Metabolism of the plant sulfolipid—Sulfoquinovosyldiacylglycerol: Degradation in animal tissues

ARCHIVES

OF BIOCHEMISTRY

AND

BIOPHYSICS

Vol. 259, No. 2, December, pp. 510-519,1987

Metabolism

of the Plant Sulfolipid- Sulfoquinovosyldiacylglycerol: Degradation in Animal Tissues’ SITA D. GUPTA AND P. S. SASTRY’

Department

of Biochemistry,

Indian

Institute

of Science, Bangalore-s-h’0 012, India

Received May 27, 1987, and in revised form August 12, 1987

Metabolism of the plant sulfolipid-sulfoquinovosyldiacylglycerol (SQDG)-was studied in animal tissues. In vivo experiments with [35S]SQDG in guinea pigs showed that this lipid is not absorbed intact in the gastrointestinal tract. In these experiments, 3 h after administration of [35S]SQDG, the intestinal mucosa contained 1 to 5% of the radioactivity as SQDG, while the remainder was in a water-soluble form. Analysis of the water-soluble components showed that about 60% of the radioactivity was present as sulfoquinovosylglycerol (SQG) and the remainder was present as free SO:-. In the blood, 99% of the radioactivity was present as SO,“-, SQG was not observed. In liver, only very little radioactivity was observed and appeared to be mainly in the form of SO:-. Experiments with everted intestinal sacs of guinea pigs confirmed the formation of SQG, SO:-, and, in addition, sulfoquinovosylmonoacylglycerol (SQMG) in this tissue. In vitro experiments with saline extracts of acetone powders of pancreas and intestinal mucosa of guinea pig, sheep, and rat showed that [35S]SQDG was deacylated to SQMG (sulfolipase A activity) and SQG (sulfolipase B activity). It is concluded that animal tissues deacylate SQDG in a stepwise manner to SQG. It is further metabolized to yield free SO:- by cleavage of the C-S bond which appears to be brought about by the intestinal microflora. Sheep pancreatic sulfolipases were characterized. Bile salts, sodium dodecyl sulfate, and Triton X-100 inhibited the pancreatic sulfolipases, while CaClz activated them. Substrate competition experiments and investigations on substrate specificity with a partially purified preparation indicated that relatively specific sulfolipase(s) may exist in pancreas. Among the species tested, guinea pig tissues showed the highest sulfolipase A and B activities followed sheep and rat tissues. Pancreatic enzymes were 18 to 60 times more active than intestinal enzymes. Q 1987 Academic Press, Inc.

1,2-Di-O-acyl-3-O-(6-deoxy-6-sulfo-a-Dglucopyranosyl)-sn-glycerol (SQDG)3 is a

unique lipid as it possesses a stable C-S bond. It is found in photosynthetic bacteria, algae, and higher plants. It accounts for about 10% of the total glycolipids in leaves (1) and contains large amounts of palmitic and linolenic acids (2,3). SQDG is ingested by herbivores through their diet; therefore, a study of absorption and metabolism of SQDG in the gastrointestinal tract is of interest. The catabolism of sulfatides, compounds that contain a C-O-S bond, has been studied in animal tissues, and sulfatases were shown to hydrolyze the sulfate ester bonds (4). In contrast,

’ This work was supported by the Council of Scientific and Industrial Research, India. ’ To whom all correspondence should be addressed. 3 Abbreviations used: DGDG, digalactosyldiacylglycerol; HVPE, high-voltage paper electrophoresis; PPO, 2’,5’-diphenyloxazole; LPC, lysophosphatidylcholine; MGDG, monogalactosyldiacylglycerol; PC, phosphatidylcholine; PE, phospbatidylethanolamine; SDS, sodium dodecyl sulfate; SQ, sulfoquinovase; SQDG, sulfoquinovosyldiacylglycerol; SQG, sulfoquinovosylglycerol; SQMG, sulfoquinovosylmonoacylglycerol; TAG, triacylglycerol. 0003-9861/8’7 $3.00 Copyright All rights

0 1987 by Academic Press, Inc. of reproduction in any form reserved.

510

METABOLISM

OF SIJLFOQUINOVOSYLDIACYLGLYCEROL

cleavage of the C-S bond, as exists in SQDG, has not been demonstrated; however, in the lower plants such as Scenedesmus obliquus (3) and Chlorella pyrenoidosa (5) as well as in higher plants (6-g), SQDG was shown to be deacylated to SQG. There are some reports suggesting further degradation to SQ, sulfolactaldehyde, and sulfoacetate in plants (10, 11). Surprisingly, except for a brief report that SQDG is susceptible to pancreatic enzymes (3), the metabolism of SQDG in the gastrointestinal tract has not been studied. Here, we report the results of our experiments on the metabolism of SQDG in the gastrointestinal tract as well as on the action of pancreatic enzymes on SQDG. MATERIALS

AND

METHODS

Sources of chemicals. a5S as sulfuric acid (carrierfree in dilute hydrochloric acid) was purchased from Bhabha Atomic Research Centre, Bombay, India. Florisil (SO-100 mesh), DEAE-cellulose, succinic acid, PPO, sodium deoxycholate, sodium tauroeholate, Triton X-100, and bovine serum albumin were purchased from Sigma Chemical Company. Silica gel G containing 1.3% CaS04. f Hz0 for TLC was obtained from ACME Chemicals, Poona, India. Animals and their tissues. Fresh sheep pancreas and small intes,tines were obtained from a slaughterhouse and transported to the laboratory in ice. Guinea pigs and albino rats (Indian Institute of Science strain) were decapitated under light anesthesia. Their small intestines and pancreas were quickly removed and kept in ice-chilled beakers. The intestines were washed with ice-cold physiological saline mucosa was scraped with the (0.9%): the intestinal help of a glass slide and collected in previously weighed chilled beakers. Preparation of[“S]SQDG. Young, tender leaves of groundnut (Arachis hvpogaea) plants grown under field conditions were collected 25-40 days after planting. The total lipids from 22.5 g of groundnut leaves were extracted (12) and a lipid concentrate of 302.4 mg was obtained. This was mixed with a 35S-labeled total lipid extract obtained as follows: 2.5 g of groundnut leaf disks (5 mm in diameter) were incubated in batches of 100 mg in 2.0 ml of distilled water containing 100 &i of carrier-free ?SOi- neutralized with dilute NaOH, for 5 h at 30°C. The disks were washed thrice with ice-cold 2 mM sodium sulfate solution and the lipids were extracted according to Bligh and Dyer (13). The labeled lipid extract (33.6 mg) contained 8.20 X lo7 cpm. From the combined preparations, [?S]SQDG was purified by the twostep column chromatographic procedure, first on

511

Florisil and then on DEAE-cellulose, as described by O’Brein and Benson (14). The preparation contained only SQDG and gave a single iodine-positive and radioactive spot on two-dimensional TLC (first N NHa, 65:30:4, direction-chloroform:methanol:7 v/v; second direction-chloroform:methanol:acetie acid:water, 170:25:25:4, v/v). This procedure usually gave an average yield of 16 pmol of [35S]SQDG with a specific activity of 2500-3000 cpm/nmol. When [35S]SQDG of a higher specific activity was required, the leaf disks were incubated with ?JOi- (200-500 &i/100 mg) and it was purified as described above. Preparation of [“S]SQMG, [S5S]SQG, and [“SISQ. [3”S]SQMG was prepared by the enzymatic hydrolysis of [35S]SQDG with sheep pancreatic acetone powder extracts as described later. [35S]SQG was prepared by methanolic alkaline hydrolysis of [%S]SQDG, essentially as described by Benson et al. (15). A single product was obtained that showed an Rf of 0.44 when subjected to ascending paper chromatography on Whatman No. 3 filter paper with phenol:water:acetic acid:ethanol(100:50:15:18, v/v) as the solvent system. [35S]SQ was obtained by ethanolic acid hydrolysis of [35S]SQDG (15) and it gave a single radioactive spot with an R, of 0.29 when chromatographed as described above. In vivo experiments in guinea pigs. [35S]SQDG (959 nmol; 1 X lo7 cpm) in chloroform was mixed with 1 ml groundnut oil and the chloroform was removed in wacuo. Guinea pigs, starved overnight, were administered the groundnut oil containing SQDG directly into the stomach (1 ml per animal) with a feeding tube. At different time intervals, blood (2.5 ml) was collected by cardiac puncture and added to tubes containing 100 ~1 of 2% potassium oxalate and the animals were sacrificed. Liver and small intestine were removed and kept immersed in cold physiological saline. The small intestine was washed gently with saline and cut open longitudinally on a cool glass plate, and the mucosa was scraped with a glass slide. The lipids from blood, mucosa, and liver were extracted (13). For the identification of the methanol-water-soluble products, the methanol-water phase was concentrated under vacuum, treated with Dowex 50W (H+), and electrophoresed on Whatman No. 3 filter paper using pyridine-acetate buffer, pH 3.5, at 2 kV for 60 min. The radioactive spots were located by autoradiography and identified with the help of standard compounds spotted on the adjacent lanes. Experiments with everted intestinal sacs of guinea pigs. Guinea pigs were anesthetized with ether and sacrificed. The intestines were quickly removed and washed twice with 50 ml of cold saline; from each intestine a segment of about 22 cm was cut from the point where the bile duct enters the intestine. This was cut into pieces of about 7-8 cm, everted with a thin Teflon tubing, and made into sacs by tying one

512

GUPTA

AND

end as described by Wilson and Wiseman (16). They were filled with 1.5 ml of Ca’+-containing KrebsRinger-phosphate buffer, pH 7.4 (17), and saturated with 02: the other end was then tied carefully. Two such pieces were transferred to a 25-ml Erlenmeyer flask containing [?S]SQDG (293 nmol; 3 X lo7 cpm) in buffer, pH ‘7.4. The 5 ml of Krebs-Ringer-phosphate flasks were stoppered and incubated at 37“C in a metabolic shaker for 1 h. The sacs were then removed and washed with cold physiological saline. The external medium was centrifuged at 5000 rpm for 10 min. The pellet obtained was added to the mucosa. Lipids from the external medium and intestinal mucosa were extracted (13). The methanol-water-soluble products were analyzed by paper electrophoresis as described above. Preparation of acetone powder and extraction of enzyme. Weighed quantities of tissues (pancreas or mucosa) were homogenized with 10 vol of chilled acetone (-20°C). The homogenate was rapidly filtered through a Buchner funnel under vacuum. The residue was again homogenized in 10 vol of chilled acetone and filtered. The final residue was air-dried for 15 min, dried over anhydrous CaCl’ in vacua, and stored in the deep-freeze at -20°C. For extraction of the enzyme(s), the acetone powder (100 mg) was suspended in 2 ml of 0.9% NaCl and kept overnight in the cold (0-5°C). The suspension was centrifuged at 30009 for 10 min and the clear supernatant was used as the enzyme source. Enzyme assay. [?S]SQDG (50-100 nmol; sp act 800 to 2000 cpm/nmol) in chloroform was taken in a test tube and the solvent was evaporated under nitrogen. The lipid was taken up in 10 ~1 of methanol and suspended in 80 ~1 of 0.25 M succinic acid-NaOH buffer, pH 5.0, on a Vortex mixer. Enzyme extract and water were added to a final volume of 0.2 ml. Following incubation at 37’C for the required time, the reaction was terminated by the addition of 0.22 ml of methanol containing 0.1 N HCI. Then, 0.28 ml of methanol and 0.25 ml of chloroform were added and the tubes were left in the cold (0-5°C). After 6-8 h, phase separation was brought about by the addition of either chloroform (0.25 ml) and water (0.25 ml) or chloroform (0.25 ml) and 0.5 M phosphate buffer, pH 7.4, containing 2 M KC1 (0.25 ml) (13, 18). The former procedure was used when the methanol-water-sohble products were analyzed by chromatography, while the latter procedure was used when the products formed were estimated quantitatively. The tubes were centrifuged at 5000 rpm for 5 min. Aliquots of the chloroform and the methanol-water phases were used for determination of radioactivity and for analysis of lipid- and water-soluble products, respectively. Analytical methods. The quantity of SQDG was determined by estimating the sugar content by the phenol-sulfuric acid method (12) with galactose as

SASTRY standard. Protein was estimated according to Lowry et al. (19) with bovine serum albumin as standard. Radioactivity measurement and autoradiography. Aliquots of lipid samples in chloroform were dried in scintillation vials. When the lipids were separated by TLC, silica gel from the area containing the lipid was scraped and transferred directly into the scintillation vials. For aqueous samples, an aliquot was spotted on a 1.5~cm-square piece of Whatman No. 3 filter paper, dried under an infrared lamp for 20 min, and placed in a scintillation vial. When samples were separated on paper chromatograms, areas of paper containing the radioactive spots were cut into small pieces and transferred into scintillation vials. Five milliliters of scintillation fluid (0.5% PPO in toluene) was added and radioactivity was determined using an LKB RackBeta liquid scintillation counter with an efficiency of 98% for a5S. For autoradiography. TLC plates containing the radioactive samples were exposed to X-ray films (Hindustan Photofilms Manufacturing Co. Ltd., Ootacamund, India) for the required length of time. Paper chromatograms and high-voltage paper electrophoregrams were dipped in 5% PPO in ether, kept in cassettes having intensifying screens (Kiran, superfast X-ray screens, 42.5 X 35 cm), and left at -70°C for the required time period. RESULTS

In vivo experiments with guinea pigs. The distribution of radioactivity in tissue lipids and methanol-water products as a function of time after administration of [35S]SQDG is shown in Fig. 1. Of the total radioactivity found in the intestinal mucosa only 1 to 5% was recovered in the lipid constituents. Similarly, the radioactivity found in the lipids of blood and liver tissues was also negligibly small; however, there was a significant amount of radioactivity in the water-soluble products in all the tissues, which increased with time up to 3 h and then declined. The amounts of radioactivity in water-soluble products from intestinal mucosa, blood, and liver at 3 h were 0.68 X 105, 1.3 X 105, and 0.58 X lo5 cpm, respectively, which account for more than 95% of the radioactivity found in these tissues. These experiments suggest extensive metabolism of SQDG prior to absorption in the gastrointestinal tract. An autoradiogram of the electrophoretie pattern of the methanol-water-soluble products obtained for the 3-h sample is shown in Fig. 2. In all the tissues, a radio-

METABOLISM

OF SULFOQUINOVOSYLDIACYLGLYCEROL

513

98.8% of the radioactivity was found as SO:- in blood, and SQG was not observed. The nature of the radioactive compounds found in liver could not be ascertained because of streaking in the electrophoregrams. Experiments

with everted intestinal

sacs.

When [35S]SQDG was incubated with everted intestinal sacs of guinea pigs for 1 h, about 10% of the radioactivity added to the incubation medium was found in the intestinal mucosa and the remaining 90% was recovered from the medium. TLC of the lipid extract of the mucosa showed the presence of SQDG and SQMG in the mucosa. Analysis of the methanol-water phase of the lipid extract of the mucosa by HVPE showed SQG and SO:- as the main components. TLC of the lipid extract of the incubation medium showed SQDG as

6.0 3.6

Time I hr) FIG. 1. Time course of metabolism of [%]SQDG in viva in guinea pigs. Guinea pigs, starved overnight, were given [%]SQDG (959 nmol/l X lo7 cpm) by intubation as described in the text. Lipids were extracted from the intestinal mucosa, blood, and liver. The weights of intestinal mucosa from guinea pigs sacrificed at 1,2,3, and 4 h were 2.2,1.9,1.5, and 2.4 g, respectively. The corresponding weights of liver were 4.3, 4.2, 2.3, and 3.5 g. Blood (2.5 ml) was obtained at all time points. Results are expressed as radioactivity/total weight of the tissue or 2.5 ml of blood. A, Chloroform phase; 0, methanol-water phase.

active spot with electrophoretic mobility almost equal to SOi- was observed. This spot was most prominent in blood. In the intestinal rnucosa another spot with mobility slightly lower than that of the standard SQG was observed. But this compound streaked, probably because of salts, and it is likely that it is SQG. Significantly, this water-soluble product was not observed in either blood or liver. An essentially similar pattern was obtained when the 2-h samples were analyzed. The quantitative estimation of the distribution of radioactivity in the various watersoluble products in the 3-h sample showed that in the intestinal mucosa, about 61% of the radioactivity was in the product which is most likely SQG, while 32.6% was found in inorganic sulfate. In contrast,

M SQG

0

c STD, STD2

IM

B

L

FIG. 2. Autoradiogram of the methanol-water-soluble products in intestinal mucosa, blood, and liver of guinea pigs given [%]SQDG. Experimental details are as described in Fig. 1 Aliquots of the methanolwater phase of the lipid extracts of intestinal mucosa, blood, and liver from the 3-h time point were subjected to HVPE at 2 kV for 1 h using pyridine-acetate buffer, pH 3.5. Standards: STDr, [a”S]SQ; STDB, [%]SQG; %O:-; IM, intestinal mucosa; B, blood; L, liver.

514

GUPTA

AND

the main compound, but a significant amount of SQMG was also present. When the products in the methanol-water phase of the lipid extract of the incubation medium were analyzed by HVPE, SQMG, SQG, nd SO:- were observed as the main components. Thus, the everted sac experiments confirmed the in vivo experiments; however, it must be pointed out that the formation of SO,“- was much higher in the in vivo experiments and it was more consistent than in the everted sac experiments. In later experiments much less radioactivity was recovered as free SOZ-, and in some experiments only negligible radioactivity was found as free SO:-. Degradation of [35S]SQDG by pancreatic and intestinal enzymes: Partitioning of [35S]SQDG and [35S]SQMG in the Bligh and Dyer extraction procedure. In the experments with everted intestinal sacs, significant amounts of SQMG and SQDG were observed to partition into the methanolwater phase during lipid extraction (13). The relative quantities of both lipids that partitioned into the methanol-water phase were time dependent and irreproducible. This phenomenon had been observed earlier by O’Brein and Benson (14); however, use of a high-molarity buffered salt solution (0.5 M phosphate buffer, pH 7.4, containing 2 M KCl) in place of water (18) in the extraction led to a near complete partitioning of SQDG into the chloroform phase as has been observed by Harwood (20). Therefore, this procedure was used in all further experiments on pancreatic and intestinal enzymes. Even with this procedure, a constant amount of SQDG (3.66 t 0.34%) partitioned into the upper phase for which a correction was made with appropriate blanks. In contrast, SQMG partitioned significantly into the upper phase under these conditions. It has been experimentally found that a constant 61% of SQMG partitioned into the upper phase and 39% remained in the chloroform phase when high-molarity buffered-salt solution was used in the extraction. Therefore, to determine the total quantity of SQMG, the amount found in the chloroform phase was multiplied by 2.56 (i.e., 100139).

SASTRY

SQMG

ORIGIN

0

10

30

50

T!ME (rmn)

FIG. 3. Autoradiogram of the TLC of the products of [35S]SQDG hydrolysis by sheep pancreatic acetone powder extract. The incubation mixture consisted of 99.6 nmol of [%]SQDG (sp act 1834 cpm/nmol), saline extract of sheep pancreatic acetone powder (100 pg protein), 100 RIM succinic acid-NaOH buffer, pH 5.0, in a total volume of 0.2 ml. Incubation was carried out at 3’7°C for 10 to 60 min. The reaction was stopped and lipids were extracted as described under Materials and Methods. An aliquot of the chloroform phase (300 ~1) was spotted on TLC and the chromatogram was developed with chloroform:methanol:water:acetic acid (130:70:12:8, v/v).

Characterization of pancreatic enFigures 3 and 4 show the lipidsoluble and methanol-water-soluble products, respectively, by the enzymatic action of the saline extract of sheep pancreatic acetone powder on [35S]SQDG. It is apparent that SQMG is formed in a time-dependent manner. Analysis of the methanolwater-soluble products (Fig. 4) showed the formation of SQG by further hydrolysis of SQMG. It may be noted that SQ is not formed in these experiments. Thus, it is clear that the sheep pancreatic enzyme(s) hydrolyzes both ester bonds in SQDG. The enzyme that hydrolyzes SQDG to SQMG is referred to as sulfolipase A and the enzyme that catalyzes further hydrolysis to SQG is called sulfolipase B. Quantitative data on the time course of C3”S]SQDG degradation are shown in Fig. 5. The amount of SQG formed was computed by determining the radioactivity in SQMG zyme(s).

METABOLISM

SQMG

SGlG

54

STO, 0 10 30

515

OF SULFOQUINOVOSYLDIACYLGLYCEROL

STO*

TIME (min )

FIG. 4. Autoradiogram of the methanol-water-soluble products of [35S]SQDG hydrolysis by sheep pancreatic acetone powder extract. The conditions of incubation were as described in Fig. 3, except that [%S]SQDG (sp act 3339 cpm/nmol) and enzyme protein (200 fig) were used. Partitioning was carried out with water. Aliquots of the methanol-water phase were treated with Dowex 50W (H+), dried under vacuum, taken up in 100 ~1 of 50% aqueous methanol, and spotted on Whatman No. 3 filter paper (54 X 18 cm). Ascending chromatography was done using phenol:water:acetic acid:ethanol(100:50:15:18, v/v) or 40 h. After it was dried, the chromatogram was dipped in 5% I?PO (in ether) and autoradiographed. Standards: STDl, [35S]SQG + r”S]SQ; STDZ; [“5SlSQMG + [=S]SQG + [‘“SISQ.

that partitioned into the upper phase (i.e., radi0activit.y in SQMG in the lower phase X 61/39). Figure 5 shows rapid hydrolysis of ester bonds in SQDG, resulting in the formation of SQMG as well as SQG. The initial rate of disappearance of SQDG was calculated to be 15 nmol/min/mg protein. Sulfolipase A activity showed an initial rate of 12 nmol/min/mg protein while that of sulfolipase B activity was of 7.2 nmol/min/mg protein. Further experiments showed that sulfolipase A and B activities were proportional to enzyme protein concentration up to 100 and 50 /*g, respectively, per reaction digest. Sulfolipase A activity showed two pH optima, at pH 5.0 and pH 7.4 (Fig. 6), and the activity was considerably higher at pH 5.0. In contrast, sulfolipase B did not show

a well-defined pH optimum. This activity was nearly the same in the pH range 4.4 to 6.0 and rose sharply at about pH 6.0. Analysis of the methanol-water-soluble products at pH 7.0 revealed the formation of SQMG and SQG only. Figure 7 shows the effect of substrate concentration on enzyme activities. From the LineweaverBurk plot (Fig. 7 inset), an apparent K, of 0.16 mM with a V,,, of 16 nmol/min/mg protein was calculated for sulfolipase A. Though sulfolipase B activity also showed a substrate dependence on SQDG, its K, and V,,, were not calculated from these data since its true substrate is SQMG. Bile salts and some detergents inhibited the sulfolipases. Sodium taurocholate (0.5 InM) inhibited sulfolipases A and B by 50 and 60%, respectively, while sodium deoxycholate (0.5 mM) inhibited both enzymes by 30%. SDS (0.5 InM) and Triton X-100 (0.5%) inhibited the enzymes by more than 90%. In contrast, CaClz (2.5-5 mM) enhanced sulfolipase A and B activities by 16 and 29%) respectively. Addition

Time lminl FIG. 5 Time course of hydrolysis of [3SS]SQDG by sheep pancreatic acetone powder extract. [3sS]SQDG (87 nmol; sp act 781 cpm/nmol) was incubated at 37°C for various time periods with the enzyme extract (100 pg protein). Other conditions of incubation and extraction were as described under Materials and Methods. Aliquots of the chloroform phase were spotted on TLC and developed as described in Fig. 3.

516

GUPTA

01

I 4

I

I

5

6

I

7

I

a

AND

1

PH FIG. 6. Effect of pH on the hydrolysis of [%]SQDG by sheep pancreatic acetone powder extract. [“%ISQDG (55 nmol; sp act 1293 cpm/nmol) was incubated with 200 fig of enzyme protein at 37°C for 5 min. pH was varied from 4.0 to 8.0 in succinic acidNaOH buffer (100 mM, pH 4-6) or sodium phosphate buffer (100 mM, pH 6-8).

of equimolar concentrations (500 PM) of TAG or PE to the reaction digests resulted in 30 to 35% inhibition of both enzyme activities. Similar experiments with PC and LPC showed much higher inhibition. Thus, PC (500 PM) inhibited sulfolipases A and B by ‘76 and 60%, respectively. LPC (500 FM) was even more effective and showed 94% inhibition of sulfolipase A and 46% of sulfolipase B. Attempts at purification of sulfolipase A activity by (NH&SO4 precipitation (O-60%) followed by DEAE-cellulose column chromatography yielded a 23-fold purified preparation. This preparation required a colipase for maximal activity and showed relative specific activities in the ratio of 1:0.74:1.6with TAG, PC, and SQDG. The rates of sulfolipases A and B in the pancreas of three animal species are compared in Table I. In all species tested (sheep, guinea pig, and rat) both sulfolipase activities are present, but the levels of activity differ widely among the species. In rat, very low sulfolipase A and B

SASTRY

activities were observed, while guinea pig showed the maximal enzyme activities, about 30-fold higher than those in rat. Sheep pancreas showed about IO-fold more activity than rat pancreas. In pancreas of all species, sulfolipase A activity is about 1.4 to 1.8 times higher than sulfolipase B activity. Intestinal enzymes. Similar experiments with saline extract of intestinal mucosa acetone powder showed the presence of sulfolipase A and B activities in this tissue (Table I) but the rates are far below those observed with pancreas. Here again, guinea pig intestinal enzymes are most active, followed by those from sheep and rat; however, in all the species, both sulfolipases A and B from intestinal mucosa exhibited only 2 to 3% of the activity found in pancreas of the same species. Analysis of the water-soluble products formed by the intestinal enzymes of the

ioCyJG 120

360

600

L3% I SQDG I pM 1 FIG. ‘7. Hydrolysis of [‘%]SQDG by sheep pancreatic acetone powder extract: Effect of substrate concentration. The reaction was carried out as described under Materials and Methods. The concentration of [%]SQDG (sp act 2800 cpm/nmol) was varied from 35 to 625 pM and incubated with protein (100 pg) at 37°C for 5 min. The inset shows the LineweaverBurk plot of [35S]SQMG formed.

METABOLISM

517

OF SULFOQUINOVOSYLDIACYLGLYCEROL TABLE

I

COMPARISON OF SULFOLIPASE A AND B ACTIVITIES OF PANCREAS AND INTESTINAL MUCOSA OF VARIOUS ANIMAL SPECIES Pancreas Species

Sulfolipase

A

Intestinal Sulfolipase

B

nmol min-’ Sheep Guinea pig Rat

12.0 39.9 1.3

7.2 21.8 0.9

Sulfolipase

A

mucosa Sulfolipase

B

mg protein’ 0.20 1.06 0.07

0.17 0.91 0.04

Note. The enzymes were assayed as described under Materials and Methods with saline extracts of acetone powder of pancreas and intestinal mucosa. [%]SQDG (92 nmol; sp act 1658 cpm/nmol) and enzyme protein (100 pg, sheep and guinea pig pancreas; 200 pg, rat pancreas; 250 ng, intestinal mucosa of all species) were incubated at 37°C for 15 min in the case of pancreatic enzymes and for 30 min in the case of intestinal mucosa enzymes.

water-soluble products in liver did not yield conclusive results, but it is likely that in liver too, the radioactivity is present mainly as SO:-. From these data, DISCUSSION it appears that SQDG is metabolized comHerbivores ingest large amounts of pletely in small intestine. Studies with lipids through forage crops and green everted intestinal sacs confirmed the revegetables. The lipids derived from these sults obtained from in vivo experiments. Here, in addition to SQG and free SO&sources originate predominantly from leaves and rnake up about 6 to 8% of the ions, the formation of SQMG was also obdry weight. These lipids consist mainly of served. Therefore, in the small intestine, SQDG is deacylated to SQG in a stepwise glycolipids and phospholipids. Glycolipids comprise 70 to 80% complex lipids (21), manner with the intermediate formation of SQMG; however, SQ was not observed and SQDG forms 10% of the total glycolipid in plants (1). Surprisingly, very little is in the degradation products of SQDG and, thus, there is no evidence for a glycosidase known about the digestion and absorption of SQDG in the gastrointestinal tract of activity. The formation of free SOi- is much higher and more consistent in the in animals. Yagi and Benson (3) reported briefly that a commercial preparation of vivo experiments than in the experiments pancreatin hydrolyzed SQDG, but the with everted sacs. It appears likely that products have not been identified. The intestinal microflora may be responsible present studies on the metabolism of [35S]- for the cleavage of C-S bonds. It has been reported that a Butyrivibrio sp. isolated SQDG in W:VOin guinea pigs showed that it is metabolized to water-soluble com- from the ovine rumen rapidly deacylates pounds in the small intestine prior to its SQDG, but its further metabolism was not absorption,, The results further showed investigated (22). However, a bacterial spp. strain belonging to Flavobacterium that SQDG is deacylated in the intestine by the lipolytic enzymes to form SQG. The produces sulfoacetate and sulfate ions formation of SO:- ions indicates that a when grown on methyl 6-[35S]sulfoquinomechanism exists in the intestine for vose (23). cleavage of the C-S bond in SQG. SignifiIn vitro assays using saline extracts of cantly, 99% of the radioactivity in blood sheep pancreatic acetone powder revealed was in the form of SO:- and SQG was not that SQDG was hydrolyzed by the pancreobserved in this tissue. Analysis of the atic enzyme(s). The products formed were three species showed the formation SQG only, as in the case of pancreas.

of

518

GUPTA

AND

identified as SQMG and SQG, which indicate only the lipolytic action of the pancreatic enzyme(s). Although our experiments do not establish whether one enzyme catalyzes the hydrolysis of both acyl ester bonds or two enzymes are present in pancreas, since SQMG was formed as an intermediate, the two activities were termed sulfolipase A and sulfolipase B. The enzymes of SQDG metabolism in Scenedesmus were similarly described by Yagi and Benson (3). The pancreatic enzymes hydrolyzed SQDG at an initial rate of 15 nmol/min/mg protein. The rates of formation of SQMG (sulfolipase A) and SQG (sulfolipase B) were 12 and 7.2 nmol/min/ mg protein, respectively. The rate of deacylation of SQDG by the total homogenate of Chlwella pyrenoidosa was reported as 0.048 nmol/min/mg protein, while it was 0.24 nmol/min/mg protein with the 30,OOOg sediment (5) and 1.4 nmol/min/mg protein with a 104,OOOg supernatant of P. multi&bus (24). Therefore, it appears that the enzyme(s) is considerably more active in pancreas than in plants. Pancreatic sulfolipase A showed two pH optima, pH 5.0 and pH 7.4, with higher activity at the acidic pH, whereas SQG was the main product at pH 7.4. A similar acidic pH optimum was also observed in P. multi&wus (6, 10). In contrast, cell-free preparations of C. pyrenoidosa catalyzed the deacylation of SQDG optimally at pH 8.2, with very little activity below pH 8.0 (5). Pancreatic sulfolipase A showed an apparent K, of 0.16 MM which agrees with the value of 0.15 mM reported for the P. multiflorus enzyme (6). Pancreatic lipases are known to be activated by bile salts and Ca2+ ions (25). Bajwa and Sastry (26) reported activation of ester hydrolysis in both MGDG and DGDG with sheep pancreatic acetone powder extracts; however, there are reports that pancreatic lipase, in the absence of colipase, is strongly inhibited by conjugated bile salts above their critical micellar concentration (27-30). In this study, bile salts inhibited sulfolipases even in crude extracts where colipase was present. SDS and Triton X-100 also inhibited the sulfolipases which agrees with

SASTRY

earlier reports of inhibition of pancreatic lipase by these detergents (31). We observed that Ca2+stimulates the hydrolysis of SQDG by pancreas, but with P. multiflorus (10) and C. pyrenoidosa (5), inhibition was reported. Pancreatic lipases are known to act on a variety of acyl esters. In an effort to determine the specificity of pancreatic sulfolipases, substrate competition experiments were carried out. TAG and PE caused only about 30% inhibition of SQDG hydrolysis. PC and LPC inhibited strongly which may be a result of their detergent property, as all the detergents tested inhibited SQDG hydrolysis. A partially purified preparation showed relatively higher activity toward SQDG but it was still active toward TAG and PC. Thus, the specificity of sulfolipases could not be ascertained, but it would not be surprising if specific enzymes exist since sulfolipids are negatively charged substrates. Sulfolipases were found in the tissues of several species. Guinea pig tissues showed much higher activity than sheep and rat tissues. Similarly, galactosylglyceride acylhydrolase activity was also found to be significantly higher in guinea pig than in sheep or rat tissues (26). REFERENCES 1. BENSON, A. A. (1963) Adv. Lipid Res. 1,387-394. 2. RADUNZ, A. (1969) Hoppe-Seyler’s Z. Physiol. Chem. 350,411-417. 3. YAGI, T., AND BENSON, A. A. (1962) Biochem. Biophys. Acta 57,601-603. 4. FAROOQUI, A. A. (1981) Adv. Lipid Rex 18, 159-202. 5. WOLFERSBERGER,

M. G., AND PIERINGER,

R. A.

(1974) J Lipid Res. 15, l-10. 6. BURNS, D. D., GALLIARD,

(1977) Phytochemistry 7. MATSUDA, AYAMA, 8. MATSUDA,

H., TANAKA,

T., AND HARWOOD,

J. L.

16, 651-654. G., MORITA,

K., AND HIR-

0. (1974) Agr. BioL Chem. 43,563-570. M., AND HIRAYAMA, 0. (1979) B&him. Biophys. Acta 573, 155-165. 9. MATSUDA, H., AND HIRAYAMA, 0. (1979) Agr. BioL Chem 43,463-469. 10. BURNS, D. D., GALLIARD, T., AND HARWOOD, J. L. (1980) Phytochemistry 19,2281-2285. 11. LEE, R. F., AND BENSON, A. A. (1972) Biochim

Biophys. Acta 261,35-37.

METABOLISM

OF SULFOQUINOVOSYLDIACYLGLYCEROL

12. ROUGHAN, P. G., AND BATT, R. D. (1968) AnaL Biochem. 22,74-88. 13. BLIGH, E. G., AND DYER, W. J. (1959) Cunad J. Biochem. PhysioL 37,911-917. 14. O’BREIN, J. S., AND BENSON, A. A. (1964) J. Lipid Res. 5, 432-436. 15. BENSON, A. A., DANIEL, H., AND WISER, R. (1959) Proc. N&l. Acad Sci. USA 45,1582-158’7. 16. WILSON, T. H., AND WISEMAN, G. (1954) J. Physiol. 123, 116-125. 17. COHEN, P. P. (1957) in Manometric Techniques and Related Methods for the Study of Tissue Metabolism (Umbriet, W. W., Burris, R. H., and Stauffer, J. F., Eds.), pp. 149-150, Burgess, Minneapolis, MN. 18. GARBUS, J., DELUCA, H. F., LOOMANS, M. E., AND STRONG, F. M. (1963) J. Biol. Chem. 238,59-63. 19. LOWRY, 0. H., ROSENBROLJGH,N. J., FARR, A. L., AND RANDALL, R. J. (1951) J. Biol. Chem. 193, 265-275. 20. HARWOOD, J. L. (1975) Biochim. Biophys. Acta 398,244-250.

519

21. ROUGHAN, P. G., AND BATT, R. D. (1969) Phytcchemistry 8,363-369. 22. HAZLEWOOD, G., AND DAWSON, R. M. C. (1979) J. Gen. MicrobioL 112,15-27. 23. MARTELLI, H. L., AND BENSON, A. A. (1964) Biochim. Biophys. Acta 93,169-171. 24. BURNS, D. D., GALLIARD, T., AND HARWOOD, J. L. (1979) Phytochemistry 18, 1793-1797. 25. MATTSON, F. H., AND VOLPENHEIN, R. A. (1966) .I Lipid Res. 7,536-543. 26. BAJWA, S. S., AND SASTRY, P. S. (1974) Biochem. J. 144,177-187. 27. MORGAN, R. G. H., AND HOFFMAN, N. E. (1971) Biochim. Biophys. Acta 248,143-148. 28. BORGSTROM, B., AND ERLANSON, C. (1971) Biochim. Biophys. Acta 242,509-513. 29. BORGSTROM, B., AND ERLANSON, C. (1973) Eur. J. Biochem. 37,60-68. 30. LARSON, A., AND ERLANSON-ALBERTSON, C. (1983) Biochim. Acta 750,171-177. 31. GARGOURI, J., JULIEN, R., BOIS, A. G., VERGER, R., AND SARDA, L. (1983) J. Lipid Res. 24, 1336-1342.