giuseppina Sandri, Maria Cristina Bonferoni, Franca Ferrari, and Carla Caramella

Pharmaceutical Development and Technology Downloaded from informahealthcare.com by University of Guelph on 04/18/13 For personal use only.

Chitosan Ascorbate as Penetration Enhancer

Department of Pharmaceutical Chemistry, University of Pavia, Viale Taramelli, Pavia, Italy

The aim of the present work was to investigate if chitosan salification with ascorbic acid could produce an increase in chitosan penetration enhancement properties towards buccal mucosa and intestinal Caco-2 cell monolayer. Three different chitosan grades were considered. Chitosan hydrochloride and lactate were used as references. Fluorescein isothiocyanate-dextran (FD4), a hydrophilic high MW molecule, was employed as model penetrant. Chitosan ascorbate showed better penetration enhancement properties towards both buccal porcine mucosa and Caco-2 cell monolayer with respect to hydrochloride and lactate salts. Cytotoxicity of chitosan ascorbate assessed on Caco-2 cells was comparable with those of chitosan hydrochloride and lactate. Keywords chitosan ascorbate, penetration enhancer, buccal mucosa, Caco-2 cell line, fluorescein isothiocyanate dextran

INTRODUCTION Chitosan, the N-deacetyl product of chitin, is a dietary fiber with potent hypocholesterolemic effects and able to interact with acidic and neutral steroids in the intestinal lumen to increase their fecal excretion. Recently it has also been reported that the addition of ascorbic acid to chitosan causes a large increase in fecal fat excretion.[1–3] In a study performed on rats Kanauchi el al. evaluated the action of different chitosan salts (chitosan ascorbate, lactate and citrate) on absorption and fat fecal excretion, using cellulose and unsalified chitosan as references.[3] The salification of chitosan with ascorbic acid resulted very promising:

Received 30 April 2008, Accepted 16 June 2008. Address correspondence to Prof. Carla Caramella, Department of Pharmaceutical Chemistry, University of Pavia, Viale Taramelli 12, 27100 Pavia, Italy; E-mail: [email protected]

chitosan ascorbate produced a larger increase in fecal excretion without affecting protein digestibility. In a subsequent work, the same authors studied the mechanism of chitosan inhibition of lipid digestion and of synergism with ascorbic acid. They concluded that chitosan dissolves into gastric fluid and becomes a gel able to entrap lipids of the intestinal lumen.[4]. The synergic effect of ascorbic acid was not acid-dependent, but due to the specificity of ascorbic acid: comparable results were obtained employing chitosan preparations enriched with sodium ascorbate. In a more recent work Tsujikawa et al. designed a pilot trial to investigate the tolerability and the amount of fecal excretion after the oral administration of a chitosan and ascorbic acid mixture for inactive Crohn’s disease.[2] The results indicated that the oral administration of such a mixture in patients with inactive Chron’s disease was tolerable and effectively increased fecal fat excretion. The mechanism for increasing fecal fat excretion by coexisting chitosan and ascorbic acid remains unclear. Some authors suggested that chitosan-ascorbic acid synergic effect is due to an increase in chitosan capability to interact and entrap dietary lipids by an emulsifying process mediated by ascorbic acid.[2] In the last years, chitosan has been proved to possess drug penetration enhancement properties towards buccal mucosa.[5–7] Senel et al. proved that a chitosan solubilized in a diluted solution of lactic acid was able to increase the absorption of a model peptide across porcine buccal mucosa.[5] Also chitosan hydrochloride have been demonstrated to possess penetration enhancement properties towards buccal mucosa.[6,7] Chitosan penetration enhancement properties were affected by the polymer molecular weight.[7] It is generally recognized that the main obstacle to drug penetration across buccal mucosa via paracellular route is represented by extracellular lipids. Some authors suggested that chitosan penetration enhancement properties are due to chitosan interaction with the intracellular lipid

513


514

S. Rossi et al.

domain.[5] Given these premises, primary aim of the present work was to investigate if chitosan salification with ascorbic acid could produce an increase in chitosan penetration enhancement properties towards buccal mucosa. In particular, three different chitosan grades having different molecular weights were considered. They were salified with ascorbic, hydrochloric and lactic acids. Chitosan hydrochloride and lactate, well-known as buccal penetration enhancers, were used as references. Fluorescein isothiocyanate dextran (FD4), a hydrophilic molecule having a high molecular weight (4400 kD), similar to that of peptidic drugs, was used as model molecule. In the past chitosan has been also proved to possess penetration enhancement properties towards intestinal epithelium.[8–11] The mechanism suggested is completely different from that supposed for buccal penetration enhancement. In fact, the main barrier to the passage of hydrophilic high molecular weight drugs in intestinal epithelium is represented by tight junctions (TJs), absent from buccal mucosa. Experimental evidences have been given that chitosan was able to open TJs.[10,11] In the present work the eventual capability of the association of chitosan with ascorbic acid to improve polymer penetration enhancement properties towards Caco-2 cell monolayer was also investigated. Moreover, cytotoxicity of chitosan ascorbate was assessed on Caco-2 cells and compared with those of chitosan hydrochloride and lactate.

EXPERIMENTAL PART Polymers The following Chitosan (CS) grades were considered: hMW (1568): deacetylation degree (DD) 91%; mMW (1961): DD 92%: lMW (1504): DD 98% (Giusto Faravelli, Milan, Italy) CS hMW was salified with: ascorbic acid (Carlo Erba Reagenti, Milan, Italy), lactic acid (80%, A.C.E.F., Piacenza, I), hydrochloric acid (Carlo Erba Reagenti, Milan, Italy). CS mMW and lMW were salified with ascorbic acid and hydrochloric acid.

Model Molecule Fluorescein isothiocyanate dextran (MW 4400) (FD4) (Sigma Aldrich, Milan, Italy) was used as model molecule.

Determination of MWv of the Three Chitosan Grades CS solutions in 0.1 mol l−1 CH3COONa – 0.2 mol l−1 CH3COOH were prepared. The following chitosan

concentrations were considered: 0.1 mg/mL, 0.2 mg/mL, 0.3 mg/mL, 0. 4 mg/mL e 0.5 mg/mL for hMW grade; 0.2 mg/mL, 0.3 mg/mL, 0.4 mg/mL, 0.0005 g/mL and 0.6 mg/mL for mMW; 1 mg/mL, 1.4 g/mL, 1.6 g/mL, 1.8 g/mL and 2 mg/mL for lMW. An Ostwald viscometer, thermostated at 30 ± 0.5°C, was used. The efflux times of each polymer solution and of hydration medium were measured. For each polymer solution the reduced viscosity (ηrid) was calculated.[12] For each CS grade, intrinsic viscosity value was calculated as intercept on x axis of the straight line ηrid vs. concentration. The mean viscosimetric molecular weight (MWv) was calculated on the basis of Mark Houwink equation as reported in [12]:

[hint ] = KMa

(1)

where: K = 1.64 × 10 −30 × DD14 , a = −1.02 × 10 −2 × DD + 1.82 Preparation of Chitosan Solutions An exact amount of each acid (to obtain a 1:1 molar ratio acid: CS deacetylated amine groups) was dissolved/ diluted in/with distilled water. For rheological, mucoadhesive and buccal permeation measurements, CS was added to obtain 2.5% (w/w) polymer solutions. pH values of CS solutions range into interval 4.5–5.5. For permeation measurements across Caco-2 cell monolayer and toxicity studies, HBSS (Hanks’ balanced salt solution: CaCl2 anhydrous 140 mg/l, MgCl2.6H2O 100 mg/L, MgSO4.7H2O 100 mg/L, KCl 400 mg/L, KH2PO4 60 mg/L, NaHCO3 350 mg/L, NaCl 8000 mg/L, Na2HPO4 48 mg/L, D-glucose 1000 mg/L, Phenol Red 10 mg/l, Gibco-BRL, NY, USA) buffered at pH 5.5 with HCl 1 N was used instead of distilled water and CS was added at 0.4% (w/w) concentration. FD4 was added at each solution at 0.2% (w/w) concentration.

Rheological Analysis Each polymer solution was subjected to a complete rheological characterization by means of a rotational rheometer (Rheostress 600, Haake, Spinea, Italy). A cone plate combination (C35) was used as a measuring system. All measurements were carried out at 37°C, after a rest time of 3 min. The apparent viscosity was measured on increasing shear rate values ranging into the interval 20–300 s−1. Dynamic oscillatory tests were performed in the linear viscoelastic range. The viscoelastic parameters, storage modulus (G′) and loss modulus (G′′), were measured at frequency values ranging from 0.1–10 Hz. Loss tangent (tgδ) was calculated as the ratio between G′′ and G′ values.



Release Measurements In vitro FD4 release was assessed by means of a Franz diffusion cell (Permegear, Bethlehem, PA, USA) with a 20 mm diameter orifice (3.14 cm2 area)). The donor and the receptor chambers were separated by a filter membrane (HA 0.45 μm, Millipore, Milan, Italy). Each polymer solution (100 mg) was spread on a circular portion (2 cm2) of the filter membrane. 500 μl of pH 6.4 phosphate buffer (USP 31) were added in the donor chamber, over the polymer solution, to simulate the buccal environment. pH 7.4 saline isotonic solution (KH2PO4 1.90 g/L; Na2HPO4 8.10 g/L; NaCl 4.11 g/L) was used as receptor phase. After 5 h, 500 μl of the acceptor phase were withdrawn. The FD4 released amount was assayed by means of a spectrofluorimetric method (LS50B spectrofluorimeter, Perkin Elmer, Milan, I, λex: 490 nm; λem: 515 nm).

Permeation Measurements Across Porcine Buccal Mucosa The penetration enhancement properties of each polymer solution were evaluated at 37°C using porcine buccal mucosa as biological substrate. A circular epithelium membrane of 5 cm2 area, obtained as described in [6], was placed between the donor and the receptor chamber of a Franz diffusion cell (Permegear, Bethlehem, PA, USA) with a 20 mm diameter orifice (3.14 cm2 area). Each polymer solution (100 mg) was applied on a circular portion (2 cm2) of the epithelium membrane. 500 μl of pH 6.4 phosphate buffer (USP 31) were added over the polymer solution in the donor chamber, to simulate the buccal environment, whereas pH 7.4 saline isotonic solution was used as acceptor phase. At fixed time intervals, 500 μl of the acceptor phase were withdrawn and replaced with fresh buffer. The drug permeated was assayed by means of the spectrofluorimetric method, above mentioned. The permeation test was also performed using 100 μl of a 0.20% (w/w) FD4 solution prepared in distilled water (reference). To take into account the different release properties of the polymer solutions, the drug amount permeated or penetrated at the end of the experiment (5 h) was normalized with respect to the drug amount released at the same time. The value obtained was expressed as percentage (% drug permeated/released).

515

Italy) FD4 penetration into porcine buccal mucosa was also evaluated (13). In particular an apparatus consisted of a circular plexiglas holder and of a Franz cell cup was used. 100 mg of each polymer solution and of FD4 solution were layered on a circular portion of a freshly excised porcine buccal epithelium (surface area: 3.14 cm2). The buccal epithelium was placed on a filter paper disc, laid on a circular holder. A Franz cell cup was put on the epithelium and then 500 μl of pH 6.4 phosphate buffer (USP 31) were placed on the sample. The apparatus was sealed with parafilm membrane and placed in an oven at 37°C for 5 h. At the end of the experiment, samples were removed from the mucosa which was rinsed twice with physiological solution. Then mucosa was included in the OTC compound (Leica Microsystem, Milan, Italy), frozen in liquid nitrogen and stored at −80°C. Slices perpendicular to the mucosa surface, 25 μm in thickness, were cut using a cryostat (Leica CM1510, Leica Microsystem, Milan, Italy) at −20°C. Each slice was placed on a microscope slide, dehydrated for 12 h and subsequently fixed by dipping the microscope slides in acetone. The nuclei of the tissue slice were stained by dipping the biological substrates into a solution (1:100000) of Propidium Iodide (Sigma Aldrich, Milan, Italy). Each microscope slide, with mucosa slice, was mounted using PVA-DABCO (polyvinyl alcohol mounting medium with DABCO antifading, BioChemika, Fluka, Milan, I) and covered with a cover glass. The slides were observed using a Confocal Laser Scanning Microscope (Leica TCS SP2, Leica Microsystems, Milan, I) using λex = 485 nm and λem = 515 nm for the visualization of FD4 and λex = 520 nm and λem = 625 nm for the visualization of Propidium Iodide. The acquired images were processed by means of a software (Leika Microsystem, Milan, Italy). To obtain depth information from specific sections (xz- and yz-sections), the confocal images of xy planes were first acquired (parallel to the plane of the slice surface). To generate the xz- and the yz-sections, two horizontal lines were drawn across the regions of interest in the z = 0 μm-xy- plane, and the digitalized image data of the successive xy-sections were optically sliced through along the z-axis. The results were the xz- and yz planar optical cross-sections.[14]

Mucoadhesion Measurements Penetration Measurements into Porcine Buccal Mucosa By means of Confocal Laser Scanning Microscopy (CLSM) (Leica TCS SP2, Leica Microsystems, Milan,

Mucoadhesive properties of polymer solutions were determined by means of a tensile stress tester previously described.[7] Porcine gastric mucin (Mucin type II, Sigma Aldrich, Milan, Italy) was employed as biological substrate.


516

S. Rossi et al.

100 mg of each polymer solution was layered on a filter paper disc (area = 2 cm2) and fixed on the movable carriage of the apparatus. 50 μl of 8% (w/w) mucin dispersion in pH 6.4 phosphate buffer (USP 31) was layered on the filter disc placed on the sample holder. A preload of 2500 mN was applied in order to allow the formation of the mucoadhesive joints. After 3 min rest, the preload was removed and the movable carriage was moved at a constant speed (4 mm/min) up to the complete separation of the two surfaces. Both the displacement of the movable carriage and force of detachment data were recorded and simultaneously collected on a personal computer. The parameter work of adhesion (AUC) was calculated as the area under the force of detachment vs. displacement curve by means of the trapezoidal rule.

Permeation Measurements Across Caco-2 Cell Monolayer Caco-2 cells (passage 37) were seeded on tissue-culturetreated polycarbonate filters (culture surface: 33.6 mm2) in 24-well culture plates (Greineger Bio-one, PBInternational, Milan, Italy) at a seeding density of 105 cells/cm2. Dulbecco’s modified Eagle’s medium (DMEM, pH 7.4, Bioindustries, Israel) supplemented with 10% foetal bovine serum, benzyl penicillin G (160 U/mL) and streptomycin sulphate (100 μg/mL (Bioindustries, Israel) and also with 1% nonessential aminoacids (Sigma Aldrich, Milan, Italy) was used as culture medium. Cell cultures were kept at 37°C in an atmosphere of 95% air and 5% CO2. Filters were used for transepithelial electrical resistance (TEER) and FD4 transport experiments 23 days after seeding. 250 μl of each sample were added on the apical side of the monolayer. At fixed time intervals, the receptor phase was taken from the basolateral side and replaced with fresh HBSS. FD4 was dosed in the receptor phase by means of the above mentioned spectrofluorimetric method. At the end of the experiment, the apical phase was replaced by 250 μl of 25 μg/mL Lucifer Yellow (LY) (Sigma Aldrich, Milan, Italy) solution in HBSS. After 1h contact, LY permeated was assayed by means of a spectrofluorimetric method ((LS50B spectrofluorimeter, Perkin Elmer, Milan, Italy) λex: 428 nm; λem: 521 nm).

In-vitro Cytotoxicity Study Caco-2 cells (passage 37) were seeded in 96-well plates with area of 0.34 cm2 at density 105 cells/cm2. Supplemented DMEM, pH 7.4 was used as culture medium.

Cell cultures were kept at 37°C in an atmosphere of 95% air and 5% CO2. After seven days, the growing cells attached to the well bottom and composed a monolayer. On the 8th day, experiments were performed. The toxicity study was performed using the neutral red (NR) assay (Tox Kit 4, Sigma Aldrich, Milano Italy) that provides to determine the accumulation of the NR supravital dye in the lysosomes of viable, uninjured cells.[15,16] Damaged cell membranes or lysosomes cause a poor capability or no capability to pick up NR. Each well was washed with saline phosphate buffer (PBS) and 280 μl of each polymer solution 0.4% (w/w) prepared in HBSS at pH 5.5 were put in contact with the cells. After 1 and 2 h the samples were removed and the cell substrates washed with PBS. 200 μl of NR solution (0.33 mg/mL in DMEM) were put in each well with 2 h of contact time. Cell substrates were then washed with PBS and then with the fixing medium (1% w/v CaCl2 and 0.5% w/v formaldehyde aqueous solution) in order to remove NR not entrapped in the cells and to fix the substrate. The fixing solution was then removed and a solubilizing solution (1% v/v of acetic acid in ethanol) was added to each cell substrate to cause cell disruption and, consequently, to produce the release of NR captured by viable cells. The NR solution absorbance was determined by means of ELISA plate reader (Perkin Elmer, Milan, Italy) at a wavelength of 490 nm with 650 nm wavelength filter. The absorbance for each sample was compared with that of HBSS, negative control (not toxic), that was considered to coincide with the maximum viability (100%).

RESULTS AND DISCUSSION The mean MWv values of the three CS grades considered were: 251000 Da for lMW grade, 1163000 Da for mMW and 1855000 Da for hMW.

Rheological Analysis In Figure 1 viscosity and loss tangent values observed for the solutions of CS hMW grades salified with ascorbic, hydrochloric and lactic acids are reported. CS lactate solution shows the highest viscosity profile, while CS hydrochloride and CS ascorbate solutions are characterized by comparable profiles. CS ascorbate solution shows loss tangent values higher than 1. Since such a parameter is calculated from the ratio between the viscoelastic parameters G′′ (loss modulus) and G′ (storage modulus), it means that CS ascorbate is characterized by a prevalence of the viscous component on the elastic one. CS hydrochloride and

Viscosity (Pa.s)

Chitosan Ascorbate as Penetration Enhancer 18

CS ascorbate

15

CS hydrocloride

12

CS lactate

Permeation Measurements Across Porcine Buccal Mucosa In Figure 2 the FD4 permeated amount vs. time profiles observed for the aqueous solutions of all the CS grades and salts and of a solution of FD4 alone (reference) are reported. lMW CS ascorbate solution shows a profile comparable to that observed for the solution of FD4 alone, it indicates the lack of penetration enhancement properties. On the contrary, mMW CS ascorbate solution is characterized by penetration enhancement properties, evidenced by a FD4 permeation profile higher than that observed for FD4 solution. Higher penetration enhancement properties are shown by hMW CS ascorbate which causes the permeation of a FD4 amount 13 times higher than that observed for the solution of FD4 alone. The solutions of lMW and mMW CS hydrochloride are capable to promote FD4 permeation across buccal mucosa. hMW CS hydrochloride is characterized by a permeation profile comparable to that observed for the solution of FD4 alone. The higher profile is observed for mMW grade which is able to increase FD4 permeated amount six times with respect to the FD4 solution without CS. Among CS hMW grades, the higher profile is observed for CS salified with ascorbic acid. The poor or absent penetration enhancement properties observed for hMW CS grades salified with lactic and hydrochloric acid could be due to the high entanglement of hydrated polymer chains, evidenced by the greater elastic behaviour of the CS lactate and hydrochloride solutions with respect to the CS ascorbate one. A high polymer chain entanglement can cause a shielding of the functional groups (protonated amine groups) responsible for the interaction of the macromolecular chains with the mucosa.

9 6 3 0

100 200 Shear rate (1/s)

300

30 hMW ascorbate

25

hMW hydrochloride loss tangent

20

hMW lactate

15 10 5 0

0

2

4 6 Frequency (Hz)

8

10

Figure 1. Viscosity and loss tangent profiles observed for the aqueous solutions of hMW CS salts (mean values ± SE; n = 3).

lactate show comparable loss tangent values, close to 1; it indicates an equilibrium between the two viscoelastic components. These results mean that CS lactate and hydrochloride are characterized by a more entangled structure with respect to CS ascorbate.

10 30 FD4 permeated amount (µg)


0

517

5

25

0

20

0

1

2

3

4

FD4 solution hMW CS ascorbate mMW CS ascorbate lMW CS ascorbate hMW CS HCl mMW CS HCl lMW CS HCl hMW CS lactate

5

15 10 5 0

0

1

2

3

4

5

Time (h)

Figure 2. FD4 permeation profiles observed for the solutions of the all CS salts and of a solution of FD4 (mean values ± SE; n = 6).

518

S. Rossi et al.

Penetration Measurements into Porcine Buccal Mucosa

% FD4 permeated/released

14 12 10 8 6 4 2


0

lMW mMW hMW lMW mMW hMW hMW CS ascorbate CS hydrochloride CS lactate

Figure 3. % FD4 permeated/released values of C salts (mean values ± SE {calculated according to the error propagation theory]).

In Figure 3, % FD4 permeated/released values of hMW CS salts are compared. Such a parameter expresses FD4 amount able to permeate mucosa with respect to FD4 amount released from polymer solution and then actually available on mucosal surface. These results confirm the best penetration enhancement properties of hMW CS ascorbate with respect to the other salts.

In Figures 4 CLSM photographs of porcine buccal mucosa placed in contact for 5 h with FD4 solution and hMW CS salts are reported. Green zones indicate the penetration depht of FD4, while colour intensity is correlated to the FD4 amount penetrated; red points are the cell nuclei coloured by propidium iodide. To better point out FD4 penetration into mucosa, CLSM and optical microscopic photographs of the same mucosa portion have been superimposed. FD4 solution produces a drug penetration into mucosa up to 40 μm depth; the low intensity of green colour indicates the presence of a poor amount of FD4 into mucosa (Figure 4a). In presence of CS ascorbate (Figure 4b), FD4 penetrates up to 100 μm depth; the intense green colour means the penetration of a higher amount of FD4 with respect to the solution of FD4 alone. CS hydrochloride and lactate show intermediate behaviours (Figures 4c, 4d), showing a penetration depth of 25 and 45 μm, respectively, accompanied by higher amounts of FD4 penetrated with respect to FD4 solution. The results obtained confirm the best penetration enhancement properties of CS ascorbate with respect to the other salts.

(a)

(b)

(c)

(d)

Figure 4. CLSM photograph of porcine buccal mucosa placed in contact for 5 h at 37°C with the solution of FD4 alone or of hMW CS salt. (a) Solution of FD4 alone; (b) CS ascorbate; (c) CS hydrochloride; (d) CS lactate. The image on the right results from superimposition of CLSM and optical microscopy photographs.


Mucoadhesion measurements

Permeation measurements across Caco-2 cell monolayer In Figure 6 TEER% values as a function of time are reported for the three hMW CS salts and for the solution of FD4 alone. All the CS solutions are characterized by a decrease in TEER % values on increasing time. It indicates the interaction of all the three salts with cell tight junctions (TJ). Such a decrease is higher for CS ascorbate solution with respect to CS hydrochloride and lactate solutions. 5000 Work of adhesion (AUC) (mN.mm)

In Figure 7, FD4 permeated amount vs. time profiles observed for the three CS salts and FD4 solution are reported. The salification with ascorbic acid produces an increase in FD4 amount permeated after 2 h with respect to the other two salts. Such an increase is lower with respect to that observed using buccal mucosa as biological substrate. CS lactate does not seem to possess penetration enhancement properties, showing a FD4 permeation profile comparable to those of FD4 solution. In Figure 8 Papp values of Lucifer Yellow (LY) calculated for all the three CS salts and FD4 solution are reported. Papp values of LY, low MW hydrophilic molecule, confirm the results obtained for CS ascorbate and hydrochloride solutions employing FD4 as penetrant molecule. Also CS lactate solution is able to promote LY absorption; it is, in fact, characterized by a Papp value higher than that observed in presence of FD4 solution (reference). This can be explained by the low molecular weight of LY with respect to FD4: CS lactate at the concentration employed is able to disturb TJ integrity (as evidenced by the decrease in TEER% [Figure 6]) producing a TJ opening sufficient to facilitate the permeation of LY but not of a high molecular weight molecule such as FD4.

4000 3000

In-vitro cytotoxicity study

°

2000

°°

1000 0 CS ascorbate CS hydrochloride

CS lactate

Figure 5. Work of adhesion (AUC) values observed for the three hMW salts (mean values ± SE; n = 6) vs. P < 0.001 Fisher test.

In Figure 9 cell viability % values observed for the three hMW CS salts are reported. Such values have been calculated by normalizing the cell viability observed in presence of the polymer solutions with respect to that observed in presence of HBSS alone. The test duration does not affect the results obtained: for each CS salt, not significant different cell viability % values are observed on increasing time. CS ascorbate solution is characterized at 1 and 2 h by a cell viability % comparable to those observed for the other two salts.

140 120

FD4 solution

100 TEER%


In Figure 5 work of adhesion (AUC) values obtained for the three hMW CS salts are reported. CS ascorbate is characterized by the highest AUC values, followed in decreasing order, by CS lactate and hydrochloride. This indicates a higher capability of CS when salified with ascorbic acid to interact with the biological substrate.

519

CS ascorbate

80

CS hydrochloride

60

CS lactate

40 20 0 0,0

0,5

1,0 Time (h)

1,5

2,0

Figure 6. TEER % vs. time profiles observed for the solutions of the three hMW CS salts (mean values ± SE; n = 6).

520

S. Rossi et al. 12

CS ascorbate

10 µg FD4 permeated

CONCLUSIONS

FD4 solution CS hydrochloride

8

CS lactate

6 4

0 0,0

0,5

1,0 Time (h)

1,5

2,0

Figure 7. Profiles of FD4 permeation across Caco-2 monolayer observed for the solutions of the three hMW CS salts (mean values ± SE; n = 6).

14

Papp.105 (cm/s)

12 10 8 6 4 2 0

FD4 solution

hMW hMW hMW lactate ascorbate hydrochloride

Figure 8. Papp values of LY calculated after 2 h for the solutions of the three hMW CS salts (mean values ± SE; n = 6).

100 80 cell viability %


2

1h

2h

60 40 20 0 CS ascorbate CS hydrochloride

CS lactate

Figure 9. % Cell viability observed after 1 and 2 h for the solutions of the three hMW CS salts (mean values ± SE; n = 6).

Depending on salt type, CS solutions are characterized by different viscosity and viscoelastic properties. In particular, hMW CS lactate solution is characterized by the highest viscosity profile, while CS ascorbate and hydrochloride solutions show comparable viscosity profiles. As for the elastic properties, the rank order between CS salts is: lactate > hydrochloride > ascorbate. These results mean a different degree of polymer chain entanglement between CS salts: lactate and hydrochloride salts are characterized by a more entangled structure whit respect to ascorbate one. FD4 permeation measurements evidenced that CS penetration enhancement properties are affected by salt type and polymer molecular weight. In particular, mMW and hMW CS ascorbate salts are able to promote the absorption of the model molecule FD4 (high MW hydrophilic molecule) across porcine buccal mucosa; the penetration enhancement properties are particularly marked for hMW grade, that permits the permeation of a FD4 amount 13 times higher with respect to that observed for the solution of FD4 alone. These results are confirmed by the values of the normalized parameter % FD4 permeated/released, which take into account the different release properties of CS solutions, and by CLSM analysis. Such analysis allowed us to evaluate the penetration depth of FD4 into porcine buccal mucosa. Mucoadhesion measurements have proved that hMW CS ascorbate is characterized by higher mucoadhesion properties with respect to the other salts. The permeation test performed on Caco-2 cell monolayer proved that hMW CS ascorbate, hydrochloride and lactate are able to interact with tight junctions: all the three CS solutions produce a decrease in TEER. hMW CS ascorbate is characterized after 2h by the highest FD4 permeated amount. Also CS hydrochloride promotes FD4 permeation, but to a lower extent; on the contrary CS lactate solutions does not possess penetration enhancement properties towards FD4. The increase in CS penetration enhancement properties in presence of ascorbic acid are less evident with respect to that observed when buccal mucosa was used as biological substrate. This is probably due to the fact that two different interaction mechanisms are involved: an interaction with extracellular lipids for buccal mucosa (poor in TJs) and an interaction with TJs for the intestinal one. The apparent coefficient values calculated for LY (hydrophilic low MW molecule) and TEER values indicate that also CS lactate is able to disturb TJ integrity; it causes a TJ opening, sufficient only to facilitate the permeation of LY but not of FD4 (high MW molecule).



Cytotoxicity tests proved CS ascorbate biocompatibility: such salt shows a cell biocompatibility comparable to those of CS hydrochloride and lactate, well-known in literature as salts characterized by low toxicity. The overall results indicate that the salification of hMW chitosan with ascorbic acid produces an increase in penetration enhancement properties of the polymer towards both buccal mucosa and Caco-2 cell monolayer. Such an increase is higher when buccal mucosa is used as biological substrate. A hypopthesis to explain such behaviour could be that ascorbic acid increases chitosan capability to interact with extracellular lipids (main barrier to drug transport across buccal mucosa). Such a hypothesis is supported by relevant literature: as mentioned in the introduction, some authors have suggested that chitosan is able to interact with buccal lipids and it has been also proved that the polymer salification with ascorbic acid produces a higher chitosan interaction with diet lipids. Work is in progress to clarify the mechanisms by which the synergic effect between chitosan and ascorbic acid acts.

10.

ACKNOWLEDGMENTS

11.

This study was supported by funding from the Italian Ministry of University and Scientific Research (MIUR) (PRIN/COFIN 2005).

12.

5.

6.

7.

8.

9.

13.

REFERENCES 1. Muzzarelli RAA. Chitosan-based dietary foods. Carbohydr. Polym. 1996; 29:309–316. 2. Tsujikawa T, Kanauchi O, Andoh A, Saotome T, Sasaki M, Fujiyama Y, Bamba T. Supplement of a chitosan and ascorbic acid mixture for Crohn’s disease: a pilot study. Appl. Nutrit. Investig. 2003;19:137–139. 3. Kanauchi O, Deuchi K, Imasato Y, Kobayashi E. Increasing effect of a chitosan and ascorbic acid mixture on fecal dietary fat excretion. Biosci. Biotechnol. Biochem. 1994;58: 1617–1620. 4. Kanauchi O, Deuchi K, Imasato Y, Shizukuishi M, Kobayashi E. Mechanism for the inhibition of fat digestion

14.

15.

16.

521

by chitosan and for the synergistic effect of ascorbate. Biosci. Biotechnol. Biochem. 1995;595:786–790. Senel S, Kremer MJ, Kas S, Wertz PW, Hincal AA, Squier CA. Enhancing effect of chitosan on peptide drug delivery across buccal mucosa. Biomaterials. 2000;21:2067–2071. Rossi S, Sandri G, Ferrari F, Bonferoni MC, Caramella C. Buccal delivery of acyclovir from films based on chitosan and polyacrylic acid. Pharmaceut. Develop. Technol. 2003;8(2): 199–208. Sandri G, Rossi S, Ferrari F, Bonferoni MC, Muzzarelli C, Caramella C. Assessment of chitosan derivatives as buccal and vaginal penetration enhancers. Eur. J. Pharmaceut. Sci. 2004;21:35–39. Artursson P, Lindmark T, Davis SS, Illum L. Effect of chitosan on the permeability of monolayers of intestinal epithelial cells (Caco-2). Pharmaceut. Res. 1994;11:1358–1361. Schipper NGM, Varurn KM, Artursson P. Chitosans as absorption enhancers for poorly absorbable drugs: influence of molecular weight and degree of acetylation on drug transport across human intestinal epithelial (Caco-2) cells. Pharmaceut. Res. 1996;13(11):1686–1692. Thanou M, Verhoef C, Junginger RE. Oral drugs absorption enhancement by chitosan and its derivatives. Adv. Drug Deliv. Rev. 2001;25:117–126. Smith J, Wood E, Dornish M. Effect of chitosan on epithelial cell tight junctions. Pharmaceut. Res. 2004;21:43–49. Jia Z, Shen D. Effect of reaction temperature and reaction time on the preparation of low-molecular-weight chitosan using phosphoric acid. Carbohydr. Polym. 2002;49: 393–396. Sandri G, Poggi P, Bonferoni M.C, Rossi S, Ferrari F, Caramella C. Histological evaluation of buccal penetration enhancement properties of chitosan and trimethyl chitosan. J. Pharm. Pharmacol. 2006;58(10):1327–1336. Alvarez-Roman R, Naik A, Kalia YN, Fessi H, Guy RH. Visualization of skin penetration using confocal laser scanning microscopy. Eur. J. Pharmaceut Biopharmaceut. 2004;58:301–316. Popiolkiewicz J, Polkowski K, Skierski JS, Mazurek AP. In vitro toxicity evaluation in the development of new anticancer drugs – genistein glycosides. Cancer Lett. 2005;229: 67–75. Fotakis G, Timbrell JA. In vitro cytotoxicity assays: Comparison of LDH, neutral red, MTT and protein assay in hepatoma cell lines following exposure to cadmium chloride. Toxicol. Lett. 2006;160:171–177.
