close

Вход

Забыли?

вход по аккаунту

?

The paranucleolar structure accessory body of cajal sex chromatin and related structures in nuclei of rat trigeminal neuronsA cytochemical and ultrastructural study.

код для вставкиСкачать
The Paranucleolar Structure, Accessory Body of
Cajal, Sex Chromatin, and Related Structures
in Nuclei of Rat Trigeminal Neurons: A
Cytochemical and Ultrastructural Study ’
JAMES H. HARDIN: SAMUEL S. SPICER AND WILLIAM B. GREENE
lnstitute of Pathobiology and Department of Pathology, Medical College
of South Carolina, Charleston, South Carolina 29401
ABSTRACT
Comparative light and electron microscopic study of nuclei in rat
trigeminal neurons identified two structures which are the ultrastructural equivalent
of the paranucloelar structure and the accessory body of Cajal, two nucleoplasmic
structures previously demonstrated in other neurons by light microscope silver staining methods. Ultrastructural evidence indicates that the dense component of the
nucleolus is converted into the paranucleolar structure, which then detaches from
the nucleolar surface to lie free in the nucleoplasm as the accessory body of Cajal.
The cytochemistry, ultrastructure, and antimonate reactivity of the paranucleolar
structure and accessory body were identical. Both structures lacked cytochemically
demonstrable DNA, RNA, or basic protein.
The neuronal nuclei also contained Feulgen-positive sex chromatin bodies that
adhered to the nucleolus, the nuclear membrane, or to both of these structures in
specimens of female but not male rats. The ultrastructure and antimonate reactivity
of these bodies closely resembled that of heterochromatin clumps but differed markedly
from that of the paranucleolar structures and accessory bodies.
Additional structures characterized ultrastructurally included patches, granular
bodies, and flakes. These structures, like the paranucleolar structure and the accessory
body of Cajal, are apparently unique to nuclei of neurons. Cytochemical methods
showed that the patches contained basic protein but no nucleic acid.
Light microscope silver staining methods have visualized structures of undetermined significance in the nucleoplasm of
neurons. One of these, “the accessory body
of Cajal,” is an argyrophilic, Feulgennegative, nonbasophilic structure. These
round bodies, measuring about 1 LI in diameter, are located between the nucleolus and
nuclear membrane (Ramon y Cajal, ’03;
Lindsay and Barr, ’55; Haggar, ’57; Thompson, Haggar and Ban, ’57). Another
argyrophilic, Feulgen-negative, nonbasophilic body in neuronal nuclei is “the paranucleolar structure,” which differs from
the accessory body of Cajal by its attachment to the nucleolus and its hemispherical or concavoconvex shape (Haggar, ’57;
Thompson, Haggar and Barr, ’57). Although it has been speculated that the
paranucleolar structure may derive from
the nucleolus and be converted to the accessory body of Cajal (Haggar, ’57), direct
evidence is lacking for such interrelationships.
ANAT. REC., 164: 403-432.
An ultrastructural counterpart of the
accessory body of Cajal or the paranucleolar structure has not been determined.
The present study describes cytochemical
and ultrastructural features of two types
of nuclear bodies in neurons of rat trigeminal ganglia. Evidence is presented
that one of these bodies corresponds to the
accessory body of Cajal and the other to
the paranucleolar structure. Observations
were also made on structures at the nucleolar surface and free in the nucleoplasm that might be confused with the
paranucleolar structure and the accessory
body of Cajal. In addition, this investigation undertakes the morphologic characterization of these and other nuclear structures as a step in the direction of delineating the structures of the nucleus to which
biochemical activities may eventually be
localized.
Received Dee. 20, ’68. Accepted Mar. 14, ‘69.
Isupported by grants 10956 and 11028 from National Institutes of Health and the National Cystic
Fibrosis Research Foundation.
2 PHS Postdoctoral Fellow.
403
404
J. H. HARDIN, S. S. SPICER AND W. B. GREENE
In addition to conventional fixation procedures for electron microscopy, the
Komnick method of fixation (Komnick,
'62), which supposedly localizes tissue sodium at the ultrastructural level, was employed in these studies. This method was
included because it produces characteristic
electron-opaque deposits in the heterochromatin (Spicer, Hardin and Greene,
'68) and in various nucleolar components
(Hardin, Spicer and Malanos, '68; Hardin
and Spicer, '69). By providing identifying characteristics for nuclear structures,
the Komnick method of fixation yields results that supplement the morphologic appearance and indicate possible relationships between structures in the nucleolus
and nucleoplasm. This method of fixation
was especially valuable in the ultrastructural identification of the sex chromatin
body. A brief report of this study has appeared previously (Hardin, Spicer and
Greene, '67).
METHODS
Trigeminal ganglia from anesthetized,
adult, albino rats were quickly removed
and fixed at 4°C by one of several procedures for electron microscopy. For cytological study, specimens were fixed for one
hour in 2% osmium tetroxide adjusted to
a pH of 7.4 with either a 0.1 M phosphate
buffer or with Millonig's buffer (Millonig,
'61); other specimens were fixed for one
hour in cacodylate buffered, 6.25% glutaraldehyde followed by one hour in collidine
buffered, 2 % osmium tetroxide. Additional
specimens were fixed one hour in Komnick's potassium pyroantimonate (antimonate) and osmium tetroxide solution
(Komnick, '62) modified as previously described (Spicer, Hardin, and Greene, '68).
As a means of determining the difference
between the inherent density of the tissue
components and the density due to antimonate deposits, unstained sections were
examined from specimens fixed in a solution similar to Komnicks solution except
for containing equimolar pyrophosphate in
place of pyroantimonate anion.
After each fixation procedure, the tissue
blocks were dehydrated with graded alcohols and inatrated and embedded in
maraglas. Ultrathin sections, with or with-
out prior lead citrate or uranyl acetate-lead
citrate staining, were examined in an AEI6B electron microscope. One-half micron
sections of maraglas-embedded specimens
were stained with a hot aqueous solution of
0.5% toluidine blue and 1% sodium borate
for examination by light microscopy. Although possessing no histochemical significance, this method of staining is widely
used for light microscopic examination of
epoxy sections. Examination of 20 such
sections cut serially permitted three dimensional reconstruction of a 10 u thick layer
in the ganglion.
For light microscope cytochemistry,
specimens were fixed at room temperature
for 24 hours in a 2% calcium acetate, 10%
formalin solution or in Bouin's or Carnoy's
fluid. Paraffin sections of all of these specimens were stained with the Feulgen
method for localizing deoxyribonucleic
acid (DNA). Sections of formalin or Bouinfixed ganglia were stained with a 0.02%
solution of azure A at pH 3.5-4.5. This
procedure specifically distinguishes blue
stained DNA from violet stained ribonucleic acid (RNA) in sites lacking acid
mucosubstances (Spicer, '61). Sections of
Bouin fixed ganglia were stained with a
direct Schiff-methylene blue sequence. Alternatively, sections of fonnalin-fixed specimens were hydrolyzed 60 minutes at 60°C
in Bouin's fluid prior to sequential staining
with Schiff reagent and methylene blue.
These two latter techniques, involving
treatment with Bouin's fluid in block or in
section, accomplish Feulgen-type hydrolysis of DNA and thus permit specific S c h 8
staining of DNA without extraction of the
RNA (Spicer, '61). RNA is then colored
blue with methylene blue as distinct from
the magenta DNA in t h i s sequence. Sections of ganglia fixed with Carnoy's fluid
were stained with Biebrich scarlet at pH
9.5 for localization of strongly cationic
protein (Spicer, '62). In this method for
basic protein, specimens were fixed with
formalin free solutions to avoid aldehyde
blockage of the protein amines. The reduced silver nitrate method utilized by
Thompson, Haggar and Barr ('57) was
used to visualize argyrophilic structures in
specimens fixed with buffered formalin.
NUCLEOPLASMIC STRUCTURES OF TRIGEMINAL NEURONS
Specimens from both male and female
rats were examined by all of the methods
employed. The results described below
apply for neurons of trigeminal ganglia
from both sexes except in regard to the
sex chromatin as noted.
405
and sensory neurons (Haggar, ’57; Thompson, Haggar and Barr, ’57) were on occasion heavily (fig. 1 ) but often lightly (fig.
2) stained with the reduced silver method.
A few nucleoli possessed two paranucleolar structures (fig. 2). However, as has
been
the experience of other investigators
LIGHT MICROSCOPIC OBSERVATIONS
(Lindsay and Barr,’55;Thompson, Haggar
As revealed by three dimensional recon- and Ban, ’57), the silver technique gave
struction from the serial half micron mara- capricious results, and in some specimens
glas sections of trigeminal ganglia stained it completely failed to reveal paranucleolar
with toluidine blue, considerable variation structures or visualized only a few of these
existed in the size of the neuron cell bodies bodies.
and their nuclei and in the size, number,
Numerous paranucleolar structures were
and location of nucleoli in each nucleus. consistently observed in 0.5 p thick, epoxy
The largest type of neuron measured up to sections stained with toluidine blue (fig.
about 55 by 45 p in longest and shortest 3 ) . They measured up to approximately
diameter and possessed the largest nucleus 2 p in width and 1 ~1 in height and exwhich measured up to 19 by 15 p. These hibited a nearly hemispherical contour
large neurons usually contained a single, with a convex outer border and a flat or
intensely stained, nearly spherical nucleo- slightly concave border attached to the
lus which was very large, measuring up to nucleolus. Paranucleolar structures, pres6 ~1 in diameter. These large nucleoli lay ent in neurons of both sexes, were espeat or near the center of the nucleus and cially prominent and distinct on the large,
usually occurred alone but on occasion centrally located, neuronal nucleoli but
were accompanied by a second, small, ec- were rarely observed on the usually smaller
centrically located nucleolus.
nucleoli contacting the nuclear membrane.
Smaller neurons measuring up to about
Paranucleolar structures lacked baso35 X 30 p in longest and shortest diameter philia and Feulgen reactivity. They failed
possessed smaller nuclei which measured to react with selective stains for RNA inup to approximately 14 by 12 p, These cluding azure A at pH 4.0 and methylene
smaller neurons usually revealed two or blue in the Schiff-methylene blue sequence.
three smaller nucleoli which measured up Lightly stained bodies suggestive of parato 4.0 ~1 but rarely exceeded 3.5 p in great- nucleolar structures were observed in secest dimension. In nuclei with two nucleoli, tions of ganglia stained with pH 9.5
one was invariably located near or at the Biebrich scarlet method. However, such
nuclear membrane and the second was acidophilic structures were infrequent and
situated more often peripherally or at the could not be identified positively as paramargin but sometimes at or near the center nucleolar structures because of other
of the nucleus. Nucleoli as a rule lay Biebrich scarlet positive material near the
widely separated from one another. In nucleolus.
nuclei with three nucleoli, they usually
Sex chromatin and clumps of heteroadhered to the nuclear margin fully re- chromatin. Paraffin sections of trigeminal
moved from one another. The large neu- ganglia stained with methods for DNA rerons generally corresponded with “light vealed clumps of deeply stained material
cells” and the smaller neurons with “dark in the nucleoplasm of many neurons. Large
cells” (Andres, ’61; Dixon, ’63). The nu- clumps attached to the nucleolar surface
clear structures described herein were ob- often possessed a planoconvex outline and,
served in both cell types.
therefore, could be mistaken for paranuParanucleolar structures.
Structures cleolar structures. Because these clumps
that protruded from the nucleolar surface were Feulgen positive and were present in
and were similar in shape, size, and stain- specimens of female but not male rats, they
ing properties to paranucleolar structures were identified as sex chromatin bodies
previously described in large primary motor (fig. 4). Sex chromatin bodies were also
406
J. H. HARDIN, S. S. SPICER AND W. B. GREENE
attached to the nuclear membrane (fig. 5 )
or to both the surface of a nucleolus and
the nuclear membrane (fig. 6). Nuclear
profiles never exhibited more than one sex
chromatin body. Feulgen reactive clumps
smaller than sex chromatin bodies or paranucleolar structures and present in specimens of male and female rats were considered nucleolar associated heterochromatin. Except for sex chromatin bodies
and the clumps of heterochromatin at the
nucleolar surface and nuclear membrane,
the nuclei of trigeminal neurons of both
sexes exhibited no large Feulgen reactive
particles. Instead, the neuronal nuclei revealed small, lightly stained, widely scattered particles which contrasted with the
large, intensely stained, crowded particles
of heterochromatin in nuclei of neighboring satellite (fig. 6 ) , Schwann, and endothelial cells. Such staining correlates precisely with the ultrastructural distribution
of heterochromatin in these cell types
(Spicer, Hardin and Greene, ’68).
Sex chromatin bodies also stained with
the azure A method. The light blue coloration of these sex chromatin bodies resembled that of the heterochromatin in nuclei
of neighboring cells and, like Feulgen reactivity, indicated the presence of DNA in
the bodies. This staining of sex chromatin contrasted with the violet metachromasia of neuronal nucleoli and cytoplasmic Nissl substance. The blue staining
with azure A is characteristic of DNA in
sites lacking acid mucosubstance, because
RNA stains metachromatically and acidic
proteins stain only at higher pH levels.
This azurophilia indicative of chromatin
thus differentiated the sex chromatin bodies from paranucleolar structures.
Sex chromatin bodies were also visible
in epoxy sections stained with toluidine
blue. In paraffin sections of ganglia fixed
with Carnoy’s fluid, structures suggestive
of sex chromatin bodies were stained with
alkaline Biebrich scarlet.
Accessory bodies of Cajal. Roughly
spherical, uniformly stained bodies that
measured up to approximately 1.5 p in
diameter and were similar to accessory
bodies of Cajal previously described in
other neurons (Lindsay and Ban, ’55;
Haggar, ’57; Thompson, Haggar and Barr,
’57) were visualized with the reduced silver staining method (fig. 7) in the nucleoplasm of rat trigeminal neurons. The silver method was inconsistent and in some
instances did not visualize the Cajal bodies.
However, the technique often revealed one
and occasionally two darkly stained bodies
that lay near or at varying distances from
the nucleolus in trigeminal neurons of
both sexes.
Structures apparently identical with the
accessory bodies were frequently observed
in the thick epoxy sections stained with
hot, alkaline toluidine blue (figs. 8, 9).
Examination of serial epoxy sections
stained with toluidine blue established that
these bodies lay free in the nucleoplasm
and were not attached to a nucleolus out
of the plane of the section.
In agreement with previous findings
(Lindsay and Ban,’55; Thompson, Haggar
and Barr, ’57), accessory bodies were
Feulgen-negative and were not basophilic.
Whether these structures were stained by
the alkaline Biebrich scarlet procedure remained undecided because of the difficulty
of differentiating them from other acidophilic material in the nucleoplasm including the patches described below.
Patches. Distinctive, irregular areas in
the nucleoplasm, measuring up to 5 p in
greatest dimension, were designated
“patches.” In epoxy sections, patches were
moderately stained with toluidine blue (fig.
10). They were not evident in paraffin
sections stained with azure A or Feulgen
methods but were visualized in Carnoy
fixed specimens with the pH 9.5 Biebrich
scarlet technique (fig. 11). An average of
about four patches lay widely distributed
in a single nuclear profile.
Electron microscopic observations
Profiles of nucleoli located free in the
nucleoplasm generally appeared round or
slightly elongated, whereas profiles of nucleoli at the periphery of the nucleus usually appeared flattened against the nuclear
membrane.
Nucleoli of rat trigeminal neurons, as
described elsewhere (Hardin and Spicer,
’69), consisted of four ultrastructurally
distinct components, including granular
component, dense (fibrillar) component,
NUCLEOPLASMIC STRUCTURES O F TRIGEMINAL NEURONS
vacuoles, and pars amorpha. The discrete
foci of pars amorpha observed in these
nucleoli closely resembled areas denoted as
pars amorpha in nucleoli of spermatogonia
and Sertoli cells (Fawcett, '66). This use
of the term clearly differs from Estable
and Sotelo's ('51) definition of pars
amorpha as the diffuse matrix of the
nucleolus.
The nucleolar surface varied to some
degree but consisted principally of large
areas of granular component and widely
separated small areas of dense component. Granular component comprised
about 60 to 95% and dense component
about 5 to 40% of the total nucleolar surface area. In about one-fourth of the nucleolar profiles, one or more small foci of
pars amorpha constituted less than 5%
of the nucleolar surface area.
Paranucleolar structures. Two types of
protrusions from the nucleolar surf ace
were seen that codd be considered the
ultrastructural equivalent of the paranucleolar structure evident at the light microscope level. One type of protrusion consisted solely of the dense component of
the nucleolus (figs. 12, 13). In about onethird of the nucleolar profiles, one or more
of the areas composed of dense component protruded from the surface of the
nucleolus. However, these protrusions were
relatively small as a rule. The few dense
protrusions that measured in the range of
the paranucleolar structures (figs. 12, 13)
were too infrequent to represent the ultrastructural counterpart of the paranucleolar
structure. The outermost portion of a few
of these protrusions of dense component
consisted of coarse granules (fig. 12).
Areas of the nucleolar surface composed
of the granular component or of pars
amorpha rarely formed protrusions
The shape, size, position, prevalence, and
distribution of the second type of protrusion served to identify it as the ultrastructural equivalent of the paranucleolar structure visible with the light microscope (figs.
13-23). Profiles of these structures exhibited a roughly hemispherical contour
with a convex outer border and a flat or
concave border of attachment to the nucleolar surface. Paranucleolar structures
varied in size, depending in part at least
407
on the plane of section, and measured up
to 1.8 cr in width at the plane of attachment to the nucleolus and 1.4 I.I in height.
A small percentage of nucleolar profiles
disclosed one of these structures and a
moderate number exhibited two (fig. 22)
which were about 20 to 180" from one
another at the nucleolar surface.
A definite relationship seemed to exist
between the prevalence of paranucleolar
structures and the position of the nucleolus
in the nucleus. Paranucleolar structures
were encountered much less frequently on
nucleoli that contacted the thin layer of
heterochromatin lining the nuclear envelope than on those nucleoli free in the
nucleoplasm. The paranucleolar structures on nucleoli at the nuclear margin
were relatively small and adhered to the
nucleolar surface at a point 90 to 180"
from the point of attachment of the nucleolus to the nuclear envelope. Nucleolar
profiles at the nuclear margin with more
than one paranucleolar strucure were not
observed, whereas about 10% of the nucleolar profiles free in the nucleoplasm
possessed two paranucleolar structures (fig.
22).
Paranucleolar structures revealed areas
of dense material that averaged about 70
mu through their smallest dimension (fig.
19). The profile of these areas vaned,
some showing a rounded contour and
others an irregular elongated contour.
Serial sections would be required to determine whether these areas constitute a
coiled thread as suggested by some profiles.
These dense areas resembled the dense
component of the nucleolus in their texture
and density (figs. 13-15). The dense areas
were partially to completely surrounded by
canalicular spaces with low density (fig.
19). Areas of intermediate density were
interspersed throughout the paranucleolar
structures, contacting both high and low
density areas.
The paranucleolar structure most often
protruded from a portion of the nucleolar
surface that consisted solely of dense component (figs. 13-17). In profile, the paranucleolar structure exactly overlay this
base of dense component; i.e., the lateral
borders of the paranucleolar structure
408
J. H. HARDIN, S. S. SPICER AND W. B. GREENE
fitted together precisely with the lateral
borders of the base of dense component.
A sharp interface provided the area of
contact between the opposing borders of
the paranucleolar structure and the solid
base of dense component. This interface
sometimes lay slightly above (fig. 14), but
usually fell even with (fig. 15) the level
of the bordering nucleolar surface composed of granular element. In many profiles, the most superficial part of the base
of dense component formed small projections at this interface (figs. 13, 14, 19).
Some of these projections were continuous with the dense areas of the paranucleolar structure, and all of such projections closely resembled these dense areas in
shape, structure, and density.
A few paranucleolar structures protruded from a discontinuous base of dense
component that was interspersed with
areas of granular component (figs. 18,
19). These areas of granular component
did not form projections toward the paranucleolar structure. However, even where
scant or interrupted, the base of dense
component with its outward projections
appeared to provide continuity with the
dense areas of the paranucleolar structure
(figs. 19, 20). Paranucleolar structures
close to an area of the nucleolar surface
consisting solely of granular component
usually appeared not to be attached to the
granular component, but rather to be
slightly removed from the nucleolus (figs.
21, 30).
The large base of dense component underlying the paranucleolar structure differed from dense component elsewhere in the
nucleolus only in lacking enclosed foci of
pars amorpha and vacuoles. The base of
dense component in some instances, however, bordered or partially enclosed a subjacent large focus of para amorpha (fig.
23).
Sex chromatin bodies and clumps of
heterochromatin.
Structures that were
considered the ultrastructural equivalent
of sex chromatin bodies seen with the light
microscope were observed in nuclear profiles from specimens of female but not
male rats. The percentage of nuclear profiles possessing sex chromatin bodies was,
of course, exceedingly small, due to the
small chance of the section passing through
such a relatively small structure. Confirming the light microscope observations of
Feulgen stained paraffn sections, sex
chromatin bodies were attached to the nucleolar surface alone (fig. 24), to the nuclear membrane alone (figs. 25, 26), or to
both sites (figs. 27-29). Those sex chromatin bodies which were attached to both
the nuclear membrane and the nucleolus
either lay beside a nucleolus in contact
with the nuclear envelope (fig. 27) or lay
interposed between the nuclear membrane and a nucleolus free in the nucleoplasm (figs. 28, 29). In those instances in
which nucleoli with attached sex chromatin contacted the nuclear envelope, the
sex chromatin itself always contacted the
nuclear envelope (fig. 27). The width of
the sex chromatin bodies at its point of
attachment to either the nucleolar surface
or the nuclear membrane measured up to
2.4 p. The height of these structures at
either location measured up to 1.7 CI.Some
sex chromatin bodies varied from the usual
roughly hemispherical shape, probably in
part as a result of the plane of section.
In addition to their differences in sex
distribution, several features served to distinguish those sex chromatin bodies attached to the nucleolar surface from paranucleolar structures. Sex chromatin bodies
were generally larger than paranucleolar
structures. Nucleolar profiles sometimes
exhibited two paranucleolar structures (fig.
22) but never showed more than one sex
chromatin body. Sex chromatin bodies always adherred fully to the nucleolus (fig.
24), showing no stages of separation as
did paranucleolar structures in their apparent transition to accessory bodies. Most
sex chromatin bodies contacted almost
exclusively that part of the nucleolar surface consisting of the granular component
(fig. 24); they seldom, if ever, made contact with the dense component at the
nucleolar surface as did the paranucleolar
structures. All sex chromatin bodies consisted of heterochromatin-like material
(figs. 24-28), which clearly differed from
the material in paranucleolar structures.
Some sex chromatin bodies contained occasional dense granules that varied from
35 to 55 mu in diameter and were each en-
NUCLEOPLASMIC STRUCTURES OF TRIGEMINAL NEURONS
closed by a thin halo of low density material (fig. 25). These granules were usually
well separated from one another and lay
at the border of the body or at boundaries
between heterochromatin and low density
foci within the body. Clumps of heterochromatin, smaller than sex chromatin
bodies, located at the nucleolar surface
and the nuclear membrane (fig. 25), and
ill-defined clumps of presumed heterochromatin scattered throughout the nucleoplasm usually contained similar dense
granules which were more numerous and
more closely aggregated than were those in
sex chromatin bodies.
Accessory bodies o f Cajal. Nucleoplasmic structures interpreted as accessory
bodies of Cajal (figs. 30-33) resembled
fully developed paranucleolar structures
morphologically (fig. 22; Cf. figs. 19, 31),
except for their separation from the nucleolus and their contour which was usually round to elliptical but occasionally
included one or more flattened or
concave surfaces. These structures corresponded in shape, size, position, prevalence,
and distribution with the argyrophilic accessory bodies of Cajal seen with the light
microscope. These frequently encountered
bodies varied in size depending in part at
least on plane of section. The somewhat
elongated bodies measured up to 1.5 CI in
the shortest and 2.2 p in the longest dimension. A few accessory bodies were partially
surrounded by a rim of low density nucleoplasm that averaged about 0.3 u in
thickness. Accessory bodies lay in the nucleoplasm at varying distances from the
nucleolus but rarely near and never in
contact with marginated heterochromatin
at the nuclear envelope. These bodies have
been encountered in a single nuclear profile containing two nucleoli. Two accessory bodies have been observed in a single
nuclear profile.
A few accessory bodies contacted
roughly rounded areas which fit the surface of the accessory body as though both
were integral parts of a single structure
(fig. 31). These areas were composed of a
flocculent material with a few indistinct
granules.
Another variant of the accessory body
(figs. 32, 33) was observed infrequently
409
in which the body partially or completely
enclosed a roughly spherical 0.3 to 0.6 CI
focus of material which resembled dense
component of the nucleolus in texture and
density and antimonate reactivity. On the
basis of their morphologic appearance,
these atypical accessory bodies apparently
could not represent tangentially sectioned
paranucleolar structures with a base of
dense component.
Granular bodies. The neuronal nucleoplasm contained another structure (figs.
34, 35) which consisted of aggregates of
dense granules and accordingly is referred
to as the “granular body.” The granular
body clearly differed from the accessory
body of Cajal in electron micrographs but
might be confused with this structure at
the light microscope level. Although it
was not proven that the granular bodies
were silver unreactive, these bodies, identified only by electron microscopy, were too
small and infrequent to account for a significant percentage of the silver reactive,
nucleoplasmic bodies observed by light microscopy. The granular bodies measured
up to 0.7 by 1.0 ~1 in shortest and longest
dimension. A partial or complete rim of
low density material which ranged from
0.1 to 0.5 p in thickness surrounded some
of the granular bodies (fig. 34). The individual granules measured about 20 mu
in diameter and were associated with a
moderately dense amorphous material and
low density areas; some granules were
aligned in rows. Granular bodies usually
had a round to elliptical profile but occasionally displayed one or more flattened
surface areas. In rare profiles, granular
bodies were observed close to or in contact
with the granular component of the nucleolus (fig. 35). One granular body was
enclosed by a membrane-like structure
(fig. 35). These bodies were not observed
contacting the nuclear membrane or marginated heterochromatin. They appeared
to be randomly distributed in the nucleoplasm but often closely bordered patches,
lying separated from these structures by a
low density area (fig. 34).
Patches. The average nuclear profile
exhibited one to four areas which were
designated “patches” (figs. 34, 36-38).
At low magnification (fig. 36), these
410
J. H. HARDIN, S. S. SPICER AND W. B. GREENE
rounded to irregular areas were identified
by their moderate density and lack of
chromatin clumps. The patches measured
up to about 5 p in greatest dimension.
Patches generally lay well separated from
one another about midway between the
center of the nucleus and its membrane;
they were seldom observed near and never
seen contacting the nuclear membrane.
Accessory bodies of Cajal were frequently
seen near one or more patches (fig. 38),
either in direct apposition or separated by
the zone of low density material that sometimes surrounds the accessory body.
At higher resolution patches were composed of punctate dense matter and associated less dense substance (figs. 34, 37,
38). The dense particulates sectioned at
an appropriate angle appeared to be arranged linearly in rows as slightly separated or contacting particles (fig. 37). Low
density material separated the rows which
occasionally lay in slightly curved parallel
arrays.
Flukes. Other ultrastructural components in the nucleoplasm of neurons included structures referred to descriptively
as “flakes” (fig. 38). Apparently depending on plane of section, the profile of the
flakes had an irregularly round or elongated contour and measured about 80 mu
in smallest dimension. They consisted of
moderately dense material with enclosed
dense granules. Some flakes lay well separated from one another, whereas others
were closely grouped or apparently linked
together; they were concentrated in some,
more than other areas of the nucleoplasm.
Although present elsewhere in the nucleoplasm, flakes appeared especially prominent in the patches (fig. 38). Flakes were
distinctive in this area where differentiation from heterochromatin posed no problem. Part or all of some patches lacked
flakes (fig. 37). Flakes were occasionally
seen in the region where patches adjoined
accessory bodies (fig. 38).
Pyroantimonate precipitates in the m u Tonal nucleus. Following fixation in
Komnick’s fluid, trigeminal neurons revealed intranuclear, electron-opaque precipitates in thin sections unstained with
heavy metals. Studies reported elsewhere
have described three different forms taken
by the antimonate deposits in particular
parts of the nucleolus (Hardin and Spicer,
’69). The granular component contained
fine deposits, the dense component moderately fine deposits and the foci of pars
amorpha coarse deposits (figs. 17, 20).
Prominent antimonate deposits were localized in the dense areas of the paranucleolar structure (figs. 17, 20). Profiles of
many of these areas were continuous with
one another and gave the appearance that
could be interpreted as a plane of section
through a coiled thread (fig. 17). Other
areas of the paranucleolar structure had
very light or no antimonate deposits. The
moderately fine deposits within the dense
areas of the paranucleolar structure resembled exactly those deposits in the dense
component of the nucleolus. The paranucleolar structures protruded from an area
of the nucleolar surface that contained
moderately fine deposits and can be identified as dense component. Projections from
the superficial area of this base of dense
component (fig. 17) corresponded precisely with the dense areas of the paranucleolar structure in regard to antimonate
reactivity - both entities possessed moderately fine antimonate deposits. At some
sites, these small projections of the base
of dense component were continuous with
the dense areas of the paranucleolar structure (fig. 17).
Antimonate deposits in the dense areas
of the accessory bodies of Cajal precisely
resembled those in the dense areas of the
paranucleolar structure. Accordingly, the
pattern of antimonate deposition was identical in these two bodies.
As reported previously (Spicer, Hardin
and Greene, ’68), the distribution of antimonate deposits in the neuronal nuclei of
trigeminal ganglia fixed by the pyroantimonate-osmium tetroxide method closely
corresponded with the distribution of heterochromatin in these nuclei of ganglia
fixed and stained by routine methods. However, some variability in the antimonate
reactivity of heterochromatin was observed
among different specimens. Thus, in specimens of both sexes, antimonate deposits
were concentrated into small clumps of
heterochromatin that were located near the
nuclear envelope, attached to the nucleolar
NUCLEOPLASMIC STRUCTURES OF TRIGEMINAL NEURONS
surface, or free in the nucleoplasm (fig.
26). In specimens of female but not male
rats, antimonate deposits were also concentrated in the large sex chromatin bodies (figs. 26, 29). The antimonate h a t i o n
method more clearly demarcated heterochromatin clumps in the nucleoplasm and
revealed more clearly than routine methods
that nucleolar associated heterochromatin
contacted only the portion of the nucleolar
surface composed of granular element.
The unstained thin sections of tissue
fixed with Komnick’s fluid also showed
several irregular areas with sparse pyroantimonate precipitates, which in most
specimens were much finer than those in
heterochromatin clumps but occasionally
were coarse. The areas with fine precipitates corresponded in size, distribution,
contour, and number with the patches
visualized in routine morphologic preparations. The fine antimonate precipitates in
the patches were distributed in a pattern
simulating the organization of these structures seen in the morphologic preparations.
Occasional areas in the nucleoplasm
that lacked pyroantimonate reactivity possibly corresponded with the granular bodies seen in the routine morphologic preparations.
DISCUSSION
On the basis of light microscopic observations with the reduced silver method
and cytochemical methods, neuronal nuclei of rat trigeminal ganglia possess structures that are identical to bodies designated as the paranucleolar structure and
the accessory body of Cajal in other neuronal nuclei (Lindsay and Ban, ’55; Haggar, ’57; Thompson, Haggar and Barr, ’57).
The ultrastructural counterpart of the paranucleolar structure and of the accessory
body of Cajd is identified in this study for
the fist time. The protuberances of the
nucleolar surface and the nucleolar satellite, both observed with phase contrast microscopy in neuronal nuclei of rat spinal
ganglia (Andres, ’61), most likely correspond to the paranucleolar structure and
the accessory body of Cajal, respectively.
The accessory body of Cajal is probably
identical to the group of dense granules
about 4 0 m ~in diameter seen in the nu-
41 1
cleoplasm adjacent to the nucleolus of rat
spinal ganglion cells “in vivo” (Hiraoka and
Van Breeman, ’63) and “in vitro” (Bunge,
Bunge, Peterson and Murray, ’67). The
nucleolar body located at one side of the
nucleolus in rat Purkinje cells (Herndon,
’63)appears to resemble the paranucleolar
structure. The relationship of the accessory
body to the aggregates of 20-30 mv dense
bodies in nuclei of the human trigeminal
neurons (Beaver, Moses and Ganote, ’65b)
and the karysosomes with 30 mw dense
granules in pea nuclei (Sankaranarayanan
and Hyde, ’65) remains undetermined.
An apparent gradation of observed configurations in trigeminal neuronal nuclei
can be arbitrarily arranged in a sequence
(figs. 12-22) strongly suggesting that the
accessory body of Cajal is derived from
the paranucleolar structure, which is itself formed from the dense component of
the nucleolus. This sequence of forms includes in order:
( 1 ) dense component within the nucleolus (fig. 12), (2) dense component at
or protruding from the nucleolar surface
(figs. 12,13), (3) paranucleolar structures
projecting into the nucleoplasm from a
base of dense component (figs. 14-18),
( 4 ) stages of separation of paranucleolar
structures from nucleoli (figs. 20,21), and
(5) accessory bodies of Cajal, free in the
nucleoplasm (figs. 22,30-32).
The idea that paranucleolar structures
are formed from the dense component of
the nucleolus is supported by considerable
evidence. First, the dense areas of the
paranucleolar structures resemble the
dense component in morphologic preparations (figs. 13-16). In pyroantimonateosmium tetroxide fixed specimens, these
two entities display a uniquely similar content of antimonate reactive substance
(figs. 17, 20). Moreover, the paranucleolar
structure comprises the upper part of an
apparent morphologic unit composed of
dense component in the deeper portion.
In such units, a distinct interface, which
might represent a metabolic front of conversion activity, separates the two halves.
Projections from the base of dense component at this interface provide continuity
with the dense areas in the paranucleolar
structure (figs. 13-15, 17, 20). If dense
412
J. H. HARDIN, S. S. SPICER AND W. B. GREENE
component is transformed into paranucleolar structures, configurations in which
these structures adhere to only a small
fragment of dense component at the nucleolar surface perhaps represent the last
phase of such transformation (figs. 20,
21), and after final conversion the paranucleolar structure breaks free of the nucleolus to become an accessory body of
Cajal (figs. 22,30-32). In those configurations where no dense component underlies
a paranucleolar structure, the latter appears to be slightly removed from the nucleolus as though perhaps in a final phase
of separation (figs. 21, 30).
Evidence favoring the thesis that paranucleolar structures give rise to accessory
bodies derives from the observation that
the paranucleolar structures mimic the accessory bodies morphologically (figs. 30,
31) and often appear in the process of detachment from the nucleolus. The similarity of antimonate precipitates in paranucleolar structures and accessory bodies
supports this concept.
Transformation proceeding in the opposite direction whereby accessory bodies
convey macromolecular substances to the
nucleolus, attaching thereto and changing
to dense component remains a possibility
but a less likely one in view of the known
intense biosynthetic activity of the nucleolus and the consequent implication of
active export of macromolecules such as
ribosomal ribonucleic acid (rRNA) from
this structure. Migration toward the nucleolus of macromolecular substances such
as messenger ribonucleic acid (mRNA) or
protein are possible but as yet biochemically undocumented transport processes.
A direction of movement out of the
nucleolus is favored by the morphologic
observation in which the flat, concave, or
irregular base of the paranucleolar structure, very slightly separated from the nucleolus, conforms in contour exactly with
the surface of the nucleolus (figs. 21, 30).
This indicates movement from the nucleolus, because an accessory body might be
expected to retain the contour of the nucleolar surface shortly after detachment
from the nucleolus but would probably not
assume such a configuration shortly before
attachment to the nucleolar surface,
Devoid of both affinity for azure A at
pH 4.0 and Feulgen reactivity, paranucleolar structures and accessory bodies afford no evidence for content of either RNA
or DNA. However, nucleoli, known to contain abundant RNA, stain intensely with
azure A. If the paranucleolar structures
derive from nucleolar dense component as
suggested, then this transition from dense
component to paranucleolar structure involves a loss in basophilia, possibly through
neutralization of RNA acid groups by the
protein linked to rRNA subunits (Perry,
'66). This seems more likely than a reverse process wherein conversion of paranucleolar structure to dense component of
the nucleolus involves addition or unmasking of RNA.
Since the only known function of the
nucleolus is the biosynthesis and export
of rRNA (Edstrom, Grampp and Schor,
'61; Perry, '62; Brown and Gurdon, '64:
Ritossa and Spiegelman, '65) and since the
nucleolar dense component contains RNA
(Marinozzi and Bernhard, '63), speculation appears justified in relating reactions
in RNA metabolism to the transformation
of nucleolar dense component into paranucleolar structures and to the fate of accessory bodies. However, the export of
rRNA takes place in non-neuronal cell nuclei devoid of paranucleolar structures and
accessory bodies and, therefore, the possibility exists that these two morphological
entities function in a capacity not directly
related to RNA synthesis. Ultrastructural
studies employing RNase digestion hopefully would provide information concerning RNA in paranucleolar structures and
accessory bodies.
The striking interface observed between
paranucleolar structure and underlying
dense component (figs. 13-17) prompts
speculation that a biochemical process of
conversion takes place along the line of
junction. Convertase, the principal enzyme thus far localized in the nucleolus,
(Perry, '66); Busch, Desjardins, Grogan,
Higashi, Jacob, Muramatsu, Ro and Steele,
'66) might be active at this site. The present state of ignorance, however, requires
consideration of other possible intranucleolar transformations as the locus for
convertase and other possible nucleolar
NUCLEOPLASMIC STRUCTURES OF TRIGEMINAL NEURONS
reactions as the chemical change effecting
formation of the paranucleolar structures.
Whatever the chemical reaction at the interface may be, it brings about an unexplained loss in cytochemically demonstrable basophilia.
The fate of the accessory body of Cajal
remains unknown. Since the accessory
body is not observed contacting the nuclear
membrane and thus is apparently not involved in transport to or from the cytoplasm, it presumably undergoes internal
transformation with or without an interaction at the site of another nuclear structure. The apparent derivation of the accessory body from the nucleolus prompts
speculation that it may transport rRNA
from the nucleolus to the nucleoplasm.
Once established that tritiated uridine
would be incorporated into the accessory
body of Cajal, the method of ultrastruct u r d radioautography might prove useful
in following the formation and fate of this
structure.
Little is known concerning the assembly
site(s) of ribosomal subunits. Recent observations concerning participation of
rRNA and protein in the release of nascent
mRNA from its DNA template (Shin and
Moldave, '66),indicate complex formation
between these two nucleic acids at the
site of mRNA synthesis in the nucleoplasm.
Although perhaps reflecting random association, the often noted proximity of accessory bodies to patches (fig. 38) suggests
that interaction might occw between components of these structures and accordingly infers the possibility of nucleic acid
and protein complexes forming in the organized structure of the patches. From
present knowledge, patches and accessory
bodies occur only in neurons.
The close topographical association of
nucleolar associated heterochromatin and
sex chromatin with a particular component of the nucleolus, namely the granular
component, is a new observation. Assuming that the nucleolar associated heterochromatin contains the cistron for rRNA
biosynthesis (Granboulan and Granboulan,
'65; Ritossa and Spiegelman, '65; Simard
and Bernhard, '67), ultrastructural autoradiographic experiments on rat trigeminal neurons might be expected to show
413
migration of labeled ribonucleic acid from
the heterochromatin to the granular component. However, autoradiographic experiments on cultured renal cells (Granboulan
and Granboulan, '65) have shown movement of newly labeled RNA from chromatin to the dense component before its
appearance in the granular component.
Although the latter experiments do not
definitely demonstrate conversion of dense
to granular component, these investigators
seems to favor this possibility. Such an
interpretation conflicts with that of the
present fmdings which indicate export of
the dense component from the nucleolus
as paranucleolar structure rather than
conversion of dense to granular component in the nucleolus. However, it is possible that the dense component of the neuronal nucleolus may contribute to the formation of both the granular component
and the paranucleolar structure.
Although sex chromatin bodies are reportedly absent from neurons of Rodentia
(Moore and Ban, '53), large intranuclear
bodies in rat trigeminal ganglia were identified as sex chromatin because of their
Feulgen reactivity, shape, size, position in
the nucleus, and sex distribution (Barr,
'66). The ultrastructure of sex chromatin
bodies, identical to that of heterochromatin, is strikingly different from the
ultrastructure of the accessory body of
Cajal and the paranucleolar structure.
Since antimonate deposits are localized in
the heterochromatin regions of the nucleus
(Spicer, Hardin and Greene, '68) and since
the sex chromatin body is a single X chromosome in a condensed state (Barr, '66),
Komnick's method of fixation is especially
useful for ultrastructural identification of
the sex chromatin body. This method is of
special value in the neuronal nuclei of rat
trigeminal neurons in which the sex chromatin body is the only large area of heterochromatin. The ultrastructure of a body
purported to be a sex chromatin body was
described in the adrenocortical cell nucleus of the cat (James, '60). However, in
contrast to the neuronal nucleus, the
adrenocortical cell nucleus possesses many
large clumps of heterochromatin that could
be easily confused with the sex chromatin
body at the electron microscopic level. The
414
J. H. HARDIN, S. S. SPICER AND W. B. GREENE
absence of configurations in which a sex
chromatin body adheres to a marginated
nucleolus without itself contacting the nuclear envelope points to a possible strong
affinity of sex chromatin for marginated
heterochromatin or the nuclear envelope.
The metabolic role of the dense granules in heterochromatin is unknown. These
granules may be identical to the indium
reactive perichromatin granules located on
the surface of chromatin clumps (Watson,
’62). Since they are apparently located in
clumps of heterochromatin, these dense
granules presumably are comparable
neither to the dense “interchromatin granules’’ in certain normal and neoplastic cells
(Bernhard and Granboulan, ‘63;Karasaki,
’65) nor to the “RNase resistant particles”
in nuclei of Erlich ascites tumor cells
(Swift, ’63). The relationship of the neuronal chromatin granules to granules in
nuclei of mouse organs, regarded by Allfrey
(’63) as the nuclear ribosomes responsible
for nuclear protein biosynthesis, is not clear
at present.
The morphologic findings afford little if
any indication as to the significance of the
granular body. Reference has not been
found to such a structure. Granular bodies seem to be unique structures to neuronal nuclei. Granular bodies have been
observed close to patches and granular
component at the nucleolar surface and
may react with or transport material between nuclear structures.
The nature and significance of the
patches remains speculative. The patches
appear similar to some of the nucleoplasmic bodies labeled “Kernplasmaschollen”
observed by phase contrast microscopy in
neuronal nuclei of rat spinal ganglia
(Andres, ’61). Lacking Feulgen reactivity
and azurophilia at pH 4.0, patches do not
contain demonstrable DNA or RNA. The
affinity of patches for alkaline Biebrich
scarlet reflects their content of basic protein. Thus, patches may be the source of
the basic protein known to comprise an
important constituent of the ribosome
(Waller, ’64). The protein portion of the
ribosome of the neuron might differ from
ribosomal protein in other cell types, accounting for the unique occurrence of the
patches in neurons.
ACKNOWLEDGMENTS
The authors express their appreciation
to Miss Joanne Wright, Mrs. Kathryn
Cowart, Mrs. Betty Hall, and Mr. George
Malanos for skilled technical assistance.
LITERATURE CITED
Allfrey, V. G. 1963 Nuclear ribosomes, messenger-RNA and protein synthesis. Exptl. Cell
Res., Suppl., 9: 183-212.
Andres, K. H. 1961 Untersuchungen iiber den
Feinbau von Spinal-ganglien. Z. Zellforsch.
Mikroscop. Anat., 55: 1-48.
Barr, M. L. 1966 The significance of the sex
chromatin. Inter. Rev. Cytol., 19: 35-95.
Beaver, D. L., H. L. Moses and C. E. Ganote
1965a Electron microscopy of the trigeminal
ganglion. I. Comparative ultrastructure. Arch.
Pathol., 79: 541-556.
1965b Electron microscopy of the trigeminal ganglion. 11. Autopsy study of human
ganglia. Arch. Pathol., 79: 557-570.
Bernard, W., and N. Granboulan 1963 The fine
structure of the cancer cell nucleus. Exptl. CeU
Res., Suppl., 9: 19-53.
Brown, D. D.,and J. B. Gurdon 1964 Absence
of ribosomal RNA synthesis in the anucleolate
mutant of Xenopus Eaeuis. Proc. Natl. Acad.
Sci., 51: 139-146.
Bunge, M. R., R. P. Bunge, E. R. Peterson and
M. R. Murray 1967 A light and electron microscope study of long-term organized cultures
of rat dorsal root ganglia. J. Cell Biol., 32:
439466.
Busch, H.,R. Desjardins, D. Grogan, K. Higashi,
S. T. Jacob, M. Muramatsu, T. S. Ro and W.
J. Steele 1966 Composition of nucleoli isolated from mammalian cells. In: International
Symposium on the Nucleolus - its Structure
and Function. W. S. Vincent and 0. L. Miller,
Jr. (eds.) United States Public Health Service,
National Cancer Institute Monograph 23. pp.
193-212.
Dixon, A. D. 1963 Fine structure of nerve-cell
bodies and satellite cells in the trigeminal ganglion. J. Dental Res., 42: 990-999.
Edstrom, J. E., W. Grampp and N. Schor 1961
The intracellular distribution and heterogeneity of ribonucleic acid i n starfish oocytes. J.
Cell Biol., 11: 549-557.
Estable, C., and J. R. Sotelo 1951 Una nueva
estructura cellular: el nucleolonema. Pub. Inst.
Invest. Cien. Biol. (Montevideo) 1 : 105-126.
Fawcett, D. W. 1966 T h e Cell. W.B. Saunders
Co., Philadelphia. pp. 28-31.
Granboulan, N., and P. Granboulan 1965 Cytochimie ultrastructurale du nuclkole. Exptl. Cell.
Ree., 38: 604-619.
Haggar, R. A. 1957 Behavior of the accessory
body of Cajal during axon reaction. J. Comp.
Neur., 108: 269-283.
Hardin, J. H., and S. S . Spicer 1969 Ultrastructure of neuronal nucleoli of r a t trigeminal
ganglia: Comparison of routine with pyroantimonate-osmium tetroxide fixation. Submitted for publication.
NUCLEOPLASMIC STRUCTURES O F TRIGEMINAL NEURONS
Hardin, J. H., S. S. Spicer and W. B. Greene
1967 Nucleolar ultrastructure clarified by
pyroantimonate-osmium fixation. J. Cell. Biol.,
35: 54A.
Hardin, J. H., S. S. Spicer and G. E. Mdanos
1968 Quantitation of nucleolar components in
rat trigeminal neurons. Anat. Rec., 160: 360361A.
Herndon, R. M. 1963 The fine structure of the
Purkinje cell. J. Cell Biol., 18: 167-180.
Hiraoka, J., and V. L. Van Breeman 1963
Ultrastructure of the nucleolus and the nuclear envelope of spinal ganglion cells. J.
Comp. Neur., 121: 69-87.
James, J. 1960 Observations o n the so-called
sex chromatin. Z. Zellforsch. Mikroscop. Anat.,
51: 597-616.
Karasaki, S. 1965 Electron microscopic examination of the sites of nuclear FlNA synthesis
during amphibian embryogenesis. J. Cell Biol.,
26: 937-958.
1962 Elektronenmikroskopishe
Komnick, H.
lokalisation von Na+ und C1- in Zellen und
Geweben. Protoplasma, 55: 414418.
Latham, J., and J. E. Darnell 1965 Entrance
of mRNA into HeLa cell cytoplasm in puromycin treated cells, J. Molec. Biol., 14: 13-22.
Lindsay, H. A., and M. L. Barr 1955 Further
observations on the behavior of nuclear structures during depletion and restoration of Nissl
material. J. Anat., 89: 47-62.
Marinozzi, V., and W. Bernhard 1963 Presence
d a m le nuclkole de deux types de ribonucleoproteines morphologiquement distinctes. Exp.
Cell Res., 32: 595-598.
Millonig. G. 1961 Advantages of a phosphate
buffer for OsOl solutions in-Exation. J. Applied
Physics., 32: 1637.
Moore, K. L.,and M. L. Barr 1953 Morphology
of the nerve cell nucleus in mammals, with
special reference to the sex chromatin. J. Comp.
Neur., 98: 213-231.
Perry, R. P. 1962 The cellular sites of synthesis of ribosomal and 4s RNA. Proc. Natl.
Acad. Sci., 48: 2179-2186.
1966 On ribosome biogenesis. In: International Symposium of the Nucleolus - its
Structure and Function. W. S. Vincent and
0. L. Miller, Jr. (eds.) United States Public
415
Health Service, National Cancer Institute Monograph 23. pp. 527-545.
Ramon y Cajal, S. 1903 Un sencillo metodo de
coloraci6n selectiva del reticulo protopl&smico
Y SUB effectos en 10s diversos organos nerviosos
de verterbrados 6 invertebrados. Trab. Lab. Invest. Biol. Univ. Madrid, 2: 129-221.
Ritossa, F. M., and S. Spiegelman 1965 Localization of DNA complementary to ribosomal
RNA in the nucleolus organizer region of
Drosophila melanagaster. Proc. Natl. Acad.
Sci., 53: 737-745.
Sankaranarayanan, K., and B. B. Hyde 1965
Ultrastructural studies of a nuclear body in
peas with characteristics of both chromatin
and nucleoli. J. Ultrastruct. Res., 12: 748761.
Shin, D. H., and K. Moldave 1966 The stimulation of RNA synthesis by ribosomes in vitro.
Biochem. Biophys. Res. Commun., 22: 232236.
Simard, R., and W. Bernhard 1967 A heat-sensitive cellular function located in the nucleolus.
J. Cell Biol., 34: 61-76.
Spicer, S. S. 1961 Differentiation of nucleic
acids by staining a t controlled pH and by a
SchiiT-methylene blue sequence. Stain Technol., 36: 337-340.
1962 Histochemically selective acidophilia of basic nucleoproteins in chromatin
and nucleoli a t alkaline pH. J. Histochem. Cytochem., 10: 691-703.
Spicer, S. S., J. H. Hardin and W. B. Greene
1968 Nuclear precipitates in pyroantimonateosmium tetroxide-hed tissues. J. Cell Biol.,
39: 216-221.
Swift, H. 1963 Cytochemical studies o n nuclear fine structure. Exptl. Cell Res., Suppl., 9:
54-67.
Thompson, B. K., R. A. Haggar and M. L. Barr
1957 The accessory body of Cajal in nerve
cell nuclei of the cat. J. Comp. Neur., 108:
253-267.
Waller, J. P. 1964 Fractionation of ribosomal
protein from Escherichia coli. J. Molec. Biol.,
10: 319-336.
Watson, M. L. 1962 Observations on a granule
associated with chromatin in the nuclei of
cells of rat and mouse. J. Cell Biol., 13: 162167.
PLATE 1
EXPLANATION OF FIGURES
Figs. 1 to 11 are photomicrographs of rat trigeminal neuronal nuclei at final magnification of x 2,400. With the exception of figure 9, all illustrations were photographed with bright-field optics. Specimens illustrated in figures 4-6 were obtained
from female rats and those shown in figures 1-3 and 7-11 were obtained from male
rats. Figures 3, 8, 9, and 10 illustrate specimens fixed sequentially with glutaraldehyde
and osmium tetroxide and embedded in maraglas. The remaining figures show
specimens fixed in formalin (unless stated otherwise) and embedded in paraffin.
1 An intensely argyrophilic paranucleolar structure protrudes from the nucleolar
surface a t about nine o’clock. Reduced silver method.
2
Two slightly stained paranucleolar structures protrude from the nucleolar surface
a t 9 and 12 o’clock. Reduced silver method.
3 A lightly stained paranucleolar structure attached to a heavily stained base
protrudes from the nucleolus a t nine o’clock. Cf. figures 13-17. Alkaline toluidine
blue.
4
A sex chromatin body is attached to the nucleolus a t nine o’clock. Cf. figure 24.
Light diffuse staining of the neuronal nucleoplasm contrasts with intense staining
of the satellite cell nucleus with condensed heterochromatin. Feulgen stain.
5 A sex chromatin body adheres to the nuclear envelope. Cf. figures 25 and 26.
Feulgen stain.
6 Sex chromatin body contacts both nuclear envelope and nucleolus. Cf. figure 27.
Feulgen stain.
7 A n argyrophilic accessory body of Cajal lies above and to the left of the lower of
two marginated nucleoli in this neuronal nucleus. Reduced silver method.
8
A lightly stained acecssory body of Cajal lies close to the nucleolus a t about ten
o’clock. Alkaline toluidine blue.
9
The accessory body of Cajal is further removed from the nucleolus than the
accessory body shown in figure 8 and is located near a patch. High density of the
accessory body is due to photography with phase contrast optics. Alkaline
toluidine blue.
10 Several moderately stained, irregular patches surround the upper right nucleolus
in this neuron. Alkaline toluidine blue.
11
416
Several moderately stained patches surround the intensely reactive nucleolus.
Fixation i n Carnoy’s fluid. Biebrich scarlet a t pH 9.5.
NUCLEOPLASMIC STRUCTURES OF TRIGEMINAL NEURONS
PLATE 1
J. H. Hardin, S. S. Spicer and W.B. Greene
417
Figs. 12 to 38 are electron micrographs of neuronal nuclei of rat trigeminal ganglia which, unless stated otherwise, were fixed with glutaraldehyde followed by osmium tetroxide. Except as noted differently, the thin
sections were stained with lead citrate. Figures 24-29 illustrate specimens
from female rats.
PLATE 2
EXPLANATION OF FIGURES
12 Dense component of the nucleolus protrudes above the level of adjacent granular
component at 12 and three o’clock on the nucleolar surface. The most supedcial
part of the dense protrusion at three o’clock has a granular appearance (arrow).
Such large dense protrusions are infrequently found and cannot therefore be
considered the ultrastructural equivalent of the paranucleolar structure. Fixation
with phosphate buffered osmium tetroxide. x 15,000.
13 A protrusion of dense component at three o’clock on the nucleolus and an area
of dense component even with the nucleolar surface at five o’clock exhibit an
irregular surface. Possibly the transformation of dense component to paranucleolar structure commences at the surface of such areas of dense component. Near 12
o’clock a paranucleolar structure protrudes from a portion of the nucleolar surface that consists solely of dense component. This base of dense component appears identical with the dense component in other areas of the nucleolus except
that it is not interspersed with low density areas. Fixation with phosphate buffered osmium tetroxide. X 12,500.
14 The interface between this paranucleolar structure and underlying base of dense
component (B) is at a level above the adjoining nucleolar surface composed of
granular component (GI. Small projections (arrows) of the most superficial
part of the base of dense component closely resemble the dense areas of the
paranucleolar structure. Phosphate buffered osmium tetroxide. x 25,000.
15 The interface between this paranucleoIar structure and base of dense component
( B ) is at a level even with the bordering nucleolar surface composed of granular
component (G). The lateral borders of the paranucleolar structure fit together
precisely with the lateral borders of the base of dense component giving the
impression that these two entities comprise a single morphological unit. Fixation
with phosphate buffered osmium tetroxide. X 30,000.
418
NUCLEOPLASMIC STRUCTURES OF TRIGEMINAL NEURONS
J. H. Hardin, S. S. Spicer and W. B. Greene
PLATE 2
419
PLATE 3
EXPLANATION OF FIGURES
16 A paranucleolar structure precisely overlies a base of dense component (B) which
is surrounded by granular component (G). This is a control preparation fixed
with a solution of potassium pyrophosphate (instead of potassium pyroantimonate) and osmium tetroxide. Electron-opaque deposits are not present in
this specimen as in the specimen shown in figure 17. X 30,000.
17 This specimen was fixed with Komnick's fluid, a solution of potassium pyroantimonate and osmium tetroxide. Electron-opaque deposits, not present in the control specimen (fig. 16), are evident in this nucleolus and paranucleolar structure.
The fine antimonate deposits (F) localized in the granular component are in
sharp contrast with the moderately fine (MP) antimonate deposits in the dense
component of the nucleolus, which are identical to those deposits in the dense
areas of the paranucleolar structure. In this profile, these dense areas of the
paranucleolar structure give the impression of being organized into a coiled and
possibly branched thread-like structure. The small, irregular projections (short
arrows) from the base of dense component possess the same type of antimonate
deposits as present in the dense areas of the paranucleolar structure, and at one
point (long arrow) such a projection is continuous with these dense areas.
Such configurations strongly suggest that the base of dense component contributes
to the formation of the paranucleolar structure. Unstained. x 30,000.
18 A large paranucleolar structure fits precisely on a thin base of dense component
at the nucleolar surface. Such a configuration suggests transformation of dense
component to paranucleolar structure may have proceeded to near completion.
Stained with uranyl acetate and lead citrate. X 30,000.
19 This infrequent configuration shows a paranucleolar structure protruding from an
area of the nucleolar surface that consists of granular and dense component. The
dense component, unlike the granular component, forms projections (arrows)
toward the paranucleolar structure. Paranucleolar structures possess areas of
dense material that closely resembles the dense component of the nucleolus and
are partially to completely surrounded by canalicular spaces with low density.
Areas of intermediate density are interspersed throughout the paranucleolar
structure. Millonig's phosphate buffered osmium tetroxide. x 38,000.
420
NUCLEOPLASMIC STRUCTURES OF TRIGEMINAL NEURONS
J. H. Hardin, S. S. Spicer and W. B. Greene
PLATE 3
42 1
PLATE 4
EXPLANATION OF FIGURES
20
This paranucleolar structure is apparently in the process of separating from the
nucleolar surface. A small area of the nucleolar surface (arrow) provides the
remaining point of contact between the nucleolus and the paranucleolar structure.
This area can be identified as dense component because it contains antimonate
deposits identical to those in the dense component elsewhere in the nucleolus.
Fine antimonate deposits ( F ) are localized in the granular component of the
nucleolus. Fixation with potassium pyroantimonate-osmium tetroxide. Unstained.
X 30,000.
422
21
The nucleolar surface beneath this partially removed paranucleolar structure consists solely of granular component except near the point of attachment (arrow).
The dense component that usually forms the base of attachment between the
paranucleolar structure and the nucleolar surface (figs. 14-18) was apparently
consumed in the production of this paranucleolar structure. The lower border of
the paranucleolar structure retains the contour of the opposed nucleolar surf ace.
Stained with uranyl acetate and lead citrate. x 30,000.
22
Two paranucleolar structures separated about 90" from one another protrude
from the nucleolar surface. An accessory body of Cajal ( C ) lies well separated
from the nucleolus but rather close to one of several patches ( P ) in the nucleoplasm. The morphology of the paranucleolar structures is identical to that of the
accessory body.
23
A paranucleolar structure protrudes from an extensive area of the nucleolar surface composed of dense component which appears stretched around part of a large
subjacent focus of pars amorpha (PA). Stained with uranyl acetate and lead
citrate. Fixation with phosphate buffered osmium tetroxide. x 15,000.
NUCLEOPLASMIC STRUCTURES OF TRIGEMINAL NEURONS
J. H. Hardin, S. S. Spicer and W. B. Greene
PLATE 4
423
PLATE 5
EXPLANATION O F FIGURES
24
The position and planoconvex shape of this sex chromatin body
attached to the nucleolar surface is identical to that of the paranucleolar structure, but its morphology closely resembles that of heterochromatin and clearly differs from that of the paranucleolar structure. Stained with uranyl acetate. X 10,000.
25 This sex chromatin body in contact with the nuclear membrane
consists largely of heterochromatin-like material and a few dense
granules. Each of these granules is enclosed by a thin halo of low
density material. These granules are more numerous in or near
small, ill-defined clumps of presumed heterochromatin (arrows) at
the nuclear envelope and in the nucleoplasm. x 20,000.
26
This nuclear profile discloses a sex chromatin body at 12 o’clock and
a nucleolus at eight o’clock, both of which adhere to the nuclear
membrane. The sex chromatin body possesses coarse, electron-opaque
0 in ~diameter and are
deposits which measure between 20 and 4
identical to those deposits in small clumps of heterochromatin at the
nuclear membrane, at the surface of the nucleolus, and free in the
nucleoplasm. Fixation with potassium pyroantimonate-osmium tetroxide. Unstained. X 10,000.
27
A sex chromatin body contacts both a marginated nucleolus and the
nuclear envelope. Sex chromatin bodies that adhered to a nucleolus
contacting the nuclear membrane never failed to adhere also to the
nuclear membrane. Stained with uranyl acetate and lead citrate.
x
424
10,000.
NUCLEOPLASMIC STRUCTURES OF TRIGEMINAL NEURONS
J. H. Hardin, S. S. Spicer and W. B. Greene
PLATE 5
425
PLATE 6
EXPLANATION OF FIGURES
28
A sex chromatin body is interposed between a nucleolus and the nuclear envelope. In this type of configuration the nucleolus is always
oriented on a n axis perpendicular to the plane of attachment of the
sex chromatin body to the nuclear envelope. Either this configuration
or that of figure 27 has been observed in every example of a sex
chromatin body contacting both the nucleolus and the nuclear membrane. x 10,000.
29
Like figure 28 except the specimen was fixed with potassium pyroantimonate-osmium tetroxide. The antimonate deposits in the sex
chromatin body and in the heterochromatin near the nuclear envelope are clearly different from those in the nucleolus. Unstained.
x 10,000.
30 A structure which may be interpreted as an accessory body of Cajal
is just barely removed from the nucleolar surface. The nucleolar
surface nearest the accessory body consists mostly of granular element which does not send projections from the surface. The apposed
surfaces of the nucleolus and the nascent Cajal body conform closely.
Stained with uranyl acetate and lead citrate. x 30,000.
31 This accessory body of
composed of flocculent
of this accessory body
tures. Cf. figure 19.
tetroxide. x 38,000.
426
Cajal is in contact with a nucleoplasmic area
material (arrow). Otherwise, the morphology
closely resembles that of paranucleolar strucFixation with phosphate buffered osmium
NUCLEOPLASMIC STRUCTURES OF TRIGEMINAL NEURONS
J. H. Hardin, S. S. Spicer and W. B. Greene
PLATE 6
427
PLATE 7
EXPLANATION OF FIGURES
32 Another variant of the accessory body of Cajal reveals a n enclosed
round area (arrow) with a texture and density resembling dense component of the nucleolus. X 45,000.
33 A n accessory body of Cajal, in a n apparent process of separation
from a small tangentially sectioned nucleolus near the nuclear
margin, partially encloses material (arrow) resembling nuclear dense
component. Such a configuration suggests that atypical accessory
bodies with similar inclusions (fig. 32) may develop as a result of
separation of a paranucleolar structure from the nucleolus prior to
complete conversion of the nucleolar dense component to paranucleolar structure. x 30,000.
34 A granular body surrounded by low density nucleoplasm lies immediately below a patch. x 25,000.
35 This granular body is enclosed by a membranoid structure except
at one corner where it lies close to granular component at the surface of a nucleolus. Granular bodies consist of dense granules, which
measure about 20 mp in diameter, associated moderately dense
amorphous material, and low density areas.
428
NUCLEOPLASMIC STRUCTURES OF TRIGEMINAL NEURONS
J. H. Hardin, S. S. Spicer and W. B. Greene
PLATE 7
429
PLATE 8
EXPLANATION OF FIGURES
36 This nuclear profile contains a nucleolus, an accessory body of Cajal,
( C ) and four moderately dense, irregularly shaped areas of the
nucleoplasm, designated patches (P). X 5,000.
37 A patch in the nucleoplasm discloses an organized pattern with
punctate dense material and associated less dense matter arranged
in slightly curved parallel rows. A clump of heterochromatin with
associated dense granules (arrow) borders the patch. X 20,000.
28
430
A zone of low density nucleoplasm largely surrounds a n accessory
body of Cajal, separating it from two closely bordering patches (P).
Flakes (arrows) are located within the patches and the low density
zone. The flakes have about the same density and size as the dense
areas in the accessory body. Uranyl acetate staining. x 25,000.
NUCLEOPLASMIC STRUCTURES OF TRIGEMINAL NEURONS
J. H. Hardin, S. S. Spicet and W. B. Greene
PLATE 8
431
Документ
Категория
Без категории
Просмотров
3
Размер файла
2 655 Кб
Теги
chromatin, nuclei, rat, cajal, cytochemical, sex, ultrastructure, accessory, structure, stud, trigeminal, body, neurons, related, paranucleolar
1/--страниц
Пожаловаться на содержимое документа