- Email: [email protected]
Acid activation of omeprazole in isolated gastric vesicles, oxyntic cells, and gastric glands
GASTROENTEROLOGY
1989;98:1453-81
Acid Activation of Omeprazole in Isolated Gastric Vesicles, Oxyntic Cells, and Gastric Glands MAGOTOSHI MORII, HIDEYUKI TAKATA, NORIAKI TAKEGUCHI Faculty of Pharmaceutical
and
SGences, Toyama Medical and Pharmaceutical
University,
Toyama,
Japan
Omeprazole, a potent inhibitor of gastric hydrogen ion transporting, potassium-stimulated adenosine triphosphatase, was found to be transformed into an SH-reactive strong fluorescent molecule (excitation and emission wavelengths of 370 and 560 nm, respectively) in an acidic medium. The addition of glutathioneor protein-containing sulfhydryl groups such as pepsin to the medium decreased the fluorescence. Also, the increase in the pH of the medium decreased the fluorescence. The fluorescent molecule was identified to be an acid-activated planar cyclic sulfenamide derivative of omeprazole. The transformation was studied in H+-preaccumulated hog gastric vesicles, which contain the hydrogen ion transporting, potassium-stimulated adenosine triphosphatase. The addition of omeprazole to the vesicle suspension induced a rapid increase in the fluorescence intensity, indicating that omeprazole was activated in the intravesicular space. Then, the intensity biphasically decreased with time. The slower small decrease was due to the reaction of the sulfenamide with sulfhydryl group(s) located on the acid secretory side of the hydrogen ion transporting, potassium-stimulated adenosine triphosphatase. Omeprazole was also activated in the acidic lumina of isolated rabbit gastric glands that were stimulated with histamine. Furthermore, direct evidence was obtained from the imaging of the fluorescence that omeprazole was activated in the acidic compartments of the isolated Xenopus oxyntic cell.
0 tive, secretion
meprazole, a substituted benzimidazole derivahas been shown to inhibit gastric acid in vivo and in vitro [l-5), proton accumulation in isolated parietal cells, gastric glands, and vesicles (6-11). Omeprazole does not inhibit hydrogen ion transporting, potassium-stimulated adeno-
sine triphosphatase (H+,K+-ATPase) activity at neutral pH, but upon treatment with acid it inhibits adenosine triphosphate (ATP) hydrolysis and proton transport (12-16). Studies on the chemistry of omeprazole in an acidic solution indicated that the cyclic sulfenamide derivative of omeprazole {a mixture of two isomers: 2,4-dimethyl-3,9and 2,4-dimethyl-3,10-dimethoxy-5H-pyrido[1’,2’:4,5][1,2,3]thiadiazino[2,3-a]benzimidazol-l3-ium chloride} and sulfenic acid derivative {3,5-dimethyl-4-methoxy-2(hydroxy-thiomethyl) - 1- (5 -methoxy- &benzimidazolyl)-pyridinium chloride} are very reactive with sulfhydryl compounds such as glutathione and 2mercaptoethanol (17,181. It has been suggested that the acid-activated compounds are involved in inactivation of the H+,K+-ATPase. However, direct experimental evidence that these compounds are active forms of omeprazole in the secretory canaliculi of parietal cells or the intravesicular medium of gastric vesicles has not yet been shown. Although the sulfenamide derivative of omeprazole has a characteristic absorbance at 370 nm (18), it is impossible to follow the activating process of omeprazole in gastric vesicles and glands from measurements of absorbance of the vesicle or gland suspension because of the high turbidity of the suspension and changes in the intensity of light scattering induced by change in the vesicle volume (19). In this study, we found that one of the active forms of omeprazole, the sulfenamide, has a strong fluorescence and that the fluorescence disappears as it reacts with sulfhydryl compounds. The fluorescence
Abbreviations used in this paper: Em, emission wavelength; Ex, excitation wavelength: H+,K+-ATPase, hydrogen ion transporting, potassium-stimulated adenosine triphosphatase. 0 1989 by the American Gastroenterological Association 0018-5085/89/$3.50
GASTROENTEROLOGYVol.96,No.6
1454 MORII ET AL.
change could be monitored at a high accuracy even in a highly turbid suspension such as gastric vesicle, oxyntic
cells,
and gastric
glands.
These
properties
total cell number tion.
as determined
by microscopic
observa-
of
the sulfenamide enabled us to follow the activation process of omeprazole in isolated gastric vesicles, oxyntic cells, and gastric glands and the sulfhydryl modification process in the gastric vesicles.
Measurement Acid-Activated
of Fluorescence Omeprazole
of
Omeprazole (10 PM) was incubated with 0.1 N HCl at 25°C. The absorbance of activated omeprazole (OD,,,) was measured by an Aminco DW-2C UV-VIS spectrophotometer. The fluorescence [excitation wavelength (Ex) = Materials and Methods 370 nm, emission wavelength (Em) = 560 nm] was meaChemicals sured with a Hitachi 650-10s fluorescence spectrophotometer equipped with a magnetic stirrer under the same Omeprazole was obtained from Fujisawa-Astra Co. conditions as absorbance measurements. The width of slit Ltd. (Osaka, Japan). Timoprazole, picoprazole, and 2-[2-(3, for excitation light in the fluorescence measurements was 5-dimethyl-4-methoxy)-pyridylmethylsulfiny1]-(5-methoxyfixed at 1 nm to minimize photobleaching. carbonyl-6-methyl)-benzimidazole (H compound) were synthesized in our laboratory (20). The tetrafluoroborate salt of the acid-activated compound of omeprazole, a cyclic Measurements of Proton Uptake and sulfenamide derivative of omeprazole, was synthesized as Fluorescence in Gastric Vesicles described elsewhere (17). Ranitidine was obtained from Proton uptake into gastric vesicles was measured Sankyo Co. Ltd. (Osaka, Japan). Pepsin from porcine stomwith a pH electrode (21,24). Gastric vesicles (100 pg/ml) ach mucosa with a digestive power of 1: 10,000 was were incubated in a solution containing 150 mM KCl, 5 obtained from Wako Pure Chemicals (Osaka, Japan). mM glycylglycine (pH 6.11), 0.2 mM MgCl,, 5 pg/ml of SCH28080 was obtained from Schering Corp. All other valinomycin, 1 mM glutathione, and 0.3 mM magnesium chemicals were of the highest purity available. salt of adenosine triphosphate (Mg-ATP) with or without 10 PM omeprazole at 25°C. The time-dependent increase in pH of the suspension was measured by a pH-meter Preparation of Hog Gastric Vesicles (Radiometer, PHM64) and recorded by a strip chart reMembrane vesicles containing the Hf,Kf-ATPase corder. The fluorescence intensity of the activated omeprain 0.25 M sucrose were prepared from hog stomachs as zole was also measured in the same incubation mixture described previously (21). Protein concentration was de(Ex = 370 nm, Em = 560 nm). At least 10 PM omeprazole termined by the method of Lowry et al. with bovine serum was required for high signal-to-noise measurements of albumin as standard (22). fluorescence. Preparation
of Rabbit
Gastric
Glands
The isolated gastric glands were prepared from rabbit gastric mucosa as described previously (23). They were suspended in a solution containing 132.4 mM NaCl, 5.4 mM KCl, 5.0 mM Na,PO,, 1.0 mM NaHPO,, 1.2 mM MgSO,, 1.0 mM CaCl,, 10 mM HEPES, 2 mg/ml of glucose, and 0.4 mg/ml of bovine serum albumin (pH 7.4). The isolated gastric glands were used within 3 h after preparation.
Preparation
of Xenopus
Oxyntic
Measurement of Fluorescence Gastric Glands
in Isolated
The isolated gastric glands suspended in HEPESNaOH buffer solution (pH 7.4) described above were preincubated with 100 FM histamine at 37°C for 15 min. Then 100 PM omeprazole at a final concentration was added, and the subsequent changes in the fluorescence intensity (Ex = 370 nm, Em = 560 nm) were measured. At least 100 PM omeprazole was required for high signal to noise measurements.
Cells
The isolated oxyntic cells were prepared from gastric mucosa of Xenopus laevis. Gastric mucosa was minced with small scissors in a buffer solution containing 87 mM NaCl, 4 mM KCl, 18 mM NaHCO,, 1 mM KH,PO,, 1 mM MgCl,, 2 mM CaCl,, and 2 mg/ml of sucrose (pH 7.4) and then filtered through a nylon mesh (20 mesh). The filtrate was centrifuged at 1000 rpm for 5 min with a Hitachi 05PR-22 refrigerated centrifuge. The sedimented fraction was washed 3-5 times with the buffer solution and suspended in the same buffer solution. The content of oxyntic cells in the cell preparation was about 50% of the
Imaging of the Distribution of Activated Omeprazole in Isolated Xenopus Oxyntic Cells The isolated oxyntic cells of Xenopus laevis were fixed on polylysine-coated slide glass in the buffer solution (pH 7.4) described above under an inverted fluorescence microscope (Nikon TMD-EFQ). Acid secretion was stimulated by preincubating with 1 mM dibutyryl-cyclic adenosine monophosphate at 25°C for 30 min. Then 100 PM omeprazole was added. Time-dependent changes of the microscopic image of fluorescence were analyzed with a
June
1989
ACID ACTIVATION OF OMEPRAZOLE
Spex Fluorolog-2 spectrofluorometer and an IMl imaging system. The excitation wavelength was 370 nm. Emitted light was passed through a dichroic mirror (cutoff wavelength = 510 nm) and a high-pass filter (cutoff wavelength = 520 nm), and the image was saved in a SIT-camera. The maximum resolution of image was 512 by 478 pixels by a-bit. Autofluorescence of cells was subtracted from the fluorescence image. The gray scale of fluorescence intensity was digitally enhanced. Pseudocoloring images of the relative fluorescence scale of activated omeprazole were hard-copied.
Ex370
1455
Em560
Kesults Acid Activation
of Omeprazole
Omeprazole in acidic solutions reacts with specific: sulfhydryl group(s) located on the intravesicular surface of the H+,K+-ATPase, resulting in inhibition of ATP hydrolysis and proton transport activity (15,16). Several degradation products are generated by the acid treatment (18,251. A cyclic sulfenamide derivative may be the main reactive degradation product in low pH because another reactive sulfenic acid is unstable in low pH and converts to the sulfenamide (17,18). The planarity of four rings of the sulfenamide derivative leads to a bathochromic shift of the long-wavelength ultraviolet absorption from 301 to 370 nm (18). Curve A in Figure 1 shows the time-dependent increase in absorbance at 370 nm induced by generation of the
=
0
Time. min
Figure
‘1. Time-dependent changes of absorbance (A) and fluorescence (B) of omeprazole activated by HCl. The methanolic solution of free base of omeprazole was added in 0.1 N HCl solution at time zero. The final concentration of omeprazole was 10 PM and that of methanol was -0.1%. A. The optical density was measured at 370 nm with the narrowest slit-width to avoid degradation of activated product(s) by the incident light. At the arrow, glutathione was added to a final concentration of 100 pM. B. The fluorescence intensity was measured at Ex = 370 nm and Em = 560 nm with the narrowest excitation slit-width. Glutathione was added to a final concentration of 100 PM.
300
400
500
Wavelength,
600
700
nm
Figure 2. Excitation and emission spectra of acid-activated omeprazole. The methanolic solution of free base of omeprazole was dissolved in 0.1 N HCI at a final concentration of 10 KM. About 20 min later, the excitation and emission spectra were measured.
sulfenamide derivative from 10 PM omeprazole in 0.1 N HCl. The rate of increase did not depend on the pH of the medium (pH 1 to pH 4). The reaction of the sulfenamide with 100 PM glutathione resulted in a decrease in absorbance at 370 nm. The reaction rate with glutathione increased as the pH of the medium was increased in the range from pH 1 to pH 4 (data not shown). The absorbance decreased instantly when the pH of the medium was neutralized by the addition of NaOH (data not shown). These results indicate that the sulfenamide was transformed to nonplanar derivatives by reaction with the sulfhydry1 group or increasing the pH of the medium. We found that the acid-treated omeprazole shows strong fluorescence in acidic medium. Figure 2 shows that the peak of the excitation wavelength is 370 nm and the peak of the emission wavelength is 570 nm for the acid-treated omeprazole in 0.1 N HCl. Curve B in Figure 1 shows the time-dependent change of fluorescence under the same conditions as the absorbance measurement. The similarity between the time-course of absorbance and of fluorescence, together with the identity between the absorbance wavelength and the excitation wavelength (370 nm), indicates that the cyclic sulfenamide derivative of omeprazole is a fluorescent product of acid-acti-
1456
GASTROENTEROLOGY
MORE ET AL.
0
I
I
I
I
10
20
30
40
Time.
Figure
mln
3. Reaction of the activated omeprazole with pepsin. The methanolic solution of free base of omeprazole was dissolved in 0.1 N HCl at a final concentration of 10 PM. The fluorescence intensity was measured at Ex = 370 nm and Ex = 560 nm. Pepsin was added to a final concentration of 2.5 mgiml at the arrow.
vated omeprazole. Furthermore, we synthesized tetrafluoroborate salt of the cyclic sulfenamide and found that the excitation and emission wavelengths in dimethylsulfoxide were 380 and 550 nm, respectively, which indicates that the fluorescent molecule in the acidic condition is the cyclic sulfenamide derivative of omeprazole. The fluorescence was gradually bleached when the intensity of the excitation light was strong. To circumvent this problem, the width of the slit for the excitation light was set at 1 nm and solutions were continuously stirred during the measurements. Similar acid-activated products of other substituted benzimidazoles, timoprazole, picoprazole, and H compound, were found not to be fluorescent molecules. However, several similar substituted benzimidazole derivatives that have a methoxy group at the 5-position in the benzimidazole ring, as in the case of omeprazole, produced fluorescent molecules by acid treatment. The fluorescence property must be due to the presence of a resonance form of four coplanar rings dependent on electron donation by the methoxy group in the benzimidazole ring. The sulfenamide also reacted with protein-containing sulfhydryl groups such as pepsin at pH 1 (Figure 3). This indicates the possibility that omeprazole modulates the peptidase activity of pepsin in vivo. Inhibition of Proton Uptake in Hog Gastric Vesicles by Omeprazole The inhibitory effects of omeprazole on proton uptake were investigated by measuring alkaliza-
Vol. 96, No. 6
tion of the extravesicular medium using a pH electrode. We could detect the small change in the pH with high sensitivity in the presence of 5 mM glycylglycine, which showed a weak pH buffering action. In a separate experiment, it was found that proton generation induced by ATP hydrolysis was negligibly small at pH 6.11, and hence the observed changes in the extravesicular pH were due to proton uptake into the vesicles. Figure 4 shows the inhibitory effect of 10 PM omeprazole on proton uptake (100 pg of protein per milliliter). To initiate the proton uptake, the vesicles were added to the reaction medium containing Mg-ATP with or without 10 PM omeprazole. In this experiment, we added 1 mM glutathione to the solution. Activated omeprazole in the extravesicular spaces at pH 6.11 was instantly quenched by glutathione. Therefore, we could measure the fluorescence only in the intravesicular space. In the control experiment (curve B), the net proton uptake rate became zero at 10 min after the addition of Mg-ATP as the rate of the proton leak became equal to that of proton uptake when the proton gradient between intravesicular and extravesicular medium became large. Omeprazole (pK, 4) is membrane permeable and can rapidly accumulate in an acidic space. However, during the first 3.5 min of the proton uptake reaction in the presence of omeprazole (curve A), no inhibition was observed and proton uptake was even slightly enhanced, indicat-
I-
I,
0
I
,.I
1,.
,
,
10
5
,
,
,
,
,
15
Time, min Figure
4. Inhibition of proton uptake of gastric vesicles by omeprazole. The reaction solution contained 150 mM KCl, 0.2 mM M&l,, 1 mM glutathione, 0.3 mM Mg-ATP, 5 &ml valinomycin, and 5 mM glycylglycine (pH 6.11) with (A) or without (B) 10 PM omeprazole. Vesicles in 250 mM sucrose were added to the reaction solution to a final concentration of 100 pg protein per milliliter at the arrow. The proton uptake was measured by a pH-meter as an alkalization of extravesicular medium.
June 198EI
Figure
5. Activation of omeprazole in the intravesicular space of gastric vesicles. The reaction mixture contained gastric vesicles (100 wg of protein per milliliter), 150 mM KCl, 0.2 mM MgCl,, 1 mM glutathione, 5 mM glycylglycine (the pH of medium was preadjusted to a value so as to shift the pH to 6.11 when Mg-ATP was added), and 5 pgiml valinomycin with (A] or without (B) omeprazole. Magnesium-stimulated adenosine triphosphate was added to a final concentration of 0.3 mM at the arrow. Fluorescence was measured with excitation and emission wavelengths at 370 nm and 560 nm. respectively.
ing that omeprazole was not activated in the initial phase. This may be due to the pH buffering activity in lipid vesicles (26,271. Omeprazole started to reduce the rate of proton uptake at about 3.5 min and the rate of proton leak exceeded that of proton uptake after about 6 min. Figure 5 shows the changes in the fluorescence intensity (Ex = 370 nm, Em = 560 nm) during the proton uptake reaction, which are similar to those shown in Figure 4. There were no changes in the fluorescence in the absence of Mg-ATP, indicating that the pH of the intravesicular medium (initially 250 mM sucrose, pH 7) was not low enough for acid activation of omeprazole in the short incubation time with glycylglycine buffer (pH 6.11). Magnesium-stimulated adenosine triphosphate was added at zero time. The control experiment performed in the absence of omeprazole (curve B) shows that the fluorescence intensity decreases during the initial proton uptake reaction. This decrease was due to interference by the light scattering changes caused by change in the vesicle volume. However, at present, we have no clear explanation for the decrease in light scattering. In the presence of omeprazole (curve A), the intensity of fluorescence transiently increased with a time lag of 3.5 min. This transient increase in the fluorescence intensity is due to the formation of the cyclic sulfenamide de-
ACID ACTIVATION OF OMEPRAZOLE
1457
rivative of omeprazole in intravesicular space. The time lag of activation may be due to a strong pH buffering action of the intravesicular medium and phospholipids of membranes (26,271. The acidity of the intravesicular medium may not be high enough to activate omeprazole during the lag phase. To clarify the reason for the later decrease of fluorescence in curve A in Figure 5, we carried out the experiment shown in Figure 6. In this experiment, omeprazole was added after the intravesicular acidification had been established (10 min after the addition of 0.3 mM Mg-ATP). As the sulfhydryl groups of the enzyme on the extravesicular side of membrane were protected by the presence of an excess amount of glutathione (1 mM) in the solution, only the activated omeprazole in the intravesicular medium could react with sulfhydryl groups of H+,K+-ATPase. The proton began to leak within 1 min after the addition of omeprazole. The fluorescence change, which reflects change in the amount of the sulfenamide in the intravesicular medium, was complicated. After the immediate increase in the fluorescence upon addition of omeprazole, the fluorescence decreased biphasically (curve A). The biphasic decrease of the fluorescence may be due to two factors: (a) transformation of the excess amount of the sulfenamide into a nonfluorescent molecule such as sulfenic acid, caused by an increase in the intravesicular pH, and (b) the reaction with sulfhydry1 group(s) located on the acid secretory side of I
.E
%
15
Figure 6. Activation of omeprazole and inhibition of proton uptake in the internal medium of the acid preaccumulated gastric vesicles. The reaction mixture contained gastric vesicles (100 pg of protein per milliliter], 150 mM KCl, 0.2 mM MgCl,, 1 mM glutathione, 5 pglml valinomycin, and 5 mM glycylglycine (pH 6.11). Proton uptake was started by addition of Mg-ATP to a final concentration of 0.3 mM. About 10 min later, 10 WM omeprazole (A and C) or 10 PM omeprazole plus 100 PM timoprazole (B) was added at the arrows. The changes in fluorescence intensity [A and B) and the proton leak (C) from gastric vesicles were measured as described in Materials and Methods.
1458
GASTROENTEROLOGY Vol. 96, No. 6
MORII ET AL.
H+,K+-ATPase. As omeprazole was added after the intravesicular acidification had been established in this experiment, formation of the activated omeprazole is rapid and a partial inactivation of enzyme would occur. This loss of activity caused the proton leak as shown in curve C in Figure 6. This proton leak induced an increase in the intravesicular pH, resulting in a decrease in the fluorescence. Thus, the faster decreasing phase would be due to transformation of the fluorescent sulfenamide into a nonfluorescent compound (17,18).In the experiments shown in curve B in Figure 6, 100 PM timoprazole was added together with 10 PM omeprazole. Timoprazole is a substituted benzimidazole derivative and its acid-activated sulfenamide inhibits H+,K+ATPase activity. However, as described above, the sulfenamide derivative of timoprazole is not a fluorescent molecule. As shown in curve B, the fluorescence of the activated omeprazole initially increased then monophasically decreased. The disappearance of the slower decreasing phase as compared with the case of curve A, indicates that most of the activated omeprazole could not react with sulfhydryl groups because the concentration of timoprazole was 10 times higher than that of omeprazole. That is, the slower decreasing phase of fluorescence in curve A would be due to the modification of the remaining sulfhydryl groups by the sulfenamide. Measurement of Fluorescence Rabbit Gastric Glands
in Isolated
The activating process of omeprazole in histamine-stimulated gastric glands was observed. The fluorescence intensity increased and then decreased (curve A in Figure 7).A potent histamine H,-receptor antagonist, ranitidine (10 PM), in the presence of histamine inhibited the rapid increase of the fluorescence (curve B in Figure 7).Therefore, the increase shown in curve A is due to activation of omeprazole. Ranitidine does not inhibit the basal acid secretion. Thus, the basal acid secretion slowly activated omeprazole (curve B). Another type of H+,K+-ATPase inhibitor SCH28080 competes with K+ at the high-affinity K+ binding site of H+,K+ATPase. The addition of SCH28080 results in inhibition of both stimulated and basal acid secretion. The addition of 2 PM SCH28080 together with 100 /.LM histamine completely inhibited the increase in the fluorescence intensity (data not shown). These results indicate that closed acidic canaliculi exist in histamine-stimulated gastric glands and that omeprazole accumulated in acidic canaliculi was activated by acid.
0
10
5
15
Time, min Figure 7. Activation of omeprazole in acidic canaliculi of isolated gastric glands. Isolated gastric glands suspended in HEPES-NaOH buffer (pH 7.4, detail in Materials and Methods] were stimulated by 100 FM histamine (A) or 100 PM histamine plus 10 FM ranitidine (B) at 37°C for 15 min. Then 100 FM omeprazole was added at the arrow, and the change in the fluorescence intensity (Ex = 370 nm, Em = 560 nm) was measured.
Imaging of the Distribution of Activated Omeprazole in Single Xenopus Oxyntic Cells In this experiment, we used amphibian oxyntic cells as they can be stimulated with dibutyrylcyclic adenosine monophosphate at room temperature. The stimulation induced a morphologic change of the cell shape from a round to an irregular shape. Time-dependent changes in pseudocoloring images of the two-dimensional distribution of activated omeprazole in an oxyntic cell of Xenopus are shown in Figure 8. Activated omeprazole was localized only in the oxyntic cells, and the fluorescence in parts of the oxyntic cell gradually increased. These results indicate that closed acidic compartments were formed in dibutyryl-cyclic adenosine monophosphate-stimulated oxyntic cells and that omeprazole was activated only in the closed acidic compartments. The increasing rate of fluorescence was slower than that in rabbit gastric glands, probably because of the slow proton accumulation by oxyntic cells at decreased temperature (25T).Also, we could detect localized acidic compartments within oxyntic cell of Xenopus by fluorescence imaging of acridine orange distribution (data not shown).
Discussion We found that acid treatment of omeprazole produced an SH-reactive strong fluorescent molecule and that the fluorescent molecule converted to a nonfluorescent molecule when reacted with the sulfhydryl group. The identity of the excitation wave-
June 1989
ACID ACTIVATION OF OMEPRAZOLE
1459
Figure 8:. Imaging of the distribution of activated omeprazole in an oxyntic cell. Isolated oxyntic cells of Xenopus laevis were stimulated with 1 mM dibutyryl-cyclic adenosine monophosphate at 25°C for 30 min. Then 100 PM omeprazole was added. Microscopic images of the fluorescence were saved in a SIT-camera and analyzed by a Spex imaging system [detail in Materials and Methods). 1, Image of an oxyntic cell by visible light. 2-4, Fluorescence distribution images at various times after the addition of omeprazole: 2, 260 s; 3, 320 s; and 4. 400 s. The bar in image 1 indicates 10 pm. The colored strip indicates the scale of relative concentration of activated omeprazole.
length and the absorbance wavelength indicates that the fluorescent molecule is a planar cyclic sulfenamide derivative of omeprazole. Furthermore, the synthesized tetrafluoroborate salt of the cyclic sulfenamide has approximately the same excitation and emission wavelengths as those of the activated compound of omeprazole. These results indicate that the
same activated omeprazole as that produced by HCl treatment in a test tube was generated in the intravesicular medium of hog gastric vesicle. The time lag of activation (curve A in Figure 5) indicates that the pH buffering action of intravesicular medium and phospholipids (26,27) is strong enough to neutralize the pH in the initial phase of proton uptake. When
1460
GASTROENTEROLOGY Vol. 96, No. 6
MORII ET AL.
omeprazole was added to the H+-preaccumulated vesicle suspension, omeprazole was activated quickly. The sulfenamide also reacted with sulfhydryl group(s) of pepsin (Figure 3). Activated omeprazole in acidic gastric juice may react with sulfhydryl group(s) of pepsin in vivo. Although pepsin is not a so-called SH-enzyme, the modification of the sulfhydry1 group modulates pepsin activity (28). Thus, sulfhydryl modification of pepsin may be part of the antiulcer effect of omeprazole in vivo (29). In spite of strong autofluorescence of the gastric glands themselves (3O), the fluorescence of acidactivated omeprazole was measurable in histaminestimulated isolated gastric glands. The increasing and decreasing phases of the fluorescence reveal that omeprazole was activated in gastric glands. Preincubation of glands with antisecretory reagents, ranitidine or SCH28080, inhibited the increase of fluorescence, indicating that the acid accumulation is necessary for activation of omeprazole. Moreover, Im et al. (12) reported that the potency of omeprazole to inhibit rat H+,K+-ATPase activity in vivo was enhanced by pretreatment with carbachol and weakened by pretreatment with cimetidine. In gastric glands obtained by treatment with collagenase, the basolateral membrane of the parietal cell faces the external medium and the luminal membrane forms closed compartments and does not face the medium. Therefore, omeprazole added to the medium permeated the basolateral membrane, moved through the cytosol without activation, and permeated the luminal membrane. Then it was accumulated in the luminal acidic compartments and activated by acid. A similar process of omeprazole accumulation may also occur in vivo from the circulating system of the blood to the luminal acidic compartments of gastric glands. Imaging of the fluorescence in an oxyntic cell of Xenopus [Figure 8) provided direct evidence that omeprazole was transformed to the active sulfenamide derivative within oxyntic cells. The distribution of the fluorescence of the sulfenamide indicates that the closed acidic compartments are localized in parts of the oxyntic cell. In this experiment, the medium did not contain glutathione. Furthermore, omeprazole was not activated in the extracellular medium adjacent to the cell surface, indicating that there was no stagnant layer of leaked proton from oxyntic cells. Therefore, the acidic compartments are not connected to the extracellular medium. Luminal openings of oxyntic cells in vivo are probably self-sealed when the cells are isolated from mucosa and freed from the restraint of the surrounding cells and connecting tissues. In conclusion, the present observations gave direct
evidence that the cyclic sulfenamide derivative of omeprazole is a key product that inhibits the H+,K+ATPase in isolated gastric glands, oxyntic cells, and gastric vesicles.
References Gustavsson S, Adami HO, Loof L, Nyberg A, Nyren 0. Rapid healing of duodenal ulcers with omeprazole: double-blind dose-comparative trial. Lancet 1983;ii:124-5. 2. Lind T, Cederberg C, Ekenved G, Haglund U, Olbe L. Effect of omeprazole-a gastric proton pump inhibitor-on pentagastrin stimulated acid secretion in man. Gut 1983;24:270-6. 3. Larsson H, Carlsson E, Junggren U, et al. Inhibition of gastric acid secretion by omeprazole in the dog and rat. Gastroenterology 1983;85:900-7. 4. Brandstrom A, Lindberg P, Junggren U. Structure activity relationships of substituted benzimidazoles. Stand J Gastroenter01 1985;108(Suppl):15-22. 5. Starlinger MJ, Hollands MJ, Rowe PH, Matthews JB, Silen W. Chloride transport of frog gastric fundus: effects of omeprazole. Am J Physiol 1986;250:G118-26. 6. Fellenius E, Berglindh T, Sachs G, et al. Substituted benzimidazoles inhibit gastric acid secretion by blocking (H+ + K+)-ATPase. Nature 1981;290:159-61. 7. Wallmark B, Jaresten BM, Larsson H, Ryberg B, Brandstrom A, Fellenius E. Differentiation among inhibitory actions of omeprazole, cimetidine, and SCN- on gastric acid secretion. Am J Physiol 1983;245:G64-71. 8. Sewing KF, Harms P, Schulz G, Hannemann H. Effect of substituted benzimidazoles on acid secretion in isolated and enriched guinea pig parietal cells. Gut 1983;24:557-60. 9. Larsson H, Carlsson E, Sundell G. Effects of omeprazole and cimetidine on gastric acid secretion and right atria1 beating frequency in isolated organ preparations from the guinea pig. Digestion 1984;29:12-8. 10. Fryklund J, Wallmark B, Larsson H, Helander HF. Effect of omeprazole on gastric secretion in H+,K+-ATPase and in pepsinogen-rich cell fractions from rabbit gastric mucosa. Biochem Pharmacol 1984;33:273-80. 11. Im WB, Blakeman DP, Mendlein J, Sachs G. Inhibition of (H+ + K+)-ATPase and H+ accumulation in hog gastric membranes by trifluoperazine, verapamil and 8-(N,N-diethylamino)octyl-3,4,5-trimethyloxybenzoate. Biochim Biophys Acta 1984;770:65-72. 12. Im WB, Sih JC, Blakeman DP, McGrath JP. Omeprazole, a specific inhibitor of gastric (H+ - K+)-ATPase, is a H’activated oxidizing agent of sulfhydryl groups. J Biol Chem 1985;260:4591-7, 13. Wallmark B, Brandstrom A, Larsson H. Evidence for acidinduced transformation of omeprazole into an active inhibitor of (H+ + K+)-ATPase within the parietal cell. Biochim Biophys Acta 1984;778:549-58. 14. Lorentzon P, Eklundh B. Brandstrom A, Wallmark B. The mechanism for inhibition of gastric (H’ + K+]-ATPase by omeprazole. Biochim Biophys Acta 1985;817:25-32. 15. Lorentzon P, Jackson R, Wallmark B, Sachs G. Inhibition of (H+ + K+)-ATPase by omeprazole in isolated gastric vesicles requires proton transport. Biochim Biophys Acta 1987;897: 1.
41-51. 16. Keeling DJ, Fallowfield
C, Underwood A. The specifity of omeprazole as an (Hf + K+)-ATPase inhibitor depends upon the means of its activation. Biochem Pharmacol 1987;36:339-
44. 17. Lindberg
P. Nordberg P, Alminger
T, Brandstrom
A, Wall-
lune 1989
18.
19. 20.
21.
22.
23.
24.
25.
mark B. The mechanism of action of the gastric acid secretion inhibitor omeprazole. J Med Chem 1986;29:1327-9. Figala V, Klemm K, Kohl B, et al. Acid activation of (H+ K’)-ATPase inhibiting 2-(2-pyridylmethyl-sulphinyl)benzimidazoles: isolation and characterization of the thiophilic ‘active principle’ and its reaction. J Chem Sot Chem Commun 1986::125-7. Rabon E, Takeguchi N, Sachs G. Water and salt permeability of gastric vesicles. J Membr Biol 1983;53:109-17. Takeguchi N, Yamazaki Y. Disulfide cross-linking of H,KATPase opens Cll conductance, triggering proton uptake in gastric vesicles. Studies with inhibitors. J Biol Chem 1986; 261:2560-6. Takeguchi N, Kashiwagura T, Morii M. A vesicle-coated particle method for the determination of electrophoretic mobility of gastric vesicles. Cell Struct Funct 1981;6:191-200. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with the Folin phenol reagent. J Biol Chem 1951;193:265-75. Berglindh T, Helander HF, iibrink KJ. Effect of secretagogues on oxygen consumption, aminopyrine accumulation and morphology in isolated gastric glands. Acta Physiol Stand 1976;‘37:401-14. Chang H, Saccomani G, Rabon E, Schackmann R, Sachs G. Proton transport by gastric membrane vesicles. Biochim Biophys Acta 1977;464:313-27. Rackur G, Bickel M, Fehlhaber HW, et al. 2-((2-pyridyl-
ACID ACTIVATION
26.
27.
28.
29.
30.
OF OMEPRAZOLE
1461
methyl)sulfinyl)benzimidazoles: acid sensitive suicide inhibitors of the proton transport system in the parietal cell. Biochem Biophys Res Commun 1985;128:477-84. Grzesiek S, Dencher NA. Dependency of pH-relaxation across vesicular membranes on the buffering power of bulk solutions and lipids. Biophys J 1986;50:265-76. Nunogaki K, Kasai M. Determination of the rate of rapid pH equilibration across isolated sarcoplasmic reticulum membranes. Biochem Biophys Res Commun 1986;140:934-t0. Shechter Y, Rubinstein M, Becker R, Bohak Z. Modulation of enzyme activity of chicken pepsin by covalent modification of its single-SH group. Eur J Biochem 1975;58:123-31. Kittang E, Aadland E, Schjdsby H. Effect of omeprazole on the secretion of intrinsic factor, gastric acid and pepsin in man. Gut 1985;26:594-8. Schettino T, Kohler M, Fromter E. Electrophysiologic studies on individual cells of isolated glands of rabbit gastric mucosa. In: Forte JG, Warnock DG, Rector FC Jr.. ed. Hydrogen ion transport in epithelia. New York: John Wiley & Sons, 1984: 281-91
Received July 25, 1988. Accepted December 24, 1988. Address requests for reprints to: Dr. Magotoshi Morii, Faculty of Pharmaceutical Sciences, Toyama Medical and Pharmaceutical University, Toyama 930-01, Japan.
1989;98:1453-81
Acid Activation of Omeprazole in Isolated Gastric Vesicles, Oxyntic Cells, and Gastric Glands MAGOTOSHI MORII, HIDEYUKI TAKATA, NORIAKI TAKEGUCHI Faculty of Pharmaceutical
and
SGences, Toyama Medical and Pharmaceutical
University,
Toyama,
Japan
Omeprazole, a potent inhibitor of gastric hydrogen ion transporting, potassium-stimulated adenosine triphosphatase, was found to be transformed into an SH-reactive strong fluorescent molecule (excitation and emission wavelengths of 370 and 560 nm, respectively) in an acidic medium. The addition of glutathioneor protein-containing sulfhydryl groups such as pepsin to the medium decreased the fluorescence. Also, the increase in the pH of the medium decreased the fluorescence. The fluorescent molecule was identified to be an acid-activated planar cyclic sulfenamide derivative of omeprazole. The transformation was studied in H+-preaccumulated hog gastric vesicles, which contain the hydrogen ion transporting, potassium-stimulated adenosine triphosphatase. The addition of omeprazole to the vesicle suspension induced a rapid increase in the fluorescence intensity, indicating that omeprazole was activated in the intravesicular space. Then, the intensity biphasically decreased with time. The slower small decrease was due to the reaction of the sulfenamide with sulfhydryl group(s) located on the acid secretory side of the hydrogen ion transporting, potassium-stimulated adenosine triphosphatase. Omeprazole was also activated in the acidic lumina of isolated rabbit gastric glands that were stimulated with histamine. Furthermore, direct evidence was obtained from the imaging of the fluorescence that omeprazole was activated in the acidic compartments of the isolated Xenopus oxyntic cell.
0 tive, secretion
meprazole, a substituted benzimidazole derivahas been shown to inhibit gastric acid in vivo and in vitro [l-5), proton accumulation in isolated parietal cells, gastric glands, and vesicles (6-11). Omeprazole does not inhibit hydrogen ion transporting, potassium-stimulated adeno-
sine triphosphatase (H+,K+-ATPase) activity at neutral pH, but upon treatment with acid it inhibits adenosine triphosphate (ATP) hydrolysis and proton transport (12-16). Studies on the chemistry of omeprazole in an acidic solution indicated that the cyclic sulfenamide derivative of omeprazole {a mixture of two isomers: 2,4-dimethyl-3,9and 2,4-dimethyl-3,10-dimethoxy-5H-pyrido[1’,2’:4,5][1,2,3]thiadiazino[2,3-a]benzimidazol-l3-ium chloride} and sulfenic acid derivative {3,5-dimethyl-4-methoxy-2(hydroxy-thiomethyl) - 1- (5 -methoxy- &benzimidazolyl)-pyridinium chloride} are very reactive with sulfhydryl compounds such as glutathione and 2mercaptoethanol (17,181. It has been suggested that the acid-activated compounds are involved in inactivation of the H+,K+-ATPase. However, direct experimental evidence that these compounds are active forms of omeprazole in the secretory canaliculi of parietal cells or the intravesicular medium of gastric vesicles has not yet been shown. Although the sulfenamide derivative of omeprazole has a characteristic absorbance at 370 nm (18), it is impossible to follow the activating process of omeprazole in gastric vesicles and glands from measurements of absorbance of the vesicle or gland suspension because of the high turbidity of the suspension and changes in the intensity of light scattering induced by change in the vesicle volume (19). In this study, we found that one of the active forms of omeprazole, the sulfenamide, has a strong fluorescence and that the fluorescence disappears as it reacts with sulfhydryl compounds. The fluorescence
Abbreviations used in this paper: Em, emission wavelength; Ex, excitation wavelength: H+,K+-ATPase, hydrogen ion transporting, potassium-stimulated adenosine triphosphatase. 0 1989 by the American Gastroenterological Association 0018-5085/89/$3.50
GASTROENTEROLOGYVol.96,No.6
1454 MORII ET AL.
change could be monitored at a high accuracy even in a highly turbid suspension such as gastric vesicle, oxyntic
cells,
and gastric
glands.
These
properties
total cell number tion.
as determined
by microscopic
observa-
of
the sulfenamide enabled us to follow the activation process of omeprazole in isolated gastric vesicles, oxyntic cells, and gastric glands and the sulfhydryl modification process in the gastric vesicles.
Measurement Acid-Activated
of Fluorescence Omeprazole
of
Omeprazole (10 PM) was incubated with 0.1 N HCl at 25°C. The absorbance of activated omeprazole (OD,,,) was measured by an Aminco DW-2C UV-VIS spectrophotometer. The fluorescence [excitation wavelength (Ex) = Materials and Methods 370 nm, emission wavelength (Em) = 560 nm] was meaChemicals sured with a Hitachi 650-10s fluorescence spectrophotometer equipped with a magnetic stirrer under the same Omeprazole was obtained from Fujisawa-Astra Co. conditions as absorbance measurements. The width of slit Ltd. (Osaka, Japan). Timoprazole, picoprazole, and 2-[2-(3, for excitation light in the fluorescence measurements was 5-dimethyl-4-methoxy)-pyridylmethylsulfiny1]-(5-methoxyfixed at 1 nm to minimize photobleaching. carbonyl-6-methyl)-benzimidazole (H compound) were synthesized in our laboratory (20). The tetrafluoroborate salt of the acid-activated compound of omeprazole, a cyclic Measurements of Proton Uptake and sulfenamide derivative of omeprazole, was synthesized as Fluorescence in Gastric Vesicles described elsewhere (17). Ranitidine was obtained from Proton uptake into gastric vesicles was measured Sankyo Co. Ltd. (Osaka, Japan). Pepsin from porcine stomwith a pH electrode (21,24). Gastric vesicles (100 pg/ml) ach mucosa with a digestive power of 1: 10,000 was were incubated in a solution containing 150 mM KCl, 5 obtained from Wako Pure Chemicals (Osaka, Japan). mM glycylglycine (pH 6.11), 0.2 mM MgCl,, 5 pg/ml of SCH28080 was obtained from Schering Corp. All other valinomycin, 1 mM glutathione, and 0.3 mM magnesium chemicals were of the highest purity available. salt of adenosine triphosphate (Mg-ATP) with or without 10 PM omeprazole at 25°C. The time-dependent increase in pH of the suspension was measured by a pH-meter Preparation of Hog Gastric Vesicles (Radiometer, PHM64) and recorded by a strip chart reMembrane vesicles containing the Hf,Kf-ATPase corder. The fluorescence intensity of the activated omeprain 0.25 M sucrose were prepared from hog stomachs as zole was also measured in the same incubation mixture described previously (21). Protein concentration was de(Ex = 370 nm, Em = 560 nm). At least 10 PM omeprazole termined by the method of Lowry et al. with bovine serum was required for high signal-to-noise measurements of albumin as standard (22). fluorescence. Preparation
of Rabbit
Gastric
Glands
The isolated gastric glands were prepared from rabbit gastric mucosa as described previously (23). They were suspended in a solution containing 132.4 mM NaCl, 5.4 mM KCl, 5.0 mM Na,PO,, 1.0 mM NaHPO,, 1.2 mM MgSO,, 1.0 mM CaCl,, 10 mM HEPES, 2 mg/ml of glucose, and 0.4 mg/ml of bovine serum albumin (pH 7.4). The isolated gastric glands were used within 3 h after preparation.
Preparation
of Xenopus
Oxyntic
Measurement of Fluorescence Gastric Glands
in Isolated
The isolated gastric glands suspended in HEPESNaOH buffer solution (pH 7.4) described above were preincubated with 100 FM histamine at 37°C for 15 min. Then 100 PM omeprazole at a final concentration was added, and the subsequent changes in the fluorescence intensity (Ex = 370 nm, Em = 560 nm) were measured. At least 100 PM omeprazole was required for high signal to noise measurements.
Cells
The isolated oxyntic cells were prepared from gastric mucosa of Xenopus laevis. Gastric mucosa was minced with small scissors in a buffer solution containing 87 mM NaCl, 4 mM KCl, 18 mM NaHCO,, 1 mM KH,PO,, 1 mM MgCl,, 2 mM CaCl,, and 2 mg/ml of sucrose (pH 7.4) and then filtered through a nylon mesh (20 mesh). The filtrate was centrifuged at 1000 rpm for 5 min with a Hitachi 05PR-22 refrigerated centrifuge. The sedimented fraction was washed 3-5 times with the buffer solution and suspended in the same buffer solution. The content of oxyntic cells in the cell preparation was about 50% of the
Imaging of the Distribution of Activated Omeprazole in Isolated Xenopus Oxyntic Cells The isolated oxyntic cells of Xenopus laevis were fixed on polylysine-coated slide glass in the buffer solution (pH 7.4) described above under an inverted fluorescence microscope (Nikon TMD-EFQ). Acid secretion was stimulated by preincubating with 1 mM dibutyryl-cyclic adenosine monophosphate at 25°C for 30 min. Then 100 PM omeprazole was added. Time-dependent changes of the microscopic image of fluorescence were analyzed with a
June
1989
ACID ACTIVATION OF OMEPRAZOLE
Spex Fluorolog-2 spectrofluorometer and an IMl imaging system. The excitation wavelength was 370 nm. Emitted light was passed through a dichroic mirror (cutoff wavelength = 510 nm) and a high-pass filter (cutoff wavelength = 520 nm), and the image was saved in a SIT-camera. The maximum resolution of image was 512 by 478 pixels by a-bit. Autofluorescence of cells was subtracted from the fluorescence image. The gray scale of fluorescence intensity was digitally enhanced. Pseudocoloring images of the relative fluorescence scale of activated omeprazole were hard-copied.
Ex370
1455
Em560
Kesults Acid Activation
of Omeprazole
Omeprazole in acidic solutions reacts with specific: sulfhydryl group(s) located on the intravesicular surface of the H+,K+-ATPase, resulting in inhibition of ATP hydrolysis and proton transport activity (15,16). Several degradation products are generated by the acid treatment (18,251. A cyclic sulfenamide derivative may be the main reactive degradation product in low pH because another reactive sulfenic acid is unstable in low pH and converts to the sulfenamide (17,18). The planarity of four rings of the sulfenamide derivative leads to a bathochromic shift of the long-wavelength ultraviolet absorption from 301 to 370 nm (18). Curve A in Figure 1 shows the time-dependent increase in absorbance at 370 nm induced by generation of the
=
0
Time. min
Figure
‘1. Time-dependent changes of absorbance (A) and fluorescence (B) of omeprazole activated by HCl. The methanolic solution of free base of omeprazole was added in 0.1 N HCl solution at time zero. The final concentration of omeprazole was 10 PM and that of methanol was -0.1%. A. The optical density was measured at 370 nm with the narrowest slit-width to avoid degradation of activated product(s) by the incident light. At the arrow, glutathione was added to a final concentration of 100 pM. B. The fluorescence intensity was measured at Ex = 370 nm and Em = 560 nm with the narrowest excitation slit-width. Glutathione was added to a final concentration of 100 PM.
300
400
500
Wavelength,
600
700
nm
Figure 2. Excitation and emission spectra of acid-activated omeprazole. The methanolic solution of free base of omeprazole was dissolved in 0.1 N HCI at a final concentration of 10 KM. About 20 min later, the excitation and emission spectra were measured.
sulfenamide derivative from 10 PM omeprazole in 0.1 N HCl. The rate of increase did not depend on the pH of the medium (pH 1 to pH 4). The reaction of the sulfenamide with 100 PM glutathione resulted in a decrease in absorbance at 370 nm. The reaction rate with glutathione increased as the pH of the medium was increased in the range from pH 1 to pH 4 (data not shown). The absorbance decreased instantly when the pH of the medium was neutralized by the addition of NaOH (data not shown). These results indicate that the sulfenamide was transformed to nonplanar derivatives by reaction with the sulfhydry1 group or increasing the pH of the medium. We found that the acid-treated omeprazole shows strong fluorescence in acidic medium. Figure 2 shows that the peak of the excitation wavelength is 370 nm and the peak of the emission wavelength is 570 nm for the acid-treated omeprazole in 0.1 N HCl. Curve B in Figure 1 shows the time-dependent change of fluorescence under the same conditions as the absorbance measurement. The similarity between the time-course of absorbance and of fluorescence, together with the identity between the absorbance wavelength and the excitation wavelength (370 nm), indicates that the cyclic sulfenamide derivative of omeprazole is a fluorescent product of acid-acti-
1456
GASTROENTEROLOGY
MORE ET AL.
0
I
I
I
I
10
20
30
40
Time.
Figure
mln
3. Reaction of the activated omeprazole with pepsin. The methanolic solution of free base of omeprazole was dissolved in 0.1 N HCl at a final concentration of 10 PM. The fluorescence intensity was measured at Ex = 370 nm and Ex = 560 nm. Pepsin was added to a final concentration of 2.5 mgiml at the arrow.
vated omeprazole. Furthermore, we synthesized tetrafluoroborate salt of the cyclic sulfenamide and found that the excitation and emission wavelengths in dimethylsulfoxide were 380 and 550 nm, respectively, which indicates that the fluorescent molecule in the acidic condition is the cyclic sulfenamide derivative of omeprazole. The fluorescence was gradually bleached when the intensity of the excitation light was strong. To circumvent this problem, the width of the slit for the excitation light was set at 1 nm and solutions were continuously stirred during the measurements. Similar acid-activated products of other substituted benzimidazoles, timoprazole, picoprazole, and H compound, were found not to be fluorescent molecules. However, several similar substituted benzimidazole derivatives that have a methoxy group at the 5-position in the benzimidazole ring, as in the case of omeprazole, produced fluorescent molecules by acid treatment. The fluorescence property must be due to the presence of a resonance form of four coplanar rings dependent on electron donation by the methoxy group in the benzimidazole ring. The sulfenamide also reacted with protein-containing sulfhydryl groups such as pepsin at pH 1 (Figure 3). This indicates the possibility that omeprazole modulates the peptidase activity of pepsin in vivo. Inhibition of Proton Uptake in Hog Gastric Vesicles by Omeprazole The inhibitory effects of omeprazole on proton uptake were investigated by measuring alkaliza-
Vol. 96, No. 6
tion of the extravesicular medium using a pH electrode. We could detect the small change in the pH with high sensitivity in the presence of 5 mM glycylglycine, which showed a weak pH buffering action. In a separate experiment, it was found that proton generation induced by ATP hydrolysis was negligibly small at pH 6.11, and hence the observed changes in the extravesicular pH were due to proton uptake into the vesicles. Figure 4 shows the inhibitory effect of 10 PM omeprazole on proton uptake (100 pg of protein per milliliter). To initiate the proton uptake, the vesicles were added to the reaction medium containing Mg-ATP with or without 10 PM omeprazole. In this experiment, we added 1 mM glutathione to the solution. Activated omeprazole in the extravesicular spaces at pH 6.11 was instantly quenched by glutathione. Therefore, we could measure the fluorescence only in the intravesicular space. In the control experiment (curve B), the net proton uptake rate became zero at 10 min after the addition of Mg-ATP as the rate of the proton leak became equal to that of proton uptake when the proton gradient between intravesicular and extravesicular medium became large. Omeprazole (pK, 4) is membrane permeable and can rapidly accumulate in an acidic space. However, during the first 3.5 min of the proton uptake reaction in the presence of omeprazole (curve A), no inhibition was observed and proton uptake was even slightly enhanced, indicat-
I-
I,
0
I
,.I
1,.
,
,
10
5
,
,
,
,
,
15
Time, min Figure
4. Inhibition of proton uptake of gastric vesicles by omeprazole. The reaction solution contained 150 mM KCl, 0.2 mM M&l,, 1 mM glutathione, 0.3 mM Mg-ATP, 5 &ml valinomycin, and 5 mM glycylglycine (pH 6.11) with (A) or without (B) 10 PM omeprazole. Vesicles in 250 mM sucrose were added to the reaction solution to a final concentration of 100 pg protein per milliliter at the arrow. The proton uptake was measured by a pH-meter as an alkalization of extravesicular medium.
June 198EI
Figure
5. Activation of omeprazole in the intravesicular space of gastric vesicles. The reaction mixture contained gastric vesicles (100 wg of protein per milliliter), 150 mM KCl, 0.2 mM MgCl,, 1 mM glutathione, 5 mM glycylglycine (the pH of medium was preadjusted to a value so as to shift the pH to 6.11 when Mg-ATP was added), and 5 pgiml valinomycin with (A] or without (B) omeprazole. Magnesium-stimulated adenosine triphosphate was added to a final concentration of 0.3 mM at the arrow. Fluorescence was measured with excitation and emission wavelengths at 370 nm and 560 nm. respectively.
ing that omeprazole was not activated in the initial phase. This may be due to the pH buffering activity in lipid vesicles (26,271. Omeprazole started to reduce the rate of proton uptake at about 3.5 min and the rate of proton leak exceeded that of proton uptake after about 6 min. Figure 5 shows the changes in the fluorescence intensity (Ex = 370 nm, Em = 560 nm) during the proton uptake reaction, which are similar to those shown in Figure 4. There were no changes in the fluorescence in the absence of Mg-ATP, indicating that the pH of the intravesicular medium (initially 250 mM sucrose, pH 7) was not low enough for acid activation of omeprazole in the short incubation time with glycylglycine buffer (pH 6.11). Magnesium-stimulated adenosine triphosphate was added at zero time. The control experiment performed in the absence of omeprazole (curve B) shows that the fluorescence intensity decreases during the initial proton uptake reaction. This decrease was due to interference by the light scattering changes caused by change in the vesicle volume. However, at present, we have no clear explanation for the decrease in light scattering. In the presence of omeprazole (curve A), the intensity of fluorescence transiently increased with a time lag of 3.5 min. This transient increase in the fluorescence intensity is due to the formation of the cyclic sulfenamide de-
ACID ACTIVATION OF OMEPRAZOLE
1457
rivative of omeprazole in intravesicular space. The time lag of activation may be due to a strong pH buffering action of the intravesicular medium and phospholipids of membranes (26,271. The acidity of the intravesicular medium may not be high enough to activate omeprazole during the lag phase. To clarify the reason for the later decrease of fluorescence in curve A in Figure 5, we carried out the experiment shown in Figure 6. In this experiment, omeprazole was added after the intravesicular acidification had been established (10 min after the addition of 0.3 mM Mg-ATP). As the sulfhydryl groups of the enzyme on the extravesicular side of membrane were protected by the presence of an excess amount of glutathione (1 mM) in the solution, only the activated omeprazole in the intravesicular medium could react with sulfhydryl groups of H+,K+-ATPase. The proton began to leak within 1 min after the addition of omeprazole. The fluorescence change, which reflects change in the amount of the sulfenamide in the intravesicular medium, was complicated. After the immediate increase in the fluorescence upon addition of omeprazole, the fluorescence decreased biphasically (curve A). The biphasic decrease of the fluorescence may be due to two factors: (a) transformation of the excess amount of the sulfenamide into a nonfluorescent molecule such as sulfenic acid, caused by an increase in the intravesicular pH, and (b) the reaction with sulfhydry1 group(s) located on the acid secretory side of I
.E
%
15
Figure 6. Activation of omeprazole and inhibition of proton uptake in the internal medium of the acid preaccumulated gastric vesicles. The reaction mixture contained gastric vesicles (100 pg of protein per milliliter], 150 mM KCl, 0.2 mM MgCl,, 1 mM glutathione, 5 pglml valinomycin, and 5 mM glycylglycine (pH 6.11). Proton uptake was started by addition of Mg-ATP to a final concentration of 0.3 mM. About 10 min later, 10 WM omeprazole (A and C) or 10 PM omeprazole plus 100 PM timoprazole (B) was added at the arrows. The changes in fluorescence intensity [A and B) and the proton leak (C) from gastric vesicles were measured as described in Materials and Methods.
1458
GASTROENTEROLOGY Vol. 96, No. 6
MORII ET AL.
H+,K+-ATPase. As omeprazole was added after the intravesicular acidification had been established in this experiment, formation of the activated omeprazole is rapid and a partial inactivation of enzyme would occur. This loss of activity caused the proton leak as shown in curve C in Figure 6. This proton leak induced an increase in the intravesicular pH, resulting in a decrease in the fluorescence. Thus, the faster decreasing phase would be due to transformation of the fluorescent sulfenamide into a nonfluorescent compound (17,18).In the experiments shown in curve B in Figure 6, 100 PM timoprazole was added together with 10 PM omeprazole. Timoprazole is a substituted benzimidazole derivative and its acid-activated sulfenamide inhibits H+,K+ATPase activity. However, as described above, the sulfenamide derivative of timoprazole is not a fluorescent molecule. As shown in curve B, the fluorescence of the activated omeprazole initially increased then monophasically decreased. The disappearance of the slower decreasing phase as compared with the case of curve A, indicates that most of the activated omeprazole could not react with sulfhydryl groups because the concentration of timoprazole was 10 times higher than that of omeprazole. That is, the slower decreasing phase of fluorescence in curve A would be due to the modification of the remaining sulfhydryl groups by the sulfenamide. Measurement of Fluorescence Rabbit Gastric Glands
in Isolated
The activating process of omeprazole in histamine-stimulated gastric glands was observed. The fluorescence intensity increased and then decreased (curve A in Figure 7).A potent histamine H,-receptor antagonist, ranitidine (10 PM), in the presence of histamine inhibited the rapid increase of the fluorescence (curve B in Figure 7).Therefore, the increase shown in curve A is due to activation of omeprazole. Ranitidine does not inhibit the basal acid secretion. Thus, the basal acid secretion slowly activated omeprazole (curve B). Another type of H+,K+-ATPase inhibitor SCH28080 competes with K+ at the high-affinity K+ binding site of H+,K+ATPase. The addition of SCH28080 results in inhibition of both stimulated and basal acid secretion. The addition of 2 PM SCH28080 together with 100 /.LM histamine completely inhibited the increase in the fluorescence intensity (data not shown). These results indicate that closed acidic canaliculi exist in histamine-stimulated gastric glands and that omeprazole accumulated in acidic canaliculi was activated by acid.
0
10
5
15
Time, min Figure 7. Activation of omeprazole in acidic canaliculi of isolated gastric glands. Isolated gastric glands suspended in HEPES-NaOH buffer (pH 7.4, detail in Materials and Methods] were stimulated by 100 FM histamine (A) or 100 PM histamine plus 10 FM ranitidine (B) at 37°C for 15 min. Then 100 FM omeprazole was added at the arrow, and the change in the fluorescence intensity (Ex = 370 nm, Em = 560 nm) was measured.
Imaging of the Distribution of Activated Omeprazole in Single Xenopus Oxyntic Cells In this experiment, we used amphibian oxyntic cells as they can be stimulated with dibutyrylcyclic adenosine monophosphate at room temperature. The stimulation induced a morphologic change of the cell shape from a round to an irregular shape. Time-dependent changes in pseudocoloring images of the two-dimensional distribution of activated omeprazole in an oxyntic cell of Xenopus are shown in Figure 8. Activated omeprazole was localized only in the oxyntic cells, and the fluorescence in parts of the oxyntic cell gradually increased. These results indicate that closed acidic compartments were formed in dibutyryl-cyclic adenosine monophosphate-stimulated oxyntic cells and that omeprazole was activated only in the closed acidic compartments. The increasing rate of fluorescence was slower than that in rabbit gastric glands, probably because of the slow proton accumulation by oxyntic cells at decreased temperature (25T).Also, we could detect localized acidic compartments within oxyntic cell of Xenopus by fluorescence imaging of acridine orange distribution (data not shown).
Discussion We found that acid treatment of omeprazole produced an SH-reactive strong fluorescent molecule and that the fluorescent molecule converted to a nonfluorescent molecule when reacted with the sulfhydryl group. The identity of the excitation wave-
June 1989
ACID ACTIVATION OF OMEPRAZOLE
1459
Figure 8:. Imaging of the distribution of activated omeprazole in an oxyntic cell. Isolated oxyntic cells of Xenopus laevis were stimulated with 1 mM dibutyryl-cyclic adenosine monophosphate at 25°C for 30 min. Then 100 PM omeprazole was added. Microscopic images of the fluorescence were saved in a SIT-camera and analyzed by a Spex imaging system [detail in Materials and Methods). 1, Image of an oxyntic cell by visible light. 2-4, Fluorescence distribution images at various times after the addition of omeprazole: 2, 260 s; 3, 320 s; and 4. 400 s. The bar in image 1 indicates 10 pm. The colored strip indicates the scale of relative concentration of activated omeprazole.
length and the absorbance wavelength indicates that the fluorescent molecule is a planar cyclic sulfenamide derivative of omeprazole. Furthermore, the synthesized tetrafluoroborate salt of the cyclic sulfenamide has approximately the same excitation and emission wavelengths as those of the activated compound of omeprazole. These results indicate that the
same activated omeprazole as that produced by HCl treatment in a test tube was generated in the intravesicular medium of hog gastric vesicle. The time lag of activation (curve A in Figure 5) indicates that the pH buffering action of intravesicular medium and phospholipids (26,27) is strong enough to neutralize the pH in the initial phase of proton uptake. When
1460
GASTROENTEROLOGY Vol. 96, No. 6
MORII ET AL.
omeprazole was added to the H+-preaccumulated vesicle suspension, omeprazole was activated quickly. The sulfenamide also reacted with sulfhydryl group(s) of pepsin (Figure 3). Activated omeprazole in acidic gastric juice may react with sulfhydryl group(s) of pepsin in vivo. Although pepsin is not a so-called SH-enzyme, the modification of the sulfhydry1 group modulates pepsin activity (28). Thus, sulfhydryl modification of pepsin may be part of the antiulcer effect of omeprazole in vivo (29). In spite of strong autofluorescence of the gastric glands themselves (3O), the fluorescence of acidactivated omeprazole was measurable in histaminestimulated isolated gastric glands. The increasing and decreasing phases of the fluorescence reveal that omeprazole was activated in gastric glands. Preincubation of glands with antisecretory reagents, ranitidine or SCH28080, inhibited the increase of fluorescence, indicating that the acid accumulation is necessary for activation of omeprazole. Moreover, Im et al. (12) reported that the potency of omeprazole to inhibit rat H+,K+-ATPase activity in vivo was enhanced by pretreatment with carbachol and weakened by pretreatment with cimetidine. In gastric glands obtained by treatment with collagenase, the basolateral membrane of the parietal cell faces the external medium and the luminal membrane forms closed compartments and does not face the medium. Therefore, omeprazole added to the medium permeated the basolateral membrane, moved through the cytosol without activation, and permeated the luminal membrane. Then it was accumulated in the luminal acidic compartments and activated by acid. A similar process of omeprazole accumulation may also occur in vivo from the circulating system of the blood to the luminal acidic compartments of gastric glands. Imaging of the fluorescence in an oxyntic cell of Xenopus [Figure 8) provided direct evidence that omeprazole was transformed to the active sulfenamide derivative within oxyntic cells. The distribution of the fluorescence of the sulfenamide indicates that the closed acidic compartments are localized in parts of the oxyntic cell. In this experiment, the medium did not contain glutathione. Furthermore, omeprazole was not activated in the extracellular medium adjacent to the cell surface, indicating that there was no stagnant layer of leaked proton from oxyntic cells. Therefore, the acidic compartments are not connected to the extracellular medium. Luminal openings of oxyntic cells in vivo are probably self-sealed when the cells are isolated from mucosa and freed from the restraint of the surrounding cells and connecting tissues. In conclusion, the present observations gave direct
evidence that the cyclic sulfenamide derivative of omeprazole is a key product that inhibits the H+,K+ATPase in isolated gastric glands, oxyntic cells, and gastric vesicles.
References Gustavsson S, Adami HO, Loof L, Nyberg A, Nyren 0. Rapid healing of duodenal ulcers with omeprazole: double-blind dose-comparative trial. Lancet 1983;ii:124-5. 2. Lind T, Cederberg C, Ekenved G, Haglund U, Olbe L. Effect of omeprazole-a gastric proton pump inhibitor-on pentagastrin stimulated acid secretion in man. Gut 1983;24:270-6. 3. Larsson H, Carlsson E, Junggren U, et al. Inhibition of gastric acid secretion by omeprazole in the dog and rat. Gastroenterology 1983;85:900-7. 4. Brandstrom A, Lindberg P, Junggren U. Structure activity relationships of substituted benzimidazoles. Stand J Gastroenter01 1985;108(Suppl):15-22. 5. Starlinger MJ, Hollands MJ, Rowe PH, Matthews JB, Silen W. Chloride transport of frog gastric fundus: effects of omeprazole. Am J Physiol 1986;250:G118-26. 6. Fellenius E, Berglindh T, Sachs G, et al. Substituted benzimidazoles inhibit gastric acid secretion by blocking (H+ + K+)-ATPase. Nature 1981;290:159-61. 7. Wallmark B, Jaresten BM, Larsson H, Ryberg B, Brandstrom A, Fellenius E. Differentiation among inhibitory actions of omeprazole, cimetidine, and SCN- on gastric acid secretion. Am J Physiol 1983;245:G64-71. 8. Sewing KF, Harms P, Schulz G, Hannemann H. Effect of substituted benzimidazoles on acid secretion in isolated and enriched guinea pig parietal cells. Gut 1983;24:557-60. 9. Larsson H, Carlsson E, Sundell G. Effects of omeprazole and cimetidine on gastric acid secretion and right atria1 beating frequency in isolated organ preparations from the guinea pig. Digestion 1984;29:12-8. 10. Fryklund J, Wallmark B, Larsson H, Helander HF. Effect of omeprazole on gastric secretion in H+,K+-ATPase and in pepsinogen-rich cell fractions from rabbit gastric mucosa. Biochem Pharmacol 1984;33:273-80. 11. Im WB, Blakeman DP, Mendlein J, Sachs G. Inhibition of (H+ + K+)-ATPase and H+ accumulation in hog gastric membranes by trifluoperazine, verapamil and 8-(N,N-diethylamino)octyl-3,4,5-trimethyloxybenzoate. Biochim Biophys Acta 1984;770:65-72. 12. Im WB, Sih JC, Blakeman DP, McGrath JP. Omeprazole, a specific inhibitor of gastric (H+ - K+)-ATPase, is a H’activated oxidizing agent of sulfhydryl groups. J Biol Chem 1985;260:4591-7, 13. Wallmark B, Brandstrom A, Larsson H. Evidence for acidinduced transformation of omeprazole into an active inhibitor of (H+ + K+)-ATPase within the parietal cell. Biochim Biophys Acta 1984;778:549-58. 14. Lorentzon P, Eklundh B. Brandstrom A, Wallmark B. The mechanism for inhibition of gastric (H’ + K+]-ATPase by omeprazole. Biochim Biophys Acta 1985;817:25-32. 15. Lorentzon P, Jackson R, Wallmark B, Sachs G. Inhibition of (H+ + K+)-ATPase by omeprazole in isolated gastric vesicles requires proton transport. Biochim Biophys Acta 1987;897: 1.
41-51. 16. Keeling DJ, Fallowfield
C, Underwood A. The specifity of omeprazole as an (Hf + K+)-ATPase inhibitor depends upon the means of its activation. Biochem Pharmacol 1987;36:339-
44. 17. Lindberg
P. Nordberg P, Alminger
T, Brandstrom
A, Wall-
lune 1989
18.
19. 20.
21.
22.
23.
24.
25.
mark B. The mechanism of action of the gastric acid secretion inhibitor omeprazole. J Med Chem 1986;29:1327-9. Figala V, Klemm K, Kohl B, et al. Acid activation of (H+ K’)-ATPase inhibiting 2-(2-pyridylmethyl-sulphinyl)benzimidazoles: isolation and characterization of the thiophilic ‘active principle’ and its reaction. J Chem Sot Chem Commun 1986::125-7. Rabon E, Takeguchi N, Sachs G. Water and salt permeability of gastric vesicles. J Membr Biol 1983;53:109-17. Takeguchi N, Yamazaki Y. Disulfide cross-linking of H,KATPase opens Cll conductance, triggering proton uptake in gastric vesicles. Studies with inhibitors. J Biol Chem 1986; 261:2560-6. Takeguchi N, Kashiwagura T, Morii M. A vesicle-coated particle method for the determination of electrophoretic mobility of gastric vesicles. Cell Struct Funct 1981;6:191-200. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with the Folin phenol reagent. J Biol Chem 1951;193:265-75. Berglindh T, Helander HF, iibrink KJ. Effect of secretagogues on oxygen consumption, aminopyrine accumulation and morphology in isolated gastric glands. Acta Physiol Stand 1976;‘37:401-14. Chang H, Saccomani G, Rabon E, Schackmann R, Sachs G. Proton transport by gastric membrane vesicles. Biochim Biophys Acta 1977;464:313-27. Rackur G, Bickel M, Fehlhaber HW, et al. 2-((2-pyridyl-
ACID ACTIVATION
26.
27.
28.
29.
30.
OF OMEPRAZOLE
1461
methyl)sulfinyl)benzimidazoles: acid sensitive suicide inhibitors of the proton transport system in the parietal cell. Biochem Biophys Res Commun 1985;128:477-84. Grzesiek S, Dencher NA. Dependency of pH-relaxation across vesicular membranes on the buffering power of bulk solutions and lipids. Biophys J 1986;50:265-76. Nunogaki K, Kasai M. Determination of the rate of rapid pH equilibration across isolated sarcoplasmic reticulum membranes. Biochem Biophys Res Commun 1986;140:934-t0. Shechter Y, Rubinstein M, Becker R, Bohak Z. Modulation of enzyme activity of chicken pepsin by covalent modification of its single-SH group. Eur J Biochem 1975;58:123-31. Kittang E, Aadland E, Schjdsby H. Effect of omeprazole on the secretion of intrinsic factor, gastric acid and pepsin in man. Gut 1985;26:594-8. Schettino T, Kohler M, Fromter E. Electrophysiologic studies on individual cells of isolated glands of rabbit gastric mucosa. In: Forte JG, Warnock DG, Rector FC Jr.. ed. Hydrogen ion transport in epithelia. New York: John Wiley & Sons, 1984: 281-91
Received July 25, 1988. Accepted December 24, 1988. Address requests for reprints to: Dr. Magotoshi Morii, Faculty of Pharmaceutical Sciences, Toyama Medical and Pharmaceutical University, Toyama 930-01, Japan.