|
|
|
|||||||
|
|||||||||
J Biol Chem, Vol. 275, Issue 1, 451-460, January 7, 2000
From the Department of Physiology and Biophysics, University of
Washington, Seattle, Washington 98195-7290 and Departments
of Interactions between growing axons and synaptic
basal lamina components direct the formation of neuromuscular junctions
during nerve regeneration. Isoforms of laminin containing SV21 is a transmembrane
keratan sulfate proteoglycan of synaptic vesicles (1, 2). It exists in
two forms, heavy (H) and light (L), that differ in glycosylation (2).
This proteoglycan has been purified and characterized from the electric
organ synapse (1-4) which is structurally and molecularly similar to
the vertebrate neuromuscular junction (5, 6). In electric organ the H
form of SV2 is identified by two monoclonal antibodies (Tor 70 and anti-SV1) that bind to the same unique keratan sulfate epitope (SV1) in
the lumen of synaptic vesicles (2, 4, 7). Both the H and L forms are
identified by a monoclonal antibody, anti-SV2, that binds a protein
epitope on the cytosolic side of synaptic vesicles (1, 2). On SDS-PAGE
the L form migrates like a protein of about 90 kDa, whereas the H form
migrates with the characteristic heterogeneous mobility of a
proteoglycan, between 90 and above 200 kDa (2). Both forms of SV2
appear to be present in vertebrate synapses, including those of
electric fish (2), rat (2), and human (8). SV2 is also found in
endocrine cells (1).
The cDNA that encodes the SV2 protein has been cloned from rat
(9-11), cow (11), and electric fish (12). Three different genes coding
for SV2 have been found in rat (13, 14). The inferred amino acid
sequence suggests that the protein has 12 transmembrane domains and a
distant sequence homology to both bacterial and eukaryotic sugar
transporters (9-11). As yet, however, no transport activity or any
other function has been found for SV2.
SV2 contains a large loop between putative transmembrane domains 7 and
8 with three highly conserved N-linked glycosylation sites
(12, 13). These are the likely locations for the attachment of the
keratan sulfate side chains recognized by anti-SV1 mAb (2). Previously
it was shown that the SV1 epitope of SV2 was not only present in
synaptic vesicles (4) but on the nerve terminal surface of electric
organ synapses as well (15). Based on this finding, Buckley et
al. (15) hypothesized that this synaptic vesicle proteoglycan was
retained on the surface of synaptosomes due to interactions with
components of the synaptic cleft. In fact, it is known that
interactions between the nerve terminal membrane and the extracellular
matrix (ECM) of the synaptic cleft are critical for nerve regeneration
at the neuromuscular junction. ECM in the synaptic cleft directs the
rebuilding of this synapse after nerve injury (16). The axon distal to
the injury degenerates and a new neurite regrows from the cut end to
form a new nerve terminal at the original synaptic site. Not only does
the regenerating neurite appear to use ECM cues to find the vacated
original synaptic site, the sites of neurotransmitter release within
the nerve terminal form at their original locations, directed by ECM
cues (17, 18). Presumably, ECM receptors of the nerve terminal
recognize synapse-specific ECM components.
Although the nerve terminal ECM receptors involved in nerve
regeneration are unknown, the ECM protein laminin is a good candidate for an ECM cue (19). Laminins are large heterotrimeric glycoproteins with molecular masses of 700-900 kDa, each containing an Here we present evidence that the SV2H on the surface of electric organ
nerve terminal may act as a laminin receptor in addition to its
function in the synaptic vesicle. On the surface of isolated synaptosomes SV2H is associated with a 900-kDa laminin that is composed
of an Materials--
The elasmobranch electric rays Narcine
brasiliensis and Torpedo californica were obtained from
Gulf Specimens (Panacea, FL) and Marinus Inc. (Long Beach, CA),
respectively. Immunobeads (goat anti-mouse immunoglobulins, heavy and
light chain-specific) were purchased from Irvine Scientific (Santa Ana,
CA). Sulfo-NHS-Biotin used for protein labeling was purchased from
Pierce. Ultrapure mouse laminin-1 (Engelbreth-Holm-Swarm sarcoma
laminin, entactin (nidogen-free) was purchased from Collaborative
Biomedical Products (catalog number 4023) of Becton Dickinson Labware
(Bedford, MA). Anti-mouse laminin-1 was prepared by injection of
ultrapure mouse laminin-1 into rabbits. The antibody reacts with all
three chains of purified mouse laminin-1 on Western blots. Rabbit
antibodies specific for electric fish synaptotagmin or synaptophysin
were prepared by immunization of rabbits with the peptide
CMKTRETHPQAFVAPMAT or CGGYSQPVPTSFTNQM, respectively, and then
conjugated to keyhole limpet hemocyanin following the methods of
Hockfield et al. (29). Except for the N-terminal cysteine
residues, these sequences are present in the N terminus of
synaptotagmin or the C terminus of synaptophysin from electric organ
synaptic vesicles (30, 31). The anti-synaptotagmin antibody was used as
an IgG fraction that was purified with protein A, as described by
Hockfield et al. (29). The anti-synaptophysin antibody was
used as an antiserum. Monoclonal antibodies, anti-SV1 and anti-SV2, are
of the IgG1 isotype and were produced as hybridoma supernatants
following the procedures of Hockfield et al. (29). MOPC-21
(an IgG1) and triacetylchitotriose were obtained from Sigma.
Anti-laminin
The affinity co-electrophoresis (ACE) gel apparatus used in these
experiments was designed after that made by Lim et al. (33). SeaPlaque low gelling temperature agarose was purchased from FMC BioProducts (Rockland, ME).
Electric Organ Synaptosomes and Immunoprecipitations of
SV2--
Synaptosomes were prepared from fresh electric organ tissue
(either N. brasiliensis or T. californica) as
described by Miljanich et al. (34) and Yeager et
al. (35) with differential centrifugation and sedimentation
equilibrium. We refer to this preparation as partially purified
synaptosomes because it contains postsynaptic membranes as well as synaptosomes.
In some cases we further purified the synaptosomes by immuno-isolation
using the anti-SV1 mAb and the methods of Buckley et al.
(15). This procedure reduces the relative abundance of acetylcholine receptor-containing membranes (postsynaptic membranes) more than 10-fold (34), so that acetylcholine receptor cannot be detected by
Western blot.2 Partially
purified synaptosomes (300 µg of protein) were washed 3 times in 400 mM NaCl, 1% BSA, 20 mM HEPES, pH 7.0 (S-buffer), by centrifugation at 10,000 × g for 10 min
followed by resuspension to a volume of 1 ml. This solution is
iso-osmotic for fish synaptosomes. Aliquots of anti-SV1 monoclonal
antibody (mAb) in a hybridoma supernatant were centrifuged 20 min at
10,000 × g at 4 °C to remove any precipitated
material and then added to the synaptosome suspension, 1 ml of
supernatant per 150 µg of synaptosomal protein. This
antibody-synaptosome suspension was incubated with gentle mixing
overnight at 4 °C. Membranes were washed 3 times in S-buffer,
centrifuged 10 min at 10,000 × g at 4 °C, and
resuspended in 1 ml of S-buffer. This mixture was then layered on 2 ml
of 12% Ficoll in S-buffer and centrifuged 5 min at 2,000 × g. The top 2 ml of this solution was withdrawn and mixed
with polyacrylamide beads to which goat anti-mouse Ig antibodies were
covalently attached (immunobeads). These immunobeads were washed in
S-buffer and resuspended to form a 10% slurry. The bead slurry was
added to the antibody-labeled membranes, 100 µl of 10% slurry, 150 µg of synaptosomal protein, and the solution was gently mixed
overnight at 4 °C. After incubation the bead solution was layered on
2 ml of 12% Ficoll and pelleted by centrifugation at 2,000 × g for 5 min. The pellet was washed 2 times in S-buffer
without BSA and prepared for SDS-PAGE (36). Control precipitations were
conducted by substituting a commercially obtained IgG1 mAb (MOPC-21)
for anti-SV1 monoclonal antibody in the initial incubation.
To immunoprecipitate Triton X-100-solubilized SV2 from the synaptosomal
surface with anti-SV1 mAb, partially purified synaptosomes were
incubated with anti-SV1 or MOPC-21 and washed as described for the
immunoprecipitation of synaptosomes. The washed synaptosomes were
solubilized in 1 ml of S-buffer containing 1.5% Triton X-100. The
solution was layered on 12% Ficoll also containing 1.5% Triton X-100
and the immunoprecipitation proceeded as described above for intact
synaptosomes with goat anti-mouse Ig immunobeads. All solutions
contained S-buffer with 1.5% Triton X-100. In addition to
immunoprecipitating the Triton X-100-solubilized SV2·900-kDa complex
with anti-SV1, we used the anti-SV2 mAb as well. The procedure was the
same as used for anti-SV1, except for the use of alkaline-stripped synaptosomes (see below).
Preparation and Analysis of Alkaline-stripped
Synaptosomes--
To test the strength of the interactions between
proteins in partially purified synaptosome preparation, we
alkaline-stripped these synaptosomes by exposure to solutions of pH
11.5. Such conditions can strip peripheral membrane proteins from their
parent membranes (37). Partially purified synaptosomes (1.5 mg) were
resuspended in 1 ml of SW-buffer (S-buffer without 1% BSA), spun down
by centrifugation at 10,000 × g for 10 min at 4 °C,
and the supernatant discarded. This was repeated once more with
SW-buffer and then once with 400 mM NaCl. The synaptosomes
were resuspended in 800 µl of 400 mM NaCl, divided into 2 aliquots of equal volume, and pelleted by centrifugation at 10,000 × g, 10 min, 4 °C. One pellet was resuspended in 100 µl of 100 mM CAPS, 300 mM NaCl, pH 11.5, and the other with 100 µl of 100 mM HEPES, 300 mM
NaCl, pH 7.5. Each aliquot was incubated at 4 °C for 30 min with
gentle agitation. The synaptosomes were pelleted as described above,
the supernatant discarded, and the pellet resuspended in 400 µl 100 mM HEPES, 300 mM NaCl, pH 7.5. The synaptosomes
were incubated at 4 °C for 30 min with gentle agitation and then
pelleted by centrifugation. These pelleted synaptosomes were washed
twice with SW-buffer by resuspension and centrifugation. The resulting
pelleted synaptosomes were solubilized with SDS-containing final sample
buffer for SDS-PAGE.
To label the proteins on the surface of alkaline-stripped synaptosomes,
we used biotinylation. All procedures were done at 4 °C. Partially
purified synaptosomes (1.3 mg of synaptosomal protein) were
alkaline-stripped (see above). These membranes were washed three times
by resuspension in 1 ml of biotinylation buffer (400 mM
NaCl, 30 mM HEPES, pH 8.5) and centrifugation at
10,000 × g for 10 min. The alkaline-stripped
synaptosomes were resuspended in 1 ml of biotinylation buffer, and 190 µl of Sulfo-NHS-Biotin (50 mg/ml in Me2SO) was added. The
reaction was allowed to proceed for 2 h with gentle agitation to
ensure that the membranes remained in suspension. The synaptosomes were
washed three times with 1 ml of 400 mM NaCl, 20 mM HEPES, pH 7.0.
Synaptic Vesicles and Immunoprecipitations of SV2--
Synaptic
vesicles were purified from electric organ by the methods of Carlson
et al. (38). To immunoprecipitate SV2 from synaptic vesicles
with anti-SV2 or SV1 mAbs the procedures of Scranton et al.
(2) were used. Briefly, 25 µg of purified electric organ synaptic
vesicles were solubilized in 250 µl of 0.28 M NaCl, 20 mM HEPES, pH 7.4, 0.1% BSA, and 1.5% Triton X-100. A 10%
slurry of goat anti-mouse Ig immunobeads (75 µl) to which the
anti-SV1, anti-SV2, or a control mAb (MOPC-21) had been previously
bound (as described in Ref. 39) was added to the solubilized vesicles. The mixture was incubated overnight at 4 °C with gentle agitation. The immunobeads were then separated from the vesicle solution by
centrifugation at 10,000 × g for 10 min at 4 °C,
and the immunobeads were washed as described above for synaptosomes.
Purification and Iodination of SV2--
SV2 was immunopurified
on Affi-Gel matrix to which the anti-SV2 mAb had been covalently
attached. The anti-SV2 mAb was purified with protein A from ascites
fluid following the methods of Hockfield et al. (29).
Affi-Gel-10 was washed with 10 ml of a MOPS buffer (100 mM
NaCl, 50 mM MOPS, pH 6.8). Purified anti-SV2 (25 mg in 2 ml
of MOPS buffer) was mixed with 1.4 ml of Affi-Gel, and the resulting
suspension was gently mixed overnight at 4 °C. The Affi-Gel was then
pelleted at 6,000 rpm, and the supernatant was withdrawn to assay
remaining protein. The Affi-Gel was then washed three times in MOPS
buffer, once in basic CAPS buffer (100 mM CAPS, pH 11.5),
and three additional times in MOPS buffer. The coupling efficiency of
this reaction was determined to be 98.3%. One ml of this Affi-Gel was
placed in a 10-ml column and equilibrated with column buffer (0.2 M NaCl, 10 mM MOPS, 1% Triton X-100, pH 7.5).
To purify SV2, we used the following procedures: synaptic vesicles were
isolated from 300 g of electric organ with the methods of Carlson
et al. (38), but the CPG-3000 column step was omitted. These
synaptic vesicles were diluted 2 times with 400 mM NaCl, 10 mM HEPES, 10 mM EGTA, pH 7.0, and pelleted by
centrifugation at 158,000 × g for 10 h at 4 °C
in a Beckman 45Ti rotor. The pellet was solubilized with 3 ml of 0.2 M NaCl, 10 mM MOPS, 2% Triton X-100, pH 7.5, and the solubilized vesicle proteins centrifuged at 4 °C for 15 min
in an Eppendorf centrifuge. The supernatant was collected and applied
to the Affi-Gel anti-SV2 mAb column. The column was washed with 20 ml
of column buffer and then eluted with 10 ml of 100 mM CAPS,
pH 11.5, where 1-ml fractions were collected. Each fraction was
neutralized with 130 ml of 1 M HEPES, pH 7.0, and assayed
for SV2 antigenicity by dot blots (4). Recovery of SV2 antigen eluted
form the column was usually about 40%.
We labeled purified SV2 with 125I-diazotized iodosulfanilic
acid as described by Scranton et al. (2). The labeled H form
of SV2 was isolated by affinity chromatography on a tomato
(Lycopersicon esculentum) lectin-agarose column. About 1 µg of 125I-labeled SV2 was applied to a 0.5-ml column of
tomato lectin-agarose equilibrated with 150 mM NaCl, 0.1 mM Ca2+ acetate, 0.5% CHAPS, 10 mM
HEPES, pH 7.0. (This buffer solution was used throughout the
chromatography.) The L form of SV2 passed through the column without
binding. After extensive washing of the column, the bound
125I-SV2H was eluted with 2 ml of 0.5 mg/ml triacetylchitotriose.
To isolate the L form of SV2 free of unbound 125I, we used
the sedimentation velocity centrifugation on 5-20% sucrose gradients. The 125I-SV2L, which passed through the tomato lectin
column without binding, was concentrated with a Centricon-30 (Amicon,
Danvers, MA) and layered on a 5-25% sucrose gradient containing 400 mM NaCl, 1.5% Triton X-100, 20 mM HEPES, pH
7.0. The gradients were centrifuged for 6 h at 300,000 × g, 4 °C in an SW-55 rotor, and fractions were collected.
Aliquots of each fraction were subjected to SDS-PAGE and
autoradiography to locate the 125I-SV2L in the gradient.
For immunoprecipitations of 125I-SV2 with anti-SV1 or
anti-SV2 mAbs, we used the methods of Scranton et al. (2),
which do not involve denaturing conditions. Here, the mAbs (anti-SV1,
anti-SV2, or MOPC-21) were bound to immunobeads before the immunobeads
were added to the antigen. Forty microliters of a 10% immunobead
slurry were used in a total reaction volume of 340 µl.
ACE--
In order to determine the strength of the interaction
between SV2H and laminin, ACE was performed according to the methods of
Lee and Lander (40). Briefly, a 75 × 100 plexiglass tray was
fitted with GelBond, and the sides were taped. A lane-forming comb and
slot-forming comb were added, and then a 1% agarose gel was poured
(1% low-gelling temperature agarose in ACE gel buffer, 1% CHAPS, 50 mM MOPS, 1 mM EDTA, 125 mM sodium
acetate, pH 7.2). The gel was allowed to solidify, and the slot-forming
comb was removed. Serial dilutions of laminin-1 (1:2 dilutions,
starting at about 360 nM laminin) were made in 350 µl of
ACE gel buffer. Each dilution was mixed with a 2% agarose solution in
ACE gel buffer (heated to 42 °C) and immediately placed in the
slots. The agarose in the slots was allowed to solidify, and the entire ACE gel was placed in a water-cooled horizontal gel apparatus. ACE
running buffer (50 mM MOPS, 125 mM sodium
acetate, pH 7.2) was added until the gel was well submerged. An aliquot
of purified 125I-SV2H was mixed with 10× ACE gel buffer
and H2O to give a final concentration of 1× ACE gel
buffer. In addition, glycerol was added to a final concentration of
10%, and 3 µl each of 0.1% bromphenol blue and phenol red were
added. The final sample volume was 120 µl. The sample comb was
removed and the diluted 125I-SV2H added to the gel. ACE was
performed at 300 mA for 4 h, until the phenol red reached the end
of the dilution lanes. The gel was then removed from the apparatus,
dried, and exposed to x-ray film for autoradiography.
In order to determine the exact concentration of laminin-1 for ACE, an
aliquot of this protein was hydrolyzed with constant boiling HCl, and
the hydrolysate was subjected to amino acid analysis (AAA Laboratory,
Mercer Island, WA). Visualization by silver stain of laminin-1 on an
SDS-polyacrylamide gel indicated that the protein was 98.8% pure and
entactin (nidogen-free) as stated by the manufacturer (Collaborative
Biomedical Products of Becton Dickinson Labware, Bedford, MA). The
molecular mass of the protein was taken to be 900 kDa (41).
SDS-PAGE and Nitrocellulose Blotting Procedures--
SDS-PAGE
was performed according to Laemmli (36). Antigen was eluted from
immunobeads for SDS-PAGE by boiling the beads in the final sample
buffer of Laemmli (36). All SDS-PAGE that was performed under reducing
conditions utilized 2.7% acrylamide stacking gels with 2.7-14%
acrylamide gradient resolving gels. For SDS-PAGE performed under
non-reducing conditions, we used 2.7% stacking gels with 3-9%,
3-7%, or 3-5% resolving gels. For molecular weight standards we
used Rainbow TM protein molecular weight markers from Amersham
Pharmacia Biotech and mouse laminin-1 purified from EHS tumors (Becton
Dickinson Labware, Bedford, MA). For laminin-1 the
Nitrocellulose blotting procedures were performed as described
previously (2) except that the transfer buffer contained 0.1% SDS, and
the transfer time was 3 h. Biotinylated proteins bound to
nitrocellulose were detected by streptavidin with the SABC kit
following the manufacturer's instructions. Horseradish peroxidase-conjugated antibody or horseradish peroxidase-streptavidin bound to blots were visualized with x-ray film and a luminol detection system (ECL from Amersham Pharmacia Biotech). Sometimes we reprobed blots a second time. By using an antibody, we reprobed a blot that had
been originally probed for biotinylated proteins. Similarly, we
reprobed with a rabbit polyclonal antibody a blot that had originally
been exposed to a mouse monoclonal antibody. To visualize the second
probe, we inactivated the horseradish peroxidase bound to the blot by
exposure overnight at room temperature to an aqueous solution of 0.02%
sodium azide. We never found the horseradish peroxidase from the first
probe to be visualized on x-ray film in the second luminol reaction.
An SV2H·Laminin Complex on the Surface of Electric Organ
Synaptosomes--
Buckley et al. (15) demonstrated that
SV2H is present on the surface of electric organ nerve terminals both
by immunoelectron microscopy and the immunoprecipitation of
synaptosomes (reviewed in Ref. 43). These workers hypothesized that
this proteoglycan is complexed on the surface of the nerve terminal
with synaptic cleft components (15). Here, we tested this hypothesis by
characterizing detergent-solubilized SV2 immunoprecipitated from the
surface of synaptosomes. For this purpose we used the anti-SV1 mAb,
because the SV1 epitope is present on the extracellular domain of SV2 (2). Anti-SV1 mAbs were first bound to the surface of synaptosomes; the
excess mAbs were washed away, and then the synaptosomes were solubilized with detergent, Triton X-100. The detergent-solubilized SV2·mAbs complexes were immunoprecipitated with polyacrylamide beads
coated with goat anti-mouse IgG antibodies (immunobeads). As expected,
the immunoprecipitated material was recognized by both anti-SV2 and
anti-SV1 mAbs (Fig. 1A, lanes
1 and 3, respectively). A control IgG1 mAb
immunoprecipitated no SV2 from the synaptosomes (Fig. 1A, lanes
2 and 4). SV1-reactive SV2 migrated as a broad band
characteristic of the H form (2) and typical of some proteoglycans (e.g. Ref. 44) due to the heterogeneity of the
glycosaminoglycan side chains. Since the anti-SV1 mAbs were used to
immunoprecipitate SV2, only the H form of SV2 is immunoprecipitated and
not the L form.
We then asked whether SV2H immunoprecipitated from the synaptosomal
surface was associated with laminin, a major component of synaptic
basal lamina. Initially, we used an antiserum to laminin-1 that
recognizes the
In addition to the anti-SV1 mAb, the solubilized SV2·900-kDa laminin
complex can be immunoprecipitated with the anti-SV2 mAb that binds a
cytosolic domain on SV2. To determine this, we first lysed synaptosomes
with a pH 11.5 buffer, which does not dissociate the 900-kDa·laminin
complex (see below). This treatment generates synaptosomes in which the
SV2 epitope is exposed. These synaptosomes were incubated with the
anti-SV2 mAb, solubilized by Triton X-100, and immunoprecipitated with
immunobeads as described for anti-SV1. Western blots show that the
900-kDa laminin was specifically immunoprecipitated by the anti-SV2 mAb
(Fig. 1C, lane 1).
SV2H in Synaptic Vesicles Is Not Associated with
Laminin--
Purified electric organ synaptic vesicles were
solubilized with detergent, and the solubilized SV2 was
immunoprecipitated with anti-SV1 (Fig. 1B, lane 1), anti-SV2
(Fig. 1B, lane 2), or a control mAb (Fig. 1B, lane
3). The resulting immunoprecipitates were subjected to SDS-PAGE
under reducing conditions, and Western blots were probed with either
anti-SV2 (Fig. 1B, lanes 1-3, upper) or
anti-laminin-1 antibodies (Fig. 1B, lanes 1-3, lower).
Unlike SV2 immunoprecipitated from the surface of synaptosomes, these Western blots probed with anti-laminin 1 reveal no 230-kDa laminin chain (Fig. 1C, lanes 1 and 2, lower). Moreover,
lysate of purified synaptic vesicles contained no detectable laminin
(data not shown). Thus, although laminin is associated with the SV2H on
the nerve terminal surface, it is not present and not associated with
this proteoglycan in electric organ synaptic vesicles.
Other Presynaptic Proteins Are Excluded from the SV2·900-kDa
Laminin Complex--
To ask whether the SV2·900-kDa laminin complex
is a general aggregate of presynaptic proteins, we tested the
immunoprecipitated complex for the presence of the
Since SV2H is associated with a laminin on the surface of synaptosomes,
we wished to determine how many laminins were present. By using the
anti-SV1 mAb, we immunoprecipitated intact electric organ synaptosomes
and subjected these membranes to Western blots probed with
anti-laminin-1 antibodies. With non-reducing SDS-PAGE two laminins are
detected with mobilities of 900 kDa (Fig. 2A, upper,
lane 1, arrow) and 740 kDa (Fig. 2A, upper, lane
1 arrowhead), respectively. This immunoprecipitation is specific:
mock immunoprecipitation of synaptosomes with control mAb yields no 900 kDa and only a very small amount (<5% of that in lane 1)
of 740 kDa (Fig. 2, upper, lane 2). Thus, not
only do synaptosomes contain the 900-kDa laminin but also a 740-kDa
laminin. However, as shown above (Fig. 1A, lane 5, upper)
and reconfirmed here (Fig. 2, upper, lane 3), only the
900-kDa laminin co-immunoprecipitates with SV2H solubilized from the
synaptosomal surface.
The synaptic basal laminin protein agrin does not co-immunoprecipitate
with the SV2H·laminin complex although we can detect agrin in
immunoprecipitated synaptosomes (data not shown). Thus, even within the
basal lamina, SV2 interacts selectively with a subset of components.
Subunit Composition of SV2H-associated Laminin--
The size of
the co-immunoprecipitating 900-kDa laminin suggested that its The SV2·Laminin Complex Is Resistant to Dissociation by pH
11.5--
Many protein-protein interactions are broken by high pH. For
example, solutions of pH 11.5 are commonly employed to elute antigens
from antibody affinity columns (29). This high pH can also strip
peripheral membrane proteins from their membranes, such as rapsyn from
acetylcholine receptor-rich membranes (37). However, interactions
between matrix proteins can require stronger conditions for
solubilization (e.g. Ref. 46). To determine whether the
interaction between SV2H and laminin was of this latter category, we
exposed partially purified synaptosomes to solutions of either pH 11.5 or pH 7.4 at 4 °C for 30 min. The membranes were pelleted by
centrifugation at 10,000 × g for 10 min. These
conditions should sediment synaptosomes and presynaptic plasma
membranes but not membranes the size of synaptic vesicles. The pelleted
membranes were then subjected to Western blot or processed for
immunoprecipitation of the Triton X-100-solubilized SV2·900-kDa
laminin complex. Since the partially purified synaptosomal preparations
contained postsynaptic membranes as well as synaptosomes, we checked
the effectiveness of our treatment by looking at whether rapsyn was
removed from acetylcholine receptor. Alkaline pH clearly stripped
rapsyn (Fig. 3A, rapsyn,
lane 2) from acetylcholine receptor (Fig. 3A,
AchR, lane 2), whereas neutral pH was without effect (Fig.
3A, rapsyn and AchR, lane 1). The amount of rapsyn decreased
~6-fold after high pH treatment. In addition, the pH 11.5 solution
lysed synaptosomes and removed synaptic vesicles. The amount of the
synaptic vesicle protein synaptophysin in synaptosomes is decreased by
the alkaline pH treatment (Fig. 3A, p38, compare
lane 1 with lane 2). Presumably the synaptophysin
that remains associated with synaptosomes represents either tightly
docked synaptic vesicles or synaptic vesicle membrane that has
undergone exocytosis.
In contrast, exposure of synaptosomes to pH 11.5 did not dissociate
laminin from SV2H. The same amount of 900-kDa laminin was isolated from
synaptosomes by immunoprecipitation of Triton X-100-solubilized SV2,
whether or not the membranes were exposed to pH 11.5 (Fig.
3B, compare lanes 2c with 1a). Thus,
the interaction between this extracellular matrix protein and SV2H is
stronger than that between the intracellular peripheral membrane rapsyn and acetylcholine receptor.
Biotinylation of Synaptosomes Indicates That the 900-kDa Laminin Is
a Major Component of the SV2H·Laminin Complex--
We next wanted to
determine how many proteins were tightly associated with SV2H on the
synaptosomal surface. We alkaline-stripped synaptosomes with pH 11.5 as
described above, biotinylated these membranes with sulfo-NHS-Biotin,
and then immunoprecipitated Triton X-100 solubilized SV2H from the
labeled membranes with anti-SV1 mAb. We used alkaline-stripped
synaptosomes so that we would only detect proteins tightly bound to
SV2H. However, the results are essentially the same if
non-alkaline-stripped synaptosomes are used (data not shown). The
immunoprecipitated proteins were subjected to SDS-PAGE under reducing
conditions and electroblotted to nitrocellulose. We used
sulfo-NHS-Biotin as a labeling reagent because we found that it reacts
poorly with purified SV2 compared with other synaptic vesicle proteins
(data not shown). It was important to minimize SV2 labeling so that
co-immunoprecipitating proteins would not be obscured on nitrocellulose
blots when stained for biotinylated proteins.
About 50% of the biotin label is present on the polypeptide chains
derived from the
The anti-
There are five major co-immunoprecipitating biotinylated polypeptides
that we have not identified (Fig. 4, lane 1, arrowheads e
and f). Three polypeptides with molecular masses between 120 and 150 kDa (Fig. 4, lane 1, arrowhead e) together contain
about 20% of the biotin label. These polypeptides might be related to the unidentified Purified SV2H Binds Laminin-1--
Since SV2H and the 900-kDa
laminin are complexed together, it seemed possible that they interact
directly. Unfortunately, methods for purifying the 900-kDa
Torpedo laminin are not yet available. However, laminin-1
contains
We developed a new procedure for the preparative purification of SV2H
from synaptic vesicles. We first partially purified synaptic vesicles
from electric organ homogenates following the methods of Carlson
et al. (38) involving differential centrifugation and
flotation equilibrium on sucrose density gradients. The synaptic vesicle proteins were solubilized with Triton X-100 and SV2 purified with the anti-SV2 mAb covalently linked to Affi-Gel-10. The SV2 was
eluted from the antibody affinity column with a pH 11.5 solution, and
the isolated proteoglycan was labeled with 125I. To isolate
the labeled SV2H from the L form, 125I-SV2 was
chromatographed on a tomato lectin-agarose column. By SDS-PAGE the
flow-through contained the L form, a protein migrating at about 90 kDa
(Fig. 5A, lane 2), whereas the H form bound the lectin
column and was eluted by triacetylchitotriose (Fig. 5A, lane
1). To confirm that H and L forms had been separated, we immunoprecipitated the 125I-labeled SV2, which had bound
the lectin column, and SV2, which had not. As expected, the SV2 that
bound lectin was immunoprecipitated by both anti-SV1 and anti-SV2 mAbs
(Fig. 5B, upper graph) indicating that it was the H form.
The SV2 that had not bound lectin had the characteristics of the L
form; it was immunoprecipitated only by the anti-SV2 mAb and not
anti-SV1 (Fig. 5B, lower graph).
The H form of SV2 as it elutes from the lectin column is at least 97%
pure (Fig. 5A, lane 1). We previously reported several unidentified proteins co-immunoprecipitating with SV2 from purified synaptic vesicles (2). However, we now believe that these were simply
fragments of SV2, since they have been eliminated by controlling proteolysis.
To measure the binding between laminin-1 and SV2H, we used ACE. ACE is
a gel retardation procedure (40) whereby the 125I-SV2 H
form is subjected to electrophoresis in the presence of laminin. An
agarose gel is poured so that a lane containing 125I-SV2H
is positioned perpendicular to several lanes containing different
concentrations of laminin-1 (Fig.
6A). During electrophoresis, the front of 125I-SV2H migrates through the
laminin-containing lanes (Fig. 6B). When the laminin
concentration is above the dissociation constant (Kd) of the binding reaction between SV2H and
laminin, the migration of SV2H in the agarose gel is retarded. If the
laminin concentration is far below the dissociation constant, the SV2H migrates unimpeded. At intermediate concentrations of laminin, SV2H has
intermediate mobilities (Fig. 6A), quantitatively reflecting the amount of SV2H bound to laminin (40).
The mobility of SV2H is greatly retarded on ACE at laminin-1
concentrations above about 100 nM (Fig. 6B),
indicating strong binding between laminin-1 and SV2H. In order to
determine the dissociation constants (Kd values) for
laminin-1 binding to SV2H, we determined the distribution of
125I-SV2H in the ACE gel lanes (for example, Fig.
6C, upper and lower graphs).
Interestingly, at high concentrations of laminin-1 (e.g. Fig. 6C, lower graph), we found a pattern
suggesting that a subfraction of SV2H bound more tightly than the
majority. In the absence of laminin-1, SV2H migrates as a single
symmetrical peak of high mobility (Fig. 6C, upper graph). In
the presence of laminin-1, a slow migrating peak of SV2H (Fig.
6C, lower graph, labeled m10%) is seen
superimposed on a larger distribution. This slowest moving peak (most
tightly bound to laminin) migrates at 10% of the most retarded SV2H.
Therefore, we determined m10% which is the mobility or the distance migrated by the most retarded 10% of SV2H
(Fig. 6C, lower graph). In addition, we also
determined m50% which is the distance migrated
for the midpoint of the SV2H distribution (Fig. 6C, lower).
This latter mobility represents the average distance migrated for all
of SV2H and is the standard measurement used by Lee and Lander (40) and
San Antonio et al. (48). From these types of measurements
with five separate ACE gels (one of which is shown in Fig.
6B), we calculated two retention coefficients (R10% and R50%), where
R = (n Hypothesis, SV2H Binds the 900-kDa Laminin--
We have shown that
SV2H is associated with a 900-kDa laminin on synaptosomes isolated from
electric organ. When solubilized in Triton X-100 from synaptosomes,
these two proteins were found to co-immunoprecipitate. The association
between these proteins is specific, since another laminin isoform on
the synaptosome is not associated with SV2. Moreover, a presynaptic
plasma membrane protein, the calcium channel, and a synaptic vesicle
protein, synaptotagmin, are not part of the SV2H·900-kDa laminin
complex. The interaction between SV2H and the 900-kDa laminin is strong in that it resists dissociation by pH 11.5, a condition that can solubilize peripheral membrane proteins. We find that the 900-kDa laminin is an
We hypothesize that SV2H is binding directly to the Characteristics of SV2H Binding to Laminin-1--
Our finding of a
higher affinity subfraction of SV2H for mouse laminin-1 suggests that
the interaction might be mediated partially or completely by the
keratan sulfate chains of SV2H. This observation is reminiscent of the
binding of another glycosaminoglycan, heparin, to laminin-1. The
g domains of the
Although the binding of SV2H to laminin-1 has some similarities to the
binding of heparin, there is also an important difference. The
subfraction of heparin that binds laminin-1 with a
Kd of 67 nM also binds the extracellular
matrix protein, fibronectin, with a Kd of 190 nM (48), a ratio of about 1:3. A subfraction of SV2H binds
to laminin-1 with a Kd of 68 nM and to fibronectin with a Kd of 1000 nM, a
ratio of about 1:15. Compared with fibronectin, SV2H has a greater
specificity for laminin-1 than does heparin.
We attempted to test whether the keratan sulfate chains are involved in
binding the Other Known Interactions of SV2H--
Bennet et al.
(53) proposed that synaptic vesicle proteins form a multimeric complex.
When rat brain synaptic vesicles were solubilized in the detergent
CHAPS, synaptophysin, SV2, synaptotagmin, VAMP, and the vacuolar
H+ pump were found in a complex. This entire complex is
probably loosely associated, since other non-denaturing detergents
(e.g. Triton X-100 and octyl glucoside) partially dissolve
it. The interaction between SV2 and synaptotagmin is most likely
stronger, since these proteins remained associated after Triton X-100
solubilization. Schivell et al. (52) confirmed and extended
this latter observation. They demonstrated that SV2 and synaptotagmin
could be selectively cross-linked in intact membranes and further
showed that the cytosolic N terminus of SV2A was binding the cytosolic
C2B region of synaptotagmin.
The protein complexes characterized by Bennet et al.
(53) and Schivell et al. (52) are distinct from the
laminin·SV2H complex we have described here as follows. 1) The
laminin·SV2H complex is immunoprecipitated by the anti-SV2 mAb
whereas the SV2·synaptotagmin complex is not. Bennet et
al. (53) and Schivell et al. (52) found that the
SV2·synaptotagmin complex was only immunoprecipitated by
anti-synaptotagmin antibody. Since the anti-SV2 mAb immunoprecipitated
only SV2 uncomplexed with synaptotagmin, Schivell et al.
(52) proposed that the association between SV2 and synaptotagmin
sterically hindered the anti-SV2 mAb from binding its antigenic site.
Indeed, the SV2 epitope and the synaptotagmin-binding site are both in
the N-terminal region (52). Similarly, with Triton X-100-solubilized
electric fish vesicles, we find no synaptotagmin immunoprecipitated by
anti-SV2 mAbs (2). 2) Bennet et al. (53) and Schivell
et al. (52) studied the interaction of SV2L with synaptotagmin and not SV2H, which we studied here. The SDS-PAGE system
they employed (4-12% gradient gels) does not allow for the detection
of SV2H, since SV2H does not enter this resolving gel (2). In fact, it
is possible that SV2H does not interact with synaptotagmin. With the
anti-SV1 mAb, we find no synaptotagmin co-immunoprecipitating with SV2H
from Triton X-100-solubilized synaptosomes or synaptic vesicles (Fig.
2A and Ref. 2).
SV2H as a Laminin Receptor--
Our finding that the
SV2H·900-kDa laminin complex resists dissociation by pH 11.5 suggests
that the binding between these proteins is of high affinity. However,
the measured dissociation constant of 68 nM for SV2H and
laminin-1 suggests a binding with moderate affinity. How might this
apparent discrepancy be reconciled? One possibility is that the
SV2H-binding site on the 900-kDa laminin is of much higher affinity
than the homologous site on laminin-1. This might not be so surprising,
since laminin-1 is not a synaptic laminin and the binding site might
have a different purpose. Alternatively, the interaction between SV2H
and the 900-kDa laminin in this protein complex might involve
multipoint attachment, which would greatly increase the strength of the
interaction. Such a possibility is similar to the binding of an IgG
molecule that is 50-100 times stronger with a bivalent antigen than
with a monovalent antigen (54). With a bivalent antigen both of the
IgG-binding sites are attached to the same antigen molecule; with
monovalent antigen only one is attached. In an analogous manner two
linked SV2H proteins might bind two associated
Our discovery that SV2H and the 900-kDa laminin are associated on the
synaptosomal surface suggests the following scenario (Fig. 8). In
synaptic vesicles SV2H is not complexed with laminin. However, once
this membrane protein becomes incorporated into the nerve terminal
plasma membrane due to exocytosis from the release of neurotransmitter,
SV2H has the opportunity to bind the 900-kDa laminin of the synaptic
basal lamina. If the extracellular domain of SV2H does not encounter
laminin, then it would follow the known path of the other synaptic
vesicle proteins. SV2 would enter a coated pit, undergo endocytosis,
and become reincorporated into synaptic vesicles. If SV2H does bind
laminin, it would form a linkage between the presynaptic membrane and
the extracellular matrix. A cytosolic domain of SV2H could serve as an
anchor for the nerve terminal cytoskeleton.
This hypothesis has several interesting implications as follows. 1)
Work at the neuromuscular junction suggests that during nerve
regeneration after injury, a basal lamina component acts as a cue for
the placement of exocytotic machinery for neurotransmitter release by
the returning axon (17). A synaptic vesicle transmembrane protein like
SV2H would be inserted into the presynaptic membrane in an ideal
position to link the exocytotic machinery to an ECM cue. 2) This
hypothesis suggests an activity-dependent mechanism for
synaptic adhesion. The more neurotransmitter that is released, the more
SV2H that cycles through the presynaptic plasma membrane, and the
greater the chance that SV2H will encounter and bind laminin in the
basal lamina. Activity-dependent synaptic adhesion could be
important in a process like synaptic competition (55).
We thank Dr. Stanley Froehner for the
monoclonal antibodies directed against the *
This work was supported by National Institutes of Health
Grant NS22367 (to S. S. C. and J. R. S.).The costs of publication of this
article were defrayed in part by the
payment of page charges. The article
must therefore be hereby marked
"advertisement" in accordance with 18 U.S.C. Section
1734 solely to indicate this fact.
¶
To whom correspondence should be addressed: Dept. of
Physiology and Biophysics, Box 357290, University of Washington,
Seattle, WA 98195-7290. Tel.: 206-543-8294; Fax: 206-695-0619; E-mail: ssc1@u.washington.edu.
2
Sunderland, W. J., Son, Y.-J., Miner, J. H.,
Sanes, J. R., and Carlson, S. S. (2000) J. Neurosci, in press.
3
S. C. Carlson, unpublished observations.
The abbreviations used are:
SV2, synaptic
vesicle protein 2;
SV2H, SV2 heavy form;
SV2L, SV2 light form;
SV1, a
keratan sulfate epitope of SV2;
anti-SV1, a monoclonal antibody that
binds the SV1 epitope;
anti-SV2, monoclonal antibody that binds the
cytosolic protein epitope of SV2;
PAGE, polyacrylamide gel
electrophoresis;
ECM, extracellular matrix;
MOPC-21, mouse myeloma
immunoglobulin;
ACE, affinity co-electrophoresis;
BSA, bovine serum
albumin;
mAb, monoclonal antibody;
CAPS, 3-(cyclohexylaminol)propanesulfonic acid;
sulfo-NHS-Biotin, sulfosuccinimidobiotin;
Me2SO, dimethyl sulfoxide;
MOPS, 4-4-morpholinepropanesulfonic acid;
CHAPS, 3-[(3-cholamidopropyl)di-methylammonio]-1-propanesulfonic acid;
AchR, acetylcholine receptor;
R, R10%, and
R50%, retention coefficients.
The Synaptic Vesicle Protein SV2 Is Complexed with an
5-Containing Laminin on the Nerve Terminal Surface*
, Medicine and of § Anatomy and
Neurobiology, Washington University School of Medicine,
St. Louis, Missouri 63110
ABSTRACT
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES
5 or 
2
chains are potential basal lamina ligands for these interactions. The nerve terminal receptors are unknown. Here we show that SV2, a synaptic
vesicle transmembrane proteoglycan, is complexed with a 900-kDa laminin
on synaptosomes from the electric organ synapse that is similar to the
neuromuscular junctions. Although two laminins are present on
synaptosomes, only the 900-kDa laminin is associated with SV2. Other
nerve terminal components are absent from this complex. The 900-kDa
laminin contains an 5, a 1, and a novel 
chain. To test
whether SV2 directly binds the 900-kDa laminin, we looked for
interaction between purified SV2 and laminin-1, a laminin isoform with
a similar structure. We find SV2 binds with high affinity to purified
laminin-1. Our results suggest that a synaptic vesicle component may
act as a laminin receptor on the presynaptic plasma membrane; they also
suggest a mechanism for activity-dependent adhesion at the synapse.
INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES
, a ,
and a chain (20). Five chains, three chains, and three chains have been described to date (20-24). Three chains, 5, 4,
and 2, have been specifically localized to the synaptic basal lamina
of the neuromuscular junction (25, 26), although the exact trimeric
structures at this synapse have not been elucidated. Mutant mice that
lack one of these chains, 2, have a pronounced defect in
neuromuscular development that includes impaired maturation of nerve
terminals (27) and also show defects in reinnervation of muscle
following nerve damage (28).
5, 1, and a novel chain. However, SV2H is not associated with a laminin protein in purified synaptic vesicles. The
interaction of terminal-associated SV2H with the 900-kDa laminin is
strong in that it resists dissociation by high pH. In addition, purified SV2H binds purified laminin-1. We hypothesize that SV2H can
act as a laminin receptor after neurotransmitter release, at which time
SV2H is still part of the presynaptic plasma membrane. Because the
amount of SV2H on the nerve terminal membrane should be directly
related to the level of vesicle exocytosis, the interaction of SV2H and
laminin could be highly activity-dependent.
EXPERIMENTAL PROCEDURES
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES
5 chain rabbit antisera were made against a fusion
protein as described in Miner et al. (22). It shows no
cross-reactivity with the 1, 1, or 1 chains of laminin-1 (22).
The anti-laminin 1 mAb (D18) was generated as described by Sanes
et al. (25) and was purchased from Developmental Studies
Hybridoma Bank (University of Iowa, Iowa City, IA). Monoclonal
antibodies directed against the subunit of acetylcholine receptor
and rapsyn were gifts from Dr. Stanley Froehner (University of North
Carolina, Chapel Hill, NC). Affinity-purified anti-Ca2+
channel (1 subunit, residues 1382-1400) antibodies were a gift from
Dr. Ruth Westenbroek and Dr. William Catterall (University of
Washington, Seattle, WA) and were prepared according to Striessnig et al. (32). The anti-agrin monoclonal antibody, 11D2, was a gift from Dr. Justin Fallon (Brown University, Providence, RI). The
SABC kit for detection of biotinylated protein was purchased from
Zymed Laboratories Inc. (San Francisco, CA). Tomato
(Lycopersicon esculentum) lectin agarose was obtained from Vector
Laboratories (Burlingame, CA).
1 chain has a
molecular mass of 440 kDa, the 1 chain 230 kDa, and 1 chain 220 kDa (42).
RESULTS
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES
View larger version (32K):
[in a new window]
Fig. 1.
SV2 is complexed with a 900-kDa laminin
isoform on the surface of synaptosomes but not in synaptic
vesicles. A, anti-SV1 mAb (lanes 1, 3, and
5) or control mAb (lane 2, 4, and 6)
were used to immunoprecipitate Triton X-100-solubilized SV2H from the
surface of synaptosomes. Western blots of the immunoprecipitates were
probed with anti-SV2 mAb (lanes 1 and 2),
anti-SV1 mAb (lanes 3 and 4), or anti-laminin-1
antibody (lanes 5 and 6, upper and
lower). SDS-PAGE was performed under reducing (lanes
1-4, 5 and 6, lower) or non-reducing conditions
(lanes 5 and 6, upper). B, SV2 was
immunoprecipitated from Triton X-100-solubilized and purified synaptic
vesicles with anti-SV1 (lane 1), or anti-SV2 (lane
2), or control mAb (lane 3). Western blots of these
antigen-antibody conjugates were probed with either anti-SV2
(upper) or anti-laminin-1 antibodies (lower). In
lane 2, the H form and L form (at about 90 kDa) are not
resolved from one another. SDS-PAGE was performed under reducing
conditions. C, the 900-kDa laminin and SV2 are
co-immunoprecipitated from Triton X-100-solubilized synaptosomes with
the anti-SV2 mAb (lane 1) but not with a control mAb
(lane 2). The Western blots were probed with anti-laminin-1,
and SDS-PAGE was performed under non-reducing conditions. Molecular
mass markers are shown at the right of the lanes and are
expressed in kDa. Laminin-1 was used to determine the mobility of a
900-kDa protein on SDS-PAGE. Abbreviations used are as follows:
Ln 1, anti-laminin-1 Abs; Ctl, control mAb;
IP, immunoprecipitation; Blot, Western
blot.
1, 1, and 1 chains of mammalian laminin. By
using Western blots following SDS-PAGE under reducing conditions, we
observed a co-immunoprecipitating immunoreactive protein (Fig. 1A, lane 5, lower) migrating about 230 kDa (228 ± 7 (S.D.)). The size of this protein is similar to that expected for a or chain (20). When the SDS-PAGE was performed under non-reducing conditions to preserve the disulfide linkages between the 3 laminin subunits, the laminin migrated at about 900 kDa (Fig. 1A, lane 5, upper). This mobility is almost identical to that of the mammalian laminin heterotrimers that contain 1, 2, or 5 chains (22).
1 subunit of the
calcium channel, a presynaptic plasma membrane protein (19, 45), and
synaptotagmin, a synaptic vesicle protein (19, 31). First, we confirmed
that these proteins were present in intact immunopurified synaptosomes, using the procedures of Miljanich et al. (34) and Buckley
et al. (15). This immunopurification method employs the
anti-SV1 mAb to bind intact synaptosomes to immunobeads. The resulting preparation is enriched for nerve terminal components and contains at
least 10 times less postsynaptic (acetylcholine receptor-containing) membranes (34). In fact, we are unable to detect acetylcholine receptor
in these membranes by Western blotting methods.2 With these
synaptosomes we find that both 1 subunit of the calcium channel
(Fig. 2A, middle, lane 1, arrow) and two isoforms of synaptotagmin (31) are present (Fig.
2A, lower, lane 1, arrows). However, unlike the 900-kDa
laminin (Fig. 2A, upper, lane 3) neither protein is
specifically co-immunoprecipitated with solubilized SV2H (Fig. 2A, middle and lower, lanes 3). A very
small amount of the 740-kDa laminin and Ca+2 channel is
detected in both anti-SV2 mAb (lane 3) and control mAb
(lane 4) due to nonspecific adsorption by immunobeads.
View larger version (42K):
[in a new window]
Fig. 2.
Solubilized SV2H co-immunoprecipitates with
an 5-containing laminin but not with other
synaptosomal proteins. A, lanes 1 and
2, intact synaptosomes were immunoprecipitated
(IP) with the anti-SV1 mAb (lane 1,
SV1) or a control mAb (lane 2, Ctl).
These synaptosomes were subjected to Western blotting (Blot)
with anti-laminin-1 (upper, Ln 1),
anti-Ca2+ channel (middle, gCa), and
anti-synaptotagmin (lower, p65) antibodies. Lanes
3 and 4, solubilized SV2H was immunoprecipitated from
synaptosomes with either anti-SV1 mAb (lane 3,
SV1) or a control mAb (lane 4, Ctl)
and subjected to Western blotting with anti-laminin-1(upper, Ln
1), anti-Ca2+ channel, 1 subunit (middle,
gCa), or anti-synaptotagmin (lower, p65) antibodies.
B, upper, SV2H was immunoprecipitated from
solubilized synaptosomes as described for lanes 3 and
4 in A, and the resulting Western blots were
probed with anti-5 antibodies (lanes 1 and 2,
respectively, 5) or with anti-laminin-1 (lanes
3 and 4, respectively, Ln 1).
Lower, SV2H was immunoprecipitated from solubilized
synaptosomes as described for lanes 3 and 4 in
A, and the resulting Western blot probed first with an
anti-1 mouse monoclonal antibody (lanes 1 and
2, respectively, 1), and then the same blot
was reprobed with anti-laminin-1 rabbit antibody (lanes 3 and 4, respectively, Ln 1). SDS-PAGE was
performed under non-reducing conditions for A,
upper, and B, and reducing conditions for
A, middle and upper.
chain
must be large, approximately 400 kDa, probably an 1, 2, or 5,
and not the smaller 3 or 4 chains (22). Indeed, we found that
this laminin was stained with an anti-5 antibody on Western blots
(Fig. 2B upper, lane 1, arrow). In addition, the presence of
a 230-kDa chain that cross-reacts with an anti-laminin-1 antibody
suggested that this laminin contains either a 1 or 1 chain.
Moreover, we were unable to stain the laminin with a mouse monoclonal
antibody directed against the 1 chain (Fig. 2B lower lane 1, arrow), even though this antibody reacts with the 740-kDa laminin
on electric organ synaptosomes (data not shown). Thus, the 230-kDa
chain that is stained by the anti-laminin-1 antibody is most likely a
1 chain and not a 1 chain. However, it is still possible that the
lack of reactivity of the 900-kDa laminin with the anti-1 mAb could
be due to epitope masking or removal of the epitope from the 1 chain
by alternative slicing. As yet we have not identified the chain in
the 900-kDa laminin. Finally, the 900-kDa laminin shows no
immunoreactivity with anti-2 laminin antibodies, whereas the 740-kDa
laminin is reactive.2
View larger version (16K):
[in a new window]
Fig. 3.
Strength of the association of SV2H with the
900-kDa laminin. Partially purified synaptosomes that contain both
pre- and postsynaptic membranes were exposed to solutions of either pH
7.4 (lanes 1) or pH 11.5 (lanes 2). A,
Western blots of these treated pH synaptosomes were probed with
antibodies directed against AchR, rapsyn, and synaptophysin
(p38). B, solubilized SV2H was immunoprecipitated
from these pH-treated synaptosomes with either anti-SV1 (lanes
1a and 2c) or a control mAb (lanes 1b and
2d). The immunoprecipitated SV2 was subjected to Western
blot and probed with anti-laminin-1 antibodies
(Ln/Sppt).
5 and 1 chains of the 900-kDa laminin. A major
biotinylated polypeptide migrates at 230 kDa (Fig.
4, lane 1, arrowhead b). When
the blot is reprobed with anti-laminin 1, a 230-kDa polypeptide also
stains (Fig. 4, lane 3). This polypeptide is most likely the
1 chain for reasons detailed above. A biotinylated polypeptide is
found migrating at 380 kDa (Fig. 4, lane 1, arrowhead a).
This polypeptide chain stains with the anti-5 antibody (Fig. 4,
lane 3) and is about the molecular weight expected for the 5 chain (22). In addition, we see several other biotinylated polypeptides that also stain with the anti-5 antibody (Fig. 4, lane 5) as follows: the major biotinylated protein migrating
at 230 kDa (Fig. 4, lane 1, arrowhead b) and minor
biotinylated proteins between 190 and 210 kDa (Fig.
5, lane 1, arrowhead c) and
170 kDa (Fig. 4, lane 1, arrowhead d). These
smaller polypeptides are quite likely fragments of the 5 chain that
are held together by disulfide bonds in the intact unreduced protein. A
similar pattern of fragmentation has been seen with laminin 5 chains from mouse lung and kidney tissue (22). With these mouse tissues, Western blots probed with the same anti-5 antibodies reveal major polypeptides migrating at about 380 and about 210 kDa, like the major
proteins seen here at 380 and 230 kDa. The finding that both anti-5
and anti-laminin-1 antibodies both identify a polypeptide at 230 kDa is
probably due to co-migration of two polypeptides and not to
cross-reactivity of the anti-5 antibody with the 1 chain. The
anti-5 antibody shows no cross-reactivity with purified mouse
laminin-1 chains (22).
View larger version (53K):
[in a new window]
Fig. 4.
Biotinylation of the SV2H·900-kDa laminin
complex. Partially purified synaptosomes were alkaline-stripped
with pH 11.5 solutions and biotinylated with sulfo-NHS-Biotin. The
Triton X-100-solubilized, SV2·laminin complex was immunoprecipitated
(IP) from these synaptosomes with anti-SV1 mAb (lanes
1, 3, and 5, SV1) or a control mAb
(lane 2, 4, and 6, Ctl). The
immunoprecipitates were subjected to SDS-PAGE and transferred to
nitrocellulose. These blots were stained for biotinylated protein
(lanes 1 and 2, Biotin), with
anti-laminin-1 antibodies (lanes 3 and 4,
Ln 1), or with anti- 5 antibodies (lanes 5 and
6, 5). Lanes 3 and 4 are
the same blot as lanes 1 and 2 reprobed with
anti-laminin-1 antibodies. Likewise, lanes 5 and
6 are from another identical blot first probed for
biotinylated proteins (data not shown) and then reprobed with anti-5
antibodies. The polypeptide b of lane 1 migrates
identically to the corresponding polypeptide of lane 3. The
polypeptides a-d of lane 1 migrate identically
to the corresponding peptide of lane 5.
View larger version (37K):
[in a new window]
Fig. 5.
Purification of SV2H. SV2 was purified
from synaptic vesicles, labeled with 125I, and the H form
separated from the L form by chromatography on tomato lectin-agarose.
A, 125I-labeled SV2, bound to tomato
lectin-agarose (lane 1), as well as SV2, passing through the
column without binding (lane 2), was subjected to SDS-PAGE.
An autoradiogram of the SDS-polyacrylamide gel is shown. Free
125I was removed from 125I-SV2L (lane
2) by sedimentation velocity on sucrose gradients before SDS-PAGE
(see "Experimental Procedures"). The arrow marks the
beginning of the resolving gel. B, the lectin-bound SV2
contains the antigenicities expected for the H form; it was
immunoprecipitated by both anti-SV1 (SV1) and anti-SV2
(SV2) mAbs (upper graph) but not by a control mAb
(Ctl). SV2 that was not bound by tomato lectin (lower
graph) behaved like the L form; it was not immunoprecipitated by
anti-SV1 (SV1) mAb, only by the anti-SV2 mAb
(SV2).
5 antibodies detect three polypeptides migrating between
550 and 650 kDa (Fig. 4, lane 5, asterisk) that are not labeled by biotin. These may be small amounts of partially reduced 5-containing laminin molecules lacking the antigenic region of the
1 chain.
chain. Interestingly, the 2 chain made by rat
and human cell lines has molecular a mass about 155 kDa
(e.g. Ref. 47). The two polypeptides of 60-66 kDa (Fig. 4,
lane 1, arrowhead f) contain about 20% of the
biotin label. The remaining 10% of the label is scattered among five
very minor polypeptides that contain about 2% each.
1, 1, and 1 chains. This laminin is similar to the
900-kDa laminin in that both contain the 1 chain, and the 5 chain
has the same domain structure as the 1 chain (21). Both chains
contain domains I-VI as well as the G region. We reasoned that
laminin-1 could contain a domain homologous to an SV2H-binding site on
900-kDa laminin. Although the domain could be tailored for another
purpose, it might be similar enough to still bind SV2H. We therefore
assayed binding between purified SV2H and commercially available
laminin-1.
View larger version (45K):
[in a new window]
Fig. 6.
The binding of SV2H by laminin-1 measured by
ACE. A, the configuration of the lanes in an agarose
gel for ACE is shown. At the beginning of the electrophoresis
125I-SV2H is placed in the long horizontal lane.
Laminin-1 is mixed with agarose and placed in the vertical
lanes, with each lane containing a different concentration of
laminin. The direction of the current flow is shown. At the end of the
electrophoresis, a front of 125I-SV2H has migrated through
the laminin-containing lanes. If laminin binds SV2H, significant
retardation of the SV2H mobility should occur in those lanes where the
concentration of laminin is near the dissociation constant of the
laminin·SV2H binding reaction. n is the distance traveled
by SV2H in the absence of laminin; m is the distance
traveled in the presence of laminin. B, a typical ACE gel
for 125I-SV2H and laminin-1. An autoradiograph of the gel
is presented. The laminin-1 concentration is shown below
each lane. "Or" marks the beginning of the
laminin-containing lanes. C, a plot of the optical density
versus distance migrated for SV2H at 0 nM
laminin-1 (upper graph) and 183 nM laminin 1 (lower graph). These data were determined from the
autoradiograph of the ACE gel in B. For each lane containing
laminin two migration distances were determined by integrating the
optical density versus migration distance curve as follows:
m10% for the most retarded 10% of the SV2H,
and m50% for the midpoint of the SV2H
(lower graph). m10% was chosen since
it tracked the peak of the most retarded SV2H subfraction. For SV2H in
the migrating in the absence of laminin, the corresponding migration
distances, n10% and
n50% were also determined (upper
graph).
m)/n
for each laminin-1 concentration. A semilog plot of
R10% and R50%
versus laminin-1 concentration yields two sigmoidal curves
(Fig. 7, filled circle and
open triangle respectively). Following the methods of San
Antonio et al. (48), we fitted each curve with the equation R = 1/(1 + Kd/[laminin-1]) to
calculate two dissociation constants:
Kd(10%) = 68 ± 9 nM
(±S.E.) and Kd(50%) of 330 ± 100 nM (±S.E.). Thus, all SV2Hs forms bind laminin-1, but a
high-affinity subfraction is probably present. In addition, no binding
is seen between the SV2H form and BSA on ACE gels, and only weak
binding (for R10%, Kd ~1
µM) with another ECM protein, fibronectin.
View larger version (16K):
[in a new window]
Fig. 7.
Determination of the dissociation constants
(Kd values) for the binding of SV2H to
laminin-1. A semilog plot of the retardation coefficient
(R) (y axis) versus laminin-1
concentration (x axis) calculated from five ACE gels where
SV2H was subjected to electrophoresis in the presence of laminin-1.
R was calculated from these ACE gels with the equation,
R = (n m)/n,
where m = distance migrated by SV2H in the presence of
a specific concentration of laminin, and n = distance
migrated by SV2H in the absence of laminin. For each laminin
concentration, two m values were measured (see Fig. 7):
m10% for the most retarded SV2H, and
m50% for the average of all SV2H. Thus, two
R values were calculated, R10%
(closed circles) and R50%
(open triangles) for each laminin concentration. By using
the relationship R = 1/(1 + Kd/[laminin]) from San Antonio et al.
(48), we fitted curves (solid lines) to the two sets of
data.
DISCUSSION
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES
51 trimer. Immunoprecipitation of the
solubilized SV2H protein complex from biotinylated, alkaline-stripped
synaptosomes reveals that the 900-kDa laminin is a major component of
the complex.
51 trimer
in the SV2H·laminin complex. This is based on the binding between
pure laminin-1 and SV2H purified from synaptic vesicles. Laminin-1 is
similar to the 51 trimer in that both share the 1 chain,
and the 1 chain has a very similar domain structure to the 5
chain. Presumably, laminin-1 contains an SV2H-binding site that is
homologous, although not necessarily identical, to the SV2H-binding
site on the 51 trimer. Such a site is not present on all ECM
molecules, since fibronectin shows very weak binding.
1 chain are known to contain heparin-binding sites (20). San Antonio et al. (48) have
isolated two subfractions of heparin that bind laminin-1 with
Kd values of about 67 and 800 nM. The
high affinity heparin subfraction is probably due to the presence of a
particular oligosaccharide sequence or to a series of related sequences
that are specifically recognized by a site on laminin-1 (48). Although
heparin is a polymerized disaccharide unit of uronic acid and
N-acetylglucosamine, a specific sequence results from a
particular set of modifications (especially sulfation) along the
polysaccharide chain (49). Likewise, keratan sulfate chains
(polymerized galactose and n-acetylglucosamine) are known to be
modified by sulfation, fucosylation, and sialylation (e.g.
Ref. 50), which could result in high-affinity laminin-binding sites on
a subfraction of SV2H keratan sulfate chains. Interestingly, the
keratan sulfate chains of SV2 are known to be modified; they contain
several unique keratan sulfate epitopes (2), one of which (1B4) is
known to be the result of extensive sulfation (51). Laminin-1 has not
been tested for binding by free keratan sulfate chains.
51 laminin to SV2H by digestion of the complex
with keratanase II and endo--galactosidase. The integrity of the
complex was unaffected by the digestion. However, this negative result
does not rule out that the keratan sulfate chains mediate the
interaction between 51 laminin and SV2H. We found previously
that the chains of SV2H are incompletely digested by keratanase I, or
keratanase II, or endo--galactosidase, or all three enzymes
(2).3 Presumably some
modification of the keratan sulfate chains makes them partially
resistant to these enzymes.
51 trimers
(Fig. 8B). Alternatively, SV2H
might be associated with another transmembrane protein in the protein
complex which also binds the 900-kDa laminin (Fig. 8C).
View larger version (43K):
[in a new window]
Fig. 8.
Hypothesis: SV2H is a laminin receptor.
We propose that SV2H in the presynaptic plasma membrane is a receptor
for a 900-kDa laminin embedded in the synaptic extracellular matrix.
Unbound SV2H is delivered to the presynaptic plasma membrane by fusion
of the synaptic vesicle membrane during neurotransmitter release. Once
SV2H is free to diffuse in this membrane, it could encounter an
unoccupied laminin binding site and bind. Alternatively it could enter
a coated pit, undergo endocytosis, and become reincorporated into
synaptic vesicles. Given that the majority of the SV2H is present in
synaptic vesicles in the nerve terminal, most of the time SV2H follows
this latter route. The SV2H which binds laminin forms a linkage between
the nerve terminal and synaptic ECM. We would expect that a cytosolic
domain of SV2H would act as a cytoplasmic anchor for the nerve terminal
cytoskeleton. In the model we show SV2H molecules associating with
laminin in three possible ways: A, as a monomer,
B, as a homodimer, and C, as a heterodimer. SV2H
may have two functions in the nerve terminal: to act as a laminin
receptor on the nerve terminal surface and to have an as yet
undetermined function in the synaptic vesicle. Alternatively, synaptic
vesicles may simply act as vehicles to place SV2H in the active zone of
the presynaptic plasma membrane, where it can bind laminin and act as
an anchor for the nerve terminal.
ACKNOWLEDGEMENTS
subunit of acetylcholine
receptor and rapsyn and Dr. Ruth Westenbroek and Dr. William Catterall
for the affinity-purified anti-Ca2+ channel antibodies. We
also thank Connie Missimer and Lauri Levi for editorial help and Don
Clifton for help with computer graphics.
FOOTNOTES
ABBREVIATIONS
REFERENCES
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES
1.
Buckley, K.,
and Kelly, R. B.
(1985)
J. Cell Biol.
100,
1284-1294 2.
Scranton, T. W.,
Iwata, M.,
and Carlson, S. S.
(1993)
J. Neurochem.
61,
29-44[CrossRef][Medline]
[Order article via Infotrieve]
3.
Stadler, H.,
and Dowe, G. H. C.
(1982)
EMBO J.
1,
1381-1384[Medline]
[Order article via Infotrieve]
4.
Carlson, S. S.,
and Kelly, R. B.
(1983)
J. Biol. Chem.
258,
11082-11091 5.
Bennett, M. V. L.
(1971)
in
Fish Physiology
(Hoar, W. S.
, and Randall, D. J., eds), Vol. 5
, pp. 347-493, Academic Press, New York
6.
Hall, Z. W.
(1992)
Molecular Neurobiology
, pp. 90-91, Sinauer Associates, Inc., Sunderland MA
7.
Kushner, P. D.
(1984)
J. Neurochem.
43,
775-786[CrossRef][Medline]
[Order article via Infotrieve]
8.
Snow, A.,
Nochlin, D.,
Sekiguchi, R.,
and Carlson, S. S.
(1996)
Exp. Neurol.
138,
305-317[CrossRef][Medline]
[Order article via Infotrieve]
9.
Bajjalieh, S. M.,
Peterson, K.,
Shinghal, R.,
and Scheller, R. H.
(1992)
Science
257,
1271-1273 10.
Feany, M. B.,
Lee, S.,
Edwards, R. H.,
and Buckley, K. M.
(1992)
Cell
70,
861-867[CrossRef][Medline]
[Order article via Infotrieve]
11.
Gingrich, J. A.,
Anderson, P. H.,
Tiberi, M.,
Mestikawy, S. E,
Jorgensen, P. N.,
Fremeau, R. T.,
and Caron, M. G.
(1992)
FEBS Lett.
312,
115-122[CrossRef][Medline]
[Order article via Infotrieve]
12.
Bindra, P. S.,
Knowles, R.,
and Buckley, K. M.
(1993)
Gene (Amst.)
137,
299-302[CrossRef][Medline]
[Order article via Infotrieve]
13.
Bajjalieh, S. M.,
Peterson, K.,
Linial, M.,
and Scheller, R. H.
(1993)
Proc. Natl. Acad. Sci. U. S. A.
90,
2150-2154 14.
Janz, R.,
Hofmann, K.,
and Sudhof, T. C.
(1998)
J. Neurosci.
18,
9269-9281 15.
Buckley, K.,
Schweitzer, E.,
Miljanich, G. P.,
Clift-O'Grady, L.,
Kushner, P. D.,
Reichardt, L. F.,
and Kelly, R. B.
(1983)
Proc. Natl. Acad. Sci. U. S. A.
80,
7342-7346 16.
Sanes, J. R.
(1983)
Annu. Rev. Physiol.
45,
581-600[CrossRef][Medline]
[Order article via Infotrieve]
17.
Sanes, J. R.,
Marshall, L. M.,
and McMahan, U. J.
(1978)
J. Cell Biol.
78,
176-198 18.
Glicksman, M. A.,
and Sanes, J. R.
(1983)
J. Neurocytol.
12,
661-671[CrossRef][Medline]
[Order article via Infotrieve]
19.
Hall, Z. W.,
and Sanes, J. S.
(1993)
Cell
72 (suppl.),
99-121
20.
Timpl, R.,
and Brown, J. C.
(1994)
Matrix Biol.
14,
275-281[CrossRef][Medline]
[Order article via Infotrieve]
21.
Miner, J. H.,
Lewis, R. M.,
and Sanes, J. R.
(1995)
J. Biol. Chem.
270,
28523-28526 22.
Miner, J. H.,
Patton, B. L.,
Lentz, S. I.,
Gilbert, D. J.,
Snider, W. D.,
Jenkins, N. A.,
Copeland, N. G.,
and Sanes, J. R.
(1997)
J. Cell Biol.
137,
685-701 23.
Koch, M.,
Olson, P. F.,
Albus, A.,
Jin, W.,
Hunter, D. D.,
Brunken, W. J.,
Burgeson, R. E.,
and Champliaud, M. F.
(1999)
J. Cell Biol.
145,
605-618 24.
Iivanainen, A.,
Morita, T.,
and Tryggvason, K.
(1999)
J. Biol. Chem.
274,
14107-14111 25.
Sanes, J. R.,
Engvall, E.,
Butkowski, R.,
and Hunter, D. D.
(1990)
J. Cell Biol.
111,
1685-1699 26.
Patton, B. L.,
Miner, J. H.,
Chiu, A. Y.,
and Sanes, J. R.
(1997)
J. Cell Biol.
139,
1507-1521 27.
Noakes, P. G.,
Gautam, M.,
Mudd, J.,
Sanes, J. R.,
and Merlie, J. P.
(1995)
Nature
374,
258-262[CrossRef][Medline]
[Order article via Infotrieve]
28.
Patton, B. L,
Chiu, A. Y.,
and Sanes, J. R.
(1998)
Nature
393,
698-701[CrossRef][Medline]
[Order article via Infotrieve]
29.
Hockfield, S.,
Carlson, S. S.,
Levitt, P.,
Evans, C.,
and Silberstein, L.
(1993)
Molecular Probes of the Nervous System
, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY
30.
Cowan, D.,
Linial, M.,
and Scheller, R. H.
(1990)
Brain Res.
509,
1-7[CrossRef][Medline]
[Order article via Infotrieve]
31.
Wendland, B.,
Miller, K. G.,
Schilling, J.,
and Scheller, R. H.
(1991)
Neuron
6,
993-1007[CrossRef][Medline]
[Order article via Infotrieve]
32.
Striessnig, J.,
Glossmann, H.,
and Catterall, W.
(1990)
Proc. Natl. Acad. Sci. U. S. A.
87,
9108-9112 33.
Lim, W. A.,
Sauer, R. T.,
and Lander, A. D.
(1991)
Methods Enzymol.
208,
196-210[Medline]
[Order article via Infotrieve]
34.
Miljanich, G.,
Braisier, A. R.,
and Kelly, R. B.
(1982)
J. Cell Biol.
94,
88-96 35.
Yeager, R. E.,
Yoshikami, D.,
Rivier, J.,
Cruz, L. J.,
and Miljanich, G. P.
(1987)
J. Neurosci.
7,
2390-2396[Abstract]
36.
Laemlli, U. K.
(1971)
Nature
227,
680-685
37.
Neubig, R. R.,
Krodel, E. K.,
Boyd, N. D.,
and Cohen, J. B.
(1979)
Proc. Natl. Acad. Sci. U. S. A.
76,
690-694 38.
Carlson, S. S.,
Wagner, J. A.,
and Kelly, R. B.
(1978)
Biochemistry
17,
1188-1199[CrossRef][Medline]
[Order article via Infotrieve]
39.
Iwata, M.,
and Carlson, S. S.
(1991)
J. Biol. Chem.
266,
323-333 40.
Lee, M. K.,
and Lander, A.
(1991)
Proc. Natl. Acad. Sci. U. S. A.
88,
2768-2772 41.
Beck, K.,
Hunter, I.,
and Engel, J.
(1990)
FASEB J.
4,
148-160[Abstract]
42.
Engel, J.,
and Furthmayr, H.
(1987)
Methods Enzymol.
145,
3-78[Medline]
[Order article via Infotrieve]
43.
Carlson, S. S.
(1996)
Perspect. Dev. Neurobiol.
3,
373-386[Medline]
[Order article via Infotrieve]
44.
Jalkanen, M.,
Nguyen, H.,
Rapraeger, A.,
Kurn, A.,
and Bernfield, M.
(1985)
J. Cell Biol.
101,
976-984 45.
Horne, W. A.,
Ellinor, P. T.,
Inman, I.,
Zhou, M.,
Tsien, R. W.,
and Schwarz, T. L.
(1993)
Proc. Natl. Acad. Sci. U. S. A.
90,
3787-3791 46.
Iwata, M.,
and Carlson, S. S.
(1993)
J. Neurosci.
13,
195-207[Abstract]
47.
Goldfinger, L. E.,
Stack, M. S.,
and Jones, J. C. R.
(1998)
J. Cell Biol.
141,
255-265 48.
San Antonio, J. D.,
Slover, J.,
Lawler, J.,
Karnovsky, M. J.,
and Lander, A. D.
(1993)
Biochemistry
32,
4746-4755[CrossRef][Medline]
[Order article via Infotrieve]
49.
Wight, T. N.,
Heinegard, D. K.,
and Hascall, V. C.
(1991)
in
Cell Biology of Extracellular Matrix
(Hay, E. D., ed), 2nd. Ed.
, pp. 45-79, Plenum Publishing Corp., New York
50.
Brown, G. M.,
Huckerby, T. N.,
Morris, H. G.,
Abram, B. L.,
and Nieduszynski, I. A.
(1994)
Biochemistry
33,
4836-4846[CrossRef][Medline]
[Order article via Infotrieve]
51.
Mehmet, H.,
Scudder, P.,
Tang, P. W.,
Hounsell, E. F.,
Caterson, B.,
and Feizi, T.
(1986)
Eur. J. Biochem.
157,
385-391[Medline]
[Order article via Infotrieve]
52.
Schivell, A. E.,
Batchelor, R. H.,
and Bajjalieh, S. M.
(1996)
J. Biol. Chem.
271,
27770-27775 53.
Bennett, M. K.,
Calakos, N.,
Kreiner, T.,
and Scheller, R. H.
(1992)
J. Cell Biol.
116,
761-775 54.
Alberts, B.,
Bray, D.,
Lewis, J.,
Raff, M.,
Roberts, K.,
and Watson, J. D.
(1994)
Molecular Biology of the Cell
, 3rd Ed.
, p. 1213, Garland Publishing Inc., New York
55.
Nguyen, Q. T.,
and Lichtman, J. W.
(1996)
Curr. Opin. Neurobiol.
6,
104-112[CrossRef][Medline]
[Order article via Infotrieve]
Copyright © 2000 by The American Society for Biochemistry and Molecular Biology, Inc.
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  K. L. Custer, N. S. Austin, J. M. Sullivan, and S. M. Bajjalieh Synaptic Vesicle Protein 2 Enhances Release Probability at Quiescent Synapses J. Neurosci., January 25, 2006; 26(4): 1303 - 1313. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Schwander, R. Shirasaki, S. L. Pfaff, and U. Muller {beta}1 Integrins in Muscle, But Not in Motor Neurons, Are Required for Skeletal Muscle Innervation J. Neurosci., September 15, 2004; 24(37): 8181 - 8191. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Teng and R. S Wilkinson 'Delayed' endocytosis is regulated by extracellular Ca2+ in snake motor boutons J. Physiol., August 15, 2003; 551(1): 103 - 114. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. R. Sanes The Basement Membrane/Basal Lamina of Skeletal Muscle J. Biol. Chem., April 4, 2003; 278(15): 12601 - 12604. [Full Text] [PDF]
|
|
|
|
 |
|
 P. Lichtlen, Y. Wang, T. Belser, O. Georgiev, U. Certa, R. Sack, and W. Schaffner Target gene search for the metal-responsive transcription factor MTF-1 Nucleic Acids Res., April 1, 2001; 29(7): 1514 - 1523. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. K. Nielsen and Y. Yamada Identification of Cell-binding Sites on the Laminin alpha 5 N-terminal Domain by Site-directed Mutagenesis J. Biol. Chem., March 30, 2001; 276(14): 10906 - 10912. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| All ASBMB Journals | Molecular and Cellular Proteomics |
| Journal of Lipid Research | ASBMB Today |