**2.1 Original findings**

Previously, in experiments monitoring not only [Ca2+] but also AVP release from HNS terminals in response to K<sup>+</sup> depolarization, it was shown that none of the classical modulators of intracellular Ca2+ release, such as caffeine, affected any of these measurements [12]. In contrast, several population (using dissociated NHTs) release experiments have been performed [1, 2, 13, 14], assessing the effects of both ryanodine and caffeine on basal neuropeptide release. Caffeine, a strong agonist of RyRs [15], induces an increase in neuropeptide release from these terminals, even in the absence of extracellular Ca2+ [1, 13, 14]. Ryanodine, at a concentration at which ryanodine is known to inactivate RyRs [15], has been shown to half-block this caffeine-induced release [14, 16]. This is similar to its effects on Ca2+ spark (syntilla) frequency in these NHTs [2]. Our results indicate that depolarization-induced neuropeptide secretion is present in the absence of external calcium, and calcium release from ryanodine-sensitive internal stores is a significant physiological contributor to neuropeptide secretion from HNS terminals.

## **2.2 Syntillas and calcium stores within NT**

More conclusive evidence for Ca2+ stores in NHTs came from our studies demonstrating the presence of spontaneous focal Ca2+ transients in mouse neurohypophysial terminals [2, 17]. Since these Ca2+ syntillas were found in the absence of extracellular Ca2+, they had to arise from intracellular stores. Additionally, the rate of syntillas is affected by agonists/antagonists of the RyR, a channel that is well known to control

the release of [Ca2+]. Dihydropyridine receptors (DHPRs) function as voltage sensors within terminals of MCNs. The DHPRs appear to be linked to type-1 RyRs in a manner bearing similarities to the mechanism in skeletal muscle [17, 18]. Data from immunocytochemistry, Western blot analysis, and electrophysiology demonstrate the existence of type-1 RyRs linking neuronal activity, as signaled by depolarization of the plasma membrane, to a rise in [Ca2+] in nerve terminals [12]. These results support previous findings [17] indicating the role of type-1 RyRs in response to depolarization and imply its possible physiological significance in depolarization secretion coupling (DSC) (**Figure 1**).

The cyclic ADP-ribose (cADPr) signaling pathway (**Figure 1**) initiates a signaling cascade leading to activation of the RyRs in vivo and subsequent release of [Ca2+] from ryanodine-sensitive stores [19–21]. Interestingly, in NHTs, blocking cADPr signaling was shown to attenuate high [K+ ] -induced rises in [Ca2+] but only inhibited OT release from isolated terminals [22, 23]. This strongly suggests that the cADPr pathway is present in OT terminals [1] and linked to neuropeptide release (**Figure 1**). Cyclic ADP ribose hydrolase (CD38) is a catalyst for the formation of cADPr and nicotinic acid adenine dinucleotide phosphate (NAADP) by ADP-ribosyl cyclase from nicotinamide adenine dinucleotide (NAD) and NAD phosphate [24]. Both are known to release Ca2+ from intracellular ryanodine-sensitive pools as part of a second messenger-signaling pathway.

#### **2.3 NH terminals compared to HNS cells**

Facilitation of neuropeptide release from magnocellular neuron (MCN) somatodendrites can be induced by increased [Ca2+] [25]. This facilitation appears to be due to the trafficking of neurosecretory granules (NSGs) toward the cell surface [26]. Such facilitation increases the somatodendritic secretory response to signals, such as OT, that mobilize intracellular Ca2+ (via IP3 receptors) in ER [25]. AVP and OT normally elicit little somatodendritic secretion, but after facilitation, each neuropeptide elicits enhanced somatodendritic secretion from their respective MCNs [25, 27]. This includes the response to somatodendritic action potentials, which typically do not result in local release, but can alter facilitation by AVP or OT.
