Received: 10 September 2018 Revised: 5 January 2019 Accepted: 1 February 2019

DOI: 10.1002/cne.24661

RESEARCH ARTICLE

Seasonal comparison of the neuromuscular junction

morphology of Bufo marinus
Dengyun Ge1 | Peter G. Noakes1,2 | Nickolas A. Lavidis1

1
Faculty of Medicine, School of Biomedical
Sciences, The University of Queensland, Abstract
Queensland, Australia At mammalian neuromuscular junctions (NMJs), prolonged inactivity leads to muscle denervation
2
Queensland Brain Institute, The University of and atrophy. By contrast, amphibian NMJs do not show such degeneration even though they can
Queensland, Queensland, Australia
remain in a state of drought-imposed dormancy (hibernation) for many years. We have previously
Correspondence
reported that during the dry season, toad (Bufo marinus) NMJs display decreased sensitivity to
Nickolas A. Lavidis, Faculty of Medicine,
School of Biomedical Sciences, The University extracellular calcium-dependent neurotransmitter release, which leads to minimal neuromuscular
of Queensland, Queensland 4072, Australia. transmission. In the present study, we examined and compared NMJ morphology of toads
Email: [email protected] obtained from the wild during the wet season (February–March) when these toads are active, to
Funding information toads obtained from dry season (October–November) when toads are inactive. Iliofibularis mus-
University of Queensland; China Scholarship
cles were isolated and prepared for immunostaining with anti-SV2, a monoclonal antibody that
Council
labels synaptic vesicle glycoprotein SV2. The corresponding postsynaptic acetylcholine receptors
were stained using Alexa Fluro-555 conjugated α-bungarotoxin. Confocal microscopy and three-
dimensional reconstructions were then used to examine the pre-and postsynaptic morphology of
toads NMJs from the dry (inactive) and wet (active) seasons. Total axon branch number, the per-
centage of axon branches with discontinuous distributions of synaptic vesicles, and further the
Pearson value of colocalization of pre and postsynaptic elements in each NMJs from both the dry
and wet season were compared. While our previous studies on dry toads revealed a significant
reduction in evoked neurotransmission, our present findings show that the structure of the NMJs
suffered limited level of remodeling, suggesting a mechanism utilized by NMJs in dry season toads
to support quick recover from their dormant state after the heavy rain in wet season.

KEYWORDS

neuromuscular junction, neurotransmission, RRID:AB_2315387, RRID:AB_2617152, season,

synapse, synaptic remodeling

1 | I N T RO D UC T I O N decreased neuronal activity could provide a period of nerve terminal

elimination, retraction to downscale, refine, consolidate, or strengthen
Synapses of the nervous system undergo continual remodeling to meet particular synapses and circuits (Nishiyama & Yasuda, 2015; Tononi &
their functional role in response to a variety of physiological conditions Cirelli, 2014). This, however, is an oversimplification as for example
(Engert & Bonhoeffer, 1999; Krause & Wernig, 1985; Leung & Wong, increased excitation in the amygdala promotes synaptogenesis while at
2017; Macleod, Dickens, & Bennett, 2001; Wernig, Anzil, & Bieser, the same time increased excitation in the hippocampus induces a
1981), for review see (Kuner & Flor, 2016; Skaper, Facci, Zusso, & reduction in synapses (Lupien, McEwen, Gunnar, & Heim, 2009). The
Giusti, 2017; Wernig & Herrera, 1986). Hence, structural remodeling is underlying mechanisms that regulate synaptic structure and function
widely viewed as the basis of functional synaptic plasticity. It was sug- are still poorly understood. Amphibian neuromuscular junctions (NMJs)
gested that synaptic remodeling generally exhibits a cyclic change in provide a good model synapse for this study for the following reasons.
both central and peripheral nervous system (Hebb, 1949). More specifi- First, the efficacy of neurotransmission in the amphibian NMJs
cally, active neurotransmission leads to nerve terminal growth, and undergoes seasonal change, with the probability of quantal

J Comp Neurol. 2019;527:1931–1939. wileyonlinelibrary.com/journal/cne © 2019 Wiley Periodicals, Inc. 1931

1932 GE ET AL.

neurotransmitter release being higher in the wet season when 2 | M A T E R I A L S A N D M ET H O D S

compared with the dry season (Bennett & Lavidis, 1991; Ge & Lavidis,
2018; Lavidis, Hudson, Choy, Lehnert, & Franklin, 2008; Pawson & 2.1 | Animals
Grinnell, 1989; Singh, 1964). Second, its structure is relatively simple
Mature cane toads (Bufo marinus) were obtained from the wild in
and widely accessible (Dorlochter, Meurer, & Wernig, 1993; Jung,
South East Queensland, Australia, during the wet season (February–
Szule, Stouder, Marshall, & McMahan, 2018; Meriney & Dittrich, 2013;
March) and dry season (October–November), and then maintained
Slater, 2015; Szule, Jung, & McMahan, 2015).
overnight in a large humidified tank with a 12 hr light–dark cycle at
As quantal size (miniature endplate potential amplitude, MEPP)
between 22 and 25  C. During the wet season, active toads can be
remains relatively constant in both dry and wet seasons (Bennett &
easily obtained along roadsides. In the late stage of the dry season
Lavidis, 1991; Ge & Lavidis, 2018; Hudson, Lavidis, Choy, & Franklin,
(October–November), toads become sluggish and could only be found
2005), which indicates the maintenance of postsynaptic structure, it is
under the logs and piles of vegetation. Compared to the wet season
suggested that synaptic remodeling is largely presynaptic (Pawson &
toads, which would escape quickly when people approach them, the
Grinnell, 1989). Previous studies utilized many different histological
dry season toads remain sluggish even when they were handled. In
methods to examine the pre and postsynaptic structure of the NMJs.
the iliofibularis muscle, quantal neurotransmitter release from the pre-
A common method used was silver staining of filaments within motor
synaptic nerve terminals was significantly decreased during the dry
nerve terminals and fluorescence conjugated α-bungarotoxin (α-BTX)
staining of the corresponding postsynaptic acetylcholine receptors
(AChRs) at the NMJs to examine morphological changes (Jans,
Salzmann, & Wernig, 1986; Krause & Wernig, 1985; Wernig et al.,
1981; Wernig, Pecotdechavassine, & Stover, 1980). The results indi-
cated that synaptic remodeling in the amphibian NMJs mostly involve
the distal end of nerve terminals, characterized by longitudinal misalign-
ment of the pre and postsynaptic elements. However, it is not clear
why the continuous long nerve terminals are sometimes opposed to
discontinuous AChR clusters in the middle part of terminal branches,
see figures in (Bennett, Lavidis, & Armson, 1987; Jans et al., 1986;
Krause & Wernig, 1985). One possibility is that highly intermittent
quantal neurotransmitter release at sections of the nerve terminal are
associated with reduced vesicle density which in turn are associated
with reductions in AChR clustering in the corresponding postsynaptic
sites. This possibility was examined in the present study by comparing
the clustering of AChRs under the nerve terminal during the wet and
dry seasons; we would expect greater intermittent clustering during the
dry season.
In the present study, presynaptic nerve terminals from different
NMJs were located with an antibody against synaptic vesicle mem-
brane glycoproteins (synaptic vesicle 2, SV2) and the postsynaptic
AChRs were labeled with Alexa Fluro-555 conjugated α-BTX. To
examine whether the decreased quantal neurotransmitter release in
the dry season induces NMJs remodeling, 114 NMJs from three dry
season animals were compared to 111 NMJs from three wet season
FIGURE 1 Examples of the short neuromuscular junctions-
animals. NMJs were sorted into three sized groups according to their
neuromuscular junctions (PTB-NMJs) in the dry (upper panel, a–c) and
longest primary terminal branch (PTB). This assessment allowed us the wet season (lower panel, d–f) amphibian iliofibularis muscle. Dry
to classify the NMJs into three categories: (a) short NMJs season nerve terminals were labeled with anti-SV2 monoclonal
(PTB < 50 μm); (b) median NMJs (50 ≤ PTB < 100 μm); and (c) long antibody (a). Dry season short PTB-NMJs postsynaptic acetylcholine
receptors (AChRs) were labeled with Alexa Fluro-555 conjugated
NMJs (PTB ≥ 100 μm). The size categories chosen allowed us to
α-bungarotoxin (α-BTX) (b). (c) Merged images from (a) and (b). Wet
compare the present data with previous publications (Bennet & Lavi- season nerve terminals were labeled with anti-SV2 monoclonal
dis, 1982; Bennett, Jones, & Lavidis, 1986; Davey & Bennett, 1982). antibody (d). Wet season AChRs were labeled with Alexa Fluro-555
For each of these NMJs, we compared the total nerve branch num- conjugated α-BTX (e). (f) Merged images from (d) and (e). Arrows in
ber, the percentage of nerve branches with discontinuous synaptic (c) and (f) indicate the first point of axon contact with the muscle
fiber. Lightning in (c) and (f) indicate the discontinuous distribution of
vesicle and AChRs distributions, and the Pearson value of colocaliza-
the presynaptic and postsynaptic elements along the nerve terminal.
tion of these two synaptic elements in NMJs from both the dry and Scale bar in c and f = 5 μm and is the scale for a–b and d–e,
wet seasons. respectively
GE ET AL. 1933

FIGURE 2 Dry season short neuromuscular junctions (NMJs) showed greater levels of presynaptic and postsynaptic colocalization and
discontinuity compared to wet season short NMJs, despite no overall change in NMJ size or primary terminal branch (PTB) number.
Quantification of the Pearson value of colocalization (a), number of nerve branches per NMJs (b), and percentage of discontinuously distributed
SVs along the nerve terminals (c) in dry and wet season short PTB-NMJs

season and the results were reported in our previous publications presynaptic nerve terminal component (synaptic vesicle glycoprotein
(Ge & Lavidis, 2017, 2018). Toads weighing between 30 and 40 g 2A, SV2) and the corresponding postsynaptic AChRs. SV2 was immuno-
(50–60 mm in length) were used. Animals were sacrificed by double localized with antibody with respect to postsynaptic AChRs labeled
pithing according to procedures approved by the Animal Ethics Com- with Alexa Fluor-555-conjugated α-bungarotoxin (α-BTX) at the NMJ.
mittee of The University of Queensland. Small bundles were blocked in PBS containing 2% bovine serum albu-
min (BSA) with 0.1–0.2% Triton X-100 (TX-100) for 30 min at room
2.2 | Isolation of nerve-muscle tissues temperature. This was followed by incubation with mouse anti-SV2
developed by Buckley (Buckley & Kelly, 1985) obtained from the Devel-
The iliofibularis muscles from wet (n = 3) and dry (n = 3) season col-
opmental Studies Hybridoma Bank, created by the National Institute of
lected animals were dissected, pinned on the bed of cured Sylgard,
Child Health and Human Development of the National Institutes of
stretched to approximately 110% of their resting length, and immersed
Health, and maintained at The University of Iowa, Department of Biol-
fixed with 4% paraformaldehyde in 0.1 M phosphate-buffered saline
ogy, Iowa City. IA 52242 (1:100 dilution) in PBS containing 2% BSA
(PBS), pH 7.4, at room temperature for 30 min. The fixed muscles were
then washed 3 × 10 min with 0.1% glycine in PBS. The muscles were and 0.1% TX-100 at 1:100 for 1 hr at room temperature. Small muscle

subsequently teased into small bundles for whole mount immunostain- bundles were then washed 3× for 5 min with PBS before incubation

ing of their NMJs. with appropriate Alexa Fluor 488-conjugated anti-mouse secondary
antibody (Molecular Probes, Invitrogen, Eugene OR, 1:500 dilution in
PBS) in combination with Alexa Fluor 555-conjugated α-BTX
2.3 | Immunohistochemistry
(Molecular Probes, Invitrogen, 1:250 dilution in PBS) at room tempera-
To avoid the possible clustering of NMJ data within a single animal ture for 2 hr. Sections were then washed with PBS and coverslip
from either wet or dry season, paraformaldehyde (PFA)-fixed iliofibu- mounted in Prolong Gold anti-fade reagent (Molecular Probes,
laris muscles from both wet and dry season toads were cut into small Invitrogen).
bundles (each bundle containing 3–5 muscle fibers) and 10 bundles
from each toad were randomly selected for immunostaining to label the
2.4 | Confocal microscopy
NMJs. Furthermore, a box-plot was plotted to examine the distribution
of nerve terminal branch numbers from each toad from both wet and Confocal image acquisitions were performed using a Leica DMi8 SP8
dry season (see Supporting Information Figure S2). The structure of inverted confocal laser scanning microscope (Leica Microsystems,
NMJs was examined after immunofluorescence double labeling of the Wetzlar, Germany) equipped with excitation laser wavelengths of
1934 GE ET AL.

488, 543, and 633. Sequential scanning of the different channels was
performed to limit fluorophore spectral crosstalk. Images were sam-
pled at a resolution of 1,024 by 1,024 pixels, using a ×63 glycerol
objective (NA1.4), a 2.5 times zoom, and a z-step size of 0.51 μm.
Saved images were reconstructed and analyzed using the Imaris 8.1.2
software (Bitplane, South Windsor, CT). The laser power, scanning
speed, photomultiplier gain levels, and pin hole size were maintained
at the same level among scanning of the different immunostained
whole mount muscle preparations.

2.5 | Imaris station for data analysis

Imaris 8.1.2 (Bitplane, South Windsor, CT) was used to analyze the
structure of nerve terminals and the corresponding postsynaptic struc-
ture as has been previously described (Caillol et al., 2012). The images
of Alexa Fluor-488 labeled nerve terminal and the corresponding Alexa
Fluor-555 stained AChRs clusters were highlighted by adjusting the sin-
gle channel brightness and contrast levels for visualization purposes.
The level of colocalization between the nerve terminal
(SV2-fluorescent) and the postsynaptic AChR-fluorescence were ana-
lyzed using Imaris' colocalization tool, and the Pearson colocalization
value for the volume for each reconstructed three-dimensional NMJs
was determined (Dunn, Kamocka, & McDonald, 2011). Briefly, the
Pearson's correlation coefficient is a quantitative measurement that
estimates the degree of overlap between fluorescence signals obtained
in two channels, the mathematical basis of this quantification can be
found in (Dunn et al., 2011; Zinchuk, Wu, & Grossenbacher-Zinchuk,
2013; Zinchuk, Zinchuk, & Okada, 2007). A Pearson colocalization value
of “1” indicates absolute colocalization and “0” indicates totally no colo-
calization. This method was previously used to quantify the extent of
colocalization of two fluorophore-labeled proteins within the same cell
(Keller et al., 2013) also at the NMJs (Caillol et al., 2012). For the pur-
pose of helping the readers to understand the application of this FIGURE 3 Examples of the median primary terminal branch-
neuromuscular junctions in the dry (upper panel, a–c) and the wet
method in the present study, we describe this quantification process
season (lower panel, d–f) toad iliofibularis muscle. Dry season nerve
with a scatterplot (see Supporting Information Figure S1). In this scat- terminals labeled with anti-SV2 monoclonal antibody (a). Dry season
terplot, suppose there are five pixels (A–E) in an eight bit digital image. postsynaptic acetylcholine receptors (AChRs) labeled with Alexa Fluro-
The intensity of red florescence is plotted against the intensity of the 555 conjugated α-bungarotoxin (α-BTX) (b). (c) Merged images from
green florescence for each pixel. Then the red and green florescence (a) and (b). Wet season nerve terminal labeled with anti-SV2
monoclonal antibody (d). Wet season postsynaptic AChRs labeled with
light density value of each pixel can be represented by its coordinate in
Alexa Fluro-555 conjugated α-BTX (e). (f) Merged images from (d) and
the axes. As the first step to quantify the colocalization of red and (e). Arrows in (c) and (f) indicate the first point of axon contact with the
green florescence in this image, a threshold level of light density for muscle fiber. Lightning in (c) and (f) indicate the discontinuous
each color was set to exclude the noise and background (as shown in distribution of the presynaptic and postsynaptic elements along the
the blue square inlet). Under this condition, even some pixels like A and nerve terminal. Scale bar in c and f = 10 μm and is the scale for a–b and
d–e, respectively
E occupy single high density of green or red florescence but very low
density of the other would be excluded. Then, the overall colocalization
The length of primary nerve branch from all the NMJs was analyzed
value of this image (B, C, and D) would be quantified with Imaris' colo-
with Image J (National Institutes of Health). The total nerve branch
calization tool. Images from single channel and merged channels were
number and the number of nerve terminals with discontinuous pre
saved as two-dimensional “Tif” format for the analysis of the number of
and postsynaptic elements in each NMJs were counted. Student's
terminal nerve branches and count for the discontinuously distributed
unpaired two-tailed t-tests were performed to compare the total
pre and postsynaptic elements.
nerve branch number, the Pearson value of colocalization and the per-
centage of nerve terminals with discontinuously distributed pre and
2.6 | Statistical analysis postsynaptic elements in each NMJs between wet and dry season
Criteria for analyzing NMJs: (a) NMJs were wholly on the surface of NMJs were also analyzed. Results are expressed as means ± SEM with
muscle fiber; (b) NMJs were positively stained for both antibodies. statistical significance accepted at p < 0.05.
GE ET AL. 1935

FIGURE 4 No difference in presynaptic and postsynaptic colocalization and continuity at median-neuromuscular junctions (NMJs) from wet and
dry season toads, but the number of terminal branches is increased at median NMJs from wet season toads. Quantification of the Pearson value
of colocalization (a), number of nerve branches per NMJs (b), and percentage of discontinuously distributed SVs along the nerve terminals (c) in
dry and wet season median primary terminal branch-NMJs

3 | RESULTS 3.2 | Median PTB-NMJs from dry season toads

showed decreased number of nerve terminal branches
3.1 | Short PTB-NMJs from dry season animals per NMJ
demonstrate an increase in the colocalization and From the 114 NMJs in dry season, 55 median PTB-NMJs were
discontinuity in pre and postsynaptic elements, when observed (48.2%). From the 111 NMJs in wet season, 48 median
compared to short PTB-NMJs from wet season PTB-NMJs were observed (43.2%), suggesting marginally higher per-
animals centage of this type of NMJs in dry (inactive) than wet (active) season
toads. Examples of median PTB-NMJs are shown in Figure 3 from dry
Among the 114 NMJs in the dry season examined, there were
season (upper panel) and wet season (lower panel). Pearson value of
47 short PTB-NMJs (41.2%). From the 111 NMJs in the wet season
colocalization did not show significant seasonal difference in the
examined, there were 54 short PTB-NMJs (48.6%), indicating slightly
median PTB-NMJs (Figure 4a). However, significantly (p < 0.05) lower
lower percentage of short NMJs in dry season (inactive) than wet
number of nerve terminal branches was seen in the dry season. There
season (active) states. We observed no significant difference in the
was a significant (p < 0.05) decrease in the percentage of nerve termi-
number of terminal branches of NMJs between wet and dry season
nals with discontinuously distributed presynaptic and postsynaptic
toads (Figures 1 and 2b). However, when we examined the distribu-
elements in the wet season NMJs when compared with the dry sea-
tion of presynaptic vesicles and postsynaptic AChRs, we did observe
son amphibian NMJs (Figure 4c).
significant differences between short NMJs isolated from wet and
dry season toads (e.g., Figure 1). The Pearson value of colocalization
3.3 | Long PTB-NMJs from dry season toads
(Figure 2a) and the percentage of nerve terminals with discontinu-
demonstrate a significant decrease in the
ously distributed presynaptic and postsynaptic elements (Figure 2c)
colocalization of SV2 and AChRs and an increase in
indicate significant differences (p < 0.05) in dry season when com-
pared with wet season. NMJs from wet season toads had more con-
discontinuity in pre and postsynaptic elements in the
tinuous distributions of synaptic vesicles along their nerve terminals,
dry season
and this distribution was also seen for the corresponding postsynap- From the 114 NMJs examined in the dry season, 12 long PTB-NMJs
tic AChRs (e.g., see Figure 1). were observed (10.5%). From the 111 NMJs examined in the wet
1936 GE ET AL.

fragmented in their distribution of synaptic vesicles along nerve termi-

nal branches compared to the NMJs from wet season toads (Figure 5).
This was supported by a lower Pearson value of colocalization of syn-
aptic vesicles to postsynaptic AChRs, and a higher level of pre and
postsynaptic discontinuity for long NMJs from dry season toads com-
pared with long NMJs from wet season toads (Figure 6a, c, respec-
tively). The number of nerve terminal branches at these long NMJ did
not differ between wet and dry seasons (Figure 6b).

4 | DI SCU SSION

In the present study, we examined the distributions and colocaliza-

tions of presynaptic vesicles and postsynaptic AChRs, along with the
number of nerve terminal branches in three morphologically classified
NMJs (short, median, and long). While we did not see any major
changes in the proportion of these NMJ classifications between wet
and dry season animals, we did observe several subtle changes such
as presynaptic and postsynaptic colocalizations, synaptic vesicle conti-
nuity, and terminal branch number between the dry and wet season
toads. Specifically, there were three main findings: first, dry season
induced a significant decrease in the Pearson value of colocalization in
long PTB-NMJs, by contrast short PTB-NMJs showed an increase.
Second, dry season gave rise to a decrease in the number of nerve ter-
minal branches in median PTB-NMJs. Third, for the three classified
NMJs in the dry season, there was a consistent trend for nerve termi-
nal branches to show a higher level of discontinuity in the distribution
of synaptic vesicles along all terminal branches.

4.1 | Seasonal change in the number of nerve

terminal branches
Although it has been shown that long terminal branches release more
neurotransmitter than smaller nerve terminal branches during low fre-
quency nerve stimulation (Kuno, Turkanis, & Weakly, 1971), longer
terminal branches release less neurotransmitter per unit length of
FIGURE 5 Long neuromuscular junctions (NMJs) from dry season nerve terminal than shorter ones (Nudell & Grinnell, 1982). These
toads appear more fragmented compared to NMJs from wet season observations are consistent with a proximal to distal decrease in the
toads. Examples of long primary terminal branch-NMJs in the dry
probability of neurotransmitter release (Bennett & Lavidis, 1979). Pre-
(upper panel, a–c) and the wet season (lower panel, d–f) toad
iliofibularis muscle. Dry season nerve terminal labeled with anti-SV2 vious studies have demonstrated that the probability of neurotrans-
monoclonal antibody (a). Dry season postsynaptic acetylcholine mitter release is easily regulated by seasonal factor(s) (Bennett &
receptors (AChRs) labeled with Alexa Fluro-555 conjugated Lavidis, 1991; Ge & Lavidis, 2017, 2018; Hudson et al., 2005; Lavidis
α-bungarotoxin (α-BTX) (b). (c) Merged images from (a) and (b). Wet et al., 2008). For example, during the dry season, the probability of
season nerve terminal labeled with anti-SV2 monoclonal antibody (d).
neurotransmitter release decreases most notably at the release sites
Wet season postsynaptic AChRs labeled with Alexa Fluro-555
conjugated α-BTX (e). (f) Merged images from (d) and (e). Arrows in that have the highest activity during the wet season, this results in the
(c) and (f) show the first point of axon contact with the muscle fiber. probability of neurotransmitter release becoming more uniformly low
Lightning in (c) and (f) indicate the discontinuous distribution of the during the dry season (Ge & Lavidis, 2017; Lavidis et al., 2008). The
presynaptic and postsynaptic elements along the nerve terminal. Scale reduction in neurotransmitter release during the dry season results in
bar in c and f = 20 μm and is the scale for a, b and d, e, respectively
subthreshold neurotransmission from nerve terminal to muscle. This,
along with spontaneous neurotransmitter release, maintains a basal
season, nine long PTB-NMJs were observed (8.1%), suggesting no level of synaptic communication between nerve terminal and muscle,
apparent change in the percentage of these type of NMJs between which may contribute to limiting the level of morphological changes
dry and wet season toads. Examples of long PTB-NMJs are shown in associated with seasonal dormancy. As a decrease in the number of
Figure 5, representative long PTB-NMJs from dry season (upper terminal branches was evident in the median-sized terminal branches
panel) and wet season (lower panel). Dry season NMJs appeared more and not in the short or long branches, it indicates minor retraction is
GE ET AL. 1937

FIGURE 6 Long neuromuscular junctions (NMJs) from wet season toads demonstrate increased colocalization of synaptic vesicles and
postsynaptic acetylcholine receptors along with a lower level of discontinuity, but no change in the number of terminal nerve branches.
Quantification of the Pearson value of colocalization (a), number of nerve branches per NMJs (b), and percentage of discontinuous distributed
SVs along the nerve terminals (c) in dry and wet season long primary terminal branch-NMJs

more pronounced than sprouting of branches in median-sized terminal pronounced in dry season nerve terminals. The absence of synaptic
branches during dormancy. vesicles in different regions along the terminal branches supports the
idea that the probability of neurotransmitter release is very low in

4.2 | Discontinuous distribution of the synaptic some regions. The deficiency in SV2 staining in some regions of termi-
nals was also associated with reduced levels of AChR labeling. What
vesicles and the corresponding postsynaptic AChRs
causes this regionalization of synaptic vesicles along the terminal
along terminal branches
branch was not investigated, although several studies of amphibian
Previous studies using silver stained amphibian nerve terminals exhibit NMJ reveal that terminal Schwann cell processes that invade the syn-
signs of sprouting and retraction mostly in the distal end of the nerve aptic cleft between the presynaptic and postsynaptic membranes may
terminal (Jans et al., 1986; Krause & Wernig, 1985; Wernig et al., be involved (Bennett et al., 1987; Werle, Jones, & Stanco, 1999).
1980). These observations indicated that the neurofibrils (cytoskeletal Discontinuous distribution of SV2 and AChR staining was evident
and associated elements detected by silver staining) remain continu- in the three different size classes of NMJs examined and increased in
ous along the length of the nerve terminals during prolonged inhibi- frequency during the dry season. These observations are consistent
tion. A pioneering investigation utilizing a combination of electron with reported nonuniform neurotransmitter release probability along
microscopy and functional labeling of SVs with FM2-dye indicated terminal branches. Currently, the cause of this synaptic vesicle rear-
that the average density of synaptic vesicles was constant along the rangement remains unclear. It is possible that a reduction in turnover
nerve branches, while small but significantly higher vesicle densities of vesicles and production of vesicles occurs during the dry season to
occurred active zones compared with nonactive zone regions (Everett, conserve limited energy resources. Previous studies have demon-
Edwards, & Etherington, 2002). FM2-dye loading, however, may be a strated that the complete inhibition of neurotransmission leads to
confounding factor with respect to vesicle distribution along the major disruption in neuromuscular communication and major dysfunc-
length of terminal branches, possibly resulting in the elimination of SV tional disruptions (Sanes & Lichtman, 1999; Witzemann, Chevessier,
discontinuities. In the present study, the presynaptic element at the Pacifici, & Yampolsky, 2013). It is likely that the amphibians have
NMJs was labeled with the monoclonal antibody that specifically developed a strategy, which preserves the structure of the NMJs
binds to the membrane glycoproteins in synaptic vesicles (SV2). Immu- while awaiting fast reactivation following the start of the wet season.
nostaining for SV2 however demonstrated a nonuniform distribution Regulation of synaptic vesicle availability and activation of voltage
of vesicles along the length of terminal branches, which was more gated calcium channels (VGCCs) are two effective mechanisms that
1938 GE ET AL.

are possibly utilized to regulate neurotransmitter release and subse- RE FE RE NC ES

quent use of limited resources. We have previously shown that a pos- Bennet, M. R., & Lavidis, N. A. (1982). Variation in Quantal secretion at dif-
sible regulatory opiate may be responsible for reducing evoked ferent release sites along developing and mature motor terminal
branches. Developmental Brain Research, 5(1), 1–9.
neurotransmitter release while sustaining miniature neurotransmitter Bennett, M. R., Jones, P., & Lavidis, N. A. (1986). The probability of
release to refrain the NMJ from gross restructuring as has been seen quantal secretion along visualized terminal branches at amphibian
in NMJ inhibition with AChR noncompetitive antagonists like α-BTX (Bufo marinus) neuromuscular synapses. The Journal of Physiology,
379, 257–274.
(Ge & Lavidis, 2018; Lavidis, 1995a, 1995b). Bennett, M. R., & Lavidis, N. A. (1979). The effect of calcium ions on the
secretion of quanta evoked by an impulse at nerve terminal release
sites. The Journal of General Physiology, 74(4), 429–456.
4.3 | Changes in the Pearson value of colocalization Bennett, M. R., & Lavidis, N. A. (1991). Probabilistic secretion of quanta
indicate conformational alternation in the pre and from the release sites of nerve-terminals in amphibian muscle modu-
lated by seasonal-changes. Neuroscience Letters, 134(1), 79–82.
postsynaptic elements Bennett, M. R., Lavidis, N. A., & Armson, F. M. (1987). Changes in the
At the NMJs, efficient and fast synaptic neurotransmission relies on dimensions of release sites along terminal branches at amphibian neu-
romuscular synapses. Journal of Neurocytology, 16(2), 221–237.
the precise colocalization of pre and postsynaptic elements (Caillol Booth, F. W. (1982). Effect of limb immobilization on skeletal-muscle. Jour-
et al., 2012), such as the VGCCs, SVs, vesicular-associated proteins, nal of Applied Physiology, 52(5), 1113–1118.
and presynaptic proteins that align structures within the terminal Buckley, K., & Kelly, R. B. (1985). Identification of a transmembrane glyco-
protein specific for secretory vesicles of neural and endocrine cells.
(Harlow, Ress, Stoschek, Marshall, & McMahan, 2001; Meriney &
The Journal of Cell Biology, 100(4), 1284–1294.
Dittrich, 2013; Wachman, Poage, Stiles, Farkas, & Meriney, 2004) Caillol, G., Vacher, H., Musarella, M., Bellouze, S., Dargent, B., & Autillo-
and the AChRs with postsynaptic proteins that align and concentrate Touati, A. (2012). Motor endplate disease affects neuromuscular junction
maturation. The European Journal of Neuroscience, 36(4), 2400–2408.
the receptors at the lips of the postsynaptic folds (Ghazanfari et al.,
Davey, D. F., & Bennett, M. R. (1982). Variation in the size of synaptic con-
2018; Plomp, Huijbers, & Verschuuren, 2018; Sanes & Lichtman, tacts along developing and mature motor terminal branches. Brain
2001). In this study, two probes used for evaluating the colocaliza- Research, 281(1), 11–22.
Dorlochter, M., Meurer, S., & Wernig, A. (1993). Acetylcholine receptor
tion of pre and postsynaptic elements of NMJs were quantified. Our
bars and transmitter release in frog neuromuscular junctions. Neurosci-
present results showed changes in the ratio of SV2 to AChR fluores- ence, 52(4), 987–999.
cence between the wet and dry seasons indicating alternations in Dunn, K. W., Kamocka, M. M., & McDonald, J. H. (2011). A practical guide
colocalization during the dry season. These changes in SV2 to AChR to evaluating colocalization in biological microscopy. American Journal
of Physiology-Cell Physiology, 300(4), C723–C742.
fluorescence ratio is most likely associated with a depletion of syn-
Engert, F., & Bonhoeffer, T. (1999). Dendritic spine changes associated
aptic vesicles followed by a decrease in neurotransmitter release and with hippocampal long-term synaptic plasticity. Nature, 399(6731),
a subsequent reduction in AChR clustering. 66–70.
Everett, A. W., Edwards, S. J., & Etherington, S. J. (2002). Structural basis
Muscle immobilization in mammals causes structural disruption
for the spotted appearance of amphibian neuromuscular junctions
in NMJs with subsequent muscle atrophy which requires consider- stained for synaptic vesicles. Journal of Neurocytology, 31(1), 15–25.
able time and effort for rehabilitation (Booth, 1982). This does not Ge, D., & Lavidis, N. (2018). Climatic modulation of neurotransmitter
release in amphibian neuromuscular junctions: Role of dynorphin-A.
occur in amphibian NMJs by mechanisms that are not clearly under-
American Journal of Physiology. Regulatory, Integrative and Comparative
stood. As the duration of the wet season is relatively short in Physiology, 314(5), R716–R723.
Australia, survival of amphibians requires quick remobilization of syn- Ge, D., & Lavidis, N. A. (2017). Seasonal factors influence quantal transmit-
aptic neurotransmission at NMJs when the rain comes following pro- ter release and calcium dependence at amphibian neuromuscular junc-
tions. American Journal of Physiology. Regulatory, Integrative and
longed drought. It is, therefore, vital that the structural integrity of
Comparative Physiology, 313(5), R202–R210.
the NMJs and muscle be maintained during prolonged inactivity Ghazanfari, N., Trajanovska, S., Morsch, M., Liang, S. X., Reddel, S. W., &
imposed by drought. Our present study has demonstrated minor Phillips, W. D. (2018). The mouse passive-transfer model of MuSK
myasthenia gravis: Disrupted MuSK signaling causes synapse failure.
NMJ changes which most likely contribute to conservation of
Annals of the New York Academy of Sciences, 1412(1), 54–61.
resources, allowing the animal to survive when the drought is broken. Harlow, M. L., Ress, D., Stoschek, A., Marshall, R. M., & McMahan, U. J.
Exactly how the NMJ is highly in tune with environmental factors (2001). The architecture of active zone material at the frog's neuro-
muscular junction. Nature, 409(6819), 479–484.
and what neuroprotective mechanisms are operating are areas that
Hebb, D. (1949). The organization of behavior. Annals of Psychology, 51,
require further investigation. 493–494.
Hudson, N. J., Lavidis, N. A., Choy, P. T., & Franklin, C. E. (2005). Effect of
prolonged inactivity on skeletal motor nerve terminals during aestiva-
ACKNOWLEDGMENTS tion in the burrowing frog, Cyclorana alboguttata. Journal of Compara-
tive Physiology A, 191(4), 373–379.
The authors are grateful to Dr. Shaun Walters (SBMS) for technical assis- Jans, H., Salzmann, R., & Wernig, A. (1986). Sprouting and nerve retraction
tance in confocal imaging and Imaris working station. D. Ge received a in frog neuromuscular junction during ontogenesis and environmental
changes. Neuroscience, 18(3), 773–781.
postgraduate scholarship from the China Scholarship Council and The
Jung, J. H., Szule, J. A., Stouder, K., Marshall, R. M., & McMahan, U. J.
University of Queensland. (2018). Active zone material-directed orientation, docking, and fusion
of dense core vesicles alongside synaptic vesicles at neuromuscular
junctions. Frontiers in Neuroanatomy, 12, 72.
ORCID
Keller, K. E., Yang, Y. F., Sun, Y. Y., Sykes, R., Acott, T. S., & Wirtz, M. K. (2013).
Dengyun Ge https://orcid.org/0000-0003-0715-8612 Ankyrin repeat and suppressor of cytokine signaling box containing
GE ET AL. 1939

protein-10 is associated with ubiquitin-mediated degradation pathways in Skaper, S. D., Facci, L., Zusso, M., & Giusti, P. (2017). Synaptic plasticity,
trabecular meshwork cells. Molecular Vision, 19, 1639–1655. dementia and Alzheimer disease. CNS & Neurological Disorders Drug
Krause, M., & Wernig, A. (1985). The distribution of acetylcholine-receptors Targets, 16(3), 220–233.
in the Normal and Denervated neuromuscular-junction of the frog. Slater, C. R. (2015). The functional organization of motor nerve terminals.
Journal of Neurocytology, 14(5), 765–780. Progress in Neurobiology, 134, 55–103.
Kuner, R., & Flor, H. (2016). Structural plasticity and reorganisation in Szule, J. A., Jung, J. H., & McMahan, U. J. (2015). The structure and func-
chronic pain. Nature Reviews Neuroscience, 18(1), 20–30. tion of 'active zone material' at synapses. Philosophical Transanctions of
Kuno, M., Turkanis, S. A., & Weakly, J. N. (1971). Correlation between the Royal Society B Biological Sciences, 370(1672), 20140189.
nerve terminal size and transmitter release at the neuromuscular junc- Tononi, G., & Cirelli, C. (2014). Sleep and the price of plasticity: From syn-
tion of the frog. The Journal of Physiology, 213(3), 545–556. aptic and cellular homeostasis to memory consolidation and integra-
Lavidis, N. A. (1995a). The effect of opiates on the secretion of transmitter tion. Neuron, 81(1), 12–34.
from amphibian motor nerve terminals. European Journal of Pharmacol- Wachman, E. S., Poage, R. E., Stiles, J. R., Farkas, D. L., & Meriney, S. D.
ogy, 276(1–2), 71–76. (2004). Spatial distribution of calcium entry evoked by single action
Lavidis, N. A. (1995b). The effect of opiates on the terminal nerve impulse potentials within the presynaptic active zone. The Journal of Neurosci-
and quantal secretion from visualized amphibian nerve terminals. Brit- ence, 24(12), 2877–2885.
ish Journal of Pharmacology, 115(3), 441–450. Werle, M. J., Jones, M. A., & Stanco, A. M. (1999). Aggregates of acetyl-
Lavidis, N. A., Hudson, N. J., Choy, P. T., Lehnert, S. A., & Franklin, C. E. choline receptors are not observed under anti-agrin staining schwann
(2008). Role of calcium and vesicle-docking proteins in remobilising cell processes at the frog neuromuscular junction. Journal of Neurobiol-
dormant neuromuscular junctions in desert frogs. Journal of Compara- ogy, 40(1), 45–54.
tive Physiology A, 194(1), 27–37. Wernig, A., Anzil, A. P., & Bieser, A. (1981). Light and electron microscopic
Leung, C. C. Y., & Wong, Y. H. (2017). Role of G protein-coupled receptors identification of a nerve sprout in muscle of normal adult frog. Neuro-
in the regulation of structural plasticity and cognitive function. Mole- science Letters, 21(3), 261–266.
cules, 22(7), 1239–1254. Wernig, A., & Herrera, A. A. (1986). Sprouting and remodeling at the nerve
Lupien, S. J., McEwen, B. S., Gunnar, M. R., & Heim, C. (2009). Effects of muscle junction. Progress in Neurobiology, 27(3), 251–291.
stress throughout the lifespan on the brain, behaviour and cognition. Wernig, A., Pecotdechavassine, M., & Stover, H. (1980). Sprouting and
Nature Reviews Neuroscience, 10(6), 434–445. regression of the nerve at the frog neuromuscular-junction in Normal
Macleod, G. T., Dickens, P. A., & Bennett, M. R. (2001). Formation and conditions and after prolonged paralysis with curare. Journal of Neuro-
function of synapses with respect to Schwann cells at the end of motor cytology, 9(3), 277–303.
nerve terminal branches on mature amphibian (Bufo marinus) muscle. Witzemann, V., Chevessier, F., Pacifici, P. G., & Yampolsky, P. (2013). The neu-
The Journal of Neuroscience, 21(7), 2380–2392. romuscular junction: Selective remodeling of synaptic regulators at the
Meriney, S. D., & Dittrich, M. (2013). Organization and function of trans- nerve/muscle interface. Mechanisms of Development, 130(6–8), 402–411.
mitter release sites at the neuromuscular junction. The Journal of Physi- Zinchuk, V., Wu, Y., & Grossenbacher-Zinchuk, O. (2013). Bridging the gap
ology, 591(13), 3159–3165. between qualitative and quantitative colocalization results in fluores-
Nishiyama, J., & Yasuda, R. (2015). Biochemical computation for spine cence microscopy studies. Scientific Reports, 3, 1365–1369.
structural plasticity. Neuron, 87(1), 63–75. Zinchuk, V., Zinchuk, O., & Okada, T. (2007). Quantitative colocalization
Nudell, B. M., & Grinnell, A. D. (1982). Inverse relationship between trans- analysis of multicolor confocal immunofluorescence microscopy
mitter release and terminal length in synapses on frog muscle fibers of images: Pushing pixels to explore biological phenomena. Acta Histo-
uniform input resistance. The Journal of Neuroscience, 2(2), 216–224. chemica et Cytochemica, 40(4), 101–111.
Pawson, P. A., & Grinnell, A. D. (1989). Seasonal changes in the normal var-
iability in release properties of motor nerve terminals in Rana pipiens.
SUPPOR TI NG I NFORMATION
Brain Research, 495(1), 182–188.
Plomp, J. J., Huijbers, M. G. M., & Verschuuren, J. (2018). Neuromuscular Additional supporting information may be found online in the Sup-
synapse electrophysiology in myasthenia gravis animal models. Annals
porting Information section at the end of this article.
of the New York Academy of Sciences, 1412(1), 146–153.
Sanes, J. R., & Lichtman, J. W. (1999). Development of the vertebrate neu-
romuscular junction. Annual Review of Neuroscience, 22, 389–442.
Sanes, J. R., & Lichtman, J. W. (2001). Induction, assembly, maturation and How to cite this article: Ge D, Noakes PG, Lavidis NA. Sea-
maintenance of a postsynaptic apparatus. Nature Reviews Neuroscience, sonal comparison of the neuromuscular junction morphology
2(11), 791–805. of Bufo marinus. J Comp Neurol. 2019;527:1931–1939.
Singh, I. (1964). Seasonal variations in the nature of neurotransmitters in a
frog vagus-stomach muscle preparation. Archives Internationales de
https://doi.org/10.1002/cne.24661
Physiologie et de Biochimie, 72(5), 843–851.