Further mechanistic study on intestinal absorption enhanced by unsaturated fatty acids: reversible effect by sulfhydryl modification

Further mechanistic study on intestinal absorption enhanced by unsaturated fatty acids: reversible effect by sulfhydryl modification

Biochimica et Bio'physicaActa, 1117 (1992) 83-89 83 © 1992 Elsevier Science Publishers B.V. All rights reserved 0304-4165/92/$05.00 BBAGEN 23685 F...

624KB Sizes 4 Downloads 53 Views

Biochimica et Bio'physicaActa, 1117 (1992) 83-89

83

© 1992 Elsevier Science Publishers B.V. All rights reserved 0304-4165/92/$05.00

BBAGEN 23685

Further mechanistic study on intestinal absorption enhanced by unsaturated fatty acids: reversible effect by sulfhydryl modification Masahiro Murakami, Kanji Takada * and Shozo Muranishi Department of Biopharmaceutics, Kyoto Pharmaceutical Unit,ersity, Kyoto (Japan)

(Received 27 January 1992) Key words: Unsaturated fatty acid; Absorption enhancer; Intestinal absorption; Diamide; N-Ethylmaleimide;Oleic acid; Carboxyfluorescein; Passive transport In order to study the relationship between the sulfhydryl (SH) modification of membrane-associated proteins and the oleic acid-induced permeability enhancement of the colonic mucosa, in vitro and in situ absorption studies were performed using rat colon and carboxyfluorescein as an impermeable dye. The pretreatment of the mucosa with diamide, a bifunctional sulfhydryl modifier, in in vitro experiments with the everted colonic loops reduced the absorption enhancing effect of oleic acid in a concentration-dependent manner, less inhibitory effect, though just a little, was observed as compared to N-ethylmaleimide. The inhibition caused by the addition of diamide was absolutely restored by exposure of the mucosa to dithiothreitol. On the other hand, these SH modifiers showed no pronounced effect on the in vivo permeability of quinine which is well-known to be absorbed by a passive transport system mainly via the membrane lipid bilayer. These results obtained in the present study have identified an important role of the functional SH groups of membrane proteins on modulating the permeability alteration of the mucosal epithelium provoked by oleic acid. Furthermore, the SH proteins have been revealed as being unimportant in the intestinal absorption of lipoid-soluble compounds. Introduction The gastrointestinal tract has an epithelial cell sheet on the luminal surface as a highly selective permeable barrier which does not permit the transfer of most polar foreign compounds. Various highly polal~ and high molecular weight therapeutic drugs are generally poorly absorbed from the intestine due to their poor mucosal permeability. Therefore, these drugs must be clinically administered by an injection. In recent years, much attention has been focused on the co-administration of absorption-enhancing agents with poorly absorbable drugs in order to improve their bioavailabilities via the enteral routes [1-3]. However, absorption mechanisms by such enhancing agents are rather speculative and have not been thoroughly clarified until now; further studies are required from a biochemical viewpoint. Unsaturated long-chain fatty acids such as oleic acid and linoleic acid have a prominent absorption-enhanc-

Correspondence to: M. Murakami, Department of Biopharmaceutics, Kyoto Pharmaceutical University, Yamashina-ku, Kyoto, 607, Japan. * Present address: Department of Pharmacokinetics, Kyoto Pharmaceutical University, Kyoto, Japan.

ing efficiency among natural substances [4]. Research efforts in our laboratory have been concerned with the evaluation of their abilities to improve the bioavailabilities of various poorly absorbable drugs including gentamicin [5], heparin [6], interferon [7], etc., in in vivo animal studies. Furthermore, we have characterized the absorption-enhancing effects of unsaturated longchain fatty acids [8] and investigated their enhancing mechanism by which these fatty acids render the polar compounds possible to p e r m e a t e through the gut epithelial cell layer [9]. It has been shown that the perturbating effects of fatty acids on a lipid bilayer are correlated with their enhancement in the permeability of the model m e m b r a n e [10]. The destabilization of the lipid bilayer by the incorporation of fatty acids is considered to be one of the essential processes involved in the mechanism of absorption enhancement. However, this cannot entirely explain the whole absorption-enhancing mechanism. Our previous study using several membrane-modifying reagents describes the contribution of membranebound proteins to th enhancement of intestinal permeability induced by oleic acid [11]. The permeation-enhancing effect of oleic acid in rat colonic mucosa was inhibited by pretreating it with m e m b r a n e - p e r m e a b l e mono-functional sulfhydryl modifiers such as N-ethyl-

84 maleimide but not with several amino-acid blockers nor with diethyl maleate. Those results suggest the possible implication of membrane-associated SH proteins in the enhancing mechanisms whereby oleic acid produces the mucosal permeability change. However, we cannot exclude the possibility that those SH modifiers may directly reduce the effect of the fatty acid by decreasing membrane permeability itself, because the inhibition on the thiol reagents is accompanied by a significant reduction in the thiol level of the whole tissue. The effect of SH modification on the intestinal absorption of passively transported drugs is also still unclear. The first aim of our present work was to obtain unambiguous evidence supporting our hypothesis that the membrane-associated SH proteins are involved in the process of the mechanism(s) of the oleic acid-induced mucosal permeability enhancement. For this purpose, diamide, a bifunctional and reversible SHmodifying reagent, was used against the enhanced effect of oleic acid refering to the irreversible effect of N-ethylmale~mide on fatty acid enhancement. An in vitro perfused system of rat colon was used in this study as described previously [11]. Carboxyfluorescein, a dye poorly permeant to the intestine, was chosen as a model compound for poorly absorbable drugs [12]. The second aim was to investigate the importance of SH modifications in a passive transport process of drugs across the intestinal barrier which is governed by the lipid partition theory. In the present study, quinine, which has been reported as a passively transported substance in the intestinal absorption study [13,14], was chosen as a lipoid-soluble drug.

carboxyfluorescein solution with the aid of a nonionic sur'factant HCO-60 which has no adjuvant effect on the intestinal absorption of carboxyfluorescein. The molar ratio of oleic acid to HCO-60 is 30 : 4.

In uitro permeation experiments In vitro permeation experiments were performed according to the method reported previously [11]. Briefly, male Wistar rats (Japan SLC, Hamamatu) weighing 230-280 g were fasted for about 16 h prior to the experiments but given water ad libitum. The rats were anesthetized with sodium pentobarbital (i.p., 30 mg/kg) and killed by exsanguination from the vena cava. The isolated colonic segment was rinsed with saline, everted on a glass rod, and connected as a loop onto the perfusion apparatus (LKB Microperpex peristaltic pump, Sweden) via two silicon canulae. The everted colonic loop (approx. 4.0 cm long) was bathed in the test solution (outer compartment of 30 ml) warmed at 37°C and perfused with 0.1 M Tris-HC1buffered solution (pH 7.4) containing 10 mM glucose at the serosal side (inner compartment) in a single-pass manner. The flow rate of perfusate was constant at 0.7 ml/min. 5-min fractions of the perfusate effluent were collected from 2.5 to 92.5 rain after dosing. Carboxyfluorescein in those aliquots was assayed after suitable dilution with 0.2 M sodium carbonate buffered solution (pH 10) as described below. For pretreatment, the everted sac was immersed in 0.1 M Tris-HCl-buffered solution containing the SH-modifying reagent for 15 min at a room temperature and then was washed three times with saline prior to the perfusion experiments. Sham operation was carried out in the plain buffered solution.

Materials and Methods

Materials Oleic acid (99.9% purity) and polyoxygenated (60 M) hydrogenated caster oil (HCO-60 ®) were supplied by Nippon Oil & Fat Co. (Tokyo, Japan) and Nikko Chemicals Co. (Tokyo, Japan), respectively. N-Ethylmaleimide and quinine were purchased from Wako Pure Chemical Industry (Osaka, Japan), and diamide was obtained from Nakarai Tesque Co. (Kyoto, Japan). 5(6)-carboxyfluorescein was purchased from Eastman Kodak (Rochester, NY) and used after purification with activated charcoal and a hydrophobic gel column, Sephadex LH-20 ® (Pharmacia, Uppsala, Sweden)[15]. All other chemicals and reagents were commercial products of the highest available grade of quality. Preparation of the test solutions Carboxyfluorescein and quinine were diluted with 0.1 M Tris-HCl-buffered solution (pH 7.4) to be 0.01 w / v % and 0.1 w/v%, respectively. Oleic acid was neutralized with 1 M NaOH and dispersed into the

In situ absorption experiments Absorption experiments were also performed by using the in situ single-pass perfusion technique in th~ anesthetized rats as described our previous report [11]. The intestinal loop was prepared by connecting two silicon perfusion cannulae to the proximal and distal ends of the entire large intestine (the colon and the rectum) by ligation. Perfusion of the test solutions into the loop was performed at the constant rate of 1 ml/min using a peristaltic pump (ATTO SJ-1220 miniperista pump, Japan). In the pretreatment studies, the intestinal lumen was exposed to the buffered solution containing the SH-modifying reagent for 15 min prior to the perfusion of the test solution. Blood samples were collected via a polyethylene catheter placed into the carotid artery at the definite time intervals during the perfusion experimental period. The blood samples were immediately centrifuged at 5500 x g for 2 min to obtain the plasma fraction. The plasma samples of 50 /xl were mixed with the same volume of Triton X-100 and diluted with sodium-bicarbonate-

85 riod of 60 m i n (Qp) was determined. The Q p values were corrected for the apparent surface area of the colonic serosa and the flow rate of the perfusate effluent as described previously [11]. The concentrations of protein-bound and non-protein sulfhydryls in the colonic segments or in the brush-border membrane preparations were determined according to the Sedlak and Lindsay [17] using 5,5'-dithiobis-2-nitrobenzoic acid (DTNB) and glutathione as standard substances.

buffered solution (pH 10) for fluorometrical assay as described below. In the studies using quinine, bile samples were also collected at appropriate time intervals via a polyethylene cannula placed into the bile duct. In the plasma and bile samples quinine was extracted with benzene according to the method of Nakae et al. [13].

Preparation of brush-border membrane from rat colon Brush-border membrane vesicles were prepared using the colon of the fasted rats according to the method of Kesseler et al. [16]. The membrane preparation obtained from ten rats contained 3-5 mg protein. The alkaline phosphate activity of alkaline phosphatase, a marker enzyme for brush-border membrane, was determined using a commercial assay kit (Alkaliphospha B test Wako ®, Japan). The activity in the final preparation was more than 10-fold higher than that in the original homogenate, and no non-protein thiols were detected in the final preparations.

Results

Effect of diamide on the in vitro mucosal permeability of carboxyfluorescein in the presence of oleic acid The effluent concentration-time profiles of carboxyfluorescein after administration with or without oleic acid to the mucosa exposed to diamide is shown in Fig. 1A. A prominent increase in the concentration of carboxyfluorescein in perfusate was observed after the administration of the oleic acid micellar solution, indicating a marked increase in the mucosal permeability as described previously [11]. However, this enhanced permeability elicited by oleic acid was partly mitigated by pre-exposure of the mucosa with 0.5 mM diamide solution. Further, the inhibitory effect caused by the presence of 0.5 mM diamide was entirely restored by subsequent treatment with 0.2 mM dithiothreitol solution as shown in Fig. lB. It was also observed that the treatment with dithiothreitol shortened the lag times in

Analytical methods The fluorescences of carboxyfluorescein and quinine were determined using a Hitachi model 650-10 S spectrofluorometer (Tokyo, Japan): the excitation and the emission wavelengths were 490 nm and 520 nm for carboxyfluorescein and 350 nm and 450 nm for quinine, respectively. In order to evaluate the extent of in vitro colonic permeation, the cumulative amount of carboxyfluorescein permeated during the perfusion pe-

B.

A.

E 0.2

0.2

tllliiiiiii

®

_c

o~

0.1

0.1

T~

!i !i

u. (9

~ ..

°o

m

30

• ...

6'o T i m e (mln)

9'o o

AA AA A A A A AA A A A A A

• •

I

I

I

SO

60

90

Time (rain)

Fig. 1. Time profile of carboxyfluorescein (CF) level in the serosal side of everted perfused colonic segments. Following the administration of CF (100 ~ g / m l ) to the mucosal side with or without oleic acid, 0.1 M Tris-HCl buffered solution containing 10 m M glucose was perfused on the serosal side at a flow rate of 0.7 m l / m i n for 90 min. Oleic acid was solubilized with 1.3 m M HCO-60 to 10 mM. (A) Effect of diamide on the transmucosal permeability of CF in the presence (©) or absence (e) of oleic acid. The mucosa was pretreated with 0.5 m M diamide for 15 min at room temperature prior to CF administration. Closed square denotes the presence of oleic acid after sham operation with the medium. Statistical comparison with the control (11) was carried out using Student's t-test: * P < 0.05, (B) effect of dithiothreitol after pretreatment with diamide on the transmucosal permeability of CF in the presence ( zx) or absence ( • ) of oleic acid. The mucosa was treated with 0.5 m M diamide for 15 min at room temperature followed by treatment with 0.2 m M dithiothreitol under the same conditions prior to CF administration. Each point represents the m e a n for a group of 3 - 4 animals and the vertical bars denote the S.E.; the absence of bar indicates that the S.E. is within the size of the symbol.

86 TABLE I

Effect o f diamide on whole mucosal tissue S H contents

3

Everted colonic sacs were incubated with diamide solution (0.5 mM) or the medium (control) for 15 min at room temperature. After washing with saline the mucosa was scraped and weighed. Proteinbound and non-protein-bound SH in the tissue were determined using DTNB according to Sedlak and Lindsay [17]. Results were calculated as the concentration of protein-bound SH or non-protein SH per g wet tissue and are represented as the mean with S.E. Statistical comparisons with the control were carried out using Student's t-test and Cochran-Cox test (a), * P < 0.05; n.s., not significant.

2

Treatment

5

}.. }.. 4

(mM)

1

0

i

i

i

i

i

0

0.5

1.0

1.5

2.0

Control Diamide

Dlamlde conch. (mM)

Fig. 2. Effect of various concentrations of diamide on the transmucosal permeability of carboxyfluorescein in the presence of oleic acid. Everted colonic segments were pretreated with various concentration of diamide ( o ) or the medium (e, Control) for 15 min at a room temperature prior to CF administration. Qp values plotted on the ordinate are indices of the permeability of CF (see text). Each value represents the mean with standard error for a group of 3-6 animals. Statistical comparisons with the control were carried out using Student's t-test: * P < 0.05; ** P < 0.01.

the elevation of carboxyfluorescein concentrations to a small extent, but the maximum concentrations both in the presence and in the absence of oleic acid changed scarcely. The effect of pretreatment with different concentrations of diamide on this enhanced permeation is shown in Fig. 2. The inhibitory effect caused by diamide was observed as being of a concentration-dependent manner. The addition of diamide significantly reduced the permeation of carboxyfluorescein enhanced by oleic acid at the relatively low concentration region from 0.1 to 1.0 mM. At a high concentration of diamide (2.0 mM), a weak but insignificant inhibition was obtained. Therefore, the maximum inhibitory effect was observed around 1.0 mM diamide. The SH contents in colonic mucosa after SH modifications were measured using DTNB, and the results are shown in Table I. After incubation of the colonic segments which were treated with 0.5 m M diamide solution for 15 min, the data did not show a significant change in either the levels of protein-bound SH groups or non-protein SH in the whole colonic mucosal tissue. This result is not in agreement with the data in our previous report [11] concerning modification by various SH-alkylating agents, though these are irreversible modifiers. Since diamide is a reversible SH-oxidizing agents, it appears to be more important to know the effects of these SH modifiers on the SH contents in the apical surface m e m b r a n e rather than the SH levels in

0.5 2.0 5.0

SH concentration ( ~ m o l / g tissue) protein-bound SH

non-protein SH

10.96 + 0.65 11.77_+0.73 n . ~ . 11.58+_0.21n . s . 11.59_+0.74 n . s .

1.42 + 0.25 1.60+_0.38 "~ 1.05+0.28 n.~. 0.52+-0.05 *'~

n

6 6 6 8

the whole tissue. Hence, we examined the effect of the oxidizing agent on the m e m b r a n e SH contents. The SH contents in the colonic brush-border membrane after treatment with the thiol reagents under the same conditions as those in the isolated tissue are shown in Table II. N-Ethylmaleimide was chosen in order to modify the SH level in comparison with diamide. Chemical modifications of the m e m b r a n e vesicles with both SH modifiers resulted in a significant decrease of the m e m b r a n e protein SH groups. NEthylmaleimide more effectively reduced the SH contents than did diamide at the same concentrations. Decrease of membrane-protein SH on diamide was

TABLE II

Membrane SH-groups in rat colonic brush-border membrane L'esicles (BBMV) treated with S H reagents Vesicles were incubated with diamide (0.5 mM) or N-ethylmaleimide (0.5 mM) for 15 min at room temperature and washed twice. Membrane protein SH groups were photometrically assayed using DTNB (Ellman's reagent) and membrane protein according to Lowry et al. [18]. In order to substantiate the formation of disulfide bonds on diamide, vesicles treated with this agent were subsequently exposed to dithiothreitol (0.2 mM) under the same conditions. Data are expressed as the mean with S.E. Statistical comparisons vs. the control were carried out using the Student's t-test, n.s., not significant. BBMV treated with 1st

2nd

Membrane SH content (/zmol/ g protein)

None Diamide

none (control) none dithiothreitol

127.0+ 13.0 96.2+ 12.8 124.9+23.0

5 4 4

0.01 n.s.

100 76 98

72.2+8.66

3

0.01

57

N-Ethyl maleimide

none

n

P

Remaining (%)

87

Effect of SH modification on the in situ large intestinal absorption of carboxyfluorescein in the presence of oleic acid

3.0

The effect of the pretreatment with diamide on the absorption-enhancing effect of oleic acid in the in situ situation is shown in Fig. 3, together with that of N-ethylmaleimide. The treatment of the mucosa with diamide as well as N-ethylmaleimide was found to suppress the increase in the plasma level of carboxyfluorescein promoted by oleic acid over the period for the in situ perfusion, being close to the inhibitory effect of N-ethylmaleimide. The suppressive effect on diamide was comparable to that of N-ethylmaleimide, particularly within initial 60 min, though not statistically significant.

I ,,N

~. 2.0 c

~5 ~

1.0

O

°o

• 30

9'o

60 Time

(rain)

Fig. 3. Plasma concentration-time curves during the in situ single-pass perfusion of carboxyfluorescein in the presence of oleic acid. Effect of pretreatment of the mucosa with 0.1 m M diamide (e); 0.l m M N-ethylmaleimide (A); sham operation with the medium (©, Control) for 15 min, followed by washing three times with 15 ml warmed saline prior to CF administration. Perfusion of CF alone without any pretreatment (t,). Each point represents the mean with S.E. for a group of 3 - 4 animals. Statistical comparisons with the control were carried out using the Student's t-test: * P < 0.05.

Effect of SH modifications on the in situ absorption of quinine The mucosal SH modifications performed in this study might affect the in situ permeation of lipoidal drugs of which absorption is governed by a passive transport mechanism. We examined the effects of diamide and N-ethylmaleimide on the colonic absorption of quinine, a typical basic drug known as being a passively absorbed compound [14]. The quinine levels in the plasma and in the bile versus time curves during the in situ mucosal perfusion of quinine after pretreatment of the mucosa with diamide and N-ethylmaleimide are shown in Figs. 4 and 5, respectively. Administration of quinine alone showed a high plasma concentration and a high recovery into the bile, being consistent with the report of Nakae et al. [13]. The results by pretreatments with diamide and N-ethylmaleimide indicated no apparent inhibitory ef-

approximately restored by the addition of dithiothreitoh These observations suggest that the SH content in brush-border membrane reversibly rose and fell and that their changes fairly correspond to the permeability increase a n d / o r decrease by oleic acid. Therefore, these results strongly suggest that certain membrane SH proteins may regulate the permeability change of the epithelial cell sheet produced by oleic acid.

3O A.

B.

A

E 1.5 t~

T

I

I

-

t~

~. 1.o

.E c

g



-g

10

~

c

o 0.5 Q) c

2

3 0 =r

0

0

!

I

80

3O Time

(mln)

9'0

o

4'

!

!

ao

80 Time

9'o

(min)

Fig. 4. Plasma (A) and bile (B) concentration-time curves during the in situ single-pass perfusion of quinine. Effect of pretreatment with 0.1 m M diamide (closed); s h a m operation with the m e d i u m (open) for 15 min, followed by washing three times with 15 ml warmed saline prior to quinine administration. Each point represents the m e a n with S.E. for a group of three animals.

88

3O

A.

A

B.

~ 1.5

~. 1.0

I °

.S C

T

o

~ 10

0.5

C

O

C

O 0

3'0

I

~o Time (mJn)

,'o

O'

o

~'o

6'o

9'o

Time (rain)

Fig. 5. Plasma (A) and bile (B) concentration-time curves during the in situ single-pass perfusion of quinine. Effect of pretreatment with 0.1 mM N-ethylmaleimide (closed); sham operation with the medium (open) for 15 min, followed by washing three times with 15 ml warmed saline prior to quinine administration. Each point represents the mean with S.E. for a group of three animals.

fect on the quinine levels neither in the plasma nor in the bile, although the latter tended to reduce the bile level of quinine slightly but not significantly. Discussion

Our previous study using monofunctional SH-modifying reagents [11] clearly showed that membrane-permeable SH blockers (e.g., N-ethylmaleimide and HgC12) effectively protect the permeability change of rat colonic mucosa elicited with oleic acid, whereas poor permeable SH blockers (e.g., PCMPS and DTNB) and diethyl maleate (a specific depleting agent of nonprotein thiols) were not effective. These results suggest that there might exist the key SH proteins responsible for the increase of permeability elicited with oleic acid within or inside of the brush-border membrane. In the present study, diamide has been mainly used as a membrane-permeable SH reagent, relating to cross-link membrane SH groups [19], on the further study of oleic acid-induced intestinal absorption enhancement. In both in vitro and in situ studies, diamide obviously inhibited the enhanced permeability of colonic mucosa induced with oleic acid (Figs. 1A and 3). This inhibitory effect of diamide was dependent of its concentration, and it was observed that the optimal effective concentration of diamide was at 1 mM (Fig. 2). Both the protein-bound and non-protein-bound SH levels in the whole mucosal tissue failed to account for the inhibitory effect of diamide and its concentrationdependence (Table I). It has been reported that relatively higher concentration of the oxidizing-reagents including diamide renders the plasma membrane leaky [20]. It was also reported by Nishihata et al. that non-protein thiol loss in the tissue causes the mueosal permeation of water-soluble compounds in rat jejunum and rectum [21]. From these data concerning SH modi-

fication, the observed concentration dependence of the effect of diamide obtained here may account for these factors. Furthermore, it is noted that the inhibitory effect of diamide was completely diminished after a subsequent treatment with dithiothreitol (Fig. 1B). This may imply the restoration of the SH group by the addition of dithiothreitol, and strongly suggests that the contribution of the SH group is concerned with oleic acid-permeability enhancement. On the contrary, the change in the protein-bound SH content in the brush-border membrane (Table II), but not in the whole tissue SH content (Table I), apparently corresponds to the effects of the thiol reagents. These results suggest that certain SH proteins in the apical surface membrane of the epithelial cells may play an important role in the change in the mucosal permeability induced by oleic acid, although localization and characterization of those SH proteins are unclear in the present study. From the previous study on macromolecule transport using different sizes of dextrans [22], it has been suggested that the enhanced mucosal permeation of the macromolecules occurred via a paracellular route as well as a transcellular route in the presence of the unsaturated fatty acid. It is known that SH blocking agents inhibit the clustering of membrane proteins and have an effect on cytoskeletal contractile proteins [23]. In addition, Madara and co-workers indicated that these contractile proteins are connected with a tightjunctional apparatus which forms a crucial permeation barrier to polar compounds via a paracellular route and may regulate its structure and permeability [24,25]. These documentations allow us to infer that the key proteins with functional SH groups which play an important role in the regulation of the transmucosal permeability of water-soluble drugs, at least, via the paracellular route, might be cytoskeletal proteins or

89 tight-junction connected proteins. However, more extensive studies are required to clarify this point. On the other hand, SH modifications by diamide and N-ethylmaleimide did not entirely affect the colonic absorption of quinine in this study (Fig. 4 and 5). Therefore, the transport process regulated by SH modification in the intestinal mucosa is conceivably unrelated to the lipid-partition-transported mechanism by which many lipoid-soluble drugs can be absorbed from the gastrointestinal tract. In summary, the present results using diamide and dithiothreitol demonstrate that certain membrane-associated proteins with free SH-groups are involved in the mechanism of enhancement of the transmucosal permeability of the colon with oleic acid. Our previous results obtained using irreversible monofunctional SH blockers are supported by these results. SH modifications have no pronounced effects on the transport across the mucosal membrane by passive diffusion. The effect of SH modifications on the mucosal permeation change elucidated with oleic acid is conceivably largely ascribed to their inhibition of the interaction between oleic acid and membrane-protein fraction rather than the interaction with lipid bilayer. The findings of reversible effect on diamide and dithiothreitol in this study may suggest that the enhancement with oleic acid occurred through functional SH groups in certain proteins existing in brush-border membrane. Further studies are necessary to clear this point.

Acknowledgement The authors wish to thank Professor Dr. T. Fujii and Dr. A. Tamura for their helpful advice and discussions.

References 1 Muranishi, S. (1985) Pharm. Res. 2, 108-118. 2 Muranishi, S. (1990) Crit. Rev. Therap. Drug Carrier Systems 7, 1-33.

3 Lee, V., Yamamoto, A. and Kompella, U.B. (1991) Crit. Rev. Therap. Drug Carrier Systems 8, 91-192. 4 Muranishi, S., Muranushi, N. and Sezaki, H. (1979) Int. J. Pharm. 2, 101-111. 5 Muranishi, S., Tokunaga, Y., Taniguchi, K. and Sezaki, H. (1977) Chem. Pharm. Bull. 25, 1159-1161. 6 Yoshikawa, H., Takada, K., Muranishi, S., Satoh, Y. and Naruse, N. (1984) J. Pharmacobio.-Dyn. 7, 59-62. 7 Muranushi, N., Kinugawa, M., Nakajima, Y., Muranishi, S. and Sezaki, H. (1980) Int. J. Pharm. 4, 271-279. 8 Fukui, H., Murakami, M., Yoshikawa, H., Takada, K. and Muranishi, S. (1987) J. Pharmacobio.-Dyn. 10, 236-242. 9 Muranishi, S. (1989) in Novel Drug Delivery and its Therapeutic Application (Prescott, L.F. and Nimmo, W.S., eds.), pp. 69-77, John Wiley & Sons, New York. 10 Muranushi, N., Takagi, N., Muranishi, S. and Sezaki, H. (1981) Chem. Phys. Lipids 28, 269-279. 11 Murakami, M., Takada, K., Fujii, T. and Muranishi, S. (1988) Biochim. Biophys. Acta 939, 238-246. 12 Hashida, N., Murakami, M., Yoshikawa, H., Takada, K. and Muranishi, S. (1984) J. Pharmacobio-Dyn. 7, 195-203. 13 Nakae, H., Sakata, R. and Muranishi, S. (1976) Chem. Pharm. Bull. 24, 886-893. 14 Hogben, C.A.M., Tocco, D.J., Brodie, B.B. and Shanker, L.S. (1959) J. Pharmacol. Exp. Ther. 125, 275-282. 15 Ralston, E., Hjelmeland, L.M., Klausner, R.D., Weinstein, J.N. and Blumenthal, R.B. (1981) Biochim. Biophys. Acta 649, 133137. 16 Kessler, M., Acuto, O., Storelli, C., Muter, H., Miiller, M. and Semenza, G. (1978) Biochim. Biophys. Acta 506, 136-154. 17 Sedlak, J. and Lindsay, R.H. (1968) Anal. Biochem. 25, 192-205. 18 Bensadoun, A. and Weinstein, D. (1976) Anal. Biochem. 25, 192-205. 19 Kosower, N.S., Kosower, E.M. and Wertheim, B. (1969) Biochem. Biophys. Res. Commun. 37, 593-596. 20 Haest, C.W.M., Kamp, D., Plasa, G. and Deuticke, B. (1977) Biochim. Biophys. Acta 469, 226-230. 21 Nishihata, T., Nghiem, B.T., Yoshitomi, H., Lee, C.S., Dillsaver, M., Higuchi, T., Choh, R., Suzuka, T., Furuya, A. and Kamada, A. (1986) Pharm. Res. 3, 345-351. 22 Masuda, Y., Yoshikawa, H., Takada, K. and Muranishi, S. (1986) J. Pharmacobio-Dyn. 9, 793-798. 23 Heller, K.B., Poser, B., Haest, C.W.M. and Deuticke, B. (1984) Biochim. Biophys. Acta 777, 107-116. 24 Madara, J.L. (1987) Am. J. Physiol. 253, C171-C175. 25 Madara, J.L., Moore, R. and Carlson, S. (1987) Am. Physiol. Soc. 253, C854-C861.