Journal of Clinical InvestigationVol. 42, No. 2, 1963

DEMONSTRATIONOF AN INTESTINAL MONOGLYCERIDELIPASE: ANENZYMEWITH A POSSIBLE ROLE IN THE INTRACELLULAR

COMPLETIONOF FAT DIGESTION *

By JOHNR. SENIORt AND KURTJ. ISSELBACHERWITH THE TECHNICAL ASSISTANCEOF DOROTHYM. BUDZ

(Froml tie DepartmtcW of Medicine, Harvard Medical School, and the Medical Services[Gastrointestinal Unit], Massachusetts General Hospital, Boston, Mass.)

(Submitted for publication August 1, 1962; accepted October 18, 1962)

The concept is well established that digestionof foodstuffs that occurs within the lumen of theintestine serves to convert macromolecules tosmaller ones that may be more assimilable andavailable for transport across the mucosal cell.Not until recently has it become widely appreci-ated that this intraluminal digestion is often in-complete and that additional digestion of the par-tially split fragments may actually occur withinthe mucosal cell during the process of absorption.Thus, it has been shown recently (2) that dipep-tides liberated from ingested proteins by the ac-tion of pancreatic proteolytic enzymes may be ab-sorbed by the cell and split by mucosal dipeptidases(3). Similarly, it has been demonstrated that af-ter the intraluminal digestion of starch to disac-charides, these sugars are hydrolyzed within theintestinal epithelial cell by specific disaccharidases(4-6). In fact, these functions appear to be car-ried out at the surface of the epithelial cells byenzymes localized at the microvilli or brush bor-ders of the cells (3, 6).

In the case of fat digestion, the ingested lipid,most of it triglyceride, is hydrolyzed in the lumenof the gut by action of pancreatic lipase. This en-zyme attacks specifically the ester bonds joiningthe fatty acyl chain to the two primary (a and a')hydroxyl groups of the glycerol molecule (7).The result is a liberation of fatty acids, but thereis also a significant accumulation of unhydrolyzedmonoglyceride, especially /3-monoglyceride. Al-

* Investigation supported in part by research grantA-3014 from the U. S. Public Health Service, and pre-sented in part before the Eastern Section meeting of theAmerican Federation for Clinical Research, January 12,1962, Philadelphia, Pa. (1).

t Work done during tenure of a Special Research Fel-lowship of the National Institutes of Arthritis and Meta-bolic Diseases. Present address: Philadelphia GeneralHospital, Philadelphia, Pa.

though some of this /8-monoglyceride may be iso-merized to an a-monoglyceride and then be sub-ject to further attack by pancreatic lipase, con-siderable amounts of monoglycerides appear toescape hydrolysis and remain available for in-testinal absorption.

The present study demonstrates that mono-glycerides absorbed by the intestinal mucosal cellmay undergo further cleavage within it by theaction of an intestinal lipase. Although intestinallipases were postulated as early as 1892 by Schiff(8) and have recently been demonstrated moreconvincingly (9, 10), the present results emphasizethat under reasonably physiologic conditions theintestinal lipase is most active towards mono-glycerides. Data are also presented concerningthe properties of the enzyme system and its lo-calization within the cell compartments. The in-testinal monoglyceride lipase may have a signifi-cant role in fat absorption, especially in facilitatingthe completion of fat digestion within the mucosalcell.

MATERIALS AND METHODS

Mono-, di-, and tripalmitin were obtained from Dr.F. H. Mattson,' as were a-monolinolein and the mono-glyceride isomers a- and p-mono-olein, a- and p-mono-palmitin, and a- and 3-monostearin. A series of DL-a-monoglycerides of varying chain length in the fatty acidportion was synthesized from the acyl chlorides and iso-propylidine glycerol under conditions established by Baerand Fischer (11). Acyl chlorides were prepared by re-fluxing the fatty acids 2 butyric, caproic, caprylic, capric,lauric, myristic, palmitic, and stearic with excess thionylchloride, and purified by distillation under reduced pres-sure of 10 to 15 mmof mercury obtained by a wateraspiration system. The DL-isopropylidine glycerol wasmade by exhaustive reflux of dry acetone and glycerol

1 Biochemical Research Division, Proctor & GambleCompany, Cincinnati, Ohio.

2 Nutritional Biochemicals Corporation, Cleveland, Ohio.

187

JOHN R. SENIOR AND KURT J. ISSELBACHER

with petroleum ether (boiling range, 38' to 58' C)through a column of glass helices over into a total-refluxcondenser fitted with a Barrett water trap (12), fol-lowed by distillation of the product. The monoglycerideswere purified by recrystallization and silicic acid chro-matography. A similar process was used to synthesizeDL-a-monopalmitin labeled in the glycerol portion bystarting with glycerol-1(3)_-C"3 diluted to a specific ac-tivity of 0.06 juc per umole. Tri- and diglycerides la-beled in the glycerol portion were prepared by directacylation of glycerol-C" in dry pyridine with acyl chlo-rides, or with the free acids in trifluoroacetic acid an-hydride (13), and were purified by elution from columnsof silicic acid. All solvents were of certified reagentquality4 and were freshly redistilled before use; glycer-ides were checked finally for purity on thin layer platesof silicic acid.

The glyceride substrates were suspended variously inaqueous solutions of bovine serum albumin,2 Tween 80,5or sodium taurocholate.6 Crude pancreatic lipase wasobtained commercially.7

Female albino rats 8 weighing 200 to 300 g and fedon Purina chow were ordinarily fasted overnight beforeuse. After cervical dislocation, the abdomens wereopened, and the small intestine was rinsed in situ withRinger's solution at room temperature and then chilledto 0' in 0.278 M mannitol solution buffered to pH 7.0with 0.01 M Tris-maleate. After the epithelial cellswere gently scraped free, 14 ml of mannitol solution perg of scraped cells was added and an homogenate preparedwith a Potter-Elvehj em tissue grinder, with a Teflonpestle. After filtration through a double layer of ab-sorbent gauze, nuclei and cell debris were spun down assediment at 1,400 X g for 10 minutes, mitochondria, at5,900 X g for 15 minutes, and the microsomal fraction,at 105,000 X g for 60 minutes. The tissue fractions wereresuspended in appropriate volumes of isotonic KCl buf-fered to pH 7.0 with 0.01 Mphosphate, and proteins weremeasured by the biuret reaction or the Lowry method(14). Isolated brush border preparations were made asdescribed by Miller and Crane (6). Tissue suspensionsboiled for 15 minutes before mixture with substrate andmedium were used for control experiments in which in-activated enzyme preparations were desired.

Assay of the lipolytic activities of the various suspen-sions was carried out by two methods: 1) liquid scintil-lation counting of glycerol-C" and lower glycerides pro-duced by hydrolysis of the glycerides labeled in theglycerol portion, and 2) direct titration of fatty acids re-leased at a constant pH in an automatic glass electrodetitrating device.9 In the first method, free glycerol wasseparated from the glyceride substrates by the chloro-

3 Nuclear-Chicago Corporation, Chicago, Ill.4 Fisher Scientific Corporation, Boston, Mass.5 Polyoxyethylene sorbitan mono-oleate.6 Organon, Incorporated, West Orange, N. J.7 Sigma Chemical Corporation, St. Louis, Mo.8 Charles River Laboratories, Boston, Mass.9 Model TTT-1, Radiometer Corporation, Copenhagen.

form-methanol extraction technique of Folch, Lees, andSloane Stanley (15), the glycerides remaining in thelower, chloroform phase, and the glycerol entering theupper, aqueous phase. Samples of the upper phase wereused for chemical and radioactive assay of the freeglycerol.

1. Radioactive assay of free glycerol-C" and glycerol-labeled glycerides. To avoid certain difficulties, bothpractical and theoretical, in the chemical assay of liber-ated glycerol from glycerides, glyceride substrates weresynthesized with the glycerol portion labeled with C'4 inthe 1- or 3-position, as described above. Glycerides weresuspended in bovine serum albumin solution, 100 mg perml of 0.01 M potassium phosphate buffer at pH 7.4,with the aid of small amounts of diethyl ether for initialdissolving of the glycerides and a glass homogenizingvessel with a rotating Teflon pestle. During homogeniza-tion, the temperature was gradually raised to 40' C byimmersion of the vessel in a beaker of warm water, un-til all of the ether was driven off and a uniform suspen-sion of 1 mg glyceride per ml of albumin solution re-mained. The incubation media consisted of 1 ml of theglyceride-albumin suspension; 0.4 ml of fivefold concen-trated, low calcium, Krebs-Ringer phosphate buffer atpH 7.4 (16) ; and 0.1 to 0.5 ml of the intestinal tissuesuspensions in isotonic buffer of the same type, whichwas also used to make the final volumes to 2.0 ml. In-cubations of the intestinal cell homogenates and resus-pended subcellular fractions were carried out in conicalPyrex centrifuge tubes for 30 to 60 minutes at 38' C inair, with occasional shaking. In some experiments, so-dium taurocholate was added to the media to a finalconcentration of 20 mMper L.

Incubations were terminated by addition of 15 ml ofice-cold methanol, followed by 30 ml of chloroform.After 30 minutes or more at room temperature, 10 mlof 0.05 MKC1 was added and the phases were allowed toseparate overnight at 4' C. The upper phase was sepa-rated and samples were taken for counting of the glycerol-C". The glycerides in the lower, chloroform phase wererecovered after removal of the solvent under a streamof nitrogen, then redissolved in chloroform and applied inspots on silicic acid thin-layer plates. After ascendingchromatography in a solvent system 30 per cent (by vol-ume) diethyl ether, 69 per cent hexane, and 1 per centglacial acetic acid, spots were located by brief exposureof the plates to iodine vapor and identified by compari-son with standard glycerides. The light brown glyceridespots were marked by needle scratches on the surround-ing silicic acid, and the iodine was allowed to disap-pear completely by sublimation. Careful scraping of thesilicic acid areas bearing the lipids was done by razorblade, and the scrapings from the glass plates were trans-ferred quantitatively into counting bottles. Scintillationcounting fluid was added, made up in 5 ml of 95 per centethanol and 10 ml of toluene containing 0.3 per cent2,5-diphenyl-oxazole and 0.01 per cent p-bis-2-(5-phenyl-oxazolyl) benzene,'0 and counting was carried out in a

10 Pilot Chemicals, Inc., Waltham, Mass.

188

INTESTINAL MONOGLYCERIDELIPASE

TABLE I

Formation of lower glycerides and free glycerol fromglycerides by rat-gut microsomes*

Labeled product isolated

Mono-Triglyc- Diglyc- glyc- Glyc-

Substrate eride eride eride erol

%original substrate radioactivityTripalmitin-C14 97.4 1.9 0.4 0.3Dipalmitin-C14 0.3 96.6 1.5 1.6Monopalmitin-C14 0.5 0.3 7.3 91.9

* Incubation conditions as described in Figure 1.

Packard liquid scintillation spectrometer. Recovery ofcounts from lipids applied to silicic acid plates or fromchromatographed lipid spots was within 5 per cent of theexpected radioactivity.

In subsequent experiments, the preparation of the tis-sue was varied to permit assay of activity in whole in-testinal epithelial cell homogenates, nuclei, mitochondria,microsomes, and cell sap, as well as in isolated brushborders and the supernatant fluid derived from them bythe technique of Miller and Crane (6). Other series ofexperiments were performed at pH levels from 5.0 to 9.5,with 0.125 M Tris-maleate or potassium phosphatebuffer.

2. Titration of fatty acids released. Suspensions of sub-cellular particles or pancreatic enzyme preparations wereincubated at a constant temperature of 38° C, and at apH maintained at 7.0 by the automatic addition of 0.2 to1 mMNaOH by a radiometer glass-electrode titratingmechanism. The volume of base required to maintain thepH per unit of time was observed, and this reflectedthe rate of acid release. Although several systems weretried as suspending media, including solutions of albu-min, gum arabic, and Tween 80, the most reproduciblemedium was found to be a solution of 0.005 M sodiumtaurocholate in 0.002 Mpotassium phosphate buffer at pH7.0. This medium permitted reasonably good dispersionof the substrates and good sensitivity to pH changeswithout undue instability of pH in control experiments.Series of experiments were carried out wvith variousglyceride substrates, including monoglycerides with vary-ing fatty acyl chain lengths, synthesized as describedabove. Comparisons were made of the rates of fattyacid release from the a- and ,-isomers of several mono-glycerides, of intestinal tissue preparations and crude pan-creatic lipase preparations, and of other pH levels. Re-sults were expressed as mAs equivalents of acid releasedper minute per milligram protein. Similar techniques ofcontinuous titration at constant pH have been previouslydescribed (17), and have the advantage of more closelyestimating the true initial reaction velocity of the ester-bound hydrolytic reaction.

RESULTS

ide products of the reaction as well as free gly-cerol, labeled glycerides were used in which theglycerol portion contained C14 in the 1- or 3-carbon. When suspensions of these radioactiveglycerides were incubated with subcellular frac-tions from intestinal epithelial cells, it was seenthat the monopalmitin-C14 was almost entirelyhydrolyzed, whereas very little hydrolysis of thetripalmitin-C14 or the dipalmitin-C14 occurred.When the glycerides obtained after incubation inthese experiments were separated on thin-layerplates of silicic acid, it was observed (Table I)that very little degradation had occurred of thetripalmitin to di- or monopalmitin, or of dipalmi-tin to monopalmitin. As might be expected inview of these results, it appeared that much of themonopalmitin produced from dipalmitin lipolysiswas further split to free glycerol and palmiticacid.

Repeated localization studies revealed consist-ently that of the cell fractions, the highest specificactivity in monoglyceride lipolysis was found inthe microsomes and the mitochondria, which wereseveral times more active than the whole homoge-nate or the cell sap. Because of the large amountof the cell protein in the cell sap, however, thetotal activity in this fraction was appreciable.When isolated brush borders were prepared, thesupernatant mixture resulting from sedimentationof the brush borders resembled the whole homoge-nate in its activity, whereas the activity in thebrush borders was slight. The specific and totalactivities are compared in Table II.

TABLE II

Intracellular localization of monoglyceride lipasein rat intestinal mucosa*

TotalSpecific Protein in activity

Fraction activity fraction in fraction

mimoles %of wholehydrolyzed homogenate

min/mgprotein

Whole homogenate 32.5 100. 100.Nuclei, unwashed 26.0 28.2 22.6Mitochondria 77.9 5.9 14.6Microsomes 67.5 10.3 21.4Cell sap 10.3 55.6 17.7

Sumof fractions 76.5

Isolated brush borders(prepared separately) 21.0 2.0 1.3

* Each incubation was carried out for 10 minutes at 38° C with 2Assay of glyceride hydrolysis with radioactive moles of monopalmitin-c'4 (112,000 cpm) and 0.4 to 1.0 mg tissuesubstrates. In order to measure the lower Oycer- protein in 5 per cent albumin at pH 7.0 in 0.15 Mpotassium phosphate,substrates. In order to measure the lower glycer- in a total volume of 2.0 ml.

189

JOHN R. SENIOR AND KURT J. ISSELBACHER

0.5

0.4

p MOLESMONOPALMITIN

SPLIT 0.3

0.2

0.1

05 6 7 8 9 10

pHFIG. 1. CURVEOF MONOGLYCERIDELIPASE ACTIVITY OF INTESTINAL MUCOSALMICRO-

SOMESAS A FUNCTION OF PH. Each point represents an incubation carried out for 30minutes at 380 C at the indicated pH, with 0.7 /Amole of mono-palmitin-C1' (40,000cpm) and 0.56 mg microsomal protein in 0.125 M Tris-maleate and 0.125 M potas-sium phosphate buffers. The total volume was 2.0 ml.

14J

IN

'4.

0.6

0.5

0.4

0.3

0.2

0.1

00 20 30

MINUTES

FIG. 2. RATE OF HYDROLYSISOF MONOPALMITINBY RATINTESTINAL EPITHELIAL CELL MICROSOMES. Each incu-bation was carried out for the indicated time at 38° C,with 1 ,umole of monopalmitin-C1' (56,000 cpm) and 0.4mg microsomal protein in 5 per cent albumin and 0.15 Mpotassium phosphate buffer, pH 7.0. The total volumewas 2.0 ml.

Although purification and isolation of the mono-glyceride lipase from the cell fractions was notextensively carried out, rough characterization ofsome of the properties of the enzyme in thesecrude fractions was attempted. The pH optimumfor the splitting of monopalmitin by intestinal epi-thelial cell microsomes was found to be about 7.8(Figure 1), although appreciable hydrolysis oc-cured between pH 7.0 and 8.5. Parallel experi-ments using no tissue preparation were carried outin order to estimate the relative amount of non-enzymatic hydrolysis; at the more alkaline pHvalues, increasing nonenzymatic hydrolysis bysaponification was noted. The addition of 20 mMsodium taurocholate did not produce a markedshift in the pH optimum of the microsomal mono-glyceride lipase, although it is known (18) tolower the optimal pH of pancreatic lipase by about2 pH units.

The rates of lipolysis were fairly linear over avariable range of enzyme concentrations whenthe amount of monopalmitin-C14 substrate wasconstant. A time study (Figure 2) revealed that

190

INTESTINAL MONOGLYCERIDELIPASE

the rate of monopalmitin-C14 hydrolysis by intes-tinal microsomes was linear for about 20 minutes.The temperature stability of the microsomal mono-glyceride lipolytic activity was noteworthy. Itwas repeatedly observed that tissue suspensionscould be kept for several days at 00 to 40 C with-out appreciable loss of activity, that only slightloss of activity occurred after 24 hours at roomtemperature, and that temperatures above 600 Cwere required for inactivation.

Addition of 20 mMsodium taurocholate pro-duced no enhancement of monoglyceride lipase ac-tivity in microsomes, mitochondria, or brush bor-ders from rat intestinal epithelial cells. In fact,some 30 to 40 per cent inhibition of the micro-somal activity was observed at this concentrationof the conjugated bile salt. Studies using tri- anddilinolein-C14 (labeled in glycerol portion) re-vealed that they were as resistant to lipolysis bymicrosomes as were the saturated higher glycer-ides, despite the much clearer dispersions yieldedby the unsaturated glycerides in albumin solution.

Study of glyceride hydrolysis by titration offatty acids released. In order to follow the lipoly-sis of glycerides over a period of time withoutterminating the reaction, it was feasible to meas-ure the fatty acids released by using an automatictitrating device at constant pH, measured by aglass electrode. This second, independent assaypermitted quantitation of the fatty acids releasedfrom ester linkage under initial reaction condi-tions, since the pH of the medium was maintainedconstant by addition of enough NaOH to neu-tralize the fatty acids liberated. Under these con-ditions, there was no accumulation of free, un-ionized fatty acids and H+ that would tend toreverse the reaction. This technique permitted as-say of a much wider range of substrates, since la-beled glyceride substrates were not necessary.Thus it was also possible to compare rates of hy-drolysis of the series of synthetic monoglycerideswith varying fatty acyl chain lengths. Titrationcould not be carried out in albumin solution, pre-sumably because of the strong buffering capacityof the protein. In the search for suitable media,10 per cent gum arabic was rejected because ofpH instability in the absence of any added enzyme;cell sap was found to have a low but definite lipo-lytic activity of its own; and Tween 80, an oleicacyl monoester of polyoxyethylene sorbitans, it-

self acted as a substrate. A dilute sodium tauro-cholate solution weakly buffered by phosphate wasfound to provide reasonable dispersions of sub-strates, nonenzymatic pH stability, and sufficientsensitivity to permit titration of the fatty acidsreleased.

At pH 7.0 and a constant temperature of 380C, the microsome-catalyzed hydrolysis of mono-glycerides was found to be maximal (Figure 3)when the fatty acyl chain length was 10 carbonatoms (monocaprin). Monoglycerides of shorteror longer side-chain lengths showed decreasedrates of splitting. Unsaturation of the side-chainseemed to have the same effect as shortening asaturated chain, a double bond having roughly theeffect of two fewer methylene groups. Thus,rates were comparable for monolinolein and mono-myristin, and for mono-olein and monopalmitin.

The effect of position of the fatty acyl groupwas estimated by comparison of the rates of hy-drolysis of the a- and ,-isomers of monopalmitin,mono-olein, and monostearin. Although theserates appeared to be about equal in suspendingmedia consisting of dilute cell sap, in experimentswith taurocholate in the medium, the ,B-isomersappeared to be hydrolyzed more slowly. Possibleisomerization of the p- to a-isomers during prepa-ration and assay of the substrates was not ex-cluded. Addition of crude pancreatic lipase, how-

N.

-4

14J

"It

CARBON ArOMS IN FArTY S/OF-CHAIN OFMONHLYCE//D

FIG. 3. RELATIVE RATES OF HYDROLYSIS OF MONO-GLYCERIDES OF VARYING FATTY ACYL CHAIN LENGTH BY

INTESTINAL MUCOSALCELL MICROSOMES. In each titrationat 380 C and constant pH 7.0, there were present 6 umolesof monoglyceride, 1.7 mg microsomal protein, 0.005 Msodium taurocholate, and 0.002 M potassium phosphate, inan initial volume of 6 ml.

60

50 -

40~~~~~~

50~~~~~~~~~~~

20

10

4 140 10 12 14 16 18

191

JOHN R. SENIOR AND KURT J. ISSELBACHER

TRIGL YCER/DES

DIGL YCERIDES

o ( R')CR") CH20-C vVWAAM

NV\AWA C-OCH110 CH20-C WAAAW

I 1i1o CR I

CH20H

RC-OC H11 IH o ol '0 CH20-CR

0

MONOGLYCER/DESCH20H CH20H

IsomerizotionHOCH - R c-C H

I~~~~~~~~11 1.CH20-CR" 0 CH20H

0

GLYCEROLCH20H

CHOH + HOOCR"

CH20H

FIG. 4. SCHEMEOF THE PROGRESSIVEHYDROLYSIS OF TRIGLYCERIDESBY PANCREATIC LIPASE. The fatty acyl side-chains are symbolized bythe zigzag lines and by the alkyl abbreviation R. The asterisks in-dicate reactions thought to be catalyzed by pancreatic lipase.

ever, accelerated the splitting of a-monoglyceridesbut not 8-isomers under similar conditions.

Results of inhibition studies revealed that themicrosomal monoglyceride lipase was essentiallyuninhibited by 0.01 Mdisodium EDTA, but onlyslightly inhibited by potassium fluoride at thesame concentration. Because of the very crudeenzyme preparations, no substrate-inhibitor curveswere attempted. No acceleration of the rate ofmonoglyceride hydrolysis was observed when cal-cium ions were added in a concentration of 0.005Mper L.

DISCUSSION

The question of whether an intestinal lipase ac-tually exists has been raised for many years, andindeed, the studies Schiff (8) reported 70 yearsago on pancreatectomized dogs suggested thatthere probably was such an enzyme. The greatabundance of highly active pancreatic lipase bath-ing the intestinal mucosa has previously obscuredthe demonstration of a separate intestinal lipase.Evidence for such an enzyme has continued to ac-

cumulate, however, and recently some of the mostconvincing work has shown pronounced lipolyticactivity in subcellular particles obtained from hogduodenum (9). In those studies, the intestinallipase was distinctly different from pancreatic li-pase in its substrate specificity and response toinhibitors, and considerable care was taken toavoid contamination of the intestinal tissue prepa-rations with pancreatic fluid material. Otherstudies by Tidwell and Johnston (10), usingeverted sacs of hamster small gut in buffered al-bumin media containing various suspended glycer-ides, have indicated that preferential splitting ofthe monoglycerides occurred, with much less hy-drolysis of the di- and triglycerides.

During digestion of dietary fats, pancreaticlipase has been shown to attack triglycerides mostrapidly, diglycerides slightly less rapidly, and tohave a sluggish effect on the monoglyceridesformed (19). Furthermore, pancreatic lipase hasbeen found to catalyze hydrolysis of the esterbonds at the outer, primary (a and a') alcoholgroups of the glycerol, and to yield /3-monoglycer-

1*+ HoocR'

I*+ HOOCR"'

192

INTESTINAL MONOGLYCERIDELIPASE

ides as a most important product (20). Althoughisomerization of the 8- to a-monoglycerides canoccur spontaneously, allowing pancreatic lipaseto complete the hydrolysis, this third stage pro-ceeds much more slowly, and considerable amountsof monoglycerides are therefore available for in-testinal absorption. This sequence is summarizedin Figure 4.

Evidence that monoglycerides may actually beabsorbed has been provided by the double-labelstudies of monoglycerides and triglycerides inwhich the glycerol portion and fatty acid portionwere labeled with different isotopes (21, 22).Whereas some of the absorbed monoglyceridesappeared in the lymph as triglycerides after re-esterification by the intestinal mucosa, consider-able amounts of the monoglycerides appeared tobe split, as indicated by a fall in the ratio ofglycerol to fatty acid isotope in the lymph tri-glycerides, compared to the original monoglycerideisotopic ratios. Furthermore, in vitro splitting ofmonoglycerides by intestinal epithelial cell prepa-rations has recently been observed in the courseof experiments on esterification of monoglycerides(23, 24).

In the present series of experiments, it washoped that methods could be developed to demon-strate, under conditions reasonably approximatingthose in the mammalian body, whether an intesti-nal lipase distinct and different from pancreaticlipase exists. Therefore, unphysiologic media,such as those containing synthetic detergents orhigh degrees of alkalinity, were avoided, as wereassays depending on such difficult-to-interprettechniques as clearing of turbidity. Efforts weremade to exclude any contamination by traces ofpancreatic lipase, insofar at least as this could beaccomplished by washing the mucosa and thenthe subcellular particles before assay. It mustbe admitted that some pancreatic lipase couldadhere to the cell fractions and especially to thebrush borders, but the latter contained a negligibleamount of the total lipolytic activity of the gutepithelial cells. Furthermore, the most highlywashed fractions, namely, microsomes and mito-chondria, showed the highest specific activities andhad a substrate specificity distinct from that ofpancreatic lipase.

Results by both of the methods used in thepresent experiments were in agreement, and indi-

cated that indeed a lipase exists within the in-testinal epithelial cells and has a great specificityfor monoglycerides. Pancreatic lipase, on theother hand, has a maximal activity towards tri-glycerides, somewhat less towards diglycerides,and is least active with monoglycerides, especiallythe /8-monoglycerides as substrates. If indeedpancreatic lipase acts strictly upon the primaryalcohol ester bonds, this substrate specificity wouldbe expected, for the triglycerides have two suchbonds, the a-f3-diglycerides one, and f3-monoglycer-ides none. These isomers of the lower glyceridesare those actually formed during digestion of tri-glycerides by pancreatic lipase (20, 25).

The intracellular localization of maximal spe-cific activity of the intestinal monoglyceride lipaseto the microsomes and mitochondria was a con-sistent finding; whether or not the lower level ofenzyme activity in the cell sap represented anartifactual "solubilization" of the enzyme fromthe membranous structures during preparationand differential centrifugation could not be de-termined. The unimpressive degree of activity ofthe isolated brush borders was another indicationof the difference between digestion of fat and thatof carbohydrates and proteins. In the latter, theproducts of intraluminal digestion are water-solu-ble, and have been shown to undergo final diges-tion at the surface of the intestinal cells by disac-charidases and oligopeptidases localized in thebrush border (3, 6). The major products of fatdigestion, however, free fatty acids and mono-glycerides, are not water-soluble to any large ex-tent, and it is not surprising that other mechanismsmay exist for the handling of these substances.In this connection, it is pertinent that several ofthe enzymes necessary for chemical transforma-tions of the absorbed fatty acids and monoglycer-ides to higher glycerides seem to be localized pri-marily in the microsomal fraction of the intestinalepithelial cell, and to some extent in the mitochon-dria (23, 24, 26).

Better characterization of the properties of theintestinal monoglyceride lipase will perhaps bepossible when isolation and purification of the en-zyme from the crude tissue fractions can be ac-complished. The definition of this enzyme as alipase depends on its substrate specificity, al-though it could be argued that in view of the maxi-mal hydrolytic specificity for monocaprin, it

193

JOHN R. SENIOR AND KURT J. ISSELBACHER

should be called an esterase. The truly water-soluble, short-chain monoglycerides, however,were less rapidly split than the intermediate-lengthmonoglycerides and the long-chain unsaturatedmonoglycerides.

The data presented correlate with observationsof Tidwell and Johnston (10) and with the veryrecent studies from the same laboratory in whichPope, Askins, and McPherson (27) have de-scribed a lipase in hamster intestinal homogenateswith a definite monoglyceride specificity.

Whether the low activity of the intestinal lipasetowards tri- and diglycerides is a function of thepoor solubility of these compounds in aqueousmedia has not been resolved by the present stud-ies. In contrast to our findings, DiNella, Meng,and Park (9) found hog intestinal mucosal lipaseto hydrolyze ester bonds of tri- and diolein asrapidly as mono-olein. Those studies, however,were carried out under quite different conditions,i.e., with media containing Span 20 11 at pH 9.0.The crucial questions yet to be settled relate tothe importance of the physical state of the sub-strate and the conditions existing in the medium.It is noteworthy that pancreatic lipase, which hasbeen purified and characterized (28) to a muchgreater degree than intestinal lipase, shows higheractivity when its substrate is water-insoluble,seems to act at lipid-aqueous interfaces, and infact is inhibited if its substrate is put into truesolution.

Whereas the physiologic importance of pan-creatic lipase in intraluminal fat digestion is wellestablished, the role and significance of the in-testinal lipase is yet to be determined. It seemsreasonable, however, that the presence of an en-zyme system in the cell active in the hydrolysis ofmonoglycerides provides a mechanism for com-pleting the digestion of glycerides after their ab-sorption from the lumen of the intestine. In viewof the present findings and the previous demon-stration of the ability of the intestinal cell to con-vert monoglycerides to higher glycerides (24), abalance appears to exist within the mucosal cell be-tween monoglyceride hydrolysis and esterifica-tion to di- and triglycerides. The mechanism(s)regulating these two pathways remains to be elu-cidated.

11 Sorbitan monolaurate.

SUMMARY

An intestinal lipase has been demonstrated in ho-mogenates and subcellular particles of rat intesti-nal epithelial cells. This lipase exhibits prominentspecificity for monoglycerides, and is much lessactive upon the di- or triglycerides of long-clhainfatty acids. A maximal hydrolytic activity wasfound for monoglycerides of intermediate chainlength, i.e., from 8 to 12 carbon atoms in thesaturated fatty acyl chain. Unsaturated long-chain monoglycerides were split at rates compara-ble to those for saturated monoglycerides ofshorter chain length. A survey of the cell frac-tions showed the monoglyceride lipase activity tobe concentrated primarily in the microsomal andmitochondrial fractions, with considerably lowerspecific activities present in the cell sap and iso-lated brush borders. Fluoride and EDTA(ethylenediamine tetraacetic acid) did not in-hibit the enzyme activity.

The intestinal monoglyceride lipase was foundto be distinct and different from pancreatic lipasein a) its specificity for monoglycerides rather thantriglycerides, b) its ability to catalyze hydrolysisof ester bonds on the ,B- as well as the a-hydroxylgroups of glycerides, and c) its activity upon dis-solved rather than suspended substrates. Otherdifferences appear to exist in the effect of inhibi-tors, pH optima, and chain-length specificity, butfurther studies are necessary when purer enzymepreparations are obtained.

The physiologic importance of the intestinalmonoglyceride lipase is not yet known. It may,however, function in the intracellular completionof fat digestion before the synthesis of chylomi-crons by the intestinal mucosa.

REFERENCES

1. Senior, J. R., and Isselbacher, K. J. Studies of anintracellular glyceride lipase of rat intestinal epi-thelial cells (abstract). Clin. Res. 1961, 9, 333.

2. Newey, H., and Smyth, D. H. Intracellular hydroly-sis of dipeptides during intestinal absorption. J.Physiol. 1960, 152, 367.

3. Holt, J. H., and Miller, D. The localization of phos-phomonoesterase and aminopeptidase in brushborders isolated from intestinal epithelial cells.Biochim. biophys. Acta (Amst.) 1962, 58, 239.

4. Fridhandler, L., and Quastel, J. H. Absorption ofsugars from isolated surviving intestine. Arch.Biochem. 1955, 56, 412.

194

INTESTINAL MONOGLYCERIDELIPASE

5. Wilson, T. H., and Vincent, T. N. Absorption ofsugars in vitro by the intestine of the golden ham-ster. J. biol. Chem. 1955, 216, 851.

6. Miller, D., and Crane, R. K. The digestive functionof the epithelium of the small intestine. II. Locali-zation of disaccharide hydrolysis in the isolatedbrush border portion of intestinal epithelial cells.Biochim. biophys. Acta (Amst.) 1961, 52, 293.

7. Mattson, F. H., Benedict, J. H., Martin, J. B., andBeck, L. W. Intermediates formed during the di-gestion of triglycerides. J. Nutr. 1952, 48, 335.

8. Schiff, M. Le suc intestinal des mammiferes, commeagent de la digestion. Arch. Physiol. norm. Path.1892, 4, 699.

9. DiNella, R. R., Meng, H. C., and Park, C. R. Prop-erties of intestinal lipase. J. biol. Chem. 1960,235, 3076.

10. Tidwell, H. C., and Johnston, J. M. An in vitrostudy of glyceride absorption. Arch. Biochem.1960, 89, 79.

11. Baer, E., and Fischer, H. 0. L. Synthesis of ahomologous series of optically active normal ali-phatic a-monoglycerides (L-series). J. Amer.chem. Soc. 1945, 67, 2031.

12. Newman, M. S., and Renoll, M. Improved prepara-tion of isopropylidene glycerol. J. Amer. chem.Soc. 1945, 67, 1621.

13. Bourne, E. J., Stacey, M., Tatlow, J. C., and Tedder,J. M. Studies on trifluoroacetic acid. I. Trifluoro-acetic acid anhydride as promoter of ester formationbetween hydroxy compounds and carboxylic acids.J. chem. Soc. 1949, 2976.

14. Lowry, 0. H., Rosebrough, N. J., Farr, A. L., andRandall, R. J. Protein measurement with theFolin phenol reagent. J. biol. Chem. 1951, 193,265.

15. Folch, J., Lees, M., and Sloane Stanley, G. H. Asimple method for the isolation and purification oftotal lipides from animal tissues. J. biol. Chem.1957, 226, 497.

16. Cohen, P. P. Methods for preparation and study oftissue in Manometric Techniques, W. W. Umbreit,

R. H. Burris, and J. E. Stauffer, Eds. Minneapo-lis, Burgess, 1959, p. 149.

17. Desnuelle, P., Constatin, M. J., and Baldy, J. Tech-nique potentiometrique pour mesure de l'activitede la lipase pancreatique. Bull. Soc. Chim. biol.(Paris) 1955, 37, 285.

18. Borgstr6m, B. On the mechanism of pancreaticlipolysis of glycerides. Biochim. biophys. Acta(Amst.) 1954, 13, 491.

19. Desnuelle, P., Naudet, M., and Rouzier, J. Sur laformation de glycerides partiels au cours de l'hy-drolyse des triglycerides par la lipase pancreatique.C. R. Soc. Biol. (Paris) 1947, 141, 1242.

20. Mattson, F. H., and Beck, L. W. The specificity ofpancreatic lipase for the primary hydroxyl groupsof glycerides. J. biol. Chem. 1956, 219, 735.

21. Reiser, R., Bryson, M. J., Carr, M. J., and Kuiken,K. A. The intestinal absorption of triglycerides.J. biol. Chem. 1952, 194, 131.

22. Skipski, V. P., Morehouse, M. G., and Deuel, H. J.,Jr. The absorption in the rat of a 1,3-dioleyl-2-deuteriostearyl glyceride-C14 and a 1-monodeu-teriostearyl glyceride-C14. Arch. Biochem. 1959,81, 93.

23. Senior, J. R., and Isselbacher, K. J. Direct esteri-fication of monoglycerides with palmityl coen-zyme A by intestinal epithelial subcellular frac-tions. J. biol. Chem. 1962, 237, 1454.

24. Johnston, J. M., and Brown, J. L. The intestinal uti-lization of doubly-labeled a-monopalmitin. Bio-chim. biophys. Acta (Amst.) 1962, 59, 500.

25. Schonheyder, F., and Volqvartz, K. Studies on thelipolytic enzyme action. III. Hydrolysis of tri-propionyl glycerol. Biochim. biophys. Acta(Amst.) 1952, 8, 407.

26. Senior, J. R., and Isselbacher, K. J. Activation oflong-chain fatty acids by rat-gut mucosa. Bio-chim. biophys. Acta (Amst.) 1960, 44, 399.

27. Pope, J. L., Askins, R. E., and McPherson, J. C.The specificity of an intestinal lipase for mono-glycerides. Fed. Proc. 1962, 21, 259.

28. Desnuelle, P. Pancreatic lipase. Advanc. Enzymol.1961, 23, 129.

195