вход по аккаунту


Perilymphatic fibrocytes in the vestibule of the inner ear.

код для вставкиСкачать
Perilymphatic Fibrocytes in the Vestibule
of the Inner Ear '
Department of Anatomy, Hantarcl Medical School,
Boston, Massachusetts
A description is given of the fine structure of perilymphatic fibrocytes
and their associated fibers in the vestibule of the inner ear in rats. The identification
of the extracellular fibers as keratin is discussed in relationship to their fine structure
and to biochemical and biophysical data of other workers. The numerous junctional
complexes between fibrocytes are described and it is shown that these do not form
zonula occludens as has been reported by other workers.
The fibrocytes that comprise the cellular portion of the perilymphatic reticulum
of the inner ear are derived embryologically from mesenchyme that condenses
around the otocyst. During development
the reticulum becomes highly specialized
structurally, apparently for the specific
function of providing support for the membranous labyrinth and of aiding in the
transmission of stimuli to the sensory receptors. In the cochlea of mammals the
function of channeling transmission of
stimuli is especially well developed, but in
the vestibule, around the utricle, saccule
and ampullae, the reticulum retains as its
primary function the support of the membranous elements. In both the vestibule
and cochlea of adult animals, fibrocytes
are embedded within a matrix of extracellular fibers and fluid which appears to
be unique to the inner ear. The fibers do
not resemble those of other connective
tissues in their chemical nature, in their
histochemical reactions, or electron microscopical appearance (Iurato, '62; Bairati
and Iurato, '59). The purpose of this paper
is to report new observations on the structure of the fibrocytes and extracellular
fibers, and on the junctional complexes between fibrocytes.
Ears from 24 white laboratory rats
(Rattus norvegicus) were fixed by intravascular perfusion with 1 or 2% osmium
tetroxide, preceded by a brief wash with
a diluted Karnovsky aldehyde mixture
(Karnovsky, '65). The perfusate and the
Rsc., 157: 627-640.
technique will be reported elsewhere. The
tissue was embedded in Araldite and Epon
and sectioned on a Huxley ultramicrotome.
Sections were doubly stained with saturated aqueous uranyl acetate and 0.1% or
0.2% lead citrate and viewed on an RCA
EMU 3G electron microscope. In one animal, the tissue was treated prior to dehydration with 0.5% uranyl acetate (pH 5.2)
as recommended by Farquhar and Palade
('65), except that maleate buffer was substituted for veronal-acetate (Karnovsky,
personal communication).
The fibrocytes of the adult rat vestibular
perilymphatic system form an elaborate
reticulum of cells that blends peripherally
into the periosteum of the otic capsule.
With the light microscope, identification
of the cells is often difficult, especially in
the thick sections used in paraffin and
celloidin embedding, because the cytoplasm is extremely attenuated except in
the immediate vicinity of the nucleus.
With the electron microscope, however, it
is clear that the very thin strands that
make up the reticulum are cell processes
that form junctions with numerous surrounding fibrocytes. In areas in which
blood capillaries and nerves are present
the fibrocyte provides support for these
elements by wrapping intimately around
them. In the case of the capillary, very
narrow extensions of fibrocyte cytoplasm
ZResearch supported by fmnt RG-06729 from the
Division of General Mehca Sciences, National Institutes of Health. United States Public Health Service
and in part by.a grant from the William F. Milton
Fund of Harvard University.
can be seen external to the basal lamina
of the capillary endothelium, and outside
of the pericytes and their processes (fig.
2). The fibrocytes tend to be more numerous near the membranous labyrinth than
elsewhere in the vestibule and they come
very close to the basal region of the epithelium. In the rat, however, there is a
very distinct separation between fibrocytes
and the epithelium of the labyrinth which
is filled by a dense accumulation of extracellular fibers that will be described below
(fig. 5).
All of the fibrocytes that have been observed (fig. 1 ) appear to be in a relatively
quiescent state and do not possess extensive development of granular endoplasmic
reticulum that is typical of actively secreting fibrocytes (Revel and Hay, '63). Instead, the endoplasmic reticulum is usually
represented by a few scattered cisternal
profiles that are bounded by smooth membranes for the most part, but which
in some areas have adherent ribosomes.
Often the lumina of the cisternae contain
a slightly electron-dense, flocculent material that is of the same appearance as
that found in the Golgi vacuoles. Occasionally, elongate cisternae become associated in packets within which both rough
and smooth membranes are found. The
juxtanuclear Golgi complex is composed of
flattened cisternae, which at the Golgi periphery, are slightly expanded and are surrounded by vacuoles containing a slightly
flocculent density. Those vacuoles toward
the nucleus are of smaller diameter and
are more regularly circular in profile than
other Golgi associated vacuoles.
The remainder of the cytoplasm is filled
with scattered ribosomes and polyribosomes, elongate mitochondria and a light
staining granular material of unknown
nature. Lysosomes are rare and usually
are small. The nucleus is of interest because it often contains large (500-550 A)
dense granules different from the heterochromatin or euchromatin (fig. 6). These
granules are invariably surrounded by
a light halo approximately 200 A wide.
Granules of this type were first reported
by Watson ('62) who found them to be
present in cells from a large number of
rat and mouse tissues. Recently Bruni and
Porter ('65) report the presence of simi-
lar granules in rat hepatocyte nuclei. The
tissues studied by Watson ('65) were
stained in block with aqueous uranyl acetate (pH 3.5-4), and in the present investigation it is in material treated in
approximately the same way (Farquhar
and Palade, '65) that the granules are
most clearly delineated. Available evidence
indicates that they contain nucleic acids
(Watson, '62), but their functional significance is unknown.
Fibrocytes have not been seen to form
junctional complexes with cells other than
fibrocytes. The type of junction that forms
between fibrocytes is of interest, however,
since in other connective tissues where
the cells are more highly dispersed one
can rarely find cell junctions. At any one
junction a number of cell processes may
take part. It is not clear whether a fibrocyte ever forms a junction with itself,
although it is possible that at the junctional complexes each process represents
a separate cell. In the tissue treated with
uranyl acetate, membrane structure is well
preserved, and in sections normal to the
membrane it is possible to trace their trilaminar unit membrane structure completely through the junction (figs. 7, 8).
Thus, as two membranes approach to
within 50 A of each other the outer leaflets
of the unit membranes persist and do not
fuse (as is found in zonulae or maculae
occludentes). The outer leaflet, however,
is different from that found in other unit
membranes (Robertson, '64) in that it is
irregularly interrupted and at times composed of punctate densities. It is approximately the same thickness as the inner
leaflet (N 35 A). As can be seen in figures
7 and 8, the intercellular space (50 A ) is
approximately of the same size and electron density as the spaces found between
outer and inner leaflets of each membrane
(35-50 A). There appear to be no specializations on the cytoplasm side of the
plasma membranes. At times, one has the
impression that membrane fusion does
occur, but in all instances seen here this
can be attributed to obliquity of the membranes with respect to the plane of section.
The extracellular fibrous matrix in which
the fibrocyte is suspended seems to be different from that described for other connective tissues (Porter, '64; Greenlee et al.,
'66; Revel and Hay, '63). Here, there are
two fibrous components (figs. 1 , 5 ) . The
first is a very light staining fiber of indefinite length (one 5 IJ long has been
seen) which varies in diameter from 20
to 50A. The light fibers can be found at
some distance from the fibrocyte and, as
mentioned above, in the vicinity of the
membranous labyrinth they form a distinct fibrous supporting band around the
epithelium (fig. 5). A periodicity is apparent in some instances in these fibers, but
it is not consistent. Cross-banded fibers
with a 640 A periodicity, typical of collagen, are never seen. In the supporting
band the fibers are more or less randomly
oriented except in areas in which the membranous labyrinth bends. In these places,
fibers become associated into dense fascicles and seem to blend into the basal
lamina of the epithelium to form a firm
support for the cells involved in change
in spacial orientation.
The most interesting aspect of the extracellular fibers is a second fiber type, that
is apparent in the vicinity of the fibrocyte, which aggregates into heterogeneous
fibrous bundles (figs. 2, 3, 4, 5). In electron micrographs of tissues fixed and prepared in a number of different ways the
structure and configuration of the bundles
is constant. This second fiber type is extremely electron-dense and much larger in
diameter (- 110 A) than the fiber first
described above. In high resolution electron micrographs (fig. 3 ) it can be shown
to be composed of four fibrils, each a p
proximately 50 A in diameter. Length
measurements are inexact since the fibers
pass in and out of the plane of section, but
it is rare to find one over 2000 A long. In
cross-section (fig. 3 ) the dense fiber appears highly angular, at times being
square and at other times being diamond
shaped, due to the packing of its fibrils.
Longitudinal sections of the dense fibers
are difficult to interpret because of their
undulant nature. Most longitudinal sections show fibers cut obliquely, thus giving
a granular appearance. In fibers where the
plane of section is favorable (fig. 4) one
can see individual 50 A fibrils that appear
to be wound upon one another. The dense
fibers are loosely packed in an hexagonal
array (figs. 2, 3) that is often disrupted,
presumably due to preparative artifacts.
Interspersed between the dense fibers is a
network of rather amorphous light-staining material that at times Iooks fibrillar
(fig. 3). Fibrous bundles are always in intimate association with fibrocytes and rarely
are seen more than a micron or so from
the cell body (figs. 1 , 2 , 5). Single dense
fibers with their characteristic angularity
are at times seen in cross-section scattered
throughout the basal supporting band of
the membranous labyrinth.
It would appear that once the definitive
adult perilymphatic reticulum is formed
there is little need for further structural
alterations, for the virtual absence of actively secreting cells would preclude much
change. That these cells are relatively quiescent is apparent in their reduced amount
of granular endoplasmic reticulum and
small Golgi complex, which fits Porter's
('64) description of mature fibrocytes compared to actively secreting fibroblasts.
Collagen is a general constituent of
other commonly studied connective tissues
(Porter, '64; Revel and Hay, '63) even
though other fiber types may be present
in specific instances (e.g., elastic fibers,
Greenlee et al., '66). The occurrence of a
connective tissue with no obvious collagen,
but with two fiber types, is thus of interest.
It is tempting to speculate that the light
fiber is a protofibril of collagen that never
aggregates with others to form a definitive
collagen fiber with 640 A banding but, as
will be discussed below, this does not fit
available evidence. These highly elongate
fibers resemble to a considerable degree
the tonofilaments commonly found in epithelial cells. This superikial resemblance
is substantiated somewhat by the reports
of Iurato ('62) that the fibrous portion of
the connective tissue in the cochlea belongs to the keratin group of proteins, and
of Mercer ('61) that the cellular tonofilaments may be composed of a keratinous
protein. Keratin, as an extracellular component, is found in the gizzard and proventriculus of birds (Mercer, '61) and can
be identified as a component of the cuticular tectorial membrane in the cochlea
(Iurato, '60). The perilymphatic cells,
however, are mesodermal derivatives and
Amino acid composition (g/100 g protein) of
collagen, wool keratin and perilymphatic fibers
Peritendon lymphatic
collagen 1
fibers 2
Aspartic acid
Glutamic acid
1 .oo
From Leach ( ' 5 7 ) .
From Iurato ('62).
From Simmonds ('55).
these cell types are nut generally implicated in extracellular keratin elaboration.
The evidence for identification of perilymphatic fibers as keratin rests upon
Iurato's ('62) data on the cochlea and on
Bairati and Iurato's ('59) data on the vestibule. Iurato ('62) has shown that a
rather well purified sample of limbus
spiralis yields x-ray diffraction patterns
with equatorial reflections at 4.6-4.7 A
and 10A. Kendrew ('54) reports similar
patterns for known keratins. Collagen, on
the other hand, besides having equatorial
reflections similar to keratin gives meridional reflections with spacings at 2.86A
(Engstrom and Finean, '58) which are
missing both from keratins and from the
perilymphatic fibers.
The amino acid analysis done by Iurato
('62) - see table 1-lends support to the
evidence above. Unfortunately, he did not
determine proline or hydroxyproline content, which would be diagnostic for collagen. The presence of cystine in the
fibers and its virtual absence in collagen
(table l), coupled with comparison of the
amounts of the other amino acids between
collagen and perilymphatic fibers, suggests
that these fibers are not collagen. Comparison with known keratins (Mercer, '61)
shows that amino acids in perilymphatic
fibers vary in quantity in a manner more
similar to known keratins than to collagen.
Iurato's ('62) determinations, of course,
were done on total perilymphatic fibers
and did not differentiate between the components. Identification of the chemical
composition of the fibrous bundles, separate from the other fibrous component, is
complicated by the fact that they comprise
only a small proportion of the extracellular fiber population. Its fine structure,
however, falls rather well into line with
the model of fibrous keratin as put forward by Birkbeck and Mercer ('57).
Thus, a fibril of keratin is considered to
be composed of groups of filaments (akeratin) which are comprised of bundles
of a-helices, the whole embedded within an
amorphous keratinous matrix (y-keratin).
The four subfibers that make up one fiber
(filament in the terminology of Mercer,
'61), as reported here, does not fit exactly
with the model of Mercer ('61), nor with
that of Pauling and Corey ('53) in which
the keratin filament is considered to be
made up either of seven or three helically
wound strands. It is possible that differences wiU exist in fine structure of keratins from different sources in the same
manner in which they differ in chemical
composition. Iurato ('62) interprets the
fibers that he sees in the cochlea as having
the seven-stranded structure. If this is true,
then i t indicates a difference between the
perilymphatic reticulum in the cochlea and
that in the vestibule. Iurato's ('62) illustrations, however, are somewhat indistinct
and are not of high resolution, and it is
probable that one should withhold judgment on the structure of the cochlear fibers
until further work has been done.
The amorphous matrix within which
the dense fibers are embedded at times
takes on a fibrillar structure. It may be
that it is the same as the elongate lightstaining fibers that form the bulk of the
extracellular fibers, for at times one can
see indications of light-staining fibrils extending beyond the limits of a fibrous
The junctional complexes found between fibrocytes are interesting in light
of recent anatomical and physiological
studies on junctions between cells in other
tissues (Farquhar and Palade, '63, '65;
Kanno and Lowenstein, '64; Robertson,
'64; Flickinger and Fawcett, '66). The
junction described here does not fit exactly into any previously described junctional type. One would predict a pTiori
that its major function is in cell to cell
adherence and, although Potter et al. ('66)
have shown in squid embryo that cells
some distance apart are electrically coupled to some degree, one would not be led
at first to think that synchronization of
activity is of much importance in this cell
system in the adult animal, for physiological and biochemical evidence implicates
perilymphatic fibrocytes neither in metabolism of perilymph, nor in audition and
equilibration. On present evidence these
cells can only be considered as supporting
elements and the junctional regions between them as adding to the integrity of
the system. Devis and James ('64) report
that in fibroblasts in tissue culture, close
apposition and apparent fusion of membranes occur, but from their micrographs
one is not able to resolve each leaflet of
the plasma membranes. They make the
point that it is possible that fusion does
not occur and that the single dense line
that they see could be a composite of narrowed extracellular space and outer leaflets
from the two closely apposed plasma membranes. The material used in this investigation would support the last contention,
for although the extracellular space is narrowed, there is never fusion of membranes
in the strict sense (Farquhar and Palade,
'63). The density within the extracellular
space suggests that perhaps something is
present acting as a binding material that
is similar in nature to, but more highly
concentrated than, the extracellular glycoprotein coating found on many cells
(Bennett, '65). The same type of density
is present within the plasma membrane
(compare, for instance, the plasma membranes in figs. 3, 7 or 8 ) which indicates
that the membrane at this point is chemically, if not structurally, different from
that seen elsewhere in the cell.
I want to thank Professor Don W. Fawcett for critically reading this manuscript.
Bairati, A., Jr., and S. Iurato
1959 Natura
e struttura submicroscopica della membrana
basale dei canali semicircolari. Monit. zool.
Ital. Suppl., 67: 242-247.
Bennett, H. S. 1965 Introductory remarks. In:
Intracellular Membranous Structure. S. Sen0
and E. V. Cowdry, eds., Japan Society for Cell
Biology, 497-503.
Birbeck, M. S. C., and E. H. Mercer 1957 The
electron microscopy of the human hair follicle.
I. Introduction and the hair cortex. J. biophys.
biochem. Cytol., 3: 203-5314.
Bruni, C., and K. R. Porter 1965 The fine structure of the parenchymal cell of the normal r a t
liver. 1. General observations. Amer. J. Path.,
46: 691-755.
Devis, R., and D. W. James 1964 Close association between adult guinea pig fibroblasts in
tissue culture, studied with the electron microscope. J. Anat., 98: 63-68.
Engstriim, A., and J. B. Finean 1958 Biological
Ultrastructure. Academic Press, New York,
Farauhar. M.. and G. Palade 1963 Junctional
complexes 'in various epithelia. J. Cell Biol.,
17: 375-412.
1965 Cell junctions in amphibian skin.
J. Cell Biol., 26: 263-291.
Flickinger, C., and D. W. Fawcett 1966 The
junctional specializations of Sertoli cells in the
seminiferous epithelium. Anat. Rec. (in press).
Greedee, T. K., R. Ross and J. L. Hartman 1966
The fine structure of elastic fibers. J. Cell Biol.,
30: 59-71.
Iurato, S. 1960 Submicroscopic structure of the
membranous labyrinth. I. The tectorial membrane. Z. f. Zellforsch., 52: 105-128.
1962 Submicroscopic structure of the
membranous labyrinth. 111. The supporting
structure of Corti's organ (basilar membrane,
limbus spiralis and spiral ligament). Z. f. Zellforsch., 56: 40-96.
Kanno, Y., and W. R. Lowenstein 1964 Low resistance coupling between gland cells. Some
observations on intercellular contact membranes and intercellular space. Nature, 201:
194-1 95.
Karnovsky, M. J. 1965 A formaldehyde-glutaraldehyde fixative of high osmolality for use i n
electron microscopy. J. Cell Biol., 27: 137A.
Kendrew, J. C. 1954 Structure proteins. I. In:
The Proteins. H. Neurath and K. Bailey, eds.,
Academic Press, New York, V. I1 (3): 845950.
Leach, A. A. 1957 The amino acid composition
of amphibian, reptile and avian gelatins. Biochem. J., 67: 83-87.
Mercer, E. H. 1961 Keratin and Keratinization.
Pergamon Press, London, XIV+316.
Pauling, L., and R. B. Corey 1953 Compound
helical configurations of polypeptide chains:
structure of proteins of the a-keratin type. Nature, 171: 59-61.
Porter, K. R. 1964 Cell fine structure and biosynthesis of intercellular macromolecules. In:
Connective Tissue: Intercellular Macromolecules. S. Farber, ed., Little, Brown and Co.,
Boston, 167-196.
Potter, D. D., E. J. Furshpan and E. W. Lennox
1966 Connections between cells of developing squid as revealed by electrophysiologicd
methods. Roc. Nat. Acad. Sci. U. S., 55: 328336.
Revel, J. P.,and E. D. Hay 1963 A n autoradiographic and electron microscopic study of col-
lagen synthesis in differentiating cartilage.
Z. f. Zellforsch., 61: 110-144.
Robertson, J. D. 1964 Unit membranes: a review with recent new studies of expcrimental
alterations and a new subunit structure in
synaptic membranes. In: Cellular Membranes
in Development. M. Locke, ed., Academic Press,
New York and London, 1-81.
Simmonds, D. H. 1955 The amino acid composition of keratins. Austral. J. Biol. Sci., 8:
Watson, M. L. 1962 Observations on a granule
associated with chromatin in the nuclei of cells
of rat and mouse. J. Cell Biol., 13: 162-167.
The general features of a perilymphatic fibrocyte are shown i n this
micrograph. The cell has little granular endoplasmic reticulum and
scattered elongate mitochondria. Note the distribution of the extracellular fibers, with fibrous bundles located close to the cell. The
upper portion of the picture is nearest to the membranous epithelium.
x 19,000.
David W. Hamilton
Here a thin layer of fibrocytes wraps around a capillary and its pericyte. Note the fibrous bundles closely applied to the fibrocyte plasma
membrane. X 44,000.
3 Enlargement of the fibrous bundle outlined in figure 3, showing the
four fibrils that comprise each dense fiber and the rather amorphous
matrix within which dense fibers are embedded. X 220,000.
Longitudinal section of a fibrous bundle. The arrows point to two
5 0 A fibrils apparently winding about each other along the length of
the fiber. x 107,000.
David W. Hamilton
5 Here a fibrocyte lies close to a labyrinth epithelial cell (containing a
lipid droplet) and is separated from it by the fibrous band composed
of light-staining fibers. Note the fibrous bundles intimately associated
with the fibrocyte. X 28,000.
David W. Hamilton
Enlargement of a portion of the nucleus from the cell in figure 1
showing the large (500 A ) dense granules with surrounding halos
found in nuclei in these cells. x 110,000.
Junction of two fibrocytes. At the arrow note that the two leaflets of
each plasma membrane can be traced into the junction with no
fusion of outer leaflets. x 210,000.
Junction of three fibrocytes. At the arrows the membranes have been
cut normally and one can see the two leaflets of the unit membrane
i n each. Note the density between the leaflets and that in the intercellular space. x 210,000.
David W. Hamilton
Без категории
Размер файла
1 170 Кб
vestibular, inner, perilymphatic, ear, fibrocytes
Пожаловаться на содержимое документа