1
2 Increased and synchronous recruitment of release sites underlies
3 hippocampal mossy fiber presynaptic potentiation
4
5 Short title:
6 Release site increase mediates presynaptic LTP
7
8 Marta Orlando
1,2,
¶
,
*, Anton Dvorzhak
1,2,
¶
, Felicitas Bruentgens
1,2,
¶
, Marta Maglione
2,3
, Benjamin R.
9 Rost
2,4
, Stephan J. Sigrist
2,3
, Jörg Breustedt
1,2
, Dietmar Schmitz
1,2,4-7,
*
10
11
12
1
Charité – Universitätsmedizin Berlin, corporate member of Freie Universität Berlin and Humboldt-
13 Universität zu Berlin, and Berlin Institute of Health, Charitéplatz 1, 10117 Berlin, Germany
14
2
NeuroCure Cluster of Excellence, Charitéplatz 1, 10117 Berlin, Germany
15
3
Department of Biology, Chemistry, Pharmacy, Freie Universität Berlin, 14195, Berlin, Germany
16
4
German Center for Neurodegenerative Diseases (DZNE) Berlin, 10117 Berlin, Germany
17
5
Bernstein Center for Computational Neuroscience (BCCN) Berlin, 10115 Berlin, Germany
18
6
Einstein Center for Neurosciences (ECN) Berlin, 10117 Berlin, Germany
19
7
Max-Delbrück-Center (MDC) for molecular medicine, 13125 Berlin, Germany
20
21 * e-mail: marta.orlando@charite.de & dietmar.schmitz@charite.de
22 Lead Contact: Dietmar Schmitz
23
24
¶
These authors contributed equally
.CC-BY 4.0 International licensemade available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is
The copyright holder for this preprintthis version posted August 21, 2020. ; https://doi.org/10.1101/2020.08.21.260638doi: bioRxiv preprint
25 ABSTRACT
26
27 Synaptic plasticity is a cellular model for learning and memory. However, the expression mechanisms
28 underlying presynaptic forms of plasticity are not well understood. Here, we investigate functional and
29 structural correlates of long-term potentiation at large hippocampal mossy fiber boutons induced by the
30 adenylyl cyclase activator forskolin. We performed two-photon imaging of the genetically encoded
31 glutamate sensor iGlu
u
that
revealed an increase in the surface area used for glutamate release at
32 potentiated terminals. Moreover, time-gated stimulated emission depletion microscopy revealed no
33 change in the coupling distance between immunofluorescence signals from calcium channels and release
34 sites. Finally, by high-pressure freezing and transmission electron microscopy analysis, we found a fast
35 remodeling of synaptic ultrastructure at potentiated boutons: synaptic vesicles dispersed in the terminal
36 and accumulated at the active zones, while active zone density and synaptic complexity increased. We
37 suggest that these rapid and early structural rearrangements likely enable long-term increase in synaptic
38 strength.
.CC-BY 4.0 International licensemade available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is
The copyright holder for this preprintthis version posted August 21, 2020. ; https://doi.org/10.1101/2020.08.21.260638doi: bioRxiv preprint
39 INTRODUCTION
40
41 The term synaptic plasticity describes the ability of synapses to change their strength and efficacy over
42 time. Long-term forms of synaptic plasticity are postulated as cellular mechanisms responsible for
43 learning and memory (Kandel, 2001; Citri and Malenka, 2008). Changes in synaptic strength are
44 paralleled by changes in the structure of neuronal contacts that underlie long-term circuit reorganization
45 (Holtmaat and Svoboda, 2009; Monday et al., 2018). The long-term increase in synaptic strength (LTP)
46 can be expressed postsynaptically, importantly by changes in postsynaptic receptor number or properties
47 (Lüscher and Malenka, 2012), but also presynaptically, by changes in neurotransmitter release (Monday
48 et al., 2018).
49 In this study we investigated presynaptic LTP at large hippocampal mossy fiber boutons (hMFB) (Nicoll
50 and Schmitz, 2005). Dentate gyrus granule cells form excitatory synapses onto spines of proximal
51 dendrites of CA3 pyramidal neurons (Amaral and Dent, 1981). hMFBs were the first synapses described
52 to undergo a NMDA receptor independent form of LTP that is both induced and expressed at the
53 presynaptic terminal (Zalutsky and Nicoll, 1990; Yang and Calakos, 2013). Here, the increase in
54 intracellular calcium following high-frequency firing activates calcium/calmodulin dependent adenylyl
55 cyclases, which leads to an increase in the intracellular concentration of cyclic adenosine
56 monophosphate (cAMP) that, in turn, drives the activation of protein kinase A (PKA). Ultimately, PKA
57 phosphorylation events result in a long-lasting increase in neurotransmission (Villacres et al., 1998;
58 Nicoll and Schmitz, 2005).
59 A variety of knock-out models provided information on potential PKA phosphorylation targets required
60 for presynaptic potentiation. Rab3A (Castillo et al., 1997), its interaction partners RIM1 and Munc13
61 (Yang and Calakos, 2011) and synaptotagmin12 (Kaeser-Woo et al., 2013) have all been shown to be
62 crucial for presynaptic LTP at hMFBs, but how exactly these proteins are involved in its induction and
63 expression is not known (Monday et al., 2018).
64 Presynaptic LTP at hMFBs has traditionally been described as the long-lasting increase in release
65 probability (P
r
) (Malinow and Tsien, 1990; Hirata et al., 1991; Yang and Calakos, 2013), but vesicle
66 availability as well as changes in the number of release sites could also play a major role in setting the
.CC-BY 4.0 International licensemade available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is
The copyright holder for this preprintthis version posted August 21, 2020. ; https://doi.org/10.1101/2020.08.21.260638doi: bioRxiv preprint
67 stage for increased neurotransmission. Indeed, at hMFBs, an increase in docked vesicles has been
68 proposed as a mechanism for post-tetanic-potentiation (Vandael et al., 2020). At cerebellar parallel and
69 climbing fiber synapses, PKA and its vesicle associated target, synapsin, dynamically control release
70 site occupancy and dictate the number of vesicles released per action potential without altering P
r
(Vaden
71 et al., 2019). Moreover, activation of silent synapses and/or addition of release sites have been suggested
72 as potential mechanisms for the expression of presynaptic LTP at hMFBs (Tong et al., 1996; Emptage
73 et al., 2003). Changes in the number and localization of docked vesicles (Sigrist and Schmitz, 2011),
74 potentially accompanied by addition of new release sites, could underlie functional changes at hMFBs.
75 The morphological complexity of mossy fiber boutons has been shown to increase in mice kept in an
76 enriched environment (Galimberti et al., 2006) and, in cryo-fixed organotypic slices treated with the
77 potassium channel blocker TEA (Zhao et al., 2012). Moreover, the transport of active zone (AZ) proteins
78 via vesicular cargo to nascent AZs likely underlies long-term plasticity in the hippocampus (Bell et al.,
79 2014).
80 Changes in AZ nano-architecture upon LTP induction have also been hypothesized to sustain the
81 increase in P
r
. Direct double patch-clamp experiments from presynaptic hMFBs and postsynaptic CA3
82 pyramidal neurons indicated a relatively long distance (70 to 80 nm) between calcium channels and
83 synaptic vesicles (SVs) and therefore a functionally “loose coupling” between calcium source and
84 calcium sensor (Vyleta and Jonas, 2014). Loose coupling is responsible for the intrinsically low P
r
of
85 this synapse (Ghelani and Sigrist, 2018). Remarkably, experiments at dissociated hMFBs suggested a
86 decreased coupling distance between calcium channels and calcium sensor as a possible mechanism for
87 LTP expression (Midorikawa and Sakaba, 2017).
88 The complexity of the phenomenon and the fact that a variety of different experimental models have
89 been used in the past decades, might explain why we currently face several diverging theories to explain
90 hMFB presynaptic LTP.
91 Our aim, in this context, was to characterize the ultrastructural and functional correlates of presynaptic
92 LTP in brain slices to clarify whether and how synapses, vesicles, or AZ reorganize to express and
93 sustain the long-term increase in neurotransmitter release. By means of two-photon fluorescent imaging
94 of glutamate release, STED microscopy and three-dimensional transmission electron microscopy (EM)
.CC-BY 4.0 International licensemade available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is
The copyright holder for this preprintthis version posted August 21, 2020. ; https://doi.org/10.1101/2020.08.21.260638doi: bioRxiv preprint
95 analysis we addressed the following questions: does the addition of release sites play a role in
96 presynaptic LTP expression? How do glutamate release dynamics change upon presynaptic
97 potentiation? Does the active zone nano-architecture rearrange to sustain long-term increase in synaptic
98 strength?
.CC-BY 4.0 International licensemade available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is
The copyright holder for this preprintthis version posted August 21, 2020. ; https://doi.org/10.1101/2020.08.21.260638doi: bioRxiv preprint