- Email: [email protected]
Comparative Biomembrane Permeation of Tacrine Using Yucatan Minipigs and Domestic Pigs as the Animal Model
Comparative Biomembrane Permeation of Tacrine Using Yucatan Minipigs and Domestic Pigs as the Animal Model ANURADHA V. GORE, ALFRED C. LIANG,
AND
YIE W. CHIEN*
Contribution from Controlled Drug-Delivery Research Center, Rutgers College of Pharmacy, 41-D Gordon Road, Piscataway, New Jersey 08854-8067 Received September 10, 1997. Final revised manuscript received January 20, 1998. Accepted for publication January 23, 1998. Abstract 0 Tacrine (THA), a centrally acting acetylcholine-esterase inhibitor, is presently administered perorally for the treatment of Alzheimer’s disease (AD). However, its low bioavailablity (i.e., 17%) and short half-life (2−4 h) demand the search for alternative routes of administration. The primary objective of this study was to assess the potential of absorptive mucosae and skin as routes for improving the systemic delivery of THA. The Yucatan minipig, which has been used increasingly in biomedical research as a useful model for humans, and the domestic pig, which is available at low cost, were evaluated for their suitability as animal model. Permeation kinetics of THA across various absorptive mucosae (nasal, buccal, sublingual, and rectal) of both species of swine were studied in the hydrodynamically wellcalibrated Valia−Chien permeation cells. For comparison, permeation through various intestinal segments (duodenum, jejunum, and ileum) was also measured. Results indicated that both species display similar permeation characteristics. However, the data obtained for the domestic pigs shows lower intra- and inter-animal variabilities than that of the Yucatan minipigs. The nasal mucosa was found to have the highest permeability, while the buccal mucosa had the lowest among the absorptive mucosae. The intrinsic permeabilities and diffusivity of THA across the four absorptive mucosae were not significantly different between species but lower than that for the intestinal segments for both species. Using dorsal skin as the model, the skin permeation of THA was also investigated and the results indicated that the domestic swine has a significantly higher skin permeability than the Yucatan minipig, with more than a 2-fold difference in intrinsic permeabilities. The intrinsic permeability, partition coefficient, and diffusivity for domestic pig skin are very similar to that for human cadaver skin. Considering the potential of bypassing the hepatic “first-pass” elimination, the absorptive mucosae may be useful routes for systemic delivery of THA to achieve improved bioavailability. With additional advantages of lower variability, ease of membrane excision, good accessibility, and lower cost, it is concluded that the domestic swine is a better animal model than the Yucatan minipig for preclinical studies on the systemic delivery of tacrine.
1. Introduction Alzheimer’s disease (AD) is a disorder associated with progressive decline in memory and cognitive function. More than 5% of the population above 65 years of age suffer from some form of dementia, and AD accounts for about half of these cases. In the United States alone, about 1.31.8 million people are affected.1 Tacrine (9-amino-1,2,3,4tetrahydroacridine, THA), a centrally acting cholinesterase inhibitor with additional pharmacological activity on monoamine levels and ion channels, was the first drug to be approved for use as a palliative treatment in patients with AD.2 AD is associated with cholinergic cell loss in © 1998, American Chemical Society and American Pharmaceutical Association
the brain, particularly those neurons communicating to the frontal and temporal cortical areas that are strongly associated with memory and cognition. Thus, an increase in acetylcholine availability is thought to enhance muscarinic effects and improve memory and cognition.3 In clinical trials, THA (80-160 mg/day) has been shown to be effective in slowing down the progression of the disease in 30-51% of patients with maximum improvements at high doses.4,5 Presently THA is marketed as tablets to be administered perorally. On peroral administration, THA has a low systemic bioavailability (17-24%), due to extensive hepatic “firstpass” metabolism, and a short half-life (2-4 h), thus requiring frequent administration and large doses. In the elderly patient population, the frequency of dosing can lead to compliance problems, and large doses can lead to hepatotoxicity. The major adverse effects of THA include changes in liver function which are found to be dosedependent.6 Autopsy data available from seven patients who were being treated with THA has shown that the liver toxicity is caused by direct damage to the liver cells.3 This hepatotoxicity severely restricts the maximum dose that can be administered orally to many patients. Thus, oral administration may not be the optimal route for delivery of THA. Other routes that can improve the systemic bioavailability by by-passing the hepatic “first-pass” metabolism need to be considered.7 Administration by alternate routes will potentially reduce the gastrointestinal side effects and hepatotoxicity since, in this case, the liver is exposed to only a fraction of the administered dose. Therefore, higher doses by these routes may be better tolerated. The primary objective of this study was to evaluate various biomembranes as potential alternate sites for the systemic delivery of THA using intestinal segments as the control. Various animal models have been studied to investigate mucosae for in vitro drug delivery studies, the most common among them being the rabbit.8 The Yucatan minipig has been used as a valuable model in biomedical research because of its similarities to humans in both structure and function.9,10 A second type of animal model that can be used in research is the domestic pig.11,12 In the present studies, Yucatan minipigs and domestic pigs were compared as animal models for a source of biomembranes for in vitro permeation studies.
2. Materials and Methods 2.1. MaterialssTacrine hydrochloride salt and triethylamine were obtained from Sigma Chemical Co. Tacrine hydrochloride was converted to tacrine base (THA) prior to use. Poly(ethylene glycol) 400 (PEG-400), disodium hydrogen phosphate, sodium dihydrogen phosphate, sodium hydroxide, phosphoric acid (reagent grade), and acetonitrile (HPLC grade) were used as obtained (Fisher Scientific).
S0022-3549(97)00359-6 CCC: $15.00 Published on Web 03/05/1998
Journal of Pharmaceutical Sciences / 441 Vol. 87, No. 4, April 1998
2.2. Sample PreparationsThe surgical method used for excision of biomembranes was similar for both species of pigs. Male Yucatan minipigs (Buckshire Corp., Perkasie, PA), weighing about 150-180 lb., were sacrificed using an overdose of sodium pentobarbital administered through the ear vein. Male domestic pigs (mixed strain), weighing about 35-45 lb., were obtained from a local abattoir, immediately after sacrifice (prior to any further processing). To obtain the biomembranes for each pig the following procedure was followed: the absorptive mucosae, viz. the nasal, buccal, sublingual, and rectal mucosae, were immediately removed surgically, and the underlying fat and connective tissue were cleaned off. The mucosae obtained were then rinsed briefly with normal saline and cut into pieces. The intestinal segments were obtained by making an incision just below the rib-cage on the abdomen of the pig. The three segments of the small intestine (i.e. duodenum, jejunum, and ileum) were identified and pieces (∼6 cm in length each) were excised from each segment. These were then cleaned of any connective tissue and cut open longitudinally. The contents were washed out with normal saline and the segments were cut into smaller pieces for use in permeation studies. The skin on the dorsal side was cleaned of any hair, using a hair clipper, and sections of skin were removed, using a dermatome set at 500 µm. Permeation studies were performed in triplicate using freshly excised biomembranes. Frozen human cadaver skin (Ohio Valley Tissue and Skin Center, Cincinnati, OH) was thawed in normal saline maintained at 37 °C, blotted dry and used in the permeation studies. 2.3. Solubility StudiessSolubility of THA in both isotonic phosphate buffer (IPB, pH 7.4) and in a cosolvent system consisting of PEG-400 and IPB (20:80) was determined. Briefly, 100 mg of the base was added to 1.5 mL of the solvent in glass vials. The vials were tightly sealed and placed in a shaking water bath maintained at 37 °C for 48 h to allow for equilibrium. The supernatant solution was then removed and filtered through a 0.22 µm nylon filter using a prewarmed syringe filter assembly. The filtrate was analyzed by HPLC after suitable dilution. 2.4. In Vitro Permeation StudiessIn vitro permeation studies were performed using the hydrodynamically well-calibrated Valia-Chien permeation cells.13 The permeation cell consists of a pair of half-cells that are mirror images. Each halfcell has a volume of 3.5 mL and the area available for permeation is 0.64 cm2. The thickness of each biomembrane sample was measured with a micrometer, by sandwiching the membrane between two glass slides. The respective biomembrane sample was then mounted between the half cells such that the mucosal side (in the case of absorptive mucosae or intestinal segments) or the stratum corneum (in the case of skin) was facing the donor half cell. The receptor solution consisted of the 20% v/v PEG-400 in IPB, while the donor solution consisted of a saturated suspension of THA in the same cosolvent system. Samples were withdrawn at predetermined intervals from each receptor halfcell for all the mucosal membranes for 12 h, and for skin for up to 48 h. The volume in all the receptor cells was immediately replaced with fresh PEG-buffer solution maintained at 37 °C. The samples were stored at -20 °C prior to analysis by HPLC. 2.5. Membrane/Buffer Partitioning StudiessMembrane/ buffer partitioning studies were performed by placing pieces of the excised and cleaned membranes (∼50 mg) in glass vials and adding donor solution containing a known concentration of THA to each vial. The vials were then placed in a shaking water bath maintained at 37 °C for 12 h. To ensure that equilibrium was reached, samples (100 µL) were removed from the supernatant solution at different time intervals and analyzed by HPLC. The amount of THA partitioning into the membrane at equilibrium was calculated by mass balance, to obtain the concentration of drug in the membrane. The partition coefficient was then determined from the ratio of the concentration of the drug in the membrane to that of the supernatant solution at equilibrium (eq 5). 2.6. Analytical MethodsAll samples were analyzed using an HPLC system (Hewlett-Packard model 1050) equipped with an autosampler. The column used was a reversed phase Hypersil C-18 column (20 cm length and 4.6 mm i.d., 5 µm particle size) maintained at 37 °C. The mobile phase consisted of 30:70 acetonitrile:water containing 0.5% of triethylamine with the pH adjusted to 5 using phosphoric acid at a flow rate of 1 mL/min. The sample volume injected for analysis was 20 µL for skin samples and 5 µL for all other biomembranes. The detector used was a variable wavelength UV detector maintained at a wave-
442 / Journal of Pharmaceutical Sciences Vol. 87, No. 4, April 1998
length of 325 nm. The limit of detection was 0.1 µg/mL and standard curves were prepared over a range of 0.5-500 µg/mL. 2.7. Data AnalysissFor each membrane, the cumulative amount of permeated drug was plotted against time and the transmembrane permeation rate was calculated from the linear portion of the plot. The lag-time for permeation was determined from the time (x) axis intercept of the linear portion of the curve. From the permeation rate values, the permeability, intrinsic permeability, and diffusivity were calculated using the solubility of the drug in the donor solution, membrane/buffer partition coefficient, and thickness of the membrane as described in the next section (2.8). Statistical analysis was performed using statistical software Sigma-Stat (Jandel Corp., San Rafel, CA). The tests used were one way analysis of variance (one way ANOVA), pairwise multiple comparisons of the Student-Newman-Keuls method and the Student’s t-test. Individual absorptive mucosae were compared with the control (intestinal segments) using the Bonferroni t-test. Differences were considered significant at the p e 0.05 level. 2.8. Theoretical ConsiderationssThe rate of permeation of a drug across a membrane is given by the eq 1
Jm ) Pm(Cd - Cr)
(1)
where Jm ) rate of permeation (in µg/cm2/h), Cd ) concentration in donor solution, Pm ) permeability in cm/h, and Cr ) concentration in receptor solution. If sink conditions are maintained throughout the permeation study, i.e., Cd . Cr, permeability can be calculated using the saturation solubility of THA as the concentration in donor solution (Cd).14
Pm )
Jm Cd
(2)
To correct for the variation in the thickness (hm) of different membranes, intrinsic permeability (Pi) is determined by multiplying the apparent permeability (Pm) by membrane thickness as follows.
Pi ) Pmhm
(3)
Diffusivity (Dm) of the drug in the membrane is given by
Dm )
Pi K
(4)
where Dm ) diffusivity and K ) (membrane/buffer) partition coefficient. The (membrane/buffer) partition coefficient can be calculated from the ratio of drug concentration in the membrane to that in the buffer solution at equilibrium.
K)
Am/Wm Cs
(5)
Am ) amount of THA in membrane, Wm ) weight of membrane, and Cs ) concentration in solution (w/w).
3. Results and Discussion 3.1. Solubility StudiessSolubility of THA base in IPB alone was found to be 1.5 mg/mL, while that in the cosolvent system was found to be 8.9 mg/mL. At pH 7.4, THA is expected to be present over 99% in its protonated form. Addition of a nonpolar solvent like PEG increases the solubility of the unprotonated form,15 which is expected to have higher permeation through the lipophilic biomembranes. The donor solution for the permeation studies was a suspension of THA base in the cosolvent system selected to maintain a high constant concentration of THA base during the course of the permeation studies. The same cosolvent system was chosen as the receptor medium to enable maintenance of sink conditions. 3.2. Permeation through Absorptive Mucosae and Intestinal SegmentssZero-order permeation kinetics
Figure 1sComparative zeroth-order permeation profiles of THA across the vaarious absorptive mucosae between Yucatan minipigs and domestic pigs.
Figure 2sComparative zeroth-order permeation profiles of THA across the various intestinal segments between Yucatan minipigs and domestic pigs.
were consistently observed for all the membranes studied. The permeation profiles of THA through the mucosae compared in Figure 1 suggest that the overall trends in permeation are similar between these species of pigs with some variations. In Yucatan minipigs, permeation profiles of THA through the nasal and rectal mucosae were fairly similar and significantly higher than that across the buccal and sublingual mucosae. In the domestic pigs, on the other
hand, the permeation profiles through the rectal and sublingual mucosae were similar and higher than that across the buccal mucosa but lower than the nasal mucosa. The data in Figure 2 indicate that while in the Yucatan minipig model, lower permeation of THA is attained for the duodenal segment than for the jejunal or ileal segments, these differences among the three segments in the domestic pig model are not statistically significant. Journal of Pharmaceutical Sciences / 443 Vol. 87, No. 4, April 1998
Table 1sComparison in Permeation Parameters of THA across the Absorptive Mucosae of Yucatan Minipigs and Domestic Pigs absorptive mucosae nasal sublingual buccal rectal
pig speciesa
permeation rate [(µg/cm2)/h ± SE]b
thickness (cm ± SE)b
partition coefficient
lag time (h ± SE]b
Y D Y D Y D Y D
640.28 (37.59)‡ 705.51(51.32)‡ 365.82(32.44) 470.07(27.63) 336.29(25.03)† 224.93(14.49)† 600.04(55.45)‡ 498.84(21.14)
0.064(0.005) 0.053(0.003) 0.074(0.006) 0.045(0.003) 0.131(0.008)‡ 0.128(0.003)‡ 0.052(0.002) 0.061(0.003)
2.338(0.216) 2.197(0.191) 2.848(0.145) 3.029(0.183) 5.533(0.500)‡ 4.095(0.446)‡ 2.478(0.140) 2.174(0.081)
NSc NS 0.803(0.117)‡ NS 1.065(0.110)†,‡ 1.837(0.148)†,‡ NS NS
a Y ) Yucatan minipigs (n ) 6), D ) domestic pigs (n ) 4). b mean(± standard error of mean). c NS ) Not significant. ‡Significantly higher than other mucosae of same pig species (p e 0.05). †Difference between corresponding mucosae of Yucatan and domestic pigs is statistically significant (p e 0.05).
Table 2sComparison in Permeation Parameters of THA across the Intestinal Segments of Yucatan Minipigs and Domestic Pigs intestinal segments
pig speciesa
permeation rate [(µg/cm2)/h ± SE]b
thickness (cm ± SE])b
partition coefficientb
lag time (h ± SE]b
duodenum
Y D Y D Y D
373.96(33.13) 420.44(39.14) 530.88(76.70) 424.12(22.92) 547.47(59.87) 442.15(55.27)
0.110(0.006) 0.090(0.002) 0.083(0.004) 0.091(0.004) 0.101(0.008) 0.127(0.011)
2.533(0.095) 2.341(0.184) 2.308(0.121) 2.084(0.095) 2.665(0.107) 2.780(0.095)
0.861(0.155) 0.919(0.107) NSc 0.578(0.137) 0.577(0.187) 0.815(0.224)
jejunum ileum a
Y ) Yucatan minipigs (n) 6), D ) domestic pigs (n ) 4). b mean(±standard error of mean). c NS ) not significant.
The rate of permeation of drugs across membranes is a function of drug permeability, which in turn is directly proportional to the partition coefficient of the drug and the diffusivity of the drug across the membrane and inversely proportional to the thickness. The experimentally determined membrane thicknesses and (membrane/buffer) partition coefficients of THA for the mucosae and intestinal segments are reported in Tables 1 and 2, respectively. The differences in the rate of permeation of THA across various membranes studied is reflected in the differences in permeability of THA, as seen in Figure 3A. Since the higher thickness of the buccal mucosa may account for the low permeability of THA across this membrane (Table 1), the intrinsic permeability of THA through the various biomembranes was compared (Figure 3B). It was observed in the case of Yucatan minipigs that the intrinsic permeability of THA through buccal mucosa was comparable to that of nasal mucosae and was significantly higher than those of both the rectal and sublingual mucosae. Thus the lower permeability of THA through the buccal mucosa indeed appears to result from its higher thickness. In the case of domestic pigs, the intrinsic permeability of THA was highest through the nasal mucosa, followed by the buccal and the rectal, with the lowest permeability for the sublingual mucosa (p < 0.05). Another factor that must be considered when studying the permeation of drugs through mucosae is the partitioning of molecules into the membrane. Diffusion coefficients were obtained to account for the differences in partition coefficient of THA in the membranes. Their rank order was observed to be nasal > rectal > buccal ∼ sublingual for both the species of pigs, as seen in Figure 3C. The buccal and sublingual mucosae also displayed the presence of a significant lag-time for the permeation of THA. This slow onset of permeation of THA across the buccal mucosa, despite the high partition coefficient, may indicate some form of binding within the mucosa which hinders permeation.16 Sublingual delivery has often been considered as an alternative route to bypass gastrointestinal absorption and metabolism. However, in the present study it was seen that the intrinsic permeability of THA across this mucosa was the lowest, which was not accounted for by either thickness or partition coefficient. It thus appears that sublingual mucosa provides a greater barrier to permeation 444 / Journal of Pharmaceutical Sciences Vol. 87, No. 4, April 1998
Figure 3sComparison of the (A) permeability, (B) intrinsic permeability, and (C) diffusivity of THA in the absorptive mucosae and intestinal segments of both species of pigs. †Differences between corresponding biomembranes of Yucatan minipigs and domestic pigs are statistically significant (p e 0.05) ‡Significantly higher than other mucosae of the same species of pig (p e 0.05).
of THA than the other mucosae and may not be suitable for chronic delivery of THA. In the case of intestinal segments, after accounting for differences in thickness and partition coefficient, no significant differences were seen in any of the permeation
Table 3sComparison of Permeation Parameters for Permeation of THA through the Dorsal Skin of Yucatan Minipigs, Domestic Pigs, and Human Cadaver Skin parametersa
Yucatan pig skin (n ) 6)
domestic pig skin (n ) 4)
human cadaver skin (n ) 5)
permeation rate [(µg/cm2)/h ± SE] permeability [(cm/h ± SE) × 103] intrinsic permeability [(cm2/h ± SE) × 103] lag time (h ± SE) thickness (cm ± SE) partition coefficient diffusivity [(cm2/h ± SE) × 103
3.536(0.44)†,‡ 0.398(0.049)†,‡ 0.030(0.004)†,‡ 19.949(0.855)†,‡ 0.063(0.003)‡ 3.686(0.002)†,‡ 0.017(0.005)†,‡
9.537(0.89)†,‡ 0.993(0.158)†,‡ 0.059(0.006)† 13.637(1.401)† 0.063(0.003)‡ 2.590(0.079)† 0.026(0.002)†
18.044(1.40) 2.031(0.158) 0.071(0.006) 6.520(1.051) 0.035(0.002) 2.519(0.317) 0.028(0.002)
a Mean (±standard error of mean). †Difference in the corresponding permeation parameters between Yucatan and domestic skin is statistically significant (p e 0.05). ‡Differences in the corresponding permeation parameters between Yucatan (or domestic) pig skin and human cadaver skin are statistically significant (p e 0.05).
parameters obtained for the Yucatan minipig model (Table 2). Similar trends were observed for the domestic pig model. The only significant difference observed in the latter was the higher intrinsic permeability values for the ileal segment when compared to the duodenal and jejunal segments. However, none of the other permeation parameters were found to differ significantly. Thus, for further comparisons with mucosal membranes, these segments could be considered together as the set of intestinal segments for the respective species of pigs. Initially, the set of absorptive mucosae was compared to the intestinal segments. The intrinsic permeability and diffusivity of THA in intestinal segments was significantly higher than the mucosal membranes. This can be explained by taking the “real” surface area available for permeation into account. The absorption of nutrients is the major function of the intestine. To facilitate absorption, the mucosal surface of the intestine is covered with numerous folds, villi, and microvilli, which increase the surface area available for absorption several 100-fold over the apparent surface area.17 The apparent surface area available for permeation using the Valia-Chien permeation cells is 0.64 cm2. Thus the large “real” surface area could account for the higher intrinsic permeability and diffusivity of THA through intestinal segments. Further, the intestinal segments were considered as the control and the individual mucosae were compared to it. The apparent permeability of THA through the nasal mucosa was found to be significantly higher than the intestinal segments for both the species of pigs, while the intrinsic permeability through nasal and buccal mucosae was comparable to the intestinal segments. Thus, despite the differences in “real” surface area, these mucosae have comparable absorption of THA and may be considered as alternate routes for the delivery of THA. To evaluate the domestic pig as a model to study biomembrane permeation, the permeation characteristics of the domestic pig biomembranes were compared to the corresponding biomembranes of Yucatan minipigs. As mentioned earlier, it was found that the trends of permeation characteristics of the biomembranes were similar in both species of pigs. Similarly, the order of intrinsic permeability of THA across absorptive mucosae and intestinal segments was also similar:
[ileum, jejunum, duodenum] > [(nasal, buccal) > (rectal, sublingual)] > skin Comparing the corresponding biomembranes of the two species of pigs, it was seen that their permeation characteristics were similar to each other, with two exceptions, viz. the buccal mucosa and the skin. The buccal mucosa of the domestic pig had a significantly lower apparent permeability and intrinsic permeability for THA (Figure
3A,B), although differences in thickness and partition coefficient of THA were not significant. After accounting for the thickness and partition coefficient, the diffusivity of THA in both was comparable (Figure 3C). For both pig species, the permeation of THA through the nasal mucosa was the highest. This observation has been reported previously for other drugs.18-20 Permeation across biological membranes can occur mainly via two main pathways: the paracellular route (aqueous pore) or the transcellular route (lipoidal).21 Hydrophilic and charged molecules that do not partition well into the cell membrane are thus transported via the intercellular space while lipophilic permeants tend to be transported via the transcellular space.21 THA, a basic drug with a pKa of 10, exists as a positively charged molecule at pH 7.4 and is thus expected to be transported via the paracellular route. The higher permeability through the nasal mucosa could be a result of its slightly more hydrophilic nature than other mucosae.22,23 Under in vivo conditions, the highly vascular nasal mucosa can be expected to provide a rich surface for rapid drug absorption. However, this mucosa suffers from the disadvantage of very short contact time due to mucociliary clearance.24 As a result, nasal delivery of drugs is useful in those situations where rapid drug effect is required for short duration25 and thus may not be suitable for chronic therapy. The buccal route has been found to be more suitable for chronic drug delivery.26-28 Similar drug delivery devices may be envisioned for the prolonged systemic delivery of THA. The potential of the rectal route in improving the bioavailability of THA was demonstrated in a clinical trial by Ahlin and co-workers.29 Thus, the mucosae most likely to be useful for development of controlled drug delivery devices appear to be the buccal and the rectal mucosae. 3.3. Permeation through the SkinsAnother biomembrane which is suitable for chronic drug delivery is the skin. In the present study, it was seen that the permeation profiles of THA through the excised dorsal skin were the lowest among all the biomembranes for both the Yucatan minipigs and domestic pigs (Figure 4) and also displayed a long lag-time. This difference cannot be accounted for by simply considering thickness or skin/buffer partition coefficient. The low permeability of drugs through the skin compared to other biomembranes may be expected since the skin is covered by the stratum corneum, which is composed of several layers of dead keratinized cells and provides an efficient barrier to the permeation of molecules. However, it may be possible to significantly increase drug permeation through the skin by using chemical enhancers30 or physical methods.31,32 Significant enhancement of permeation of THA and reduction in the lag-time for permeation through the skin using solvent mixtures has been previously demonstrated both in vitro and in vivo in rats.33 Journal of Pharmaceutical Sciences / 445 Vol. 87, No. 4, April 1998
permeation characteristics of THA. In addition, the domestic pigs are more easily available and more economical than the Yucatan minipigs, and may thus serve as a suitable alternative to the Yucatan minipig model to study biomembrane permeation.
References and Notes
Figure 4sPermeation profiles of THA through the dorsal skin excised freshly from Yucatan minipigs and domestic pigs in comparison with human cadaver skin.
Thus, the evaluation of skin as a possible route for THA delivery merits further investigation. Comparing the THA permeation profiles through the dorsal skin of Yucatan minipigs and domestic pigs, all permeation parameters obtained were significantly different, although the thickness of the membranes obtained was similar (Table 3). The permeation of THA through the domestic pig skin was found to be almost 2-fold higher than that across the Yucatan minipig skin (Figure 4). To further evaluate the pig skin as a model to study THA permeation, it was compared to human cadaver skin which has been commonly considered as a model to study permeation of drugs in vitro. However, the human cadaver skin obtained was lower in thickness than both the pig species. The permeation profile of THA through the human cadaver skin was significantly higher than both the Yucatan and domestic pig skin (Figure 4). However, it was observed that intrinsic permeability and diffusivity of THA across human cadaver skin was comparable to that of domestic pig dorsal skin (Table 3). Thus, these data seems to suggest that the permeation characteristics of domestic pig skin are closer to human cadaver skin than the Yucatan pig skin and may thus be a better model to study transdermal permeation of drugs in vitro.
4. Conclusions The results of the present study indicate that despite the differences in the “real” surface area available for permeation, some of the absorptive mucosae show comparable THA permeability to the intestinal segments. The nasal route, though highly permeable, may not be suitable for chronic drug therapy, while the traditionally used sublingual route suffers from the disadvantage of lower permeability. Thus the mucosae that may be suitable for the development of controlled drug delivery devices for the chronic delivery of THA may be the buccal and the rectal mucosae. Controlled delivery of THA using these routes can be expected to increase bioavailability by avoiding hepatic “first-pass” metabolism and also reduce frequency of dosing. Another potential route for delivery of THA may be the transdermal route, although permeation enhancers would be required to achieve the required drug delivery rate. Last, it was observed that the biomembranes of the two species of swine were comparable with respect to 446 / Journal of Pharmaceutical Sciences Vol. 87, No. 4, April 1998
1. Briggs, R. S. J. Alzheimer’s disease: The clinical context. In Current Problems in Neurology; Davies, D. C., Ed.; John Libbey and Co. Ltd.: Paris, France, 1987; Vol. 11, pp 1-8. 2. Wagstaff, A. J.; McTavish, D. Drugs Aging 1994, 4 (6), 510540. 3. Crimson, M. L. Tacrine: The first drug approved for Alzheimer’s disease. Ann. Pharmacother. 1994, 28, 744-51. 4. Knapp, M. J.; Knopman, D. S.; Solomon, P. R.; Pendlebury, W.; Davies, C. S. J. A 30 week randomized controlled trial of high dose tacrine in patients with Alzheimer’s disease. Am. Med. Assoc. 1993, 271, 985-991. 5. Kauffer, D. I.; Cummings, J. L.; Christine, D. Effect of tacrine on behavioral symptoms in Alzheimer’s disease: An open label study. J Geriatric Psych. Neurol. 1996, 9 (1), 1-6. 6. Farlow, M.; Gracon, S. I.; Hershey, L. A.; Lewis, K. W.; Sadowsky, C. H.; Dolan-Urano, J. A controlled trial of tacrine in Alzheimer’s disease. J. Am. Med. Assoc. 1992, 268, 25232529. 7. Chien, Y. W. Recent advances in noninvasive systemic delivery of pharmaceuticals and biopharmaceuticals. Drug Dev. Ind. Pharm. 1994, 20 (4), 417-468. 8. Chien, Y. W. Biopharmaceutics basis for transmucosal delivery. S. T. P. Pharma Sci. 1995, 5 (4), 257-275. 9. Kurali, T. T. Comparison of the gastrointestinal anatomy, physiology and biochemistry of humans and commonly used laboratory animals. Biopharm. Drug Dispos. 1995, 16, 351380. 10. Swindle, M. M. Porcine models in surgical research. In Swine in Biomedical Research; Tumbleson, M. E., Ed.; Plenum Press: New York, 1986; Vol. 1, pp 235-239. 11. Qiao, G. L.; Williams, P. L.; Riviere, J. E. Percutaneous absorption, biotransformation, and systemic disposition of parathion in vivo in swine. I. Comprehensive pharmacokinetic model. Drug. Metab. Dispos. 1994, 22, 459-471. 12. Reifernrath, W. G.; Hawkins, G. S.; Kurtz, M. S. Percutaneous penetration and skin retention of topically applied compounds: An in vitro-in vivo study. J. Pharm. Sci. 1991, 80, 526-532. 13. Chien, Y. W.; Valia, K. Development of a dynamic skin permeation system for long-term permeation studies. Drug Dev. Ind. Pharm. 1984, 10, 575-599. 14. Chien, Y. W. Transdermal drug delivery and delivery systems. In Novel Drug Delivery Systems, 2nd ed.; Marcel Dekker Inc.: New York, 1992; pp 301-380. 15. Martin, A. Solubility and distribution phenomena. In Physical Pharmacy, 4th ed.; Lea and Febiger: Philadelphia, 1993; pp 212-250. 16. Harris, D.; Robinson, J. R. Drug delivery via the mucous membranes of the oral cavity. J. Pharm. Sci. 1992, 81 (1), 1-10. 17. Caspary, W. F. Physiology and pathophysiology of intestinal absorption. Am. J. Clin. Nutr. 1992, 55, 299S-308S. 18. Nair, M.; Chien, Y. W. Effect of steriochemistry of substituent groups on permeation across biological membranes. S. T. P. Pharma Sci. 1993, 3 (6), 425-430. 19. Corbo, D. C.; Liu, J. C.; Chien, Y. W. Drug absorption through mucosal membranes: Effect of mucosal route and penetrant hydrophilicity. Pharm. Res. 1989, 6 (10), 848-852. 20. Rojansakul, Y.; Wang, L. Y.; Bhat, M.; Glover, D. D.; Malanga, C. J.; Ma, J. K. H. The transport barrier of epithelia: A comparative study on membrane permeability and charge selectivity in the rabbit. Pharm. Res. 1992, 9 (8), 1029-1034. 21. Sayani, A. P.; Chien, Y. W. Systemic delivery of proteins and peptides across absorptive mucosae. Crit. Rev. Ther. Drug Carrier Syst. 1996, 13 (1&2), 85-184. 22. Corbo, D. C.; Liu, J.; Chien, Y. W. Characterization of the barrier properties of mucosal membranes. J. Pharm. Sci. 1990, 79 (3), 202-206. 23. Nair, M. K.; Chetty, D. J.; Ho, H.; Chien, Y. W. Biomembrane permeation of nicotine: Mechanistic studies with porcine mucosae and skin. J. Pharm. Sci. 1997, 86 (2), 257-262. 24. Edman, P.; Bjork, E. (D) Routes of delivery: Case studies (1) Nasal delivery of peptide drugs. Adv. Drug Del. Rev. 1992, 8, 165. 25. Ladau, A. J.; Eberhardt, R. T.; Frishman, W. H. Intranasal delivery of cardiovascular agents: An innovative approach
26.
27.
28. 29.
30.
to cardiovascular pharmacotherapy. Am. Heart J. 1994, 127 (6), 1594-1599. Cassidy, J.; Berner, B.; Chan, K.; John, V.; Toon, S.; Holt, B.; Rowland, M. Human transbuccal absorption of diclofenac sodium from a prototype hydrogel device. Pharm. Res. 1993, 10 (1), 126-129. DeGrande, G.; Benes, L.; Horrie`re, F.; Korsenty, H.; Lacoste, C.; McQuinn, R.; Guo, J.; Scherrer, R. Specialized oral mucosal drug delivery systems: patches. In Oral mucosal drug delivery; Rathbone, M. J., Ed.; Marcel Dekker inc.: New York, 1996; pp 285-318. de Vries, M. E.; Bodde´, H. E.; Verhoef, J. C.; Junginger, H. E. Developments in buccal drug delivery. Crit. Rev. Ther. Drug Carrier Syst. 1991, 8 (3), 271-303. Ahlin, A.; Adem, A.; Junte, T.; Ohman, G.; Nyback, H. Pharmacokinetics if tetrahydroaminoacridine: Relations to clinical and biochemical effects in Alzheimer patients. Int. J. Clin. Psychopharmacol. 1992, 7, 29-36. Williams, A. C.; Barry, B. W. Skin absorption enhancers. Crit. Rev. Ther. Drug Carrier Syst. 1992, 9 (3, 4), 305-353.
31. Prausnitz, M. R. Effects of electric current applied to skin: Review for transdermal drug delivery. Adv. Drug Del. Rev. 1996, 18 (1), 395-425. 32. Tyle, P.; Agrawala, P. Drug delivery by phonophoresis. Pharm. Res. 1989, 6 (5), 355-361. 33. Sathyan, G.; Ritschel, W. A.; Hussain, A. S. Transdermal delivery of tacrine: I. Identification of a suitable delivery vehicle. Int. J. Pharm. 1995, 114, 75-83.
Acknowledgments Authors wish to acknowledge Schering-Plough Foundation for the graduate research fellowship which funded this research. We would also like to thank Dr. Senshang Lin and Dr. Li-lan Chen for their help in the surgical techniques for excision of the biomembranes.
JS970359U
Journal of Pharmaceutical Sciences / 447 Vol. 87, No. 4, April 1998
AND
YIE W. CHIEN*
Contribution from Controlled Drug-Delivery Research Center, Rutgers College of Pharmacy, 41-D Gordon Road, Piscataway, New Jersey 08854-8067 Received September 10, 1997. Final revised manuscript received January 20, 1998. Accepted for publication January 23, 1998. Abstract 0 Tacrine (THA), a centrally acting acetylcholine-esterase inhibitor, is presently administered perorally for the treatment of Alzheimer’s disease (AD). However, its low bioavailablity (i.e., 17%) and short half-life (2−4 h) demand the search for alternative routes of administration. The primary objective of this study was to assess the potential of absorptive mucosae and skin as routes for improving the systemic delivery of THA. The Yucatan minipig, which has been used increasingly in biomedical research as a useful model for humans, and the domestic pig, which is available at low cost, were evaluated for their suitability as animal model. Permeation kinetics of THA across various absorptive mucosae (nasal, buccal, sublingual, and rectal) of both species of swine were studied in the hydrodynamically wellcalibrated Valia−Chien permeation cells. For comparison, permeation through various intestinal segments (duodenum, jejunum, and ileum) was also measured. Results indicated that both species display similar permeation characteristics. However, the data obtained for the domestic pigs shows lower intra- and inter-animal variabilities than that of the Yucatan minipigs. The nasal mucosa was found to have the highest permeability, while the buccal mucosa had the lowest among the absorptive mucosae. The intrinsic permeabilities and diffusivity of THA across the four absorptive mucosae were not significantly different between species but lower than that for the intestinal segments for both species. Using dorsal skin as the model, the skin permeation of THA was also investigated and the results indicated that the domestic swine has a significantly higher skin permeability than the Yucatan minipig, with more than a 2-fold difference in intrinsic permeabilities. The intrinsic permeability, partition coefficient, and diffusivity for domestic pig skin are very similar to that for human cadaver skin. Considering the potential of bypassing the hepatic “first-pass” elimination, the absorptive mucosae may be useful routes for systemic delivery of THA to achieve improved bioavailability. With additional advantages of lower variability, ease of membrane excision, good accessibility, and lower cost, it is concluded that the domestic swine is a better animal model than the Yucatan minipig for preclinical studies on the systemic delivery of tacrine.
1. Introduction Alzheimer’s disease (AD) is a disorder associated with progressive decline in memory and cognitive function. More than 5% of the population above 65 years of age suffer from some form of dementia, and AD accounts for about half of these cases. In the United States alone, about 1.31.8 million people are affected.1 Tacrine (9-amino-1,2,3,4tetrahydroacridine, THA), a centrally acting cholinesterase inhibitor with additional pharmacological activity on monoamine levels and ion channels, was the first drug to be approved for use as a palliative treatment in patients with AD.2 AD is associated with cholinergic cell loss in © 1998, American Chemical Society and American Pharmaceutical Association
the brain, particularly those neurons communicating to the frontal and temporal cortical areas that are strongly associated with memory and cognition. Thus, an increase in acetylcholine availability is thought to enhance muscarinic effects and improve memory and cognition.3 In clinical trials, THA (80-160 mg/day) has been shown to be effective in slowing down the progression of the disease in 30-51% of patients with maximum improvements at high doses.4,5 Presently THA is marketed as tablets to be administered perorally. On peroral administration, THA has a low systemic bioavailability (17-24%), due to extensive hepatic “firstpass” metabolism, and a short half-life (2-4 h), thus requiring frequent administration and large doses. In the elderly patient population, the frequency of dosing can lead to compliance problems, and large doses can lead to hepatotoxicity. The major adverse effects of THA include changes in liver function which are found to be dosedependent.6 Autopsy data available from seven patients who were being treated with THA has shown that the liver toxicity is caused by direct damage to the liver cells.3 This hepatotoxicity severely restricts the maximum dose that can be administered orally to many patients. Thus, oral administration may not be the optimal route for delivery of THA. Other routes that can improve the systemic bioavailability by by-passing the hepatic “first-pass” metabolism need to be considered.7 Administration by alternate routes will potentially reduce the gastrointestinal side effects and hepatotoxicity since, in this case, the liver is exposed to only a fraction of the administered dose. Therefore, higher doses by these routes may be better tolerated. The primary objective of this study was to evaluate various biomembranes as potential alternate sites for the systemic delivery of THA using intestinal segments as the control. Various animal models have been studied to investigate mucosae for in vitro drug delivery studies, the most common among them being the rabbit.8 The Yucatan minipig has been used as a valuable model in biomedical research because of its similarities to humans in both structure and function.9,10 A second type of animal model that can be used in research is the domestic pig.11,12 In the present studies, Yucatan minipigs and domestic pigs were compared as animal models for a source of biomembranes for in vitro permeation studies.
2. Materials and Methods 2.1. MaterialssTacrine hydrochloride salt and triethylamine were obtained from Sigma Chemical Co. Tacrine hydrochloride was converted to tacrine base (THA) prior to use. Poly(ethylene glycol) 400 (PEG-400), disodium hydrogen phosphate, sodium dihydrogen phosphate, sodium hydroxide, phosphoric acid (reagent grade), and acetonitrile (HPLC grade) were used as obtained (Fisher Scientific).
S0022-3549(97)00359-6 CCC: $15.00 Published on Web 03/05/1998
Journal of Pharmaceutical Sciences / 441 Vol. 87, No. 4, April 1998
2.2. Sample PreparationsThe surgical method used for excision of biomembranes was similar for both species of pigs. Male Yucatan minipigs (Buckshire Corp., Perkasie, PA), weighing about 150-180 lb., were sacrificed using an overdose of sodium pentobarbital administered through the ear vein. Male domestic pigs (mixed strain), weighing about 35-45 lb., were obtained from a local abattoir, immediately after sacrifice (prior to any further processing). To obtain the biomembranes for each pig the following procedure was followed: the absorptive mucosae, viz. the nasal, buccal, sublingual, and rectal mucosae, were immediately removed surgically, and the underlying fat and connective tissue were cleaned off. The mucosae obtained were then rinsed briefly with normal saline and cut into pieces. The intestinal segments were obtained by making an incision just below the rib-cage on the abdomen of the pig. The three segments of the small intestine (i.e. duodenum, jejunum, and ileum) were identified and pieces (∼6 cm in length each) were excised from each segment. These were then cleaned of any connective tissue and cut open longitudinally. The contents were washed out with normal saline and the segments were cut into smaller pieces for use in permeation studies. The skin on the dorsal side was cleaned of any hair, using a hair clipper, and sections of skin were removed, using a dermatome set at 500 µm. Permeation studies were performed in triplicate using freshly excised biomembranes. Frozen human cadaver skin (Ohio Valley Tissue and Skin Center, Cincinnati, OH) was thawed in normal saline maintained at 37 °C, blotted dry and used in the permeation studies. 2.3. Solubility StudiessSolubility of THA in both isotonic phosphate buffer (IPB, pH 7.4) and in a cosolvent system consisting of PEG-400 and IPB (20:80) was determined. Briefly, 100 mg of the base was added to 1.5 mL of the solvent in glass vials. The vials were tightly sealed and placed in a shaking water bath maintained at 37 °C for 48 h to allow for equilibrium. The supernatant solution was then removed and filtered through a 0.22 µm nylon filter using a prewarmed syringe filter assembly. The filtrate was analyzed by HPLC after suitable dilution. 2.4. In Vitro Permeation StudiessIn vitro permeation studies were performed using the hydrodynamically well-calibrated Valia-Chien permeation cells.13 The permeation cell consists of a pair of half-cells that are mirror images. Each halfcell has a volume of 3.5 mL and the area available for permeation is 0.64 cm2. The thickness of each biomembrane sample was measured with a micrometer, by sandwiching the membrane between two glass slides. The respective biomembrane sample was then mounted between the half cells such that the mucosal side (in the case of absorptive mucosae or intestinal segments) or the stratum corneum (in the case of skin) was facing the donor half cell. The receptor solution consisted of the 20% v/v PEG-400 in IPB, while the donor solution consisted of a saturated suspension of THA in the same cosolvent system. Samples were withdrawn at predetermined intervals from each receptor halfcell for all the mucosal membranes for 12 h, and for skin for up to 48 h. The volume in all the receptor cells was immediately replaced with fresh PEG-buffer solution maintained at 37 °C. The samples were stored at -20 °C prior to analysis by HPLC. 2.5. Membrane/Buffer Partitioning StudiessMembrane/ buffer partitioning studies were performed by placing pieces of the excised and cleaned membranes (∼50 mg) in glass vials and adding donor solution containing a known concentration of THA to each vial. The vials were then placed in a shaking water bath maintained at 37 °C for 12 h. To ensure that equilibrium was reached, samples (100 µL) were removed from the supernatant solution at different time intervals and analyzed by HPLC. The amount of THA partitioning into the membrane at equilibrium was calculated by mass balance, to obtain the concentration of drug in the membrane. The partition coefficient was then determined from the ratio of the concentration of the drug in the membrane to that of the supernatant solution at equilibrium (eq 5). 2.6. Analytical MethodsAll samples were analyzed using an HPLC system (Hewlett-Packard model 1050) equipped with an autosampler. The column used was a reversed phase Hypersil C-18 column (20 cm length and 4.6 mm i.d., 5 µm particle size) maintained at 37 °C. The mobile phase consisted of 30:70 acetonitrile:water containing 0.5% of triethylamine with the pH adjusted to 5 using phosphoric acid at a flow rate of 1 mL/min. The sample volume injected for analysis was 20 µL for skin samples and 5 µL for all other biomembranes. The detector used was a variable wavelength UV detector maintained at a wave-
442 / Journal of Pharmaceutical Sciences Vol. 87, No. 4, April 1998
length of 325 nm. The limit of detection was 0.1 µg/mL and standard curves were prepared over a range of 0.5-500 µg/mL. 2.7. Data AnalysissFor each membrane, the cumulative amount of permeated drug was plotted against time and the transmembrane permeation rate was calculated from the linear portion of the plot. The lag-time for permeation was determined from the time (x) axis intercept of the linear portion of the curve. From the permeation rate values, the permeability, intrinsic permeability, and diffusivity were calculated using the solubility of the drug in the donor solution, membrane/buffer partition coefficient, and thickness of the membrane as described in the next section (2.8). Statistical analysis was performed using statistical software Sigma-Stat (Jandel Corp., San Rafel, CA). The tests used were one way analysis of variance (one way ANOVA), pairwise multiple comparisons of the Student-Newman-Keuls method and the Student’s t-test. Individual absorptive mucosae were compared with the control (intestinal segments) using the Bonferroni t-test. Differences were considered significant at the p e 0.05 level. 2.8. Theoretical ConsiderationssThe rate of permeation of a drug across a membrane is given by the eq 1
Jm ) Pm(Cd - Cr)
(1)
where Jm ) rate of permeation (in µg/cm2/h), Cd ) concentration in donor solution, Pm ) permeability in cm/h, and Cr ) concentration in receptor solution. If sink conditions are maintained throughout the permeation study, i.e., Cd . Cr, permeability can be calculated using the saturation solubility of THA as the concentration in donor solution (Cd).14
Pm )
Jm Cd
(2)
To correct for the variation in the thickness (hm) of different membranes, intrinsic permeability (Pi) is determined by multiplying the apparent permeability (Pm) by membrane thickness as follows.
Pi ) Pmhm
(3)
Diffusivity (Dm) of the drug in the membrane is given by
Dm )
Pi K
(4)
where Dm ) diffusivity and K ) (membrane/buffer) partition coefficient. The (membrane/buffer) partition coefficient can be calculated from the ratio of drug concentration in the membrane to that in the buffer solution at equilibrium.
K)
Am/Wm Cs
(5)
Am ) amount of THA in membrane, Wm ) weight of membrane, and Cs ) concentration in solution (w/w).
3. Results and Discussion 3.1. Solubility StudiessSolubility of THA base in IPB alone was found to be 1.5 mg/mL, while that in the cosolvent system was found to be 8.9 mg/mL. At pH 7.4, THA is expected to be present over 99% in its protonated form. Addition of a nonpolar solvent like PEG increases the solubility of the unprotonated form,15 which is expected to have higher permeation through the lipophilic biomembranes. The donor solution for the permeation studies was a suspension of THA base in the cosolvent system selected to maintain a high constant concentration of THA base during the course of the permeation studies. The same cosolvent system was chosen as the receptor medium to enable maintenance of sink conditions. 3.2. Permeation through Absorptive Mucosae and Intestinal SegmentssZero-order permeation kinetics
Figure 1sComparative zeroth-order permeation profiles of THA across the vaarious absorptive mucosae between Yucatan minipigs and domestic pigs.
Figure 2sComparative zeroth-order permeation profiles of THA across the various intestinal segments between Yucatan minipigs and domestic pigs.
were consistently observed for all the membranes studied. The permeation profiles of THA through the mucosae compared in Figure 1 suggest that the overall trends in permeation are similar between these species of pigs with some variations. In Yucatan minipigs, permeation profiles of THA through the nasal and rectal mucosae were fairly similar and significantly higher than that across the buccal and sublingual mucosae. In the domestic pigs, on the other
hand, the permeation profiles through the rectal and sublingual mucosae were similar and higher than that across the buccal mucosa but lower than the nasal mucosa. The data in Figure 2 indicate that while in the Yucatan minipig model, lower permeation of THA is attained for the duodenal segment than for the jejunal or ileal segments, these differences among the three segments in the domestic pig model are not statistically significant. Journal of Pharmaceutical Sciences / 443 Vol. 87, No. 4, April 1998
Table 1sComparison in Permeation Parameters of THA across the Absorptive Mucosae of Yucatan Minipigs and Domestic Pigs absorptive mucosae nasal sublingual buccal rectal
pig speciesa
permeation rate [(µg/cm2)/h ± SE]b
thickness (cm ± SE)b
partition coefficient
lag time (h ± SE]b
Y D Y D Y D Y D
640.28 (37.59)‡ 705.51(51.32)‡ 365.82(32.44) 470.07(27.63) 336.29(25.03)† 224.93(14.49)† 600.04(55.45)‡ 498.84(21.14)
0.064(0.005) 0.053(0.003) 0.074(0.006) 0.045(0.003) 0.131(0.008)‡ 0.128(0.003)‡ 0.052(0.002) 0.061(0.003)
2.338(0.216) 2.197(0.191) 2.848(0.145) 3.029(0.183) 5.533(0.500)‡ 4.095(0.446)‡ 2.478(0.140) 2.174(0.081)
NSc NS 0.803(0.117)‡ NS 1.065(0.110)†,‡ 1.837(0.148)†,‡ NS NS
a Y ) Yucatan minipigs (n ) 6), D ) domestic pigs (n ) 4). b mean(± standard error of mean). c NS ) Not significant. ‡Significantly higher than other mucosae of same pig species (p e 0.05). †Difference between corresponding mucosae of Yucatan and domestic pigs is statistically significant (p e 0.05).
Table 2sComparison in Permeation Parameters of THA across the Intestinal Segments of Yucatan Minipigs and Domestic Pigs intestinal segments
pig speciesa
permeation rate [(µg/cm2)/h ± SE]b
thickness (cm ± SE])b
partition coefficientb
lag time (h ± SE]b
duodenum
Y D Y D Y D
373.96(33.13) 420.44(39.14) 530.88(76.70) 424.12(22.92) 547.47(59.87) 442.15(55.27)
0.110(0.006) 0.090(0.002) 0.083(0.004) 0.091(0.004) 0.101(0.008) 0.127(0.011)
2.533(0.095) 2.341(0.184) 2.308(0.121) 2.084(0.095) 2.665(0.107) 2.780(0.095)
0.861(0.155) 0.919(0.107) NSc 0.578(0.137) 0.577(0.187) 0.815(0.224)
jejunum ileum a
Y ) Yucatan minipigs (n) 6), D ) domestic pigs (n ) 4). b mean(±standard error of mean). c NS ) not significant.
The rate of permeation of drugs across membranes is a function of drug permeability, which in turn is directly proportional to the partition coefficient of the drug and the diffusivity of the drug across the membrane and inversely proportional to the thickness. The experimentally determined membrane thicknesses and (membrane/buffer) partition coefficients of THA for the mucosae and intestinal segments are reported in Tables 1 and 2, respectively. The differences in the rate of permeation of THA across various membranes studied is reflected in the differences in permeability of THA, as seen in Figure 3A. Since the higher thickness of the buccal mucosa may account for the low permeability of THA across this membrane (Table 1), the intrinsic permeability of THA through the various biomembranes was compared (Figure 3B). It was observed in the case of Yucatan minipigs that the intrinsic permeability of THA through buccal mucosa was comparable to that of nasal mucosae and was significantly higher than those of both the rectal and sublingual mucosae. Thus the lower permeability of THA through the buccal mucosa indeed appears to result from its higher thickness. In the case of domestic pigs, the intrinsic permeability of THA was highest through the nasal mucosa, followed by the buccal and the rectal, with the lowest permeability for the sublingual mucosa (p < 0.05). Another factor that must be considered when studying the permeation of drugs through mucosae is the partitioning of molecules into the membrane. Diffusion coefficients were obtained to account for the differences in partition coefficient of THA in the membranes. Their rank order was observed to be nasal > rectal > buccal ∼ sublingual for both the species of pigs, as seen in Figure 3C. The buccal and sublingual mucosae also displayed the presence of a significant lag-time for the permeation of THA. This slow onset of permeation of THA across the buccal mucosa, despite the high partition coefficient, may indicate some form of binding within the mucosa which hinders permeation.16 Sublingual delivery has often been considered as an alternative route to bypass gastrointestinal absorption and metabolism. However, in the present study it was seen that the intrinsic permeability of THA across this mucosa was the lowest, which was not accounted for by either thickness or partition coefficient. It thus appears that sublingual mucosa provides a greater barrier to permeation 444 / Journal of Pharmaceutical Sciences Vol. 87, No. 4, April 1998
Figure 3sComparison of the (A) permeability, (B) intrinsic permeability, and (C) diffusivity of THA in the absorptive mucosae and intestinal segments of both species of pigs. †Differences between corresponding biomembranes of Yucatan minipigs and domestic pigs are statistically significant (p e 0.05) ‡Significantly higher than other mucosae of the same species of pig (p e 0.05).
of THA than the other mucosae and may not be suitable for chronic delivery of THA. In the case of intestinal segments, after accounting for differences in thickness and partition coefficient, no significant differences were seen in any of the permeation
Table 3sComparison of Permeation Parameters for Permeation of THA through the Dorsal Skin of Yucatan Minipigs, Domestic Pigs, and Human Cadaver Skin parametersa
Yucatan pig skin (n ) 6)
domestic pig skin (n ) 4)
human cadaver skin (n ) 5)
permeation rate [(µg/cm2)/h ± SE] permeability [(cm/h ± SE) × 103] intrinsic permeability [(cm2/h ± SE) × 103] lag time (h ± SE) thickness (cm ± SE) partition coefficient diffusivity [(cm2/h ± SE) × 103
3.536(0.44)†,‡ 0.398(0.049)†,‡ 0.030(0.004)†,‡ 19.949(0.855)†,‡ 0.063(0.003)‡ 3.686(0.002)†,‡ 0.017(0.005)†,‡
9.537(0.89)†,‡ 0.993(0.158)†,‡ 0.059(0.006)† 13.637(1.401)† 0.063(0.003)‡ 2.590(0.079)† 0.026(0.002)†
18.044(1.40) 2.031(0.158) 0.071(0.006) 6.520(1.051) 0.035(0.002) 2.519(0.317) 0.028(0.002)
a Mean (±standard error of mean). †Difference in the corresponding permeation parameters between Yucatan and domestic skin is statistically significant (p e 0.05). ‡Differences in the corresponding permeation parameters between Yucatan (or domestic) pig skin and human cadaver skin are statistically significant (p e 0.05).
parameters obtained for the Yucatan minipig model (Table 2). Similar trends were observed for the domestic pig model. The only significant difference observed in the latter was the higher intrinsic permeability values for the ileal segment when compared to the duodenal and jejunal segments. However, none of the other permeation parameters were found to differ significantly. Thus, for further comparisons with mucosal membranes, these segments could be considered together as the set of intestinal segments for the respective species of pigs. Initially, the set of absorptive mucosae was compared to the intestinal segments. The intrinsic permeability and diffusivity of THA in intestinal segments was significantly higher than the mucosal membranes. This can be explained by taking the “real” surface area available for permeation into account. The absorption of nutrients is the major function of the intestine. To facilitate absorption, the mucosal surface of the intestine is covered with numerous folds, villi, and microvilli, which increase the surface area available for absorption several 100-fold over the apparent surface area.17 The apparent surface area available for permeation using the Valia-Chien permeation cells is 0.64 cm2. Thus the large “real” surface area could account for the higher intrinsic permeability and diffusivity of THA through intestinal segments. Further, the intestinal segments were considered as the control and the individual mucosae were compared to it. The apparent permeability of THA through the nasal mucosa was found to be significantly higher than the intestinal segments for both the species of pigs, while the intrinsic permeability through nasal and buccal mucosae was comparable to the intestinal segments. Thus, despite the differences in “real” surface area, these mucosae have comparable absorption of THA and may be considered as alternate routes for the delivery of THA. To evaluate the domestic pig as a model to study biomembrane permeation, the permeation characteristics of the domestic pig biomembranes were compared to the corresponding biomembranes of Yucatan minipigs. As mentioned earlier, it was found that the trends of permeation characteristics of the biomembranes were similar in both species of pigs. Similarly, the order of intrinsic permeability of THA across absorptive mucosae and intestinal segments was also similar:
[ileum, jejunum, duodenum] > [(nasal, buccal) > (rectal, sublingual)] > skin Comparing the corresponding biomembranes of the two species of pigs, it was seen that their permeation characteristics were similar to each other, with two exceptions, viz. the buccal mucosa and the skin. The buccal mucosa of the domestic pig had a significantly lower apparent permeability and intrinsic permeability for THA (Figure
3A,B), although differences in thickness and partition coefficient of THA were not significant. After accounting for the thickness and partition coefficient, the diffusivity of THA in both was comparable (Figure 3C). For both pig species, the permeation of THA through the nasal mucosa was the highest. This observation has been reported previously for other drugs.18-20 Permeation across biological membranes can occur mainly via two main pathways: the paracellular route (aqueous pore) or the transcellular route (lipoidal).21 Hydrophilic and charged molecules that do not partition well into the cell membrane are thus transported via the intercellular space while lipophilic permeants tend to be transported via the transcellular space.21 THA, a basic drug with a pKa of 10, exists as a positively charged molecule at pH 7.4 and is thus expected to be transported via the paracellular route. The higher permeability through the nasal mucosa could be a result of its slightly more hydrophilic nature than other mucosae.22,23 Under in vivo conditions, the highly vascular nasal mucosa can be expected to provide a rich surface for rapid drug absorption. However, this mucosa suffers from the disadvantage of very short contact time due to mucociliary clearance.24 As a result, nasal delivery of drugs is useful in those situations where rapid drug effect is required for short duration25 and thus may not be suitable for chronic therapy. The buccal route has been found to be more suitable for chronic drug delivery.26-28 Similar drug delivery devices may be envisioned for the prolonged systemic delivery of THA. The potential of the rectal route in improving the bioavailability of THA was demonstrated in a clinical trial by Ahlin and co-workers.29 Thus, the mucosae most likely to be useful for development of controlled drug delivery devices appear to be the buccal and the rectal mucosae. 3.3. Permeation through the SkinsAnother biomembrane which is suitable for chronic drug delivery is the skin. In the present study, it was seen that the permeation profiles of THA through the excised dorsal skin were the lowest among all the biomembranes for both the Yucatan minipigs and domestic pigs (Figure 4) and also displayed a long lag-time. This difference cannot be accounted for by simply considering thickness or skin/buffer partition coefficient. The low permeability of drugs through the skin compared to other biomembranes may be expected since the skin is covered by the stratum corneum, which is composed of several layers of dead keratinized cells and provides an efficient barrier to the permeation of molecules. However, it may be possible to significantly increase drug permeation through the skin by using chemical enhancers30 or physical methods.31,32 Significant enhancement of permeation of THA and reduction in the lag-time for permeation through the skin using solvent mixtures has been previously demonstrated both in vitro and in vivo in rats.33 Journal of Pharmaceutical Sciences / 445 Vol. 87, No. 4, April 1998
permeation characteristics of THA. In addition, the domestic pigs are more easily available and more economical than the Yucatan minipigs, and may thus serve as a suitable alternative to the Yucatan minipig model to study biomembrane permeation.
References and Notes
Figure 4sPermeation profiles of THA through the dorsal skin excised freshly from Yucatan minipigs and domestic pigs in comparison with human cadaver skin.
Thus, the evaluation of skin as a possible route for THA delivery merits further investigation. Comparing the THA permeation profiles through the dorsal skin of Yucatan minipigs and domestic pigs, all permeation parameters obtained were significantly different, although the thickness of the membranes obtained was similar (Table 3). The permeation of THA through the domestic pig skin was found to be almost 2-fold higher than that across the Yucatan minipig skin (Figure 4). To further evaluate the pig skin as a model to study THA permeation, it was compared to human cadaver skin which has been commonly considered as a model to study permeation of drugs in vitro. However, the human cadaver skin obtained was lower in thickness than both the pig species. The permeation profile of THA through the human cadaver skin was significantly higher than both the Yucatan and domestic pig skin (Figure 4). However, it was observed that intrinsic permeability and diffusivity of THA across human cadaver skin was comparable to that of domestic pig dorsal skin (Table 3). Thus, these data seems to suggest that the permeation characteristics of domestic pig skin are closer to human cadaver skin than the Yucatan pig skin and may thus be a better model to study transdermal permeation of drugs in vitro.
4. Conclusions The results of the present study indicate that despite the differences in the “real” surface area available for permeation, some of the absorptive mucosae show comparable THA permeability to the intestinal segments. The nasal route, though highly permeable, may not be suitable for chronic drug therapy, while the traditionally used sublingual route suffers from the disadvantage of lower permeability. Thus the mucosae that may be suitable for the development of controlled drug delivery devices for the chronic delivery of THA may be the buccal and the rectal mucosae. Controlled delivery of THA using these routes can be expected to increase bioavailability by avoiding hepatic “first-pass” metabolism and also reduce frequency of dosing. Another potential route for delivery of THA may be the transdermal route, although permeation enhancers would be required to achieve the required drug delivery rate. Last, it was observed that the biomembranes of the two species of swine were comparable with respect to 446 / Journal of Pharmaceutical Sciences Vol. 87, No. 4, April 1998
1. Briggs, R. S. J. Alzheimer’s disease: The clinical context. In Current Problems in Neurology; Davies, D. C., Ed.; John Libbey and Co. Ltd.: Paris, France, 1987; Vol. 11, pp 1-8. 2. Wagstaff, A. J.; McTavish, D. Drugs Aging 1994, 4 (6), 510540. 3. Crimson, M. L. Tacrine: The first drug approved for Alzheimer’s disease. Ann. Pharmacother. 1994, 28, 744-51. 4. Knapp, M. J.; Knopman, D. S.; Solomon, P. R.; Pendlebury, W.; Davies, C. S. J. A 30 week randomized controlled trial of high dose tacrine in patients with Alzheimer’s disease. Am. Med. Assoc. 1993, 271, 985-991. 5. Kauffer, D. I.; Cummings, J. L.; Christine, D. Effect of tacrine on behavioral symptoms in Alzheimer’s disease: An open label study. J Geriatric Psych. Neurol. 1996, 9 (1), 1-6. 6. Farlow, M.; Gracon, S. I.; Hershey, L. A.; Lewis, K. W.; Sadowsky, C. H.; Dolan-Urano, J. A controlled trial of tacrine in Alzheimer’s disease. J. Am. Med. Assoc. 1992, 268, 25232529. 7. Chien, Y. W. Recent advances in noninvasive systemic delivery of pharmaceuticals and biopharmaceuticals. Drug Dev. Ind. Pharm. 1994, 20 (4), 417-468. 8. Chien, Y. W. Biopharmaceutics basis for transmucosal delivery. S. T. P. Pharma Sci. 1995, 5 (4), 257-275. 9. Kurali, T. T. Comparison of the gastrointestinal anatomy, physiology and biochemistry of humans and commonly used laboratory animals. Biopharm. Drug Dispos. 1995, 16, 351380. 10. Swindle, M. M. Porcine models in surgical research. In Swine in Biomedical Research; Tumbleson, M. E., Ed.; Plenum Press: New York, 1986; Vol. 1, pp 235-239. 11. Qiao, G. L.; Williams, P. L.; Riviere, J. E. Percutaneous absorption, biotransformation, and systemic disposition of parathion in vivo in swine. I. Comprehensive pharmacokinetic model. Drug. Metab. Dispos. 1994, 22, 459-471. 12. Reifernrath, W. G.; Hawkins, G. S.; Kurtz, M. S. Percutaneous penetration and skin retention of topically applied compounds: An in vitro-in vivo study. J. Pharm. Sci. 1991, 80, 526-532. 13. Chien, Y. W.; Valia, K. Development of a dynamic skin permeation system for long-term permeation studies. Drug Dev. Ind. Pharm. 1984, 10, 575-599. 14. Chien, Y. W. Transdermal drug delivery and delivery systems. In Novel Drug Delivery Systems, 2nd ed.; Marcel Dekker Inc.: New York, 1992; pp 301-380. 15. Martin, A. Solubility and distribution phenomena. In Physical Pharmacy, 4th ed.; Lea and Febiger: Philadelphia, 1993; pp 212-250. 16. Harris, D.; Robinson, J. R. Drug delivery via the mucous membranes of the oral cavity. J. Pharm. Sci. 1992, 81 (1), 1-10. 17. Caspary, W. F. Physiology and pathophysiology of intestinal absorption. Am. J. Clin. Nutr. 1992, 55, 299S-308S. 18. Nair, M.; Chien, Y. W. Effect of steriochemistry of substituent groups on permeation across biological membranes. S. T. P. Pharma Sci. 1993, 3 (6), 425-430. 19. Corbo, D. C.; Liu, J. C.; Chien, Y. W. Drug absorption through mucosal membranes: Effect of mucosal route and penetrant hydrophilicity. Pharm. Res. 1989, 6 (10), 848-852. 20. Rojansakul, Y.; Wang, L. Y.; Bhat, M.; Glover, D. D.; Malanga, C. J.; Ma, J. K. H. The transport barrier of epithelia: A comparative study on membrane permeability and charge selectivity in the rabbit. Pharm. Res. 1992, 9 (8), 1029-1034. 21. Sayani, A. P.; Chien, Y. W. Systemic delivery of proteins and peptides across absorptive mucosae. Crit. Rev. Ther. Drug Carrier Syst. 1996, 13 (1&2), 85-184. 22. Corbo, D. C.; Liu, J.; Chien, Y. W. Characterization of the barrier properties of mucosal membranes. J. Pharm. Sci. 1990, 79 (3), 202-206. 23. Nair, M. K.; Chetty, D. J.; Ho, H.; Chien, Y. W. Biomembrane permeation of nicotine: Mechanistic studies with porcine mucosae and skin. J. Pharm. Sci. 1997, 86 (2), 257-262. 24. Edman, P.; Bjork, E. (D) Routes of delivery: Case studies (1) Nasal delivery of peptide drugs. Adv. Drug Del. Rev. 1992, 8, 165. 25. Ladau, A. J.; Eberhardt, R. T.; Frishman, W. H. Intranasal delivery of cardiovascular agents: An innovative approach
26.
27.
28. 29.
30.
to cardiovascular pharmacotherapy. Am. Heart J. 1994, 127 (6), 1594-1599. Cassidy, J.; Berner, B.; Chan, K.; John, V.; Toon, S.; Holt, B.; Rowland, M. Human transbuccal absorption of diclofenac sodium from a prototype hydrogel device. Pharm. Res. 1993, 10 (1), 126-129. DeGrande, G.; Benes, L.; Horrie`re, F.; Korsenty, H.; Lacoste, C.; McQuinn, R.; Guo, J.; Scherrer, R. Specialized oral mucosal drug delivery systems: patches. In Oral mucosal drug delivery; Rathbone, M. J., Ed.; Marcel Dekker inc.: New York, 1996; pp 285-318. de Vries, M. E.; Bodde´, H. E.; Verhoef, J. C.; Junginger, H. E. Developments in buccal drug delivery. Crit. Rev. Ther. Drug Carrier Syst. 1991, 8 (3), 271-303. Ahlin, A.; Adem, A.; Junte, T.; Ohman, G.; Nyback, H. Pharmacokinetics if tetrahydroaminoacridine: Relations to clinical and biochemical effects in Alzheimer patients. Int. J. Clin. Psychopharmacol. 1992, 7, 29-36. Williams, A. C.; Barry, B. W. Skin absorption enhancers. Crit. Rev. Ther. Drug Carrier Syst. 1992, 9 (3, 4), 305-353.
31. Prausnitz, M. R. Effects of electric current applied to skin: Review for transdermal drug delivery. Adv. Drug Del. Rev. 1996, 18 (1), 395-425. 32. Tyle, P.; Agrawala, P. Drug delivery by phonophoresis. Pharm. Res. 1989, 6 (5), 355-361. 33. Sathyan, G.; Ritschel, W. A.; Hussain, A. S. Transdermal delivery of tacrine: I. Identification of a suitable delivery vehicle. Int. J. Pharm. 1995, 114, 75-83.
Acknowledgments Authors wish to acknowledge Schering-Plough Foundation for the graduate research fellowship which funded this research. We would also like to thank Dr. Senshang Lin and Dr. Li-lan Chen for their help in the surgical techniques for excision of the biomembranes.
JS970359U
Journal of Pharmaceutical Sciences / 447 Vol. 87, No. 4, April 1998