Clinical Science and Molecular Medicine (1978) i4,495-501
Effect of hyperosmolar stimuli and coeliac disease on the
permeability of the human gastrointestinal tract
P. G. WHEELER, I. S. MENZIES AND B. CREAMER
The Gastrointestinal Laboratory and Department of Chemical Pathology, St Thomas's Hospital, London
(Received 28 June 1977; accepted 8 November 1977)
Summary
1. Oral loads have been used to assess the
permeability of the human gastrointestinal tract,
with lactulose (mol. wt. 342), rafnnose (mol. wt.
504), stachyose (mol. wt. 666) and a fluoresceinlabelled dextran (mol. wt. 3000) as marker
substances. Timed urinary recovery of these
substances, which are not metabolized, was
measured by quantitative paper chromatography
and directfluorimetry,and the results were used as
an indication of passive intestinal permeability.
2. Results in healthy adults showed that per­
meability to these markers was dependent on
molecular size, even after correction for aqueous
diffusion differences, such that a profile of restric­
ted permeability could be described for this range
of markers. Interpretation in terms of conventional
pore theory suggested the presence of more than
one population of pores.
3. Ingestion of solutions made hyperosmotic by
inclusion of glycerol resulted in a large increase in
permeability, in a pattern that suggested an
increase in either the size or frequency of a range of
smaller pores.
4. A similar increase in permeability and
alteration in the profile of restriction was found in
patients with coeliac disease.
5. The possible location of such pores in the
gastrointestinal mucosa is discussed in relation to
the cell membrane, the intercellular junction, and
the sites of cell exfoliation.
Correspondence: Dr P. G. Wheeler, Liver Unit, King's
College Hospital Medical School, London.
35
495
Key words: coeliac disease, diffusion, gastro­
intestinal tract, hyperosmolarity, permeability.
Abbreviation: FITC, fluorescein-isothiocyanate.
Introduction
The mucosa of the gastrointestinal tract is known
to be variably permeable to a wide range of
molecular species. Substances are able to cross this
barrier by simple diffusion, independently of active
or passive mechanisms. However, only small
amounts of substances with a molecular weight of
342 (disaccharide size) and above pass across by
diffusion, too small to be of much nutritional
significance, although possibly sufficient to be
active biologically.
Menzies (1974) has demonstrated that small
amounts of metabolically inert oligosaccharides
cross the human intestine after oral ingestion,
apparently by simple diffusion. These are fully
excreted in the urine, and this excretion is greatly
increased when the ingested solution is made
hyperosmolar, by adding a second solute (Laker &
Menzies, 1977). Although raised luminal
osmolarity as a cause of increased permeability is
not otherwise well described, it is clearly recognized
for many other epithelia (Ussing, Erlij & Lassen,
1974). Small-intestinal transfer of large-proteinsized molecules such as insulin (Danforth &
Moore, 1959) and horseradish peroxidase (Walker,
Cornell, Davenport & Isselbacher, 1972) have been
studied in animals, although the latter substance
may be largely absorbed by an active pinocytotic
P. G. Wheeler, I. S. Menzies and B. Creamer
496
mechanism as well as passively. Loehry, Axon,
Hilton, Hider & Creamer (1970) have demon­
strated intestinal permeability to polyvinylpyrrolidone, inulin, cyanocobalamin and creatinine
in animals, by measuring plasma clearance into
intestinal perfusate at high plasma concentrations.
Estimations of intestinal mucosal pore size by
Lindemann & Solomon (1962) in isolated rat
intestine, and also by Fordtran, Rector, Ewton,
Soter & Kinney (1965) in man, using hydrophilic
markers of small molecular size, indicate a pore
radius of 0-4 nm in the rat, and a radius of from
0-7 to 0-85 nm in the human jejunum.
We have studied human intestinal permeability
to a wide range of medium-sized non-electrolyte
markers and examined the effect of hyperosmolar
solutions and coeliac disease on this permeability.
We interpret our results in terms of diffusion
through postulated water-filled pores which may
vary in size and therefore in degree of restriction to
the passage of such molecules.
Methods
Permeability markers
We used as permeability markers: lactulose, a
disaccharide (mol. wt. 342, radius 0-50 nm;
Duphar Laboratories, Basingstoke, Hants., U.K.);
raffinose, a trisaccharide (mol. wt. 504, radius
0·59 nm; Sigma Chemical Co. Ltd, Kingston,
Surrey, U.K.); stachyose, a tetrasaccharide (mol.
wt. 666, radius 0-62 nm; Koch-Light Laboratories,
Colnbrook, Bucks., U.K.); a commercially
available fluorescein-labelled (FITC) dextran of
very narrow molecular-weight range (mean mol.
wt. 3000, radius 1-25 nm: Pharmacia Fine
Chemicals AB, Uppsala, Sweden). Molecular
radii were obtained from the Stokes-Einstein
equation, corrected for the smaller oligosaccharide
molecules according to the formula proposed by
Schultz & Solomon (1961). These three oligosaccharides are insignificantly metabolized in vivo,
and this was also established for the FITC-dextran
in man (Fig. 1). After intravenous injection of a
bolus of 500 mg of each of the oligosaccharides
(1-5 mmol of lactulose, 1 mmol of raffinose and
0-75 mmol of stachyose) intofivehealthy subjects
and of 2 mg of FITC-dextran (0-66 μπιοΐ) into
three subjects, all with normal renal function,
excretion in the urine was almost complete within
12 h, was of equivalent degree for each marker
during each consecutive 2\ h period, and tended to
follow an exponential pattern (Fig. 1). These oligo­
saccharides are not hydrolysed by human intestinal
hydrolases (EC group 3.2.1) (Dahlquist &
Gryboski, 1965; Udupihille, 1974), and there is no
evidence of mediated absorption (Menzies, 1974).
Although some dextranase (EC 3.2.1.11) activity
may exist in homogenates of jejunal mucosa,
1-7
0
LL
R St
F
25
LL
R
St
F
5
LL
R St
F
7-5
LL
R St
F
10
LL
R St
F
12 5
Time (h)
FIG. 1. Incremental percentage urinary recovery of permeability markers after intravenous administration in man. LL,
lactulose: R, raffinose: St, stachyose; F, FITC-dextran. Mean values + 1 SD are shown. Cumulative percentage
recoveries are shown as the increasing values at the top of the figure.
Human intestinal permeability
activity in vivo has not been established for intact
human intestine, and hydrolysis of a large orally
administered load is unlikely to be sufficient to
affect the very small percentage which permeates
across the intestinal wall. Since these molecules,
including dextran (Ogston & Woods, 1953), are
uncharged and have a spherical shape in aqueous
solution, they are satisfactory markers for assess­
ing the permeability of biological membranes. They
are also hydrophilic with negligible lipid solubility.
Permeability experiments
Standard oral loading solutions containing 14-6
mmol of lactulose, 9-9 mmol of raffinose, 7-5
mmol of stachyose (5 g of each) and 0-33 mmol (1
g) of the FITC-dextran, dissolved in 100 ml of
water, were given as a bolus to 12 fasting healthy
volunteer subjects after the collection of a 2 h
'baseline' urine sample. Nothing was allowed by
mouth for 2\ h after the load, after which fluids
were encouraged and food was permitted (barring
soya and baked beans, which contain some
raffinose and stachyose). Urine was collected for 5
h from the time of oral loading, into a bottle
containing 0-5 ml of 10% merthiolate as a sugar
preservative (Menzies, 1973). The urine volume
was measured, and a portion stored at 4°C for
subsequent analysis, the oligosaccharide and
FITC-dextran content being a measure of that
amount of each substance which had crossed the
intestinal wall.
A further six subjects were given identical loads
under the same conditions, except that the ingestion
solutions contained added glycerol to produce an
osmolarity of 2000 mosmol/1, and did not include
stachyose. Two other subjects were given hyperosmolar oral loads containing all four markers.
Five patients with biopsy-proven coeliac disease
were given identical loads as for the standard oral
loading experiments, excluding stachyose, and all
urine collections were made in the same way.
Permission for these studies was obtained from the
Ethics Committee of this Hospital and all patients
gave informed consent. Three of these patients
were known to have subtotal villous atrophy at the
time of study, one being newly diagnosed and
untreated. The remaining two had clinical signs of
active disease, having relapsed after a previous
period of gluten withdrawal.
Analytical methods
The three oligosaccharides, lactulose, raffinose
and stachyose, were measured in urine by sensitive
497
paper chromatography (Menzies, 1973; Menzies &
Seakins, 1976). In this technique the absorbance,
and hence oligosaccharide content, of 'test' and
'standard' zones developed on the same chromatogram were compared. Integrated peak areas were
estimated by means of a Joyce-Loebl scanning and
integrating densitometer (Joyce-Loebl Co. Ltd,
Princes Way, Team Valley, Gateshead, U.K.).
Sources of error were carefully controlled by
several modifications of traditional paperchromatographic technique, the coefficient of
variation for urinary sugar estimation being be­
tween 1 · 6 and 3% with both sample and standard
analysed in duplicate. Standard solutions were
prepared from the original oral loading solutions
and prepared for chromatography in the same way
as the 'test' samples. A standard curve was
obtained on the same chromatogram as 'test'
values.
Urinary FITC-dextran concentration was
measured by spectrophotofluorimetry, with adequ­
ate correction for background urinary fluorescence
and quenching. Urine samples, collected just before
oral loading, were matched for concentration with
test urine and three standard solutions prepared by
addition of small, accurate volumes of a suitable
dilution of the oral loading solution. Both standard
and test urines were diluted 1:10 with a
glycine/NaOH (0-1 mmol/1) buffer, pH 9-2 op­
timum for fluorescence, this being measured at an
excitation wavelength of 495 nm and emission
wavelength of 520 nm. Ether extraction of test and
standard urine samples, which removed all free
fluorescein, gave similar values, indicating that free
fluorescein had not been absorbed and excreted in
urine. It is likely that any absorbed free fluorescein
binds rapidly to plasma proteins.
Values are expressed as a percentage (R) of the
amount of each substance given in the oral load,
and represent between 5 and 10 mg, approxi­
mately, for the oligosaccharides, and 0-5 mg for
the FITC-dextran, in the standard experiments,
amounts easily measured by the techniques em­
ployed.
Passive diffusion of hydrophilic permeability
markers across a biological membrane will follow
an aqueous route and will therefore obey the laws
of simple diffusion (Davson & Danielli, 1952;
Davson, 1970). For unrestricted diffusion R oc D
for each marker, where D is the diffusion
coefficient. Diffusion in free solution is inversely
proportional to the square root of the molecular
weight (M) of the substance concerned, so that D
oc M'05, and thus R = K.Af-0'. For a number of
P. G. Wheeler, I. S. Menzies and B. Creamer
498
also in those patients with coeliae disease. This is
significant for all markers in each case (P < 0-01
Wilcoxon), and is of the order of 400-600% for the
oligosaccharides and 100-300% for the dextran.
When corrected for diffusion (R.M"'S), values
for recovery continue to show a significant reg­
ression on molecular size for the standard oral
loads (P < 0-01), the hyperosmolar oral loads (P <
0-05) and for the patients with coeliae disease (P <
0-01), indicating a true physical restriction to the
passive absorption of these permeability markers in
each group.
Calculated mean diffusion ratios (R.M0'5 ratios)
are shown in Table 1. Each marker is related to the
smallest, i.e. lactulose, and alteration in the ratio
indicates a change in permeability profile induced
by hyperosmolarity or coeliae disease.
permeability markers diffusing simultaneously
without restriction K is constant, so that values of
R.M0i will be equal and the ratios between these
values (hereafter called diffusion ratios) will be
unity. If there is physical restriction to the passage
of permeability markers then K is no longer
constant, R.M0'5 values decrease with increasing
molecular weight to give a true permeability profile,
and diffusion ratios are less than unity (Renkin,
1954). The significance of differences between
ratios and between values of R for different groups
was assessed by Wilcoxon's sum of rank test.
Results
Analysis of variance on the 5 h percentage
recoveries (R) in urine of each permeability marker
after standard oral loads in 12 normal subjects,
hyperosmolar loads in eight normal subjects, and
standard loads in five patients with coeliae disease
(Fig. 2), shows a significant regression of recovery
on molecular size for all three groups: standard
loads (P < 0-001), hyperosmolar loads (P < 0-01)
and the coeliae patients (P < 0-001). The results
also indicate a large increase in marker recovery,
and hence in passive permeability of the gastro­
intestinal tract, after a hyperosmolar stimulus, and
Discussion
Oral loads of permeability markers of this size are
probably chiefly absorbed in the small intestine
rather than in the less-permeable stomach (Daven­
port, Warner & Code, 1964). Low-molecularweight polyethylene glycol (PEG 400), a marker
similar in size to the oligosaccharides used here,
was shown by Chadwick, Phillips & Hofmann
•
20
•
•
10
•
•
•
•
•
•
•
•
•
•
•
•
-
•
•
•
•
•
•
t
t
•
1
•
•
»
•
0-1
•
:
t
f
•
%
•
•
•
•
005
Stand ard
002
-
•
•
•
•
•
•
•
%
•
•
•
•
•
•
•
Hyperosmolar
LL
R
St
Coeliae
F-D
Molecular size
Fio. 2. Values for percentage urinary recovery of permeability markers over 5 h in normal subjects (standard),
hyperosmolar experiments and coeliae disease, after oral ingestion. LL, Lactulose; R, raffinose; St, stachyose;
F-D, FITC-dextran.
Human intestinal permeability
499
TABLE 1. Mean diffusion ratios after standard and hyperosmolar oral loads, and in coeliac disease
For definition of diffusion ratios see the Methods section. Mean values + 1 SD are shown. ** P < 0-01
compered with standard values (Wilcoxon's sum of rank test). Numbers of paired observations are
shown in parentheses.
Diffusion ratio
Raffinose/lactulose
Standard
Hyperosmolar
Coeliac
0-76 ±0-06 (12)
0-91 ±0-07*· (8)
0-89 ±0-04« (5)
(1977) to have little difference in percentage
absorption after oral loads or loads instilled directly
into the upper jejunum.
Our purpose in using oral loading solutions as
opposed to intestinal infusions was to develop a
non-invasive method for studying the permeability
of the gastrointestinal tract in both patients and
normal subjects. The marked increase in perme­
ability to large molecules after a hyperosmolar
stimulus may result from an effect on the small
intestine, although an effect on the stomach might
contribute
(Altamirano,
1969). However,
Udupihille (1974) showed that lactulose recovery
after ingestion of a hyperosmotic load tended to
correlate with duodenal rather than gastric
osmolality, suggesting that the effect arose distal to
the pylorus. Other results (J. Michael & I. S.
Menzies, unpublished work) have shown that direct
infusion of hyperosmotic loads into the duodenum
produces a similar increase in permeability to that
after an oral load. Furthermore, patients with
coeliac disease, whose pathology is confined to the
small intestine, characteristically show an exag­
gerated response to the ingestion of hyper­
osmolar lactulose solutions (Menzies, 1974), sug­
gesting that the effect is of small-intestinal rather
than gastric origin.
Hyperosmolar solutions might appear to
increase overall intestinal permeability by other
mechanisms. Luminally directed osmotic water
flow might disrupt and reduce the thickness of the
epithelial unstirred water layer, or alternatively
separate any obliterated intervillous spaces in the
jejunum, thus increasing the total available
mucosal surface area. Also the solute responsible
for the hyperosmolarity of the load solution will
permeate across the intestinal wall and may drag
with it a proportion of the smaller amounts of
constituent permeability markers by solute-solute
interaction (Franz, Galey & Van Bruggen, 1968),
thus increasing their transfer. However, the stron­
gest argument for a direct effect of these hyper­
Stachyose/lactulose
0-68 + 0-08 (6)
0-86 + 0-04 (2)
FITC-dextran/lactulose
0-57 ± 0-06 (10)
0-33 ± 0-07** (6)
0-49 ± 0-07 (5)
osmolar solutions on the epithelial barrier is our
demonstration that they produce a significantly
different permeability profile.
The configuration of the diffusion-corrected
permeability profile with standard oral loads
indicates that the passage of the larger marker
molecules across the gastrointestinal epithelium is
restricted, as compared with the smaller molecules.
This is presumably due to the limiting physical
dimensions of the pathway taken by such hydrophilic markers in relation to their molecular size,
and fits in with the classical view (Solomon, 1968)
that the transfer of lipid-insoluble non-electrolytes
through biological membranes is limited to actual,
or theoretical, water-filled pores in the membrane.
The very small recoveries of our markers indicate
that the total surface area of available pores is
small, as expected, whereas substances able to
partition into membrane lipid would show much
greater transfer.
The increased permeability to the oligosaccharides in the presence of a hyperosmolar
solution is associated with an increase in
the diffusion ratios raffinose/lactulose and
stachyose/lactulose, implying an increase in the
mean size of the pores through which such
molecules are passing. The reduction in the
diffusion ratio for the larger substance, FITCdextran, however, paradoxically suggests a
diminution in pore size despite a marked overall
increase in permeability to this substance. The
hyperosmolar solution may therefore increase the
size, or incidence, of one group of smaller pores
that favour the oligosaccharides, without affecting
a further group of larger pores of lower incidence
capable of admitting the dextran freely. This
observation, implying two or more populations of
pores of different mean size in the intestinal
mucosa, is supported by studies in the dog's
stomach (Altamirano & Martinoya, 1966) and
other epithelia (Arturson & Granath, 1972; Grotte,
1956). Any attempt to calculate a mean pore size
500
P. G. Wheeler, I. S. Menzies andB. Creamer
from our profile data would therefore be of limited
value.
Aqueous pores in the mucosa may represent
either defects within the cell membrane, or an inter­
cellular (tight-junction) pathway. These molecular
markers are too large to pass through the pores of
approximately 0·4 nm radius which exist in
erythrocyte membranes (Solomon, 1968; Red­
wood, Rail & Perl, 1974), or rat intestinal cells
(Lindemann & Solomon, 1962), and they may
therefore pass through tight-junction pores. Such
conclusions have been made for sucrose, of
equivalent size to lactulose, in epithelia of the rabbit
gallbladder and the toad urinary bladder (Wright &
Pietras, 1974; Bindslev & Wright, 1976). The
potential of this tight-junction route in the human
gastrointestinal tract is emphasized by our findings
with hyperosmolar oral loads, as such a hyper­
osmotic stimulus profoundly increases perme­
ability in many other epithelia (Ussing, 1966;
Franz et al., 1968), and may often cause tight
junctions to open (Erlij & Martinez-Palomo, 1972,
1973; Brightman, Hori, Rapoport, Reese & Westergaard, 1973).
The results after standard oral loads in active
coeliac disease show a definite increase in intestinal
permeability to the whole range of molecular
markers, and are consistent with the possibility that
greater exudation of some nutritionally important
substances occurs through such a diffusely
damaged intestinal wall (Creamer, 1971). Weser &
Sleisenger (1965) also attributed increased lactosuria and sucrosuria in coeliac disease to increased
permeability, as well as reduced disaccharidase
activity. Increased lactulose absorption in coeliac
disease has already been described by Menzies
(1974).
The alteration in permeability profile and dif­
fusion ratios in the patients with coeliac disease is
somewhat similar to that after a hyperosmolar
stimulus. The increased raffinose/lactulose ratio,
but unchanged or slightly decreased FITCdextran/lactulose ratio, again suggests a multiple
pore system, with an increase in size or incidence of
only those smaller pores which are freely available
to the oligosaccharides but less so to the dextran.
Such pores may again represent an intercellular
route, more permeable with altered epithelial
morphology, or possibly cell extrusion zones,
whose incidence might be greater with the
increased epithelial cell turnover which is charac­
teristic of coeliac disease (Pink, Croft & Creamer,
1970).
This increased permeability to large molecules is
interesting in view of the villous atrophy and
reduced absorptive area found in coeliac disease,
usually associated with decreased absorption of the
smaller D-xylose (mol. wt. 150). The two findings
may be reconciled by the proposal that D-xylose is
mainly transferred by small pores of high incidence
within the mucosal cell membrane, which are not
available to the larger oligosaccharides. This route,
and that for actively mediated D-xylose absorption
(Caspary, 1972), would become impaired by
reduction of intestinal surface area and function.
Fordtran, Rector, Locklear & Ewton (1967) found
a reduced passive absorption of [3H]water,
[14C]urea, L-xylose and [3H]erythritol in coeliac
disease, which suggested a reduction in both
permeability and pore size. These markers are also
much smaller than oligosaccharides, and the
findings may again reflect a reduced surface area.
An explanation for the reduced intestinal perme­
ability to polyethylene glycol 400 in coeliac disease
(Chadwick, Phillips & Hofmann, 1977) must lie in
the contrasting properties of this marker with the
oligosaccharides. Approximately 20% of poly­
ethylene glycol 400 is absorbed in the normal
subject after an oral load, as against 0-2% for
lactulose, over a similar period of time. This 100fold difference might be due in part to a difference
in molecular shape, but is probably due mainly to
greater lipid solubility, and although little is known
about this latter property for low-molecular-weight
polyethylene glycol, it is soluble in many organic
solvents. Also, the olive oil:water partition
coefficient of diethylene glycol is 5 x 10 -3 , 170
times greater than the 3 x 10 - 5 for sucrose, which
is a very similar molecule to lactulose (Davson &
Danielli, 1952). Therefore absorption of poly­
ethylene glycol 400 may also depend partly on
intestinal surface area, and thus be affected by
coeliac disease.
The significance of intestinal permeability must
relate to loss of metabolically important plasma
constituents or absorption of antigenic, carcino­
genic or other toxic substances. A hyperosmotic
stress, or coeliac disease, might considerably
enhance this possibility, and increased accessibility
to gliadin or its toxic fractions might, in the latter
case, provide a self-perpetuating pathogenic
stimulus.
Acknowledgments
We are grateful to Professor R. Brooks, Depart­
ment of Chemical Pathology, St Thomas' Hospital,
and to Professor M. Bradbury, Department of
Human intestinal permeability
Physiology, King's College, for advice. This work
was supported by the Endowment Fund of St
Thomas' Hospital.
References
ALTAMIRANO, M. (1969) Action of concentrated solutions of
non-electrolytes on the dog gastric mucosa. American
Journal of Physiology, 216,33-40.
ALTAMIRANO, M. & MARTINOYA, C. (1966) The permeability of
the gastric mucosa of dog. Journal of Physiology {London),
184,771-790.
ARTURSON, G. & GRANATH, K. (1972) Dextrans as test
molecules in studies of the functional ultrastructure of
biological membranes. Clinica Chimica Ada, 37,309-322.
BINDSLEV, N. & WRIGHT, E.M. (1976) Effect of temperature on
non-electrolyte permeation across the toad urinary bladder.
Journal of Membrane Biology, 29,265-289.
BRIGHTMAN, M., HORI, M., RAPOPORT, S.I., REESE, T.S. &
WESTERGAARD, E. (1973) Osmotic opening of tight-junctions
in cerebral endothelium. Journal of Comparative Neurology,
152,317-325.
CASPARY, W.F. (1972) On the mechanisms of D-xylose
absorption from the intestine. Gastroenterology, 63,531-532.
CHADWICK, V.S., PHILLIPS, S.F. & HOFMANN, A.F. (1977)
Measurements of intestinal permeability using low molecular
weight polyethylene glycols (PEG 400). I. Chemical analysis
and biological properties of PEG 400. II. Application to
normal and abnormal permeability states in man and animals.
Gastroenterology, 73,241-251.
CREAMER, B. (1971) Loss from the small intestine. Journal of
the Royal College of Physicians (London), 5,323-332.
DAHLQUIST, A. & GRYBOSKI, J.D. (1965) Inability of the
human small intestinal lactase to hydrolyze lactulose.
Biochimica Biophysica Ada, 110,635-636.
DANFORTH, E. & MOORE, R.D. (1959) Intestinal absorption of
insulin in the rat. Endocrinology, 65,118-123.
DAVENPORT, H.W., WARNER, H.A. & CODE, C.F. (1964)
Functional significance of gastric mucosal barrier to sodium.
Gastroenterology, 47,142-152.
DAVSON, H. (1970) Permeability and the structure of the plasma
membrane. In: A Textbook of General Physiology, 4th edn.,
vol. 1, chapter 8. J. and A. Churchill, London.
DAVSON, H. & DANIELLI, I.F. (1952) The Permeability of
Natural Membranes, 2nd edn. Cambridge University Press,
Cambridge.
ERLIJ, D. & MARTINEZ-PALOMO, A. (1972) Opening of tight
junctions in frog skin by hypertonic urea solutions. Journal of
Membrane Biology, 9,220-240.
ERLU, D. & MARTINEZ-PALOMO, A. (1973) Opening of tight
junctions in toad urinary bladder by hypertonic solutions.
Federation Proceedings, 32,218.
FORDTRAN, J.S., RECTOR, F.C., EWTON, M.F., SOTER, N. &
KINNEY, J. (1965) Permeability characteristics of the human
small intestine. Journal of Clinical Investigation, 44, 1935—
1944.
FORDTRAN, J.S., RECTOR, M.F., LOCKLEAR, T.W. & EWTON,
M.F. (1967) Water and solute movement in the small
intestine of patients with sprue. Journal of Clinical
Investigation, 46,287-298.
501
FRANZ, T.J., GALEY, W.R. & VAN BRUGGEN, I.T. (1968)
Further observations of asymmetrical solute movement
across membranes. Journal of General Physiology, 51,1-12.
GROTTE, G. (1956) Passage of dextran molecules across the
blood-lymph barrier. Ada Chirurgica Scandinavica, Suppl.
211.
LAKER, M.F. & MENZIES, I.S. (1977) Increase in human
intestinal permeability following ingestion of hypertonic
solutions. Journal of Physiology (London), 265,881-894.
LINDEMANN, B. & SOLOMON, A.K. (1962) Permeability of
luminal surface of intestinal mucosal cells. Journal of General
Physiology, 45,801-810.
LOEHRY, C A . , AXON, A.T.R., HILTON, P.J., HIDER, R.C. &
CREAMER, B. (1970) Permeability of the small intestine to
substances of different molecular weight. Gut, 11,466-470.
MENZIES, I.S. (1973) Quantitative estimation of sugars in blood
and urine by paper chromatography using direct densitometry. Journal of Chromatography, 81,109-127.
MENZIES, I.S. (1974) Absorption of intact oligosaccharide in
health and disease. Biochemical Society Transactions, 2,
1042-1047.
MENZIES, I.S. & SEAKINS, J.W.T. (1976) In: Chromatography
and Eledrophoretic Techniques, vol. 1, chapter 9. Ed.
Smith & Seakins. Heinemann, London.
OGSTON, A.G. & WOODS, E.F. (1953) Molecular considerations
of dextrans in solution. Nature (London), 171,221-222.
PINK, I.J., CROFT, D.N. & CREAMER, B. (1970) Cell loss from
the small intestinal mucosa: a morphological study. Gut, 11,
217-222.
REDWOOD, W.R., RALL, E. & PERL, W. (1974) Red cell
membrane permeability deduced from bulk diffusion
coefficients. Journal of General Physiology, 64,706-729.
RENKIN, E.M. (1954) Filtration, diffusion and molecular sieving
through porous cellulose membranes. Journal of General
Physiology, 38,225-243.
SCHULTZ, S.G. & SOLOMON, A.K. (1961) Determination of the
effective hydrodynamic radii of small molecules by viscometry. Journal of Genera! Physiology, 44,1189-1199.
SOLOMON, A.K. (1968) Characterisation of biological mem­
branes by equivalent pores. Journal of General Physiology,
51,336-364.
UDUPIHILLE, M. (1974) Factors affecting human intestinal
permeability to non-metabolised oligosaccharides. M.Phil.
Thesis, University of London.
USSING, H.H. (1966) Anomalous transport of electrolytes and
sucrose through the isolated frog skin induced by hypertonicity of the outside bathing solution. Annals of the New
York Academy of Sciences, 137,543-555.
USSING,
H.H., ERLIJ, D. & LASSEN, U. (1974) Transport
pathways in biological membranes. Annual Reviews in
Physiology, 36,17-48.
WALKER,
W.A.,
CORNELL,
R.,
DAVENPORT,
L.M.
&
ISSELBACHER, K.J. (1972) Macromolecular absorption:
mechanism of horseradish peroxidase uptake and transport in
adult and neonatal rat intestine. Journal of Cell Biology, 54,
195-205.
WESER, Ε. & SLEISENGER, M.H. (1965) Lactosuria and lactase
deficiency in adult coeliac disease. Gastroenterology, 48,
571-578.
WRIGHT, Ε.Μ. & PIETRAS, RJ. (1974) Routes of non-electro­
lyte permeation across epithelial membranes. Journal of
Membrane Biology, 17,293-312.