close

Вход

Забыли?

вход по аккаунту

?

617

код для вставкиСкачать
MICROSCOPY RESEARCH AND TECHNIQUE 46:265–280 (1999)
Peptidergic Peripheral Nervous Systems in the Mammalian
Pineal Gland
SHOJI MATSUSHIMA,* YUKO SAKAI, AND YOSHIKI HIRA
Department of Anatomy, Asahikawa Medical College, Nishikagura, Asahikawa, 078-8510, Japan
KEY WORDS
synapse; immunohistochemistry; sympathetic nerve fiber; superior cervical ganglion
ABSTRACT
The distribution and density of tyrosine hydroxylase (TH) and neuropeptide Y
(NPY)-immunoreactive, sympathetic fibers and calcitonin gene-related peptide (CGRP)-, substance
P (SP)-, and vasoactive intestinal polypeptide (VIP)-immunoreactive, non-sympathetic fibers in the
pineal gland, the effects of superior cervical ganglionectomy (SCGX) on these fibers, and the location
of their terminals in the pineal gland were compared between rodents and non-rodents. A dense
network of TH/NPY-positive fibers is present all over the pineal gland. A less dense network of
CGRP/SP- or VIP-positive fibers occurs in the whole pineal gland of non-rodents, but these fibers are
usually confined to the superficial pineal gland in rodents. After SCGX, some TH/NPY-fibers remain
only in the deep pineal gland in rodents, whereas considerable numbers of these fibers persist
throughout the gland in non-rodents. Thus, the remaining fibers, probably originating from the
brain, may be more numerous in non-rodents. Since CGRP-, SP- or VIP-immunoreactive fibers in
the pineal capsule can be traced to those in the gland, and since these fibers are ensheathed by
Schwann cells, it is concluded that these fibers belong to the peripheral nervous system. However,
the existence of SP-positive central fibers cannot be denied in some species. In the superficial pineal
gland of rodents, sympathetic terminals are mostly localized in perivascular spaces, whereas the
parenchymal innervation by sympathetic fibers in the pineal gland is more dense in non-rodents
than in rodents. Synapses between sympathetic nerve terminals and pinealocytes occur occasionally
in non-rodents, but only rarely in the superficial pineal gland of rodents. The occurrence of the
synapses may depend on the frequency of intraparenchymal sympathetic terminals. Microsc. Res.
Tech. 46:265–280, 1999. r 1999 Wiley-Liss, Inc.
INTRODUCTION
Until recently, it has been considered that the pineal
gland of mammals is innervated by the three groups of
nerve fibers, i.e., sympathetic, parasympathetic, and
commissural central fibers (Vollrath, 1981). Sympathetic fibers have been studied most extensively among
these pinealopetal fibers, and it has been established
that these fibers are commonly found in the mammalian species, and are derived from the superior cervical
ganglia (Bowers et al., 1984; Kappers, 1960; Vollrath,
1981). However, the origin and nature of commissural
fibers have not clearly been understood, and the available information on the parasympathetic innervation
has been fragmentary. During the last 20 years, our
knowledge of the innervation of the mammalian pineal
gland has increased enormously by the use of immunohistochemical techniques. As a result, it has been
found, in the pineal gland of many mammals, that
sympathetic fibers contain neuropeptide Y (NPY)
(Møller et al., 1996; Schon et al., 1985), and that
non-sympathetic fibers immunoreactive for a variety of
peptides including calcitonin gene-related peptide
(CGRP) (Reuss, 1996; Shiotani et al., 1986), substance
P (SP) (Reuss, 1996; Rønnekleiv and Kelly, 1984; Shiotani et al., 1986), and vasoactive intestinal polypeptide
(VIP) (Møller et al., 1996; Uddman et al., 1980) are
present. The existence of choline acetyltransferaseimmunoreactive fibers is also described in the pineal
r 1999 WILEY-LISS, INC.
gland of the cow (Phansuwan-Pujito et al., 1991). Such
non-sympathetic fibers belong to either the peripheral
or central nervous system (Møller, 1992; Møller et al.,
1996; Reuss, 1996). CGRP/SP- and VIP-positive fibers
in the pineal gland of gerbils have been demonstrated,
using retrograde tracings combined with immunohistochemistry, to originate from trigeminal and pterygopalatine ganglia, respectively (Shiotani et al., 1986). In
addition, the data concerning possible sites of origin of
peptidergic central fibers are gradually emerging
(Møller, 1992; Reuss, 1996).
Although the determination of the distribution and
density of sympathetic and non-sympathetic fibers in
various regions of the pineal gland may be of fundamental importance, such studies are rather scanty. For
example, the distribution of sympathetic and nonsympathetic fibers in the three regions of the rodent
pineal gland, i.e., the superficial component lying directly underneath the skull, the deep component found
adjacent to the habenular and posterior commissures
and the third ventricle, and the stalk connecting the
This paper is dedicated to Professor Russel J. Reiter who advised one of the
authors (S.M.) to initiate the comparative anatomical study of the mammalian
pineal gland.
*Correspondence to: S. Matsushima, Department of Anatomy, Asahikawa
Medical College, Nishikagura, Asahikawa, 078-8510, Japan.
Accepted 3 March 1999
266
S. MATSUSHIMA ET AL.
TABLE 1. Distribution and density of TH-immunoreactive nerve fibers in the pineal gland and effects of superior cervical ganglionectomy
(SCGX) in rodent (A) and non-rodent (B) species1
Distribution and density
Species (A)
Rat
Rat
Guinea-pig
Syrian hamster
Gerbil
Cotton rat
European hamster
Mouse
Chinese hamster
Syrian hamster
Effect of SCGX
Superficial
Stalk
Deep
Superficial
Stalk
Deep
Reference
⫹⫹⫹
⫹⫹⫹
⫹⫹⫹
⫹⫹⫹
⫹⫹⫹
⫹⫹⫹
⫹⫹⫹
⫹⫹⫹
⫹⫹⫹
⫹⫹⫹
⫹⫹⫹
⫹⫹⫹
⫹
⫹⫹⫹
⫹⫹⫹
(⫺)
(⫺)
(⫾ or ⫹)
(⫾)
(⫾ or ⫹)
(⫾)
⫹⫹⫹
⫹⫹⫹
⫹⫹⫹
⫹⫹⫹
⫹⫹⫹
⫹⫹⫹
⫹⫹⫹
(⫺)
(⫺)
(⫺)
(⫺)
(⫾)
(⫺)
(⫾)
(⫺)
(⫺)
(⫾)
(⫺)
(⫾)
(⫾)
Zhang et al. (1991)
Matsuura et al. (1994)
Romeo et al. (1993)
Li and Welsh (1991)
Li and Welsh (1991)
Matsushima et al. (1994)
Møller et al. (1998)
Matsushima et al. (unpublished data)
Matsushima et al. (unpublished data)
Matsushima et al. (unpublished data)
⫹⫹⫹
⫹⫹⫹
⫹⫹⫹
Species (B)
Distribution and density
Effect of SCGX
Sheep
Pig
⫹⫹⫹
⫹ (proximal and central),
⫹⫹⫹ (distal and peripheral)
⫹⫹⫹
(⫾)
Cozzi et al. (1992)
Kaleczyc et al. (1994)
(⫾)
Matsushima et al. (unpublished data)
Tree shrew
Reference
listed below were used throughout tables. ⫹⫹⫹ high; ⫹⫹ moderate; ⫹ low; ⫺ no immunoreactive fiber; (⫺) disappeared; (⫹) unchanged; (⫾) decreased but
remained.
1Symbols
two components, has not until very recently been
examined. Furthermore, comparisons of the distribution of pinealopetal fibers have not so extensively been
made between intact animals and those subjected to
superior cervical ganglionectomy. This comparison is
essential for the study of the pineal innervation for the
following reasons: (1) tyrosine hydroxylase (TH) and
NPY are found to be contained also in some nonsympathetic fibers in the pineal gland of several mammals (Cozzi et al., 1992; Matsuura et al., 1994; Møller et
al., 1990; Møller et al., 1998; Møller et al., 1994; Zhang
et al., 1991); and (2) neuronal cell bodies in the superior
cervical ganglia are reported to contain peptides, such
as CGRP, SP, and VIP (Ariano and Kenny, 1985; Baffi et
al., 1992; Kummer and Heym, 1988; Lundberg et al.,
1979; Sasek and Zigmond, 1989; Schmitt et al., 1988)
and, thus, it is possible that these peptides are present
in pineal sympathetic fibers. In addition, immunohistochemical observations on the innervation of the pineal
gland of Syrian and Chinese hamsters and mice have
scarcely been made, the available data on the pineal
innervation in non-rodents are much fewer than those
in rodents, and immunoelectron microscopic techniques have hardly been employed for the study of the
innervation of the mammalian pineal gland. For a
number of reasons mentioned above, we selected some
rodent (Syrian and Chinese hamsters, mice, and cotton
rats) and one non-rodent (tree shrews, Tupaia glis)
species, and the immunohistochemical and immunocytochemical data obtained from these animals, whose
superior cervical ganglia were left intact or removed,
are also considered in this paper. Diagrams and immunoelectron micrographs were specially prepared for this
review.
In this review, we attempted to compare the distribution and density of TH/NPY-immunoreactive, sympathetic fibers and CGRP/SP-, and VIP-immunoreactive,
non-sympathetic, fibers in the pineal gland in rodents
with those in non-rodents. In rodents, central fibers are
confined to the stalk and the deep pineal gland corresponding to a minor division of the gland, and it is
generally believed that these fibers are smaller in
number in rodents than in non-rodents (Møller, 1992).
Thus, comparative studies on the distribution of sympathetic and non-sympathetic fibers in rodents and nonrodents appear to be important for understanding their
significance in the mammalian pineal gland.
TH-IMMUNOREACTIVE NERVE FIBER
The distribution of TH-immunoreactive fibers in various regions of the pineal gland has been examined in
several rodent species (Table 1). In all animals, except
guinea-pigs, TH-immunoreactive fibers form a dense
network everywhere throughout the gland. In the
guinea-pig, the density of TH-fibers is high in the
superficial pineal gland, but these fibers are scarce in
the stalk (Romeo et al., 1993). Our observations on the
Syrian hamster (Fig. 1; Table 1) agree with the data
obtained by Li and Welsh (1991) (Table 1), and THimmunoreactive fibers were densely distributed in all
regions of the gland, also in mice and Chinese hamsters
(Figs. 2,3A; Table 1).
The effects of superior cervical ganglionectomy have
been reported to differ considerably among animals
(Table 1). Following ganglionectomy, TH-immunoreactive fibers almost disappear from the superficial pineal
gland but persist in the stalk and the deep pineal gland
in rats (Matsuura et al., 1994; Zhang et al., 1991) and
European hamsters (Møller et al., 1998), whereas THpositive fibers are entirely abolished in cotton rats
(Matsushima et al., 1994). Our results of the denervation studies also varied according to the species (Table 1).
TH fibers were abolished in the entire gland in mice
(Table 1). These fibers were nearly lost in the superficial
pineal gland, but persisted in the deep pineal gland in
Chinese and Syrian hamsters (Table 1). Taking the
above data together, it is assumed that the rodent
pineal gland is innervated by two kinds of THimmunoreactive fibers: postganglionic sympathetic fibers distributed throughout the entire gland, and extrasympathetic fibers localized in the stalk and the deep
pineal gland. The latter fibers are supposed to be
derived from the brain (Matsuura et al., 1994; Zhang et
al., 1991). In the Syrian (Fig. 1) and Chinese (Fig. 3)
PERIPHERAL INNERVATION OF THE MAMMALIAN PINEAL
267
Fig. 1. TH-, NPY-, CGRP-, and VIP-immunoreactive fibers in the
superficial and deep pineal glands, and the habenular (HC) and
posterior (PC) commissures of unoperated Syrian hamsters in the
parasagittal plane. Profiles of the deep pineal gland were depicted
under those of the superficial pineal gland. Black and white arrows
indicate immunolabeled fibers seen in the capsule surrounding the
distal region of the superficial pineal gland and the connecting stalk,
respectively. SO, subcommissural organ.
hamsters having a pineal complex where its superficial
and deep components are separated, two bundles of
TH-positive fibers are found to enter the distal region of
both the superficial and deep components. One bundle
entering the distal end of the superficial component
corresponds to the nervi conarii and the other may
participate in the formation of the so-called connecting
stalk intervening between the two components (Gregorek et al., 1977). Judging from their disappearance
following ganglionectomy (Figs. 3B, 4), both bundles
may consist of postganglionic sympathetic fibers. As
indicated in Figures 1 and 3, TH-immunoreactive fibers
are present in the habenular and posterior commissures in the Chinese and Syrian hamsters. Some THpositive fibers that remain in the deep pineal gland
after ganglionectomy are continuous with commissural
fibers immunoreactive for TH (Fig. 4). These observations suggest that TH-positive fibers enter the deep
pineal gland also by way of the commissures in these
animals.
In the Syrian hamster, TH-immunoreactive fibers
drastically decrease in number, but a few of them
remain after superior cervical ganglionectomy (Fig. 4).
In the pineal complex of ganglionectomized Syrian
hamsters, immunoreactive profiles generally appear as
small dots in the superficial portion, unlike in the deep
portion, where those apparently have the shape of
nerve fibers (Fig. 4). Such TH-positive structures as
seen in the superficial pineal gland may represent
processes of TH-positive cell bodies (Jin et al., 1988).
In non-rodents, the distribution of TH-positive fibers
has been described in sheep and pigs (Table 1). THfibers are distributed densely and evenly throughout
the pineal gland in sheep (Cozzi et al., 1992), whereas
in pigs the distribution of these fibers exhibits regional
differences in the gland. The density of TH-fibers is
higher in the distal and peripheral regions than in the
proximal and central regions (Kaleczyc et al., 1994).
The effects of superior cervical ganglionectomy on
TH-positive fibers have previously been examined only
in the sheep (Cozzi et al., 1992). TH-fibers decrease
markedly in number after sympathetic denervation,
but some fibers remain in the pial capsule and the
pineal gland itself. Our observations on the tree shrew
demonstrate that considerable numbers of TH-positive
fibers persist following ganglionectomy (Table 1). Since
bundles of TH-fibers entering the gland at its distal end
totally disappear in ganglionectomized tree shrews,
TH-fibers that remain in the gland are of extrasympathetic origin. Such fibers may originate from the brain
268
S. MATSUSHIMA ET AL.
Fig. 2. TH-, NPY-, CGRP-, and VIP-immunoreactive fibers in the pineal gland of unoperated mice in
the parasagittal plane. See Figure 1 legend for abbreviations and symbols.
because of the presence of TH-positive fibers in the
posterior commissure of this animal.
NPY-IMMUNOREACTIVE NERVE FIBER
The distribution and density of NPY-immunoreactive
fibers are about the same as those of TH-positive fibers
(Table 2). Thus, NPY-immunoreactive fibers are usually
distributed densely and evenly throughout the gland,
whereas in the guinea-pig these fibers are present
moderately in the superficial pineal gland and are
scarce in the stalk (Romeo et al., 1993). Our data on
mice and Syrian and Chinese hamsters correspond to
the above observations in that TH- and NPY-positive
fibers are similar in their distribution (Table 2). NPYcontaining fibers entering the distal end of the superficial or deep pineal glands are also observed in these
animals (Figs. 1–3). The response of NPY-containing
fibers to superior cervical ganglionectomy is also similar to that of TH-fibers, and thus the ganglionectomy
results in the complete disappearance of NPY-positive
fibers in cotton rats but in their persistence in the stalk
and the deep pineal gland in rats and European hamsters (Table 2). From the above data obtained from
cotton rats, it is evident that both TH and NPY are
exclusively contained in postganglionic sympathetic
fibers in the pineal gland of this animal. Indeed, NPY is
PERIPHERAL INNERVATION OF THE MAMMALIAN PINEAL
269
Fig. 3. TH- (A, B) and NPY- (C, D) immunoreactive fibers in the
superficial pineal gland of unoperated Chinese hamsters (A,C), and in
the deep pineal gland and the habenular (HC) and posterior (PC)
commissures of unoperated (A,C) and ganglionectomized (B,D) Chi-
nese hamsters in the parasagittal plane. Profiles of the superficial and
deep pineal glands in A and C are arranged in the same way as in
Figure 1. See Figure 1 legend for an additional abbreviation and
symbols.
found to coexist with TH in pineal sympathetic fibers
(Fig. 5). In ganglionectomized Chinese hamsters, NPYpositive fibers, like TH-fibers, disappear from the superficial pineal gland, its capsule, and the connecting stalk
but remain in the deep pineal gland (Fig. 3D; Table 2).
NPY-positive fibers exist in the posterior commissure in
this animal (Fig. 3C,D). Thus, NPY-containing fibers
that survive after ganglionectomy may also originate
from the brain in the Chinese hamster. The observations of adjacent sections stained with anti-TH or
270
S. MATSUSHIMA ET AL.
Fig. 4. TH-, NPY-, CGRP-, and VIP-immunoreactive fibers in the
superficial and deep pineal glands, and the habenular (HC) and
posterior (PC) commissures of ganglionectomized Syrian hamsters in
the parasagittal plane. Profiles of the superficial and deep pineal
glands are arranged in the same way as in Figure1. See Figure 1
legend for an additional abbreviation and symbols.
TABLE 2. Distribution and density of NPY-immunoreactive nerve fibers in the pineal gland and effects of superior cervical ganglionectomy
(SCGX) in rodent (A) and non-rodent (B) species1
Distribution and density
Species (A)
Rat
Guinea-pig
Syrian hamster
Gerbil
Cotton rat
European hamster
Mouse
Chinese hamster
Syrian hamster
Species (B)
Monkey
Sheep
Cow
Cat
Mink
Pig
Tree shrew
1See
Effect of SCGX
Superficial
Stalk
Deep
Superficial
Stalk
Deep
Reference
⫹⫹⫹
⫹⫹
⫹⫹⫹
⫹⫹⫹
⫹⫹⫹
⫹⫹⫹
⫹⫹⫹
⫹⫹⫹
⫹⫹⫹
⫹⫹⫹
⫹
⫹⫹⫹
(⫺)
(⫾ or ⫹)
(⫾ or ⫹)
⫹
⫹⫹⫹
⫹⫹⫹
⫹⫹⫹
⫹⫹⫹
⫹⫹⫹
⫹⫹⫹
(⫺)
Zhang et al. (1991)
Romeo et al. (1993)
Li and Welsh (1991)
Li and Welsh (1991)
Matsushima et al. (1994)
Møller et al. (1998)
Matsushima et al. (unpublished data)
Matsushima et al. (unpublished data)
Matsushima et al. (unpublished data)
⫹⫹⫹
⫹⫹⫹
⫹⫹⫹
Distribution and density
⫹⫹
⫹⫹⫹
⫹⫹⫹
⫹⫹⫹
⫹ (distal), ⫹⫹⫹ (proximal)
⫹ (proximal and central),
⫹⫹⫹ (distal and peripheral)
⫹⫹⫹
(⫺)
(⫺)
(⫾)
(⫺)
(⫾)
(⫺)
(⫺)
(⫾ or ⫹)
(⫺)
Effect of SCGX
(⫾)
(⫾)
(⫹)
(⫺)
(⫺)
(⫾ or ⫹)
(⫺)
(⫾)
(⫺)
Reference
Mikkelsen and Mick (1992)
Cozzi et al. (1992)
Phansuwan-Pujito et al. (1993)
Møller et al. (1994)
Møller et al. (1990)
Kaleczyc et al. (1994)
Matsushima et al. (unpublished data)
Table 1 for definitions.
anti-NPY indicate that TH- and NPY-immunoreactive
fibers remaining in the deep pineal gland of ganglionectomized Chinese hamsters are apparently different in
their distribution (Fig. 3B,D). This suggests that THand NPY-positive, extrasympathetic fibers originate
from different regions of the central nervous system.
PERIPHERAL INNERVATION OF THE MAMMALIAN PINEAL
Consequently, it can be said that, in many rodent
species except cotton rats, TH- or NPY-positive fibers,
probably originating from various regions of the brain,
are localized in the deep pineal gland and the stalk.
Such presumptive central fibers distributing mainly in
a small portion of the gland may be of limited significance in the rodent pineal gland.
It is noteworthy that the effects of superior cervical
ganglionectomy on NPY-containing fibers in the pineal
gland in mice and Syrian hamsters differ markedly
from those in the above-mentioned rodents. After ganglionectomy, a small to moderate number of NPYpositive fibers remain not in the deep pineal gland but
in the distal region of the superficial pineal gland (Figs.
4,6). Our immunoelectron microscopic study demonstrates the presence of NPY-containing nerve endings
in pericapillary spaces in the superficial pineal gland of
ganglionectomized mice (Fig. 7). NPY-fibers remaining
after ganglionectomy are more abundant in mice than
in Syrian hamsters. In ganglionectomized mice, a few
NPY-fibers are also observed in the capsule covering the
distal region of the superficial pineal gland (Fig. 6). Also
in non-rodents, NPY-immunoreactive fibers enter the
pineal gland at its distal end (Cozzi et al., 1992; Møller
et al., 1994; Phansuwan-Pujito et al., 1993). As will be
mentioned, some NPY-fibers remain in the capsule and
the pineal gland of the sheep after ganglionectomy
(Cozzi et al., 1992). Since NPY is known to coexist
with VIP and acetylcholine in cranial parasympathetic
neurons (Cavanagh et al., 1990; Kuwayama et al., 1988;
Leblanc et al., 1987; Suzuki et al., 1990), Cozzi et al.
(1992) and Møller et al. (1994) speculate that nonsympathetic, NPY-positive fibers originate from parasympathetic ganglia. This may also be true for mice
and Syrian hamsters. This subject will be dealt with
later in connection with a relationship between NPYand VIP-immunoreactive fibers in the pineal gland of
mice.
The distribution and density of NPY-immunoreactive
fibers have been studied in several non-rodent species
(Table 2). In many animals (sheep, Cozzi et al., 1992;
monkey, Mikkelsen and Mick, 1992; cat, Møller et al.,
1994; cow, Phansuwan-Pujito et al., 1993), NPYimmunoreactive fibers are evenly distributed moderately or heavily throughout the gland, whereas in the
mink (Møller et al., 1990) and pig (Kaleczyc et al., 1994)
NPY-fibers exhibit regional differences in their distribution; NPY-fibers are more dense in the proximal (rostral) than distal (caudal) regions in the former, and
these fibers show a distribution similar to that seen in
TH-fibers in the latter (Table 1). The effects of superior
cervical ganglionectomy have been examined in the cat,
sheep and mink (Table 2). Unlike in rodents, considerable numbers of NPY-immunoreactive fibers remain
throughout the gland after ganglionectomy, especially
in the sheep (Cozzi et al., 1992). NPY-immunoreactive
fibers that persist after ganglionectomy are believed to
derive from the central nervous system (Cozzi et al.,
1992; Mikkelsen and Mick, 1992; Møller et al., 1994;
Phansuwan-Pujito et al., 1993). From the observations
on the intrapineal distribution of extrasympathetic
fibers immunoreactive for NPY, it is believed that these
fibers play an important role in the pineal gland of
non-rodent species.
271
Unlike in rodents, pineal NPY-immunoreactive fibers
in non-rodents are encountered not only perivascularly
but also intraparenchymally (Cozzi et al., 1992; Møller
et al., 1994; Phansuwan-Pujito et al., 1993). Abundant
intraparenchymal NPY-fibers form a diffuse network
among pinealocytes in these animals. According to our
data, NPY-fibers disappear from the pineal gland of the
tree shrew after superior cervical ganglionectomy
(Table 2), suggesting that NPY-positive fibers in the
pineal gland of this animal mostly originate from this
ganglia. NPY-fibers are often observed among pinealocytes, and their endings occasionally make synaptic
contact with pinealocytes (Fig. 8).
CGRP- AND SP-IMMUNOREACTIVE
NERVE FIBER
The first systematic study on the distribution and
density of CGRP-immunoreactive fibers in various regions of the pineal gland and a course of these fibers to
the gland in mammals have been made in our laboratory (Matsushima et al., 1994). This study is based on
the observations of myelinated fibers in the pineal
gland of the cotton rat (Matsushima et al., 1991). We
found that myelinated fibers were abundantly distributed in the distal region of the pineal gland and its
capsule of this animal. Examination of serial sections
revealed that intracapsular myelinated fibers continued into the bilateral bundles of myelinated fibers,
which run under the transverse sinuses and thus
correspond to the nervi conarii, and also into intrapineal myelinated fibers. Based on the observation that
intracapsular myelinated fibers and those traced between the capsule and the parenchyma bifurcate in a
proximal direction, these fibers were assumed to be
pinealopetal fibers. Myelinated fibers in the pineal
gland of cotton rats are unusual because in many
mammals these fibers are mainly found in the proximal
region of the gland; such fibers are generally believed to
originate from the central nervous system (Korf and
Møller, 1984, 1985). Since myelinated fibers in the
pineal gland of cotton rats are known to persist following superior cervical ganglionectomy (Karasek et al.,
1983; Sakai et al., 1996), it is evident that myelinated
fibers are not derived from this ganglia. In order to
determine the nature of the myelinated fibers, we tried
to examine whether these fibers are immunoreactive
for CGRP.
On examination of serial sections, it became apparent that CGRP-immunoreactive fibers, together with
TH- or NPY-positive ones, contribute to the formation of
the nervi conarii, and that CGRP-fibers in the nervi
conarii were able to be traced to those in the distal
region of the pineal gland (Matsushima et al., 1994). In
the nervi conarii, fibers immunoreactive for TH and
NPY disappeared after superior cervical ganglionectomy, but those immunoreactive for CGRP persisted.
CGRP-immunoreactive cell bodies were not detected in
the gland (Matsushima et al., 1994). Thus, it seems
reasonable to conclude that non-sympathetic CGRPfibers, together with sympathetic fibers, enter the
pineal gland at its distal end by way of the nervi conarii.
The course of CGRP-immunoreactive fibers, as far as
their entrance into the pineal gland, was found to be
similar in some additional rodent species, i.e., mice
272
S. MATSUSHIMA ET AL.
Figs. 5, 7, 8, 10–14.
PERIPHERAL INNERVATION OF THE MAMMALIAN PINEAL
(Fig. 2), rats (Fig. 9) and Syrian (Fig. 1) and Chinese
(Matsushima et al., 1995) hamsters. Since CGRPimmunoreactive fibers and myelinated fibers are very
similar in their course to the pineal gland and intrapineal distribution in cotton rats (Matsushima et al.,
1991, 1994) and Chinese hamsters (Matsushima et al.,
1995), both fibers may belong to the same group of
fibers. CGRP-fibers, like TH- or NPY-fibers, are seen to
enter the distal end of the deep pineal gland in the
Syrian hamster (Fig. 1). A considerable number of
CGRP-immunoreactive fibers are present in the superficial pineal gland, but these fibers are rare in the stalk
and the deep pineal gland in rats (Matsuura et al.,
1994). Our observations on rats are almost identical
with the above data (Fig. 9; Table 3). Although CGRPimmunoreactive fibers are much sparser in the other
rodents than in rats and cotton rats, the distribution of
CGRP-positive fibers in the various regions of the
pineal gland is similar among the rodents thus far
examined; these fibers are present in the superficial
pineal gland, but scarce or absent in the deep pineal
gland (Table 3).
Careful examination of adjacent sections stained
with anti-CGRP and anti-SP shows that CGRP- and
SP-immunoreactive fibers are very similar in their
distribution in the dura mater, the pineal capsule, and
the pineal gland in the cotton rat (Matsushima et al.,
1994) and Chinese hamster (Matsushima et al., 1995).
Thus, it is possible that some CGRP-immunoreactive
fibers also contain SP. In the above rodents, the density
of both CGRP- and SP-immunoreactive fibers is high or
moderate in the superficial pineal gland and the stalk,
but the latter fibers are generally less abundant than
the former (Matsushima et al., 1994, 1995; Tables 3
and 4). CGRP-positive fibers were reported to be more
numerous than SP-positive fibers also in the pineal
gland of gerbils (Shiotani et al., 1986). In mice and
Syrian hamsters, the pineal gland contains no SPcontaining fibers (unpublished observations). It is supposed, therefore, that SP contained in CGRP-positive
fibers is usually small in amount or none at all in the
pineal gland of rodents except rats and cotton rats.
Sympathetic fibers in the pineal gland of cotton rats
may be unique in that these fibers perhaps contain
CGRP or SP. As shown in Tables 3 and 4, CGRP- and
Fig. 5. Electron micrograph showing an NPY-positive sympathetic
nerve ending labeled by DAB in a pericapillary space (*) in the
superficial pineal gland of an unoperated cotton rat. This ending was
also labeled for TH (arrows) by the gold-silver method. Bar ⫽ 0.5 µm.
Fig. 7. NPY-positive ending in a pericapillary space (C, capillary
lumen) in the superficial pineal gland of a ganglionectomized mouse. P,
pinealocyte. Bar ⫽ 0.5 µm. Fig. 8. Synapse (arrow) between an
NPY-positive ending and the cell body of a pinealocyte (PN, pinealocyte nucleus) of an unoperated tree shrew. Bar ⫽ 0.5 µm. Fig. 10.
CGRP-positive, myelinated fiber in a pericapillary space (*) in the
superficial pineal gland of a ganglionectomized cotton rat. Bar ⫽ 0.5
µm. Fig. 11. CGRP-positive, unmyelinated fiber in a pericapillary
space (*) of a ganglionectomized cotton rat. SN, Schwann cell nucleus.
Bar ⫽ 0.5 µm. Fig. 12. SP-positive ending embedded between pinealocytes (PN, pinealocyte nucleus) of an unoperated tree shrew. Bar ⫽ 0.5
µm. Fig. 13. VIP-positive ending and unlabeled fibers with Schwann
cell sheath in a pericapillary space (*) in the superficial pineal gland of
a ganglionectomized mouse. Bar ⫽ 0.5 µm. Fig. 14. VIP-positive
ending and unlabeled sympathetic endings (arrows) containing small
granulated vesicles present in a pericapillary space (C, capillary
lumen) in the superficial pineal gland of an unoperated mouse. P,
pinealocyte. Bar ⫽ 0.5 µm.
273
SP-immunoreactive fibers decrease considerably in
number after superior cervical ganglionectomy. Neuronal cell bodies in the superior cervical ganglia in some
mammals have been reported to contain CGRP or SP
(Ariano and Kenny, 1985; Baffi et al., 1992; Kummer
and Heym, 1988; Schmitt et al., 1988). In the superficial
pineal gland of the cotton rat and Chinese hamster,
CGRP- or SP-immunoreactive fibers are distributed
along blood vessels, together with TH- or NPY-positive
sympathetic fibers (Matsushima et al., 1994, 1995). In
addition, electron microscopic observations indicate
that myelinated and unmyelinated fibers ensheathed
by Schwann cells are exclusively located in perivascular spaces in the superficial pineal gland of these
animals subjected to superior cervical ganglionectomy
(Matsushima et al., 1995; Sakai et al., 1996). Some
unmyelinated fibers have the feature of peptidergic
fibers because of having numbers of large granulated
vesicles (Matsushima et al., 1995). In view of the
presence of CGRP-immunoreactive fibers in the dura
mater and the pineal capsule, their entrance through
the pineal capsule into the gland, and their preferential
localization in perivascular spaces, it is reasonable to
assume that these fibers in the pineal gland of the
above-mentioned rodents are peripheral fibers. This
assumption has been fully justified by our immunoelectron microscopic observations that myelinated (Fig. 10)
and unmyelinated (Fig. 11) fibers immunoreactive for
CGRP, present in pericapillary spaces in the superficial
pineal gland of the cotton rat, are surrounded by
Schwann cell sheath.
Among rodents except cotton rats and Chinese hamsters, only in the rat has the intrapineal distribution of
SP-immunoreactive fibers been described (Rønnekleiv
and Kelly, 1984). SP-immunoreactive fibers are observed in all regions of the pineal gland, and also in the
habenular commissure. Thus, it has been assumed that
SP-immunoreactive fibers enter the gland through the
habenular commissure in the rat. As mentioned, CGRPor SP-immunoreactive fibers are scarce in the deep
pineal gland, and are absent in the habenular and
posterior commissures in the cotton rat (Matsushima et
al., 1994). This is also true for mice and Chinese and
Syrian hamsters (Tables 3, 4). In rats examined by us,
SP-positive fibers are few in the superficial pineal gland
and are absent in the deep pineal gland. Thus, our
results on several rodent species including rats are
incompatible with the data obtained by Rønnekleiv and
Kelly in rats. The reason for this discrepancy remains
unknown.
CGRP-immunoreactive fibers in the pineal gland of
non-rodent species have not been reported in the literature. However, the presence of SP-immunoreactive
fibers has been described in the monkey (Rønnekleiv,
1988) and cow (Møller et al., 1993); these fibers were
distributed throughout the gland abundantly in the
former and moderately in the latter. Since SP-immunoreactive fibers are found in the habenular or posterior
commissure in the above-mentioned animals, it is
supposed that fibers immunoreactive for SP enter the
pineal gland from the brain by the commissures. However, a few SP-immunoreactive fibers appear in the
pineal capsule in the cow. This suggests that some
SP-positive fibers of extracerebral origin also partici-
274
S. MATSUSHIMA ET AL.
Fig. 6. NPY-, and VIP-immunoreactive fibers in adjacent parasagittal sections of the pineal glands of
two ganglionectomized mice. Immunolabeled fibers are few in one animal (A), but a moderate number of
these fibers are present in the other (B). See Figure 1 legend for abbreviations and a symbol.
pate in the innervation of the pineal gland of this
animal (Møller et al., 1993).
Our observations reveal that the pineal gland of the
tree shrew contains CGRP- or SP-immunoreactive fibers (Tables 3, 4), and that bundles of fibers immunoreactive for CGRP or SP enter the gland at its distal end.
In addition, some myelinated fibers running in the
meningeal tissue surrounding the great cerebral vein
were traced to the distal end of the pineal gland. In
spite of repeated investigations, intrapineal neuronal
cell bodies were not detected. In this animal, fairly
abundant, SP-immunoreactive fibers are distributed
evenly throughout the gland (Table 4). CGRP-positive
fibers, on the other hand, are scarce (Table 3). CGRP- or
SP-immunoreactive fibers are very few or absent in the
habenular and posterior commissures. Based on these
observations, we speculate that the pineal gland of the
tree shrew is mainly innervated by CGRP- or SPpositive fibers entering the gland at its distal end.
Myelinated and unmyelinated fibers with Schwann cell
275
PERIPHERAL INNERVATION OF THE MAMMALIAN PINEAL
Fig. 9. CGRP-immunoreactive fibers in parasagittal sections of the superficial pineal glands of an
unoperated (A) and a ganglionectomized (B) rat. See Figure 1 legend for abbreviations and symbols.
TABLE 3. Distribution and density of CGRP-immunoreactive nerve fibers in the pineal gland and effects of superior cervical ganglionectomy
(SCGX) in rodent (A) and non-rodent (B) species1
Distribution and density
Species (A)
Rat
Rat
Cotton rat
Mouse
Chinese hamster
Syrian hamster
Species (B)
Tree shrew
1See
Effect of SCGX
Superficial
Stalk
Deep
Superficial
⫹⫹
⫹⫹
⫹⫹⫹
⫹
⫹
⫹
⫺ or ⫹
⫺
⫹⫹⫹
⫺
⫺ or ⫹
⫺
⫹
⫺
⫺
⫺⬃⫾
(⫹)
(⫹)
(⫾)
(⫹)
(⫹)
(⫹)
Distribution and density
⫹
Stalk
(⫾)
Deep
(⫾)
(⫹)
Reference
Matsuura et al. (1994)
Matsushima et al. (unpublished data)
Matsushima et al. (1994)
Matsushima et al. (unpublished data)
Matsushima et al. (unpublished data)
Matsushima et al. (unpublished data)
Effect of SCGX
Reference
(⫹)
Matsushima et al. (unpublished data)
Table 1 for definitions.
sheath lying in the connective tissue space in the distal
region of the pineal gland are found, by immunoelectron microscopy, to be immunoreactive for SP. Thus,
pineal CGRP- or SP-containing fibers also appear to
belong to the peripheral nervous system in nonrodents. SP-immunoreactive nerve fibers and endings
are occasionally present in the parenchyma interposed
between pinealocytes (Fig. 12). No SP-positive endings
are so far observed in perivascular spaces.
VIP-IMMUNOREACTIVE NERVE FIBER
Although the existence of VIP-immunoreactive fibers
in the mammalian pineal gland has been demonstrated
in many species (Møller et al., 1996), the distribution
and density of these fibers in the different regions of the
pineal gland have hitherto been investigated in only a
few species (Cozzi et al., 1994; Mikkelsen, 1989; Mikkelsen et al., 1994; Table 5). VIP-immunoreactive fibers
276
S. MATSUSHIMA ET AL.
TABLE 4. Distribution and density of SP-immunoreactive nerve fibers in the pineal gland and effects of superior cervical ganglionectomy
(SCGX) in rodent (A) and non-rodent (B) species1
Distribution and density
Species (A)
Cotton rat
Rat
Chinese hamster
Species (B)
Stalk
Deep
Superficial
Stalk
Deep
Reference
⫹⫹⫹
⫹⫹
⫹
⫹⫹⫹
⫹⫹
⫹
⫹
⫺
(⫾)
(⫹)
(⫾)
(⫾)
Matsushima et al. (1994)
Rønnekleiv and Kelly (1984)
Matsushima et al. (unpublished data)
Distribution and density
Effect of SCGX
Reference
(⫹)
Rønnekleiv (1988)
Møller et al. (1993)
Matsushima et al. (unpublished data)
⫹⫹⫹
⫹⫹
⫹⫹
Monkey
Cow
Tree shrew
1See
Effect of SCGX
Superficial
Table 1 for definitions.
TABLE 5. Distribution and density of VIP-immunoreactive nerve fibers in the pineal gland and effects of superior cervical ganglionectomy
(SCGX) in rodent (A) and non-rodent (B) species1
Distribution and density
Species (A)
Rat
Mouse
Mouse
Syrian hamster
Species (B)
Sheep
1See
Superficial
Stalk
Deep
⫹ or ⫹⫹
⫹ or ⫹⫹
⫹ or ⫹⫹
⫺ or ⫹⫹
⫺ or ⫹
⫺
⫺
⫺ or ⫹
⫺
⫺
⫺
Effect of SCGX
Superficial
Distribution and density
⫹
Stalk
Deep
Reference
Mikkelsen (1989)
Mikkelsen et al. (1994)
Matsushima et al. (unpublished data)
Matsushima et al. (unpublished data)
(⫹)
(⫹)
Effect of SCGX
(⫹)
Reference
Cozzi et al. (1994)
Table 1 for definitions.
have been found in the superficial and deep pineal
glands in gerbils, but no mention has been made of
their density in both regions (Møller et al., 1985). In
rats (Mikkelsen, 1989) and mice (Mikkelsen et al.,
1994), VIP-immunoreactive fibers are mainly distributed in the superficial pineal gland, whereas only a few
of them or none are present in the stalk and the deep
pineal gland. Our observations confirm the above results on the mouse (Table 5). As shown in Figure 2, a
small to moderate number of VIP-immunoreactive fibers are distributed in the distal region of the superficial pineal gland, whereas these fibers are absent in the
proximal region of the superficial pineal gland, the
stalk and the deep pineal gland. In addition, the data
obtained in our laboratory indicate that there are
considerable individual variations in the presence of
VIP-positive fibers in the pineal gland of the Syrian
hamster. In the superficial pineal gland of the three
Syrian hamsters examined, VIP-containing fibers are
moderately present in one animal (Fig. 1), but few of
these fibers or none are found in the others; no immunolabeled fibers are detected in the deep pineal gland in
all animals (Table 5). Consequently, it is generally
supposed that VIP-containing fibers have their significance mainly in the superficial pineal gland in rodents.
In the sheep, however, these fibers are distributed
throughout the gland (Cozzi et al., 1994). Thus, the
distribution pattern of VIPergic fibers in the various
regions of the gland appears to vary from rodent to
non-rodent species.
As mentioned, VIP-immunoreactive fibers are mainly
distributed in the superficial pineal gland, and these
fibers are absent in the habenular and posterior commissures in rats (Mikkelsen, 1989) and mice (Mikkelsen et
al., 1994). The above observations on mice are in good
agreement with our results (Fig. 2, Table 5), and we
were unable to detect VIP-positive fibers in the habenular and posterior commissures also in Syrian hamsters
(Fig. 1). In addition, it has been shown that VIP-fibers
are distributed in the pineal capsule in several mammals (Cozzi et al., 1990; Mikkelsen, 1989; Mikkelsen et
al., 1994; Uddman et al., 1980), and that some intracapsular VIP-fibers enter the gland via the connective
tissue spaces (Cozzi et al., 1994; Mikkelsen, 1989). The
capsule on the distal region of the superficial pineal
gland of mice (Fig. 2) and Syrian hamsters also contains
VIP-immunoreactive fibers. These findings support the
view that VIP-fibers do not represent central fibers but
belong to the peripheral nervous system (Mikkelsen,
1989; Mikkelsen et al., 1994; Møller and Mikkelsen,
1989). Our immunoelectron microscopic observations
substantiate this to be true at least in the mouse;
VIP-positive nerve endings in the pineal gland of this
animal are ensheathed by Schwann cells (Fig. 13).
Whether VIP-immunoreactive fibers are present
mainly in the connective tissue spaces or the parenchyma is inconsistent in the previous results thus far
obtained. In many mammalian species, VIP-immunoreactive fibers are distributed mostly in the perivascular
spaces and occasionally in the parenchyma (Mikkelsen,
1989; Møller et al., 1985; Uddman et al., 1980), whereas
a reverse relationship is observed in the mouse (Mikkelsen et al., 1994). Mikkelsen et al. (1994) have
mentioned that VIP-immunoreactive fibers are mainly
encountered in the parenchyma, and some are found in
relation to blood vessels in the superficial pineal gland
of this animal. We demonstrate, immunoelectron microscopically, that VIP-positive nerve endings are localized
almost exclusively in perivascular spaces; solitary VIPpositive nerve endings are scattered among a number of
PERIPHERAL INNERVATION OF THE MAMMALIAN PINEAL
sympathetic nerve endings (Fig. 14). So far no VIPcontaining endings are found between pinealocytes.
Sympathetic nerve fibers and endings are almost entirely present in pericapillary spaces in the superficial
pineal gland of several rodents including mice (Fig. 14).
Thus, it may be concluded that VIPergic fibers, together
with sympathetic ones, are predominantly distributed
in perivascular spaces in the rodent pineal gland.
VIP-immunoreactive nerve endings, like sympathetic
ones, mainly appear as free endings (Fig. 14), but there
are some VIP-positive endings ensheathed by Schwann
cells (Fig. 13). VIP-containing nerve endings are much
sparser than sympathetic ones. The difference in the
density between both endings can easily be understood
by comparison of the density of NPY- and VIP-positive
fibers (Fig. 2).
The effect of superior cervical ganglionectomy on
VIP-immunoreactive fibers in the mammalian pineal
gland has previously been examined only in the sheep
(Cozzi et al., 1994). Since the density of VIP-immunoreactive fibers does not change after the removal of the
ganglia, it is apparent that intrapineal VIP-fibers do
not originate from the ganglia at least in the sheep
(Cozzi et al., 1994). The density of VIP-positive fibers
was found to be unaffected by superior cervical ganglionectomy in mice (Fig. 2 vs. Fig. 6; Table 5). In addition,
in one Syrian hamster among two ganglionectomized, a
moderate number of VIP-immunoreactive fibers are
present in the superficial pineal gland, and a few nerve
fibers immunoreactive for VIP are recognized in the
capsule of its distal region (Fig. 4). Thus, intrapineal
VIP-containing fibers may not originate from this ganglia in both species.
Examination of adjacent sections stained with antiVIP and anti-NPY reveals that the intrapineal distribution of nerve fibers immunoreactive for these peptides
is very similar in ganglionectomized mice (Fig. 6). Thus,
it is possible that pineal VIPergic fibers may contain
NPY in this animal. Double immunoelectron microscopic studies are necessary to examine the colocalization of these peptides in extrasympathetic fibers in the
pineal gland of mice.
COMPARISON OF THE DISTRIBUTION
AND DENSITY OF THE PERIPHERAL AND
CENTRAL FIBERS IN THE PINEAL GLAND
OF RODENTS AND NON-RODENTS
TH/NPY-positive, sympathetic fibers are distributed
densely and evenly throughout all regions of the pineal
gland in many rodent species. By contrast, the density
of CGRP/SPergic or VIPergic, peripheral fibers is apparently lower than that of sympathetic fibers, and their
distribution exhibits regional differences. These fibers
are more or less confined to the superficial pineal gland.
The significance of such differences in the distribution
of these peripheral fibers in the rodent pineal gland
remains to be studied.
TH/NPY-positive fibers that remain mainly in the
stalk and the deep pineal gland in rodents after superior cervical ganglionectomy are believed to originate
from the central nervous system. The above observations are compatible with the results of neuronal tracing experiments. Møller and his co-workers have demonstrated, using anterograde tracing techniques, that
in some rodents nerve fibers from the hypothalamus or
277
the lateral geniculate nucleus terminate in the stalk
and the deep pineal gland, but not in the superficial
pineal gland (Møller, 1992).
The distribution and density of sympathetic fibers
and non-sympathetic peripheral fibers in the pineal
gland of non-rodents are almost the same as those in
rodents with few exceptions (see Tables 1–5). However,
the central innervation of the pineal gland shows
marked differences between the two groups of animals.
In non-rodents, unlike in rodents, TH/NPY-positive
fibers remaining after superior cervical ganglionectomy
are usually distributed throughout the gland and, thus,
these fibers appear to be more numerous than in
rodents (sheep, Cozzi et al., 1992; tree shrew, Matsushima et al., unpublished data; cat, Møller et al.,
1994; cow, Phansuwan-Pujito et al., 1993). Based on the
reasons mentioned before, these fibers are supposed to
arise from the brain. Although it is generally believed
that the pineal gland of non-rodents receives a rich
supply of central fibers from the habenular and posterior commissures (Korf and Møller, 1984, 1985; Møller,
1992; Reuss, 1996; Vollrath, 1981), comparative studies
on the central innervation in the pineal gland of rodents
and non-rodents have up to now not been conducted.
For such studies, pinealopetal fibers containing substances peculiar to the central fibers appear to be more
appropriate than TH/NPY-positive fibers. Such substances may include vasopressin (VP), oxytocin (OXY),
and histamine. Nerve fibers immunoreactive for them
are found to be confined to the deep pineal gland of the
rat (Buijs and Pévet, 1980; Matsuura et al., 1994;
Mikkelsen et al., 1992), whereas in the dog VP/OXYpositive fibers are distributed everywhere throughout
the pineal gland, though decreasing in number toward
the distal region (Matsuura et al., 1983). Thus, the
distribution of VP/OXY-immunoreactive fibers in the
pineal gland is similar to that of TH/NPY-positive,
central fibers in rodent and non-rodent species. However, since the available data are still very fragmentary,
further observations on the central fibers should be
made in more mammalian species.
RELATIONSHIPS BETWEEN THE
PERIPHERAL FIBERS AND THEIR ENDINGS
AND PARENCHYMAL CELLS IN THE PINEAL
GLAND OF RODENTS AND NON-RODENTS
From conventional electron microscopic observations, it is generally accepted that sympathetic fibers
are localized in both perivascular spaces and the parenchyma in the mammalian pineal gland (Møller, 1992;
Pévet, 1983; Vollrath, 1981). In rodents, the intrapineal
localization of sympathetic fibers appears to be different between the superficial and deep components of the
pineal complex, though the available information concerning their location in the latter component is rather
limited. Sympathetic fibers and endings are mainly
encountered in perivascular spaces in the superficial
pineal gland (Matsushima and Reiter, 1975, 1977;
Sakai et al., 1996; Vollrath, 1981). On the other hand,
these fibers are exclusively located in the parenchyma
in the deep pineal gland, even though blood vessels are
present (Masson-Pévet et al., 1987; Sakai et al., 1996).
Although the results of the conventional electron microscopic studies on the intrapineal localization of sympathetic fibers and endings in non-rodents are still frag-
278
S. MATSUSHIMA ET AL.
mentary (Vollrath, 1981), recent immunohistochemical
data indicate the presence of perivascular and intraparenchymal, TH/NPY-positive fibers, probably consisting of sympathetic fibers and some central fibers (Cozzi
et al., 1992; Mikkelsen and Mick, 1992; Møller et al.,
1994; Phansuwan-Pujito et al., 1993). In the above
observations, TH/NPY-containing fibers are found to
form a diffuse network around pinealocytes. Thus, the
deep pineal gland of rodents is similar to the gland of
non-rodents with respect to the presence of abundant
intraparenchymal sympathetic fibers.
As noted, CGRP/SPergic and VIPergic fibers are
predominantly distributed in the superficial pineal
gland in rodents. Both fibers are also recognized mainly
along blood vessels (Mikkelsen, 1989; Matsushima et
al., 1994; Matsushima et al., unpublished data, see
Figs. 13, 14; Møller et al., 1985). In non-rodents, on the
contrary, SP- or VIP-positive fibers are detected perivascularly as well as between pinealocytes (Cozzi et al.,
1990; Matsushima et al., unpublished data, see Fig. 12;
Møller et al., 1993; Rønnekleiv, 1988). It is considered,
therefore, that the intrapineal location of CGRP/SP- or
VIP-positive fibers, probably consisting mostly of peripheral fibers, resembles that of sympathetic fibers in
rodent and non-rodent species.
The above-mentioned location of the peripheral fibers
is closely associated with the occurrence of synapses
between their terminals and pinealocytes. In some
rodents, synapses or synaptic-like contacts between
sympathetic nerve terminals and pinealocytes or glial
cells occur in the deep pineal gland, but not in the
superficial pineal gland (Masson-Pévet et al., 1987;
Sakai et al., 1996). Synapses are rarely observed between sympathetic nerve endings and pinealocytes in
the superficial pineal gland of the rat (Huang and Lin,
1984), but nothing has been found in the literature
concerning such synapses in the deep pineal gland of
this animal. Conventional electron microscopic studies
demonstrate the presence of synapses between presumptive sympathetic nerve endings and pinealocytes in
some non-rodents; such synapses occur occasionally in
the monkey (Ling et al., 1989, 1990) and frequently in
the tree shrew (Hwang, 1982). Synaptic-like junctions
have recently been observed immunoelectron microscopically between nerve fibers immunoreactive for
C-terminal flanking peptide of NPY and pinealocytes in
the pig (Przybylska-Gornowicz et al., 1997). Synaptic
nerve terminals mentioned above can not always be
identified as sympathetic nerve endings, because the
pineal gland of non-rodents is also innervated, to some
extent, by TH/NPY-positive, non-sympathetic fibers.
However, since NPY-positive fibers are nearly lost in
the pineal gland of the tree shrew by superior cervical
ganglionectomy (Table 2), NPY-immunoreactive nerve
endings synapsing with pinealocytes of this animal
(Fig. 8) can undoubtedly be regarded as the terminals of
sympathetic fibers. The evidence thus far obtained,
though not sufficient to be conclusive, suggests that
synapses between sympathetic nerve endings and pinealocytes occur commonly in the deep pineal gland of
rodents and the pineal gland of non-rodents, but scarcely
in the superficial pineal gland of rodents.
In the superficial pineal gland of rodents, processes of
pinealocytes extend into pericapillary spaces (Matsushima and Morisawa, 1982; Matsushima and Reiter,
1975, 1977; Matsushima et al., 1979, 1989; for review
see Pévet, 1983; Vollrath, 1981) and, therefore, come to
lie close to sympathetic fibers. This arrangement may
permit neurotransmitters from sympathetic nerve endings to reach pinealocytes promptly through pericapillary spaces (Pévet, 1983; Redecker, 1993; Vollrath,
1981).
The deep pineal gland and the stalk of rodents (Calvo
and Boya, 1985; Hewing, 1981; Luo et al., 1984; Sakai
et al., 1996; Sheridan and Reiter, 1970) as well as the
pineal gland of non-rodents (Papasozomenos, 1983;
Vollrath, 1981; Zang et al., 1985) contain abundant
astrocytes. In the deep portion of the pineal gland of
cotton rats, unlike in the superficial portion, pericapillary spaces are surrounded completely by astrocytic
processes (Sakai et al., 1996), thus preventing pinealocytes from being exposed to the spaces. The same
appears to be the case in non-rodents (Papasozomenos,
1983; Vollrath, 1981; Zang et al., 1985). Our electron
microscopic observations (unpublished) have revealed
that the pericapillary area in the pineal gland of the
tree shrew is very similar to that of the deep pineal
gland of the cotton rat; no pinealocyte processes are
found in the spaces. Taking all things together, it may
be suggested that the occurrence of pericapillary or
intraparenchymal sympathetic fibers depends on the
presence or absence of pinealocyte processes in the
pericapillary spaces. In the deep pineal gland of rodents
and the pineal gland of non-rodents, where pinealocytes are separated from the pericapillary spaces by a
glial barrier, sympathetic fibers are mainly observed in
the parenchyma, and some of their endings make
synaptic contact with pinealocytes or astrocytes. This
speculation is, of course, regarded as tentative as yet,
and additional observations on more mammalian species are necessary to establish such relations between
sympathetic fibers and pinealocytes in rodents and
non-rodents.
Non-sympathetic nerve endings synapsing with pinealocytes have also been described in the pineal gland of
rodents (Korf et al., 1990; Matsushima and Reiter,
1978; Matsuura et al., 1994; Møller, 1985) and nonrodents (Hwang, 1982; Ichimura, 1992; Ichimura et al.,
1986). These nerve endings contain small clear vesicles
and a few large granulated vesicles. Although such
synaptic nerve endings are supposed to derive from
extrapineal sources or from intrapineal neurons or
neuron-like cells, their exact nature and origin are
unknown.
ORIGINS OF CGRP/SP- AND
VIP-IMMUNOREACTIVE FIBERS
Retrograde tracing combined with immunohistochemistry has revealed that CGRP/SPergic fibers in the
pineal gland of gerbils (Shiotani et al., 1986) and rats
(Reuss et al., 1992) have the common source, i.e., the
trigeminal ganglia. The only study on the origin of
VIPergic fibers in the mammalian pineal gland has
been made by Shiotani et al. (1986), who have demonstrated, using combined tracing and immunohistochemistry, that these fibers in the pineal gland of gerbils
arise from the pterygopalatine ganglia. Due to technical difficulties, similar attempts to examine the origin
of these peripheral fibers have not hitherto been made
in non-rodents. Thus, our knowledge of the origin of
PERIPHERAL INNERVATION OF THE MAMMALIAN PINEAL
CGRP/SPergic and VIPergic peripheral fibers is still
incomplete.
For determining the origin of CGRP/SP- or VIPpositive fibers in the mammalian pineal gland, mice
and cotton rats may be expected to be good experimental animals for the following reasons: (1) in both
species, unlike in many other rodents, TH/NPYimmunoreactive central fibers are absent in the pineal
gland, because of the complete disappearance of TH/
NPY-positive fibers after superior cervical ganglionectomy; (2) a moderate or large number of CGRP/SP- and
VIP-immunoreactive fibers are present in the pineal
gland of the cotton rat and mouse, respectively; (3) a
moderate number of NPY-positive, non-sympathetic
peripheral fibers are found in the pineal gland of the
mouse; and (4) presumptive central fibers synapsing
with pinealocytes are distributed in the deep pineal
gland of the mouse (Korf et al., 1990). Studies are now
in progress, using retrograde tracing and immunohistochemistry, on the origins of CGRP/SPergic and VIPergic
nerve fibers in the pineal gland of these animals.
REFERENCES
Ariano MA, Kenny SL. 1985. Peptide coincidence in rat superior
cervical ganglion. Brain Res 340:181–185.
Baffi J, Görcs T, Slowik F, Horváth M, Lekka N, Pásztor E, Palkovits
M. 1992. Neuropeptides in the human superior cervical ganglion.
Brain Res 570:272–278.
Bowers CW, Dahm LM, Zigmond RE. 1984. The number and distribution of sympathetic neurons that innervate the rat pineal gland.
Neuroscience 13:87–96.
Buijs RM, Pévet P. 1980. Vasopressin- and oxytocin-containing fibers
in the pineal gland and subcommissural organ of the rat. Cell Tissue
Res 205:11–17.
Calvo J, Boya J. 1985. Ultrastructure of the rat pineal stalk. Acta Anat
123:172–177.
Cavanagh JFR, Mione MC, Burnstock G. 1990. Use of enhanced silver
staining combined with electron microscopical immunolabelling to
demonstrate the colocalization of neuropeptide Y and vasoactive
intestinal polypeptide in cerebrovascular nerves. Neuroscience 39:
775–785.
Cozzi B, Mikkelsen JD, Merati D, Capsoni S, Møller M. 1990.
Vasoactive intestinal peptide-like immunoreactive nerve fibers in
the pineal gland of the sheep. J Pineal Res 8:41–47.
Cozzi B, Mikkelsen JD, Ravault J-P, Møller M. 1992. Neuropeptide Y
(NPY) and C-Flanking Peptide of NPY in the pineal gland of normal
and ganglionectomized sheep. J Comp Neurol 316:238–250.
Cozzi B, Mikkelsen JD, Ravault J-P, Locatelli A, Fahrenkrug J, Zhang
E-T, Møller M. 1994. Density of peptide histidine-isoleucine- and
vasoactive intestinal peptide-immunoreactive nerve fibers in the
sheep pineal gland is not affected by superior cervical ganglionectomy. J Comp Neurol 343:72–82.
Gregorek JC, Seibel HR, Reiter RJ. 1977. The pineal complex and its
relationship to other epithalamic structures. Acta Anat 99:425–434.
Hewing M. 1981. Topographical relationships of synaptic ribbons in
the pineal system of the vole (Microtus agrestis). Anat Embryol
162:313–323.
Huang H-T, Lin H-S. 1984. Synaptic junctions between the adrenergic
axon varicosity and the pinealocyte in the rat. J Pineal Res
1:281–291.
Hwang BH. 1982. Fluorescence and electron microscopic study of the
tree shrew pineal organ. J Neural Transm 53:193–212.
Ichimura T. 1992. The ultrastructure of neuronal-pinealocytic interconnections in the monkey pineal. Microsc Res Tech 21:124–135.
Ichimura T, Arikuni T, Hashimoto PH. 1986. Fine-structural study of
the pineal body of the monkey (Macaca fuscata) with special
reference to synaptic formations. Cell Tissue Res 244:569–576.
Jin KL, Shiotani Y, Kawai Y, Kiyama H. 1988. Immunohistochemical
demonstration of tyrosine hydroxylase (TH)-positive but dopamine
␤-hydroxylase (DBH)-negative neuron-like cells in the pineal gland
of golden hamsters. Neurosci Lett 93:28–31.
Kaleczyc J, Przybylska B, Majewski M, Lewczuk B. 1994. Immunohistochemical studies on the coexistence of catecholamine-synthesizing
enzymes and neuropeptide Y in nerve fibers of the porcine pineal
gland. J Pineal Res 17:20–24.
279
Kappers JA. 1960. The development, topographical relations and
innervation of the epiphysis cerebri in the albino rat. Z Zellforsch
52:163–215.
Karasek M, Petterborg LJ, King TS, Hansen JT, Reiter RJ. 1983.
Effect of superior cervical ganglionectomy on the ultrastructure of
the pinealocyte in the cotton rat (Sigmodon hispidus). Gen Comp
Endocrinol 51:131–137.
Korf H-W, Møller M. 1984. The innervation of the mammalian pineal
gland with special reference to central pinealopetal projections.
Pineal Res Rev 2:41–86.
Korf H-W, Møller M. 1985. The central innervation of the mammalian
pineal organ. In: Mess B, Rúzsá Cs, Tima L, Pévet P, editors. The
pineal gland. Current state of pineal research. New York: Elsevier. p
47–69.
Korf H-W, Sato T, Oksche A. 1990. Complex relationships between the
pineal organ and the medial habenular nucleus-pretectal region of
the mouse as revealed by S-antigen immunocytochemistry. Cell
Tissue Res 261:493–500.
Kummer W, Heym C. 1988. Neuropeptide distribution in the cervicothoracic paravertebral ganglia of the cat with particular reference to
calcitonin gene-related peptide immunoreactivity. Cell Tissue Res
252:463–471.
Kuwayama Y, Emson PC, Stone RA. 1988. Pterygopalatine ganglion
cells contain neuropeptide Y. Brain Res 446:219–224.
Leblanc GG, Trimmer BA, Landis SC. 1987. Neuropeptide Y-like
immunoreactivity in rat cranial parasympathetic neurons: Coexistence with vasoactive intestinal peptide and choline acetyltransferase. Proc Natl Acad Sci USA 84:3511–3515.
Li K, Welsh MG. 1991. Tyrosine hydroxylase and neuropeptide-Y
immunoreactivity in pineal glands developing in situ and in pineal
grafts. Cell Tissue Res 264:515–527.
Ling EA, Tan SH, Yick TY, Wong WC. 1989. Ultrastructure of the
pineal gland of the monkey, Macaca fascicularis, with special
reference to the presence of synaptic junctions on pinealocytes. Anat
Embryol 180:151–158.
Ling EA, Tan SH, Wong WC. 1990. Synaptic junctions between
sympathetic axon terminals and pinealocytes in the monkey Macaca
fascicularis. Anat Embryol 182:21–27.
Lundberg JM, Hökfelt T, Schultzberg M, Uvnäs-Wallensten K, Köhler
C, Said SI. 1979. Occurrence of vasoactive intestinal polypeptide
(VIP)-like immunoreactivity in certain cholinergic neurons of the
cat: evidence from combined immunohistochemistry and acetylcholinesterase staining. Neuroscience 4:1539–1559.
Luo ZR, Schultz RL, Whitter EF, Vollrath L. 1984. Ultrastructural
characterization of glial cells in the rat pineal gland with special
reference to the pineal stalk. Anat Rec 210:663–674.
Masson-Pévet M, Pévet P, Gourmelen S. 1987. Monoaminergic synaptic contacts on pinealocytes in the deep pineal of the European
hamster (Cricetus cricetus, L.). In: Trentini GP, DeGaetani C, Pévet
P, editors. Fundamentals and clinics in pineal research. New York:
Raven Press. 83–86.
Matsushima S, Morisawa Y. 1982. Ultrastructural observations on the
pineal gland of the Chinese hamster, Cricetulus griseus. I. The
superficial pineal. Cell Tissue Res 222:531–546.
Matsushima S, Reiter RJ. 1975. Comparative ultrastructural studies
of the pineal gland of rodents. In: Hess M, editor. Ultrastructure of
endocrine and reproductive organs. New York: John Wiley and Sons.
p 335–356.
Matsushima S, Reiter RJ. 1977. Fine structural features of adrenergic
nerve fibers and endings in the pineal gland of the rat, ground
squirrel and chinchilla. Am J Anat 148:463–477.
Matsushima S, Reiter RJ. 1978. Electron microscopic observations on
neuron-like cells in the ground squirrel pineal gland. J Neural
Transm 42:223–237.
Matsushima S, Morisawa Y, Petterborg LJ, Zeagler JW, Reiter RJ.
1979. Ultrastructure of pinealocytes of the cotton rat, Sigmodon
hispidus. Cell Tissue Res 204:407–416.
Matsushima S, Sakai Y, Hira Y. 1989. Twenty-four-hour changes in
pinealocytes, capillary endothelial cells and pericapillary and intercellular spaces in the pineal gland of the mouse. Cell Tissue Res
255:323–332.
Matsushima S, Sakai Y, Hira Y, Kawai Y, Saito Y. 1991. Postnatal
changes in nerve cells and myelinated fibers in the pineal gland of
cotton rats (Sigmodon hispidus). A light microscopic study. Arch
Histol Cytol 54:379–389.
Matsushima S, Sakai Y, Hira Y, Oomori Y, Daikoku S. 1994. Immunohistochemical studies on sympathetic and non-sympathetic nerve
fibers and neuronal cell bodies in the pineal gland of cotton rats,
Sigmodon hispidus. Arch Histol Cytol 57:47–58.
280
S. MATSUSHIMA ET AL.
Matsushima S, Sakai Y, Hira Y. 1995. Sprouting of non-sympathetic
myelinated and unmyelinated fibres in response to chronic sympathetic denervation in the pineal gland of the Chinese hamster,
Cricetulus griseus. J Neurocytol 24:519–531.
Matsuura T, Kawata M, Yamada H, Kojima M, Sano Y. 1983. Immunohistochemical studies on the peptidergic nerve fibers in the pineal
organ of the dog. Arch Histol Jpn 46:373–379.
Matsuura T, Kumamoto K, Ebara S. 1994. Nerve fibers originating
from the brain in the rat pineal complex. J Electron Microsc
43:255–263.
Mikkelsen JD. 1989. Immunohistochemical localization of vasoactive
intestinal peptide (VIP) in the circumventricular organs of the rat.
Cell Tissue Res 255:307–313.
Mikkelsen JD, Mick G. 1992. Neuropeptide Y-immunoreactive nerve
fibres in the pineal gland of the macaque (Macaca fascicularis). J
Neuroendocrinol 4:681–688.
Mikkelsen JD, Panula P, Møller M. 1992. Histamine-immunoreactive
nerve fibers in the rat pineal gland: evidence for a histaminergic
central innervation. Brain Res 597:200–208.
Mikkelsen JD, Møller M, Larsen PJ, Fahrenkrug J. 1994. The
presence of nerve fibers immunoreactive for vasoactive intestinal
peptide (VIP), peptide histidine isoleucine (PHI), and prepro VIP
(111–122) in the mouse pineal gland. J Pineal Res 16:50–56.
Møller M. 1985. Non-sympathetic synaptic innervation of the pinealocyte of the Mongolian gerbil (Meriones unguiculatus): an electron
microscopic study. J Neurocytol 14:541–550.
Møller M. 1992. Fine structure of the pinealopetal innervation of the
mammalian pineal gland. Microsc Res Tech 21:188–204.
Møller M, Mikkelsen JD. 1989. Vasoactive intestinal peptide (VIP) and
peptide histidine isoleucine (PHI) in the mammalian pineal gland.
Adv Pineal Res 3:1–10.
Møller M, Mikkelsen JD, Fahrenkrug J, Korf H-W. 1985. The presence
of vasoactive intestinal polypeptide (VIP)-like-immunoreactive nerve
fibres and VIP-receptors in the pineal gland of the Mongolian gerbil
(Meriones unguiculatus). An immunohistochemical and receptorautoradiographic study. Cell Tissue Res 241:333–340.
Møller M, Mikkelsen JD, Martinet L. 1990. Innervation of the mink
pineal gland with neuropeptide Y (NPY)-containing nerve fibers. An
experimental immunohistochemical study. Cell Tissue Res 261:477–
483.
Møller M, Phansuwan-Pujito P, Govitrapong P, Schmidt P. 1993.
Indications for a central innervation of the bovine pineal gland with
substance P-immunoreactive nerve fibers. Brain Res 611:347–351.
Møller M, Phansuwan-Pujito P, Pramaulkijja S, Kotchabhakdi N,
Govitrapong P. 1994. Innervation of the cat pineal gland by neuropeptide Y-immunoreactive nerve fibers: an experimental immunohistochemical study. Cell Tissue Res 276:545–550.
Møller M, Ravault J-P, Cozzi B. 1996. The chemical neuroanatomy of
the mammalian pineal gland: Neuropeptides. Neurochem Int 28:23–33.
Møller M, Masson-Pévet M, Pévet P. 1998. Annual variations of the
NPYergic innervation of the pineal gland in the European hamster
(Cricetus cricetus): a quantitative immunohistochemical study. Cell
Tissue Res 291:423–431.
Papasozomenos SC. 1983. Glial fibrillary acidic (GFA) proteincontaining cells in the human pineal gland. J Neuropathol Exp
Neurol 42:391–408.
Pévet P. 1983. Anatomy of the pineal gland of mammals. In: Relkin R,
editor. The pineal gland. New York: Elsevier Biomedical. p 1–75.
Phansuwan-Pujito P, Mikkelsen JD, Govitrapong P, Møller M. 1991. A
cholinergic innervation of the bovine pineal gland visualized by
immunohistochemical detection of choline acetyltransferase-immunoreactive nerve fibers. Brain Res 545:49–58.
Phansuwan-Pujito P, Pramaulkijja S, Govitrapong P, Møller M. 1993.
An immunohistochemical study of neuropeptide Y in the bovine
pineal gland. J Pineal Res 15:53–58.
Przybylska-Gornowicz B, Lewczuk B, Møller M. 1997. Demonstration
of nerve fibers containing the C-terminal flanking peptide of neuropeptide Y (CPON) in the pig pineal gland (Sus domesticus): an
immunocytochemical study by light and electron microscopy. Anat
Rec 248:576–582.
Redecker P. 1993. Close microtopographical relationships between
sympathetic nerve terminals and bulbous process endings of pinealocytes in the pineal gland of the Mongolian gerbil. J Pineal Res
15:199–207.
Reuss S. 1996. Components and connections of the circadian timing
system in mammals. Cell Tissue Res 285:353–378.
Reuss S, Riemann R, Vollrath L. 1992. Substance P- and calcitonin
gene-related peptide-like immunoreactive neurons in the rat trigeminal ganglion- with special reference to meningeal and pineal innervation. Acta Histochem 92:104–109.
Romeo HE, Weihe E, Müller S, Vollrath L. 1993. Protein gene product
(PGP) 9.5 immunoreactivity in nerve fibers and pinealocytes of
guinea-pig pineal gland: interrelationship with tyrosine-hydroxylase- and neuropeptide-Y-immunoreactive nerve fibers. Cell Tissue
Res 271:477–484.
Rønnekleiv OK. 1988. Distribution in the Macaque pineal of nerve
fibers containing immunoreactive substance P, vasopressin, oxytocin, and neurophysins. J Pineal Res 5:259–271.
Rønnekleiv OK, Kelly MJ. 1984. Distribution of substance P neurons
in the epithalamus of the rat: an immunohistochemical investigation. J Pineal Res 1:355–370.
Sakai Y, Hira Y, Matsushima S. 1996. Regional differences in the
pineal gland of the cotton rat, Sigmodon hispidus: light microscopic,
electron microscopic, and immunohistochemical observations. J
Pineal Res 20:125–137.
Sasek C, Zigmond RE. 1989. Localization of vasoactive intestinal
peptide- and peptide histidine isoleucine amide-like immunoreactivities in the rat superior cervical ganglion and its nerve trunks. J
Comp Neurol 280:522–532.
Schmitt M, Kummer W, Heym C. 1988. Calcitonin gene-related
peptide (CGRP)-immunoreactive neurons in the human cervicothoracic paravertebral ganglia. J Chem Neuroanat 1:287–292.
Schon F, Allen JM, Yeats JC, Allen YS, Ballesta J, Polak JM, Kelly JS,
Bloom SR. 1985. Neuropeptide Y innervation of the rodent pineal
gland and cerebral blood vessels. Neurosci Lett 57:65–71.
Sheridan MN, Reiter RJ. 1970. Observations on the pineal system in
the hamster. II. Fine structure of the deep pineal. J Morphol
131:163–178.
Shiotani Y, Yamano M, Shiosaka S, Emson PC, Hillyard CJ, Girgis S,
MacIntyre I. 1986. Distribution and origins of substance P (SP)-,
calcitonin gene-related peptide (CGRP)-, vasoactive intestinal polypeptide (VIP)- and neuropeptide Y (NPY)-containing nerve fibers in
the pineal gland of gerbils. Neurosci Lett 70:187–192.
Suzuki N, Hardebo JE, Kåhrström J, Owman C. 1990. Neuropeptide Y
co-exists with vasoactive intestinal polypeptide and acetylcholine in
parasympathetic cerebrovascular nerves originating in the sphenopalatine, otic, and internal carotid ganglia of the rat. Neuroscience
36:507–519.
Uddman R, Alumets J, Håkanson R, Lorén I, Sundler F. 1980.
Vasoactive intestinal peptide (VIP) occurs in nerves of the pineal
gland. Experientia 36:1119–1120.
Vollrath L. 1981. The pineal organ. In: Oksche A, Vollrath L, editors.
Handbuch der mikroskopischen Anatomie des Menschen, VI/7.
Berlin: Springer. p 186–212, 220–222.
Zang X, Nilaver G, Stein BM, Fetell MR, Duffy PE. 1985. Immunocytochemistry of pineal astrocytes: species differences and functional
implications. J Neuropathol Exp Neurol 44:486–495.
Zhang E-T, Mikkelsen JD, Møller M. 1991. Tyrosine hydroxylase- and
neuropeptide Y-immunoreactive nerve fibers in the pineal complex
of untreated rats and rats following removal of the superior cervical
ganglia. Cell Tissue Res 265:63–71.
Документ
Категория
Без категории
Просмотров
3
Размер файла
534 Кб
Теги
617
1/--страниц
Пожаловаться на содержимое документа