close

Вход

Забыли?

вход по аккаунту

?

Differential expression of annexins I-VI in the rat dorsal root ganglia and spinal cord

код для вставкиСкачать
THE JOURNAL OF COMPARATIVE NEUROLOGY 368:356-370 (1996)
Differential Expression of Annexins I-VI
in the Rat Dorsal Root Ganglia
and Spinal Cord
JORGE M. NACIFF, MARCIA A. KAETZEL, MICHAEL M. BEHBEHANI,
AND JOHN R. DEDMAN
Department of Molecular and Cellular Physiology, University of Cincinnati
College of Medicine, Cincinnati, Ohio 45267-0576
ABSTRACT
The annexins are a family of Ca2+-dependentphospholipid-binding proteins. In the present
study, the spatial expression patterns of annexins I-VI were evaluated in the rat dorsal root
ganglia (DRG) and spinal cord (SC) by using indirect immunofluorescence. Annexin I is
expressed in small sensory neurons of the DRG, by most neurons of the SC, and by ependymal
cells lining the central canal. Annexin I1 is expressed by most sensory neurons of the DRG but is
primarily expressed in the SC by glial cells. Annexin I11 is expressed by most sensory neurons,
regardless of size, by endothelial cells lining the blood vessels, and by the perineurium. In the
SC, annexin I11 is primarily expressed by astrocytes. In the DRG and the SC, annexin IV is
primarily expressed by glial cells and at lower levels by neurons. In the DRG, annexin V is
expressed in relatively high concentrations in small sensory neurons in contrast to the SC,
where it is expressed mainly by ependymal cells and by small-diameter axons located in the
superficial laminae of the dorsal horn areas. Annexin VI is differentially expressed by sensory
neurons of the DRG, being more concentrated in small neurons. In the SC, annexin VI has the
most striking distribution. It is concentrated subjacent to the plasma membrane of motor
neurons and their processes. The differential localization pattern of annexins in cells of the SC
and DRG could reflect their individual biological roles in Ca2+-signaltransduction within the
central nervous system. G 19% WiIey-Liss, Inc.
Indexing terms: Caz+-bindingproteins, sensory neurons, motor neurons, glial cells,
immunolocalization
Transient increases in cytoplasmic free Ca2+ levels are
regulatory signals that allow neurons to respond to a
variety of extracellular stimuli including growth factors,
hormones, neurotransmitters, and electrical activity across
the plasma membrane. An increase in intracellular Ca2+
must be translated into responses as varied as changes in
secretion, ion transport, metabolic activity, and gene expression. Ca2+ is also a major second messenger involved in
neuronal development by participating in cellular proliferation, differentiation, and apoptosis. The multiplicity of
cellular processes activated by changes in the cytosolic Ca2+
levels involves multiple Ca2'-regulated pathways and suggests differential expression of independent mediator proteins that discriminate the Ca2+signal. The understanding
of the cellular pathways used to mediate and discriminate
the intracellular Ca2+signal in the nervous system requires
the identification of Ca2+-bindingproteins with regard to
cellular specificity and subcellular localization.
The annexins are a gene family of Ca2+-dependentphospholipid-binding proteins, with the potential to participate
O
1996 WILEY-LISS, INC.
in the transduction of the intracellular Ca2+ signal. The
annexin family encompasses members found in diverse
eukaryotic organisms, including fruit flies, sponges, slime
molds, and higher plants (Crumpton, 1992). They constitute a significant amount of total cellular protein. Ten
unique annexins have been identified in mammalian tissues
(Moss, 1992; Raynal and Pollard, 1994).These proteins are
water soluble, amphipathic, and have the common property
to bind to phospholipids when complexed with Ca2+.The
primary structure of the annexins is composed of highly
homologous repeats of approximately 70 amino acid residues (Geisow et al., 1986; Barton et al.,1991; Moss, 1992;
Raynal and Pollard, 1994).The Ca2+-bindingaffinity of the
annexins is enhanced when they are bound to phospholipid
membranes (Haigler et al., 1987; Tait et al., 1989; Bazzi and
Accepted December 6,1995.
Address correspondence to John R. Dedman, Department of Molecular
and Cellular Physiology, University of Cincinnati College of Medicine,
Cincinnati, OH 45267-0576. E-mail: dedmanjr(ajucbeh.san.uc.edu
ANNEXINS IN DORSAL ROOT GANGLIA AND SPINAL CORD
Nelsestuen, 1991). Ca2+-bindinginduces large conformational changes in the annexin molecules (Concha et al.,
1993). The N-termini are diverse and unique for each
annexin. This domain can be modified by alternative splicing and tyrosine and serinelthreonine phosphorylation
(Johnson et al., 1988; Gerke, 1992). In annexin 11, the
N-terminal domain also contains the binding site for a
unique subunit, p l l (Johnson et al., 1988). Annexin I1
forms a heterotetramer that binds F actin and phospholipid
vesicles (Weber, 1992). The conserved repeating segments
of the annexins contain the common Caz+-and phospholipidbinding sites, whereas the various N-termini are involved in
the mediation of the biological function(s) specific to each
annexin (Moss, 1992).
Knowledge of the individual expression pattern of each
annexin at the cellular and subcellular levels would allow
for the correlation of localization with cellular function.
There has been no thorough analysis of annexins I-VI in
the spinal cord and the dorsal root ganglia of any species
reported thus far. In the present study we identify the
subcellular localization of annexins I-VI in the mature rat
spinal cord and dorsal root ganglia (DRG).
MATERIALS AND METHODS
Purification of individual annexins
The purification of individual annexins was performed
according to Kaetzel et al. (1989) by using rat liver obtained
from PelFreeze. Crushed frozen tissue, 100 g, was homogenized 1:5 in a buffer containing (in mM): 20 Tris-HC1 (pH
7.41, 5 ethylenediamine-tetraaceticacid (EDTA), 7 p-mercapto-ethanol, 400 NaC1, 0.02% NaN3, and 1phenylmethyl
sulfonyl fluoride (PMSF) in a Waring blender. The homogenate was centrifuged at 8,000g for 20 minutes. The
supernatant was passed through chilled glass wool to
remove solidified surface lipid and was centrifuged at
100,OOOg for 75 minutes. The supernatant was dialyzed
overnight into column buffer consisting of (in mM): 20
Tris-HC1(pH 7.4), 1ethylene glycol-bis(P-aminoethyl ether)
N,N,N',N'-tetraacetic acid (EGTA), 1Mg-acetate, 7 p-mercapto-ethanol, and 0.02% NaN3. The dialyzate was clarified
by centrifugation at 8,OOOg for 20 minutes and applied to a
DEAE cellulose column and washed with column buffer.
The column was batch eluted with buffer containing 300
mM NaC1. Annexins I and I1 are present in the DEAE
unbound material, and annexins 111-VI are present in the
0-300 mM NaCl eluted material. Each pool was adjusted to
200 pM of free Caz+,applied to a phenyl-Sepharose column,
washed, and eluted by Caz+chelation (Kaetzel et al., 1989).
Annexins 111-VI were separated from one another by
chromatography on a Mono Q FPLC. Annexins I and I1
were separated on a Mono S FPLC (Pharmacia-LKB).
Monospecific antibody production
Sheep polyclonal antibodies were produced against rat
liver annexins (I, 111-VI) and against pig lung annexin I1 (C.
Balch, University of Cincinnati, unpublished) according to
Kaetzel et al. (1989). Rabbit polyclonal antibodies were
produced against chicken heart annexin VI by using pure
lyophilized protein to immunize rabbits (Naciff, unpublished). Annexin family members share amino acid sequence homology, particularly within the repeated domains, and antibodies that recognize these consensus
sequences must be eliminated to obtain monospecific anti-
357
bodies. Monospecific antibodies to each annexin were
achieved by affinity chromatography. To remove antibodies
that cross react with other annexins, affinity columns of
each respective annexin were prepared by coupling purified
protein to cyanogen-bromide-activated Sepharose (Pharmacia-LKB) according to the manufacturer's instructions.
Antibodies that could cross react with other annexins were
removed by passing the immune serum through columns of
each annexin (I-VI) in such an order so that the last column
contained the annexin against which the serum had been
produced. This final column was washed with sodium
bicarbonate buffer (0.1 M NaHC03, pH 8.3, 0.5 M NaCl)
until the ODzsowas less than 0.005. The affinity column
was then eluted with 0.2 M glycine (pH 2.7). Fractions were
collected into tubes containing a predetermined amount of
1M Tris base immediately to neutralize the antibody to pH
7.4. The pooled fractions were dialyzed against borate
saline (100 mM boric acid, 25 mM sodium borate, and 75
mM NaCl, pH 8.4) for storage at 4°C. The high-affinity
monospecific antibodies obtained through this procedure
recognize only the specified annexin when tested against
annexins I-VI and detect a single antigen by immunoblot of
total extracts prepared from rat brain, rat and mouse spinal
cord, chicken heart, and mouse intestine and lung.
Indirect immunofluorescence of tissue sections
Adult Sprague-Dawley rats (250 g) were deeply anesthetized with pentobarbital and perfused transcardially with
10% formalin in phosphate buffer saline (PBS), pH 7.4.
Spinal cord and DRG were dissected and postfixed by
immersion in 10%formalin in PBS, pH 7.4, for 48 hours at
4"C, washed with seven changes of PBS, and embedded in
paraffin, after which 4-pm-thick sections were obtained.
For immunostaining, sections were sequentially deparaffinized with xylene and ethanol, hydrated with PBS, and
incubated with 1:10 dilution of nonimmune rabbit serum (1
hour at 37°C) to block nonspecific binding. The sections
were then incubated with 2 pg of the respective sheep
anti-annexin antibody (2 hours at 37"C), washed in PBS,
and incubated further with fluorescein-conjugated rabbit
anti-sheep IgG or goat anti-rabbit IgG (30 minutes at 37°C).
The anti-bovine glial fibrillary acidic protein (GFAP) antibody (Dakopatts; Glostrup, Denmark) was used to identify
astrocytes in the same tissue sections that were used for the
immunolocalization of annexin 111. These sections were
first washed extensively in PBS, then in 200 mM glycine, 75
mM NaCl, pH 2.7 for 10 minutes to eliminate the immunocomplexes from the tissue. The sections were then washed
three times with PBS. Nonspecific binding was blocked
with nonimmune goat serum before the sections were
incubated with 0.3 pg of purified immunoglobulin fraction
of rabbit anti-bovine GFAP for 2 hours at 37°C. The
sections were then washed with PBS and incubated further
with fluorescein-conjugated goat anti-rabbit IgG (30 minutes at 37°C). Records of the immunostaining pattern
found in tissue sections were photomicrographed on Kodak
T-MAX ( M A 100) and Kodachrome film by using a Nikon
Optiphot epifluorescence microscope.
Sodium dodecyl sulfate-polyacrylamide
gel electrophoresis
The spinal cord was dissected from male Sprague-Dawley
adult rats under halothane anesthesia and cut into ventral
and dorsal areas at the level of the central canal. The tissues
were then homogenized 1:20 (w/v) in tissue solubilization
J.M. NACIFF ET AL.
358
Total
Protein
Stain
V
D
AI
V
D
A 11
L
V
D
A 111
I
V
D
A IV
V
D
AV
V
D
AVI
V
D
Fig. 1. Immunoblot detection of annexins of the rat spinal cord.
Total spinal cord homogenates (10 kg of protein per lane) from the
ventral (lanes indicated by V) or the dorsal (lanes indicated by D) areas
were applied to a 12% sodium dodecyl sulfate-polyacrilamide minigel.
Following electrophoresis, the first two lanes of the gel were stained
with Coomassie Blue (total protein stain). The proteins were electroblotted onto a sheet of nitrocellulose. Six pairs ofequivalent lanes (representing the ventral and the dorsal areas of the spinal cord) were cut, and
each strip was then probed with affinity-purified monospecific antibodies anti-annexin I-VI, respectively, followed by peroxidase-conjugated
rabbit anti-sheep IgG. The immunocomplexes were visualized by using
4-chloro-1-naphthol as a color substrate. The monospecificity of antiannexin I and I1 antibodies was evaluated by immunoblot analysis of
richer sources of each annexin, mouse lung (L) and intestine (I). In each
case, each antibody recognized one single antigen.
buffer consisting of 0.25 M Tris-HC1 (pH 6.8), 5 mM EDTA,
6% sodium dodecyl sulfate (SDS), 20% glycerol, and 5%
2-P-mercapto-ethanol at 60°C. The homogenate were further heated at 95°C for 10 minutes and stored at -70°C
until use. SDS-polyacrylamide gel electrophoresis was performed by using the buffer system of Laemmli (1970) on
12%(wiv)polyacrylamide gels. Samples were electroblotted
onto nitrocellulose filters after electrophoresis, as described
by Towbin et al. (1979). Individual annexins were immunodetected from blotted proteins with affinity-purified monospecific sheep anti-annexin antibodies, followed by peroxidase-conjugated rabbit anti-sheep IgG antibodies. The
immune complex was visualized with 4-chloro-1-naphthol
as a color substrate.
the spinal cord. The amount of annexins I and I1 in the
spinal cord (per mg of total protein) is below the immunodetection level in the blot (Fig. 1)when using 10 pg of total
protein per lane. The monospecificity of the anti-annexins I
and I1 antibodies was evaluated by immunoblot analysis of
two rich sources of these annexins, mouse lung and intestine. In each case, each antibody recognized a single antigen
(Fig. 1). These monospecific antibodies to each annexin
were used to determine the cellular and subcellular distribution patterns of annexins I-VI in rat DRG and spinal cord.
RESULTS
The affinity-purified monospecific antibodies used in this
study recognize only the specified annexin by immunoblot
analysis of total tissue extract of rat spinal cord. The
monospecificity was also obtained by immunoblot analysis
of whole rat brain, mouse spinal cord, intestine, and lung
(Fig. 1).The tissue distribution of the different annexins is
similar at the level of the ventral versus the dorsal areas of
Differential expression of annexins in the DRG
It has been established that the DRG consist of a
neuronal population with wide functional diversity. In any
given species, every DRG contains over 20 functionally
distinct types of neurons that innervate different tissues
(Perl, 1992). The morphological classification of sensory
neurons has favored two main subgroups: large, light
type-A neurons and small, dark type-B neurons (Lawson,
1992). These two main classes of primary sensory neurons
have been subdivided into several subclasses based on
ultrastructural and immunocytochemical criteria (Rambourg et al., 1983; Philippe and Droz, 1988; Lawson, 1992).
For example, in the rat DRG, Rambourg et al. (19831, by
ANNEXINS IN DORSAL ROOT GANGLIA AND SPINAL CORD
359
evaluating the three-dimensional organization of mem- Both types of glial cells in the DRG are rich in annexin IV,
brane-bound organelles, described six subclasses of sensory where it is mainly localized in the cytoplasm. Satellite cells
neurons. In the present study, we use the classification of (Fig. 3B, white arrow) surrounding the sensory neurons are
large (35-50 km in diameter) and small (15-30 pm in richer in annexin IV than the neurons are (Fig. 3B, black
diameter) sensory neurons.
arrow). In the sensory neurons, this annexin is localized in
Annexin I. As shown in Figure 2, annexin I is present at high concentrations within the nucleus (Fig. 3C) but absent
a relatively high concentration in small sensory neurons from the nucleolus. In the cytoplasm of the neurons,
throughout the DRG, with a perinuclear distribution that annexin IV is found in lesser amounts than in the nucleus
follows a pattern similar to that shown by the endoplasmic and is distributed evenly in a granular pattern (Fig. 3B,
reticulum (Fig. 2B,C, black arrows). Annexin I is also black arrow). It is primarily confined to the cellular body,
concentrated along the neurolemma (Fig. 2C). This annexin although there is weak immunoreactivity in the peripheral
seems to be restricted to the soma; it is not immunodetected and central branches of the axons. Annexin IV immunostainat the axon length, neither toward the central branch nor ing is also seen in cells of the connective tissue layer that
toward the peripheral branch of the DRG neurons (Fig. 2A), surrounds the DRG (Fig. 3A,C), although it is not detectalthough annexin I may be present at the end of the able in either the perineurium surrounding axon bundles or
peripheral branches, which could terminate in skin, muscle, endothelial cells lining the capillary walls (Fig. 3C).
or other tissues as free nerve endings or in association with
Annexin V. This annexin is highly expressed in most of
specialized connective tissue or epithelial cells that consti- the small sensory neurons of the DRG (Fig. 3D). In very few
tute the sensory afferent. The same is true for all the other large neurons, this protein is also expressed in relatively
annexins evaluated in our study. Annexin I is not present in high amounts (Fig. 3F). In those cells, annexin V is found
glial cells, endothelial cells lining the capillary tubes, cells heavily concentrated around the nucleus and follows a
that form part of the connective tissue layer surrounding distribution pattern similar to that of the endoplasmic
the DRG, or in the perineurium that surrounds axon reticulum. It is also found at the plasma membrane. In
bundles within the DRG (Fig. 2A,B). Some large sensory some small sensory neurons, annexin V is within the
neurons have low levels of annexin I (Fig. 2B,C, white vs. nucleus although segregated from the nucleolus (Fig. 3E,F,
black arrows) distributed in a punctuate pattern through- arrows vs. asterisks). In most of the large sensory neurons,
out the cytoplasm and in some neurons being more concen- annexin V is found in relatively low amounts, with a
discrete punctuate distribution pattern throughout the
trated at the plasma membrane (Fig. ZC, arrowhead).
Annexin II. Figure 2 shows the distribution pattern of cytoplasm. It appears more concentrated around the nucleus
annexin I1 in the DRG. Although most sensory neurons (Fig. 3F, cell with the asterisk). Again, in some large
express this annexin, regardless of size (Fig. 2D,E), annexin neurons, this annexin is found inside the nucleus (Fig. 3F).
I1 is highly concentrated in small sensory neurons (Fig. 2F, Annexin V immunoreactivity is restricted to the cellular
white vs. black arrowhead). In some neurons, annexin I1 body and is not detected at either the peripheral or the
has a preferential perinuclear distribution; in most, how- central branches of the sensory axons. It is also absent from
ever, it is found throughout the cytoplasm in a granular endothelial cells, glial cells, and the perineurium. Cells from
pattern. Endothelial cells lining the blood vessels and the outer connective tissue layer of the DRG also express
capillaries and smooth muscle cells forming the walls of annexin V (Fig. 3D,F).
Annexin VZ. As can be seen in Figure 3, most of the
blood vessels are also rich in annexin I1 (Fig. 2F, arrows).
Some satellite cells also express annexin I1 (Fig. 2E, arrow- sensory neurons throughout the DRG express annexin VI
head). This annexin is expressed by Schwann cells ensheath- in relatively high concentrations, regardless of cellular size.
ing the axons and by cells of the connective tissue layer Some of the large sensory neurons (Fig. 3G, arrowhead),
surrounding the DRG (Fig. 2D,F). Annexin I1 is absent however, have less annexin VI immunoreactivity than
from the axons and from the cells of the perineurium neighboring smaller cells (Fig. 3G, arrow). In the cytoplasm, this annexin is also found with a perinuclear distrisurrounding the axon bundles (Fig. 2D, arrow).
Annexin 111. This annexin is localized in most of the bution, following the pattern of the endoplasmic reticulum.
sensory neuronal bodies, regardless of cell size (Fig. 2G-I), It is also concentrated at the plasma membrane (Fig. 3H,
although it is highly concentrated in small neurons (Fig. arrowhead).Annexin VI is found along the axolemma of the
ZH, white vs. black arrowhead). With the exception of the central and peripheral branches of the sensory axons,
nucleus, annexin I11 is distributed evenly throughout the particularly the ones of large caliber (Fig. 31, arrow and
cytoplasm of small neurons and in a granular pattern in arrowhead). The perineurium and the Schwann cells enlarge neurons. It is also concentrated in close proximity to sheathing the axons (Fig. 31) contain annexin VI. Satelite
the neurolemma (Fig. 21). This annexin is also highly glial cells and endothelial cells do not show annexin VI
concentrated in endothelial cells lining small blood vessels immunoreactivity .
(capillaries; Fig. 21, arrow) and in the perineurium that
surrounds axon bundles (Fig 2G, arrow). In contrast with
Differential expression of annexins
annexins I and 11, annexin I11 is present within the axons
in the spinal cord
(Fig. 21) but at a lesser amount than in the soma. Glial cells
Although the spinal cord segments at various levels
also contain annexin 111, particularly Schwann cells ensheathing the axon (Fig. 21) but at relatively lower amounts including the cervical, thoracic, lumbar, and sacral regions
than in the sensory neurons. Cells from the connective show variations in size, shape, and topography of gray and
tissue layer surrounding the DRG are also rich in annexin white matter, the distribution pattern of each annexin at
the cellular and subcellular levels is conserved along the
I11 (Fig 21, asterisk).
Annexin ZV. The distribution pattern of annexin IV in different segments of the cord.
Annexin 1. This annexin is expressed in relatively high
DRG is very different than that of the other annexins (Fig.
3). Peripheral glial cells are separated into two distinct amounts in small and large neurons at their soma and
groups: satellite cells and Schwann cells (Pannese, 1974). processes found throughout the gray matter (Fig. 4A-C).
A1
A I1
A I11
Figure 2
ANNEXINS I N DORSAL ROOT GANGLIA AND SPINAL CORD
Annexin I is also expressed by the smooth muscle cells of
the blood vessels and by some cells of the meningeal
membranes that surround the spinal cord (Fig. 4A, arrow).
In the gray matter, annexin I immunoreactivity was also
localized in ependymocytes lining the central canal (Fig. 4A,
CC) and in smooth muscle cells of blood vessels. In the
neurons, this annexin is found throughout the cytoplasm
around the nucleus (Fig. 4B,C). It is excluded from vesiclelike structures (Fig. 4B,C). In the ventral horn areas,
mostly large motor neurons express annexin I in relatively
high concentrations throughout the cytoplasm of the cellular bodies (Fig. 4C) and the cytoplasm of neuronal processes
(Fig. 4C, arrow). Annexin I is also expressed by oligodendrocytes ensheathing the axons, particularly axons entering
the spinal cord through the dorsal roots. This annexin is
not immunodetected in glial cells and astrocytes located
along the different levels of the spinal cord.
Annexin ZZ. This annexin is expressed in high concentration by glial cells throughout the spinal cord (Fig. 4D-F),
although the strongest immunoreactivity is found in the
white matter, particularly at the funiculus dorsalis and
funiculus lateralis (Fig. 4E) and in endothelial and smooth
muscle cells of blood vessels and capillaries (Fig. 4D). Some
large neurons located at the ventral horn areas (Fig. 4F)
contain annexin I1 within the nucleus (Fig. 4F, arrow). It is
excluded from the nucleolus and the plasma membrane.
Annexin I1 is also expressed by oligodendrocytes surrounding axons and by ependymal cells lining the central canal
(Fig. 4D, CC). This annexin is not expressed by astrocytic
processes that surround the fascicles of axons.
Annexin ZIZ. This protein has a striking distribution
along the spinal cord. It is highly concentrated in astrocytes
located throughout the cord (Fig. 4G-I). The identity of
these cells was confirmed by coimmunodetection of the
GFAP (Fig. 4G and inset, 1 4 ) .This antigen is an astrocytic
marker (Eng and Ghirnikar, 1994; Sajin and Steindler,
1994). Annexin I11 is also abundant in endothelial cells
lining the blood vessels and capillaries (Fig. 4G,H, arrowheads) and in the astrcocytic processes that surround axon
bundles (Fig. 4G and inset, arrows). Small amounts of this
annexin were also found at the neurolemma of large spinal
cord neurons of the ventral horn (Fig. 4H). In these
neurons, annexin I11 is not restricted to their soma as it is
present within the axons and dendrites. This immunostain-
361
ing pattern could be due to the annexin I11 being present in
thin astroglial lamellae that surround motor neurons.
Axons grouped in bundles contain annexin I11 in relatively
high concentrations, subjacent to the axolemma (Fig. 4G).
Neither ependymocytes lining the central canal nor oligodendrocytes showed annexin I11 immunoreactivity.
Annexin IV. As shown in Figure 5 , annexin IV is found
primarily in oligodendrocytes ensheathing the axons. This
annexin is also present in ependymal cells of the central
canal and in glial cells throughout the white and gray
matter of the spinal cord (Fig. 5A,B). Annexin IV,although
in low levels, is also expressed by large neurons of the
ventral horn areas, where it is mainly localized along the
plasma membrane of the cell body and associated processes
(Fig. 5C,arrow). Astrocytic processes that surround axon
bundles contain annexin IV (Fig. 5B, arrow). Blood vessels
do not express annexin IV at immunodetectable levels.
Annexin V. As shown in Figure 5, this annexin is highly
expressed by ependymal cells lining the central canal of the
spinal cord (Fig. 5D). This annexin is also expressed by
cellular elements located in the superficial laminae of the
dorsal horns (Fig. 5E, SL). We did not observe the same
staining pattern with the anti-GFAP antibody (data not
shown), which suggests that the immunostaining may be
localized in fine-diameter axons of the superficial dorsal
horn, which are abundant in this area. Among the small
elements expressing annexin V at the superficial laminae of
the dorsal horn, there are larger cells (presumably neurons
or GFAP-negative cells) that do not express this annexin.
Annexin V is also expressed by oligodendrocytes surrounding the axons entering the spinal cord through the dorsal
root (Fig. 5E, arrow). Annexin V is found in smaller
amounts at the plasma membrane of large neurons located
at the ventral horns, where it seems to be confined to the
cellular body (Fig. 5F). Not all large neurons located in this
area are immunoreactive with the annexin V monospecific
antibody (Fig. 5F, white vs. black arrow). Meningeal elements ofthe spinal cord express annexin V (Fig. 5D, arrow).
Within the gray matter and outside of the superficial
laminae of the dorsal horns, this annexin is not immunodetected in astrocytes or in other glial cells a t any spinal cord
level.
Annexin VI. Annexin VI is highly concentrated at the
plasma membrane of most of the large neurons of the spinal
cord, particularly in the ventral horn regions, at all levels of
the cord (Fig. 5G, lumbar region, and 5H, thoracic region).
The identity of the annexin VI immunoreactive neurons
Fig. 2. Immunofluorescent localization of annexins 1-111 in rat
dorsal root ganglia (DRG). Monospecific antibodies against annexins located at the ventral horn regions was confirmed by
retrograde labeling with the fluorescent dye True Blue
1-111 were used to stain 4-km tissue sections obtained from dorsal root
ganglia and were processed as described in Methods. Annexin I (A I) is
(Naciff et al., manuscript submitted). The colocalization of
expressed in relatively high concentrations by small sensory neurons
True Blue and annexin VI in the same neurons of the
(A,B,C, black vs. white arrow) that are found in groups (B, black
ventral horn confirmed these cells to be motor neurons.
arrow). Large neurons lack annexin I or have very low amounts of it
(white arrows in B,C); some cells segregate this annexin in close Annexin VI is not restricted to the cellular bodies of the
proximity with their plasma membrane (C, arrowhead). Annexin I1 (A motor neurons; it is also abundant in their dendrites and
11) is expressed by most sensory neurons, regardless of cellular size (D), axons, mostly confined at the neurolemma (Fig. 5G,H).
although it is relatively more concentrated in small neurons (F,white Axons of the DRG sensory neurons that enter the spinal
vs. black arrowhead). It is also highly expressed by endothelial cells cord through the dorsal root contain high levels of annexin
lining the blood vessels and capillaries (F, arrows) and in Schwann cells
ensheathing the axons. Some satellite cells also express this annexin (E, VI (Fig. 51). Ependymal cells lining the central canal and
arrow). The perineurium lacks annexin 11 (D, arrow). Annexin 111 (A cells from the meninges show weaker annexin VI immunoreactivity when compared with neurons and their pro111) is also expressed widely in sensory neurons, regardless of cellular
size ( G ) ,although it is relatively more concentrated in small neurons cesses. This annexin is not immunodetected in astrocytes,
(H, white vs. black arrowhead). It is also expressed by epithelial cells oligodendrocytes, or in other glial cells or in endothelial
lining the blood vessels and capillaries (I, arrow) and by cells from the
cells lining the blood vessels.
perineurium (G, arrow) and from the connective tissue layer surroundThe spatial expression pattern of annexins I-VI in the rat
ing the DRG (I, asterisk). Scale bars = 100 pm for A,D,E,G, 50 p m for
DRG and spinal cord is summarized in Table 1.
B,C,F,H,I.
A IV
AV
A VI
Figure 3
ANNEXINS IN DORSAL ROOT GANGLIA AND SPINAL CORD
DISCUSSION
363
in the spinal cord. McKanna and Cohen (1989) found that
annexin
I was expressed mostly by primitive glial ependyWe have identified annexins I-VI in the rat DRG and
spinal cord by using monospecific antibodies. This study is mal cells of the floor plate of the embryonic rat central
the first report that addresses the identification of these nervous system (CNS). They also found that the raphe
annexins in cells of the DRG and spinal cord in any species. annexin I immunostaining disappears by postnatal day 5 .
These data offer the first approach to elucidate the role of Using immunohistochemical techniques, these investigaindividual annexins in the intracellular pathways used by tors did not detect annexin I in any other region of the CNS
neurons to mediate and discriminate the intracellular Ca2+ in either rat embryos or adults. In contrast to our results
signal. The identification of the molecules that constitute (Fig. 21, Hamre et al. (1995) did not find annexin I
the cells of the nervous system and insights into their immunoreactivity in any neuronal structure of the adult
function are requirements for understanding normal and rat or mouse CNS. They found that the only structure
pathological processes specific to the nervous system. To within the developing or mature murine CNS labeled by
date, more than 1,000 different neurological syndromes their anti-annexin I antibody was the midline raphe of the
have been described. Of these only 40 genetic disorders brainstem; they did not detect annexin I in the spinal cord
affecting the central and peripheral nervous systems have or the DRG. Bolton et al. (1990) and Elderfield et al. (1993)
been characterized (Heizmann and Braun, 1992; Coyle and found annexin I immunoreactivity in the cervical spinal
Puttfarcker, 1993; Martin, 1993). The identity of the cord of Lewis rats, which increased with acute experimental
proteins involved in most neurological disorders, however, allergic encephalomyelitis (EAE). In these studies, normal
is still unknown. Modifications in specific Ca2+-binding animals showed moderate annexin I immunoreactivity in
proteins have been found in human neurodegenerative the walls of blood vessels and capillaries and in heavily
disorders such as Alzheimer’s disease, Parkinson’s disease, stained cells within the lumen of blood vessels. In contrast,
and epilepsy (Heizman and Braun, 1992). In this context, the EAE-diseased animals had increased amounts of anthe identification of members of the annexin family in nexin I at the inflammatory lesions, which contained many
spinal cord and DRG may provide insight into the patho- lymphocytes and macrophages, cells rich in annexin I
(Blackwell et al., 1982; Flower, 1988; Johnson et al., 1988).
physiology of some neurodisorders.
We did not observe any enrichment of annexin I immunoreAnnexin I
activity within the blood vessels, either in the spinal cord or
In the DRG, annexin I was found expressed in relatively in the DRG, because the animals that we used were healthy,
high concentrations in small sensory neurons, with a and they were saline perfused transcardially before fixing
perinuclear distribution that follows a pattern similar to the tissues; hence, cellular elements of the blood were
that shown by the endoplasmic reticulum, and concen- absent. Also using immunohistochemistry, Johnson et al.
trated at the neurolemma of the cellular bodies. In the (1989a,b) reported that, in the normal human CNS, anspinal cord, annexin I is expressed in relative high amounts nexin I immunoreactivity is localized primarily in ependyin the neurons and in relative lower concentrations in glial ma1 cells lining the lateral and fourth ventricles and the
cells, endothelial and smooth muscle cells of blood vessels, central canal throughout the spinal cord, in scattered
and ependymal cells lining the central canal. Although the subependymal astrocytes throughout the ventricular syspresence of annexin I in rat spinal cord has been reported tem, and in choroid plexus epithelia. In addition, in studies
previously, this is the first time that annexin I has been on brain tissue from patients with various CNS diseases,
identified in sensory neurons of the DRG and in neurons these investigators also detected annexin I in reactive
from the spinal cord, particularly motor neurons. Fava et astrocytes and infiltrating macrophages surrounding inal. (1989) detected annexin I in different rat organs, farcted areas in patients with lesions caused by various
including the brain; they were unable to detect this annexin conditions. Annexin I has been found to have pharmacological activity in the CNS. When exogenous annexin I (human
recombinant lipocortin I) was administrated intravenously
into the rabbit, it prevented febrile reactions (Davidson et
al., 1991). Furthermore, intracerebroventricular injection
Fig. 3. Immunofluorescent localization of annexins IV-VI in rat
of annexin I inhibits cytokine induced fever in the rat
dorsal root ganglia. Monospecific antibodies against annexins IV-VI
(Carey et al., 1990). Relton et al. (19911, using a model of
were used to stain 4-pm tissue sections obtained from dorsal root
ganglia. Annexin IV (A IV) is mainly expressed by satellite glial cells (B, cerebral ischemia, observed increased expression of anwhite arrow) surrounding the sensory neurons (A,B) and by Schwann nexin I around infarcted areas: a cerebral injection of an
cells ( C , arrowhead), where it is evenly distributed throughout the exogenous N-terminal fragment of annexin I was able to
cytoplasm. In sensory neurons (B, black arrow), this annexin is
expressed in relatively lower amounts than in satellite cells, and it is reduce infarct size, and the administration of neutralizing
antibody against annexin I increased the extent of ischemic
mainly segregated into the nucleus (C) but absent from the nucleolus.
Annexin V (A V) is heavily expressed by small sensory neurons (D). injury and edema. The same group (Black et al., 1991,1992)
Although very few large neurons express this annexin in high concentra- claimed that injection of the N-terminal fragment of antions, most of them express it in lower amounts (El, with a distinctive nexin I inhibited neuronal damage mediated via the NMDAgranular pattern throughout the cytoplasm, and concentrated around receptor in the striatum of the rat and suggested that
the nucleus (F). Some neurons segregate annexin V within their
annexin I could be used as a therapeutic agent in the
nucleus (arrow in E,F), but most of them do not express it within this
organelle (asterisk in E,F). Most of the sensory neurons express treatment of excitotoxic cell death or damage. Recently,
annexin VI (A VI) in relatively high concentrations, regardless of Mullens et al. (1994) found that the annexin I and its
cellular size (GI, although smaller cells seem to be richer in this protein
mRNA are upregulated in reactive astrocytes in kainate( G , arrow vs. arrowhead). This annexin is also found along the axon lesioned rat cerebellum, but not in neurons. The identificalength and is particularly abundant in axons of large diameter (I, tion of annexin I function(s) and its pharmacological proparrowhead). In the soma and axons, annexin VI is located at the plasma
membrane or very close to it (arrow in HJ). Scale bars = 100 pm for erties in CNS may have some relevance for some human
neurological disorders.
A,D,G, 50 km for B,C,E,F,H,I.
A1
A I1
A I11
Figure 4
ANNEXINS IN DORSAL ROOT GANGLIA AND SPINAL CORD
Annexin I1
In the mature rat, annexin I1 is expressed by most of the
sensory neurons of the DRG, although it is more concentrated in small neurons. In contrast, in the spinal cord,
annexin I1 is expressed mostly by glial cells throughout the
white matter and in very low amounts by neurons, where it
is localized mostly within the nucleus and excluded from
the nucleolus. This annexin is also highly expressed by
endothelial cells of the blood vessels and capillaries, in
spinal cord and in DRG, and by ependymal cells lining the
central canal. Hamre et al. (1995) immunodetected the
expression of annexin I1 in the floor plate of the spinal cord,
brainstem, mesencephalon, and in most dorsal root and
sensory ganglion cells of the developing mouse CNS. In
agreement with our results in rat DRG, Hamre et al.
immunodetected annexin I1 in DRG sensory neurons of all
size classes, in the dorsal horn of the spinal cord, in the
solitary nucleus, and in blood vessels in adult mouse. Also
in agreement with the present study, Greenberg et al.
(1984) immunodetected the expression of annexin I1 in
chicken DRG sensory neurons, in specific brainstem sensory nuclei, and in endothelial cells of blood vessels and
fibroblasts in the meninges. They did not detect annexin I1
in adult avian nervous tissue. Annexin I1 has been the focus
of considerable interest due to the fact that its amino
terminal domain contains phosphorylation sites for ~60"'"
tyrosine kinase and protein kinase C (Glenney and Tack,
1985; Gould et al., 19861, kinases that are important in
multiple signal transduction pathways. This annexin is able
to interact, through its amino terminal domain, with a
small subunit, p l l , to form tetramers. The annexin 11-pll
heterotetramer has a strong affinity for the plasma membrane and the subjacent cytoskeleton (Powell and Glenney,
1987; Johnson et al., 1988; Semich et al., 1989; Thiel et al.,
1992). This complex has been associated with regulation of
Caz+-dependent exocytosis (Sarafian et al., 1991) and has
been identified as a major component of fusogenic endosoma1 vesicles (Emans et al., 1993). Recently, it has been
shown that the heterotetramer contributes to early endo-
Fig. 4. Immunoflurescent localization of annexins 1-111 in rat spinal
cord. Monospecific antibodies against annexins 1-111 were used to stain
4-pm tissue sections obtained from the lumbar enlargement of adult rat
and processed as described in Methods. The distribution of each
annexin, at the cellular and subcellular levels, is conserved along the
different segments of the cord. Annexin I (A I) is primarily expressed by
neurons located throughout the gray matter (A-C). Ependymal cells
lining the central canal (CC in A) and meningeal membranes (A, arrow)
also express annexin I. Large neurons of the ventral horn areas express
this annexin in relatively high levels (C) not only at their soma but also
at their processes (C, arrow). Annexin I1 (A 11)is expressed primarily by
ependymal cells lining the central canal (D) and by glial cells throughout the white and gray matter (D,E).This annexin is also expressed by
endothelial cells lining the blood vessels. Large neurons of the ventral
horns also express this annexin although in smaller amounts and
mostly localized within the nucleus (F, arrow), being excluded from the
nucleolus (F, arrowhead). This annexin is not associated with the
plasma membrane of these neurons at immunodetectable levels (F).
Annexin I11 (A 111) is expressed in relatively high concentrations by
astrocytelike cells found throughout the white and gray matter; these
cells express the glial fibrillary acidic protein (inset in G and J), an
astrocytic marker. Annexin 111is also expressed by astroglial processes
that surround axon bundles (G and inset, arrows) and by endothelial
cells lining the blood vessels (arrowheads in G,H). Large neurons of the
ventral horn (VH) areas also express annexin 111 (HI but in lower
amounts than the astrocytes do (I).Scale bars = 100 pm for A,B,D,
E,G(inset),H,50 pmfor C,F,G,I,J.
365
some trafficking in the endocytic pathway (Harder and
Gerke, 1993). The selective expression of annexin I1 that we
have found in the spinal cord and DRG cells makes it
difficult to assign its role in a single indispensable cellular
function that would occur in every mammalian cell. More
likely, annexin I1 and other individual annexins may have
unique cellular functions required by specialized cellular
types.
Annexin I11
In the DRG, annexin I11 is expressed by most sensory
neurons, endothelial cells lining the blood vessels, and cells
that form part of the perineurium. In the spinal cord, this
annexin is expressed in high concentrations by astrocytes
found throughout the white and the gray matter and in
very small amounts by large motor neurons. The identity of
the cells expressing annexin I11 was confirmed by immunolocalization of the astrocytic marker GFAP (Eng and
Ghirnikar, 1994; Sajin and Steindler, 1994) on the same
cells immunostained by the anti-annexin I11 antibody (compare Fig. 4G and inset with Fig. 41,J). This protein is a good
astrocytic marker because GFAP is a major component of
the intermediate filaments of astrocytes. In the astrocytes,
annexin I11 is evenly distributed throughout their cytoplasm, even in the cellular processes, whereas, in large
neurons, annexin I11 is only immunodetected in close
proximity to their neurolemma. This pattern suggests that
annexin I11 may not be present in the motor neurons
themselves because this annexin could actually be in the
thin astroglial lamellae that surround motor neurons. No
report on the cellular localization of annexin I11 in the CNS
has been published. Annexin I11 has been described as being
an enzyme involved in phosphoinositide metabolism. Ross
et al. (1990) reported that the amino acid sequence of
polypeptide fragments from purified fractions of the inosito1 1,2-cyclicphosphate 2-phosphohydrolase were identical
to annexin I11 sequence. This enzyme converts inositol
1,2-cyclic phosphate into a conventional noncyclic inositol
1-phosphate. Sekar et al. (19941, however, reported that
annexin I11 could be separated from cyclic inositol phosphohydrolase activity in the rat spleen and in guinea pig
kidney. They concluded that annexin I11 and phosphohydrolase are different proteins. Although these results remain to
be verified by others, they indicate that annexin I11 does not
necessarily have to be expressed with the universality of the
phosphoinositide system. Our results show that annexin I11
has a restricted expression pattern throughout the rat
spinal cord and DRG, which may indicate a more specific
function for this annexin.
Annexin IV
In the spinal cord and DRG, annexin N is primarily
expressed by oligodendrocytes (spinal cord) and by Schwann
cells ensheathing the axons and by glial cells found throughout the white and gray matter. It is also expressed at lower
levels by neurons. In glial and ependymal cells, annexin N
is mostly a cytoplasmic protein; in sensory neurons, annexin IV is particularly enriched within their nucleus but
excluded from their nucleoli. In the spinal cord neurons,
this annexin is located in close proximity to the plasma
membrane but not within the nucleus. The immunodetection of annexin IV in the nucleus of sensory neurons but not
in the nucleus of spinal cord neurons indicates that the
compartmentalization of this annexin is not due to an
artifact of the affinity-purified antibody used. No report
A IV
AV
A VI
Figure 5
ANNEXINS IN DORSAL ROOT GANGLIA AND SPINAL CORD
TABLE 1. Localization and Relative Annexin Expression Level in the Rat
Dorsal Root Ganglia (DRG) and Spinal Cord'
Annexin
Cell type
I
I1
I11
IV
V
VI
+t
+++
+++
++
+
++
+
+
+++
+++
+++
+
+
+
+++
++
+
+
+
-
+
+
-
DRG
sensory neurons2
small
large
satelite cells
Schwann cells
meningeal elements
endothelial cells
Spinal cord
neurons
motor neurons
glial cells3
ependymal cells
meningeal elements
endothelial cells
+
t
~
+
+t
++
++
++
++
+
+
++
+
+
+++
+
+
+
~
+++
+
+++
-
++
+++
-
+++
++
+
-
~
-
-
+
+++
+
+
+
+
+++
+++
+
+
'Relative expression level of each annexin was evaluated by judging the immunostaining
(+++)high.
intensity,notedby- (~)absent,(+)low,(++)moderate,
2The size of the sensory neurons is given according to their diameter. small ( 15 to 30
pm), and large (35 to 50 pmj.
301igodendrocytes express preferentially annexin IV, while astrocytes express mostly
annexin 111.
~
exists relating annexin IV and nuclear localization. Immunocytochemical and biochemical studies have, however,
described a nuclear localization of annexins I (Raynal et al.,
19921, I1 (Jindal et al., 19911, V (Sun et al., 1992; Koster et
al., 1993), and X (Mizutani et al., 1992). As in the present
report, Hamre et al. (1995) found that annexins I1 and IV
are expressed by neurons and glia, although they did not
describe any nuclear localization of any of the annexins.
Elderfield et al. (1992) detected annexins I, 11, IV and V
by Western blotting of postmortem CNS tissue (brain and
spinal cord) from multiple sclerosis (MS) patients and
normal controls. They pooled CNS samples into gray and
white matter, and both types of samples showed annexin I,
11,IV, and V reactivity. In white matter samples from MS
patients, all of these annexins were found to be significantly
elevated. This annexin increase might be due to perivascular infiltrate, which is rich in macrophages, lymphocytes,
and plasma cells in the samples from MS patients. Annexin
IV is also expressed in relatively high concentrations by
ependymal cells lining the central canal of the spinal cord. It
has been demonstrated that this annexin is expressed in a
Fig. 5. Immunoflurescent localization of annexins IV-VI in rat
spinal cord. Monospecific antibodies against annexin IV-VI were used
to stain 4-pm tissue sections obtained from the lumbar enlargement of
adult rat. Annexin IV (A IV)is expressed by glial cells and by ependymal
cells lining the central canal (A). Oligodendrocytes (B) and the astrocytic processes surrounding axon bundles (B, arrow) are also rich in this
annexin. Some large neurons from the ventral horns also express
annexin IV in lower amounts than glial cells do ( C ) .In these neurons,
annexin IV is only found at or very close to the plasma membrane (C,
arrow), with a punctuate pattern, and it is excluded from the nucleus
(C). Annexin V (A V) is expressed in relatively high concentrations by
elements located at the superficial laminae (SL) of the dorsal horns (El
that seem to be fine-diameter axons. It is also abundant in ependymal
cells lining the central canal (D) and in meningeal elements (D, arrow).
This annexin is also found in some large neurons of the ventral horns in
relatively lower concentrations (F, black vs. white arrow), where it is
mostly segregated to the plasma membrane. Annexin VI (A VI) is
primarily expressed by large neurons of the ventral horns (G,H) at all
levels of the cord. Lumbar (G) and thoracic (HI areas are shown. In
those neurons, annexin VI is heavily concentrated along the plasma
membrane of the soma and the cellular processes. The axons entering
in bundles into the spinal cord, through the dorsal root, are rich in this
annexin (I).Scale bars = 100 km for A,D-F, 50 krn for B,C,G-I.
367
variety of epithelia (Kaetzel et al., 1989, 1994), whose
functions include selective absorption and secretion across
a polarized cell layer. Kaetzel et al. (1994) reported that
annexin IV inhibits a calcium-activated chloride conductance in cultured epithelial cells. The expression of annexin
IV in glial and ependymal cells may suggest a similar
regulatory role of this protein in these cells. Ependymal
cells lining the central canal are epithelial cells with selective transport functions, which maintain a constant extracellular environment for neurons and glia along the cord.
While the apical surface of ependymal cells is in contact
with the cerebrospinal fluid, their basolateral surface is in
contact with the extracellular fluid that bathes neurons and
glial cells. Thus, these cells serve as a selective barrier to the
flow of ions and molecules across the two compartments. It
has been established that the chloride concentration in the
cerebrospinal fluid is higher than in the serum (Fishman,
1980). Glial cells are thought to serve different functions
including the buffering of potassium ions in the extracellular space, and some remove chemical transmitters released
by neurons during synaptic transmission. When neurons
fire repeatedly, K+ accumulates in the extracellular space.
Because of their high permeability, some glial cells absorb
excess K+, thereby protecting neighboring neurons from
the depolarization that might result from accumulated K+
ions. To maintain electrical neutrality, glial cells must also
transport C1-. These ion fluxes through the plasma membrane of glial cells may well be regulated by annexin IV.
Annexin V
In the DRG, annexin V is expressed in relatively high
concentrations in small sensory neurons. In the spinal cord,
annexin V is expressed in high concentrations in elements
located in the superficial laminae of the dorsal horns. This
annexin is also abundant in oligodendrocytes, ependymal
cells lining the central canal, and in some cells from the
meningeal coverings of the spinal cord and the DRG. In the
ventral horn areas, only large neurons express annexin V,
mostly localized in close proximity with the plasma membrane and at relatively low levels. As is the case with
annexin IV, some sensory neurons have annexin V concentrated within their nucleus; it is segregated from the nuclei
in spinal cord neurons. The fact that only some sensory
neurons showed immunostaining of their nuclei, but not of
their nucleoli, either with anti-annexin 11, IV,or V antibodies, suggests that these annexins are faithfully confined to
this compartment in specific cellular types. Furthermore, a
nuclear localization of annexin V has been reported in other
cell types (Sun et al., 1992; Koster et al., 1993). Annexin V
expression has been demonstrated in nervous tissues by
other researchers, although not in the spinal cord or DRG.
Woolgar et al. (1990) reported that annexins V and VI are
the two major annexins in porcine brain. In this tissue,
annexin V was found mainly associated with glial cells
bodies in the white matter of the cerebellar folia and the
molecular layer. They suggested that this annexin could be
the exclusive annexin of glial cells in the cerebellum. Spreca
et al. (1992) investigated the ultrastructural localization of
annexin V in the rat cerebellum and other tissues and found
that this annexin is restricted to glial cells, where it was
found diffusely in the cytoplasm and associated with the
plasma membrane. They reported that Bergmann glial cell
bodies and their processes, astrocytes in the cerebellar
cortex, oligodendrocytes in the cerebellar white matter, and
Schwann cells in the sciatic nerve express annexin V. In
368
J.M. NACIFF ET AL.
these studies, neuronal cell bodies, their dendrites, and budding of clathrin-coated pits, present in bovine brain
their axons did not show annexin V immunoreactivity. cytosol, is annexin VI. These researchers showed that
Giambanco et al. (1993) reported that annexin V accumu- annexin VI is not only an active component in the detachlates in the cytoplasm and the plasma membrane of a ment of the coated pits from the membrane but also a site
human-derived glioma cell line (GL15) during differentia- for regulation of the formation of clathrin-coated vesicles.
tion. Our results show that annexin V is not expressed Thus, if annexin VI is a key component of the mechanisms
exclusively by glial cells. Some sensory neurons are richer involved in the internalization of macromolecules by recepin this annexin than glial cells are, and motor neurons tor-mediated endocytosis, then all cells performing this
express low levels of annexin V. The physiological func- process must express annexin VI; this was not found in our
tion(s) of annexin V has not been established. There are study. The recent findings of Smythe et al. (1994) agree
some in vitro studies in which concrete functions have been with our results. They showed by several techniques that
proposed. It has been demonstrated that, when purified human A431 squamous carcinoma cells do not express
annexin V is incorporated into artificial acidic phospholipid annexin VI but are able to perform the internalization of
bilayers, it has the ability to act as a voltage-gated cation- transferrin, which is known to occur via the budding of
selective channel (Rojas et al., 1990; Pollard et al., 1992). clathrin-coated pits. The selective cellular expression of
Based on the electrophysiological data and the crystal- annexin VI in spinal cord and DRG may indicate a more
structure data (Huber et al., 1992a,b),it has been proposed specialized role for this protein than the universal endocytic
that annexin V may act as a cation channel in vivo, although process. Watanabe et al.(1994) identified the brain spectrin
this function remains to be verified in intact cells.
calspectin, a membrane skeletal protein, as one of the
proteins with the ability to interact with annexin VI in
Annexin VI
vitro. They found that this annexin inhibited the F actin
Our immunohistochemical studies indicate that annexin cross-linking activity of calspectin in a Ca2+-and phosphoVI is expressed by most sensory neurons of the DRG. In lipid-dependent manner. Their results suggest that ancellular bodies, this annexin is found in close proximity with nexin VI may be involved in the regulation of membrane
the plasma membrane and with intracellular structures skeletons of cells from the CNS in response to Ca2+.We
that are concentrated around the nucleus, with a distribu- have shown that annexin VI is not expressed by all cells of
tion pattern similar to that of the endoplasmic reticulum. the spinal cord and DRG, and this selective expression
The staining of sensory neurons with anti-annexin VI suggests specialized function(s)of annexin VI.
In conclusion, we have shown immunohistochemical
antibodies resembles the staining of Nissl bodies (data not
shown). It has been demonstrated that annexin VI is evidence for the cellular and subcellular localizations of
associated with endoplasmic reticulum membranes in lym- annexins I-VI in the rat spinal cord and DRG. At the
phocytes (Owens et al., 1984) and with sarcoplasmic reticu- cellular level, none of these annexins have similar distribulum in rat skeletal muscle (Hazarika et al., 1991). Further- tion patterns (Table 1). Our data show that none of the
more, Diaz-Mufioz et al. (1990)demonstrated that annexin annexins studied here is exclusive to a single cell type,
VI is a potent regulator of the calcium-release channel, although there is preferential expression of individual
isolated from sarcoplasmic reticulum of skeletal muscle and annexins depending on the individual cell. At the subcelluincorporated into artificial phospholipid bilayers. Whether lar level, all the annexins studied here are associated with
this annexin has a role in the modulation of intracellular the plasma membrane or structures adjacent to it. This
calcium handling in neurons or other cell types remains to finding suggests an active role for annexins in the Ca2+be established. Annexin VI does not seem to be expressed by dependent regulation of plasma membrane functions. In
glial cells throughout the DRG. We have demonstrated that addition, the annexins are also located in the cytoplasm,
annexin VI immunoreactivity is found primarily in the preferentially associated with structures that follow the
large motor neurons located in the ventral horn of the distribution pattern of the endoplasmic reticulum, suggestspinal cord. This apparent selective expression of annexin ing that these proteins could also participate in the regulaVI by motor neurons is accentuated by its absence in tion of cytoplasmic processes. The nuclear localization of
normal astrocytes, oligodendrocytes, and other glial cells annexins 11, IV, and V also suggest a possible role of these
and neurons of the spinal cord. The fact that this annexin is proteins in this cellular compartment. The differential
localized mainly at the neurolemma of the cellular bodies localization pattern of annexins in neurons and in glial cells
and in their dendrites and axons suggests a possible role of of the spinal cord and DRG could reflect different biological
annexin VI in plasma membrane activity. It is well known roles of individual annexins and their requirement in
that intracellular Ca2+ regulates multiple plasma mem- particular cell types for signal transduction of Ca2+.
brane processes, some of which might well be mediated by
The morphological information presented in this paper
individual annexin family members. Woolgar et al. (1990) allows for the design of experiments to investigate the
claimed that annexin VI immunoreactivity in cerebellum is physiological roles of individual annexins in spinal cord and
restricted to Purkinje cells. Clark et al. (1991) studied the DRG cells. The use of monospecific antibodies against
expression of annexin VI in normal human tissues by particular annexins to block their intracellular function by
immunohistochemical techniques and reported that an- whole-cell perfusion has been performed with annexin N
nexin VI was not present in the cortex, brainstem, or (Chan et al., 1994; Kaetzel et al., 1994). This technique can
cerebellum. They did not test other nervous tissues. Inter- be used to address the function of individual annexins in
estingly, Clark et al. (1991) found that annexin VI expres- nervous tissue cells. Another option is to functionally
sion is usually confined to specialized cell types within each neutralize individual annexins by using genetically engitissue, being highly expressed in most endocrine cells, and neered dominant mutants, as has been successfully done
that the staining is generally diffuse and cytoplasmic in with annexin I1 and its subunit p l l (Harder and Gerke,
those cells that express it. Lin et al. (1992) reported that a 1993). Another approach is to target overexpression, or
purified and active factor required for the final steps in the downregulation, of individual annexins in transgenic mice.
ANNEXINS IN DORSAL ROOT GANGLIA AND SPINAL CORD
Our findings provide a rationale to direct the appropriate
physiological evaluation of individual annexins in specific
cells of the spinal cord and DRG.
ACKNOWLEDGMENT
We are grateful to Dr. Curtis Balch (University of
Cincinnati) for his generous gift of monospecific antiannexin I1 antibodies.
LITERATURE CITED
Barton, G.J., R.H. Newman, P.S. Freemont, and M.J. Crumpton (1991)
Amino acid sequence analysis of annexin super-gene family of proteins.
Eur. J. Biochem. 198:749-760.
Bazzi, M.D., and G.L. Nelsestuen (1991) Proteins that bind calcium in a
phospholipid-dependent manner. Biochemistry 30:971-978.
Black, M.D., J.K. Relton, F. Carey, A.R. Crossman, and N.J. Rothwell(1991)
Lipocortin-1 inhibits NMDA-Induced neuronal damage. Br. J. Pharmacol. 104:C34.
Black, M.D., F. Carey,A.R. Crossman, J.K. Relton, andN.J. Rothwell(1992)
Lipocortin-1 inhibits NMDA receptor-mediated neuronal damage in the
striatum of the rat. Brain Res. 585:135-140.
Blackwell, G.J., R. Carnuccio, M. Di Rosa, R.J. Flower, C.S.J. Langham, L.
Parente, P. Persico, N.C. Russell-Smith, and D. Stone (1982) Glucocorticoids induce the formation and release of anti-inflammatory and antiphospholipase proteins into the peritoneal cavity of the rat. Br. J.
Pharmacol. 7:185-194.
Bolton, C., A.J. Elderfield, and R.J. Flower (1990) The detection of lipocortins 1, 2 and 5 in central nervous tissues from Lewis rats with acute
experimental allergic encephalomyelitis. J. Neuroimmunol. 29; 173-181.
Carey, F.O., R. Forder, M.D. Edge, A.R. Greene, M.A. Horan, P.J.L.M.
Strijbos, and N.J. Rothwell (1990) Lipocortin 1 fragment modifies
pyrogenic actions of cytokines in rats. Am. J. Physiol. 259:R266-R269.
Chan, H.C., M.A. Kaetzel, A.L. Gotter, J.R. Dedman, and D.J. Nelson (1994)
Annexin IV inhibits calmodulin dependent protein kinase 11-activated
chloride conductance. A novel mechanism for ion channel regulation. J.
Biol. Chem. 269:32464-32468.
Clark, D.M., S.E. Moss, N.A. Wright, and M.J. Crumpton (1991) Expression
of annexin VI (p68, 67 kDa-calelectrine) in normal human tissues:
evidence for developmental regulation in B- and T-lymphocytes. Histochemistry 96.405412.
Concha, N.O., J.K. Head, M.A. Kaetzel, J.R. Dedman, and B.A. Seaton
(1993) Rat annexin V crystal structure: Ca2+-inducedconformational
changes. Science 261:1321-1324.
Coyle, J.T., and P. Puttfarcker (1993) Oxidative stress, glutamate, and
neurodegenerative disorders. Science 262:689-695.
Crumpton, M.J. (1992) Preface. In S.E. Moss (ed): The Annexins. Chapel
Hill, NC: Portland Press, pp. v-vii.
Davidson, J., R.J. Flower, A S . Milton, S.H. Peers, and D. Rotondo (1991)
Antipyretic actions of human recombinant lipocortin-1. Br. J. Pharmacol. 102.7-9.
Diaz-Mufioz, M., S.L. Hamilton, M.A. Kaetzel, P. Hazarika, and J.R.
Dedman (1990) Modulation of Ca2+ release channel activity from
sarcoplasmic reticulum by annexin VI (67-kDa calcimedin). J. Biol.
Chem. 26515894-15899.
Elderfield, A.J., J. Newcombe, C. Bolton, and R.J. Flower (1992) Lipocortins
(annexins) 1, 2, 4, and 5 are increased in the central nervous system in
multiple sclerosis. J. Neuroimmunol. 39.91-100.
Elderfield, A.J., C. Bolton, and R.J. Flower (1993) Lipocortin 1 (annexin I)
immunoreactivity in the cervical spinal cord of Lewis rats with acute
experimental allergic encephalomyelitis.J. Neurol. Sci. 119.146-153.
Emans, N., J.P. Gorvel, C. Walter, V. Gerke, R. Kellner, G. Griffiths, and J.
Gruenberg [ 1993) Annexin I1 is a major component of fusogenic endosoma1 vesicles. J. Cell. Biol. 120,1357-1369.
Eng, L.F., and R.S. Ghirnikar (1994) GFAP and astrogliosis. Brain Pathol.
4:229-237.
Fava, R.A., J. McKanna, and S. Cohen (1989) Lipocortin I ( ~ 3 5is) abundant
in a restricted number of differentiated cell types in adult organs. J. Cell.
Physiol. 141984-293.
Fishman, R.A. (1980) Cerebrospinal Fluid in Diseases of the Nervous
System. Philadelphia: W.B. Saunders.
Flower, R.J. (1988) Lipocortin and the mechanism of action of the glucocorticoids. Br. J. Pharmacol. 102.987-1015.
369
Geisow, M.J., U. Fritsche, J.M. Hexham, B. Dash, and T. Johnson (1986) A
consensus amino-acid sequence repeat in Torpedo and mammalian
Ca”-dependent-membrane binding proteins. Nature 320:636-638.
Gerke, V. (1992) Evolutionary conservation and three-dimensional folding of
the tyrosine kinase substrate annexin 11. In S.E. Moss (ed): The
Annexins. Chapel Hill, NC: Portland Press, pp. 47-60.
Giambanco, I., G. Sorci, S. Antonioli, M.G. Rambotti, A. Spreca, V. Bocchini,
and R. Donato (1993) Immunocytochemical analysis of annexin V
(CaBP331, in a human-derived glioma cell line. Expression of annexin V
depends on cellular growth state. FEBS Lett. 328:45-50.
Glenney, J.R. Jr., and B.F. Tack (1985) Aminoterminal sequences of p36 and
associated p10: Identification of the site of tyrosine phosphorylation and
homology with S100. Proc. Natl. Acad. Sci. USA 823884-7888.
Gould, K.L., J.R. Woodgett, C.M. Isacke, and T. Hunter (1986) The protein
tyrosine kinase substrate, p36, is also substrate for protein kinase C in
vivo and in vitro. Mol. Cell. Biol. 6:2738-2744.
Greenberg, M.E., R. Brackenbury, and G.M. Edelman (1984) Changes in the
distribution of the 34-kdalton tyrosine kinase substrate during differentiation and maturation of chicken tissues. J. Cell. Biol. 98:9473:486.
Haigler, H.T., D.D. Schlaepfer, and W.H. Burgess (1987) Characterization of
lipocortin I and an immunologically unrelated 33-KDa protein as
epidermal growth factor receptorikinase substrates and phospholipase
A2 inhibitors. J. Biol. Chem. 262.6921-6930.
Hamre, K.M., K.P. Chepenik, and D. Goldowitz (1995) The annexins:
Specific markers of midline structures and sensory neurons in the
developing murine central nervous system. J. Comp. Neurol. 352.421435.
Harder, T., and V. Gerke (1993) The subcellular distribution of early
endosomes is affected by annexin 112~112
complex. J. CeU. Biol. 123:111%
1132.
Hazarika, P., M.A. Kaetzel, A. Sheldon, N.J. Karin, S. Fleischer, T.E.
Nelson, and J.R. Dedman (1991) Annexin VI is associated with calciumsequestering organelles. J. Cell. Biochem. 46:78-85.
Heizmann, C.W., and K. Braun (1992) Changes in Ca2+-bindingproteins in
human neurodegenerative disorders. Trends Neurosci. 1.5.259-264.
Huber, R., R. Berendes, A. Burger, H. Luecke, and A. Karshikov (1992a)
Annexin V Crystal structure and its implications on function. In S.E.
Moss (ed): The Annexins. Portland Press, pp. 105-124.
Huber, R., R. Berendes, A. Burger, M. Schneider, L. Karshikov, H. Luecke, J.
Romisch, and E. Paques (1992b) Crystal and molecular structure of
human annexin V after refinement. Implications for structure, membrane binding and ion channel formation of the annexin family of
proteins. J. Molec. Biol. 223.683-704.
Jindal, H.K., W.G. Chaney, C.W.Anderson, R.G. Davis, and J.K. Vishwanatha
(1991) The protein-tyrosine kinase substrate, calpactin I heavy chain
( ~ 3 6 1is
, part of the primer recognition protein complex that interacts
with DNA polymerase 01. J. Biol. Chem. 266.5169-5176.
Johnson, M.D., J.M. Kamso-Prait, W.O. Whetsell, and R.B. Pepinsky
(1989a) Lipocortin-l immunoreactivity in the normal central nervous
system and lesions with astrocytes. Am. J. Clin. Pathol. 92.424429.
Johnson, M.D., J.M. Kamso-Pratt, R.B. Pepinsky, and W.O. Whetsell
(1989b) Lipocortin-1 immunoreactivity in central and peripheral nervous system glial tumors. Hum. Pathol. 20,772-776.
Johnson, N., G. Marriott, and K. Weber (1988) P36, the major cytoplasmic
substrate of src tyrosine kinase, binds to its p l l regulatory subunit via a
short amino-terminal amphipathic helix. EMBO J. 7.2435-2442.
Kaetzel, M.A., P . Hazarika, and J.R. Dedman (1989) Differential tissue
expression of three 35-kDa annexin calcium-dependent phospholipidbinding proteins. J. Biol. Chem. 264.14463-14470.
Kaetzel, M.A., H.C. Chan, W.P. Dubinsky, R.D. Dedman, and D.J. Nelson
(1994) A role for annexin IV in epithelial cell function. Inhibition of
calcium-activated chloride conductance. J. Biol. Chem. 269.5297-5302.
Koster, J.J.,C.M. Boustead, A.M. Middleton, and J.H. Walker (1993) The
sub-cellular localization of annexin V in cultured chick-embryo fibroblast. Biochem. J. 291:595-600.
Laemmli, U.K. (1970) Cleavage of structural proteins during the assembly of
the head of bacteriophage T4. Nature 227:680-685.
Lawson, S.N. (1992) Morphological and biochemical cell types of sensory
neurons. In S.A. Scott (ed):Sensory Neurons: Diversity, Development,
and Plasticity. Oxford: Oxford University Press, pp. 27-59.
Lin, H.C., T.C. Sudhof, and G.W. Anderson (1992) Annexin VI is required for
budding of clathrin-coated pits. Cell 70283-291.
Martin, J.B. (1993) Molecular genetics of neurological diseases. Science
262674-676.
370
McKanna, J.A., and S. Cohen (1989) The EGF receptor kinase substrate p35
in the floor plate of embryonic rat CNS. Science 243:1477-1479.
Mizutani, A,, N. Usuda, H. Tokomitsu, H. Minami, K. Yasui, R. Kobayashi,
and H. Hidaka (1992) CAP-50, a newly identified annexin, localizes in
nuclei of cultured fibroblast 3Y1 cells. J. Biol. Chem. 267:13498-13504.
Moss, S.E. (1992) The annexins. In S.E. Moss (ed): The Annexins. Chapel
Hill, NC: Portland Press, pp. 1-9.
Mullens, L., D.R. Marriott, K.A. Young, L. Tannahill, S.L. Lightman, and
G.P. Wilkin (1994) Up-regulation of lipocortin-1 and its mRNA in
reactive astrocytes in kainate-lesioned rat cerebellum. J. Neuroimmunol.
50:25-33.
Owens, R.J., C.J. Gallagher, and M.J. Crumpton (1984) Cellular distribution
of p68, a new calcium-binding protein from lymphocytes. EMBO J.
3.945-952.
Pannese, E. (1974) The histogenesis of the spinal ganglia. In Brodal A., W.
Hild, J. Van Limborgh, R. Ortmann, T.H. Schiebler, G. Tondury, and E.
Wolf (eds): Advances in Anatomy, Embryology and Cell Biology. Berlin:
Springer. 47.3-97.
Perl, E.R. (1992) Function of dorsal root ganglion neurons: An overview. In
S.A. Scott (ed): Sensory Neurons: Diversity, Development, and Plasticity. Oxford: Oxford University Press, pp. 3-23.
Philippe, E., and B. Droz (1988) Calbindin D-28k-immunoreactive neurons
in chick dorsal root ganglion: Ontogenesis and cytological characteristics
of the immunoreactive sensory neurons. Neuroscience 26.915-224.
Pollard, H.B., H.R. Guy, N. Arispe, M. de IaFuente, G. Lee, E.M. Rojas, J.R.
Pollard, M. Sriwastava, Z.Y. Zhang-Keck, N. Merezhinskaya, H. Coahuy,
A.L. Burns, and E. Rojas (1992) Calcium channel and membrane fusion
activity of synexin and other members of the annexin gene family.
Biophys. J. 6215-18.
Powell, M.A., and J.R. Glenney (1987) Regulation of calpactin I phospholipid
binding by calpactin I light-chain binding and phosphorylation by
p60'-sr~.Biochem. J. 427:321-328.
Rambourg, A,, Y. Clermont, and A. Beaudet (1983) Ultrastructural features
of six types of neurons in rat dorsal root ganglia. J. Neurocytol. 12:47-66.
Raynal, P., and H.B. Pollard (1994) Annexins: The problem of assessing the
biological role for a gene family of multifunctional calcium- and phospholipid-binding proteins. Biochem. Biophys. Acta 1197:63-93.
Raynal, P., P.M. VanBergen en Henegouwen, F. Hullin, J.M. RagahThomas, J. Fauvel, and H. Chap (1992) Morphological and biochemical
evidence for partial nuclear localization of annexin I in endothelial cells.
Biochem. Biophys. Res. Commun. 186:432-439.
Relton, J.K., P.J.L.M. Strijbos, C.T. O'Shaughnessy, F. Carey, R.A. Forder,
F.J.H. Tilders, and N.J. Rothwell (1991) Lipocortin-1 is an endogenous
inhibitor of ischemic damage in the rat brain. J. Exp. Med. 174:305-310.
Rojas, E., H.B. Pollard, H.T. Haigler, C. Parra, and A.1,. Burns (1990)
J.M. NACIFF ET AL.
Calcium-activated endonexin I1 forms calcium channels across acidic
phospholipid bilayer membranes. J. Biol. Chem. 265.21207-21215.
Ross, S., J.F. Tait, and P.W. Majerus (1990) Identity of inositol 1,Z-cyclic
phosphatase 2-phosphohydrolase with lipocortin 111. Science 248:605607.
Sajin, B., and D.A. Steindler (1994) Cells on the edge: Boundary astrocytes
and neurons. Persp. Dev. Biol. 2275-289.
Sarafian, T., L.A. Pradel, J.P. Henry, D. Aunis, and M.F. Bader (1991) The
participation of annexin I1 (calpactin I) in calcium evoked exocytosis
requires protein kinase C. J. Cell Biol. 114:1135-1147.
Sekar, M.C., V. Sambandam, J.M. McDonald, W.E. Grizzle, R.C. Stockard,
and R. Berry (1994) Dissociation of cyclic inositol phosphohydrolase
activity from annexin 111binding. Mol. Cell. Biol. 5r144a.
Semich, R., V. Gerke, H. Robenek., and K. Weber (1989) The p36 substrate
of pp60*rc kinase is located a t the cytoplasmic surface of the plasma
membrane of fibroblasts; an immunoelectron microscopic analysis. Eur.
J. Cell Biol. 50:313-323.
Smythe, E., P.D. Smith, S.M. Jacob, J. Theohald, and S.E. Moss (1994)
Endocytosis occurs independently of annexin VI in human A431 cells. J.
Cell Biol. 124:301-306.
Spreca, A,, M.G. Rambotti, I. Giambanco, G. Pula, R. Bianchi, P. Ceccarelli,
and R. Donato (1992) Immunocytochemical localization of annexin V
(CaBP33), a Ca2+-dependentphospholipid- and membrane-binding protein, in the rat nervous system and skeletal muscle and in the porcine
heart. J. Cell. Physiol. 152:587-598.
Sun, J., H.H. Salem, and P. Bird (1992) Nucleolar and cytoplasmic localization of annexin V. FEBS Lett. 314:425429.
Tait, J.F., D. Gibson, and K. Fujikawa (1989) Phospholipid binding properties of human placental anticoagulant protein-I, a member of the
lipocortin family. J. Biol. Chem. 264:7944-7948.
Thiel, C., M. Osborn, and V. Gerke (1992) The tight association of the
tyrosine kinase substrate annexin I1 with the submembranous cytoskeleton depends on intact p l l and Ca2+binding sites. J. Cell Sci. 103:733742.
Towbin, H.T., T. Staehlin, and J. Gordon (1979) Electrophoretic transfer of
proteins from polyacrylamide gels to nitrocellulose sheets, procedure and
some applications. Proc. Nat. Acad. Sci. USA 76:4350-4354.
Watanabe, T., M. Inui, B-Y.Chen, M. Iga, and K. Sobue (1994) Annexin
VI-binding proteins in brain. Interaction of annexin VI with a membrane skeletal protein, calspectin (brain spectrin or fodrin). J. Biol.
Chem. 26: 17656-17662.
Weber, K. (1992) Annexin I1 interaction with p l l . In S.E. Moss (ed): The
Annexins. Portland Press, pp. 61-68.
Woolgar, J.A., O.M. Boustead, and J.H. Walker (1990) Characterization of
annexins in mammalian brain. J. Neurochem. 54.62-71.
Документ
Категория
Без категории
Просмотров
7
Размер файла
5 853 Кб
Теги
expressions, spina, dorsal, cord, differential, annexins, rat, ganglia, roots
1/--страниц
Пожаловаться на содержимое документа