вход по аккаунту



код для вставкиСкачать
Original Papers
© 1987 S. Karger AG, Basel
0012-282 3/87/0371-0001 $2.75/0
Digestion 37: 1-9 (1987)
Metal Absorption and the Intestinal Mucus Layer
J. Quarterman
Rowett Research Institute, Bucksburn, Aberdeen, UK
Key Words. Intestinal absorption ■Mucus ■Glycocalyx • Zinc • Iron • Starvation
The mucus lining of the gastrointestinal
tract has lubricative and protective func­
tions, but it may also have a role in the
breakdown and absorption of nutrients from
the digesta in the lumen. Evidence for a role
in digestion has been described by Ugolev
[1], who found both digesta and digestive
enzymes in this mucus layer by filling an
intestinal loop with agar jelly. The mucus
which was removed by this technique was
described by them as the glycocalyx, the fila­
mentous extrusions from the microvilli visi­
ble in the electron microscope. However, it
is apparent from the work described below
in which this technique has been used that
the amount of material found in the agar gel
is too great to be identified with the glycoca­
lyx and must include the amorphous mucus
layer which lies above it and is much thicker
than it. This technique makes it possible to
separate cleanly the mucus layer from the
mucosa and permits observations on these
separately. Previous work has used mucosal
scrapings which include both mucus and
The structure of the intestinal mucosa
and mucus production and secretion by it
are known to be influenced by a number of
dietary treatments and feeding procedures
[2-4], Our interest was drawn to the possibil­
ity that the increase in absorption of several
trace elements after an overnight fast [5]
may be related to the observed increase in
metal-binding capacity of the mucus layer
Downloaded by:
Vanderbilt University Library - 10/26/2017 3:39:29 PM
Abstract. The luminal mucus layer of the rat’s small intestine was separated from the
mucosa by filling the lumen with agar jelly. When the rats were fasted overnight and given
59Fe or 65Zn by stomach tube the mucus isolated by this technique contained more isotope
and more sialic acid than fed controls, and more of the isotope was absorbed into the tissues.
Saturation and time-course studies showed that the mucus layer has metal-binding activity
distinct from that of the mucosa and it is postulated that the mucus layer may play a role in
the process of metal absorption.
after such a fast. These findings have been
reported briefly [6] and are presently de­
scribed in detail with some observations on
the kinetics of zinc uptake by the mucus
layer and the mucosa.
of sialic acid [11]. It was found that the presence of
agar had an insignificant effect on the estimate of
DNA. The effect of warm and cold agar on the stabil­
ity of sialic acid was estimated by homogenizing por­
cine intestinal mucus, alone, with cold agar gel or with
warm (50 *C) agar gel. Cold agar gel had no effect, but
warm gel caused the loss of about 10% of the sialic
Materials and Methods
Preparation o f Agar Cast
A length of small intestine, usually the first 10-20
cm from the pylorus, was removed from an anaesthe­
tized rat, washed out with saline, filled with 3 % agar
or agarose solution at about 45 °C, tied at each end
and cooled immediately in ice. The length of gut was
then slit open longitudinally and the tissue carefully
peeled off the agar cast. The mucosa was removed
from the gut wall by gently scraping with a micro­
scope slide. This method is based on that of Ugolev et
al. [7] who claim that the surface of the mucosa was
undamaged by treatment with agar; even the micro­
villi are intact. Electron microscopic observations of
the mucosa (carried out by Dr. T.P. King of this Insti­
tute) in preparations from the present experiments
have confirmed that agarose does not damage the
Isotopic Measurements
The activity of tissues containing 59Fc or 65Zn was
measured in a -/-counter (Gamma Guard 400; ICN
Tracerlab., Hersham. Surrey, UK) and of gut-free rat
carcasses in a whole-body counter (Nuclear Enter­
prises, Sighthill, Edinburgh, UK). [3-Activity of tis­
sues containing ,5S was measured in a Beckman LS345 liquid scintillation counter after agar or mucosal
samples had been treated with 1 N NaOH and mixed
with a scintillator.
Chemical Measurements
Protein was estimated with Folin reagent, phos­
phate with ammonium molybdate, sialic acid by the
method of Warren [8] after hydrolysis of samples at
8 0 0C for 1 h in 0.1 N H2S 0 4 and hexosamine by that
of Lewy and McAllen [9]. DNA was estimated with
diphenylamine [10] with a correction for the presence
The rats were given a semipurified diet containing
casein, sucrose, arachis oil, minerals and vitamins
[12]. Iron or zinc supplements were included or omit­
ted from the diet as described in the individual exper­
Experiments and Results
Composition o f the Mucus Removed by
Agar and the Effect o f Fasting
Two groups of 6 rats were used in this
experiment. One group was given food up to
the time the rats were anaesthetized and
killed and the other group was given water
but no food for 16-18 h before that time.
Three hours before being killed, each rat
received 5 pCi [32S] NajSO^ (specific activ­
ity 30 Ci/mg) in 0.1 ml saline intraperitoneally. Agar casts were made of the first 20 cm
of the small intestine from the pylorus and
the mucosa was removed by scraping. Casts
and mucosae were homogenized and dia­
lysed against distilled water with sodium
azide (500 mg/1).
The analyses of the casts and the mucosa
for protein, DNA, hexosamine, phosphate,
sialic acid and P-radioactivity per centimetre
length of intestine are given in table 1. About
a fifth of the DNA in the gut loop was found
in the agar cast. This DNA may be derived
from white blood cells which migrate
through the mucosa and from sloughed-off
mucosal cells embedded in the mucus and
Downloaded by:
Vanderbilt University Library - 10/26/2017 3:39:29 PM
Hooded Lister female rats (Rowctt Institute
strain) weighing about lOOg were used in all the
Metal Absorption and Mucus Layer
Table I. Effect of fasting on the composition of the agar cast and mucosa (units/cm gut length, mean ± SE)
Agar gel
fed rats
DNA, pg
Protein, mg
35 ± 3
1.00 ±0.09
fasted rats
fed rats
fasted rats
42 ± 7
101 ± 9
125 ± 16
1.66 ±0.23*
Phosphate, pg
24 + 2
31 ±5
57 ± 6
83± 12
Hexosamine, pg
18 + 2
27 ± 5
25 ± 3
28 ±4
50 ±6
109 ±22*
762 ±169
911 ± 151
Sialic acid, pg
35SOj, cpm
0.68 ±0.12
1,593 ±389
1.67 ±0.28**
1,161 ± 110
not removed by the preliminary saline rinse.
There was no evidence that such loss of mu­
cosal cells was increased in these experi­
ments since there was no electron micro­
scopically visible evidence of damage to the
mucosa. Similar distributions of protein and
phosphate were found, but hexosamine was
distributed about equally between agar cast
and mucosa. This distribution of hexos­
amine suggests that, if nondialysable hexos­
amine can be taken as a measure of mucus,
the surface mucus layer contains as much
mucus as the underlying mucosa which se­
cretes it. It is not known, however, how
much of the estimated hexosamine is de­
rived from glycoproteins and how much
from glycosaminoglycans. A surprising result
was that the sialic acid content of the agar
cast was only one hundredth of that of the
mucosa. Some sialic acid may be lost during
contact with warm agar (see Materials and
Methods section) but it is unlikely that such
a large difference was entirely artefactual
since the duration of the contact of warm
agar with the mucus was similar in the trials
described above and in the intestinal prepa­
rations. If the difference is real then the
mucus secreted from the mucosa is different
in composition from the bulk of that present
in the mucosa. Either a small component
with a rapid rate of turnover is secreted to
form the mucus layer or the mucus secreted
is desialated as it leaves the mucosal cells.
The latter possibility is unlikely as mucus is
stored in large vesicles in goblet cells before
rapid expulsion into the lumen.
A greater proportion of the nondialysable
35S-[SO.i] injected was found in the agar cast
than in the mucosa. Since the activity was
determined 3 h after injection it is likely that
most of the activity was present as sulphated
glycoproteins and glycosaminoglycans and
that by this time more than half of the label
had been incorporated and secreted by the
mucosa into the surface mucus layer.
The effect of the overnight fast is seen
only in the sialic acid content of the agar and
the mucosa and in the protein content of the
Downloaded by:
Vanderbilt University Library - 10/26/2017 3:39:29 PM
Fasted rats significantly different from fed rats: *p < 0.05; **p < 0.01.
Table II. Effect of fasting on 59Fe uptake
cpm IO~3/ml
cpm 10'Vcm loop
cpm 10 '/cm loop
Gut-free carcass
cpm 1 0 '3
4.96± 1.15
0.63 ±0.15
0.24 ±0.05
22.08 ±1.07
5.45 ±0.58
3.15 ± 0.77
163.3 ± 37.8
25.31 ±2.64
6.20 ±0.92
208.8 ±13.1
Table III. Effect of fasting on sialic acid content of
agar cast and mucosa
Period of
fasting, h
Agar sialic acid
pg/cm gut length
Mucosa sialic acid
pg/cm gut length
0.87 ±0.21
405 ±43
1.90 ±0.21“
485 ±89
3.33 ± 0.28b
409 ±36
Significantly different from 0 h fast, p < 0.01.
Significantly different from 16 h fast, p < 0.01.
agar. The sialic acid content, in spite of the
large differences in absolute content, was
more than doubled by fasting.
Effect o f Fasting on Intestinal Mucus
Experiments with 59Fe. In this experi­
ments. three groups of 5 rats weighing 100120 g were given the semipurified diet with­
out iron for 2 days. The use of this diet was
to reduce the concentration of iron in the
lumen of the gut and minimize the differ­
ences in lumen mucosal iron contents among
rats. Food but not water was withdrawn
from one group 40 h before the experiment,
another group was fasted overnight (1620 h) and a third group was used with no
period of fasting. One hour before being
killed, each rat was given 0.2 ml of saline
containing 2 pCi 59FeCl3, 1 pg Fe as
F e S O ^ fF O and 0.2 mg ascorbic acid intu­
bated into the stomach. The rats were anaes­
thetized with sodium barbital, blood was
taken from the heart and a loop of small
intestine was isolated and filled with agar as
described above. Care was taken to prevent
any gut contents escaping into the carcass.
The length of the gut loop filled with agar
was recorded. (3-Radioactivity was measured
in the agar cast and the mucosa scraped from
the loop and in blood and the gut-free car­
cass. Sialic acid was estimated in the agar
and the mucosa from each loop.
The absorption of the oral dose of iron, as
indicated by the activity in blood and car­
cass, was increased more than 4-fold by
overnight fasting. A longer period of fasting
(40 h) produced a further nonsignificant in­
crease of absorption (table II.) At the same
time the 59Fe activity in the agar cast con­
taining the mucus and in the mucosa was
increased by an even greater factor by these
periods of fasting. The sialic acid content
was found to be increased by fasting in the
agar cast but not in the mucosa (table III).
Experiments with 6SZn. The experiments
using 65Zn were conducted in a similar man­
ner to the experiment with 59Fe. Rats were
Downloaded by:
Vanderbilt University Library - 10/26/2017 3:39:29 PM
Period of fasting
Metal Absorption and Mucus Layer
Table IV. Effect of fasting on the 65Zn uptake of mucus in the agar cast and the mucosa: content of agar cast
and mucosa (units/g wet mucosa)
Agar cast
65Zn uptake
cpm 10"3
13.1 ±0.4
4.6 ± 1.4
0.8 ±0.2
24.9 ±4.7*
11.3± 1.2*
7.6 ±1.0*
1.5 ± 0.2*
Sialic acid
8.06 ± 1.24
7.75 ±2.48
9.92 ± 3.41
31.62 ±7.13*
18.91 ± 1.89*
12.71 ±1.86
1.8 ±0.2
3.1 ±0.6
2.7 ±0.7
2.7 ±0.7
0.4 ±0.1
1,705 ± 112
2,480± 527
1,240 ±279
620 ±124
4.8 ±0.9*
2.2 ±0.3
3.5 ±0.4
0.6 ±0.1
1.953 ± 124
1.674 ± 124
1.364 ±155
651 ±159
given the semipurifted diet without zinc for
2 days (to minimize zinc concentrations in
the lumen and gut tissues), and food but not
water was withdrawn from one group 16 h
before the experiment. Each rat was given by
intubation into the stomach about 3 pCi
65Zn in 0.2 ml saline containing 5 pg Zn/ml
1 h before killing and agar casts were made
of sections of the small intestine, in some
cases a duodenal section 10-15 cm long from
the pylorus and an ileal section using the last
10-15 cm before the ileo-caecal junction, in
other cases only the duodenal section. The
55Zn activity in the agar casts and the muco­
sal scrapings was measured and both of these
samples were analysed for protein, silaic
acid, hexosamine, phosphate and DNA. In
order to present the data for the agar casts
and the mucosa scrapings on a comparable
basis, the parameters were estimated for ma­
terial from the whole of the intestinal section
and then divided by the wet weight of muco­
sal scrapings from that section. The results
are shown in table IV.
As expected, fasting increased the net ab­
sorption of 65Zn. Thus, for example, in ex­
periment A, heart blood taken at the time of
killing had 292 ± 46 cpm of 65Zn activity/ml
in fasted rats and 154 ± 14 cpm/ml in fed
rats; in experiment C, liver of fasted rats had
4.8-103 ± 1.1-103 cpm/g and fed rats
2.6-103 ± 0.3-103 cpm/g. The activity of
65Zn in the agar casts of duodenal sections
was increased in every case by fasting by
from 3 to 8 times, whereas that of ileal casts
was only doubled in one experiment and
decreased in another. 65Zn activity in the
mucosa was increased by fasting to a very
much smaller extent, reaching significance
in only one experiment.
The sialic acid content of the agar casts
was increased by fasting, to a greater extent
Downloaded by:
Vanderbilt University Library - 10/26/2017 3:39:29 PM
* Fasted rats significantly different from fed rats, p < 0.05.
1 Three experiments were conducted in a similar manner in experiment A, only 65Zn uptake from duodenal
loops was measured.
Fig. 1. The relationship between 65Zn concentration in the mucosa and in a lumenal agar cast and time after
an oral dose of 65Zn in 5 pg Zn as Z 11SO4 given to rats: (a) in the proximal 15 cm of the small intestine and (b) in
the next 15 cm.
Distribution o f 65Zn in Intestinal Agar
Casts and Mucosa at Intervals o f Time
after Dosing
In this experiment there were six groups
of 6 rats weighing about lOOg. They were
given the low zinc diet and all were starved
overnight as described previously. They
were then intubated with 2 pCi 65Zn in 0.2
ml saline containing 5 pg Zn/ml into the
stomach and agar casts of the small intestine
were made 0, 0.5, 1, 2, 4 and 6 h later. Two
casts were made for each rat, one of the first
15 cm from the pylorus and another of the
next 15 cm caudally. Zn activity was mea­
sured in the agar casts and the mucosa
scraped from the sections and expressed as
activity per centimetre of the intestinal sec­
tion. Figure 1 shows these data with stan­
dard errors on a semilogarithmic plot. In the
duodenal sections the increase and decrease
of 65Zn concentration is seen to be exponen­
tial in both compartments and the maximum
concentration occured in the agar casts half
an hour before it did in the mucosa. The
uptake and release of 65Zn by the agar casts
was more rapid than by the mucosa. In the
second 15-cm sections (fig. lb) a generally
similar relationship of activity and time was
observed with maximum 65Zn contents oc­
curring in the agar casts before those in the
mucosa. Although the standard errors of the
Zn contents in this section were generally
smaller than in the duodenal section, the
mean values for 65Zn contents in the mucosa
were erratic.
Concentration of Zn in Intestinal Agar
Casts and Mucosa after Different Oral
Doses o f Zn
This experiment was conducted in the
same way as the last except that instead of
giving rats the same dose of zinc at different
intervals of time before the agar cast was
made, they were intubated into the stomach
with 2 pCi 65Zn in 0.2 ml saline with 1,5, 10,
25 or 50 pg Zn and agar casts made after 1 h.
Downloaded by:
Vanderbilt University Library - 10/26/2017 3:39:29 PM
in the duodenum than in the ileum, but the
sialic acid content of the mucosa was unaf­
Metal Absorption and Mucus Layer
Fig. 2. The relationship be­
tween 65Zn concentration in the
ntucosa and in a luminal agar
cast and the amount of ^Znlabelled Zn given as an oral dose
to rats. Observations were made
on the proximal 10 cm of the
small intestine.
This work is based on the technique of
filling a loop of small intestine with warm
agar solution, and allowing it to cool and set
during which time it traps material on the
luminal surface of the gut. Ugolev and co­
workers [I, 7] who used this technique to
develop their ideas of membrane digestion
assumed that the material removed by the
agar was the glycocalyx. They believe that
the glycocalyx provides a structural environ­
ment for the absorption of enzymes of pan­
creatic and mucosal origin, and organic di­
gestion products, and for their progressive
absorption. It is clear from this work, how­
ever, that the material removed by the agar
must also contain a considerable part, if not
all of the amorphous mucus layer overlying
the glycocalyx and some cell debris. Mucus
from the stomach [13, 14] and the small
Downloaded by:
Vanderbilt University Library - 10/26/2017 3:39:29 PM
If the weight of Zn in the agar and the
mucosa per centimetre gut is a measure of
the velocity of uptake of the metal into the
two compartments, the results can be de­
scribed as a plot of velocity of uptake against
concentrations of zinc in the intubated dose
(fig. 2).
The rate of uptake of zinc by the agar cast
and the mucosa increased linearly with the
quantity of zinc placed in the ligated loop,
with some evidence of saturation of the mu­
cosal uptake above a load of 25 pg. From
double reciprocal plots of the same data, Km
values of 29 and 50 pg/ml can be calculated
for the agar cast (i.e. mucus layer) and mu­
cosa respectively. This experiment was re­
peated (results not shown) with zinc loads up
to 200 pg and gave similar results with satu­
ration of the glycocalyx and the mucosa
above 50 pg Zn and Km values of 14 and
67 pg/ml.
intestine [15, 16] has been shown to bind
several metal ions. Digesta normally passes
rapidly through the duodenum and jejunum
and the luminal mucus layer is its only con­
tact with the tissue wall. This mucus layer
may absorb metals rapidly during the pas­
sage of the digesta, and then transfer of met­
als into and transport across the mucosa can
take place more slowly without loss of the
metals by passage down the gut. These exper­
iments do not provide any further informa­
tion about the nature of metal binding by
intestinal mucus although data from the sat­
uration and time-course studies are compat­
ible with the idea. They showed that the
maximum amounts of zinc taken from the
lumen was about the same (in terms of pg
Zn/unit length of gut) for both agar cast and
mucosa and that this was achieved in the
agar 30 min before the mucosa. The strength
of binding of zinc for the agar, as judged
from the K.m values, was greater than for the
mucosa, but zinc was both taken up by and
lost from the agar more rapidly than the
It has been reported that fasting increased
the amount of mucus glycoprotein in the cor­
pus of rats’ stomachs but did not affect that in
the antrum, and there was evidence of a
change of composition of the glycoproteins
[17]. There was a decrease of sulphate and
sialic acid relative to fucose. In the rat small
intestine the amount of mucus present after
fasting was increased considerably at the vil­
lus tips though not towards the base [3]. In
the present work no fractionation of the mu­
cosa was attempted and no similar change
was observed in the analyses of the whole tis­
sue. However, the secreted mucus in the agar,
as suggested by the raised content of sialic
acid, hexosamine and protein, was increased
(table I). The sialic acid data are noteworthy.
Fasting consistently induced a change in the
composition of the mucus; in both the mu­
cosa and the agar the sialic acidihexosamine
ratio was doubled by fasting. The ratios them­
selves were very different however in the two
compartments, due to a 100-fold decrease in
sialic acid content in the agar. In the mucosa
this ratio was rather lower than that found for
rat goblet cell mucin [18] or pig small intes­
tine mucus [19].
The mucus secreted from the mucosa is a
mixture of that derived from goblet cells and
that secreted from the brush border. Autora­
diographic studies using 35S0 4 have shown
that there is much less labelling of the mucus
secreted at the tips of the brush border than
of goblet cell mucus [20], In the present
work, 35S activity was found in the agar and
its amount was not influenced by fasting,
especially it did not vary in the way the sialic
acid content varied between agar and mu­
cosa and fed and fasted treatments. It is not
known how the composition of mucus glyco­
proteins from these two sources differs,
apart from its sulphate content, or what pro­
portion of each is secreted into the lumen.
Mucus which has been isolated from mu­
cosa and may be a mixture has often been
called ‘goblet cell mucus’. This work which
separated the mucosa from its overlying mu­
cus layer suggests that the mucus secreted
into the lumen may not be the same as the
bulk of mucus present in the mucosa and
that starvation may influence the composi­
tion of its components in different ways.
The significant fact has emerged that the
change in composition is indicated by a
change in sialic acid content and this is ac­
companied by an increased quantity of mu­
cus, a change in the degree of metal binding by
the mucus and the extent of metal transport
across the mucosa into the tissues.
Downloaded by:
Vanderbilt University Library - 10/26/2017 3:39:29 PM
Metal Absorption and Mucus Layer
1 Ugolev, A.M.: Membrane (contact) digestion; in
Smyth, Biomembranes, vol. 4A, pp. 285-362 (Ple­
num Press, London 1978).
2 Biol, M.-C.; Marlin, O.; Ochninger, C.; Lonisot,
P.; Richard, M.: Biosynthesis of glycoproteins in
the intestinal mucosa. II. Influence of diets. A.
Rev. Nutr. 25: 269-280 (1981).
3 Quarterman, J.; Feams, L.M.: The influence of
diet on intestinal mucin production. Proc. Nutr.
Soc. 43: 129A (1984).
4 Smith, M.W.; Expression of digestive and absorp­
tive function in differentiating enterocytes. A.
Rev. Physiol. 47: 247-260 (1985).
5 Quarterman, J.; Morrison, E.: The effects of short
periods of fasting on the absorption of heavy
metals. Br. J. Nutr. 44: 277-287 (1981).
6 Quarterman, J.; A possible role for the glycocalyx
in metal absorption. J. Physiol. 322: 23P (1981).
7 Ugolev, A.M.; Smirnova, L.F.; Iezuitova, N.N.;
Timofeena. N.M.; Mityushova, N.M.; Egorova,
V.V.; Parshkov, E.M.: Distribution of some ab­
sorbed and intrinsic enzymes between the muco­
sal cells of the rat small intestine and the apical
glycocalyx separated from them. FEBS Lett. 104:
35-38 (1979).
8 Warren, L.: The thiobarbituric acid assay of sialic
acids. J. biol. Chem. 234: 1971-1975 (1959).
9 Levvy, G.A.; McAllen, A.; The N-acetylation and
estimation of hexosamincs. Biochem. J. 73: 127—
132 (1959).
10 Burton, K.: A study of the conditions and mecha­
nism of the diphenylaminc reaction for the calori­
metric estimation of deoxyribonucleic acid. Bio­
chem. J. 62: 315-323 (1956).
11 Croft, D.N.; Lubran, M.: The estimation of deoxy­
ribonucleic acid in the presence of sialic acid:
application to analysis of human gastric washings.
Biochem. J. 95: 612-620 (1965).
12 Davies, N.T.; Reid, H.: An evaluation of the phytate, zinc, copper, iron and manganese contents
of, and Zn availability from, soya-based textured
vegetable-protein meat substitutes or meal exten­
ders. Br. J. Nutr. 41: 579-589 (1979).
13 Jacobs, A.; Miles, P.M.: The iron-binding proper­
ties of gastric juice. Clinica chim. Acta 24: 87-92
14 Bella, A.; Kim. Y.S.: Iron binding of gastric mu­
cins. Biochim. biophys. Acta 304: 580-585
15 Forstner, J.F.; Forstner, G.G.: Calcium binding to
intestinal goblet cell mucin. Biochim. biophys.
Acta 386: 283-292 (1975).
16 Quarterman. J.: Metal binding properties of intes­
tinal mucus. Proc. X lllth Int. Congress of Nutri­
tion, 1985, p. 221.
17 Ohara, S.; Kakei, M.; Ishihara. K.; Katsuyama, T.;
Hotta, K.: Effects of fasting on mucus glycopro­
tein in rat stomach. Compar. Biochem. Physiol.
79B: 325-329 (1986).
18 Forstner, J.F.; Jabbal, I.; Forstner, G.G.: Goblet
cell mucin of rat small intestine. Chemical and
physical characterization. Can. J. Biochem. 51:
1154-1166 (1973).
19 Mantle. M.; Allen. A.: Isolation and characteriza­
tion of the native glycoprotein from pig small
intestine mucus. Biochem. J. 195: 267-275
20 Ito, S.; Revel, J.P.: Autoradiographic studies of
the enteric surface coat; in Sullivan, Gastrointesti­
nal radiation injury (Excerpta Medica Founda­
tion, Amsterdam 1967).
Received: June 12, 1986
Received in revised form: September 8. 1986
J. Quarterman, MD,
Rowett Research Institute,
Aberdeen AB2 9SB (UK)
Downloaded by:
Vanderbilt University Library - 10/26/2017 3:39:29 PM
Без категории
Размер файла
1 155 Кб
Пожаловаться на содержимое документа