Presynaptic endoplasmic reticulum and neurotransmission

Abstract

Synaptic transmission relies on rapid calcium (Ca²⁺) influx into presynaptic terminal via voltage-gated Ca²⁺ channels. However, smooth ER is present in presynaptic terminals and accumulating evidence indicate that ER Ca²⁺ signaling may play a modulatory role in synaptic transmission. Most recent publication by Lindhout and colleagues (EMBO J, 38 (2019) e101345) suggested that the fragmentation state of the ER affects synaptic vesicle release. Here we discuss these results as well as several key publications that addressed a connection between ER Ca²⁺ signaling and synaptic transmission.

The vast majority of synaptic transmission literature is focused on the role of voltage-gated Ca²⁺ channels in evoked neurotransmitter release. Rapid and local Ca²⁺ influx via voltage-gated Ca²⁺ channels elicits synchronized fusion of synaptic vesicles [1]. In this context, slow and delocalized Ca²⁺ release from the endoplasmic reticulum (ER) does not appear to have a significant role. However, smooth ER is present in presynaptic terminals [2] and a number of experimental observations suggest that ER Ca²⁺ stores may play a modulatory role in mediating synaptic transmission (Fig. 1). Several lines of evidence over the last two decades support the premise that Ca²⁺ release from intracellular stores can increase baseline spontaneous release frequency in hippocampal synapses [3] as well as in the cerebellum [4]. This pathway can be activated by classical modulatory neurotransmitters such as acetylcholine [3] or by unconventional neuromodulators such as Reelin [5].

Fig. 1. Presynaptic ER Ca²⁺ stores.

Presynaptic terminal is filled with synaptic vesicles (SVs). Canonical evoked neurotransmission is driven by Ca²⁺ influx through voltage gated Ca²⁺ channels (VGCCs). Smooth ER is present in many presynpatic terminals. Ca²⁺ release from ER is mediated by ryanodine receptors (RyR) by a mechanism of Ca²⁺-induced Ca²⁺ release (CICR). ER Ca²⁺ depletion causes clustering of STIM1 Ca²⁺ sensor protein and activation of store-operated Ca²⁺ entry pathway (SOCE) via Ca²⁺ influx through Orai channels. There is increasing evidence that both CICR and SOCE can impact neurotransmitter release.

In addition to the impact of Ca²⁺ release from stores on baseline spontaneous release, evoked release probability was also proposed to be affected by ER Ca²⁺ levels. Emptage and colleagues [6] performed electrophysiological and imaging experiments in organotypic hippocampal slices and showed that action potential-induced Ca²⁺ influx was followed by Ca²⁺-induced Ca²⁺ release (CICR) in presynaptic terminals. In addition, they discovered that ryanodine (inhibitor of ryanodine receptors) or CPA (inhibitor of SERCA Ca²⁺ pump) caused significant reduction in paired-pulse facilitation. Moreover, they reported that store-operated Ca²⁺ entry (SOCE) in presynaptic terminals had major influence on the rate of spontaneous neurotransmitter release. These results suggested that not only Ca²⁺ release from presynaptic ER store but also Ca²⁺ entry activated by depletion of these stores (i.e. SOCE process) can impact neurotransmission. Although the direct impact of presynaptic ER Ca²⁺ on rapid evoked release remains controversial [7], the role of ER-mediated Ca²⁺ signaling in regulation of slower forms of release has been observed in multiple systems. It is important to note that these earlier studies were performed before molecular identity of SOCE components became known and as such they solely relied on pharmacological tools that may have non-specific effects.

More recently, the role of store-operated Ca²⁺ entry in neurotransmission was revisited with the advantage of molecular tools and advanced imaging methods. A recent study by de Juan-Sanz and colleagues, performed in primary hippocampal neuronal cultures, proposed an interesting and provocative idea [8]. These investigators reasoned that presynaptic ER Ca²⁺ stores modulate synaptic transmission not by changing Ca²⁺ levels in the terminal but by affecting clustering state of STIM1 (ER Ca²⁺ sensor that mediates SOCE) (Fig. 1). Results obtained in this paper suggested existence of a negative feedback loop that links ER Ca²⁺ stores in the presynaptic terminal and action potential-evoked synaptic transmission events. Mechanistically, these authors argued that ER Ca²⁺ store depletion causes enhanced clustering of STIM1 and that clustered STIM1 exerts inhibitory effect on synaptic transmission. The authors speculated that STIM1 may exert direct inhibitory effect on voltage-gated Ca²⁺ channels or impair coupling between Ca²⁺ influx and synaptic vesicle fusion. These are very provocative and interesting findings, in future studies exploration of their potential mechanism will be extremely informative.

A more recent study by Lindhout et al. [9] approached the same problem from a cell biological perspective. Using primary hippocampal neuronal cultures, these investigators evaluated a role of ER integrity and remodeling in presynaptic function. In these studies, they focused on ER membrane receptors VAPA and VABP and their binding partner SCRN1. Based on a series of gain-of-function and loss-of-function experiments, they demonstrated that VAP-SCRN1 association affects ER structure and dynamics. Disruption of VAP-SCRN1 association resulted in fragmentation of ER and also caused decreased cycling of synaptic vesicles as detected by diminished uptake of an antibody targeting luminal region of synaptic vesicle protein synaptotagmin1. The authors concluded that VAP-SCRN1 association and local ER Ca²⁺ morphology modulate steady state Ca²⁺ homeostasis at presynaptic terminals and affect action potential-driven Ca²⁺ influx. The authors argue that fragmentation of ER leads to elevated steady-state Ca²⁺ levels and reduced driving force for Ca²⁺ entry into presynaptic terminal. This explanation is not likely as steady-state Ca²⁺ levels remain in the submicromolar range in these experiments. Another suggestion is that elevated steady state Ca²⁺ levels lead to compensatory changes in presynaptic terminal that leads to downregulation of Ca²⁺ channel activity and synaptic vesicle cycling. This explanation appears more plausible. In particular, there is evidence that elevated steady state Ca² levels can slow synaptic vesicle retrieval and trafficking [10]. However, it is difficult to propose a clear explanation for these findings based on synaptotagmin1 antibody uptake alone. Future experiments using electrophysiological approaches combined with optical tools to monitor synaptic vesicle trafficking are needed to pinpoint the direct impact of VAP-SCRN1 function and presynaptic ER integrity on exocytosis and/or endocytosis in presynaptic terminals.

Overall, the study by Lindhout and colleagues adds to the increasing lines of evidence that presynaptic ER Ca²⁺ stores do in fact play a modulatory role in control of synaptic vesicle cycling and synaptic transmission. Although, at this time it is difficult to reconcile several proposed models and ideas, future studies should provide additional insights into the role of ER integrity and various forms of ER-dependent Ca²⁺ signaling in presynaptic function.