вход по аккаунту


Morphology of particulate glycogen in guinea pig liver revealed by electron microscopy after freezing and drying and selective staining en bloc.

код для вставкиСкачать
Department of Anatomy, The University of Chicago, Chicago, Illinois
Glycogen has been stated to occur in liver cells as flakes
(Lazarovitch-Hrebeljanovich, ' 3 6 ) , granules (Pfuhl, '32 ; and
Aterman, '52)' and irregular masses (Pfuhl, '32)' or in a
homogeneous form (M. R. Lewis, '21 ;W. H. Lewis, '26 ;Nordmann, '29 ; Bensley and Gersh, '33 ; and Mancini, '48). Morgan
and Mowry ('51) attempted with the electron microscope to
identify glycogen in human liver fixed with alcohol-formalin
and shadowed with palladium.
The subject has been restudied in liver specimens fixed by
ultrarapid freezing and drying (Gersh, Isenberg, Stephenson
and Bondareff, '57). The same staining method (periodic acidleucofuchsin) was used en bloc to color the glycogen for viewing with the light microscope and to produce enhanced contrast for viewing with the electron microscope. The specificity
of the method of staining was tested extensively and found
to be satisfactory. Glycogen appeared in paraffin sections as
'This work was aided by grants from the Wallace C. and Clara A. Abbott
Memorial Fund of the Universityy of Chicago, and the Commonwealth f i n d .
Present address: National Institute of Mental Health; National Institutea of
Health ; Public Health Service; Department of Health, Education, and Welfare;
Bethesda 14, Maryland.
small granules, 1-2 p in diameter, which were resolvable into
two classes of submicroscopic particles with the electron microscope.
Six male guinea pigs were fed a standard chow diet (Purina
Rabbit Chow) supplemented by lettuce and water ad libiturn
for 4 days. Six other guinea pigs were treated the same way
and after 4 days were removed to individual cages with widemesh screen bottoms; for 24 hours before being killed, the
animals had free access to water but no food.
All animals were killed by a sharp blow to the back of the
head and bled by opening the thoracic cavity and severing
the major vessels.
A portion of the left lobe of the liver was removed and
placed in a moist chamber. Minute pieces, about 0.1mm to
0.2 mm on a side, were cut by means of a double-bladed, modified Valentine knife. These were transferred to small pieces
of metal foil which were rapidly removed from the chamber
and immersed in liquid propane, precooled to -1175°C. by
immersion of a beaker in liquid nitrogen.a No more than
6 minutes elapsed between killing the animal and the freezing
of the specimens. The pieces of foil were then transferred
t o gelatin capsules containing liquid nitrogen. These were
placed in the drying chamber, maintained at -30°C.,
the specimens were dried at a low pressure (about .001 mm
Hg) over anhydrous phosphorous pentoxide for 24 hours.
The drying chamber was brought to room temperature, air
was admitted through a desiccant (silica gel), and the specimens were rapidly transferred to micro test tubes (Feigl)
containing 95% ethyl alcohol. The specimens remained in
these tubes throughout all subsequent manipulations until they
were embedded in methacrylate. An outline of subsequent
ew bloc procedures is given below.
' I t is important that the temperature be maintained above -180°C. to avoid
the condensation of atmospheric oxygen and the possible formation of an explosive mixture (Stevenson, '54).
I. Staining of glycogen by the periodate-leucofuchsin
method (modified after Hotchkiss, '48):
70% alcohol f o r 1 hour.
Alcoholic periodic acid (Solution A) for 30 minutes.
70% alcohol, 6 to 8 changes, for a total of 2 hours.
Leucofuchsin for 3 hours.
Sulfurous acid, 8 changes, for a total of 4 hours.
95% alcohol, 1 2 changes, for a total of 24 hours.
11. Salivary control :
1. 70% alcohol, 2 hours; 50% alcohol, 2 hours; distilled water
(pH 6 . 5 ) , 2 hours.
2. 50% filtered human saliva in M/10 phosphate buffer at pH
7.0 at 37°C. for 2 hours.
3. M/10 phosphate buffer a t pH 7.0 for 1 hour.
4. Distilled water, 2 or 3 changes, for 1 hour.
5. 50% alcohol, 6 changes, for 2 hours.
6. 70% alcohol, 6 changes, for 1 hour.
7. Steps 2 through 6 under I.
111. Phosphate buffer control f o r salivary digestion :
1. Step 1 under 11.
2. M/10 phosphate buffer a t pH 7.0 a t 37°C. for 3 hours.
3. Steps 4 through 7 under 11.
IV. Leucofuchsin control of the staining procedure :
Steps 1 through 6 under I except for step 2. Here the same
solution was used but for the omission of periodic acid.
V. Periodic acid control of the staining procedure:
Steps 1 through 6 under I, except that in step 4 the dye was
From the 95% alcohol, the tissues were dehydrated in absolute alcohol for two hours and then gradually infiltrated
with methacrylate monomer. The latter was a solution of 5
parts methyl methacrylate monomer and 95 parts n-butyl
methacrylate monomer which was polymerized at 50°C. in
the presence of 0.25% catalyst (Luperco CDB, Newman et
al., '49). Ultra thin sections were prepared, mounted on formvar films, and viewed with a Philips E M 100 (rated resolution :
50B) at initial magnifications ranging from 1,000 to 60,000.
For details of the above procedures see Gersh et al. ('57).
I n addition to the above, some specimens, treated as described above in Procedures I, 11, and 111, were cleared in
xylene, embedded in paraffin, and sectioned at 6 p . These
preparations were studied and photographed under oil immersion with a light microscope.
A p p e a r a m e of glycogen zoith the light microscope. Glycogen
appears in the livers of fed guinea pigs as numerous granules,
about 1 or 2 p in diameter, stained brilliant red. These are
present throughout the cytoplasm but are absent from the
nucleus (fig. 1).After prior treatment with phosphate buffer,
the picture is essentially unaltered, except that the stain is
not as intense and there are indications of swelling (fig. 2).
As a result of treatment with saliva diluted with the same
buffer, no glycogen particles are visible (fig. 3). I n the livers
of starved guinea pigs, glycogen is greatly reduced in amount.
Where present, it occurs as densely-staining granules confined
to the cytoplasm of a small number of cells.
Electron microscopic appearance of glycogen in liver cells
of f e d guiFnea pigs. At low magnification glycogen appears
as small granules (3,000 to 9,000 or larger) distributed
throughout the cytoplasm. The granules may appear circular,
oval o r irregular in shape (fig. 4). These have been designated
third order granules. The smaller size of the granules as
compared with those viewed with the light microscope is attributable, as least in part, to the thinness of the sections.
At higher magnifications, the granules appear more irregular and are generally paler in the peripheral than in the central
parts. They consist of smaller aggregates, chiefly elliptical,
between 600 and 1,500 A in diameter. These smaller units
are called second order particles. Their peripheries are paler
than their central portions (fig. 7) and because of this they
are difficult to measure accurately.
At still higher magnification, the second order particles are
seen to contain the smallest, roughly spherical particles, which
are approximately 130 A in diameter (range : 106 to 152 a)
when viewed with the Philips E M 100 (fig. 8). The average
axial ratio (minimum diameter/maximum diameter) of the
central part of these first order particles is about 0.8. Each
is separated from its neighbors by what appears as a continuous paler region.
A few sections from one specimen were examined with the
Siemens Elmiskop I by Frl. Dr. C. Weichan (fig. 9). I n these
high resolution electron micrographs the diameter of the third
order particle was found to range between 66 b and 96 and
the average axial ratio was 0.46.
Salivary digestion of glycogen. No granules are visible.
The protoplasm shows little contrast and the image shows
very little detail (fig. 5) though mitochondria and nuclei are
Phosphate buffer control f o r saliuary digestion. Second
and third order granules are clearly visible. They are paler,
less sharply defined and somewhat larger than in specimens
stained directly without prior treatment (fig. 6). No first
order particles are visible.
Leucofuchsin and periodic acid controls of t h e staining
procedure. No granules are visible. The protoplasm shows
little contrast and the image shows very little detail.
Other ilztracellular colzstituents. I n untreated sections of
livers from fed animals viewed with the light microscope,
mitochondria are not visible and nuclei appear as clear, completely unstained areas. I n electron micrographs, however,
both nuclei and mitochondria are visible. Their appearance
remains unchanged following the various procedures described in the section on Materials and Methods.
Contrast in the electron microscope results primarily from
differential scatter of electrons. To observe a detail of structure, therefore, the electrons must be scattered more strongly
by it than by its surroundings. I n addition, if a material such
as a dye were to be specifically bound to a cellular component,
the density (and the electron scattering) of the region of
binding would be increased, and a correspondingly contrasty
image would appear. Such a material need not be a heavy
metal, but as has been shown by Isenberg ('56), may be
any of the organic dyes commonly used in cytochemical procedures. I n the material described here, fuchsin sulfite is
bound specifically to glycogen. The fuchsin-glycogen complex is denser than glycogen (approximate ratio 3 : l), and
because of this casts an enhanced shadow on the screen or
photographic film.
It should be mentioned here that heavy metal forms (iodinated and brominated compounds) of basic fuchsin were
prepared and found to be unsatisfactory for several reasons;
namely, they were not as soluble, in both alcohol and water,
as is common basic fuchsin, and they were not histochemically
specific for glycogen.
It is clear from the above that glycogen particles which
are stained deep red should be expected to be, and in fact
are identifiable as electron scattering objects. This specificity of glycogen staining as viewed with the light microscope
and with the electron microscope is shown empirically by the
results described in the prior section. F o r example, it could
be seen with both instruments that glycogen was removed
by salivary amylase and was not stained by periodic acid
alone nor by leucofuchsin alone. That glycogen particles are
identifiable as electron scattering objects is further supported
by the removal of only part of the stainable glycogen by
P-amylase and a corresponding reduction in contrast with
the electron microscope and finally, by the great reduction
of glycogen in the liver cells of starved as compared with
fed animals. These findings lead to the conclusion that the
periodic acid-Schiff method of staining glycogen f o r the light
microscope, as well as for the electron microscope, is specific
under the conditions employed.
In sections observed with the light microscope, as well as
in others viewed with the electron microscope, glycogen is
distributed in the cytoplasm in particulate form. There are
three orders of particles: the granules visible in the light
microscope or the third order granules ; the submicroscopic
particles which range from about 600 A to about 1500 or
the second order particles; the smallest particles of about
130B diameter, which are regarded as the central part of
the first order particles. The remainder of the discussion is
conflned to a consideration of these three orders of glycogen
Glycogen occurs as flakes or granules in liver cells only
after the use of fluid fixatives. Lison ('53) has this comment
on the use of such fixatives for the preservation of glycogen
in protoplasm :
I1 est certain que toutes les reclierches sur la distribution intracellulaire du glycoghne qui ont Qti:effectu6es au moyen de miithodes
comportant une fixation en phase liquide sont pratiquement sans
valeur, a cause de l'importance des artefacts de fixation. Ni 1'6tat
sous lequel le glycog&ne se trouve dans la cellule (&at granulaire
ou diffus), ni sa repartition intracellulaire ne peuvent 6tre apprgcies
correctement au moyen de telles mbthodes.
I n earlier works in which liver was prepared by freezing and
drying (Bensley and Gersh, '33 ; Mancini, '48)' the formation
of ice crystals may have caused a compression of the protoplasm between the glycogen granules resulting in such a
redistribution of these fine particles as to give an appearance
of homogeneity with the light microscope. The homogeneous
distribution of intracellular glycogen described by M. R. Lewis
( '21)' W. H. Lewis ( '26), and Nordmann ( '29) may possibly
be attributed to poor resolution or the effects of fixation by
iodine vapors. On the other hand, it is possible that under
some conditions second order particles may predominate, in
which case the glycogen would appear homogeneous with
the light microscope.
The second order particles probably correspond t o those
isolated by differential centrifugation by Lazarow ( '42) and
described in highly purified form by Claude ( '54). Lazarow
reported that the submicroscopic particles contained 0.16%
fat and 1 to 1.4% protein. Claude reported that the protein
content was reduced to 0.03% and presented an electron micro-
graph of shadowed glycogen granules. Measurements taken
from Claude's picture show that the diameter of the particles ranges from 500A to 1,500B. Such particles may be
assumed to have been subjected to some abrasion in the
course of preparation. It is clear that the particles described
above must represent aggregates of smaller units. Molecular weight determinations for relatively undegraded glycogen
have been reported to vary between 2 X los and 6 X lo6
(Greenwood, '52). Such values are too small to characterize
a 500 A glycogen particle. The diameter of the glycogen molecule has been estimated to be about 250
(Meyer, '43) and
from 150 A to 300 (Husemann and Ruska, '40). The results
presented in this report show smaller particles to be present
within the second order particles. It is probable that the
highly purified glycogen particulates of Claude represent aggregates of molecules.
It has not been possible to demonstrate photographically
the intimate relation of the second order granules to cytoplasmic structure. From a consideration of the size of the
granules, it would seem that they are located within submicroscopic vacuoles. Such vacuoles may be stained by
numerous metal-containing compounds and are of the same
order of size as the granules (Gersh et nl., '57). This is particularly clear when the frozen-dried liver of a carbohydratefed animal is stained with platinum tetrabomide.
The first order particles probably represent the central or
denser part of the glycogen molecule. They appear as dark
granules about 130 A in a less dense background which appears
to be homogeneous. Recent studies from the laboratory of
G. T. Cori and C. H. Cori have shown that the glycogen molecule consists of a more dense interior and a less dense periphery (Illingworth et al., '52; Larner et al., '52). Stetten
and Stetten ('54) have reported that the peripheral layers
of rat liver glycogen are metabolically far more active than
the interior. Their work is in support of the suggested structure of glycogen, whose peripheral parts are said t o consist
of branched chains of glucose residues which are more spread
out or separated than in the central part.
Hydroxyl groups of 1, 2 glycol groupings such as occur
on carbon atoms 2 and 3 of the glucose residues in glycogen,
are known to be selectively oxidized to the aldehyde stage by
periodic acid. It has been shown by Jackson and Hudson
('37) that one molecular equivalent of periodic acid is reduced per glucose molecule when corn starch is treated with
a 0.58 M aqueous periodic acid; this amount is that which is
required to oxidize the starch to the dialdehyde stage. It is,
therefore, evident that in an excess of periodic acid, the
number of 1, 2 glycol groups converted to aldehyde is directly
related to the number of glucose molecules present in the
starch particle. These aldehyde groups are capable of reacting with fuchsin sulfite t o produce a red color. It is probable
that the same mechanisms of oxidation and staining apply to
the glycogen molecule (Hotchkiss, '48 ; Lillie, '51).
Assuming that all the residues react (or that the proportion of rcactive residues is the same in the central and peripheral parts of the glycogen molecule) and that none are
lost (or that the periphcral loss is proportional to that of
the interior), one might find the denser interior of the stained
glycogen molecule to be darker in the electron microscope
than the peripheral portion, as the concentration of fuchsin
sulfite molecules added in the former site should be greater.
In accordance with this increased density, electron scattering in the central portion would be correspondingly greater.
According to this interpretation the dense and measurable
portion of the first order granule would represent the central
part of the glycogen molecule, while the less dense material
surrounding the central portion would represent the peripheral region of the glycogen molecule. It is possible also that
the process of staining resulted in some hydrolysis. If the
products of this rcaction were sufficiently large to diffuse
away only slowly (prior to being precipitated), they would
tend to increase the homogeneous background.
Liver cells of the guinea pig were fixed by freezing and
drying in such a way as to reduce or abolish ice crystal artifacts. Glycogen was stained in situ by the periodic acidleucofuchsin method and the distribution of glycogen was
studied with the light microscope and the electron microscope.
Glycogen occurs in particulate form. The central portion
of the first order particle is about 130 in diameter; it is
surrounded by a less dense portion which is not measurable.
I t is suggested that the central portion of these first order
particles corresponds to the denser, central portion of the
glycogen molecule. The first order particles are aggregated
into second order glycogen particles of about 600 ik to 1,500 ik.
These probably correspond to the submicroscopic particulates of Lazarow and Claude. It seems probable that the
second order particles are contained in submicroscopic cytoplasmic vacuoles, whose walls are stainable by other methods.
The second order particles are frequently aggregated to form
larger, third order particles which are visible with the light
I wish to thank Dr. Isidore Gersh and Dr. Irvin Isenberg
for their aid and advice.
ATERYAN,K. 1952 Some local factors in the restoration of the r a t liver after
partial hepatectomy. I. Glycogen; the Golgi apparatus ; sinusoidal cells ;
the basement membrane of the sinusoids. Arch. Path., 53: 197-208.
BENSLEY, R. R., AND I. GERSH 1933 Studies on cell structure by the freezingdrying method. I. Introduction. Anat. Rec., 57 .- 205-216.
CLAUDE,A. 1954 Cell morphology and the organization of enzymatic systems
in cytoplasm. Proc. Roy. Soc. London, B., 142: 177-186.
CORI, G. T., C. F. CORIAND G. SCHMIDT 1939 The role of glucose-1-phosphate
in the formation of blood sugar and synthesis of glycogen i n the liver.
J. Biol. Chem., 169: 629-640.
AND W. BONDAREFF1957 Submicroscopic structure of frozen-dried liver specifically stained for electron
microscopy. Part I. Technical. Anat. Rec., If!&?.- 91-112.
C. T. 1952 The size and shape of some polysaccharide molecules.
Advances in Carbohydrate Chemistry, 7 : 289-332.
R. D. 1948 A microchemical reaction resulting i n the staining of
polysaccharide structures in fixed tissue preparations. Arch. Biochem.,
16: 131-142.
E., AND H. RUSKA 1940 Versuche zur Sichtbarmachung von Glykogen-molekiilen. J. Prakt. Chem., 256: 1-10.
AND G. T. CORI 1952 Structure of glycogens and
amylopectins. 1. Enzymatic determinations of chain length. J. Biol.
Chem., 199: 631-640.
I. 1956 The use of organic dyes in electron microscopy. J. Histochem. Cytochem., 4: 416-418.
E. L., AND C. S. HUDSON1937 Application of the cleavage type of
oxidation by periodic acid to starch and cellulose. J. Am. Chem. SOC.,
69: 2049-2050.
G. T. CORI AND C. F. CORI 1952 Structure of
glycogens and amylopectins. 111. Analysis by stepwise enzymatic
degradation. J. Eiol. Chem., 199: 641-652.
M. C. 1936 a b e r histologische Veriinderungcn in Leber und Herz der hungernden Ratte. Arch. exp. Pathol.
Pharmakol., 180: 670-717.
A. 1942 Particulate glycogen : A submicroscopic component of the
guinea pig liver cell; its significance i n glycogen storage and the
regulation of blood sugar. Anat. Rec., 84: 31-50.
LEWIS,M. R. 1921 The presence of glycogen in the cells of embryos of Fundulus
heteroelitus studied i n tissue culture. Biol. Bull., 41 : 241-247.
LEWIS, W. H. 1926 Cultivation of embryonic heart muscle. Carnegie Institution Washington, Contrib. to Embryol., 90: 1-22.
LILLIE,R. D. 1951 Histochemical comparison of the Casella, Bauer, and periodic
acid oxidation-Schiff leucofuchsin technics. Stain Technol., 26 : 123136.
LISON, L. 1953 Histoehimie et Cytochimie Animales : Principes e t m6thodes.
Dcuxieme edition. Gxuthier-Villars, Paris.
MAKCINI,R. E. 1948 Histochemical study of glycogen in tissues. Anat. Rec.,
101 : 149-160.
MEYER,K. H. 1943 The chemistry of glycogen, Advances in Enzymology, 3:
C., AND R. W. MOWRY 1951 Demonstration of glycogen in the human
liver by the electron microscope. Proc. SOC.Exp. Biol. and Med., 76:
AND M. SWERDLOW1949 New sectioning techniques
of light and electron microscopy. Science, 110: 66-68.
M. 1929 Wachstum und Stoffwechsel der Leberzellen in der Gewebskultur. Arch. exp. Zellforsch., 8: 371-414.
PFUHL,W. 1932 Die Leber. Randbuch der mikroskopischen Anatomie des
Menschen (von Mollendorff). Julius Springer, Berlin. Vol. v, Part 2,
pp. 296-603.
M. R., AND DEWITTSTEWEN,JR. 1954 A study of the nature of
glycogen in the intact animal. J. Biol. Chem., ,907: 331-340.
J. L. 1954 Caution in the use of liquid propane for freezing biological specimens. Nature, 174: 235.
1 Photomicrograph of section of untreated liver from a normal, fed animal,
stained en bloc b y the periodic acid-leucofuchsin method. Numerous, deeply
stained, glycogen granules, about 1to 2 p in diameter, are distributed throughout the cytoplasm. Nuclei are unstaine(1. x 2700.
Photomicrograph of section of liver from a normal, fed animal treated with
fifth molar phosphate buffer (pH 7.0) a t 37" C. and subsequently stained
b y the periodic acid-leucofuclrsin method. Somewhat swollen glycogen particles appear essentially the same as i n untreated tissue. x 2700.
Photomicrograph of section of liver from n normal, fcd animal treated with
a 50% solution of saliva in fifth molar phosphate buffer (pH 7.0) a t 37°C.
and subsequently stained by t h e periodic arid-lcurofurhdn method. No glycogen is visible. X 2700.
4 Electron micrograph of section of liver from a normal, fed animal stained
r n bloc by the periodic acid-leucofuchsin method. Third order glycyogen
particles, 3,000 A to 9,000 A, appear as small dnrk granules, distributed iii
the cytoplasm. X 5000.
Electron micrograph of a section of liver from a normal, f e d animal treated
with a 50% solution of saliva in fifth niolnr phosphate buffer (pH 7.0) a t
37°C. and subsequently stained by the periodic acid-leucofuchsin mcthod.
h o glycogen particles are visible. X 25,000.
Electron micrograph of a section of liver from a normal, fed animal treated
with fifth molar phosphate buffer ( p H 7.0) a t 37°C. Somewhat swollen glycogen particles appear essentially the same as in untreated tissue. x 5000.
Electron micrograph of highrr initial magnification of portion of the same
section of figure 4. Larger masses represent third order particles, within
which smaller, secoud order particles (GOO ti to 1,500 ii) are visible. x 25,000.
Second order glycogen particles appear as large diffuse masses with darkrr
centers a t a higher initial magnification of a portion of the same section as
figure 4. Within these second order particles minute first order particles are
visible. The central part of the first order particle, about 130 A in diameter,
appears as a black granule around which is a diffuse peripheral arm. Thc
central parts of some of the first order particles are marked by arrows.
X 75,000.
High resolution electron micrograph, of high initial magnification, of a
section of liver from a normal, fed animal stained by the periodic acid-leucofuchsin method. Third order particles appear as minute dense areas, between
66 A and 96 A in diameter. X 150,000.
Без категории
Размер файла
743 Кб
morphology, block, selective, pig, electro, live, particular, guinea, freezing, microscopy, revealed, drying, staining, glycogen
Пожаловаться на содержимое документа