Journal of Clinical InxestigationVol. 45, No. 3, 1966

Intestinal Absorption of Sucrose in Man: Interrelation ofHydrolysis and Monosaccharide Product Absorption *

GARYM. GRAYt ANDFRANZJ. INGELFINGER(From the Evans Memorial Department of Clinical Research, University Hospital, and the

Department of Medicine, Boston University School of Medicine, Boston UniversityMedical Center, Boston, Mass.)

Disaccharides are hydrolyzed by their specificenzymes present in the intestinal mucosa (1-3).Although some current textbooks still state thatthese enzymes are secreted into the intestinal lu-men (4-7), the concentrations of monosaccharideproducts free in intestinal contents during disac-charide absorption in vitro (8-10) have been foundinsufficient to support the concept of intraluminalhydrolysis. In addition, the low disaccharidaseactivity of intestinal contents during the absorp-tion process in vivo (11-13) strongly suggeststhat the disaccharide either enters the cell beforebeing hydrolyzed or else is hydrolyzed at the cellsurface by mucosa-bound enzyme. The releasedmonosaccharide products presumedly are thentransported across the intestinal cell (8, 14, 15).

Little information is available that relates disac-charide hydrolysis to absorption of the componentmonosaccharides. Wilson and Vincent (14) com-mented on the accumulation of monosaccharides inthe mucosal medium during the process of disac-charide absorption in hamster gut sac preparations,and Dahlqvist and Thomson have reported thatlarge amounts of free fructose accumulate intra-luminally during sucrose absorption in the intactrat (16). Despite these findings in animals,Dahlqvist and Borgstr6m found little free intra-luminal monosaccharides during the process ofdisaccharide absorption in man (12). It was re-cently demonstrated, however, that appreciable

*Submitted for publication June 10, 1965; acceptedDecember 2, 1965.

Supported in part by U. S. Public Health Service re-search grant AM03560-04 and 05 and by training grantT1 AM 5025-07 and 08 from the National Institute ofArthritis and Metabolic Diseases.

Presented in part at the Annual Meeting of the Amer-ican Society for Clinical Investigation, Atlantic City,N. J., May 3, 1964.

t Address requests for reprints to Dr. Gary M. Gray,U. S. Army Tropical Research Medical Laboratory,APO, New York, N. Y. 09851.

amounts of the hydrolysis products accumulateintraluminally during the process of sucrose ab-sorption in man and that these monosaccharidesappear to move back from their mucosal site of hy-drolysis to the lumen (13).

The present work is concerned with the rela-tion of hydrolysis of sucrose to the absorption ofits monosaccharide components, glucose, an ac-tively absorbed monosaccharide (14, 15, 17), andfructose, which is passively absorbed (14, 15).

Methods

Thirty-two normal young subjects were studied on 105occasions by use of a double-lumen tube placed at variouslevels of the intestine (13, 18). Polyethlene glycol 4000(PEG) was used as the nonabsorbable water-solublemarker and was determined by a modification of Hyd&i'smethod (19). Sucrose solution or an equivalent mix-ture of glucose and fructose made isotonic with NaCl(290 ± 10 mOsmper L) was infused at 15 ml per minutethrough the proximal orifice of the tube. In some stud-ies, galactose was also infused. Intestinal samples werecollected by siphonage from the distal orifice, which waslocated 15 or 30 cm from the site of infusion. The ini-tial 30 minutes of an infusion period allowed steady stateconditions to be approached so that, thereafter, concen-tration of PEG and the test sugar or sugars in succes-sively collected samples showed little variation (13).When successive experiments were performed in thesame intestinal segment, the equilibration period alsoserved to prevent the contamination of one test by theresiduals of a preceding test. Absorption rates deter-mined by infusing a specific solution changed little (mean± 10%) when experiments were repeated during thecourse of a 10-hour period; thus there was no evidence of"fatigue" of the intestinal segment under study, and ab-sorption of different carbohydrate solutions could becompared. The test solutions were infused in randomorder, and specimens obtained from the distal orifice werecollected by use of the precautions previously outlined inorder to insure stability of the carbohydrates (13).

Assay of carbohydrates in intestinal samples. AfterBa(OH)a-ZnSO4 deproteinization (20), sucrose,' glu-

1 a-D-Glucopyranosyl-P-D-fructofuranoside, analyticalgrade, Merck Co., Rahway, N. J.

388

INTESTINAL ABSORPTIONOF SUCROSEIN MAN

cose,2 and total hexose (glucose plus fructose) 3 weredetermined by the specific enzymatic methods previouslydescribed (13). Glucose was substracted from totalhexose to determine fructose.

In some of the samples containing only fructose andglucose, fructose was analyzed by Dische and Devi'sketohexose method (21); values obtained were within3% of those found by use of the enzymatic method.

Galactose 4 was assayed by a modification of the galac-tose oxidase assay described by Avigad, Amaral, Asensio,and Horecker (22). The reagent consisted of 125 Ugalactose oxidase,5 3 mg peroxidase,6 0.6 ml of 1%o-dianisidine in 95% ethanol, 100,000 U catalase,5 and 65ml of 0.5 M Tris buffer (23) at pH 7.0. It was neces-sary to prepare the reagent daily since the chromogenrapidly became colored. To 1 ml of deproteinized sample2.5 ml of the reagent was added. The reaction mixturewas incubated at 370 C for 20 minutes and the reactionstopped with 0.1 ml of 1 N H2SOs Extinction was de-termined in a spectrophotometer at 395 m/A and a 1-cmlight path used. A linear relationship of optical densityto concentration occurred when 50 to 125 ,ug of galactosewas in the cuvette. Amounts of galactose less than 50/Ag produced little absorbance. Sucrose, glucose, andfructose in 25 times the concentration of galactose readthe same as the reagent blank and did not interfere withthe reaction. Recovery of galactose added to intestinalcontents was 100 ± 4%.

Calculations. Sucrose disappearance over the intesti-nal segment may be considered as equal to the sucrosehydrolyzed, since appreciable amounts of sucrose do notdisappear by absorption into blood of the intact disac-charide; this was discussed in a previous report (13).Considerable quantities of the monosaccharide hydroly-sis products of sucrose accumulate intraluminally duringthe absorption process, and therefore amounts of themonosaccharide found at the collecting orifice must besubtracted from the sucrose hydrolyzed (sucrose thatdisappeared) to determine absorption (13).Sucrose hydrolysis=2[S1 V-Se-V-(P1/P.)], and [1]sucrose absorption = 2 [S. V - S.eV- (P1/P.) ]

- [(Ge + F) V (P1/Pe)J, [2]where hydrolysis and absorption are expressed as mil-limoles monosaccharide per hour, and symbols refer tomillimolar concentrations as follows: St = sucrose in in-fusion; S. = sucrose in collected samples; P1 = PEG ininfusion; P. = PEG in collected samples; G. = glucosein collected samples; F. =fructose in collected samples.V= volume (liters infused in 1 hour).

2 D-Glucose, analytical grade, Merck Co., Rahway, N. J.3.D-Fructose, C. P. Pfanstiehl Co., Waukegan, Ill.4 D-Galactose, C. P. Pfanstiehl Co., Waukegan, Ill.

Although all other carbohydrates used were chromato-graphically pure (less than 1% impurity), D-galactosecontained 5% D-glucose (as measured by glucose oxidaseand hexokinase reagents) and a trace of an oligosac-charide.

6 Worthington Biochemical Corp., Freehold, N. J.INutritional Biochemicals Corp., Cleveland, Ohio.

Sucrose absorption can also be expressed in terms ofthe individual monosaccharides:Glucose product absorption

= [Si V-S,-V-(Pi/P,)J - [G..V'(Pi/P,)], [3]andfructose product absorption

= [S1-V-Se-V-(P1/P.) ]-[F.*V (P1/P.)]. [4]The use of Equations 3 and 4 is necessary if absorption

of sucrose is to be compared with that of an equivalentglucose-fructose mixture.

Results

Comparison of absorption of sucrose versus glu-cose-fructose mixture. To investigate the relationbetween the hydrolysis of the disaccharide andtransport of its monosaccharides, we undertookpaired experiments comparing absorption from su-crose (73 mM) with that from an equivalent mix-ture containing glucose plus fructose (73 mMeachmonosaccharide). Experiments at a given intesti-nal level were performed over successive 60-minuteintervals and the solutions infused in randomorder. A 30-cm distance between infusion orificeand collection orifice was used. As shown in Ta-ble I, glucose and fructose absorption rates werenot significantly different whether the disaccharidewas infused (mean: glucose, 35 mmoles per hour;fructose, 21 mmoles per hour) or the monosac-charide mixture was infused (mean: glucose, 39mmoles per hour; fructose, 25 mmoles per hour).Also, a similar relation prevailed in 13 pairedstudies when 29 mMsugars were infused. Thisis summarized in Table I.

Figure 1 relates glucose absorption rates in in-dividual paired experiments and demonstrates thatthere is close correlation to linearity (r = 0.87,p < 0.001) as well as to the theoretical line repre-senting identical absorption from either infusionsolution. A significant correlation, but of some-what lower order, was also found for fructose ab-sorption (r = 0.62, p < 0.01) (Figure 2).

Effect of galactose on sucrose hydrolysis andabsorption. In our experiments the rate of su-crose hydrolysis was greatly different in jejunumand ileum (Table I) (13). Despite this, similaramounts of monosaccharide products accumulatedat all levels of intestine (Figure 3), suggesting thepossibility that the hydrolytic process was inhibitedin both jejunum and ileum once the intraluminalconcentration of monosaccharides reached a criti-

389

GARYM. GRAYAND FRANZ J. INGELFINGER

TABLE I

Comparison of absorption rates of sucrose with equimolar mixture of glucose plus fructose*

Glucose 73 mM+ fructose§Sucrose 73 mM§ 73 mM

Intestinal Sucrose Glucose Fructose Glucose FructoseSubjectt level: hydrolysis absorption absorption absorption absorption

cm mmokes disac- mmoks/hr mmoles/hrcharide/hr

P.C. 100 23 22 7 24 18G.Ha.a 95 32 30 26 33 22G.Ha.b 135 54 52 38 49 33D.Z. 120 33 27 13 52 36G.Ha.IIa 112 37 34 22 50 32G.Ha.IIb 240 41 25 11 32 20W.C.a 160 48 46 34 46 30W.C.b 232 22 20 9 26 18O.H.a 115 48 47 38 48 33O.H.a 130 51 49 41 53 17O.H.b 240 20 19 11 31 16D.Z.a 165 38 32 12 32 21D.Z.b 210 30 26 7 22 14D.Go.a 145 38 36 19 43 28D.Go.b 210 34 31 22 38 26H.L.a 150 50 45 26 41 27H.L.b 200 32 29 15 28 19H.L.c 255 31 27 12 26 23L.R. 140 52 51 27 48 34A.Q.a 140 63 62 46 61 38A.Q.b 230 31 25 9 27 16

Mean i 2 SE 39 4t 6 35 4 6 21 4 6 39 i 5 25 4 4

* Thirty-cm tube distance from infusion to collection orifice.t Roman numerals after subjects' initials refer to a second study session 4 or more weeks after the initial experiment,

and letters indicate first (a), second (b), or third (c) day of a study session.T Centimeters from incisor teeth to collecting orifice: 110 to 170 cm = jejunum; 170 to 260 cm = ileum.§ Glucose and fructose absorption not significantly different whether disaccharide or monosaccharide mixture in-

fused (p > 0.3).

90-

80-

Omw zV 70-LL

X 60

I-°

mD = 40-

30

(IW

OU)

() 0 20-

J D10_

.0

0SOD0 A

0 0

lb io 3o 40 50 6o io go go

GLUCOSE ABSORPTION FROM SUCROSE.

(mMoles/hr)

FIG. 1. CORRELATIONOF GLUCOSEABSORPTIONFROMSU-CROSE (73 MM) WITH ABSORPTIONFROMAN EQUIVALENTMONOSACCHARIDEMIXTURECONTAINING GLUCOSE(73 MM)AND FRUCTOSE (73 MM) (r =.87, p < 0.001). Each plotrepresents results of a paired experiment in a single sub-ject, as related to the theoretical line representing equalabsorption from each solution.

cal level. Preliminary experiments comparing theeffect of added glucose and fructose on hydrolysisrates of sucrose by use of the 30-cm infusion-to-collection distance did not show definite inhibition,and the added glucose made it impossible to de-termine glucose product absorption from the su-

crose. Further, these 30-cm segment studiesshowed very efficient hydrolysis (greater than80%o) when sucrose (29 mM) alone was infused,and it was possible that the intestine's maximalcapacity was not really being determined. If thiswere the case, inhibition by the added monosac-

TABLE II

Comparison of sucrose to glucose-fructose in29 mMstudies*

Sucrose Glucose +fructose

No. of Glucose Fructose Glucose Fructosestudies absorption absorption absorption absorption

mmoles/hr mmoles/hr13 19-42 1143 22:142 11:4:2

* A 30-cm segment was used. Values expressed as mean -I- 2 SE.

390

INTESTINAL ABSORPTIONOF SUCROSEIN MAN

91

cO =

LL0

Z EIs Wp w

Cc °

00iL 0

a

71

6

5

4

2

0-

00

10

io-oo

io-

0

0

0~~~~~0

00 0

000

0

00

0 20 30 40 0 60 70 80 90

FRUCTOSE ABSORPTION FROM SUCROSE(mMoles/hr)

FIG. 2. Comi TION OF FRUCTOSE ABSORPTION IN THESAMEMANNERAS IN FIGURE4 (r .62, p < 0.001).

charides might have been compensated for by thereserve hydrolytic activity of the intestine. There-fore a shorter intestinal segment (15 cm) wasperfused, and galactose was substituted for glu-cose as a possible inhibitor, for galactose has achemical structure closely related to glucose andmakes, use of the same mechanism for active trans-port across intestinal mucosa (14, 15, 24-27).

Paired experiments demonstrated that hydroly-sis rates of sucrose were 20% lower when galac-tose was in the test solution than when sucrosewas the sole sugar (Table III). The differencesare statistically significant (paired analysis, t test,p < 0.02). Glucose product absorption was also

IJJ

°IaJ

<a.

I 20

o0z0

36-

34.

32-

30-

28-

26-

24-

22-

20.

is.

16-

4-

2-

10-

8-

6-

4.

2-

.62

IOFUSKNs SWUROSE73 8446 MM.sFRUlCTOSE

° *.GLUGOSE

@ 0 0

00

0

0 0 00

0 0 0 0 0

0 0

0o0oTo

000 0 OS 0

O~~~ ~~~*0 0 0 0 0 0 0

0

000 @0 0 0*0

*

*9

*.

**

.

*

*

50o 00 20 140 60 10o 260 220 240LEVEL OF INTESTINE (GM. FROM TEETH)

260

FIG. 3. INTRALUMINAL ACCUMULATIONOF MONOSAC-CHARIDES DURING SUCROSEABSORPTIONAS RELATED TO IN-TESTINAL SITE. Both 73 mMand 146 mMstudies aregrouped, since values were not significantly different

decreased (p < 0.02), but fructose product absorp-tion did not change (p > 0.2).

Effect of infusion concentration on absorptionrates. The process of active absorption of glucoseand galactose in animal preparations appears toconform to Michaelis-Menten kinetics (15, 25,28), and some information concerning the kineticsof glucose absorption in man has been recently

TABLE HI

Effect of galactose on sucrose hydrolysis and absorption*

Sucrose 40 mM Sucrose 40 mM+ galactose 40 mMGlucose Fructose Fructose Galactose

Intestinal absorp- absorp- Glucose ab- absorp- absorp-Subjectt levelt Hydrolysis tion tion Hydrolysis: sorptiont tion tion

cm mmois mmoics/hr imoaes %A mmoks/hr %A mmin es/hrdisac- disac-

charide/hr charide/hrD.Go.IIa 105 25 20 21 18 -28 16 -20 14 16D.Go.IIb 210 12 10 3.7 11 - 8 7.6 -24 6.0 9.5V.W. 130 26 23 13 22 -is 20 -13 8.0 8.5J.C.a 145 26 23 17 22 -15 19 -17 17 15J.C.b 220 9.0 8.5 5.8 7.9 -12 6.5 -24 5.6 10A.B.a 160 13 12 3.4 10 -23 6.4 -47 4.3 7.0A.B.b 250 27 21 5 17 -37 13 -38 I 10Mean 20 17 11 15 -20 13 -20 95 11

* ;z-u n A- An. _ 4ron gnun r_a i ection onnce

391

- iviteen-cm tube custance from ifusion to collection orifice.vSee Table I for explanation.

sBoth sucrose hydrolysis and glucose absorption were significantly lower when galactose was in the infusion (paired analysis, t test, p <0.02).Fructose absorption did not change significantly (p > 0.2).5 Specimen lost; mean fructose values therefore based on six studies only.

0o a

0o

To

To T 0o

0

p

0 0 0

0o 0

0o

GARYM. GRAYAND FRANZ J. INGELFINGER

.0) - C4 C4 - C~~0 00 C4

N~~~~

m NmNt

...N m00N 0400- N N4fo

.0) -4: C4 4m4 -4C 4WC400o r4

7.0

N-

2-

E~ N

.ll.N

02~2

0 00 0 000 C . e' e

asU N W) InX 04 - or On X_ Co00 00000ok '4C'.00 C' -00O+C'40s

_ It _i i IN -i ll

_%OC')es4 00

C'j%0r OO Oo + e o +o '

C0 It V) R)e e U t% CuAie

0NO 00 0 0 00 0 0o 0t

'oo14 t-t0'000

O co NO W) mu0

0sC Ct C4 -0 - 04

O Cd0 0 0 0 0 14 0 0 )t')000%lo0000000000' 0v t,.te.

C'I. 00 C00- oN0 'C' 00 0

Cs4

-- N _ - N _ - - _ ^ -_C8-- O. .4-.

C.

reported (29-31). To determine whether thereis any kinetic relation of concentration of disac-charide to its rates of hydrolysis and absorption,we perfused solutions containing 10, 20, 40, 80,and 160 mMsucrose through 15-cm study seg-ments. These solutions were given in random or-der to a subject over a single day. The relationsobtained are shown in Table IV and representativedata plotted in Figure 4. There appeared to be a

> limiting velocity for both hydrolysis and absorp-S tion in all experiments. Figure 5 shows the Line-"° weaver-Burk plot (32) for sucrose hydrolysis;R there is no significant deviation from linearity

(analysis of variance, p < 0.001). An almostidentical curve can also be shown for glucose prod- -

*20 uct absorption. The kinetics of these two proc-esses are compared at different intestinal sites inTable V. Fructose product absorption conformedpoorly to Michaelis-Menten kinetics and is notrepresented graphically.

930:0CdO

06bODiscussion

UI i~When the absorption of sucrose was consideredin terms of its component monosaccharides,7 glu-cose was absorbed considerably more rapidly thanfructose, and sucrose hydrolysis exceeded glucoseabsorption only slightly (Table I). Figure 6 dia-grams what appears to occur during sucrose ab-sorption in man, the mean results of infusing a 73

S mMsolution being shown. Of 66 mmoles of su-.2 crose infused in 1 hour, 39 mmoles was hydrolyzed

to the monosaccharide products. Of the 39mmoles of glucose released, 35 was absorbed,

08 whereas 4 was present or accumulated in the in-7 Estimation of the sucrose absorption process in these

* terms depends upon the assumption that each molecule< of sucrose is split by sucrase to one molecule of glucosem and one of fructose. As well as having hydrolytic ac-, tivity, all sucrases that have been purified thus far alsor have been found to catalyze the formation of small^ amounts of oligosaccharides when another sugar, ratherg than water, is used as a receptor (33-35). With mam-a malian sucrase, this transglycosylation process may causeE a net release of smaller amounts of free glucose than

fructose (36). Table VI demonstrates that such sucraseB activity as there is in intestinal contents results in re-

lease of essentially equal amounts of glucose and fruc-tose. Furthermore, analysis of collected intestinal sam-

2 ples by paper chromatography revealed no sugars other>, than sucrose, glucose, and fructose. Thus sucrose is split* to essentially equal amounts of glucose and fructose un-

der the conditions of our study.

392

2

I1

'00)

20)

oa

S.2

en

X

0

v

0).

b

U2

0)

0

00

.0)4

4.)

A,82.0

CO

.0)

(.0

,u X

.00

u0)0.0

CO

* 0

CO

o

0

.0

i) b-..

cO

000.0

0),v00.0

04-)

esd

0

.4-

e 0)O0

00.

INTESTINAL ABSORPTIONOF SUCROSEIN MAN

MA ]a30-

20-

10-

b '-o' do io o' b 'ik 1,Ao5

40

30-

20-

10-

-.* SUCROSE HYDROLYSIS

-- GLUJCOSE PRODUCT ABSORPTION

o FRUCTOSE PRODUCT ABSORPTION

DG llb

10 30 io70 90 1 130 150

AH10 30o 7070g I9 130 110

50-

40-

30-

20-

10-

RK

10 90i 50. 1 7'0 1 9'0 111 1 0 15

DG ia

SP

I

10 50 T0 90 110 130 150

MEAN INTRALUMINAL SUCROSE (mMolar)

FIG. 4. RELATION OF MEANINTRALUMINAL SUCROSECONCENTRATIONTO HYDROLY-

SIS OF SUCROSEAND ABSORPTION RATES OF ITS GLUCOSEAND FRUCTOSEPRODUCTS. A15-cm test segment was used.

traluminal contents. In contrast, of the 39 mmoles,of fructose released, only 21 was absorbed, and 18accumulated intraluminally. Thus, nearly half-of the fructose released from hydrolysis of sucrose

was still present intraluminally at the point of thecollection orifice. This reflects the slow absorp-tion rate of fructose as compared to glucose (14,17, 31, 37). These values for glucose and fruc-

tose accumulation are minimal values, since ab-sorption of released monosaccharides presumablyoccurs over the whole length of the segment under-study.

The magnitude of this intraluminal accumula-tion of monosaccharides is interesting in view of

the current belief that the disaccharides enter theintestinal cell before being hydrolyzed (1, 9, 10,12, 16, 38-41). Eichholz and Crane have recentlylocalized sucrase to the plasma-membrane of themicrovillus (42), and Crane has set up a hypo-thetical model for intestinal transport of sugars

in which hydrolysis of disaccharides occurs withinthe outer portion of brush border plasma-mem-brane whereas the released monosaccharides are

transported across a .permeability barrier locateddeeper in the cell, in the inner portion of the same

membrane (43). Transport across this barrierpresumedly requires utilization of a carrier mech-anism (15, 44, 45), since carbohydrates have a

393

0wma:

0cn

0wN

0:

a]

o 21

'I

-r

02

E

0-

o110-

60-

50-

40

30-

20-

10-

GARYM. GRAYAND FRANZ J. INGELFINGER

V Xlo' 100 X

50-

0 20 40 60 80 100

IX goFIG. 5. LINEWEAVER-BURKPLOT FOR SUCROSEHYDROLY-

sis. Values are calculated from Table IV and expressedas mean ± 2 SE. Units for S (substrate concentration)and V (rate of hydrolysis) are millimoles per liter andmillimoles per hour, respectively. The correlation withlinearity is significant (analysis of variance, p < 0.001).

hydrated molecular diameter larger than the cal-culated pore size of the permeability barrier andhence cannot cross it by simple diffusion (46, 47).Our findings of large amounts of monosaccharideproducts within the lumen are consistent withCrane's hypothesis, since the disaccharide couldbe split in the outer brush border and some of thereleased monosaccharides could then move backinto the lumen by diffusion without crossing a per-meability barrier. However, the exact location ofthis barrier in the intestinal cell is uncertain, andit may well be located at the outermost portion ofthe cell, i.e., at the limiting plasma-membrane ofthe microvillus. If this were the case, intracellu-lar hydrolysis would result in an inefficient disac-

TABLE V

Kinetic data from sucrose saturation studies*

Sucrose Glucose prod-hydrolysis uct absorption

No. of Intestinalstudies site Vmax Km Vmax Km

Mmoles/ mmoles/ mmoles/ mmoles/hr/ L hr/ L

15-cm 15-cmsegment segment

8 Jejunum 83 142 83 1546 Ileum 36 74 25 50

* Derived from data in Table IV. Vmax = maximal velocity; KmMichaelis constant.

TABLE VI

Comparison of amounts of monosaccharide released fromsucrase activity of intestinal contents sn vitro*

Glucose/Sample Glucose Fructose Fructose

no. released released ratio

mmoles/hr/L in-testinal contents

A 3 3 1.0B 5 6 0.84C 5 6 0.84D 9 9 1.0E 8 9 0.89F 7 8 0.88G 6 7 0.86H 19 19 1.0I 9 8 1.1

* Incubation at 370 C for 1 hour in sucrose substrate.

charide absorption process because both entry ofsucrose into the cell and subsequent exit of re-leased glucose and fructose from the cell back intothe lumen would require carrier-mediated trans-port across the permeability barrier. Thus, ourresults, although compatible with the intracellularhydrolysis hypothesis, are perhaps more consistentwith the hypothesis that sucrose is hydrolyzed bybrush border enzyme while the disaccharide isstill within the lumen, especially if the intestinalpermeability barrier were shown to be located atthe limiting plasma-membrane of the cell.

The finding of equal absorption rates for su-crose and the monosaccharide mixture at two dif-

G35mlM

39mlIPA GLUOSE4mM

SUCROSE66mM FRUCTOSE18mM

SUCROSE27mM

FIG. 6. SUCROSEABSORPTION IN HUMANINTESTINALPERFUSION EXPERIMENTS. A diagrammatic representationof what occurs over a 30-cm segment of intestine whensucrose (73 mM) is infused. Values refer to millimolesper hour. See text for elaboration.

394

INTESTINAL ABSORPTIONOF SUCROSEIN MAN

ferent infusion concentrations suggests that thehydrolytic step is not rate limiting in the over-allprocess of sucrose absorption. This is in accordwith the in vitro guinea pig experiments of Frid-handler and Quastel (8) as well as with the workof Dahlqvist and Thomson, who fed sucrose andan equivalent glucose-fructose mixture to rats(16). It is tempting to conclude that the indi-vidual transport rates for glucose and fructosedefine the sucrose absorption process. However,little is known about the importance of local physi-cal conditions in the absorption process, and it isconceivable that rate of contact between sugarmolecules and mucosal surface may occur moreslowly than either the hydrolytic or transportprocesses. Whether physical conditions are ratelimiting in the over-all process of absorption can-not be determined from our experiments.

In these comparative studies, a measure of thedependence of absorption upon monosaccharideconcentrations within the lumen is provided by de-termining the ratio of absorbed to intraluminalmonosaccharide (Table VII). It is notable thatsucrose has a marked advantage over the mono-saccharide mixture since this ratio is three to sixtimes greater for glucose and two times greaterfor fructose than that found for the monosac-charide mixture, indicating that absorption ratesof the released glucose and fructose products do notdepend on their concentrations within the luminalfluid. Instead, it is probable that a local highconcentration, comparable to the luminal concen-tration of an equivalent monosaccharide mixture,is provided from splitting of sucrose at the brushborder membrane and maintained for the trans-port mechanism at the outermost portion of theintestinal cell. This finding is analogous to theintracellular-medium ratios found by Miller andCrane in gut sac preparations (9). Unfortu-

TABLE VII

Ratio of absorbed to intraluminal monosaccharides*

Infusion:Infusion: glucose

Infusion sucrose +fructoseconcen-tration Glucose Fructose Glucose Fructose

29 mM 19 1.2 5.5 0.7373 mM 8.8 1.2 1.4 0.61

* Derived from mean values of paired studies (Tables I and II).

nately, it does not distinguish intracellular hy-drolysis from hydrolysis at the cell surface. If thedisaccharide were split at the cell surface, theremight be a high concentration of monosaccharide inthe fluid immediately adjacent to the mucosal sur-face with a gradient of decreasing concentrationstoward the center of the lumen. Analysis of thecollected luminal fluid reveals only the mean in-traluminal concentration.

The fact that the presence of galactose in theinfusing solution appeared to cause an inhibitionof sucrose hydrolysis (Table IV) must be inter-preted with caution. Galactose competes with glu-cose for the active transport process (24-27), andhence any glucose displaced from this process bygalactose might accumulate and inhibit sucrase,thereby retarding sucrose hydrolysis. It seemsunlikely that the trace impurities of glucose oroligosaccharide in the galactose would have sucha marked effect on sucrose hydrolysis. Appar-ently, then, sucrose hydrolysis in man may be in-hibited in the presence of galactose by 1) galac-tose itself, 2) glucose released from sucrose anddisplaced from the active transport process by thegalactose (product inhibition), or 3) feedbackinhibition of sucrase by a saturated active trans-port mechanism that must transport the addedgalactose as well as glucose. The last hypothesiswould require an intimate interrelation betweendisaccharide hydrolysis and the active transportprocess for glucose. Such a relation has indeedbeen suggested by Crane (43), and more recentlyby Semenza, Tosi, Vallotton-Delachaux, and Mill-haupt (48), who found that Nat is not only im-portant for active glucose transport (49), but alsoactivates sucrase.

When sucrose in 10 to 160 mMconcentrationswas infused in 15-cm segments, every study ap-peared to show hydrolysis and absorption ratesthat approached a maximum (Table IV, Figure4), but the relation of these rates to mean intra-luminal sucrose concentration varied from subjectto subject. Sucrose hydrolysis and glucose prod-uct absorption did appear to conform reasonablywell to Michaelis-Menten kinetics when meanvalues of the studies were plotted. Figure 5 dem-onstrates this for sucrose hydrolysis. A nearlyidentical relation prevailed for glucose productabsorption, suggesting a close interrelation betweenhydrolysis of sucrose and absorption of the re-

395

GARYM. GRAYANDFRANZJ. INGELFINGER

leased glucose. When jejunal and ileal studieswere evaluated separately, values for the maxi-mal velocity and the Michaelis constant werehigher for jejunum than ileum (Table V). Thisprobably indicates that jejunum has a greater ca-pacity for absorption of sucrose at high concen-trations of sugar, whereas the ileum performs quiteefficiently at the low concentrations that still re-main after the major portion has been absorbedfrom jejunum.

The restricted surface area used in intestinal per-fusion experiments may result in saturation phe-nomena irrespective of the presence or absence ofspecific binding sites for hydrolysis or transport,and demonstration of a maximal rate is not, ofitself, evidence for enzymatic hydrolysis or car-rier-mediated transport. However, the apparentMichaelis constant for sucrose hydrolysis in thesestudies in man is remarkably close to that re-cently found by Dahlqvist and Thomson for in-tact rabbit intestine in vitro (50). It is importantto note that the kinetic relationships shown forthe hydrolytic process should not be assumed toequal those for sucrase itself, since the geometricrelations important for contact between sugar andenzyme are undoubtedly quite different for intactintestinal tissue as compared to a homogenate orpurified tissue extract. Thus it has been shownthat Michaelis constants for intact intestine are afull order of magnitude higher than those forhomogenates (50).

Summary

1) Sucrose absorption was studied in man andexpressed in terms of sucrose hydrolysis and ab-sorption of its component monosaccharides, glu-cose and fructose.

2) Sucrose hydrolysis rates exceeded the mono-saccharide product absorption rates, and the glu-cose component was absorbed considerably morerapidly than the fructose component. Moreover,paired experiments demonstrated the same ab-sorption rates of these monosaccharides from su-crose as from an equivalent glucose-fructose mix-ture. Hence, hydrolysis does not appear to berate limiting in the process of sucrose absorption,and the individual absorption rates for glucose andfructose may define the rate of the disaccharideabsorption process.

3) Comparison of absorption from solutions hav-ing different concentrations of sucrose demon-strated the kinetic relationships to be variablefrom subject to subject, but Lineweaver-Burkplots revealed saturation kinetics that were nearlyidentical for sucrose hydrolysis and glucose prod-uct absorption, suggesting an interdependence ofthese two processes.

4) The addition of galactose to the sucrose solu-tion infused caused an appreciable decrease in su-crose hydrolysis and glucose product absorption,but had no effect on fructose product absorption.Since galactose is known to compete with glucosefor the active transport process, this suggests thatintestinal sucrase is inhibited either by the glu-cose product or by a feedback inhibition by a satu-rated active transport mechanism.

Acknowledgments

The authors are grateful to Dr. Robert K. Crane forhis interest and suggestions and to Mrs. Priscilla Taggartfor technical assistance.

References1. Miller, D., and R. K. Crane. The digestive function

of the epithelium of the small intestine. II. Local-ization of disaccharide hydrolysis in the isolatedbrush border portion of intestinal epithelial cells.Biochim. biophys. Acta (Amst.) 1961, 52, 293.

2. Dahlqvist, A. Specificity of the human intestinaldisaccharidases and implications for hereditarydisaccharide intolerance. J. clin. Invest. 1962, 41,463.

3. Auricchio, S., A. Rubino, R. Tosi, G. Semenza, M.Landolt, H. Kistler, and A. Prader. Disacchari-dase activities in human intestinal mucosa. En-zymol. biol. clin. (Basel) 1963, 3, 193.

4. White, A., P. Handler, E. L. Smith, and D. W.Stetten. Principles of Biochemistry. New York,Blakiston, 1959, p. 381.

5. Thomas, J. E. Secretion and absorption in the in-testine in The Physiologic Basis of Medical Prac-tice, 7th ed., C. H. Best and N. B. Taylor, Eds.Baltimore, Williams & Wilkins, 1961, p. 669.

6. Cantarow, A., and B. Schepartz. Biochemistry, 3rded. Philadelphia, W. B. Saunders, 1962, p. 270.

7. Hoffman, W. S. The biochemistry of clinical medi-cine, 3rd ed. Chicago, Year Book, 1964, p. 65.

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

9. Miller, D., and R. K. Crane. The digestive functionof the epithelium of the 'small intestine. I. An

396

INTESTINAL ABSORPTIONOF SUCROSEIN MAN

intracellular locus of disaccharide and sugar phos-phate ester hydrolysis. Biochim. biophys. Acta(Amst.) 1961, 52, 281.

10. Newey, H., P. A. Sanford, and D. H. Smyth. Loca-tion of function in the intestinal epithelial cell inrelation to carbohydrate absorption. J. Physiol.(Lond.) 1963, 168, 423.

11. Cajori, F. A. The enzyme activity of dogs' intestinaljuice and its relation to intestinal digestion. Amer.J. Physiol. 1933, 104, 659.

12. Dahlqvist, A., and B. Borgstrom. Digestion and ab-sorption of disaccharides in man. Biochem. 3.1961, 81, 411.

13. Gray, G. M., and F. J. Ingelfinger. Intestinal ab-sorption of sucrose in man: the site of hydrolysisand absorption. J. clin. Invest. 1965, 44, 390.

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

15. Crane, R. K. Intestinal absorption of sugars. Phys-iol. Rev. 1960, 40, 789.

16. Dahlqvist, A., and D. L. Thomson. The digestionand absorption of sucrose by the intact rat. J.Physiol. (Lond.) 1963, 167, 193.

17. Fisher, R. B., and D. S. Parsons. Glucose move-ments across the wall of the rat small intestine.J. Physiol. (Lond.) 1953, 119, 210.

18. Fordtran, J. S., K. H. Soergel, and F. J. Ingelfinger.Intestinal absorption of D-xylose in man. NewEngl. J. Med. 1962, 267, 274.

19. Hyden, S. A turbidmetric method for the determina-tion of higher polyethylene glycols in biologic ma-terials. Ann. roy. agr. Coll. Sweden 1955, 22,139.

20. Somogyi, M. Determination of blood sugar. J.biol. Chem. 1945, 160, 69.

21. Dische, Z., and A. Devi. A new colorimetric methodfor the determination of ketohexoses in presence ofaldoses, ketoheptoses and ketopentoses. Biochim.biophys. Acta (Amst.) 1960, 39, 140.

22. Avigad, G., D. Amaral, C. Asensio, and B. L. Hor-ecker. The D-galactose oxidase of Polyporus cir-cinatus. J. biol. Chem. 1962, 237, 2736.

23. Dahlqvist, A. Determination of maltase and iso-maltase activities with a glucose-oxidase reagent.Biochem. J. 1961, 80, 547.

24. Cori, C. F. The rate of absorption of a mixture ofglucose and galactose. Proc. Soc. exp. Biol.(N. Y.) 1926, 23, 290.

25. Fisher, R. B., and D. S. Parsons. Galactose ab-sorption from the surviving small intestine of therat. J. Physiol. (Lond.) 1953, 119, 224.

26. Jorgensen, C. R., B. R. Landau, and T. H. Wilson.A common pathway for sugar transport in ham-ster intestine. Amer. J. Physiol. 1961, 200, 111.

27. Annegers, J. H. Intestinal absorption of hexose inthe dog. Amer. J. Physiol. 1964, 206, 1095.

28. Crane, R. K. Studies on the mechanism of the in-testinal absorption of sugars. III. Mutual in-

hibition, in vitro, between some actively trans-ported sugars. Biochim. biophys. Acta (Amst.)1960, 45, 477.

29. Schedl, H. P., and J. A. Clifton. Kinetics of intes-tinal absorption in man: normal subjects and pa-tients with sprue (abstract). J. clin. Invest. 1961,40, 1079.

30. Schedl, H. P., and J. A. Clifton. Kinetics of in-testinal absorption in man: normals and patientswith sprue. Proceedings of the Second WorldCongress of Gastroenterology, Munich, 1962, II.1963, 728.

31. Holdsworth, C. D., and A. M. Dawson. The ab-sorption of monosaccharides in man. Clin. Sci.1964, 27, 371.

32. Lineweaver, H., and D. Burk. The determinationof enzyme dissociation constants. J. Amer. chem.Soc. 1934, 56, 658.

33. Bacon, J. S. D., and J. Edelman. The action of in-vertase preparations. Arch. Biochem. 1950, 28,467.

34. Blanchard, P. H., and N. Albon. The inversion ofsucrose; a complication. Arch. Biochem. 1950,29, 220.

35. Edelman, J. The formation of oligosaccharides byenzymatic transglycosylation. Advanc. Enzymol.1956, 17, 189.

36. Dahlqvist, A. "Substrate inhibition" of intestinalglycosidases. Acta chem. scand. 1960, 14, 1797.

37. Groen, J. The absorption of hexoses from the upperpart of the small intestine in man. J. clin. In-vest. 1937, 16, 245.

38. Miller, D., and R. K. Crane. The digestion of car-bohydrates in the small intestine. Amer. J. clin.Nutr. 1963, 12, 220.

39. Prader, A., G. Semenza, and S. Auricchio. In-testinale Absorption und Malabsorption der Di-saccharide. Schweiz. med. Wschr. 1963, 93, 1272.

40. Isselbacher, K. J., and J. R. Senior. The intestinalabsorption of carbohydrate and fat. Gastroenter-ology 1964, 46, 287.

41. Sonntag, W. M., M. L. Brill, W. G. Troyer, Jr., J.D. Welsh, G. Semenza, and A. Prader. Sucrose-isomaltose malabsorption in an adult woman. Gas-troenterology 1964, 47, 18.

42. Eichholz, A., and R. K. Crane. Personal communi-cation.

43. Crane, R. K. Hypothesis for mechanism of intes-tinal active transport of sugars. Fed. Proc. 1962,21, 891.

44. Cirillo, V. P. Sugar transport in microorganisms.Ann. Rev. Microbiol. 1961, 15, 197.

45. Csaky, T. Z. Intestinal absorption: changing con-cepts and ideas. Gastroenterology 1964, 47, 201.

46. Solomon, A. K. Measurement of the equivalent poreradius in cell membranes in Membrane Transportand Metabolism, A. Kleinzeller and A. Kotyk,Eds. London and New York, Academic Press,1961, p. 94.

397

GARYM. GRAYAND FRANZ J. INGELFINGER

47. Lindemann, B., and A. K. Solomon. Permeabilityof luminal surface of intestinal mucosal cells. J.gen. Physiol. 1962, 45, 801.

48. Semenza, G., R. Tosi, M. C. Vallotton-Delachaux,and E. Mfilhaupt. Sodium activation of humanintestinal sucrase and its possible significance inthe enzymatic organization of brush borders. Bio-chim. biophys. Acta (Amst.) 1964, 89, 109.

49. Riklis, E., and J. H. Quastel. Effects of cations on

sugar absorption by isolated surviving guinea pigintestine. Canad. J. Biochem. 1958, 36, 347.

50. Dahlqvist, A., and D. L Thomson. The hydrolysisof sucrose by intact and homogenized cells of ratsmall intestine. Influence of pH and substrate con-centration. Biochim. biophys. Acta (Amst.) 1964,92, 99.

398