Comparative Assessment of Intestinal Transport of Hydrophilic Drugs Between Small Intestine and Large Intestine Hiroaki Yuasa, Kenji Matsuda, Yukie Kimura, Naomi Soga, and Jun Watanabe
Drug Delivery Downloaded from informahealthcare.com by CDL-UC Davis on 10/27/14 For personal use only.
Faculty of Phamuzceutical Sciences, Nagoya City Universizy, Nagoya, Japan
Intestinal passive transport of several hydrophilic drugs (and probe compounds) was examined in the large intestine (colon), in comparison with that in the small intestine, in an effort to obtain basic information for developing rational colonic drug delivery strategies. The drugs tested were polyethylene glycol (PEG 900), L-glucose, D-XylOSe, 5-fluorouracil (5-FU) and urea. In everted intestinal sacs, the uptake of every drug was larger in the small intestine than in the large intestine, although by various extents. The uptake of urea was larger than those of D-XylOSe and L-glucose in both the small intestine and large intestine and associated with a larger large intestine (LI)/small intestine (SI) uptake ratio. Assuming that passive transport via the paracellular route (or aqueous pore) is predominant for them, the large intestine may have smaller paracellular (or aqueous) pores, restricting the transport of those monosaccharides compared with smaller molecules such as urea by a larger extent in the large intestine than in the small intestine. The passive transport of 5-FU was significantly larger than those of the monosaccharides in both the small intestine and large intestine and associated with a larger LYSI uptake ratio, even though 5-FU has a molecular weight close to that of the monosaccharides. 5-FU may be transported predominantly by transcellular diffusion, because its oil-to-water partition coefficient is about 200 times larger than those of the monosaccharides. Although transport mechanisms, including transport pathways, are yet to be fully clarified, drugs with physicochemical properties similar to those of 5-FU or urea may be more feasible for colonic drug delivery than those with physicochemical properties similar to those of monosaccharides. Keywords Colon, Intestinal Absorption, Midgut, Passive Transport, Rat
With recent advances in colon-specific controlled-release techniques, drug delivery via colon has been of increasing interest. For example, for peptide drugs that are easily degraded by intestinal peptidases, the colon, in which peptidase activities have been suggested to be lower than in the small intestine, may Received 11 June 1997; accepted 16 July 1997. This study was supported in part by a Grant-in-Aid for Scientific Research from the Ministry of Education, Science and Culture, Japan. Address correspondence to Hiroaki Yuasa, PhD, Faculty of Pharmaceutical Sciences, Nagoya City University, 3- 1 Tanabe-dori, Mizuhoku, Nagoya 467, Japan. Drug Delivery, 4269-272, 1997 Copyright 0 1997 Taylor & Francis 1071-7544/97$12.00 .00
+
be a potential and promising absorption site (Ritschel, 1991) if the colonic membrane is permeable to these drugs. However, information about the permeability of the colonic membrane is still scarce. Although several studies have suggested for passively absorbed compounds that, surprisingly, the apparent membrane permeability based on a cylindrical and smooth surface approximation of the intestinal tract may be comparable between the small intestine and large intestine (Hosoya et al. 1993; Kimura et al. 1994; Sasaki et al. 1994; Sawada et al. 1991; Tomita et al. 1992), our preceding studies (Yuasa et al. 1996b, 1997a) have suggested consistently in vitro, in situ, and in vivo that D-xylose, a passively absorbed pentose with a hydrophilic nature, is far less absorbable (or permeable) in the large intestine than in the small intestine, in agreement with the conventional notion that the large intestine is disadvantageous, compared with small intestine, for drug absorption because of the smaller surface area. We, therefore, further examined the membrane permeability of the large intestine in comparison with that of the small intestine in everted sacs in vitro and the intestinal loop in situ, using several hydrophilic drugs (and probe compounds) to obtain basic information for developing rational colonic drug delivery strategies. MATERIALS A N D METHODS Chemicals ~ - [l- ~ ~ C] Glucose (1.7 GBq/mmol), [14C]urea (2.1 GBq/ mmol), 3-0-[glucose-14C(U)]-methyl-~-glucose (3-0-MG, 11.7 GBqImmol), [ 1,2-14C]PEG 4000 (0.67 GBq/g), 5-[6-3H]FU (555 GBq/mmol), [ 1,2-3H]PEG 900 (2.05 GBq/g), and [ 1,23H]PEG 4000 (0.088 GBq/g) were purchased from DuPontNEN (Boston, MA, USA). ~-['~C(U)]Xylose (3.44 GBq/mmol) was purchased from Amersham International (Buckinghamshire, UK). Unlabeled 5-FU, urea, and PEG 900 were purchased from Wako Pure Chemical Industries (Osaka, Japan). Unlabeled L-glucose, 3-0-MG, and D-xylose were purchased from Sigma Chemical Co. (St. Louis, MO, USA). All other reagents were of analytical grade and commercially obtained.
269
Drug Delivery Downloaded from informahealthcare.com by CDL-UC Davis on 10/27/14 For personal use only.
270
H. YUASA ET AL.
Uptake Experiments in Everted Sacs Everted sacs (2 cm in length) were prepared from the small intestine (jejunum to midgut) and the large intestine (colon) of male Wistar rats, weighing about 300 g and without fasting, and uptake experiments were conducted as described previously (Yuasa et al. 1986, 1996a). Briefly, the initial uptake into the tissue was measured by radioactivity determination after incubating everted sacs at 37°C with a shaking rate of 100 strokeslmin in Krebs-Ringer-bicarbonate buffer (pH 7.4) containing an appropriate amount of an unlabeled drug (or probe compound) and a trace amount of the I4C- or 3H-labeled drug in combination with a trace amount of 3H- or I4C-labeled PEG 4000 as a nonabsorbable marker. The uptake was estimated by subtracting the amount in the adherent fluid and the initial adsorption to the everted sacs. The uptake rate was calculated by dividing the uptake by time in the initial uptake phase (1 min for 5-FU, urea, and 3-0-MG in the small intestine and 5 min for the others), where uptake was proportional to time, and then dividing by the concentration in the medium to obtain the uptake clearance (CL,,). The values (mean ? SE) of wet tissue weight for a unit length were 114 5 1 and 150 ? 4 mgkm, respectively, for the small intestine and the large intestine.
2o 1
d
0 0
5
10
15
20
Time (min) FIG. 1. Time courses of the uptake of urea in rat intestinal everted sacs. Results are represented as the mean ? SE ( n = 3) for the small intestine (0) and large intestine (a). The concentration of urea was 0.1 mM.
RESULTS The tested drugs (or probe compounds) of PEG 900, L-glucose, D-XylOSe, and urea have been generally considered to be transported passively. For 5-FU, uptake was measured at a Absorption in Intestinal loop The absorption of urea was evaluated in the 5-cm intestinal high concentration of 10 mM, at which transport by pyrimidine loop of the small intestine (midgut) or large intestine (colon) of carriers in the small intestine is saturated and passive transport male Wistar rats, weighing about 300 g and fasted overnight is predominant (Yuasa et al. 1996a). 3-O-Methyl-~-glucose prior to experiments, as described previously (Yuasa et al., (3-0-MG), which is transported by D-glucose carriers in the 1997a). The urea solution (0.1 mM) was prepared in phosphate small intestine (Hopfer 1987), was included for comparison. Figure 1 shows the time courses of the uptake (normalized buffer (pH 6.4) with trace amounts of ['4C]urea and L3H]PEG 4000 added as a nonabsorbable marker. Although surgical by concentration) of urea, one of those that showed the fastest operation was conducted under light ether anesthesia to cannu- uptake in the everted sacs. The uptake of urea was proportional late into the right jugular vein for blood sampling and inject the to time up to 1 min in the small intestine and 5 min in the large urea solution (0.5 mL) into the loop, the rats regained conscious- intestine. Therefore, in the large intestine, the initial uptake was ness shortly after the operation, being unanesthetized for the measured at 5 min for the all drugs. In the small intestine, the most of the experimental period. Plasma samples were obtained initial uptake was measured at 1 min for urea and those (5-FU periodically and the luminal solution was recovered at the end and 3-0-MG) whose uptake clearances were comparable to that of experiments (30 min after administration). The concentra- of urea, and 5 min for those (PEG 900, L-glucose, and D-xylose) tions of urea and PEG 4000 were measured by radioactivity whose uptake clearances were comparable to or smaller than that of urea in the large intestine. determination. The uptake of every drug was smaller in the large intestine The fraction absorbed (F,) was estimated as the fraction that disappeared from the intestinal lumen, correcting for minor than in the small intestine, although by various extents (Figure volume changes based on changes in PEG 4000 concentrations. 2). In the large intestine, the uptake of D-xylose, a pentose, was Assuming first-order absorption (disappearance), the absorp- not at a detectable level, in agreement with our preceding study (Yuasa et al. 1997a); the uptake of L-glucose, a hexose, was also tion rate constant (k,) was estimated as follows: insignificant; urea showed the largest uptake. For urea, we further examined its absorption in the intestinal ln(1 - Fa) k =(1) loop (Figure 3A and Table 1). On the basis of both appearance t in plasma and disappearance from the lumen after administrawhere t represents the absorption (experimental) period. The tion into the intestinal loop, the absorption of urea was smaller apparent membrane permeability clearance (CL,,,) was esti- in the large intestine than in the small intestine by about 60 to mated as the product of k, and the luminal volume (100 pL/cm 85%, consistent with the results in everted sacs. The absorption of D-xylose in the loop was, from our preceding report (Yuasa et as administered volume).
27 1
INTESTINAL TRANSPORT OF HYDROPHILIC DRUGS
TABLE 1 Large intestine (LI)/small intestine (SI) ratio of intestinal membrane permeability in rats
T
B
n
Drug Delivery Downloaded from informahealthcare.com by CDL-UC Davis on 10/27/14 For personal use only.
0
m
CL (uL/min/cm)"
L
Drug Urea D-Xylose
51
FIG. 2. Comparison of uptake clearances in rat everted sacs between small intestine and large intestine. Results are represented as the mean _f SE (n = 3). The open and hatched bars represent the results for the small intestine and the large intestine, respectively. ND, Not detected. Initial uptake was measured at 1 rnin for 5-W, urea, and 3-0-MG in the small intestine and at 5 min for the others. The concentrations were as follows: PEG 900, 3 mM; L-glucose and o-xylose, 1 mM; 5-FU, 10 mM; urea, 0.1 mM; 3-0-MG, 0.3 pM.
al. 1997a), more than an order of magnitude smaller in the large intestine than in the small intestine (Figure 3B and Table l), also consistent with the results for everted sacs. In the small intestine perfused in situ, the absorption of L-glucose was previously shown to be comparable to that of D-xylose (Yuasa et al. 1997b), and those of 5-FU and 3-0-MG have been suggested to be comparable to that of urea (Yuasa et al. 1989, 1996a). It should also be noted that the uptake clearances in everted sacs are comparable to disappearance clearances in the loop in situ. Thus, the initial uptake in everted sacs can be assumed to be quite relevant to the transepithelial transport in situ and in vivo. The only exception is PEG 900, as it was 0.4
1 (*)
0.4
7 (B)
T
0.3
0.2 0.1
0.0
0.0 0
20
40
Time (min)
60
0
20
40
60
Time (min)
FIG. 3. Plasma concentrations of urea (A) and D-XylOSe (B) after administration into the loop of rat small intestine and large intestine. Results are represented as the mean & SE ( n = 3) for the small intestine ( 0 )and large intestine (a). The data for o-xylose are from our preceding report (Yuasa et al., 1997a).
Method
LI
SI
LUSI
Everted sacs Loop Everted sacs Loop"
1.72 ? 0.38 0.60 ? 0.04 ND 0.09 2 0.05
4.74 & 0.37 4.02 2 0.53 1.44 C 0.07 1.44 C 0.22
0.363 0.149 0 0.063
"The CL values, uptake clearance (CL,,) for everted sacs and apparent membrane permeability clearance (CL,,) for the loop are represented as the mean 2 SE ( n = 3). ND, Not detected. "The data for D-xylose in the loop are cited from our preceding report (Yuasa e t a]., 1997a).
suggested to penetrate gradually into the tissue before being transported into the blood stream only slowly. Although the uptake of PEG 900 was comparable to those of D-xylose and L-glucose, its fraction absorbed (disappeared) from the loop of the small intestine was, as evaluated in rats anesthetized with urethane, 3.9 2 0.4% (mean 5 SE, n = 3) in 30 min, giving a CL,, value of 0.134 5 0.014 pL/min/cm, which is only about 10% of that of D-xylose in unanesthetized rats. Although the CL,,, of PEG 900 may be lowered by about 50% by the effect of anesthesia (Yuasa et al. 1993), PEG 900 still seems to be transported into the blood stream far more slowly than those monosaccharides. DISCUSS1ON The large intestine (LI)/small intestine (SI) ratio of the uptake clearance tended to be smaller for L-glucose and D-xylose, being below 0.1, than for urea, being above 0.1. Assuming that these hydrophilic drugs are passively transported mainly by the paracellular route (or aqueous pore), the large intestine may have smaller paracellular (or aqueous) pores, restricting the transport of those monosaccharides, compared with smaller molecules such as urea, to a larger extent in the large intestine than in the small intestine. The passive transport of 5-FU was significantly larger than those of the monosaccharides in both the small intestine and large intestine and associated with a larger LI-to-SI uptake ratio, even though 5-FU has a molecular weight close to that of monosaccharides. Because the oil-to-water partition coefficient (0.11) of 5-FU is about 200 times larger than those (about 0.0005) of monosaccharides (Leo et al. 1971), the larger uptake may be attributable to predominant transcellular diffusion. There is also a possibility that the monosaccharides may have a weak affinity for D-glucose transporters in the small intestine, resulting in larger transport in the small intestine than in the large intestine. 3-0-MG, which is transported by D-glucose carriers in the small intestine, also showed a relatively small LI-to-SI uptake ratio (below 0.1) as L-glucose and D-xylose.
Drug Delivery Downloaded from informahealthcare.com by CDL-UC Davis on 10/27/14 For personal use only.
272
H. YUASA ET AL.
To maintain cylindrical geometry, both small intestine and large intestine require 100 pLkm of solution inside the lumen or everted sacs, suggesting that they have comparable surface areas ( S ) with the cylindrical and smooth surface approximation. Therefore, LI-to-SI uptake ratios are maintained even if uptake clearances (CL,, = Pup.S) are converted to conventional permeability coefficients (Pup) based on the smooth surface approximation. The result that LI-to-SI ratios are smaller than 1 is in agreement with the conventional notion that the large intestine is disadvantageous, compared with the small intestine for drug absorption because of a smaller surface area, although not only differences in anatomical surface areas but also potential differences in transcellular and paracellular permeabilities need to be clarified in relation to the LI-to-SI uptake ratio. In conclusion, the present study has revealed, for several hydrophilic drugs (and probe compounds), that intestinal transport (membrane permeability) is consistently and significantly smaller in the large intestine than in the small intestine. Although transport mechanisms, including transport pathways, are yet to be fully clarified, drugs with physicochemical properties similar to those of 5-FU or urea may be more feasible for colonic drug delivery than those with physicochemical properties similar to those of monosaccharides. It requires more extensive investigations to clarify transport mechanisms and to devise rational colonic drug delivery strategies. As discussed in our preceding report (Yuasa et al. 1997a), suggestions of comparable membrane permeability between small intestine and large intestine from the measurement of transport across the intestinal tissue, typically in an Ussing chamber, may need to be reexamined for their relevance to in vivo absorption.
REFERENCES Hopfer, U. 1987. Membrane transport mechanisms for hexoses and amino acids in the small intestine. In Physiology of the Gastrointestinal Tract, ed. L. R. Johnson et al., 1499-1526. New York: Raven Press.
Hosoya, K., Kubo, H., Natsume, H., Sugibayashi, K., Morimoto, Y.,and Yamashita, S. 1993. The structural barrier of absorptive mucosae: Site difference of the permeability of fluorescein isothiocyanate-labelled dextran in rabbits. Biopharm. Drug Dispos. 14:685-696. Kimura, T., Sudo, K., Kanzaki, K., Miki, K., Takeichi, Y., Kurosaki, Y., and Nakayama, T. 1994. Drug absorption from large intestine: Physicochemical factors governing drug absorption. Biol. Pharm. Bull. 17:327-333. Leo, A., Hansch, C., and Elkins, D. 1971. Partition coefficients and their uses. Chem. Rev. 71:525-616. Ritschel, W. A. 1991. Microemulsions for improved peptide absorption from the gastrointestinal tract. Methods Find. Exp. Clin. Pharmacol. 13:205220. Sasaki, I., Fujita, T., Murakami, M., Yamamoto, A,, Nakamura, E., Imasaki, H.. and Muranishi, S. 1994. Intestinal absorption of azetirelin, a new thyrotropinreleasing hormone (TRH) analogue. I. Possible factors for the low oral bioavailability in rats. Biol. Pharm. Bull. 17: 1256-1261. Sawada, T., Ogawa, T., Tomita, M., Hayashi, M., and Awazu, S. 1991. Role of paracellular pathway in nonelectrolyte permeation across rat colon epithelium enhanced by sodium caprate and sodium caprylate. Pharm. Res 8:1365-1371. Tomita, M., Sawada, T., Ogawa, T.,Ouchi, H., Hayashi, M., and Awazu, S. 1992. Differences in the enhancing effects of sodium caprate on colonic and jejunal drug absorption. Pharm. Res. 9:648453. Yuasa, H., Miyamoto, Y., Iga, T., and Hanano, M. 1986. Determination of kinetic parameters of a carrier-mediated transport in the perfused intestine by two-dimensional laminar flow model: Effects of the unstined water layer. Biochim. Biophys. Acta 856:219-230. Yuasa, H., Iga, T., Hanano, M., and Watanabe, J. 1989. Relationship between the first-order intestinal absorption rate constant in vivo and the membrane permeability clearance in a perfusion system: An intragastric administration method in vivo. J. Pharmacobiodyn. 12:264-271. Yuasa, H., Matsuda, K., and Watanabe, J. 1993. Influence of anesthetic regimens on intestinal absorption in rats. Pharm. Res. 10884-888. Yuasa, H., Matsuhisa, E., and Watanabe, J. 1996a. Intestinal brush border transport mechanism of 5-fluorouracil in rats. Biol. Pharm. Bull. 19:94-99. Yuasa, H., Kuno, C., and Watanabe, J. 199613. Evaluation of the fractional absorption of D-XylOSe by the analysis of gastrointestinal disposition after oral administration in rats. Biol. Pharm. Bull. 19:604-607. Yuasa, H., Kuno, C., and Watanabe, J. 1997a. Comparative assessment of D-XylOSe absorption between small intestine and large intestine. J. Pharm. Pharmacol. 49:26-29. Yuasa, H., Soga, N., Kimura, Y., and Watanabe, J. 1997b. Effect of aging on the intestinal transport of hydrophilic drugs in the rat small intestine. Biol. Pharm. Bull., in press.
Drug Delivery Downloaded from informahealthcare.com by CDL-UC Davis on 10/27/14 For personal use only.
Faculty of Phamuzceutical Sciences, Nagoya City Universizy, Nagoya, Japan
Intestinal passive transport of several hydrophilic drugs (and probe compounds) was examined in the large intestine (colon), in comparison with that in the small intestine, in an effort to obtain basic information for developing rational colonic drug delivery strategies. The drugs tested were polyethylene glycol (PEG 900), L-glucose, D-XylOSe, 5-fluorouracil (5-FU) and urea. In everted intestinal sacs, the uptake of every drug was larger in the small intestine than in the large intestine, although by various extents. The uptake of urea was larger than those of D-XylOSe and L-glucose in both the small intestine and large intestine and associated with a larger large intestine (LI)/small intestine (SI) uptake ratio. Assuming that passive transport via the paracellular route (or aqueous pore) is predominant for them, the large intestine may have smaller paracellular (or aqueous) pores, restricting the transport of those monosaccharides compared with smaller molecules such as urea by a larger extent in the large intestine than in the small intestine. The passive transport of 5-FU was significantly larger than those of the monosaccharides in both the small intestine and large intestine and associated with a larger LYSI uptake ratio, even though 5-FU has a molecular weight close to that of the monosaccharides. 5-FU may be transported predominantly by transcellular diffusion, because its oil-to-water partition coefficient is about 200 times larger than those of the monosaccharides. Although transport mechanisms, including transport pathways, are yet to be fully clarified, drugs with physicochemical properties similar to those of 5-FU or urea may be more feasible for colonic drug delivery than those with physicochemical properties similar to those of monosaccharides. Keywords Colon, Intestinal Absorption, Midgut, Passive Transport, Rat
With recent advances in colon-specific controlled-release techniques, drug delivery via colon has been of increasing interest. For example, for peptide drugs that are easily degraded by intestinal peptidases, the colon, in which peptidase activities have been suggested to be lower than in the small intestine, may Received 11 June 1997; accepted 16 July 1997. This study was supported in part by a Grant-in-Aid for Scientific Research from the Ministry of Education, Science and Culture, Japan. Address correspondence to Hiroaki Yuasa, PhD, Faculty of Pharmaceutical Sciences, Nagoya City University, 3- 1 Tanabe-dori, Mizuhoku, Nagoya 467, Japan. Drug Delivery, 4269-272, 1997 Copyright 0 1997 Taylor & Francis 1071-7544/97$12.00 .00
+
be a potential and promising absorption site (Ritschel, 1991) if the colonic membrane is permeable to these drugs. However, information about the permeability of the colonic membrane is still scarce. Although several studies have suggested for passively absorbed compounds that, surprisingly, the apparent membrane permeability based on a cylindrical and smooth surface approximation of the intestinal tract may be comparable between the small intestine and large intestine (Hosoya et al. 1993; Kimura et al. 1994; Sasaki et al. 1994; Sawada et al. 1991; Tomita et al. 1992), our preceding studies (Yuasa et al. 1996b, 1997a) have suggested consistently in vitro, in situ, and in vivo that D-xylose, a passively absorbed pentose with a hydrophilic nature, is far less absorbable (or permeable) in the large intestine than in the small intestine, in agreement with the conventional notion that the large intestine is disadvantageous, compared with small intestine, for drug absorption because of the smaller surface area. We, therefore, further examined the membrane permeability of the large intestine in comparison with that of the small intestine in everted sacs in vitro and the intestinal loop in situ, using several hydrophilic drugs (and probe compounds) to obtain basic information for developing rational colonic drug delivery strategies. MATERIALS A N D METHODS Chemicals ~ - [l- ~ ~ C] Glucose (1.7 GBq/mmol), [14C]urea (2.1 GBq/ mmol), 3-0-[glucose-14C(U)]-methyl-~-glucose (3-0-MG, 11.7 GBqImmol), [ 1,2-14C]PEG 4000 (0.67 GBq/g), 5-[6-3H]FU (555 GBq/mmol), [ 1,2-3H]PEG 900 (2.05 GBq/g), and [ 1,23H]PEG 4000 (0.088 GBq/g) were purchased from DuPontNEN (Boston, MA, USA). ~-['~C(U)]Xylose (3.44 GBq/mmol) was purchased from Amersham International (Buckinghamshire, UK). Unlabeled 5-FU, urea, and PEG 900 were purchased from Wako Pure Chemical Industries (Osaka, Japan). Unlabeled L-glucose, 3-0-MG, and D-xylose were purchased from Sigma Chemical Co. (St. Louis, MO, USA). All other reagents were of analytical grade and commercially obtained.
269
Drug Delivery Downloaded from informahealthcare.com by CDL-UC Davis on 10/27/14 For personal use only.
270
H. YUASA ET AL.
Uptake Experiments in Everted Sacs Everted sacs (2 cm in length) were prepared from the small intestine (jejunum to midgut) and the large intestine (colon) of male Wistar rats, weighing about 300 g and without fasting, and uptake experiments were conducted as described previously (Yuasa et al. 1986, 1996a). Briefly, the initial uptake into the tissue was measured by radioactivity determination after incubating everted sacs at 37°C with a shaking rate of 100 strokeslmin in Krebs-Ringer-bicarbonate buffer (pH 7.4) containing an appropriate amount of an unlabeled drug (or probe compound) and a trace amount of the I4C- or 3H-labeled drug in combination with a trace amount of 3H- or I4C-labeled PEG 4000 as a nonabsorbable marker. The uptake was estimated by subtracting the amount in the adherent fluid and the initial adsorption to the everted sacs. The uptake rate was calculated by dividing the uptake by time in the initial uptake phase (1 min for 5-FU, urea, and 3-0-MG in the small intestine and 5 min for the others), where uptake was proportional to time, and then dividing by the concentration in the medium to obtain the uptake clearance (CL,,). The values (mean ? SE) of wet tissue weight for a unit length were 114 5 1 and 150 ? 4 mgkm, respectively, for the small intestine and the large intestine.
2o 1
d
0 0
5
10
15
20
Time (min) FIG. 1. Time courses of the uptake of urea in rat intestinal everted sacs. Results are represented as the mean ? SE ( n = 3) for the small intestine (0) and large intestine (a). The concentration of urea was 0.1 mM.
RESULTS The tested drugs (or probe compounds) of PEG 900, L-glucose, D-XylOSe, and urea have been generally considered to be transported passively. For 5-FU, uptake was measured at a Absorption in Intestinal loop The absorption of urea was evaluated in the 5-cm intestinal high concentration of 10 mM, at which transport by pyrimidine loop of the small intestine (midgut) or large intestine (colon) of carriers in the small intestine is saturated and passive transport male Wistar rats, weighing about 300 g and fasted overnight is predominant (Yuasa et al. 1996a). 3-O-Methyl-~-glucose prior to experiments, as described previously (Yuasa et al., (3-0-MG), which is transported by D-glucose carriers in the 1997a). The urea solution (0.1 mM) was prepared in phosphate small intestine (Hopfer 1987), was included for comparison. Figure 1 shows the time courses of the uptake (normalized buffer (pH 6.4) with trace amounts of ['4C]urea and L3H]PEG 4000 added as a nonabsorbable marker. Although surgical by concentration) of urea, one of those that showed the fastest operation was conducted under light ether anesthesia to cannu- uptake in the everted sacs. The uptake of urea was proportional late into the right jugular vein for blood sampling and inject the to time up to 1 min in the small intestine and 5 min in the large urea solution (0.5 mL) into the loop, the rats regained conscious- intestine. Therefore, in the large intestine, the initial uptake was ness shortly after the operation, being unanesthetized for the measured at 5 min for the all drugs. In the small intestine, the most of the experimental period. Plasma samples were obtained initial uptake was measured at 1 min for urea and those (5-FU periodically and the luminal solution was recovered at the end and 3-0-MG) whose uptake clearances were comparable to that of experiments (30 min after administration). The concentra- of urea, and 5 min for those (PEG 900, L-glucose, and D-xylose) tions of urea and PEG 4000 were measured by radioactivity whose uptake clearances were comparable to or smaller than that of urea in the large intestine. determination. The uptake of every drug was smaller in the large intestine The fraction absorbed (F,) was estimated as the fraction that disappeared from the intestinal lumen, correcting for minor than in the small intestine, although by various extents (Figure volume changes based on changes in PEG 4000 concentrations. 2). In the large intestine, the uptake of D-xylose, a pentose, was Assuming first-order absorption (disappearance), the absorp- not at a detectable level, in agreement with our preceding study (Yuasa et al. 1997a); the uptake of L-glucose, a hexose, was also tion rate constant (k,) was estimated as follows: insignificant; urea showed the largest uptake. For urea, we further examined its absorption in the intestinal ln(1 - Fa) k =(1) loop (Figure 3A and Table 1). On the basis of both appearance t in plasma and disappearance from the lumen after administrawhere t represents the absorption (experimental) period. The tion into the intestinal loop, the absorption of urea was smaller apparent membrane permeability clearance (CL,,,) was esti- in the large intestine than in the small intestine by about 60 to mated as the product of k, and the luminal volume (100 pL/cm 85%, consistent with the results in everted sacs. The absorption of D-xylose in the loop was, from our preceding report (Yuasa et as administered volume).
27 1
INTESTINAL TRANSPORT OF HYDROPHILIC DRUGS
TABLE 1 Large intestine (LI)/small intestine (SI) ratio of intestinal membrane permeability in rats
T
B
n
Drug Delivery Downloaded from informahealthcare.com by CDL-UC Davis on 10/27/14 For personal use only.
0
m
CL (uL/min/cm)"
L
Drug Urea D-Xylose
51
FIG. 2. Comparison of uptake clearances in rat everted sacs between small intestine and large intestine. Results are represented as the mean _f SE (n = 3). The open and hatched bars represent the results for the small intestine and the large intestine, respectively. ND, Not detected. Initial uptake was measured at 1 rnin for 5-W, urea, and 3-0-MG in the small intestine and at 5 min for the others. The concentrations were as follows: PEG 900, 3 mM; L-glucose and o-xylose, 1 mM; 5-FU, 10 mM; urea, 0.1 mM; 3-0-MG, 0.3 pM.
al. 1997a), more than an order of magnitude smaller in the large intestine than in the small intestine (Figure 3B and Table l), also consistent with the results for everted sacs. In the small intestine perfused in situ, the absorption of L-glucose was previously shown to be comparable to that of D-xylose (Yuasa et al. 1997b), and those of 5-FU and 3-0-MG have been suggested to be comparable to that of urea (Yuasa et al. 1989, 1996a). It should also be noted that the uptake clearances in everted sacs are comparable to disappearance clearances in the loop in situ. Thus, the initial uptake in everted sacs can be assumed to be quite relevant to the transepithelial transport in situ and in vivo. The only exception is PEG 900, as it was 0.4
1 (*)
0.4
7 (B)
T
0.3
0.2 0.1
0.0
0.0 0
20
40
Time (min)
60
0
20
40
60
Time (min)
FIG. 3. Plasma concentrations of urea (A) and D-XylOSe (B) after administration into the loop of rat small intestine and large intestine. Results are represented as the mean & SE ( n = 3) for the small intestine ( 0 )and large intestine (a). The data for o-xylose are from our preceding report (Yuasa et al., 1997a).
Method
LI
SI
LUSI
Everted sacs Loop Everted sacs Loop"
1.72 ? 0.38 0.60 ? 0.04 ND 0.09 2 0.05
4.74 & 0.37 4.02 2 0.53 1.44 C 0.07 1.44 C 0.22
0.363 0.149 0 0.063
"The CL values, uptake clearance (CL,,) for everted sacs and apparent membrane permeability clearance (CL,,) for the loop are represented as the mean 2 SE ( n = 3). ND, Not detected. "The data for D-xylose in the loop are cited from our preceding report (Yuasa e t a]., 1997a).
suggested to penetrate gradually into the tissue before being transported into the blood stream only slowly. Although the uptake of PEG 900 was comparable to those of D-xylose and L-glucose, its fraction absorbed (disappeared) from the loop of the small intestine was, as evaluated in rats anesthetized with urethane, 3.9 2 0.4% (mean 5 SE, n = 3) in 30 min, giving a CL,, value of 0.134 5 0.014 pL/min/cm, which is only about 10% of that of D-xylose in unanesthetized rats. Although the CL,,, of PEG 900 may be lowered by about 50% by the effect of anesthesia (Yuasa et al. 1993), PEG 900 still seems to be transported into the blood stream far more slowly than those monosaccharides. DISCUSS1ON The large intestine (LI)/small intestine (SI) ratio of the uptake clearance tended to be smaller for L-glucose and D-xylose, being below 0.1, than for urea, being above 0.1. Assuming that these hydrophilic drugs are passively transported mainly by the paracellular route (or aqueous pore), the large intestine may have smaller paracellular (or aqueous) pores, restricting the transport of those monosaccharides, compared with smaller molecules such as urea, to a larger extent in the large intestine than in the small intestine. The passive transport of 5-FU was significantly larger than those of the monosaccharides in both the small intestine and large intestine and associated with a larger LI-to-SI uptake ratio, even though 5-FU has a molecular weight close to that of monosaccharides. Because the oil-to-water partition coefficient (0.11) of 5-FU is about 200 times larger than those (about 0.0005) of monosaccharides (Leo et al. 1971), the larger uptake may be attributable to predominant transcellular diffusion. There is also a possibility that the monosaccharides may have a weak affinity for D-glucose transporters in the small intestine, resulting in larger transport in the small intestine than in the large intestine. 3-0-MG, which is transported by D-glucose carriers in the small intestine, also showed a relatively small LI-to-SI uptake ratio (below 0.1) as L-glucose and D-xylose.
Drug Delivery Downloaded from informahealthcare.com by CDL-UC Davis on 10/27/14 For personal use only.
272
H. YUASA ET AL.
To maintain cylindrical geometry, both small intestine and large intestine require 100 pLkm of solution inside the lumen or everted sacs, suggesting that they have comparable surface areas ( S ) with the cylindrical and smooth surface approximation. Therefore, LI-to-SI uptake ratios are maintained even if uptake clearances (CL,, = Pup.S) are converted to conventional permeability coefficients (Pup) based on the smooth surface approximation. The result that LI-to-SI ratios are smaller than 1 is in agreement with the conventional notion that the large intestine is disadvantageous, compared with the small intestine for drug absorption because of a smaller surface area, although not only differences in anatomical surface areas but also potential differences in transcellular and paracellular permeabilities need to be clarified in relation to the LI-to-SI uptake ratio. In conclusion, the present study has revealed, for several hydrophilic drugs (and probe compounds), that intestinal transport (membrane permeability) is consistently and significantly smaller in the large intestine than in the small intestine. Although transport mechanisms, including transport pathways, are yet to be fully clarified, drugs with physicochemical properties similar to those of 5-FU or urea may be more feasible for colonic drug delivery than those with physicochemical properties similar to those of monosaccharides. It requires more extensive investigations to clarify transport mechanisms and to devise rational colonic drug delivery strategies. As discussed in our preceding report (Yuasa et al. 1997a), suggestions of comparable membrane permeability between small intestine and large intestine from the measurement of transport across the intestinal tissue, typically in an Ussing chamber, may need to be reexamined for their relevance to in vivo absorption.
REFERENCES Hopfer, U. 1987. Membrane transport mechanisms for hexoses and amino acids in the small intestine. In Physiology of the Gastrointestinal Tract, ed. L. R. Johnson et al., 1499-1526. New York: Raven Press.
Hosoya, K., Kubo, H., Natsume, H., Sugibayashi, K., Morimoto, Y.,and Yamashita, S. 1993. The structural barrier of absorptive mucosae: Site difference of the permeability of fluorescein isothiocyanate-labelled dextran in rabbits. Biopharm. Drug Dispos. 14:685-696. Kimura, T., Sudo, K., Kanzaki, K., Miki, K., Takeichi, Y., Kurosaki, Y., and Nakayama, T. 1994. Drug absorption from large intestine: Physicochemical factors governing drug absorption. Biol. Pharm. Bull. 17:327-333. Leo, A., Hansch, C., and Elkins, D. 1971. Partition coefficients and their uses. Chem. Rev. 71:525-616. Ritschel, W. A. 1991. Microemulsions for improved peptide absorption from the gastrointestinal tract. Methods Find. Exp. Clin. Pharmacol. 13:205220. Sasaki, I., Fujita, T., Murakami, M., Yamamoto, A,, Nakamura, E., Imasaki, H.. and Muranishi, S. 1994. Intestinal absorption of azetirelin, a new thyrotropinreleasing hormone (TRH) analogue. I. Possible factors for the low oral bioavailability in rats. Biol. Pharm. Bull. 17: 1256-1261. Sawada, T., Ogawa, T., Tomita, M., Hayashi, M., and Awazu, S. 1991. Role of paracellular pathway in nonelectrolyte permeation across rat colon epithelium enhanced by sodium caprate and sodium caprylate. Pharm. Res 8:1365-1371. Tomita, M., Sawada, T., Ogawa, T.,Ouchi, H., Hayashi, M., and Awazu, S. 1992. Differences in the enhancing effects of sodium caprate on colonic and jejunal drug absorption. Pharm. Res. 9:648453. Yuasa, H., Miyamoto, Y., Iga, T., and Hanano, M. 1986. Determination of kinetic parameters of a carrier-mediated transport in the perfused intestine by two-dimensional laminar flow model: Effects of the unstined water layer. Biochim. Biophys. Acta 856:219-230. Yuasa, H., Iga, T., Hanano, M., and Watanabe, J. 1989. Relationship between the first-order intestinal absorption rate constant in vivo and the membrane permeability clearance in a perfusion system: An intragastric administration method in vivo. J. Pharmacobiodyn. 12:264-271. Yuasa, H., Matsuda, K., and Watanabe, J. 1993. Influence of anesthetic regimens on intestinal absorption in rats. Pharm. Res. 10884-888. Yuasa, H., Matsuhisa, E., and Watanabe, J. 1996a. Intestinal brush border transport mechanism of 5-fluorouracil in rats. Biol. Pharm. Bull. 19:94-99. Yuasa, H., Kuno, C., and Watanabe, J. 199613. Evaluation of the fractional absorption of D-XylOSe by the analysis of gastrointestinal disposition after oral administration in rats. Biol. Pharm. Bull. 19:604-607. Yuasa, H., Kuno, C., and Watanabe, J. 1997a. Comparative assessment of D-XylOSe absorption between small intestine and large intestine. J. Pharm. Pharmacol. 49:26-29. Yuasa, H., Soga, N., Kimura, Y., and Watanabe, J. 1997b. Effect of aging on the intestinal transport of hydrophilic drugs in the rat small intestine. Biol. Pharm. Bull., in press.
