Mucus glycoprotein secretion by duodenal mucosa in response to luminal arachidonic acid

Biochimica et Biophysica Acta 884 (1986) 419-428

419

Elsevier BBA22381

Mucus glycoprotein secretion by duodenal mucosa in response to luminal arachidonic acid

Malgorzata Kosmala a, Steven R. Carter a, Stanislaw J. Konturek b, Amalia Slomiany a and Bronislaw L. Slomiany a Gastroenterology Research Laboratory, Department of Medicine, New York Medical College, Valhalla, N Y 10595 (U.S.A.) and b Institute of Physiology Medical Academy, Cracow (Poland)

(Received 2 September 1986)

Key words: Mucin composition; Arachidonic acid; Mucus secretion; (Dog duodenum)

The effect of luminal application of arachidonic acid on the alkaline secretion, prostaglandin generation, and mucus glycoprotein output and composition was studied in proximal and distal duodenum of conscious dogs. Surgically prepared duodenal loops were instilled in vivo for up to 2 h with saline (control) followed by various concentrations (12.5-100 # g / m l ) of arachidonic acid. The experiments were conducted with and without intravenous pretreatment with indomethacin. The recovered instillates were assayed for the content of prostaglandin and H C 0 3 , and used for the isolation of mucus glycoprotein. Exposure of duodenal mucosa to arachidonic acid led to concentration-dependent increase in the output of HCO 3 and prostaglandin generation. In both cases this response was greater in the proximal duodenum. Pretreatment with indomethacin caused reduction in the basal HCO3 and prostaglandin output, and prevented the increments evoked by arachidonic acid. The proximal and distal duodenum displayed similar basal output and composition of mucus glycoprotein. Comparable increases in these giycoproteins were also obtained with arachidonic acid, the effect of which was abolished by indomethacin. Compared to basal conditions, mucus glycoproteins elaborated in response to arachidonic acid exhibited higher contents of associated lipids and covalently bound fatty acids, and contained less protein. The associated lipids of mucus glycoproteins elaborated in the presence of arachidonic acid showed enrichment in phospholipids and decrease in neutral lipids. The carbohydrate components in these glycoproteins also exhibited higher proportions of sialic acid and sulfate. The changes brought about by arachidonic acid were prevented by indomethacin pretreatment, and in both cases the glycoprotein composition returned to that obtained under basal conditions. The enrichment of mucus glycoprotein in lipids, sialic acid and sulfate in response to endogenous prostaglandin may be of significance to the function of this giycoprotein in the hostile environment of the duodenum.

Introduction Mucus layer covering the epithelial surfaces of the gastrointestinal tract constitutes the first line of mucosal defense against mechanical, chemical, Correspondence address: Dr. B.L. Siomiany, Gastroenterology Research Laboratory, New York Medical College, Munger Pavilion, Valhalla, NY 10595, U.S.A.

enzymatic and bacterial insults [1,2]. The component primarily responsible for these protective functions of the mucus gel is a highly glycosylated mucus glycoprotein. This high molecular weight glycoprotein existing in the mucus layer in an expanded form, imbibed with water, is capable of interacting with other constituents of the gel such as proteins and lipids [3-5]. The continuous renewal and its resilient nature assures the viscoelas-

420

tic and permselective properties of the gel, and provides an ideal milieu that confines the reaction between secreted HCO 3 and H ÷ entering the gel in such a way that a neutral pH is maintained on the mucosal side [4,5-8]. The process of HCO 3 secretion involves the prostaglandin mechanism, and the alkaline output of the mucosa varies depending upon the location along the gastrointestinal tract with the highest being in duodenum [8-10]. The duodenum, in particular its proximal portion, is also exposed to the greatest pH fluctuation due to the frequent contact with acid leaving the stomach. While the capacity of the duodenal mucosa to cope with gastric acid has been examined with respect to bicarbonate output and the level of prostaglandin [8,10], less attention has been paid to the nature of duodenal mucus gel which provides the mixing barrier for H ÷ and HCO~-. We report here on the content and composition of mucus glycoproteins elaborated by dog proximal and distal duodenum in response to topical application of prostaglandin precursor, arachidonic acid.

experiments, were placed in cloth slings and the cannula of each loop was connected by a rubber tube (10 mm in diameter) to a barostat to keep a constant intraluminal pressure of 10 cm H 2 0 in the loop lumen. Each loop was instilled with 0.15 M NaC1 (pH 6.0) applied alone or in combination with arachidonic acid. At the end of each 30 min period, the bathing fluid was removed, its volume measured to the nearest 0.1 ml and fresh saline or arachidonic acid solution was instilled. In all cases the recovery of the instilled fluids ranged from 98 to 104%. The instillation procedures were continued at 30 min intervals for up to 2 h. The experiments were conducted without and with pretreatment with indomethacin (2.5 m g / k g given intravenously 30 min prior to instillation). Test solutions of arachidonic acid (Sigma), 12.5, 25, 50 and 100/zg/ml, were prepared in saline (pH 6.0) immediately prior to use with brief (30 s) sonication [10]. Animals were used twice a week and three experiments were performed with each test solution on each animal.

Animal preparation The studies were conducted with ten mongrel dogs of 18-20 kg weight. Five animals were used for the preparation of proximal duodenal loops and five for distal duodenal loops. The isolated loops, each 7 cm long, were fashioned from the proximal or distal portion of the duodenum and equipped with Gregory-type metal cannula for drainage [10]. In the case of proximal duodenal loops, the common bile duct and accessory pancreatic duct were ligated and transected, and the gallbladder was anastomosed to the jejunum about 10 cm distal to the ligament of Treitz. Following surgery, the gastrointestinal continuity was restored by end-to-end gastroduodenal or duodenojejunal anastomosis [10], and the dogs were allowed a 6-week recovery period prior to beginning the experiments. The volume of the prepared loops, measured at a distention pressure of 10 cm H20, was about 13 ml.

Isolation of mucus glycoprotein The recovered instillates from each experiment were made into separate pools, dialyzed against distilled water and lyophilized. The powder was dissolved at 8 m g / m l in 6 M u r e a / 1 0 mM sodium phosphate buffer (pH 7.0) and applied in 5 ml portions to Bio-Gel P-100 columns (2.0 × 120 cm). Elution was achieved with buffered 6 M urea and the eluted fractions (4 ml) were monitored for protein and carbohydrate [11]. The tubes containing the excluded mucus glycoprotein peak were pooled, dialyzed against distilled water and lyophilized. The lyophilizate from each preparation was dissolved in 0.05 M phosphate/0.15 M NaC1 (pH 7.0) containing 42% (w/w) CsC1 at a loading density of 1.43 g / m l and centrifuged for 48 h at 12°C and 46000 rpm in a Beckman 50Ti rotor [12]. The resultant gradient was fractionated into 1 ml fractions using a Beckman fraction recovery system and each fraction was assayed for protein and neutral sugar [11]. The fractions containing mucus glycoprotein were pooled, dialyzed against distilled water and lyophilized.

Perfusion The animals, deprived of food for 18 h prior to

Isolation of associated and covalently bound lipids Extraction of associated lipids from the pre-

Experimental procedures

421 pared mucus glycoprotein samples was performed with chloroform/methanol (2 : 1 and 1 : 1, v/v) followed by c h l o r o f o r m / m e t h a n o l / w a t e r (65:35:8, v/v) [13]. The lipids contained in the combined solvent extracts were dried, dissolved in a small volume of chloroform, and individually fractionated on silicic acid columns (0.5 × 15 cm) into neutral lipid, glycolipid and phospholipid fractions [12]. The neutral lipids were separated into individual components by thin-layer chromatography, identified by comparison with chromatograms of authentic standards, and quantitated [14,15]. The glycolipids were chromatographed on a DEAE-Sephadex column (0.5 × 10 cm) into neutral and acidic fractions, separated into individual components by thin-layer chromatography, and quantitated by measuring their carbohydrate and lipid components [14,15]. The phospholipids, eluted from the silicic acid columns with methanol, were separated by two-dimensional thin-layer chromatography, identified by co-chromatography with appropriate standards, and quantitated by measuring their phosphorus content [16]. Removal of the covalently bound fatty acids from the delipidated glycoprotein samples was accomplished by alkali. The glycoprotein was incubated for 30 min at 37 ° C with 0.3 M methanolic KOH and the released fatty acid methyl esters were recovered by three consecutive extractions with hexane [13].

Analytical methods The output of HCO~- in the instillates recovered from proximal and distal duodenal loops was measured by backtitrating the samples to the initial pH level (6.0) with 0.05 M HC1 [10]. Prostaglandin generation was assayed in the instillates by radioimmunoassay procedure [17], using a P G / R I A Kit (Seragen). The transmucosal potential difference was measured with a high-input impedance voltmeter connected with matched calomel electrodes immersed in saturated KCI solution. The potential difference bridges were 80 cm long polyethylene tubing filled with saturated KCI in 4% agar. The recording electrode was placed in the fluid bathing the mucosa of the loop, and the reference electrode was connected to the peripheral vein [18]. The protein content of sam-

pies was measured by the method of Lowry et al. [19], and sulfate was measured turbidimetrically [20]. The phenol/H2SO 4 method was used for monitoring the carbohydrate in column and density gradient fractions. The content and composition of carbohydrates in various glycoprotein preparations was determined by gas-liquid chromatography following methanolysis, re-N-acetylation, and derivation with silylating reagent [11,12]. Gas-liquid chromatography analyses of methyl glycosides, glycerol and fatty acid methyl esters were performed on 3% SE-30 columns [11,15]. Gel electrophoresis in 1% SDS was performed in 4% polyacrylamide gels [21]. Samples of glycoprotein (250-300 #g) were incubated for 3 min at 100*C in the sample buffer (pH 6.8) devoid of flmercaptoethanol and then applied to the gels. After electrophoresis, the gels were stained for protein with Coomassie brilliant blue, and for carbohydrate by the periodate-Schiff method [22]. Results of experiments were expressed as means + S.D. Student's t-test was used to test significance, and P values of 0.05 or less were considered as significant. Results

Alkaline output and prostaglandin generation Measurements of transmucosal potential difference of the loops exposed to saline or studied concentrations of arachidonic acid indicated that the values remained virtually unchanged throughout the entire period of experiments. The potential difference for proximal duodenum averaged - 9 . 8 mV and -4.1 mV for distal duodenum. Under basal conditions, the HCO~- output of proximal duodenum was about 2-times higher than that of distal duodenum. The proximal duodenum also exhibited 1.7-times greater content of prostaglandin (Table I). Topical application of arachidonic acid led to concentration-dependent increase with the output of HCO~ and prostaglandin generation. With 12.5 pg/ml arachidonic acid, the HCO~output in proximal and distal duodenum increased about 2-fold while the prostaglandin content increased by 50 and 33%, respectively. At the highest concentration of arachidonic acid (100 #g/ml), the HCO~ output increased about 3-fold, and the content of prostaglandin by 4-fold in the proximal

422 TABLE I A L K A L I N E O U T P U T A N D P R O S T A G L A N D I N (PGE2) C O N T E N T OF A R A C H I D O N I C A C I D P E R F U S A T E S F R O M D O G P R O X I M A L A N D DIST AL D U O D E N U M

Each value represents the mean-+ S.D. of analyses performed on three perfusates from each of five dogs. * P < 0.01 ; ** P < 0.001 as compared with control. Type of perfusion

Saline Saline + indomethacin Arachidonic acid (/~ g/ml) 12.5 25.0 50.0 100.0

Indomethacin + arachidonic acid (12.5 # g / m l )

Proximal duodenum

Distal duodenum

HCO 3 ( # m o l / 3 0 min)

PGE 2 ( n g / 3 0 min)

105 _+16 56 + 6

180_+ 23 45 _+ 7

198-+10"* 1 9 5 + 1 2 ** 227-+19 ** 283-+21 **

275-+42" 390-+48 ** 542_+77 ** 712-+84 **

61 _+ 6

HCO 3 (~ m o l / 3 0 min) 57 _+12 36 _+ 5

duodenum and 4.8-fold in the distal duodenum. Pretreatment with indomethacin led to a 47% reduction in the basal HCO 3- output in the proximal duodenum and a 37% reduction in HCOf output in the distal duodenum, while the content of prostaglandin decreased by 75% and 61%, respectively. Indomethacin also caused reduction in the HCO 3 and prostaglandin production evoked by arachidonic acid (Table I).

Mucus glycoprotein content Gel filtration of the instillates recovered with saline and arachidonic acid experiments from proximal and distal duodenal loops, followed by CsC1 equilibrium density gradient centrifugation, afforded mucus glycoprotein preparations which on SDS-polyacrylamide gel electrophoresis gave in each case single bands on staining for carbohydrate and protein. Under basal conditions, the output of mucus glycoprotein from proximal and distal duodenum was similar and averaged about 10 mg/100 mg of the dialyzed and lyophilized instillate (Table II). Exposure of the mucosa to arachidonic acid produced also in each case a similar increase in the mucus glycoprotein content of the recovered instillates. With 12.5 # g / m l arachidonic acid, a 2-fold increase in the glycoprotein output occurred in both areas of the duodenum, while an approximately 3-fold increase in the glycoprotein output was observed at 100

107 + 12 42-+ 10

110_+ 6 " * 112-+11 ** 154_+12 ** 174_+18 **

108_+ 17

PGE 2 ( n g / 3 0 min)

143+12" 225_+26 ** 3 2 2 + 4 3 ** 514_+76 **

40_+ 5

83 + 14

/~g/ml of arachidonic acid. Pretreatment with indomethacin caused a slight reduction in mucus glycoprotein content of the saline instillates from both regions, and completely prevented the increments evoked by arachidonic acid (Table II).

Mucus glycoprotein composition Table III shows the composition of mucus glyT A B L E II E F F E C T OF A R A C H I D O N I C A C I D ON T H E M U C U S G L Y C O P R O T E I N O U T P U T IN D O G P R O X I M A L A N D D IS TA L D U O D E N U M Each value represents the mean_+ S.D. of analyses performed on three perfusates from each of five dogs. * P < 0.001 as

compared with control. Type of perfusion

Mucus glycoprotein ( m g / 1 0 0 mg of dialyzed

and lyophilized perfusate) proximal duodenum Saline 10.1 + 1.2 Saline + indomethacin 9.5 + 0.9 Arachidonic acid (/~g/ml) 12.5 25.0 50.0 100.0

20.6__+2.2 * 23.3_+2.1 * 25.8 -+ 2.6 * 32.9+3.3 *

Indomethacin + arachidonic acid (12.5 # g / m l ) 11.3 + 1.2 *

distal duodenum 9.5 + 0.7 9.1 + 0.7 18.4_+2.1 21.5+2.2 23.9 _+2.5 31.8_+3.0

* * * *

10.6__+0.9 *

423 TABLE III CHEMICAL COMPOSITION OF MUCUS GLYCOPROTEIN FROM ARACHIDONIC ACID PERFUSATES OF DOG PROXIMAL AND DISTAL DUODENUM Each value represents the mean+ S.D. of analyses performed on three perfusates from each of five dogs. * P < 0.05; ** P < 0.01; • ** P < 0.001 as compared with control. Location and type of perfusion Proximal duodenum saline saline + indomethacin arachidonic acid (/~g/ml) 12.5 25.0 50.0 100.0 indomethacin + arachidonic acid (12.5/~g/ml) Distal duodenum saline saline + indomethacin arachidonic acid (/~g/ml) 12.5 25.0 50.0 100.0 indomethacin + arachidonic acid (12.5/~g/ml)

Component (mg/lO0 mg of glycoprotein) protein

carbohydrate

associated lipids

covalently bound fatty acid

23.1 + 2.4 23.9 + 2.5

59.7 _+6.1 59.8 + 6.3

12.1 _+1.2 11.2 + 1.1

0.3 _+0.1 0.2 -+0.1

19.4 _+2.2 * 17.1 + 2.0 ** 15.8_+1.7 *** 10.1 _+1.2 ***

62.2 + 6.2 62.9 _-4-6.3 62.5_+6.1 65.0 _+6.3

15.5 _+1.3 ** 17.5 + 1.8 ** 21.1_+2.0 *** 23.7 _+2.4 ***

0.3 _+0.1 0.4 _+0.1 0.4_+0.1 0.6 _+0.2

23.6 _+2.4

61.1 _+5.8

12.5 _+1.2

0.2 _+0.1

24.6 + 2.5 25.3 -+2.6

58.9 + 6.0 58.5 _+5.7

11.5 + 1.2 10.8 _+0.9

0.2 -+0.1 0.2 _+0.1

20.8-+2.1 ** 18.5 _+2.1 ** 16.6 _+1.8 *** 12.3 _+1.4 ***

60.1-+5.7 61.0 _+5.9 62.5 _+6.1 64.1 _+6.3

14.1_+1.2 ** 16.3 _+1.4 ** 18.4_+1.8 *** 22.1 _+2.3 ***

0.3_+0.1 0.3 _+6.1 0.4_+0.1 0.5 _+0.2

24.0 + 2.5

59.3 _+6.1

12.0 _+6.1

0.2 _+0.1

coproteins from saline a n d arachidonic acid perfusates of dog proximal a n d distal d u o d e n a l loops. T h e glycoproteins from b o t h regions exhibited similar c o m p o s i t i o n a n d were comprised of protein, carbohydrate, associated lipids a n d covalently b o u n d fatty acids. L u m i n a l application of a r a c h i d o n i c acid led to c o n c e n t r a t i o n - d e p e n d e n t changes in the p r o p o r t i o n of the glycoprotein constituents. I n b o t h cases, arachidonic acid caused a n increase i n the c o n t e n t of associated lipids a n d covalently b o u n d fatty acids of the glycoproteins, a n d a decrease i n the c o n t e n t of protein. Less a p p a r e n t differences were observed in the c o n t e n t s of carbohydrates. C o m p a r e d to saline control, the glycoprotein from proximal d u o d e n u m showed a 28% increase i n associated lipid c o n t e n t with 12.5 / ~ g / m l of arachidonic acid a n d 96% increase with 100 # g / m l arachidonic acid, while 23% a n d 92% increases i n the associated lipids were f o u n d in the glycoprotein from distal d u o d e n u m . The highest

c o n c e n t r a t i o n of arachidonic acid (100 # g / m l ) in the instillates also evoked in the glycoproteins from b o t h regions at least a 2-fold increase i n the c o n t e n t of covalently b o u n d fatty acids. I n b o t h cases the highest c o n c e n t r a t i o n of arachidonic acid i n the instillates p r o d u c e d m u c u s glycoproteins with a significantly ( P < 0.001) lower c o n t e n t of protein. W i t h b o t h proximal a n d distal d u o d e n u m , the changes i n m u c u s glycopr0tein c o n t e n t of proteins a n d lipids evoked b y arachidonic acid were prevented w h e n the a n i m a l s were pretreated (intravenously) with i n d o m e t h a c i n .

Changes in carbohydrate and lipids T h e c a r b o h y d r a t e c o m p o s i t i o n of m u c u s glycoproteins from p r o x i m a l a n d distal d u o d e n u m is given i n T a b l e IV. T h e m u c u s glycoproteins of b o t h regions exhibited similar c a r b o h y d r a t e comp o s i t i o n which consisted of fucose, galactose, Nacetylgalactosamine, N-acetylglucosamine and

424

TABLE IV CARBOHYDRATE COMPOSITION OF MUCUS GLYCOPROTEIN FROM ARACHIDONIC ACID PERFUSATES OF DOG PROXIMAL AND DISTAL DUODENUM

Each value represents the mean ± S.D. of analyses performed on three perfusates from each of five dogs. * P < 0.01; ** P < 0.001 as compared with control. Location and type of per±us±on Pro~maldu~enum saline saline+indomethacin arac~domcacid(~g/ml) 12.5 25.0 50.0 1~.0 indomethacin + arac~domc acid (12.5 ~g/ml) Distal d u ~ e n u m sMine saline+indomethacin arac~domcacid(~g/ml) 12.5 25.0 50.0 1~.0 indomethacin+arac~domc acid(12.5~g/ml)

Carbohydrate (mg/lO0 mg glycoprotein) Fuc

Gal

GalNAc

GlcNAc

NeuAc

SO2

8.3±0.7 8.0±0.6

14.9±1.1 15.8±1.2

16.7±1.2 16.2±1.2

13.8±1.1 14.1±1.1

3.9±0.3 3.7+0.3

2.1+0.2 2.0+0.1

8.5±0.7 8.2±0.6 8.4±0.6 8.5±0.7

16.3±1.5 17.1±1.6 15.1±1.4 16.5±1.4

16.5±1.3 16.9±1.3 16.4±1.2 16.7±1.5

14.0+1.1 13.7+1.2 14.5±1.1 14.5±1.2

4.3±0.3 4.5+0,4 5.2+0.4 * 5.7+0.5 *

2,6±0.2* 2,5+0.2* 2,9±0.3** 3,1+0.3"*

8.2±0.7

16.4±1.6

16.4±1.3

14.3+1.2

3.7±1.2

2,1±0.1

8.0±0.6 7.2±0.5

14.5±1.2 14.7±1.6

16.3±1.4 16.0±1.5

13.5+1.1 14.7±1.5

4.2±0.3 3.8±0.4

2.4±0.2 2.1±0.2

8.5±0.7 8.3±0.7 8.7±0.9 8.5±0.9

13.8±1.4 13.3±1.5 15.5±1.6 15.6±1.5

16.2±1.5 16.7±1.6 15.9±1.4 16.5±1.6

14.2±1.3 14.7±1.5 13.8±1.4 14.2±1.5

4.6+1.3 4.9±0.5 5.5±0.5 * 5.8±0.6 *

2.9±0.2" 3.1+0.3" 3.1±0.3" 3.5±0.3*

7.8±0.6

14.7±1.4

16.2±1.5

14.4±1.5

3.9+0.4

2.3±0.2

sialic acid. Both glycoproteins also contained comparable amounts of sulfate. The glycoproteins elaborated in response to the topical application of arachidonic acid, especially at its higher concentrations, showed significant increases in the contents of sialic acid and sulfate, but less evident differences were detected in other carbohydrates. The increments in sialic acid and sulfate content of mucus glycoprotein from both regions caused by arachidonic acid were abolished by indomethacin pretreatment. Table V shows the effect of arachidonic acid application of the composition of lipids associated with duodenal mucus glycoproteins. Under basal conditions and with arachidonic acid, the lipids associated with mucus glycoproteins from proximal and distal duodenum were comprised of neutral lipids, glycolipids and phospholipids. The neutral lipids consisted of free fatty acids, cholesterol and its esters, and glycerides, and their proportions remained essentially unchanged with the

type of per±us±on (Table VI). The phospholipids in all preparations were rich in phosphatidylcholine, phosphatidylethanolamine and sphingomyelin, while the glycolipids consisted of neutral and sulfated glyceroglucolipids (about 95%), and of simple glycosphingolipids (mainly glucosyl- and lactosylceramides). While the mucus glycoproteins from proximal and distal duodenum exhibited similar composition of the major lipid classes, considerable changes were observed with the application of arachidonic acid (Table V). Compared to saline controls, the arachidonic acid-elicited mucus glycoproteins showed an increase in the proportions of phospholipids and a decrease in neutral lipids, while the levels of glycolipids remained essentially unchanged. The changes in the proportions of lipid classes were most pronounced at the highest concentration of arachidonic acid (100 #g/ml), at which concentration an approximately 32% increase in phospholipids and 23% decrease in neu-

425 TABLE V E F F E C T OF A R A C H I D O N I C A C I D ON T H E LIPID C O M P O S I T I O N OF M U C U S G L Y C O P R O T E I N P R O X I M A L A N D DISTAL D U O D E N U M

FROM DOG

Each value represents the mean 4- S.D. of analysis performed on three perfusates from each of five dogs. * P < 0.05; ** P < 0.01 as compared with control. m g / l O 0 mg of total lipids

Location and type of perfusate Proximal duodenum saline saline + indomethacin arachidonic acid (# g / m l ) 12.5 25.0 50.0 100.0 indomethacin + arachidonic acid (12.5 # g / m l ) Distal d u o d e n u m saline saline + indomethacin arachidonic acid (~ g / m l ) 12.5 25.0 50.0 100.0 indomethacin + arachidonic acid (12.5 t~g / m l )

neutral

glycolipids

phospholipids

43.4 + 4.2 45.7 +_4.8

31.3 _+2.9 31.0 4- 2.9

25.3 + 2.2 23.3 4- 2.1

39.4 + 4.1 36.8 + 3.8 * 34.4+3.2 ** 33.4+3.2 **

31.8 + 2.8 32.6 ___2.9 33.2+2.9 33.1+3.1

28.7 + 2.5 30.6 4- 2.6 32.4+2.6 33.5+2.7

41.8 4- 3.9

32.5 _+3.0

25.7 4- 2.4

44.3 4- 4.2 45.2 + 4.3

31.1 + 2.8 31.8 _+2.9

24.6 + 2.3 23.0 4- 2.1

40.2 _+3.9 36.3 _+3.3 ** 35.4 4- 4.3 ** 34.1+3.5 **

32.7 4- 2.8 33.9 + 3.1 34.4 4- 3.2 33.24-3.1

27.1 4- 2.4 29.8 + 2.5 * * 30.2 + 2.6 '~* 32.74-2.6 **

42.0 4- 4.1

32.9 4- 3.1

25.1 + 2.3

* * ** **

T A B L E VI N E U T R A L LIPID C O M P O S I T I O N OF M U C U S G L Y C O P R O T E I N F R O M D O G P R O X I M A L A N D DISTAL D U O D E N U M Each value represents the mean 4- S.D. of analyses performed on three perfusates from each of five dogs. FFA, free fatty acids; C, cholesterol; CE, cholesteryl esters; MG, monoglycerides; DG, diglycerides; TG, triglycerides. Location and type

m g / 1 0 0 mg of neutral lipids

of perfusion

FFA

C

CE

MG + DG

TG

49.5 + 5.1

21.5 __.2.3

8.9 _+1.0

0.5 4- 0.2

19.6 + 2.2

50.4 4- 5.3 51.2 _+5.3

21.8 4- 2.3 19.1 _+2.1

8.5 4- 0.9 9.1 4-1.0

0.6 _+0.2 0.5 4- 0.2

18.7 + 1.9 20.1 4- 2.2

49.7 _+5.2

22.6 4- 2.4

8.3 _+0.9

0.4 4- 0.1

19.0___2.1

52.3 4- 5.4

19.7 _+1.1

7.2 _+0.8

0.4 +_0.1

20.4 +_2.0

50.6 _+5.5 51.1 _+5.4

22.0 + 2.3 20.9 + 2.3

8.0 + 0.9 8.8 + 1.0

0.4 + 0.1 0.6 + 0.2

19.0 + 2.1 18.6 + 1.9

52.0 4- 5.5

19.6 + 2.1

7.9 + 1.1

0.5 + 0.1

20.0 + 2.2

Proximal d u o d e n u m saline arachidonic acid (/x g / m l ) 12.5 100.0 indomethacin + arachidonic acid (12.5/~g/ml) Distal d u o d e n u m saline arachidonic acid (# g / m l ) 12.5 100.0 indomethacin + arachidonic acid (12.5 # g / m l )

426

tral lipids occurred in both duodenal regions. Pretreatment with indomethacin had no apparent effect on the contents of glycolipids, but caused a reduction in phospholipids and increase in neutral lipids to the levels found under basal conditions. Discussion

Among the regions of gastrointestinal tract challenged by various noxious agents, the most severely tested is duodenum, in particular its proximal portion, the lumen of which confronts the changing tides of gastric acid and biliary and pancreatic secretions [8,10,23,24]. This hostile environment puts considerable demand on the mechanism responsible for the protection of duodenal epithelium. Consequently, the brunt of severe insult falls on the mucus layer which constitutes the first line of mucosal defense. To cope with this situation, the duodenal mucosa and particularly its proximal portion is equipped with an efficient system for acid disposal. Indeed, studies with several animal species indicate that the rate of H C O 3- secretion in duodenum is considerably higher than in stomach and other intestinal areas [8-10,25]. It is estimated that about 50-70% of the duodenal acid load undergoes neutralization by the HCO 3 secreted from the surface epithelial cells of the duodenum [26,27]. Efficient utilization of the secreted HCO~- requires a large surface area and a hydrated milieu to provide for a slow diffusion. Such a mixing medium is supplied by the layer of mucus gel, the elaboration of which as H C O 3 secretion remains under prostaglandin control [7-10,25]. While prostaglandins have been shown to elicit a rapid HCO3 and mucus secretion, and increase in thickness of the mucus layer [28-30], the changes in composition of the mucus evoked by prostaglandins are not well explored. As the component of mucus responsible for its protective functions is mucus glycoprotein, the composition and content of this glycoprotein secreted by the proximal and distal duodenum in response to endogenous prostaglandins generated by the luminal application of arachidonic acid were examined. The data obtained herein with dog duodenal loops revealed that the output of HCO 3- and prostaglandin content in proximal duodenum,

which more frequently comes in contact with acid, is considerably greater than in the distal duodenum. These values are much higher than those obtained for fundic or antral stomach areas of the same animal [10]. Although both proximal and distal duodenum responded to luminal arachidonic acid application by increased prostaglandin generation and H C O f secretion, the response by proximal duodenum was consistently greater. This suggests that the proximal portion of duodenum has more active prostaglandin metabolism than the distal portion. The proximal duodenum also showed somewhat higher but not significant increase in the mucus glycoprotein output. Both areas of duodenum responded to arachidonic acid by increased production of mucus glycoprotein, the output of which was concentration-dependent and at 100 # g / m l of arachidonic acid reached a value 3-fold that of basal. The increased production of mucus glycoprotein in response to prostaglandin was also reported for gastric mucosa [31]. The duodenal mucosa, like that in stomach, also responded to prostaglandin synthesis inhibitor, indomethacin, by completely arresting the arachidonic acid-induced mucus glycoprotein output [32,33]. The increased mucosal prostaglandin levels evoked considerable changes in the composition of the elicited mucus glycoproteins. In both areas of the duodenum these changes were reflected in the glycoprotein content of protein, associated lipids and covalently bound fatty acids. Mucus glycoproteins elaborated in response to arachidonic acid exhibited a significantly lower content of protein, and higher contents of associated and covalently bound lipids than the glycoproteins secreted under basal conditions. That these changes were brought about by prostaglandin was demonstrated by the results of the experiments with indomethacin, which was capable of abolishing completely the effect of arachidonic acid. As associated lipids and covalently bound fatty acids strongly influence such functional properties of mucus glycoproteins as viscosity, retardation of acid diffusion, regulation of peptic aggression and control of bacterial proliferation [4,6,13,34], the prostaglandin-induced increases in the mucus glycoprotein lipid contents may be of direct relevance to the ability of duodenum to meet successfully

427

the hostile environment to which it is continuously exposed. The lipid-rich mucus glycoprotein could impart greater resistance to the mucus layer against the diffusion of acid and other noxious agents. That hydrophobicity of the mucosal surface plays an important role in acid impedance has been demonstrated by studies of the mucosa with fluorescent probes and by goniometric procedures [35,36]. The changes in composition of the duodenal mucus glycoproteins due to elevated levels of prostaglandin generation were also reflected in the individual carbohydrate and lipid Components. The carbohydrates of mucus glycoproteins from both areas showed increased content of sialic acid and sulfate, while the associated lipids were significantly enriched in phospholipids and exhibited lower levels of neutral lipids. The fact that the changes in carbohydrate and lipid composition of the glycoproteins evoked by arachidonic acid were abolished by the pretreatment with indomethacin points towards the involvement of prostaglandins in the regulation of mucus glycoprotein synthesis. Indeed, the stimulation by prostaglandin of the synthesis of the protein core of mucin, its acylation with fatty acids and glycosylation have been recently recognized [29,37]. The increase in sialic acid and sulfate content of mucus glycoproteins would certainly improve the integrity of mucus gel, as these acidic groups are known to enhance the resistance of gastrointestinal mucins to proteolytic degradation [38,39]. The stimulation of the mucus glycoprotein acylation with fatty acids, which occurs in the regions devoid of carbohydrates, could offer the glycoprotein greater ability for interaction with mucus gel lipids, particularly phospholipids, which are known to interact with the nonglycosylated regions of mucins [12,40]. Taken together, the features imparted by prostaglandin to the mucus glycoproteins certainly would benefit the protective qualities of the mucus gel. Since the mucosal response to HC1 irritation is mediated by a prostaglandin mechanism [9,10,25], the frequent confrontation of the duodenal mucosa with gastric acid could actually maintain the protective quality of elaborated mucus at its best.

Acknowledgement This work was supported by USPH Grant AM No. 21684-09 from the National Institute of Diabetes and Digestive and Kidney Diseases and Grant HL No. 32553-02 from the National Heart, Lung and Blood Institute, NIH.

References 1 Glass, G.B.J. and Slomiany, B.L. (1977) in Mucus in Health and Disease (Elstein, M. and Parke, D.V., eds.), pp. 311-347, Plenum Press, New York 2 Allen, A. (1981) in Physiology of the Gastrointestinal Tract (Johnson, L.R., ed.), Vol. I, pp. 153-179, Academic Press, New York 3 Slomiany, B.L., Piasek, A. and Slomiany, A. (1985) Scand. J. Gastroenterol. 20, 1191-1196 4 Sarosiek, J., Slomiany, A., Takagi, A. and Slomiany, B.L. (1984) Biochem. Biophys. Res. Commun. 118, 523-531 5 Fostner, J.F. (1978) Digestion 17, 234-263 6 Murty, V.L.N., Sarosiek, J., Slomiany, A. and Slomiany, B.L. (1984) Biochem. Biophys. Res. Commun. 121,521-529 7 Flemstrom, G. and Garner, A. (1982) Am. J. Physiol. 242, G183-193 8 Flemstrom, G., Kivilaakso, E., Briden, S., Nylander, O. and Jedstedt, G. (1985) Dig. Dis. Sci. 30, $63-68 9 Miller, T.A. (1983) Am. J. Physiol. 245, G601-623 10 Konturek, S.J., Bilski, J., Tasler, J. and Laskiewicz, J. (1984) Am. J. Physiol. 247, G149-154 11 Slomiany, B.L., Aono, M., Murty, V.L.N., Piasek, A. and Slomiany, A. (1984) J. Appl. Biochem. 6, 308-313 12 Witas, H., Slomiany, B.L., Zdebska, E., Kojima, K., Liau, Y.H. and Slomiany, A. (1983) J. Appl. Biochem. 5, 16-24 13 Slomiany, A., Jozwiak, Z., Takagi, A. and Slomiany, B.L. (1984) Arch. Biochem. Biophys. 229, 560-567 14 Slomiany, A., Yano, S., Slomiany, B.L. and Glass, G.B.J. (1978) J. Biol. Chem. 253, 3785-3791 15 Slomiany, B.L., Murty, V.L.N., Aono, M. and Mandel, I.D. (1983) J. Dent. Res. 62, 866-869 16 Lowry, O.H. and Tinsley, I.J. (1974) Lipids 9, 491-492 17 Konturek, S.J., Mikos, E., Pawlik, W. and Walus, K. (1979) J. Physiol. 286, 15-28 18 Sarosiek, J., Slomiany, B.L., Kojima, K., Swierczek, J., Slomiany, A. and Konturek, J. (1983) J. Appl. Biochem. 5, 429-436 19 Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J. (1954) J. Biol. Chem. 193, 265-275 20 Clarke, A.E. and Denborough, M.A. (1971) Biochem. J. 121,811-817 21 Laemmli, U.K. (1970) Nature 227, 680-685 22 Fairbanks, G., Steck, T.L. and Wailach, D.F.H. (1971) Biochemistry 10, 2606-2617 23 Konturek, S.J. (1985) Dig. Dis. Sci. 30, S105-108 24 Slomiany, A., Siomiany, B.L. and Glass, G.B.J. (1980) J. Appl. Biochem. 2, 336-341

428 25 Johansson, C. and Bergstrom, S. (1982) Scand. J. Gastroenterol. 17, Suppl. 77, 21-46 26 Harmon, J.W., Woods, M. and Gurll, N.J. (1978) Am. J. Physiol. 235, E629-698 27 Dorricot, N.J., Green, F. and Silen, W. (1975) Am. J. Physiol. 228, 269-275 28 Bolton, J.P., Palmer, D. and Cohen, M.M. (1978) Am. J. Dig. Dis. 23, 359-364 29 Robert, A. (1985) in Peptic Ulcer Disease: Basic and Clinical Aspects (Nelis, G.F., Boeve, J. and Misiewicz, J.J., eds.), pp. 297-328, Martinus Nijhoff Publishers, Boston 30 Bickel, M. and Kauffman, G.L. (1981) Gastroenterology 80, 770-775 31 Lamont, J.T., Ventola, A.S., Maull, EA. and Szabo, S. (1983) Gastroenterology 84, 306-313

32 33 34 35 36 37 38 39 40

Rainsford, K.D. (1978) Biochem. Pharmacol. 27, 877 885 Parke, D.V. (1978) Br. Med Bull. 34~ 89-94 Beachy, E.H. (1981) J. Infect. Dis. 143, 325-345 Gwozdzinski, K., Murty, V.L.N., Liau, Y.H., Slomiany, A. and Slomiany, B.L. (1986) Fed. Proc. 45, 1985 Hills, B.A., Butler, B.D. and Lichtenberger, L.M. (1983) Am. J. Physiol. 244, G561-568 Slomiany, A., Takagi, A., Kosmala, M. and Slomiany, B.L. (1985) gastroenterology 88, 1590 Takagaki, Y.M. and Hotta, K. (1977) Biochim. Biophys. Acta 481,631-637 Takagaki, Y.H. and Hotta, K. (1979) Biochim. Biophys. Acta 584, 288-297 Witas, H., Sarosiek, J., Aono, M., Murty, V.L.N., Slomiar~y, A. and Slomiany, B.L. (1983) Carbohydr. Res. 120, 67-,'6