| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J Neurophysiol (March 1, 2003). 10.1152/jn.00690.2002
Submitted on Submitted 16 August 2002; accepted in final form 31 October 2002
1Department of Psychiatry and Human Behavior, University of Mississippi Medical Center, Jackson, Mississippi 39211; and 2Harold W. Siebens Hearing Research Center, Central Institute for the Deaf, Departments of 3Otolaryngology and 4Anatomy and Neurobiology, Washington University, Saint Louis, Missouri 63110
 |
ABSTRACT |
|---|
 TOP
 ABSTRACT
 INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Si, Xiaohong, Mridha Md. Zakir, and J. David Dickman. Afferent Innervation of the Utricular Macula in Pigeons. J. Neurophysiol. 89: 1660-1677, 2003. Biotinylated dextran amine (BDA) was used to retrogradely label afferents innervating the utricular macula in adult pigeons. The pigeon utriclar macula consists of a large rectangular-shaped neuroepithelium with a dorsally curved anterior edge and an extended medioposterior tail. The macula could be demarcated into several regions based on cytoarchitectural differences. The striola occupied 30% of the macula and contained a large density of type I hair cells with fewer type II hair cells. Medial and lateral extrastriola zones were located outside the striola and contained only type II hair cells. A six- to eight-cell-wide band of type II hair cells existed near the center of the striola. The reversal line marked by the morphological polarization of hair cells coursed throughout the epithelium, near the peripheral margin, and through the center of the type II band. Calyx afferents innervated type I hair cells with calyceal terminals that contained between 2 and 15 receptor cells. Calyx afferents were located only in the striola region, exclusive of the type II band, had small total fiber innervation areas and low innervation densities. Dimorph afferents innervated both type I and type II hair cells with calyceal and bouton terminals and were primarily located in the striola region. Dimorph afferents had smaller calyceal terminals with few type I hair cells, extended fiber branches with bouton terminals and larger innervation areas. Bouton afferents innervated only type II hair cells in the extrastriola and type II band regions. Bouton afferents innervating the type II band had smaller terminal fields with fewer bouton terminals and smaller innervation areas than fibers located in the extrastriolar zones. Bouton afferents had the most bouton terminals on the longest fibers, the largest innervation areas with the highest innervation densities of all afferents. Among all afferents, smaller terminal innervation fields were observed in the striola and large fields were located in the extrastriola. The cellular organization and innervation patterns of the utricular maculae in birds appear to represent an organ in adaptive evolution, different from that observed for amphibians or mammals.
| |
INTRODUCTION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
The vestibular otolith system
functions to detect translational motion and position of the head with
respect to gravity. In birds, the otolith system is extensive, with
three different endorgans including the utricular, the saccular, and
the lagenar maculae. Since Retzius (1884)
,
Ramon y Cajal (1909) and Lorente de No
(1926) first described differences in afferent fiber size, as
well as differences in each fibers terminal field in the sensory
epithelia of the maculae, questions concerning the functional utility
of the observed morphological variations have been posed. Anatomical descriptions have shown that receptor hair cells differ in structure and number throughout the maculae but are organized with a common plan
that spans several animal classes. For example, all vertebrate utricular maculae have a specialized region: the striola, which differs
in receptor cell density and otoconial formation (Werner 1933). In the approximate center of the striola lies an
imaginary line, demarcated by the opposing reversal of the hair cell's
morphological polarization (Flock 1964; Lindeman
1969). The striola typically runs longitudinally through the
neuroepithelium; however, its exact shape and position differs among
species. In fish (Platt 1975; Popper and
Northcutt 1983), amphibians (Baird and Schuff 1994), and birds (Jørgensen and Andersen 1973),
the utricular striola is eccentrically placed near the lateral border
of the macula, whereas in mammals (Flock 1964;
Fernandez et al. 1990), it lies more medially toward the
center of the macula. In chinchillas, a utricular striola was described
that comprised 10% of the surface epithelium and divided the rest of
the macula into a medial extrastriola (40%) and a lateral extrastriola
(50%) (Fernandez et al. 1990).
In amniote vertebrates, two distinct types of vestibular receptor cells
exist based on differing morphologies (Lysakowski and Goldberg
1997; Wersäll 1956), innervations
(Kevetter et al. 1994; Lysakowski and Goldberg
1997), and channel properties (Correia and Lang
1990; Eatock et al. 1994; Rennie and
Correia 1994; Ricci et al. 1997;
Rüsch et al. 1998). Mature type I hair cells are characterized by their amphora shape and the calyceal afferent terminal
that engulfs the hair cell. Type II hair cells are generally thinner,
tube shaped, and are innervated by bouton terminals. In birds, a
differential distribution of the two receptor types in the utricular
sensory epithelium has been shown with a higher concentration of type I
hair cells contained within the striolar region and more numerous type
II hair cells outside the striola (Jorgensen and Andersen
1973; Rosenhall 1970). In mammals, the utricular
macula can be proportioned into separate areas: the striola region with
a high concentration of type I hair cells, the juxtastriola region with
a mixture of type I and type II hair cells, and the extrastriola
region, which lies in the peripheral macula and also contains a mixture
of type I and II hair cells (Fernandez et al. 1990). In
birds, the striolar region of the utricle and lagena otolith organs are
substantially different from those described for the mammalian
utricular macula (Fernandez et al. 1990) or avian
saccule (Zakir et al. 2003). In these bird organs, a
narrow band of six to eight type II hair cells in width (type II band)
runs through the striola with the morphological reversal line in the
center of the band. On either side of the type II band, larger zones of
mixed type I and type II hair cells are found (Jørgensen and
Anderson 1973; Rosenhall 1970).
Primary afferents that innervate receptor cells of the utricular
neuroepithelium have generally been characterized based on their
regional innervation patterns. For vertebrates that possess both type I
and II cells, only the chinchilla has been thoroughly examined, and
three classes of afferents were described (Fernandez et al.
1990). Calyx afferents exclusively innervated either one or
multiple type I hair cells and were found only in the striola region.
Dimorph afferents innervated both type I and type II hair cells and
were distributed throughout the macula. Although few in number, bouton
afferents innervated only type II hair cells and were found only in the
extrastriola regions. In rats, both calyx and dimorphic afferents have
been described using serial section electron micrographs (Ross
1985). In frogs and fish, only bouton afferents exist, yet
differences among hair cell type and afferent innervation in the
maculae have been reported (Baird and Lewis 1986;
Baird and Schuff 1994; Chang et al. 1992;
Lewis and Li 1975).
To date, the afferent innervation patterns in bird vestibular organs
have not been examined except for our recent companion study of the
pigeon saccular macula (Zakir et al. 2003). Because the
utricular macula of birds appears to contain similar morphological properties to those of both amphibians and mammals, it is of interest to characterize the afferent innervation patterns and terminal distribution in these animals. The present study utilized neural tracers, image analysis, and three-dimensional reconstruction techniques to quantify the morphological parameters of utricular afferents in pigeons. Differences in the anatomical organization, regional distribution of afferent types, and innervation profiles were
determined and then compared with these same properties described for
other species.
| |
METHODS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Animals
The experiments were conducted in adult male pigeons (Columba livia) that ranged in age from 1 to 3 yr. The methods of study were approved by both the University of Mississippi Medical Center and the Central Institute for the Deaf Institutional Animal Care and Use Committees (work was performed at both locations), and all animal procedures were performed in accordance with the American Physiological Society guidelines.
Tracer injections
Biotinylated dextran amine (BDA, 10,000 MW; Molecular Probes)
was injected into the vestibular nuclei of the pigeon brain stem. Each
animal was initially anesthetized with pentobarbital sodium (16 mg/kg
iv), then supplemented with ketamine hydrochloride (10 mg/kg im) given
once every hour or as needed. Body temperature (40°C) was maintained
with a thermal pad. Once anesthetized, the animal was placed in a
stereotaxic device with the beak angled downward ~13° so that the
horizontal semicircular canals were aligned with the stereotaxic
horizontal plane. A small opening was made in the parietal bone, and a
glass micropipette filled with BDA (10% in saline) was lowered into
the brain using predetermined coordinates for the vestibular nuclei
(about posterior, 3.5 mm; lateral, 2.5 mm; and dorsal, 2.5 mm relative
to a point intersecting the interaural axis and the midline of the
bird). BDA was passed into the brain using ionophoresis with positive
current (6-8 µA, 50% duty cycle) over a 10-min duration. Using
slightly different injection site coordinates for each animal, BDA was
deposited into different regions of the vestibular nuclei. After
cessation of injection current, the electrode was allowed to remain in
position for at least 2 min with a small negative current (
0.04 µA)
applied. The electrode was then retracted. Gelfoam was placed over the brain surface followed by bonewax, and the skin was sutured closed. A
single dose of butorphanol (0.5 mg/kg im, Stadol) was given to
alleviate postoperative pain, and the animal was returned to its home cage.
Histology
After 10-14 days of postinjection survival, the pigeon was anesthetized with pentobarbital sodium (20 mg/kg iv), the mastoid bone was opened, and an intralabyrinthine perfusion was performed using 1% glutaraldehyde, 1.25% paraformaldehyde, and 1% acrolein in 0.1 M phosphate buffer. Next, the chest cavity was opened, and the animal was transcardially perfused with sodium nitrite (1%) in saline (250 ml), followed by 750 ml of 1% glutaraldehyde-1.25% paraformaldehyde fixative. The membranous labyrinth was excised and dissected free into individual endorgans (3 semicircular canals, utricle, saccule and lagena). The brain was removed and along with the vestibular endorgans, placed in the aldehyde fixative solution for 24 h. The utricle was then trimmed so that the sensory neuroepithelium was visible. This tissue was treated with a protease (type XXIV; Sigma) 30 µg/ml for 10 min to remove the otolith membrane and otoconia by gentle agitation. The brain was blocked, frozen, and serially sectioned (60 µm) in the transverse plane.
Both the endorgans and the brain sections were processed for BDA using
a modified 3-3' diaminobenzidine (DAB) procedure (Brandt and
Apkarian 1992). Briefly, the tissue was incubated for
12 h in a solution of phosphate buffer, 1% Triton-X100, and
0.25% avidin D-HRP (Vector Labs). The tissue was then
reacted using a solution of DAB and 1% nickel-cobalt in 0.1 M
phosphate buffer as the chromogen and 0.3%
H2O2 as the initiator until
a dense reaction product was visualized. The tissue was then rinsed in
phosphate buffer. The brain sections were dehydrated using a series of
graded alcohols and xylenes, mounted on glass slides, and
counterstained with neutral red. After the reaction procedure, the
whole maculae were placed into a depression slide, viewed using a Nikon
light microscope, and the apical surface of the neuroepithelium was
photographed. The utriclar maculae were then embedded into plastic
(Durcupan) and serially sectioned (10 µm thickness) using a rotary
microtome. The sections were mounted on glass slides and counterstained
(Richardson et al. 1960).
SEM preparation
In some animals, the utriclar maculae (n = 14)
were prepared for scanning electron microscopy (SEM). First, the
otoconia and otolith membranes were removed by treatment with protease.
The tissue was placed into a 0.5% osmium and 0.1 M phosphate buffer solution for 2 h. After six serial washings (10 min each) in
distilled water, the tissue was dehydrated using a graded series of
acetones (10 min each in 70, 90, and 95% acetone), followed by three
washings of 15 min in 100% acetone. The tissue was then incubated in a 1:1 mixture of tetramethylsilane (TMS) and acetone for 45 min followed
by an incubation of 45 min in a 3:1 mixture of TMS and acetone. The
tissue was then incubated twice in 100% TMS for 1 h. Finally, the
tissue was dried in the oven for 30 min, allowing the TMS to sublimate
(Dye et al. 1999). The tissue was then mounted onto aluminum studs and gold coated. The maculae were scanned and
photographed using either a JEOL TSM-3000 or a Hitachi S800 scanning
electron microscope (15 KV). Measurement comparisons between freshly
fixed utricular maculae and processed tissues showed that an average
34% shrinkage occurred in our SEM-prepared tissues.
Afferent reconstruction
The location of the BDA injection site for each animal and the
sectioned tissue from each utriclar macula was examined using video
microscopy on a Nikon E600 microscope with DIC infinity optics and
drawn using an image analysis and reconstruction program (Neurolucida,
MicroBrightfield). There was no correction for tissue shrinkage, which
has generally been shown to be between 5 and 10% for aldehyde fixation
and plastic embedding (Kushida 1962). Several parameters
for each section were measured, including the section width, the
relative location and width of the striola region, the type II band,
and the morphological reversal point. For each pigeon utricular macula,
a planar map of the epithelial surface was plotted using the measured values.
For reconstruction of labeled afferents, only fibers that were darkly stained and sufficiently isolated from other afferents were traced in an effort to establish the complete morphological structure of the fiber. Afferents with partial staining (ghost fibers), or those that overlapped other afferents were discarded from analysis. In practice, many more BDA-stained fibers were observed in each utricular macula than could be quantitatively analyzed as was true for the anterior striola region shown in Fig. 3 where numerous labeled afferents overlapped and could not be discriminated with adequate certainty. The regional location of each reconstructed afferent within the utricular macula was made, using relevant distance measurements (anterior, medial, lateral edges, and reversal line). The three-dimensional reconstruction of the identified afferents was performed using ×60 (Infinity Optics, NA 0.95) magnification. A number of morphological parameters for each of the reconstructed afferents were quantified. Axonal diameter was defined as the average diameter of the last 5 µm of fiber prior to the point where the neuroepithelium was penetrated. Fiber length and volume were calculated as the total values of the all branches of the fiber within the neuroepithelium (above the stroma). Branch number indicated the total number of branches. The number of boutons was defined as the number of both terminal boutons and en passant boutons. The number of type I hair cells included all type I cells contained in calyceal terminals, as identified by visualization of an apical neck and top. The innervation area of the afferent was measured by drawing a contour around the reconstructed surface terminal field, then obtaining the area of the contour with the three-dimensional reconstruction software. The innervation density for an afferent was derived by dividing the total number of terminals (type I hair cells + bouton terminals) by the innervation area. The innervation angle was also calculated from the reconstructed dorsal view (clockwise rotation). A line was drawn through the major innervation axis of the terminal field. The innervation angle corresponded to the angle between the innervation axis and an imaginary axis line drawn through the utricular neuroepithelium from the anterior border to the posterior border. Thus 0° corresponded to a vector directed from the anterior border to the posterior border of the macula and 90° corresponded to an angle directed from lateral to medial.
Statistical analyses
The afferents were divided into three main groups based on their innervation patterns. Calyx, dimorph, and bouton afferents were distinguished as described in the following text. Comparisons between a number of morphological parameters, both within and between these groups of the traced afferents, were made using ANOVA. All post hoc comparisons were made using the Sheffè follow-up test. Proportional relationships were analyzed using Pearson product-moment correlations. All statistical comparisons were performed using Statistica (Statsoft).
| |
RESULTS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Neuroepithelium of the utricle
The general morphology of the utricular macula was examined in 14 specimens using SEM views of the dorsal surface epithelium. In vivo,
the macula is curved, with the anterior quarter of the neuroepithelium
turning dorsally at nearly a right angle to the remaining macular
surface. When processed for SEM, the epithelium was laid flat on the
mounting stud allowing the anterior edge to be visualized but also
producing stress fractures in the surface. As shown in Fig.
1, with the otoconia and otolith membrane
removed, the extent of the neuroepithelium was determined by measuring the boundaries of the hair cell distribution. At lower magnifications, the macula surface was observed to be rectangular with a rounded anterior edge (Fig. 1A). The posterior edge of the utricular
macula contained a pointed tail at the medial border, formed by an
anterior projected indentation that contained no hair cells. At
moderate magnifications (×1,000-5,000), the stereocilia of individual
hair cells were distinctly visible allowing the boundary line (Fig. 1A, 
) for the receptor neuroepithelium to be drawn. The
size of the utriclar macula in five pigeons as taken from the sectioned material (not adjusted for shrinkage) was measured across the center of
the epithelium to an average of 1.51 ± 0.22 (SD) mm along the
anterior-posterior axis and 1.19 ± 0.09 mm along the medial-lateral axis (hair cell borders). The location of the reversal line, as defined by the hair cell morphological polarization, was also
determined from the surface SEM views using moderate magnifications
(Fig. 1B). The reversal line coursed in an eccentric C-shaped manner to run from the mid posterior macula, along the epithelial margins, to the end of the medial posterior tail. The reversal line ran through the entire macula, extending to the edges of
the posterior epithelium (Fig. 1, A and E). As
shown in Fig. 1, B-D, two different morphological
polarization patterns of opposing hair cells were observed along the
reversal line. For hair cells in the lateral and anterior macula
regions, polarizations were directed toward the reversal line (Fig. 1,
B, D, and E). In contrast, hair cells located in
the medial macula had polarizations that were directed parallel to the
reversal line, with cells on either side of the line being organized in
an anti-parallel manner (Fig. 1, B and C). Hair
cells located near the medial border were polarized with their
kinocilia directed toward the posterior tail, forming a zone ~50-µm
wide (Fig. 1B). Hair cells located lateral to the
anti-parallel reversal line were polarized anteriorly. The majority of
the hair cells in the central macula were mostly polarized toward the
lateral edge but systematically varied toward the appropriate opposing
directions when located near to the reversal line (Fig. 1B).
|
To examine the regional variations in cellular architecture of the utricular macula, transverse sections through the medial-lateral axis of the neuroepithelium were cut in several animals that had no BDA label. Three major zones were demarcated in the macula based on the concentrations of type I and type II cells and their relationship to the reversal line. The striola region was defined by the high concentration of type I cells with a lower density of type II hair cells. In fact, type I hair cells were exclusively located in the striola region. The striola was located eccentrically in the lateral and anterior portions of the macula and varied in width between 150 and 400 µm (~30-40% of the epithelium). Coursing through the striola region was a narrow 30-50 µm wide band of four to eight exclusive type II cells. In the center of the type II band was the opposing polarization point that defined the reversal line. Thus the striola region actually consisted of three separate zones, including the medial and lateral type I hair cell rich areas (medial and lateral striola zones) and the type II band. Outside the striola region lay the extrastriola zone that contained only type II cells. The extrastriola formed a circumferential band around the macula, as well as occupied nearly the entire medial half of the epithelium. Together the extrastriola regions comprised ~60% of the epithelium.
Using the transverse sectioned tissue for all animals, contours were drawn of the apical and basal surfaces serially throughout the neuroepithelium using the reconstruction program. The traced sections were then rotated so that a top view of the macula was obtained for each utricle. As shown in Fig. 2 for one macula, the outline of the neuroepithelium was drawn from the top view sections. Along with the outline, the striolar region, type II band, and extrastriola were identified and plotted. This drawing then served as a schematized map of the utricular surface for all later analyses.
|
Injection sites
BDA injections into nine vestibular nuclear regions (2 animals had
bilateral injections) produced labeled afferents in the utricular
epithelium that were utilized for quantitative analyses. Seventeen
additional injections produced labeled afferents innervating other
receptor organs and were included as part of additional studies.
Injection sites centered in the more lateral portions of the vestibular
nuclei primarily produced tracing in afferent fibers that innervated
the utricular and saccular maculae, while medial injection sites
primarily labeled canal afferents. Regional differences in the density
of utricular labeled afferents were noted as the position of the
injection site varied within the vestibular nuclear complex. Among the
seven animals, utricular afferents were predominantly labeled with
injections sites located in the lateral portions of the SVN
(n = 4) and lateral vestibular nucleus (LVN;
n = 5). For example, as shown in Fig.
3, an ipsilateral injection into the
lateral portion of the rostral superior vestibular nucleus (SVN; Fig.
3A) of one animal (PN63) produced 138 labeled utricular
fibers (Fig. 3B). Although labeled fibers were observed throughout the macula surface, most were located in the striola region
of the right utricle with a dense concentration of fibers in the
anterior quadrant. These findings are consistent with our previous
anterograde tracing study in pigeons that demonstrated the central
terminal fields of afferents from each individual vestibular receptor
organ (Dickman and Fang 1996).
|
BDA uptake and axon diameter
As reported in our companion study of pigeon saccule innervation
(Zakir et al. 2003), the degree of uptake by the BDA
tracer into afferents of various size was determined by examining three branches of the vestibular nerve distal to Scarpa's ganglion in two of
the animals used for the utricle studies. As shown in Fig. 4A, the BDA tracer (dark
fibers) was absorbed and transported by axons of various diameters. The
internal diameter (lumens, not measuring the myelin) of all axons in
the three branches were measured (×100 oil), and the BDA containing
fibers were compared against the total population of afferents (Fig.
4B). In our sample of 2,351 total fibers in three branches,
axons varied between 0.54 and 8.2 µm, with a mean value of 2.54 ± 1.6 µm. BDA-filled fibers (n = 533) ranged between
0.56 and 6.99 µm with a mean of 2.4 ± 1.7 µm, which was
actually smaller than the mean 2.6 ± 1.6 µm for the non-BDA
fibers [n = 1818; F(1,2349) = 7.4, P < 0.006]. As shown in Fig. 4B, the
distribution of BDA-filled fibers and the total population of fibers
were nearly equivalent, suggesting that little or no bias in BDA uptake
by axons of either small or large diameter was present in our
experiments. The vestibular afferent axon diameters observed in our
sample agree well with those reported by Landolt et al.
(1973) for pigeons, who found a mean of 2.7 µm for
8,720 fibers in all branches (exclusive of posterior canal) and a
similar distribution of fiber diameters.
|
In addition, the diameters of all BDA-labeled fibers (n = 290 from 2 utricles; average of the last 5 µm of fiber
beneath basement membrane) at the point where they entered the
utricular neuroepithelia in these same animals were obtained. The
distribution of the BDA-labeled afferents was plotted against that for
the total fiber population in Fig. 4B. The results show that
fibers increase in diameter as they approach the basement membrane, as
the percentage of the 1- to 2-µm diam fibers decreased and the 2- to
3-µm diam fibers nearly doubled that observed in the distal nerve
branches of the same animals (
2 = 51.2, P < 0.00001). Still, many of the smallest diameter
fibers (0.5-1 µm) were also found to contain BDA at the level of the neuroepithelial crossing.
Afferent innervation patterns
In nine utricular maculae, the innervation patterns of 288 afferents were identified, with their locations in the macula being shown in Fig. 2. Of these 288 fibers, 208 met our criteria for quantitative analyses and were anatomically reconstructed. Three major classes of afferent innervation patterns were observed, including calyx, dimorph, and bouton fibers based on their terminal profile. For the majority of the reconstructed afferents, top view photomicrographs of the whole-mount macula were obtained prior to sectioning to aid in visualization of structure.
CALYX AFFERENTS. The calyx afferents were characterized by their specialized large calyceal terminal that contained several type I hair cells, as shown in Fig. 5. All of the calyx afferents (n = 82) observed were positioned within the striola region. Seventy-one calyx afferents were reconstructed for quantitative comparison and their locations are shown in Fig. 2. As represented by the four afferents in Fig. 5, a narrow continuum of innervation size and complexity was observed. For all calyx afferents, a single unbranched parent axon entered the neuroepithelium and was typically large, ranging between 1.0 and 6.6 µm (Fig. 4) with a mean diameter of 3.0 ± 1.0 µm (Table 1). On entering the neuroepithelium, >90% of the calyx afferents terminated with no branch point (Fig. 5, B, I, and K). Thus the calyx fibers exhibited the simplest arborization patterns of all afferents (Table 1). Some (7/71) calyx afferents did contain short daughter branches (Fig. 5, E and F); however, there was only a single calyceal terminal per branch. Corresponding to small arborization patterns, both the length and the volume of the calyx axons within the neuroepithelium were the smallest of all afferents (Table 1).
|
|
|
|
|
|
DIMORPH AFFERENTS. The dimorph afferents were characterized by their terminal structure that contained both calyceal and bouton terminals that innervated both type I and II hair cells. Eighty dimorph afferents were identified, and 70 of these were reconstructed for quantification (Fig. 2). Most (95%) of the dimorph afferents observed were located in the striola region, intermixed with the calyx afferents. A few dimorphs (5%) were observed to enter the neuroepithelium outside the striola zone, where fiber branches arborized to innervate a number of type II afferents, then continued into the striolar zone where calyceal terminals developed. A few dimorph afferents also had terminal branch fields that innervated the type II band, while none were located in the peripheral extrastriolar zones.
The dimorph afferents varied greatly in innervation structure, as shown in Fig. 9. Similar to the calyx afferents, dimorph afferents typically consisted of a single axon that entered the neuroepithelium, although 5.7% (4/70) of the afferents branched once prior to penetrating the base layer. The diameters of the dimorph parent axons were similar to calyx afferents in both their range distribution (Fig. 4) and size (Tables 1 and 2). Simple dimorph afferents contained a single branch off the parent axon, while the most complex afferent quantified contained 119 branch fibers (out to the 25th order). The mean number of branches/fiber for the dimorph afferents was 23.3 (Table 1), which was significantly higher than calyx afferents (Table 2). The simplest innervation pattern observed consisted of afferents with a single calyx and one bouton terminal, while the most complex afferent quantified contained five calyces and 129 bouton terminals. One feature sometimes noted on simple dimorphs was a small fiber branch that appeared to extend off of a calyceal terminal body, instead of the parent axon or another fiber branch (Fig. 9, G-I). These small branch extensions typically contained several en passant boutons and finished with one, or occasionally a cluster of several, terminal boutons. As the dimorph innervation patterns increased in complexity, the number of calyces, their locations in the arborization field, and the number of terminal bouton branches also increased. Representative examples of the diversity of dimorph afferent structures observed are shown in the reconstructions of Fig. 6. Due to the longer parent axons and the more extensive arborizations of dimorphs, the total fiber lengths and fiber volumes were significantly larger than those of calyx afferents (Tables 1 and 2). Similar to the innervation patterns of the calyx afferents, the structure of the terminal fields observed for dimorph afferents could generally be classified into flower and linear profile types. The flower profile consisted of a central calyx, or group of calyces, with small branch fibers ending in terminal boutons extending away from the calyx core (Figs. 6, F-G, and I, and 9, A-C and G-I). For some afferents, the extending bouton terminal branches surrounded the calyceal core, yet for other afferents, the extensions were unilateral. Many of the small fiber branches extended upward away from the basement membrane, giving rise to en passant and terminal boutons that often extended to within 1 µm of the apical surface of the epithelium (Figs. 6, H and I, and 9, B, C, K, and L). Dimorph afferents with flower profiles appeared to be densely innervating hair cells that lay in close approximation to each other. Dimorph afferents with linear profiles were characterized by a long parent axon that coursed for tens of microns through the neuroepithelium with few, if any, major branch fibers (Figs. 9, D-F, J, and L, and 6, H and J). In dimorphs, one to five small calyceal terminals were attached to the parent axon, and these could be located anywhere along the fiber tree but were most often positioned at either the beginning (Figs. 6H and 9, D-F) or the end (Fig. 9, J-L) of the terminal field.
|
|
BOUTON AFFERENTS. The innervation patterns of bouton afferents were characterized by en passant and bouton terminals that exclusively terminated on type II hair cells. These afferents (n = 126) were located in the extrastriolar regions and the type II band (Fig. 2). Sixty seven of the 126 identified bouton afferents were reconstructed for quantitative comparisons, and their locations are shown in Fig. 2. Bouton afferents generally entered the macula without branching, although 6% (4/67) of the fibers had one branch beneath the neuroepithelium basement membrane. The diameters of the bouton afferents were significantly smaller (Fig. 4) than those of the calyx and dimorphic afferents, with a mean value of 2.4 µm (Table 1). After entering the neuroepithelium, bouton afferents varied greatly in the size and complexity of their arborization patterns, as shown in Figs. 6 and 11. Similar to the other two groups of afferents, the bouton fibers could be classified into one of two innervation profile types. Flower profiles consisted primarily of radiating branched patterns extending from the parent axon entering the basement membrane (Figs. 6, K and M-O, and 11, A-H and M-P). Linear profiles were comprised of branching patterns that extended along a narrow width of macula surface (Figs. 6, L and P, and 11, I-L). The simplest afferent quantified contained only three branch fibers and a few bouton terminals, whereas the most complex afferent had 150 branches out to the 24th order. With a mean number of 54.4 branches per fiber, the bouton afferents were significantly more arborous than either the dimorph or calyx afferents (Tables 1 and 2). Measures of the overall fiber length for the bouton afferents were compared against the two other classes of fibers according to regional location in the utricular macula, as shown in Fig. 12. As noted in the preceding text, the calyx units had very short fiber lengths, the dimorph afferents had larger fiber patterns, but the bouton afferents had the largest terminal structures of all (Table 2).
|
|
Regional differences in innervation patterns
To more closely examine the regional differences in morphological properties of the three classes of afferents, differences for the reconstructed afferents were analyzed according to afferent location. First, all afferents were plotted for terminal fiber length and innervation area as a function macula location, as shown in Fig. 12. The utricular epithelium was then divided into anterior, central, and posterior thirds for comparisons. Examination of the regional distribution of calyx, dimorph, and bouton afferents shows that some differences in the patterns across the macula surface exist (Figs. 10 and 12). For example, innervation areas were significantly larger for anterior calyx afferents (377 µm2), as compared with either centrally (286 µm2) or posteriorly (227µm2) located fibers [F(2,68) = 3.6, P < 0.05]. Conversely, many dimorph and bouton afferents with the largest innervation areas were located in the posterior regions of the macula (Fig. 12), although no overall significant difference for the three regional locations was found. For dimorph afferents, there was a moderate trend for more type I hair cells per fiber in the posterior third of the macula (6.5), as compared with either the central (4.4) or anterior (4.7) fibers [F(2,67) = 2.98, P < 0.06]. No other comparisons for regional location along the anterior-posterior axis were significant. Comparisons between calyx and dimorph afferents located on the medial and lateral side of the reversal line in the striola were also made, with no significant differences found. These results indicate that the morphology of the calyx and dimorph afferents are quite homogeneous throughout the striolar region.
For the bouton afferents, examination of the distribution plots (Fig. 12) suggests that fibers innervating the more central regions of the macula may differ from those innervating the more peripheral utricular regions. To test the observation, all of the bouton afferents that were located within 100 µm of the peripheral border were grouped (n = 28) and compared with those that were centrally located (n = 39) on several morphological measures. As shown in Table 3, peripherally located bouton afferents had larger diameter axons, higher branch orders, longer fibers, more bouton terminals, and larger innervation areas than central bouton afferents. Finally, it was of interest to examine the bouton afferents that innervated the type II band, as they represent a terminal field in a unique portion of the macula. In the distribution plots of Fig. 12, it appears that these afferents had relatively small terminal fields. Although only 16 type II band bouton afferents were quantified, statistical comparisons showed that these fibers had smaller fiber lengths, fewer bouton terminals and smaller innervation areas that afferents located elsewhere in the macula (Table 3).
|
Efferent fibers of the utricle
In the present study, central BDA injections into the vestibular
nuclei were utilized to retrogradely label vestibular afferents. Indeed, labeled cell bodies in Scarpa's ganglion were observed for all
animals. However, with central BDA injections it is also possible that
some of the labeled fibers and terminals in the utricular
neuroepithelium were anterogradely labeled efferent fibers.
Vestibular efferents are known to have their cell bodies located in the
brain stem, ventral and slightly lateral to the abducens nucleus in
birds (Dickman and Fang 1996; Eden and Correia 1982). Thus for each animal utilized in the study, the efferent nuclei were bilaterally examined for BDA labeled cell bodies. In two
birds, several efferent nuclei neurons were in fact labeled, and these
animals were eliminated from further study.
As a further control, two animals were given large BDA injections into
the contralateral brain stem in and surrounding the region that
contains efferent somas. The contralateral nucleus was chosen since in
all species studied to date, no vestibular primary afferent has ever
been shown to cross the midline using either nerve branch or
intracellular tracing experiments (Carleton and Carpenter
1984; Dickman and Fang 1996; Gacek
1969). For the two control animals, a number of known efferent
fibers were obtained and their innervation patterns were examined. As
shown in Fig. 13, the efferent fibers
exhibited innervation patterns that were very different from afferent
fibers and were readily distinguishable by visual inspection. For
example, the efferent fibers typically branched several times beneath
the neuroepithelium and had thin parent axons. Although only three
efferent fibers were quantified, the largest axonal diameter of the
three was 0.78 µm as compared with the smallest afferent diameter
(from the 208 units) of 0.86 µm. On penetrating the basement
membrane, the parent axon for the efferents extended several hundred
microns, much longer than any of the afferents observed. The shortest
axonal length of the efferent fibers was 2,474 µm as compared with
1,306 µm for the longest afferent quantified. Numerous very fine
branch fibers extended into the epithelium with a high density of en
passant and terminal boutons (Fig. 13), and no calyceal terminals were ever observed. The number of terminals on the smallest of the three
efferents were 377 as compared with 182 terminals on the largest
afferent (bouton) quantified. Although more numerous, the efferent
terminals were noticeably smaller than those typically observed for
utricular afferents. Consistent with the longer fiber and more numerous
branches for efferents, the innervation area of 15,405 µm2 for the smallest unit was nearly twice the
largest afferent quantified with an innervation area of 7,085 µm2. These characteristics for pigeon
vestibular efferents were consistent with the findings obtained for
semicircular canal efferents in gerbils (Purcell and Perachio
1997).
|
| |
DISCUSSION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
The present results describing the morphology and innervation
patterns of the utricular macula in pigeons clearly show that there are
a number of similarities with other amniote species and yet a number of
interesting differences. Our study sought to quantitatively examine
morphological properties of afferents that innervate the utricle so
that a more complete understanding of afferent structure and terminal
field patterns could be obtained for birds, as compared with the
otolith organs in other animal classes. To that end, it is unfortunate
that very few species have been thoroughly examined, with the major
works regarding utricular innervation being reported only for
chinchillas (Fernandez et al. 1990) and frogs
(Baird and Lewis 1986; Baird and Schuff 1994).
Morphology of the pigeon utricular macula
The utricular receptor epithelium in adult pigeons is quite large
for an animal of its size. With average dimensions of 1.5 × 1.2 mm, the pigeon utricular macula appears to be comparably larger than
that of several other animals studied including frogs (Baird and
Schuff 1994), turtles (Fontilla and Peterson
2000), chinchillas (Fernandez et al. 1990),
guinea pigs (Werner 1933), and three species of fish
(Chang et al. 1992; Platt 1975;
Popper and Northcutt 1983). If one compares across
animal classes, several morphological features are present in the
pigeon utricular macula that may represent general adaptations refined
through parallel evolutionary paths for motion detection in
terrestrial/aerial environments. For example, as classically described,
type I hair cells are present in reptiles, birds, and mammals
(Jørgensen 1970, 1974; Vinnikov et al.
1965; Wersäll 1956; Wersäll
and Bagger-Sjöbäck 1974). In adult amniotes, type I
and type II hair cells can be differentiated by a number of
morphological (Kevetter et al. 1994; Lysakowski
and Goldberg 1997; Ricci et al. 1996, 1997;
Rüsch et al. 1998) and physiological
characteristics (Correia and Lang 1990; Eatock et
al. 1994; Rennie and Correia 1994;
Rüsch and Eatock 1996). Perhaps two of the most
defining are the presence of a distinct calyceal terminal
(Wersäll 1956) and a signature delayed rectifier
potassium conductance (gKI)
(Correia and Lang 1990) for type I hair cells that are
not present in type II hair cells. In fish and amphibians, it has been
classically thought that only type II hair cells exist
(Wersäll and Bagger-Sjöbäck 1974;
Wersäll et al. 1965); however, "type I hair
cell-like" morphological features have been described for cells in
fish striola (Chang et al. 1992; Lanford and
Popper 1996). To date, neither signature
gKI currents (Steinacker and
Romero 1992; Steinacker et al. 1997) nor true
calyceal terminals have been identified in either fish or amphibians,
so functional analogues of type I cells in these classes remain in
question. Still, in both frogs and fish, heterogeneity in receptor
cells, cell density, kinocilia formation and macula surface extension
have been utilized to define separate macular regions including a
striola, juxtastriola, and extrastriola (Baird and Schuff
1994; Chang et al. 1992). Based on the
distribution of type I and type II receptor cells, the pigeon utriclar
macula can be divided into two major organizational regions, including
the striola and extrastriola, with no clear juxtastriola present. In
fact, in birds type I hair cells appear to be exclusively located in
the striola. In mammals, type I hair cells are located with the highest
density (2:1) in the striola but are also found throughout the macula
at lower densities (Fernandez et al. 1990). As suggested
by Goldberg and Brichta (1998) for the cristae
ampullares, evolution appears to favor the wider distribution of type I
hair cells and calyceal terminals throughout the receptor epithelium of
the otolith organs as animals adapted from an aquatic to terrestrial
environment. Of particular interest to this point, is the observation
that in bird saccular maculae, type I hair cells and calyceal terminals
are not confined to the striola but populate nearly the entire
epithelium (Jørgensen and Andersen 1973;
Rosenhall 1970; Zakir et al. 2003).
Apparently unique to birds, is the type II band of cells that sandwich
the reversal line in the striola region (Jørgensen
1970; Rosenhall 1970). Interestingly, the type
II band is present in the utricular and lagenar maculae but not in the
saccular macula (Jørgensen and Andersen 1973;
Zakir et al. 2003).
There are some further cytoarchitecture regional differences among the
various animal classes. For example, similar to frog and fish utricular
maculae, the pigeon striola eccentrically traverses the lateral portion
of the epithelium with a narrow lateral extrastriola region
(Baird and Schuff 1994, Chang et al.
1992; Platt 1975; Popper and Northcutt
1983), whereas in mammals, the striola is more centrally
located (Fernandez et al. 1990). In the bird utriclar macula, the reversal line of opposing polarizations extends along the
outer margin of much of the macula, similar to fish and amphibians (Baird and Schuff 1994; Lewis and Li
1975; Platt 1975; Wersäll et al.
1965). Although most of the polarizations were perpendicular for the lateral and anterior macular regions, the antiparallel polarizations of the medial macula reversal line appear unique to birds
and to our knowledge have not been previously observed. In mammals, the
reversal line is ce
indicating direction of hair cell morphological
polarization. C-E: Higher magnification of
insets in A. - - -, antiparallel reversal line
(C). · · · , perpendicular reversal
line (D and E). Scale bars = 10 µm.
, location of injection site in right
brain stem; 
, tracer halo. Serial sections (60 µm)
are arranged from rostral to caudal order and consecutively numbered.
L, locus coeruleus; LV, lateral vestibular nucleus; M, medial
vestibular nucleus; MLF, medial longitudinal fasciculus; PM, paramedial
nucleus; PT, principal trigeminal nucleus; SV, superior vestibular
nucleus; T, tangential nucleus; V, dorsal trigeminal nucleus; VI,
abducens nucleus; VII, facial nucleus; VIIn, facial nerve; VIII,
Scarpa's ganglion and VIIIth nerve root. B: a dorsal view
of the whole right utricle with retrogradely labeled afferents produced
from injection site in A. Striola region is outlined (
), and bouton
(
) type afferents are visible (not all are labeled for
clarity). Anterior, rightward; medial, upward.
, n = 2351, BDA and non-BDA
filled), and BDA filled at base of utricular epithelia in animal PN76
(
, n = 290) are shown. Line plots
for calyx (
), and bouton
(
) afferents as a function of innervation area.
Each bar represents the number of afferents per group for each of the
binned innervation areas.
