ClinicalScience (1970) 39, 769-779.
THE INFLUENCE OF pH, BICARBONATE AND
HYPERTONICITY ON THE ABSORPTION OF
AMMONIA FROM THE RAT INTESTINE
J. D. SWALES, J. D. TANGE AND O. M. WRONG
Royal Postgraduate Medical School, Hammersmith Hospital
(Received 11 February 1970)
SUMMARY
1. Ammonia absorption has been studied in closed loops of rat jejunum, ileum and
colon.
2. Ammonia absorption was significantly greater from solutions buffered at higher
pH. Absorption was also greater in the presence of the bicarbonate ion, and this
effect was not due solely to the influence of bicarbonate upon the initial pH of the
solutions studied. In the presence of ammonia net bicarbonate absorption was
increased and net bicarbonate secretion was decreased.
3. The above effects are compatible with the hypothesis that non-ionic diffusion
plays an important role in the absorption of ammonia.
4. Ammonia absorption was significantly less from hypertonic solutions than from
hypotonic or isotonic solutions. This effect persisted even after the hypertonic
solution had been removed, and therefore appears to have been due to a change in
ileal wall permeability rather than an effect of these solutions on water movement.
The absorption of hydrazine was similar to that of ammonia in these circumstances;
as there is evidence that hydrazine is absorbed by a process of non-ionic diffusion
in the lipid phase, this finding is a further indication that hypertonicity exerts its effect
through a change in membrane permeability.
Intestinal ammonia is formed mainly from bacterial breakdown of nitrogenous substances
(Folin & Denis, 1912; Wilson, lng, Metcalfe-Gibson & Wrong, 1968) of which urea is the
most important (Walser & Bodenlos, 1959). Absorption of intestinal ammonia is an intermediary process in the re-cycling of urea nitrogen into plasma albumin (Richards, MetcalfeGibson, Ward, Wrong & Houghton, 1967; Jones, Smallwood, Craigie & Rosenoer, 1969).
Intestinal ammonia may playa part in the production of hepatic coma (Sherlock, 1958) and
possibly contributes to the production of uraemic colitis (Bourke, Milne & Stokes, 1966).
Partition of ammonia between blood and gastric juice (FleshIer & Gabuzda, 1965) like
Correspondence: Dr J. D. Swales, Department of Medicine, Royal Infirmary, Manchester.
769
770
J. D. Swales, J. D. Tange and O. M. Wrong
diffusion of ammonia into the renal tubule (Orloff & Berliner, 1956) and distribution of
ammonia between blood and cerebrospinal fluid (Stabenau, Warren & RaIl, 1959), has been
attributed to pH-dependent non-ionic diffusion, but there is no general agreement on the
method of absorption of ammonia by the rest of the gastro-intestinal tract. Mossberg (1967)
claimed that the isolated hamster ileal sac transported ammonia by an active, non-pH dependent mechanism, but Price, Schwartz, Molavi, Britton & Vorhees (1967) obtained evidence
for pH-dependent absorption. Kettering & Summerskill (1967) failed to find any influence of
pH upon ammonia movement between jejunum and blood in man.
If ammonia is absorbed from the ileum by a process of non-ionic diffusion in the lipid
phase the absorptive process should be independent of osmotically determined movements of
water through the ileal wall (Fordtran & Dietschy, 1966). It is conceivable that hypertonicity
may modify the absorptive process by producing changes in the permeability of the gut wall.
The absorption of glucose from rat ileum and jejunum is impaired when glucose is infused in
hypertonic mannitol or urea solutions (Kameda, Abei, Nasrallah & Iber, 1968) and this
impairment persists in subsequent experiments with hypotonic solutions.
This paper is concerned with absorption of ammonia from the small and large intestine of
rats and with the effects of hydrogen and bicarbonate ions and hypertonicity on this absorption.
To elucidate further the effect of tonicity, hydrazine has been incorporated into the solutions
as another weak base. Hydrazine is not a normal physiological intermediary, but nevertheless
exhibits pH-dependent movement in the kidney (Coe & Korty, 1967).
MATERIALS AND METHODS
The effects ofpH and bicarbonate
Eight buffered and four unbuffered solutions were made up (Tables I and 2).
These buffered solutions were used in rotation in individual animals, and prepared by
titrating two aliquots of a stock solution of 0·2 M Tris (Tris-(Hydroxymethyl)-Methylamine,
24·2 gjl) with 0·2 N hydrochloric acid so that the final pH was 7·0 and 9'0 respectively, as
measured by a glass electrode pH meter. Osmolalities of the resultant solutions were measured
by a Knauer osmometer and adjusted to 160 mosmoljkg by diluting with distilled water. The
appropriate salts were then added (Table I) and the final osmolality and pH measured. If
necessary the osmolality was further adjusted to 300 mosmoljkg by adding a measured quantity of sodium chloride. Fresh solutions were prepared for each group of experiments.
Four unbuffered solutions containing ammonium chloride (150, 75, 37·5 and 18'75 mEqjl)
were used in two animals (Table 2) starting with the low concentration of ammonia in one and
the high concentration· in the other. Final osmolality was adjusted to 300 mosmoljkg by the
addition of sodium chloride.
Four consecutive experiments were performed on the one loop of intestine in each rat.
Forty-eight experiments with buffered solutions and eight experiments with unbuffered
solutions were performed for each region of intestine.
The effects of hypertonicity
Solutions of different tonicity were produced by incorporating mannitol and urea in a
solution of ammonium chloride. Hydrazine was added to some solutions. Three groups of
experiments were carried out.
Absorption of ammonia from the rat intestine
771
Group A (three rats). Five solutions of ammonium chloride (NH 3 37'5mEq/l) were prepared.
The osmolality of the most hypotonic solution was thus 75 mosmol/kg and the osmolalities
of the remaining solutions adjusted to 150, 225, 300 and 375 mosmol/kg by the addition of
calculated amounts of mannitol. The solutions were used consecutively on three animals so
that a total of fifteen experiments was performed.
Group B (six rats). Two solutions of ammonium chloride (NH 3 37·5 mEq/l) with hydrazine
4 mg/l were prepared and their final osmolalities adjusted to 225 and 375 mosmol/kg with
TABLE 1. Buffered solutions
Tris buffer
Salt added
(70 mEq cation/!) - - - - - pH 7·0 pH 9·0
(Resultant pH)
NH 4CI
NaCl
NH4HCO a
NaHCO a
7'18
7·19
7-39
7'42
8·43
9·06
8·42
8·80
TABLE 2. Unbuffered solutions
Salt added NH 4Cl
Resultant pH
(Concentration NH4 mEq/!)
150
75
37·5
18·75
5·13
5·19
5·20
5·26
calculated amounts of mannitol. The hypotonic solution was left in the ileum for 30 min,
followed by the hypertonic solution for the same period, and then by the hypotonic fluid,
which was used for three periods of 30 min.
Group C (six rats). Solutions contained similar amounts of ammonia and hydrazine to those
in Group B, but urea was substituted for mannitol to produce similar osmolalities. The same
sequence of experiments was performed as in Group B.
Procedures
Under methoxyflurane (Penthrane) anaesthesia, a 4 em mid-line abdominal incision was
made in Wistar rats weighing 200-300 g. The small bowel or caecum, as appropriate, was
delivered through the incision. For small bowel studies either the proximal 16 em of jejunum
or the distal 16 em of ileum was marked off. This length represents approximately the amount
supplied by a single fold of mesentery. The proximal end of this segment wasopened and a
plastic intravenous cannula (Braunula No.1) was inserted and secured by ligature. The distal
F
772
J. D. Swales, J. D. Tange and O. M. Wrong
end of the loop was then opened and isotonic saline run through the intestine from the proximal
to the distal end until the washings were clear. A second cannula was fixed in the distal end.
For studies on the large bowel, a cannula was inserted in the caecum and the entire large bowel
washed out with isotonic saline; a second cannula was then inserted at the junction of colon
and rectum.
In all experiments the isolated segment of bowel was next washed through with the solution
to be studied and then 3 ml of this solution were introduced into the loop by means of syringes
attached to the cannulae. After 30 min the solution was withdrawn into the syringes and the
volume noted. It was transferred to a specimen tube, under oil, and analysed for total carbon
dioxide (Peters & Van Slyke, 1932)and ammonia (Conway, 1962). Bicarbonate was calculated
from the total carbon dioxide concentration and pH by means of the Henderson-Hasselbalch
equation, assuming a pK of 6·10. Hydrazine was measured by the method of Dambrauskas
& Cornish (1962).
RESULTS
Effects ofpH and bicarbonate
In these experiments ammonia was readily absorbed by the rat intestine. Individual animals
did not differ greatly in their capacity to handle these solutions as demonstrated by the small
standard deviations (Table 3). Ammonia absorption from unbuffered solutions expressed as a
percentage of initial concentration showed no tendency to fall with time or with increasing
ammonia concentration in the fluid infused (Table 4).
At the end of each 30 min period the sodium chloride and sodium bicarbonate solutions
contained O'S-l'O mEq/l of ammonia. The final pH of the sodium chloride solutions ranged
from 6,48-7,02 (initial pH 7,19) and 7·00-7·46 (initial pH 9,06) and the final pH of the sodium
bicarbonate solutions from 7'17-7'S3 (initial pH 7,42) and 7,81-8,14 (initial pH 8,80).
The final pH of the ammonium chloride solutions ranged from 6,69-7,04 (initial pH 7,18)
and 7·4S-7·79 (initial pH 8,43). The corresponding ranges for ammonium bicarbonate solutions
were 6·84-7·24 and 7·76-7·81 (initial pH 7·39 and 8,42).
For each group of experiments (Table 3) significantly more ammonia was absorbed from the
solutions buffered at a higher pH (P<O·OS; Student's t-test); likewise there was significantly
greater absorption of ammonia from the solutions of ammonium bicarbonate than from the
corresponding ammonium chloride solutions (P<O·OS).
More bicarbonate accumulated in the saline solutions than in the corresponding solutions
of ammonium chloride, except in the jejunum where the difference between secretion of
bicarbonate into the sodium chloride solution buffered at pH 7·19 and the corresponding
ammonium chloride solution was not statistically significant (P>O·OS). At the higher pH
however, there was greater bicarbonate secretion into the sodium chloride solution (P<O·OS).
The absorption of bicarbonate was greater from each of the ammonium bicarbonate solutions
than from the corresponding sodium bicarbonate solutions (P<O·OS), and was also greater
from both solutions at the lower pH (P<O·OS).
The relationship between pH, ammonia absorption and bicarbonate movement was qualitatively similar in the jejunum, ileum, and colon, though there were quantitative differences.
It is noteworthy that there was a net secretion of bicarbonate from the ileum into sodium
bicarbonate solutions. Ammonia absorption was less from the colon than from the other two
Absorption of ammonia from the rat intestine
TABLE
3. Mean net ammonia absorption, bicarbonate secretion and absorption from rat intestine as calculated
from changes in concentration and volume of instilled solutions (SD in parenthesis)
NH4Cl
Initial pH:
Jejunum Ammonia absorption (%)
Net bicarbonate absorption (%)
Net bicarbonate secretion
(mEq/1)
Ileum
Ammonia absorption (%)
Net bicarbonate absorption (%)
Net bicarbonate secretion
(mEq/1)
Colon
773
Ammonia absorption (%)
Net bicarbonate absorption (%)
Net bicarbonate secretion
(mEq/l)
NH4HC0 3
7·18
8'43
7·39
8·42
62·7
(7'5)
74·4
(7'7)
77·1
83·1
(H)
(2-8)
65·7
62·3
(7'3)
(%)
4·6
(2'7)
11·8
63·2
(4'7)
76·7
(3'1)
(2-3)
81·7
(3-8)
42·0
(3-9)
7·6
(1'8)
42·7
(4'8)
6·6
(2'2)
20·2
(4'9)
57·5
(7-3)
NaCl
7·19
9·06
4·9
(0'9)
18·1
12·7
29·1
(2'7)
8'80
52·0
(5'9)
20·1
(11'5)
2'5
(1'3)
(4-2)
(2-8)
74·5
(6'3)
25·7
(8'8)
7-7
(6'5)
12·7
14·5
(1'9)
7·42
86·2
(3'7)
22·2
(5'4)
(2-4)
60·8
(10'7)
36·9
(7'7)
NaHC0 3
9·5
0·1
(3'5)
19·8
(3'2)
(2-8)
4. Ammonia absorption at different concentrations from the ileum
of two rats. Concentrations were used in ascending order in Rat 1 and
descending order in Rat 2
TABLE
Ammonia concentration (mEq/l)
Rat 1 (% absorption)
Rat 2 (% absorption)
18·75
37·5
75·0
150·0
54·3
46·8
57·1
47-3
60·6
51·2
54·8
47-4
sites (P<O·05). On the other hand, more bicarbonate was absorbed from the jejunum than
from the other two sites, whilst bicarbonate secretion was greater into the ileum than into the
jejunum or colon (P<O·05).
Effects of hypertonicity
The absorption of ammonia from hypotonic solutions was constant in all three rats studied
in Group A. Absorption from the final, hypertonic solution, however, was markedly impaired
(Fig. 1). Mean ammonia absorption from this solution was significantly less than from the
mean of the previous experiments (P<O·05).
In Group B ammonia absorption was significantly less when the second, hypertonic solution
J. D. Swales, J. D. Tange and
774
o. M.
Wrong
JOO
90
80
Q)
c
'"
C
,_ ~~..::._ ---- - __ -e---- --_
70
Q)
<,
--::::::~"
Q)
a.
....
.=-<.::::.::::..:-----:~--
~
60
.•
~
50
40
J·O
~
.s
0·5
Q)
u
c
c
"5
0-0
.0
"0
~ -0·5
-1·0
FIG. 1. Absorption
of ammonia and fluid by the ileum of three rats from solutions of progressively
increasing osmolality. Each plotted line represents one animal.
100
80
Q)
60
c
'"
C
Q)
~
Q)
a.
40
20
t
225
t
t
375
225
t
225
t
225
(rnosrnolz l)
o
2
Hcurs
2. Percentage absorption of ammonia and hydrazine (dotted line) from hypotonic and hypertonic mannitol solutions instilled into the rat ileum (Mean and 1 SD plotted).
FIG.
Absorption of ammonia from the rat intestine
775
was instilled (P<0·05). In the next two experiments with hypotonic solutions ammonia absorption was still significantly depressed (P<0·05) but subsequently returned towards initial values
(Fig. 2). Hydrazine absorption from the hypertonic and subsequent solutions was likewise
reduced (Fig. 2). The absorption of hydrazine and ammonia was significantly correlated in
these experiments (r = +0'845).
In Group C there was a fall in the mean absorption of ammonia from 67·5% to 51·8 % when
the hypertonic urea solution was instilled (Fig. 3). The difference was statistically significant
(P < 0,05). The absorption of ammonia in the third, fourth and fifth experiments was significantly less than in the first experiment (P<0·05).
100
80
.
60
rn
....e
E
c
0..
~/
"<~l- - ---1-------{------i
40
20
1- -I
t
225
t
t
375
225
t
225
t
225
(mosmot/t)
o
2
Hours
FIG. 3. Percentage absorption of ammonia and hydrazine (dotted line) from hypotonic and hypertonic urea solutions instilled into the rat ileum.
Mean hydrazine absorption fell from 64·5 %to 39·3 %when a hypertonic solution of urea was
used, and remained at a lower level for the remaining three experiments (Fig. 3). The difference
between the first and each of the subsequent experiments was statistically significant (P<0·05).
The correlation between ammonia and hydrazine absorption in the thirty experiments with
urea was also significant (r = +0'628).
DISCUSSION
The concentrations of ammonia used in these experiments were greater than those used by
Summerskill (1966) and were chosen to approximate to the concentrations which might be
expected to occur in uraemic patients ifmost of the urea in the gut was converted into ammonia;
for instance a concentration of urea in the intestinal fluid of 240 mg/IOO ml would yield 80
776
J. D. Swales, J. D. Tange and O. M. Wrong
mEqjl of ammonia, though simultaneous absorption of ammonia would reduce this figure to
the lower values for faecal ammonia found by Wilson et al. (1968). The faeces of uraemic
patients have been shown to contain minimal traces of urea, unless antibiotics are given, when
the concentration approached that of plasma urea (Wilson et aI., 1968). This finding suggested
that bacterial breakdown of urea in the intestine is normally complete; as many colonic microorganisms produce urease, it is probable that this is the most important route of urea breakdown.
Our results support the hypothesis that non-ionic diffusion plays an important role in
ammonia absorption from the intestine not only in our experimental model but also at the
concentrations likely to arise in uraemic patients. It is true that absorption will reduce the
concentrations which might otherwise be expected to develop in uraemic patients, but Table 4
shows that absorption of ammonia is, in fact, independent of initial concentrations within a
wide range.
10 all three parts of the gastro-intestinal tract examined, absorption was greater from solutions of higher pH, in which more ammonia is present as the lipid-soluble unionized form
(Milne, Scribner & Crawford, 1958). It is possible that pH-dependent ammonia absorption
might be due to pH-induced enzyme changes enhancing active ammonia transport or synthesis.
This explanation is unlikely since absorption of some basic drugs which are lipid soluble in the
unionized phase exhibits similar pH dependence (Schanker, Tocco, Brodie & Hogben, 1959).
Hydrazine, which is similar to ammonia in relevant physical properties, but dissimilar chemically, also shows pH-dependent absorption from the ileum (Swales, unpublished observations).
Ammonia absorption was greater from bicarbonate solutions. This finding reinforces the
observation of Mossberg (1967) that bicarbonate facilitates ammonia transport by the ill vitro
ileal sac. We have demonstrated that the reciprocal effect also obtains, i.e, the final concentration of bicarbonate in ammonia-containing fluids was less than that of non-ammonia containing fluids. This was true whether net bicarbonate absorption or net bicarbonate secretion
occurred. The comparatively small effect of ammonia upon the initial pH of the solutions used
(Table 3) was insufficient to account for its facilitatory effect upon bicarbonate absorption.
If the facilitatory influence of bicarbonate upon ammonia absorption was due to an effect upon
trans-membrane potential (Mossberg, 1967), it is difficult to explain this reciprocal effect. A
more likely explanation is afforded by the work of Rosenfeld, Aboulafia & Schwartz (1963)
upon the absorption of ammonia from the dog bladder in vivo. These workers demonstrated
greater disappearance of ammonia and bicarbonate from solutions of ammonium bicarbonate,
than from solutions where each one of these was present as the salt of a strongly ionized
electrolyte (i.e. ammonium chloride and sodium bicarbonate respectively). They suggested
that paired non-ionic diffusion was taking place with simultaneous, rapid diffusion ofNH 3 and
CO 2 and continuous regeneration of each non-ionic species at the expense of the ionic member
of each buffer pair.
NH~ +HCO; =<== NH 3+C02+H20
Our observations thus support the hypothesis that non-ionic diffusion plays an important
role in the absorption of ammonia from the intestine. Other evidence is contradictory. Summerskill (1966) found no evidence for pH-dependence in the absorption of ammonia from the
human jejunum. However, he did not estimate bicarbonate secretion into his solutions, and
in the jejunum this may be sufficiently great to obscure any pH effect. Mossberg (1967) found
that pH had no influence in vitro upon ammonia transport by the hamster ileum, although the
Absorption of ammonia from the rat intestine
777
differencesin pH used by him were comparatively small. He did not use buffered preparations,
so that the local changes at the mucosal surface induced by bicarbonate secretion may have
been overlooked. Mossberg also claimed that ammonia transport can be inhibited by the use
of the metabolic inhibitors 2: 4 dinitrophenol and sodium cyanide. These metabolic poisons
may, however, have reduced ammonia absorption because they inhibited mucosal secretion of
bicarbonate. In isolated loops of bowel which were constantly perfused by bicarbonate-rich
blood, these inhibitors influenced ammonia absorption only in the direction to be expected
from the change in pH which they produced (Swales, unpublished observations).
Where ammonia is being absorbed by a process of non-ionic diffusion in the lipid phase, the
rate of absorption should be independent of water movement through the ileal wall. In these
experiments ammonia absorption was significantly impaired when solutions of only moderate
hypertonicity were infused. This effect cannot be attributed to solvent drag (i.e. movement of
water in the opposite direction impairing ammonia movement along a concentration gradient)
since the impairment persisted for at least 1 h after the hypertonic solution was removed.
This finding suggests that hypertonic solutions produce a change in the ileal wall, reducing its
permeability.
The absorption of hydrazine from mannitol solutions was closely correlated with changes in
ammonia absorption. A smaller decrease in ammonia absorption was noted when urea was
used to adjust osmolalities. There was a significant, although smaller correlation between
absorption of ammonia and hydrazine from these solutions. Two explanations are possible for
the similarity between ammonia and hydrazine absorption. Both substances could share the
same active transport process, or, alternatively both could be absorbed by a process of diffusion.
In favour of the former hypothesis, it could be argued that hydrazine, like ammonia, can combine with glutamic acid, and form glutamine analogues. However, these compounds are much
less readily formed in biological systems than the corresponding ammonia compounds (Willis,
1966).In favour of absorption by diffusion are the close physical similarities of these substances:
both ammonia and hydrazine are weak bases (pK a 9·1 and 7·9 respectively) and the unionized
species of each has a moderate degree oflipid solubility. Both show pH dependence of urinary
excretion (Orloff & Berliner, 1956; Coe & Korty, 1967), absorption from the buccal mucosa
and from the ileal mucosa (unpublished observations).
The results presented here, therefore, support the view that non-ionic diffusion plays an
important role in the absorption of ammonia from the small and large intestine. This conclusion is in keeping with the findings of Price et al. (1967) in both the isolated hamster ileum
and the dog jejunum, and also with observations on absorption of ammonia from gastric juice
(Fleshler & Gabuzda, 1965) and on absorption of ammonia from saliva (Swales, Kopstein &
Wrong, unpublished observations). Additional evidence is provided by Elkington, Floch &
Conn (1969), who found the synthetic disaccharide lactulose lowered the colonic pH and the
arterial ammonia levels in patients with the neurological complications of cirrhosis.
Our results indicate that the inhibiting action of hypertonic solutions upon ammonia and
hydrazine absorption resembles the action of these solutions on glucose absorption from the
ileum and jejunum (Kameda et al., 1968). In both experiments it is due to changes in permeability caused by hypertonicity rather than changes in the direction of water flow through the
ileal wall. If this is so, estimates of intestinal pore size made by studies of the osmotic effect of
mannitol and urea solutions (Fordtran, Rector, Ewton, Soter & Kinney, 1965)are of questionable validity. An analogous situation may exist in the renal tubule where mannitol and other
778
J. D. Swales, J. D. Tange and O. M. Wrong
non-resorbable osmotically active substances produce histological tubular changes (Schreiner
& Maher, 1965).
ACKNOWLEDGMENT
The authors acknowledge with gratitude the assistance of Mrs Kate Brown. This work was
supported by M.R.C. Grant No. G 968j42jC.
REFERENCES
BOURKE, E., MILNE, M.D. & STOKES, G.S. (1966) Caecal pH and ammonia in experimental uraemia. Gut, 7'
558-561.
COE, F.L. & KORTY, P.R. (1967) Mechanism of hydrazine excretion by the mammalian kidney. American Journal
ofPhysiology, 212, 394-399.
CONWAY, E.J. (1962) Micro-diffusion Analysis and Volumetric Error, 5th revised edn, pp. 98-100. Crosby,
Lockwood and Son, London.
DAMBRAUSKAS, T. & CORNISH, H.H. (1962) A modified spectrophotometric method for the determination of
hydrazine. American Industrial Hygiene Association Journal, 23, 151-156.
ELKINGTON, S.G., FLOCH, M.H. & CONN, H.O. (1969) Lactulose in the treatment of chronic porta-systemic
encephalopathy. New England Journal of Medicine, 281, 408-412.
FLESHLER, B. & GABUZDA, G.T. (1965) Effect of ammonium chloride and urea infusions on ammonium levels
and acidity of gastric juice. Gut, 6, 349-356.
FOLIN, O. & DENIS, W. (1912) Protein metabolism from the standpoint of blood and tissue analysis. Second
paper. The origin and significance of the ammonia in the portal blood. Journal ofBiological Chemistry, 11,
161-167.
FORDTRAN, J.S. & DIETSCHY, J.M. (1966) Water and electrolyte movement in the intestine. Gastroenterology,
50, 263-285.
FORDTRAN, J.S., RECTOR, F.C., JR, EWTON, M.F., SOTER, N. & KINNEY, J. (1965) Permeability characteristics of
the human small intestine. Journal of Clinical Investigation, 44, 1935-1944.
JONES, E.A., SMALLWOOD, R.A., CRAIGIE, A. & ROSENOER, V.M. (1969) The enterohepatic circulation of urea
nitrogen. Clinical Science, 37, 825-836.
KAMEDA, H., ABEl, T., NASRALLAH, S. & IBER, F.L. (1968) Functional and histological injury to intestinal
mucosa produced by hypertonicity. American Journal ofPhysiology, 214, 1090-1095.
KETIERING, R.F. & SUMMERSKILL, W.H.J. (1967) Comparison of effects of pH and concentration gradients on
ammonia transport in the human jejunum. Medicine (Baltimore), 46, 91-96.
MILNE, M.D., SCRmNER, B.H. & CRAWFORD, M.A. (1958) Non-ionic diffusion and the excretion of weak acids
and bases. American Journal of Medicine, 24, 709-729.
MOSSBERG, S.M. (1967) Ammonia absorption in hamster ileum: effects of pH and total CO 2 on transport in
everted sacs. American Journal ofPhysiology, 213, 1327-1330.
ORLOFF, J. & BERLINER, R.W. (1956) The mechanism of the excretion of ammonia in the dog. Journal ofClinical
Investigation, 35, 223-235.
PETERS, J.P. & VAN SLYKE, D.D. (1932) Quantitative Clinical Chemistry, Vol. II Methods, pp. 285-296. Ballliere,
Tindall and Cox, London.
PRICE, J.B., SCHWARTZ, G.F., MOLAVI, A., BRITION, R.C. & VORHEES, AB. (1967) Mechanism and clinical
significance of intestinal ammonia transport. Surgical Forum, 18, 331-333.
RICHARDS, P., METcALFE-GmsoN, A, WARD, E.E., WRONG, O. & HOUGHTON, B.J. (1967). Utilisation of ammonia nitrogen for protein synthesis in man and the effect of protein restriction and uraemia. Lancet, ii,
845-849.
ROSENFELD, J.B., ABOULAFIA, E.D. & SCHWARTZ, W.B. (1963) Influence of non-ionic diffusion on absorption of
NH;j: and HCO'3 from the bladder. American Journal ofPhysiology, 204, 568-572.
SCHANKER, L.S., Tocco, D.J., BRODIE, B.B. & HOGBEN, C.AM. (1959) Absorption of drugs from the rat small
intestine. Journal ofPharmacology and Experimental Therapeutics, 123, 81-88.
Absorption of ammonia from the rat intestine
779
SCHREINER, G.E. & MAHER, J.F. (1965) Toxic nephropathy. American Journal of Medicine, 38, 409-449.
SHERLOCK, S. (1958) Pathogenesis and management of hepatic coma. American Journal of Medicine, 24, 805813.
STABENAU, J.R., WARREN, K. & RALL, D.P. (1959) The role of pH gradient in the distribution of ammonia
between blood and cerebrospinal fluid, brain and muscle. Journal of Clinical Investigation, 38, 373-383.
SUMMERSKlLL, W.H.J. (1966) On the origin and transfer of ammonia in the human gastro-intestinal tract.
Medicine (Baltimore), 45, 419-496.
WALSER, M. & BODENLOS, L.J. (1959) Urea metabolism in man. Journal of Clinical Investigation, 38,16171626.
WILLIS, J.E. (1966) Substitution of I-methylhydrazine for ammonia in the glutamine synthetase system. Biochemistry (Washington), 5, 3557-3563.
WILSON, D.R., lNG, T.S., METCALFE-GIBSON, A. & WRONG, O.M. (1968) The chemical composition of faeces in
uraemia, as revealed by in-vivo faecal dialysis. Clinical Science, 35, 197-209.