вход по аккаунту


The passage of exogenous peroxidase from blood capillaries into the intestinal epithelium.

код для вставкиСкачать
The Passage of Exogenous Peroxidase from Blood
Capillaries into the Intestinal Epithelium
Department of Anatomy, Northwestern University Medical und
Dental Schools, Chicago, Illinois
Horseradish peroxidase was injected into the spleens of mice and the
animals were sacrificed ten minutes after injection. The tissues were reacted with
3-3' diaminobenzidine hydrochloride and the distribution of the reaction product was
studied with both the light and electron microscope. The peroxidase was localized
between epithelial cells up to the region of the tight junction and within vacuoles in
the absorptive cells. Granules ranging in size from ca. 40A to 600A were observed in
the cytoplasm of epithelial cells in numbers far in excess of that found in control
specimens. It appeared that the diffuse light brown staining observed in epithelial
cells with the light microscope could be attributed to large numbers of granules of
reaction product free in the cytoplasm. When corn oil was given by stomach tube and
an intravascular injection of perioxidase was given ten minutes later, absorbed lipid
was found to pass from interepithelial cell spaces to lamina propria at the same time
that peroxidase was traversing the same compartments in the reverse direction. Hence,
it was shown that exogenous peroxidase and probably other substances of vascular
origin required for the metabolism of epithelial cells are exposed to both the basal and
lateral epithelial cell membranes, even when absorbed lipid is traversing the same
spaces in the opposite dircction.
The fine structure of the intestinal epithelium and the pathway of absorption and
transport of lipid from gut lumen to lamina
propria and lacteals have been studied in
considerable detail. The results of such experiments showed that lipid was absorbed
in the form of micelles (Rostgaard and
Barrnett, '65) at the striated border of absorptive cells, esterified to triglycerides and
packaged for export as chylomicrons in the
endoplasmic reticulum and Golgi complex
(Isselbacher, '61; Isselbacher and Budz,
'63; Strauss, '63; Jersild, '66a,b). The Lipoprotein particles, or chylomicrons, were
discharged into intercellular spaces at
about the level of the nucleus (Palay and
KarLin, '59) from which they passed
through the epithelial basement membrane
into the lamina propria, where they were
pickcd up by lacteals (Casley-Smith, '62).
Karnovsky ('65) demonstrated that exogenous peroxidase administered intravenously can be used as an excellent lowmolecular weight tracer to follow transport
through capillary endothelium and into
very small extracellular spaces such as the
T-system of muscle.
This paper reports results of experiments
intended to determine whether peroxidase
enters intercellular spaces of the intestinal
ANAT. REC.,159: 159-170.
epithelium, where it could be expected that
diffusion into such spaces might have to
occur against the flow of absorbed substances in transit to the lamina propria.
Three C57 Brown mice weighing about
20 gm each were anesthetized and 5 mg of
CI type 2 horseradish peroxidase dissolved
in 0.5 ml of saline (Karnovsky, '65a) was
injected into the spleens through small abdominal incisions. The spleen was used
because of the relatively large amount of
material to be injected and the ease with
which post-injection hemorrhage could be
controlled. After an interval of ten minutes, segments of jejunum were excised
and fixed for two hours in a mixture of
formaldehyde and glutaraldehyde (Karnovsky, '65b) in 0.2 M phosphate buffer.
The tissues were then washed 18 hours in
0.2 M phosphate buffer, incubated for ten
minutes in 3-3' diaminobenzidine hydrochloride (Karnovsky, '65a), fixed one hour
in 2% Os01 in 0.2 M phosphate buffer, dehydrated in ethanol and embedded in
Araldite 502 in accordance with the method
of Luft ('61). Two additional animals were
fed 0.5 ml of corn oil by stomach tube and,
ten minutes later, were given intrasplenic
injections of peroxidase. At 20 minutes
after the administration of corn oil and
ten minutes after peroxidase injection, the
mice were killed and the tissues were
treated as described above.
The cytoplasm of epithelial cells contained a higher population of small granules than was observed in tissues prepared
in the same way except that peroxidase
was omitted (fig. 2). The size of the granules varied considerably, from ca. 40A to
600A and usually the larger granules apExamination of sections 1 p thick with peared to be aggregates of smaller ones.
the light microscope showed the pattern of The larger granules free in the cytoplasm
capillaries in the intestine in exquisite de- appeared to be similar in size and density
tail. Moreover, it was clearly evident that to large granules attached to membranes
the peroxidase had permeated intercellular lining the large cytoplasmic vacuoles (fig.
spaces of the epithelium at least as far as 2). The methods used do not distinguish
the tight junctions. Intracellular vacuoles between ribosomes and other normally occontaining the black reaction product could curing granules but it is presumed that
be seen in the cytoplasm of epithelial cells, the diffuse light brown staining reaction
observed with the light microscope can be
particularly on the distal half of the villus traced to large numbers of reaction product
and a diffuse light brown stain was present granules in the cells.
in the basaI cytoplasm. The muscularis
Figure 3 shows reaction product in interexterna also showed a heavy infiltration of cellular spaces as far as the region of the
reaction product but the lamina propria junctional complex but no evidence was
and epithelium in and around the crypts obtained that peroxidase was discharged
showed very little.
into the gut lumen. Even though large
In figure 1, a fenestrated capillary and vacuoles were observed in epithelial cells,
its basement membrane are shown and, at such as the one shown in figure 3 , no direct
least at the 10-minute interval after injec- evidence was found that such vacuoles
tion there seems to be no piling-up of reac- opened into the gut lumen. The beaded
tion product in either of these structures. appearance of microvilli shown in figure
This would suggest that neither the capil- 3 was found to be a function of aldehyde
lary endothelium nor the basement mem- fixation and not a specific reaction to
brane constituted a serious barrier to peroxidase.
transport and/or diffusion. A heavy conThe knowledge that absorbed lipid flows
centration of reaction product is present from intercellular spaces of the epithelium
in the region of the epithelial basement to the lamina propria and the demonstramembrane at the left of the figure and in tion that peroxidase of vascular origin can
the tortuous intercellular spaces of the go in the reverse direction raised the quesepithelium. Occasional aggregates of small tion as to whether peroxidase will follow
dense granules can be seen in the basal intercellular routes in the epithelium when
region of one of the epithelial cells and in lipid absorption is in progress. Figure 4
close proximity to the lateral cell mem- shows that, when corn oil was given by
branes of all of the cells shown in the stomach tube and this was followed by an
figure. We have not encountered similar intravascular injection of peroxidase, the
granules in the basal regions in any of the same routes were followed as when each
many epithelial cells we have studied over is given separately. It was noted that
the years. The distribution of reaction large lipid droplets rather than the smaller
product in intercellular spaces can be seen ones characteristic of chylomicrons were
better in figure 2. It is clear that the prod- always found. It turned out that the large
uct fills the spaces but it cannot be stated irregular droplets occurred as a result of
with certainty whether all of it is bound aldehyde fixation and that the presence of
to plasma membranes. The binding of peroxidase played no part in determining
reaction product to membrane lining intra- their size and morphology. Dense granules
cellular vacuoles (fig. 2) suggests that of reaction product were commonly found
nearly all of the membrane surface was in the lipid droplets between cells. It is
coated with bound peroxidase.
noteworthy, as shown in figure 4, that reac-
tion product in the epithelia and lamina
propria of mice which had received corn
oil was almost always associated with lipid
and that very little could be found attached
to plasma membranes and intercellular
spaces in regions in which lipid could not
be demonstrated.
The results of these experiments have
shown that horseradish peroxidase administered intravascularly could be traced
into intercellular spaces of the intestinal
epithelium at least as far as the junctional
complex and that some of it was taken into
epithelial cells where it could be identified
in granular form attached to membranes
lining vacuoles and probably free in the
cytoplasm. By implication, this would
mean that substances which leave blood
capillaries in the mucosa would be exposed
to the lateral cell membranes of intestinal
epithelial cells as well as to the basal cell
membrane. The reaction product was not
detected in the gut lumen but the presence
of it in epithelial cells, particularly i n large
vacuoles near the luminal surface, does
suggest a secretory pathway. In all probability, the agitation in processing fluids
would have removed any peroxidase discharged into the lumen.
When peroxidase was administered after
corn oil had been fed to the mice, the reaction product was found mostly between the
lipid and cell membranes. It looked as
though the peroxidase was quickly bound
to the membrane. But since the tissues
had to be fixed in aldehyde and since such
a fixative does not seem to stabilize lipid,
it is possible that chylomicrons might have
coalesced to form large droplets, and that
in the process, peroxidase seemed to be
forced to a more peripheral location,
namely, between the lipid and the plasma
membrane. It would appear that tissue
fluid was not miscible with the lipoprotein
complex comprising chylomicrons. It is
significant that movement from gut lumen
to lamina propria and the reverse can proceed simultaneously through the same tissue compartments.
Casley-Smith, J. R. 1962 The identification of
chylomicra and lipoproteins i n tissue sections
and their passage into jejunal lacteals. J. Cell
Biol., 15: 259-277.
Isselbacher, K. J. 1961 Fat absorption and the
esterification of fatty acids by intestinal mucosa.
Gastroenterology, 40: 259-260.
Isselbacher, K. J., and D. M. Budz 1963 Synthesis of lipoproteins by rat intestinal mucosa.
Nature, 200: 364-365.
Jersild, R. A. 1966a A time sequence study of
f a t absorption in the rat jejunum. Am. J. Anat.,
118: 135-162,
19661, A radioautographic study of
glyceride synthesis in vivo during intestinal
absorption of fats and labeled glucose. J. Cell
Biol., 31: 413-427.
Karnovsky, M. J. 1965a Vesicular transport of
exogeneous peroxidase across capillary endothelium into the T-system of muscle. J. Cell
Biol., 27: 49A.
fixation of high osmolality for use in electron
microscopy. J. Cell Biol., 27: 137A.
Luft, J. H. 1961 Improvements in.cpoxy resin
embedding methods. J. 3iophys. Biochem.
Cytol,. 9: 409-414.
Palay, S. L., and L. J. Karlin 1959 An electron
microscopic study of the intestinal villus. 11.
The pathway of fat absorption. J. Biophys.
Biochem. Cytol., 5: 373-384.
Rostgaard, J., and R. J. Barrnett 1965 Fine
structural observations of the absorption of
lipid particles in the small intestine of the rat.
Anat. Rec., 152: 325-350.
Strauss, E. W. 1963 The absorption of fat by
intestine of the golden hamster in vitro. J. Cell
Biol., 17: 597-607.
C, capillary
BNI, basement membrane
P, reaction product
G, granules
V, vacuoles
I, intercellular space
J, junctional complex
L, lipid
I A portion of a fenestrated capillary (C), the basal regions of several
epithelial cells. The capillary and its surrounding basement membrane
branes (BM). A large amount of reaction product (P) can be seen
in the region of the epithelial basement membrane and in intercellular
spaces. Small dense granules ( G ) can be seen in the cytoplasm of the
epithelial cells. The capillary and its surrounding basement membrane appear to be relatively clear of reaction product. x 10,000.
James C. Hampton and Benjamin Rosario
Reaction product can be seen in intercellclar spaces ( I ) and attached
to the iiiner lamellae of membranes lining several vacuoles ( V ) in the
basal region of an epithelial cell. The cytoplasm of the epithelial cell
contains many granules ( G ) , the smallest of which are ca. 40A and
the largest ca. 600A. x 80,000.
James C. Hampton and Benjamin Rosario
Reaction Froduct can be seen between epithelial cells as far as the
junctional complex ( J ) and within a large vacuole ( V ) in the apical
part of one of the cells. Occasional fragments of granular material ( G )
resembling that shown within the vacuole could be found i n close
association with microvilli but no lzrge accumulations were seen in
the gut lumen. x 10,000.
James C. Hampton and Benjamin Rosario
This figure illustrates movement of horseradish peroxidase ( P ) from
lamina propria to intercellular spaces in the epithelium and the
simultaneous movement of absorbed lipid ( L ) in the reverse direction
within the same compartments. x 13,500.
James C. Hampton and Benjamin Rosario
Без категории
Размер файла
978 Кб
capillaries, exogenous, passages, epithelium, intestinal, blood, peroxidase
Пожаловаться на содержимое документа