Glycerolipid Biosynthesis in Isolated Rat Intestinal Epithelial Cells

Depart~~zent of Bioclzemistry, and Ba~zti~zg and Best Depart~nent of Medical Research, C. H. Best Institute, University of Toronto, Toronto, Ontario M5G lL6

Received February 17, 1975

O'Doherty, P. J. A. & Kuksis, A. (1975) Glycerolipid Biosynthesis in Isolated Rat Intestinal Epithelial Cells. Can. J . Biochem. 53, 1010-1019

Intestinal epithelial cells were prepared from fasted rats by dispersion with collagenase (EC 3.4.24.3 ). The structural and metabolic integrity of the cells was verified by electron microscopy, a high percentage of Trypan Blue exclusion, a low degree of release of lactate dehydrogenase (EC 1.1.1.27) in the medium, and by the retention of sensitivity to agents known to modify metabolic and transport activity in everted sacs of intestinal mucosa. The isolated intestinal epithelial cells were used to study glycerolipid biosyn- thesis from glucose, glycerol, 2-monoacylglycerol, and free fatty acids. The cells actively incorporated the labeled precursors into glycerolipids without specific cofactor require- ments. Addition of fatty acids stimulated the incorporation of both glucose and glycerol into triacylglycerols and glycerophospholipids, the greatest effect being observed with palmitate. The stimulation of monoacylglycerol acylation appeared to depend on both the nature of the monoacylglycerol and fatty acid supplied. Stereospecific analyses of the diacylglycerols formed from 2-monoacylglycerols and free fatty acids showed that 1,2- diacyl-sn-glycerols (62-70% ) were the major and that 2,3-diacyl-s~z-glycerols (30-38% ) the minor intermediates in triacylglycerol biosynthesis. The data indicate that isolated intestinal epithelial cells exhibit a total capacity of glycerolipid synthesis and a stereo- chemical course of reaction which is comparable to that observed for triacylglycerol formation in everted sacs of intestinal mucosa, but much less specific than that seen in microsomnl preparations of intestinal mucosa.

O'Doherty, P. J. A. & Kuksis, A. (1975) Glycerolipid Biosynthesis in Isolated Rat Intestinal Epithelial Cells. Can. J . Biochem. 53, 10 10-101 9

Nous obtenons des cellules kpithkliales intestinales de rats B jeCln par dispersion avec la collagCnase (EC 3.4.24.3). L'intkgritk structurale et mktabolique des cellules est vCrifike par microscopie klectronique, par un pourcentage klevk d'exclusion du Trypan Bleu, par un faible degrk de libkration de la lactate dkshydrogknase (EC 1.1.1.27) dans le milieu et par rktention de la sensibilitk aux agents reconnus pour modifier I'activitk mktabolique et l'activitk de transport dans des sacs retournks de muqueuse intestinale. Les cellules CpithCliales isolkes sont utiliskes pour Ctudier la biosynthkse des glyckrolipides B partir du glucose, du glyckrol, du 2-n~onoacylglyckrol et des acides gras libres. Les cellules incor- porent activement les prCcurseurs marquis dans les glyckrolipides sans que soient requis des cofacteurs spkcifiques. L'addition d'acides gras stimule l'incorporation du glucose et du glycCrol dans les triacylglycCrols et les glyckrophospholipides, le palmitate Ctant le plus actif. La stimulation de l'acylation des monoacylglyckrols semble dkpendre de la nature du monoacylglyckrol et de l'acide gras fournis. L'analyse stCrCospCcifique des diacyl- glyckrols B partir des 2-monoacylglyckrols et des acides gras montre que les 1,2-diacyl-sn- glyckrols (62-70%) sont les intermkdiaires majeurs et les 2,3-diacyl-sn-glyckrols (30- 3 8 % ) , les intermkdiaires mineurs dans la biosynthkse des triacylglyckrols. Les rksultats montrent que les cellules CpithCliales intestinales isolkes conservent un plein pouvoir de synthkse des glycCrolipides et une voie rkactionnelle stkrkochimique comparable B celle observCe lors de la formation des triacylglycCrols dans les sacs retournis de muqueuse intestinale. Cette voie est cependant mois spkcifique que celle des prkparations microso- miques de la muqueuse intestinale. [Traduit par le journal]

Introduction in recent years by this (1-5) and other labora-

Preparations of fat-filled isolated intestinal tories ( 6 9 7 ) to study the processes of chylomi-

epithelial cells of rat intestine have been used C r ~ n formation and secretion under various physiological conditions. To date, no studies

'Present address: Department of Biochemistry, Uni- have been reported on the specificity and capa- versity of Wisconsin, Madison, Wisc. 53706. city of isolated intestinal cells of fasted animals

"To whom correspondence should be addressed. to carry out total glycerolipid synthesis.

Can

. J. B

ioch

em. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y U

nive

rsity

of

Wat

erlo

o on

11/

30/1

4Fo

r pe

rson

al u

se o

nly.

O'DOMBRTY AND KUKSIS: GLYCEROLlPPD BIOSYNTWESIS 1011

Previous work with everted sacs of rat in- the mesenteric fat. The intestine was then flushed with

testinal mucosa has shown (8) that the synthesis of triacylglycerols from 2-monoacylglycerols and free fatty acids proceeds via both 12- and 2,3-dixyl-sn-glycerols as intermediates. The total yield and the proportions of the enantio- rneric diacylglycerols were observed to vary with the nature of the supplied monoacylglycerols and free fatty acids. These observations with everted sacs are at variance with the results of studies in microsornes of intestinal mucosal cells. Thus, preparations of microsomes of in- testinal mucosa of hamster (9) and of rat (10) have yielded largely the 1,2-diacyl-siz-glycerols when incubated with 2-monoacylglycerols and free fatty acids. However, the microsomes of rat intestine were shown (11) to effectively acylate both 1,2- and 2,3-diacyl-sn-glycerols to triacyl- glycerols when the former were added to the incubation indium.

The present study describes a preparation of isolated epithelial cells of intestinal mucosa of fasted rats which has a total capacity for gly- cerolipid synthesis from glycerol, glucose, 2- monoacylglycerol, and free fatty acids, and shows that, unlike microsomes, isolated intes- tinal cells have not lost the capability to acylate 2-monoacylglycerols to 2,3-diacylglycerols.

Materials and Methods

Substrates [U-14CC]Glucose (specific activity (sa) 50 mCi/mmol),

[2-%]glycerol (sa 200 mCi/rnmol) , [9, 10-SH]pahitic acid (sa 200 mCi/mmol ) , and glycerol (tri[l-"61- oleate) (sa 60 mC!i/mol) were obtained from New England Nuclear, Boston, Mass. 2-Monsoleoyl[2-"H]- glycerol (sa 118 dpm/mol) was synthesized as de- scribed ( 12). 2-Mono ([1-14C]01eoyl) glycerol was obtained from gly~erof(tri[1-~%e]oleate) by digestion with triacylglycerol lipase (EC! 3-1.1.3 ) as described by Brockerhoff ( 13 ) . Unlabeled myristic, palmitic, oleic, and linoleic acids were obtained from Applied Science Laboratories, Inc., State College, Pa. Phospho- lipase Aa (EC 3.1 .I .4) (Crstalus atrox) was obtained from the Ross Allan Reptile Institute, Silver Springs, Pla. Triacylglycerol lipase was purchased from Nutri- tional Biochemicals Corp., CJevelmd, Ohio. Colla- genase (EC 3.4.24.3 ) (Clostridium histoly ticum, type III) was from Sigma Chemical Corp., St. Louis, Mo.

Issdadion of Intestinal EpithelB1 Cells Male Wistar rats (275-300 g ) were maintained on

laboratory chow prior to experimentation. After an overnight fast, the animals were sacrificed under diethyl ether anesthesia, and the upper one third of the intes- tine was excised, care being taken to remove most of

ice cold Krebs-bicarbonate buffer (pH 7.4), contain- ing 6 pmol glucose per millilitre. Suitable lengths (10- 15 cm) of the jejunum were everted over a stainless steel probe and the mucosal side was washed free of the luminal contents. These intestinal segments were tied off at one end and then gently passed between a pair of smooth flattened forceps proceeding from the open end of the sac. During these manipulations (3-4 min), the segments of the intestine remained sub- merged in the cold buffer. The mucosal scrapings were incubated at 37 "C in 10 ml of the Krebs-bicarbonate buffer containing 5-6 mg collagenase (EC 3.4.24.3). The incubations were carried out for 20 min in a Dubnoft metabolic shaker with moderate shaking speed (100 cycles per minute). At the end of this time, the suspension produced by the mechanical agitation was poured through nylon stocking material to remove intact intestinal tissue and to further disperse the cell population. This cell filtrate was poured into poly- ethylene tubes, made up to I5 ml, and centrifuged at 100 x g for 2 min to sediment the intact cells. After removing the supernatant, the pellet was completely dispersed and more buffer was added to bring the volume to 15 ml. The suspension was again centrifuged for 2 min at 100 x g. This washing procedure is nor- mally repeated twice to completely remove collagenase and until the cells sediment rapidly and uniformly, leaving a minimum of debris in the supernatant. After a final wash the cells were suspended in 5 ml of fresh media. This suspension is ready for experimental use. One million cells are approximately equivalent to 2.0 mg protein. Cell suspensions were routinely examined microscopically for Trypan Blue exclusion (14). A high percentage (85-98%) of unstained cells was always observed.

Metabolic aild Enzymic Assays Cellular glycolytic activity was measured by estimat-

ing the production of lactic acid from glucose. Isolated epithelial cells, equivalent to 20 rng cellular protein, were suspended in 5 ml of Krebs-bicarbonate buffer to which was added 1.5 mM glucose. The flasks were incubated at 37 "C in a Denbnoff metabolic shaker. At varions time intervals, 1.8 rnl aliquots were removed and transferred to an equal volume of 25% trichloro- acetic acid to stop metabolic activity. The protein was sedimented by centrifugation and lactic acid was deter- mined on an aliquot of the supernatant by the methd of Barker and Summerson (15). Lactate dehydro- genase (EC 1.1.1.27) was assayed as previously de- scribed (16). Protein was determined according to the method of Huebscher et al. ( 17 ) .

Measurement of carbon dioxide production from labeled glucose was performed as described by Kim- mich (1 8). A cell suspension of 20 rng protein in f ml was added to 1.8 ml of the Krebs-bicarbonate buEer with [l-14C]glucose (2 mM, 0.2 pCi) in a silicsnized 25-ml Erlenmyer flask containing a center well. A small glass vial containing 0.2 ml of hydroxide of Hyamine was placed on the center, the flask top stop- pered, and the flask incubated at 37 "C! in a Denbnoff metabolic shaker. At various time intervals, a small

Can

. J. B

ioch

em. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y U

nive

rsity

of

Wat

erlo

o on

11/

30/1

4Fo

r pe

rson

al u

se o

nly.

1012 CAN. J . BIOCHEM. VOL. 53. 1975

amount of 2 N HPSOc was injected into the flask to stop metabolic activity, and the vessels were shaken an additional 1 h to allow time for complete diffusion of liberated COz to the Hyamine. After this time, the Hyamine vial was removed from each flask, dropped directly into the scintillation fluid containing 0.3% 2,5-diphenyloxazole and 0.05 % 1,4-bis-2 (5-phenyl- oxazolyl) benzene in toluene, and counted for radio- activity.

Biosynthesis of triacylglycerols and phospholipids was determined by incubating radioactive glycerol and glucose with micellar solutions (19) of fatty acids with the isolated intestinal epithelial cells, as described in the footnotes to the appropriate tables and figures given in Results and Discussion. In some experiments, albumin-bound fatty acid complexes (20) were incu- bated with the cells. In other experiments, the cells were incubated with micellar solutions (19) of labeled monoacylglycerols and free fatty acids. The metabolic reactions were terminated by adding chloroform- methanol 2: 1 to the cell suspensions, homogenizing, and completing a total lipid extraction.

Lipid Analyses Total lipid extracts were prepared according to the

method of Folch eb al. (21). The lipids were resolved into fractions corresponding to triacylglycerols, free fatty acids, diacylglycerols, and monoacylglycerols by TLCa on silica gel H (Merck and Co.) using heptane - isopropyl ether - acetic acid 60:40:3 (v/v/v) as the developing solvent (2). The phospholipids remaining at the origin were recovered and resolved into lyso- phosphatidylcholine, phosphatidylcholine, and phos- phatidylethanolamine fractions by TLC using chloro- form - methanol - glacial acetic acid - water 26:15:4:2 (v/v/v/v) as the developing solvent (22). In all cases the lipid components were identified by cochromatog- raphy with standards. In some experiments, stereo- specific analyses were performed on the X-1,2-diacyl- glycerols. After dilution with carrier sn- 1,2-(2,3 ) - diacylglycerols ( 10 mg ) from triacylglycerol lipase hydrolysis of lard triacylglycerols, the X-1,2-diacyl- glycerols were converted into phosphatidylphenols which were analyzed stereospecifically by treatment with phospholipase Az according to the procedure of Brockerhoff ( 13 ) . The resulting lysophosphatidyl- phenols and residual phenols were chromatographed on silica gel H using chloroform - methanol - 3 % aqueous ammonia 63 : 30: 7 (v/v/v) as the developing solvent ( 13). The radioactivity of the fractions was assessed after elution and counting in a liquid scintil- lator. In some experiments the labeled diacylglycerols were resolved into classes differing in degree of un- saturation by argentation TLC (8) prior to stereo- specific analysis. The lipid masses were quantitated by GLC of the fatty acids in the presence of heptadecanoic acid as internal standard as previously described (2) . The radioactivity of the lipid fractions was measured in a Nuclear Chicago liquid scintillation counter using the channel ratio method (23 ) .

3Abbreviations used are, GLC, gas-liquid chroma- tography; TLC, thin-layer chromatography.

Two tests were made of the efficiency of the stereo- chemical method. In one, 508 mg of corn oil mixed with 180 pl of [2-W]glycerol trioleate (0.05 pCi) was digested with triacylglycerol lipase (24) and the mixture of sn-1,2-(2,3)-diacylglycerols isolated by neutral lipid chromatography as described above. After stereo- specific analysis (13) the percentage of distribution of the srr-1,2- to 2,3-isomers, determined from the radio- activity of the corresponding lyso- and residual phos- phatides, was 52:48, approximating the 1 : 1 ratio expected. In the other test, phosphatidylphenol forma- tion and phospholipase A, digestion (13) was per- formed on the sn-1,2-diacylglycerols derived from egg yolk phosphatidylcholine by degradation with phospho- lipase C (EC 3.1.4.3 ) (25 ) . Only lysophosphatides, indicating the presence of pure sit-1,2-diacylglycerols were recovered.

Electron Microscopy Electron microscopy was performed as described

previously ( 1 ) .

Results and Discussion

General Characteristics sf Isolated Intestinal Epithelial Cells

The initial objective was to determine the structural and metabolic integrity of the isolated epithelial cells. This was done by electron micro- scopic, biochemical, and physiological methods.

Figure 1 represents an electron micrograph of an isolated intestinal epithelial cell. The various cell components appear to have survived the isolation procedure without significant alteration. In some cells, part of the lateral cell margin appeared to lack a recognizable membrane pro- file, as also noted by Marsh et a!. (26) in in- testinal cells prepared by treatment with phos- pholipase C or EDTA. Nevertheless, 85-90y6 of the cells appeared to exclude Trypan Blue at all times. If the isolated intestinal cells remained structurally and functionally intact during incu- bation, they should retain soluble cytoplasmic enzymes. The leakage of lactate deh yd rogenase from the isolated intestinal cells into the extra- cellular fluid was therefore determined follow- ing incubation at 37 "C for 30 and $0 min. A maximum of 12-167, of the total enzyme ac- ti\ ity was released into the medium. The amount of lactate dehydrogenase released into the me- dium could therefore be correlated with the proportion of damaged cells as determined by Trypan Blue exclusion.

The glycolytic rate of the isolated cells was measured by lactic acid production. It was linear for approximately 2 h, and the yield of lactate

Can

. J. B

ioch

em. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y U

nive

rsity

of

Wat

erlo

o on

11/

30/1

4Fo

r pe

rson

al u

se o

nly.

O'BOHERTY AND KUKSIS : GLYCEROLIPID BIOSYNTNESIS

FIG. 8 . Electron micrograph of an isolated intestinal epithelial cell. Magnification X 12 000. The marker represents 5 pm.

averaged about 250 nmol/h per milligram of cell protein. This is comparable to the 270 nmol/ h per milligram of cell protein reported by Leslie and Rowe (27) for hyaluronoglucosidase-dis- persed cells and to the 360 nmol/h per milligram of cell protein reported by Stern and Reilly (28) for trypsin-dispersed cells. Mechanically dis- persed intestinal cells, as reported by Iemhoff et al. (29), had a lactate production of 120 nmol/h per milligram of cell protein.

Figure 2 shows the rate of carbon dioxide production from [14C]glucose by the isolated epithelial cells, and the effects sf several agents known to modify metabolic and transport ac- tivity in everted sacs of intestinal mucosa (30, 3 1 ). The carbon dioxide production is essential- ly linear for the 2-h incubation interval. Further- more, stimulation by 2,4-dinitrophenol indicates

that the cells have retained their ability to couple metabolic activity to energy production. The severe inhibitory effects of ouabain and oligo- mycin on carbon dioxide production also indi- cate that the isolated cells retain sensitivity to those agents which are inhibitory before isola- tion (30, 31). On the basis of the above results, it may be concluded that much of the structural and characteristic metabolic integrity of the in- testinal epithelial cells has indeed been retained in the isolated cell preparations.

Conditions of Glycerolipid Biosynthesis To avoid the release of chylomicrons into the

incubation medium, the isolated intestinal epi- thelial cells were incubated with the various lipid precursors in Krebs-bicarbonate buffer in the absence of albumin or serum (1). Under these

Can

. J. B

ioch

em. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y U

nive

rsity

of

Wat

erlo

o on

11/

30/1

4Fo

r pe

rson

al u

se o

nly.

CAN. J. BIOCHEM. VOL. 53, 1975

Time (min? FIG. 2. Effect of various agents on carbon dioxide production from [I-W]glucose by isolated intestinal epithelial

cells. Incubations were carried out as described in Materials and Methods. Where appropriate, 400 pM 2,4-dinitro- phenol (IDNP), 208 pM ouabain, and 15 pg oligomycin were added to the incubation medium.

conditions, essentially linear rates of incorpora- TABLE 1 . Percentage distribution of label in mucaa8 cell tion were obtained with all precursors tested glycerolipids after incubating for I h * over a period of I k without the addition of m y cofactors. After 1 h, the rate of incorporation Lipid classes%

of the precursors tended to level off. Sunder Labeled precursors TG DG MG FFA PL et u1. (32) have recently reported constant rates of glycerolipid biosynthesis for 1 h from labeled glycerol in isolated parenchymal cells of rat liver.

Table I shows the percentage distribution of radioactivity in the free fatty acids and gIyc- erolipids of isolated intestinal epithelial cells of the rat 1 h after incubation with a variety of labeled precursors. 'In all instances, the bulk of the label was found in the triacylglycerols (50- 8OyO). Phospholipids were much more ex- tensively labeled from glucose (45y0) than from glycerol ( 1 Oyc). Very little label appeared in the phospholipids from monoacylglycerols la- beled in either the glycerol or fatty acid moiety. In all instances the diacylglycerols represented 10-157c of the total radioactivity taken up by the cells. Since only trace amounts of radio-

-

*Data represent average of two experiments. TTG, triacylglycerols; DG, diacylglycerols; MG, monoacyl-

glycerols; FFA, free fatty acids; PL, phospholipids.

activity could be recovered in the free fatty acid fraction following incubations with labeled glyc- erol or glycerol esters, the incorporation of the labeled substrate occurred almost exclusively in the glycerol portion of the molecule. This pat- tern of the overall distribution of radioactivity among the various glycerolipids of the isolated cells is quite similar to that noted previously using everted sacs of intestinal mucosa (12, 43). There was no uptake of diacylglycerols or tria-

Can

. J. B

ioch

em. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y U

nive

rsity

of

Wat

erlo

o on

11/

30/1

4Fo

r pe

rson

al u

se o

nly.

O'BBHERTY AND KUKSIS : GEYCEROEIPIB BIOSYNTHESIS

Time (min) Time (minl

FIG. 3. Effect of fatty acids on the incorporation of [2-3H]glycerol into the triacylglycerols (A) and glycerophospho- lipids (B) of intestinal epithelial cells. Isolated cells (20 mg cell protein) were incubated at 37 "C in Krebs-bicarbonate buffer, pH 7.4. Final volume in all incubations was 5 ml. [2-aH]Glycerol (0.304 pmol, 1.8 pCi) and 4 pmol of each fatty acid as a micellar solution were added to the incubation mixture. The results represent an average of three to four experiments.

cylglycerols by the isolated intestinal cells, which in the selection and positional placement of the further attested to the integrity of their cellular fatty acids than in the phospholipids in the liver membranes. cell. The latter possibility is supported by the

Eflect of Diflerent Fatty Acids on Glycerolipid Synthesis

Figure 3 demonstrates that the extent of in- corporation of glycerol into the glycerolipids was greatly stimulated by the addition of mi- cellar solutions of free fatty acids to the incuba- tion medium. At a concentration of 4 pmol per 20 mg of cellular protein, palmitic and oleic acids had the greatest stimulating effect, while stearic and myristic acids had the lowest, with linoleic acid showing an intermediate effect. The relative order of stimulatory activity was the same for these fatty acids for both triacylglyc- irol and phospholipid biosynthesis from glyc- erol. These observations are in agreement with those of Sundler et al. (32) who noted a stimu- latory effect of free fatty acids on the incorpora- tion of glycerol into the glycerolipids by isolated rat hepatocytes. These workers, however, found that the effect on phospholipid was more de- pendent on fatty acid structure. The comparable order of stimulatory activity of the different acids in the present study indicates that phos- pholipid biosynthesis in isolated intestinal epi- thelial cells may be subject to a lesser specificity

studies of Arvidson and Nilsson (34) who noted that during fat absorption in vivo the fatty acid composition of the mucosal phospholipids tended to approximate that of the dietary fat, including high proportions of phosphatidylcho- lines containing two molecules of the same dietary fatty acid in the same molecule.

We also observed (not shown) that the stimu- lation of glycerol incorporation into glycero- lipids was greater with micellar solutions of fatty acids than with albumin-bound fatty acids. This is consistent with the findings of Johnston and Borgstrom (19) who used everted sacs of in- testinal mucosa. There was a maximum of 7-8y0 lipolysis in these cells, as determined by the rates of hydrolysis of labeled trioleoylglycerol and dio%eoylglycerol added to cell homogenates and by determining the difference in the specific activity of medium glycerol and free fatty acids at the beginning and at the end of incubation.

Figure 4 shows the effect of free fatty acids upon the incorporation of radioactivity frona glucose into the glycerolipids of isolated, epi- thelial cells. As seen in the glycerol experiments above, the addition of fatty acids markedly

Can

. J. B

ioch

em. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y U

nive

rsity

of

Wat

erlo

o on

11/

30/1

4Fo

r pe

rson

al u

se o

nly.

1016 CAN. J . BHOCHEh¶. VOL. 53, 1995

0 - .g .- C r ig

8 = ." $

15 30 45 " 1 2 0

Time (rnin) Time BrninB

FIG. LFa Effect of fatty acids on the incorporation of [U-14C]glucose into the triacylglycerols (A) and glycerophospho- lipids (B) sf intestinal epithelial cells. Incubation conditions were as in Fig. 3 except that [U-IIC]glucose (0.146 Pmol, 0.5 pCi) and 4 pmol of each fatty acid as a micellar solution were added to the incubation medium. The results represent an average of two to three experiments.

Time (rnin) FIG. 5. Time course of ii~eorporation sf [2-3H]glycerol (A) and [U-la@]glucose (B) into glycerophosgholipids of

intestinal cells. Isolated cells (28 mg cell protein) were incubated at 37 OC in Krebs-bicarbonate buffer, pH 7.4. Final volume in all incubations was 5 ml. [2-3H]Glycerol (0.304 Prnol, I .O pCi) and [U-14C]glucose (0.146 Pmol, 0.5 PCi) were added to the respective incubation media. The results represent the means + SD of four experiments.

stimulated the incorporation of radioactivity into both triacylglycerols md phospholipids. Of the fatty acids tested, palmitate gave the highest degree of stimulation in both instances. Am earlier study by Dawson and Isselbacher (35) had shown that palmitate stimulated the in- corporation of labeled glucose into the total lipids of everted sacs of intestinid mucosa, but no differentiation was made between neutral

lipids and phospholipids. The present study shows that the stimulation of glycerslipid label- ing from glucose was much lower than that from glycerol (compare Figs. 3 and 4). This preferential incorporation s f labeled glycerol in isolated epithelial cells is in agreement with the in vim observations of Saunders and Bawson (33), who have argued in favor of an acylation sf free glycerol, despite the apparent absence of

Can

. J. B

ioch

em. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y U

nive

rsity

of

Wat

erlo

o on

11/

30/1

4Fo

r pe

rson

al u

se o

nly.

0,BOHERTY AND KUKSIS : GLYCEROLPID BIOSYNTHESIS

+ 18:O

Control Con! rol

Time (rnin) Time (rn~n)

FIG. 6. Effect of fatty acids on the incorporation of 2-mono([l-1~C]oleoyl)g~ycerol into diacylglycerols (A) and triacylglycerols (B) in isolated intestinal epithelial cells. Incubation conditions were as described in Fig. 3, except that 4 firno1 momoacylglycerol and 8 firno1 of each fatty acid were added to the incubation mixture as a micellar solution. The results represent an average of three to four experiments.

TABLE 2. Incubation of intestinal epithelial @ells with labeled 2-rnonoacylglycerol and free fatty acids*

Lipid precursorst (76 total X-1,2-isomer)

16:O + MO 1 8 2 + MO 18:l + MP 1 8 ~ 2 + MP

Diacylglycerol isomer A B A B A B A B

*The incubation flasks contained 28 mg cell protein in Krebs-bicarbonate buffer to which was added 4 ptnol monoacylglyccrol and 8 pmol fatty acid as a miceliar solution. After 1 h the incubations were terminated and the acylglycerols annljzed. The yield of diacylglywmls was 80-110 nmol. A and B represent separate incubations.

?Ms . 2-monoolesyl-sn-glycerol; MP, 2-monopalmitoyl-sn-glycerol; (X-l,3-diacylglycerols = 12-17%).

a glycerol kinase (EC 2.7.1.30) from the intes- tinal mucosa. The mechanism of utilization of free glycerol by this tissue requires further examination.

Figure 5 gives the time course of incorpora- tion of radioactivity from glycerol and glucose into the choline and ethanolamine phospha- tides, which are the two major phospholipids of intestinal mucosal cells. With both precursors, a greater amount of phosphatidylcholine was synthesized.

Figure 6 shows the stimulatory effect of vari- ous fatty acids on the incorporation of labeled 2-monoolesylglycerol into the di- and triacyl- glycerols. The highest stimulation occurred with palmitic and oleic acids. When labeled 2-mono- palmitoylglycerol was used as the monoacyl- glycerol, the greatest stimulation occurred with oleic and linoleic acids. It would thus appear that the stimuiatory effect of fatty acids on the

incorporation of monoacylglycerols depends on the nature of both the monoacylglycesol and the free fatty acid supplied, and that this is also reflected in the yields of the corresponding di- and triacylglycerols. This latter observation is consistent with that made with both everted sacs (8) and microsomes (1Q) of rat intestinal mucosa.

Stereochenrical Course of Diacy bglycerol Formation

Table 2 indicates the proportions of 1,2- and 2,3-diacyl-sut-glycerols isolated from the cells following incubation with 2-monopalmitoylglyc- erol and oleic or linoleic acid, and with 2-mono- oleoylglycerol and palmitic and linoleic acids. In all instances the 1,2-isomer was found to predominate but significant amounts of 2,3- diacyl-sn-glycerol were also formed. Table 3 gives the amounts of isomeric diacylglycerols

Can

. J. B

ioch

em. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y U

nive

rsity

of

Wat

erlo

o on

11/

30/1

4Fo

r pe

rson

al u

se o

nly.

CAN. J. BHBCHEM. VOL. 53, 1975

TABLE 3. Incubation of intestinal epithelial cells with labeled 2-monoacylglycerol and a mixture sf free fatty acids*

Lipid precur~ors~ (% total %-I ,2-isomer)

1 4 9 + 16:O + MO 24:0 + %8:1 + MO

Diarcylglycerol isomer A B A B

*Incubation conditions were. as in Table 2 except that a mixture of fatty acids were used. The yield of diacylglycerols was 110-120 nmol. A and B represent separate incubations.

fMO, 2-mclnoolesyk-sn-glycerol; (X-1,3-diacylglycerols = 10-16%).

TABLE 4. Chemical classes of diacylglycerols formed in intestinal epithelial cells with labeled Zmsaoacy~glycerol and a mixture of free fatty acids*

Lipid precursorsf (76 total X-B ,%isomer)

16:1 + 18:1 + MO 16:0 + 1 8 2 + MO

DiacyBgBycersl isomer Total Monoenes Dienes Total Monoenes Trienes

*Incubation conditions were as in TaMe 3. The yield of diacylglycesols was 110-125 nmol. t MO = 2-moncsoltoyl-~n-glycer~9.

formed when labeled 2-monooBeoylglycero1 was incubated with mixtures of myristic and pal- mitic, and of myristic a d oleic acids, Again, both 1 $2- and 2,3 -diacyI-sn-glycerols were formed with the 1,2-isomer predominating to about the same extent as in the previous experiments. The chain Bength of the fatty acids appeared to have little effect on the ratio of the isomers found, although it is known to effect the molecular weight distribution and the molecular associa- tion of the fatty acids in a diacyfglycerol mole- cule (12).

Table 4 shows that the degree of unsaturatism of the free fatty acids also does not significantly affect the ratios of the enantiomers forn~ed. There was a preferential synthesis of the 13- isomers in the monoene, diene, and triene frac- tions, but the ratios were about the same as those noted above under conditions of a more limited choice of fatty acids. Since the diacylglycerols derived from the monoacylglycerol pathway were not utilized for phospholipid biosynt%zesis, there apparently was no sekctive removal sf the 1,2-diacyl-sn-glycerol isomers for this purpose in any s f the unsaturation classes of the m i x 4 diacylglycerols. The Hack of utilization of even the unsaturated diacylglycerol products of the monoacyIglycero% pathway in phospholipid bis-

synthesis is consistent with the earlier findings sf Johnston. et a/. (36) and Breckenridge (I%), who noted that labeled msnoacylg8ycerols failed to appear in the diacylglycerol moieties of the glycerophosphslipids of intestinal mucosa. It was subsequently shown by Johnston et d. (9) that even the 1,2-diacyl-sa-glycerols arising from the monoacylglycerol pathway were not utilized in phosphatidylcho%ine biosynthais in mucosal microsomes.

The finding that both sn-1,2= and 23-diacyH- glycerols are formed in significant amounts by isolated epithelial cells is in agreement with the observation of Breckenridge and Kuksis (18) who noted comparable proportions of the enantiomers in the lipids recovered from everted sacs of intestinal mucosa. The propor- tions sf 1,2- and 2,3-diacyl-sn-glycersHs found in intact cells, however, are markedly diEerent from those recovered from similar incubations in rat mucosal microsomes, which largely yield 1,2-diacyl-sa-glycerols ( 1 0). Hn none of these instances, however, was the specificity for the synthesis of B ,2-diacyB-sn-glycero1~ as high as that in mucosal microsomes of the hamster (9) which gave nearly 100% of the 12-isomer. Both Breckenridge and Kuksis (8) and Gal'lo and Treadwell (37) have suggested that the enzyme

Can

. J. B

ioch

em. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y U

nive

rsity

of

Wat

erlo

o on

11/

30/1

4Fo

r pe

rson

al u

se o

nly.

O'BOPIERTY AND KUKSIS: GLYCEROLIPID BIOSYNTHESIS 1819

responsible for the acylation of the sn-3-position of the 2-monoacylglyceroI may be lost during: subcellular fractionation sf the intestinal mu- cosa. Clearly, the mucosaf acylation of sn- glycerol-3-phosphate and of the mono- and di- acylglycerols must be mediated by acyltransfer- ases of varying substrate specificity and subsel- lular distribution. In such a case, the production sf both 1,2- and 2,3 -diacyl-sn-glycerols could k taken as a further indication of the structural and metabolic integrity sf the isolated cell preparation.

The occurrence of %,3diasyl-sn-glycerols in extrajejunal tissues is as yet unknown, but it may be pointed out that the ability to acylate %,3-isomers of diacylglycerols to triacylglycero~s has k e n demonstrated in fiver microsomes of rat (10, 111, chicken (38), m d pig (39), in adipose tissue microsomes of chicken ($01, and in microsomes of Ehrlich ascites cells (10).

Preparations of viable intestinal epithelial cells, as described in this report, have been used to study the synthesis and secretion of chylo- microns as documented elsewhere (41). The ability to prepare suspensions of intestinal epi- thelial cells which retain their metabolic md structural integrity offers a productive avenue of approach to the study of intestinal lipid and lipoprotein biosynthesis and transport. The ease of manipulation and reproducibility of sampling provide significant advantages over other sys- tems employing more intact tissue preparations. More important, the use of cell suspension al- lows a degree sf homogeneity in cell popula- tions not possible to be obtained with classical techniques.

These studies were supported by the Ontario Heart Foundation, and the Medical Research Council of Canada. We thank Mr. E. F. Whitter for performing the electron microscopy.

1. Yousef, I. M. & Kuksis, A. (6972) Lipids 7, 380-386 2. OYBoberty3 P. J. A., Yousd, I. M. & Kuksis, A.

(1973) Arch. Biochem. BdopB1)'s. 156, 586-594 3. O'Doherty, P. J. A., Yousef3 8. M. & Kuksis, A.

(6973) J. Nutr. 103, xxviii 4. Youse& I. M., 09Doherty, P. J. A. & Kuksis, A.

(1973) J. Ngtr. 183, xvii 5. O'Doherty, P. J. A. & Kuksis, A. (1975) Int. J. Bio-

c!l~m. in press 6. Kessler, J. I. (1972) J. Clhr. IIIYPSP. 51, 50a 7. Kesler, J. I. & Mauldin, D. P. (1973) Gastroenterol-

ogy 64, A175 8. Breckenridge, W. C. & Kuksk, A. (1 972) Lipids 7,

256259

9. Johnston, 3. M., Paultauf, F., Schiller, @. M. & Schultz, L. D. (1970) Biochim. Bioplzj~s. Acda 218, 124-1 33

10. O'Doherty, P. J. A. & Kuksis, A. (1974) Can. J . Bioehem. 52, 514-524

If . O'Doherty, P. 3. A., Kuksis, A. & Buchnea, D. (1972) Car?. J. Biocllem. 50, 88 1-887

12. Breckenridge, W. C. (1970) Ph.D. thesis, Univasity of Toronto, Toronto, Ontario, pp. 1-262

13. Brockerhoff, H. (1 965) J. Lipid Res. A, 10-8 5 14. Girardi, A. J., Michael, M. M. & Henle, W. (1956)

Virolagy 2, 534-544 15. Barker, S. B. & Summerson, W. H. (1941) 1. Bid.

Chem. 138,535-954 16. O'Doherty, P. J. A. & Feltham, L. A. W. (1971)

Cotmp. Biochem. Physisl. 38, 543-551 17. Huebscher, G., West, @. R. CG Brindley, D. N. (1965)

BiocAem. J. 97,629-642 18. Kimmick, G. A. (1970) Bioc/le~listry 9, 3659-3668 19. Johnston, J. M. & Borgstrom, B. (1964) Bioehiw.

Biahys. ACPQ 84,412-423 20. Fillerlip, D. L., Migtiore, J. C. & Mead, J. F. (1958)

J. Biol. C11em. 233,98-101 21. Foleh, J., Lees, M. & Sloane Stanley, G. H. (1957)

J. Biol. Clzem. 226, 497-509 22. Skipski, V. P., Petterson, R. F. & Barclay, M. (1964)

Bioehem. J. 90, 374-378 23. Baillie, E. A. (1960) brzt. J . Appj. Radiat. Bsot. 8, 1-8 24. Mattson, F. H. & Vslpnhein, R. A. (1961) J. LipEd

Res. 2,5842 25. Kuksis, A. & Marai, L. (1967) Lipih 2,217-224 26- Marsh, M. N., Peters, T. J. & Brown, A. 6. (1971)

Gut 12, 499-508 27. Leslie, G. I. & Rowe, 6. B. (197%) BicAetnisPry 11,

1696-1 703 28. Stern, B. K. & Reilly, R. W. (1965) Nature 205, 563-

565 29. femhoff, W. G. J., Van den Berg, J . W. C., DePijper,

A. M. & Hulsrnann, W. C. (1970) Bimltim. Biopl?ys. Acfa 21.5, 229-241

30. Csaky, T. Z . & Hara, Y . (1969) Am. J. P l ~ y s i l . 209, 467-472

31. Parsons, B. J., Smith, D. H. & Taylor, C. B. (1958) J. Physiol. (Lolmdorl) 144, 387-398

32. Sunder, R., Akesson, B. & Nibson, A. (1974) J . Bid. @/ien?. 249, 51832-5107

33. Saunders, D. R. & Dawson, A. M. ( 1962) Bimhem. 9. 82,477-483

34. Arvidson, G. A. E. & Nilsson, A. (1972) Lipids 7, 344-348

35. Dawson, A. M. & Isseibacher, K. J . (1960) J. C h . Illvest. 39, 730-740

36- Johnston, J. M., Rm, 6. A. & Lswe, P. A. (1967) Bioe!?im. Biaplgvs. Aetn 137, 578-580

37. Gallo, L. & Treadwell, C. R. (1970) Am&. Biochem. BiopI~ys. 141, 614-621

38. Webs, S. B., Kennedy, E. la. $s Kiyasu, J. Y. (1960) J. Biol. Chern. 235, 40-44

39. Akesson, B. (1969) Eur. J. Biochem. 9,436414 40. Goldman, P. & Vagelos, P. R, (1961) J. Bid. Chem.

236,2623-2623 41. O'kherty, P. J. A., Yousef$ 1. M. & Kuksis, A.

(1972) J. Am. Oil Cl~em. S06.49,306A

Can

. J. B

ioch

em. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y U

nive

rsity

of

Wat

erlo

o on

11/

30/1

4Fo

r pe

rson

al u

se o

nly.