Single point mutants in the itpr gene were generated in an EMS (ethyl methane sulfonate) screen; detailed molecular information on these alleles has been published (Joshi et al., 2004; Srikanth et al., 2004). The UAS transgenic strains used have been published and the appropriate references are included in the results. Elav^C155GAL4 (pan neuronal) was obtained from Bloomington Stock Center, Indiana University, Bloomington, IN, USA. Canton S (CS), in the background of which all mutant and transgenics were back-crossed, was used as the wild-type control. RNAi lines for itpr (1063) was obtained from National Institute of Genetics, Japan, and for dSTIM (47073), cac (104186), Ca-β (102188), Ca-α1D (51491) and Ca-α1T (48008, 31961) were obtained from Vienna Drosophila Resource Centre, Austria. Fly strains used in this study were generated by standard genetic methods using individual mutant and transgenic fly lines described above.

The protocol for pupal neuronal culture was adapted from Sicaeros and O’Dowd (2007). The brain and ventral ganglion complexes were dissected from different pupal stages of the appropriate genotypes. Dissected brain tissues were incubated for 15 min at RT with 50 U/ml papain activated by 1.32 mM cysteine in dissecting saline (5.4 mM KCl/137 mM NaCl/0.22 mM KH₂PO₄/0.17 mM NaH₂PO₄/43.8 mM sucrose/33.3 mM glucose/9.9 mM HEPES, pH 7.3 with NaOH). The brain tissue was dissociated with gentle pipetting. The lysate containing a mixture of tissue clumps and single cells was spun down, re-suspended and plated onto poly lysine-coated glass coverslip mounted to the bottom of a petri dish. Cells were resuspended and cultured in DMEM/F12-1065 (Life Technologies, Carlsbad, CA, USA), containing Glutamax-I, 2.438 sodium bicarbonate and sodium pyruvate, supplemented with 50 U/ml penicillin (Life Technologies, Carlsbad, CA, USA), 50 μg/ml streptomycin (Life Technologies, Carlsbad, CA, USA), and 10 μg/ml amphotericin B (Life Technologies, Carlsbad, CA, USA), 1 mg/ml sodium bicarbonate, 20 mM HEPES, 100 μM putrescine, 20 ng/ml progesterone, 50 μg/ml insulin, 1 μg/ml 20-hydroxyecdysone. The cells were incubated at 25°C in a humidified incubator with 5% CO₂ for 14–16 h. All chemicals for cell culture were obtained from Sigma-Aldrich (St. Louis, MO, USA) unless otherwise stated.

For measurement of spontaneous calcium influx and SOCE cells were incubated with 2.5 μM Fluo-4 AM and 0.02% Pluronic F-127 in M1 media for 30 min at room temperature in the dark. Cells were washed twice with M1 before and after dye incubation and finally covered with M1 (30 mM HEPES/150 mM NaCl/5 mM KCl/2 mM MgCl₂/35 mM sucrose, pH 7.2 with NaOH) with (1 mM CaCl₂) or without (2 mM EGTA) Ca²⁺. Data were acquired using 488 nm excitation and 520 nm emission filter sets at 15 s intervals. An Olympus IX81-ZDC2 inverted wide field microscope with epifluorescence and Z-drift compensation and a 60×/1.35 NA (oil) objective lens, was used for calcium imaging. Excitation of fluorescent Ca²⁺ indicator dyes was performed using specific wavelength illuminations from a halogen arc lamp with TILL Polychrome 5000 monochromator (TILL Photonics, Graefelfing, Germany) for variable bandwidth and intensity. Emitted light was detected through band pass filter sets (Chroma, Brattleboro, VT, USA). Image acquisition was performed using the Andor iXON 897E EMCCD camera and Andor iQ 2.4.2 imaging software. The time lapse acquisition mode of the software was used to follow fluorescence changes over time.

Images acquired through Andor iQ 2.4.2 on Olympus IX81-ZDC2 were analyzed with ImageJ 1.43m (NIH, Bethesda, MD, USA). Each cell was marked separately and mean fluorescence intensity was calculated in arbitrary units. Arbitrary units of fluorescence for each cell were converted to ΔF/F₀ values and the highest ΔF/F₀ values were tabulated; ΔF/F₀ values for each assay and each genotype were plotted as a box plot or cumulative frequency distribution plot in Origin 8.0 software (Origin Lab, Northampton, MA, USA). To compare data between genotypes or assay conditions Kruskal-Wallis test for variance followed by Wilcoxon post hoc test was performed.

Total RNA was isolated from 5 to 10 dissected CNS from suitably aged Drosophila pupae with TRIzol reagent (Invitrogen, Life Technologies, Carlsbad, CA, USA) following manufacturer’s instructions. RNA was dissolved in nuclease free water and quantified using a nanodrop machine (Thermo Scientific, Wilmington, DC, USA) and the integrity was checked on a 1.5% TAE gel. Approximately 200 ng RNA was used for cDNA preparation by Reverse Transcriptase as described in Pathak et al. (2015). Quantitative PCR (qPCR) was performed by ABI 7500 fast machine operated by ABI 7500 software using KAPA^™ SYBR^® FAST qPCR master mix (Kapa Biosystems, Wilmington, MA, USA) for SYBR assay I dTTP (Eurogentec, Belgium). For each genotype, three biological replicates and two technical replicates were included in the experiment. rp49 was used as internal control. A melt curve was performed after the assay to check for specificity of the reaction. The fold change of gene expression in test genotype relative to control was determined by the comparative DDCt, where DDCt = (Ct_target − Ct_p49)_test − (Ct_target − Ct_rp49)_control.

Pupal CNS lysates were extracted in 50 mM Tris (pH 8), 150 mM NaCl, 1 mM EGTA and 1% (v/v) Triton X-100 with 1 mM PMSF, 0.5 μM E64 Calpain inhibitor and 100 μg/ml leupeptin. Protein extract was boiled for 5 min at 95°C in 125 mM Tris pH 6.8, 5% (v/v) Glycerol, 0.25% (w/v) SDS, 2% (v/v) β-mercaptoethanol, 10% (w/v) Bromophenol blue and resolved in 8% SDS polyacrylamide gel electrophoresis. Proteins were detected with 1:5000 anti-TRP (kind gift from Prof. Raghu Padinjat, NCBS).

Flies were anesthetized in ice before tethering to a tungsten wire between the head and thorax and allowed to recover for 30 min before recording flight. Flight was stimulated with a puff of air and recorded for 30 s.

To explore the role of IP₃-mediated Ca²⁺ signaling in spontaneous Ca²⁺ influx and SOCE in Drosophila pupal neurons, primary neurons from the central nervous systems of 48–50 h pupae were cultured for 16–18 h and Ca²⁺ signals of single neurons were recorded. In presence of 2 mM extracellular CaCl₂ wild type and itpr mutant neurons displayed spontaneous Ca²⁺ signals in culture (Figures 1A,B). The amplitude of signals was significantly higher in neurons with mutant IP₃Rs (S224F/G1891S and S224F/G2117E; heteroallelic combinations) as compared with wild type (P < 0.01, Kruskal-Wallis test of variation followed by Wilcoxon post hoc test; Figures 1A,B,D, 2D). The spontaneous Ca²⁺ signals observed in both genotypes were sustained rather than transients. Spontaneous Ca²⁺ signals were not observed in minimal extracellular Ca²⁺ (8–11 nM) and after application of either 50 μM lanthanum chloride (Prakriya and Lewis, 2001) or 2 mM cobalt chloride (Jiang et al., 2005) in presence of 2 mM extracellular CaCl₂ (Figure 1C), suggesting that the spontaneous Ca²⁺ signals observed were due to a Ca²⁺ influx across the PM and not from intracellular stores. This observation is consistent with published data reporting an absence of spontaneous Ca²⁺ transients in the absence of extracellular Ca²⁺ in cultured Drosophila pupal neurons of mushroom body Kenyon cells (Jiang et al., 2005). However, the frequency and amplitude of spontaneous Ca²⁺-signals observed in our study differed from what was reported in mushroom body Kenyon cells (Jiang et al., 2005), possibly due to variations in spontaneous Ca²⁺-signals in specific subtypes of neurons and culture conditions.

Drosophila larval neurons cultured from itpr mutants (S224F/G1891S; heteroallelic combination) displayed reduced SOCE after passive depletion of intracellular store-Ca²⁺ with 10 μM thapsigargin (Venkiteswaran and Hasan, 2009; Chakraborty and Hasan, 2012a, b; Chakraborty et al., 2016). To understand if SOCE is also compromised in pupal neurons from itpr mutants, intracellular store-Ca²⁺ was depleted with 10 μM thapsigargin in minimal extracellular Ca²⁺ (8–11 nM) followed by measurement of SOCE by replenishing 2 mM extracellular CaCl₂. Because itpr mutant pupal neurons displayed higher spontaneous Ca²⁺ influx as compared to wild type in presence of extracellular Ca²⁺, SOCE was determined by subtracting spontaneous Ca²⁺ influx from Ca²⁺ influx induced by store-Ca²⁺ depletion with thapsigargin (Figure 1E). As in larval neurons, SOCE was significantly (P < 0.01, Kruskal-Wallis test of variation followed by Wilcoxon post hoc test) reduced in itpr mutant pupal neurons (Figure 1F).

To understand if spontaneous Ca²⁺ influx in pupal neurons is affected by SOCE, spontaneous Ca²⁺ influx and SOCE was measured in pupal neurons over-expressing dSTIM and dOrai in presence or absence of mutant IP₃Rs (Figure 2). Higher spontaneous Ca²⁺ influx was significantly reduced by over-expression of dSTIM (Figures 2A,D) and dOrai (Figure 2D) in itpr mutant neurons, suggesting that their higher spontaneous Ca²⁺ influx is a compensatory increase for reduced SOCE. This idea was tested further, by measuring spontaneous Ca²⁺ influx and SOCE in pupal neurons with knock down of dSTIM. Knock down of dSTIM (dsdSTIM) reduced SOCE in pupal neurons (Figures 2C,D); however, it did not result in significant up regulation of spontaneous Ca²⁺ influx (Figures 2A,D). Thus the higher spontaneous Ca²⁺ influx in itpr mutant neurons appears to be an adaptive effect of recurrent loss of IP₃R function rather than a response to loss of SOCE. Loss of SOCE in itpr mutant larval neurons can be restored by over-expression of dSTIM and dOrai (Agrawal et al., 2010). Over-expressing dSTIM and dOrai in wild type and itpr mutant pupal neurons also increased SOCE after store-Ca²⁺ depletion with thapsigargin (Figures 2B,C,E). As published in larval neurons (Chakraborty et al., 2016), over-expression of dSTIM and dOrai in itpr mutant neurons did not alter store-Ca²⁺ depletion in pupal neurons. However, knock down of dSTIM attenuated intracellular store Ca²⁺ release presumably due to reduced store [Ca²⁺] (Figures 2B,C,E). These data suggest that loss of SOCE, in absence of a compensatory increase in spontaneous Ca²⁺ influx as seen in neurons with dSTIM knock-down (Figure 2A), leads to reduced Ca²⁺ in intracellular stores. This reiterates that increased spontaneous Ca²⁺ entry in itpr mutant neurons helps refill ER store in the absence of SOCE.

To identify PM Ca²⁺ channels that contribute to higher spontaneous Ca²⁺ influx in Drosophila itpr mutant pupal neurons, we performed quantitative PCR analysis of mRNA encoding subunits of VGCCs (Ca-β, cacophony/cac, Ca-α1D and Ca-α1T), TRP channels (trp, trpl) and NCX (calx) in extracts derived from 48 h to 50 h pupal central nervous systems. The choice of candidates investigated was based on existing literature (Gu et al., 2009; Iniguez et al., 2013; Kanamori et al., 2013). The time window selected was based on previous data on the requirement of IP₃R function and SOCE for the development of the Drosophila flight circuit (Banerjee et al., 2004; Richhariya et al., 2017) and on the onset of spontaneous Ca²⁺ transients in Drosophila pupal neurons (Jiang et al., 2005). Spontaneous Ca²⁺ transients in distinct subsets of Drosophila pupal neurons (in vivo and in 2 day old cultures) have been attributed to PLTX-II sensitive P/Q-type VGCC, cac (Jiang et al., 2005; Gu et al., 2009) or to T-type VGCC, Ca-αT (Iniguez et al., 2013). An increased activity of voltage-activated transient K⁺ current I_A was observed in cac mutant neurons, associated with an increased expression of Shaker, gene for I_A, and a corresponding reduction in the expression of Slowpoke, gene for [IK_(Ca)] (Peng and Wu, 2007). Hence expression levels of K⁺ channel Slowpoke and Shaker were also analyzed as reporters for VGCC activity. Surprisingly, a significant reduction in cacophony/cac, Ca-α1D and Ca-αT mRNA levels was observed in itpr mutant neurons with a corresponding reduction in Slowpoke mRNA. Expression of Ca-β and Shaker, however, was not significantly altered (P < 0.05, Student’s t-test; Figure 3A). These data suggest that higher spontaneous Ca²⁺ influx in itpr mutant neurons is very likely not mediated by VGCCs. On the contrary, the normal expression of VGCCs correlates with normal IP₃R function.

The TRPC class of PM Ca²⁺ channels are responsible for Ca²⁺ spike activity in developing Xenopus spinal cord, where Shh signaling induces synchronous Ca²⁺ spikes and IP₃ transients at the primary cilium (Belgacem and Borodinsky, 2011). NCX (Na⁺/Ca²⁺ exchanger) expression is indirectly modulated by SOCE, through ERK1/2, in neuronal PC12 (Sirabella et al., 2012) and rat parotid acinar cells (Soltoff and Lannon, 2013). Therefore, the levels for mRNAs encoding two Drosophila TRPC proteins; TRP and TRPL (Thebault et al., 2005) and a Drosophila NCX homolog, CalX (Wu et al., 2011) were tested. A significant increase (P < 0.05, Student’s t-test) in mRNA levels of trp was observed in itpr mutant neurons, whereas expression levels of trpl and calx remained unchanged (Figure 3A). Over-expression of TRP in the pupal nervous system of itpr mutants was confirmed by measuring expression of TRP protein directly (Figures 3B,C).

Interestingly, rescue of SOCE by over-expression of dSTIM and dOrai in itpr mutant neurons reverted the change in mRNA levels of cacophony/cac, Ca-α1D, Slowpoke and trp (P < 0.05, Student’s t-test; Figure 3A). Moreover, expression of Ca-α1T and trpl (P < 0.05, Student’s t-test; Figure 3A) was also altered significantly suggesting that raising SOCE in itpr mutant neurons, by overexpression of dSTIM and dOrai, altered gene expression of VGCCs and TRPs. To understand if expression levels of cacophony/cac, Ca-α1D, Slowpoke and trp were directly sensitive to SOCE, mRNA levels were analyzed from pupal neurons with down regulation of dSTIM (Figure 3D). A significant (P < 0.05, Student’s t-test) down regulation of cac and Slowpoke was observed in dSTIM knocked down neurons whereas expression levels of Shaker, Ca-β, Ca-α1D and trp, remained unchanged. These data suggest that the expression of cac and slo are sensitive to SOCE. The reduced expression of Slowpoke possibly reports reduced activity of the P/Q-type VGCC encoded by cac, as reported earlier for cac mutant neurons (Peng and Wu, 2007). However, it is also possible that Slowpoke expression in this case is directly regulated by SOCE rather than through reduced activity of VGCCs. The absence of TRP up-regulation, as well as normal spontaneous Ca²⁺ influx in neurons with dSTIM knock-down supports higher expression of TRP in itpr mutant neurons as a likely cause for higher spontaneous Ca²⁺ influx. Over-expression of TRP thus appears to be sensitive to recurrent loss of IP₃R function rather than loss of SOCE in Drosophila pupal neurons. These data support a central role for the IP₃R in maintaining calcium homeostasis in Drosophila pupal neurons.

Previous results have demonstrated that knock down of either itpr or dSTIM in Drosophila neurons during pupal development leads to flight deficits in adults (Agrawal et al., 2013; Richhariya et al., 2017). Reduced expression of cac in itpr mutant neurons and in neurons with knock-down of dSTIM suggested that cac down regulation may be associated with the observed deficits in Drosophila flight. To test this hypothesis, we measured flight times in animals with knock-down of cac and other subunits of VGCCs in either all or subsets of Drosophila neurons (Figure 4A). Pan-neuronal knock-down of cac (104186) and Ca-α1D (51491) resulted in 100% lethality in late stage 3rd instar larvae and thus these animals could not be tested for flight. Pan-neuronal knock-down of Ca-β (102188) and Ca-α1T (48008, 31961) did not affect either viability or flight. We also tested flight after knock-down of these VGCC subunits in specific neuronal subsets. The RNAi lines used were based on their efficacy as described in published literature (Gu et al., 2009; Iniguez et al., 2013; Kanamori et al., 2013). Knockdowns were performed in dopaminergic, aminergic, glutamatergic, peptidergic and GABA-ergic neuronal subsets, which include neuronal classes in which IP₃R and SOCE function has been previously implicated as required for adult flight (Banerjee et al., 2004; Venkiteswaran and Hasan, 2009; Agrawal and Hasan, 2015; Pathak et al., 2015). Flight times were tested up to 30 s after air-puff stimulation of tethered flight and appeared normal in all the knock down genotypes tested. Thus, the observed down regulation of VGCCs (Figures 3A,D) correlates with reduced SOCE (Figure 2E) but does not affect flight circuit function (Figure 4A). Knock-down of TRP in itpr mutant neurons and its effect on spontaneous Ca²⁺ influx and flight in Drosophila could not be assessed as the recombinant flies with TRP RNAi and itpr mutants didn’t survive.

Recent work investigating the role of intracellular calcium signaling in flight has shown that SOCE is required in dopaminergic neuronal subsets for transcriptional maturation of the Drosophila flight circuit during pupal development (Pathak et al., 2015). Therefore, we tested spontaneous Ca²⁺ influx specifically in dopaminergic neurons of itpr mutant and dSTIM knockdown pupae (Figures 4B,C). As observed in all pan-neuronal populations (Figures 1, 2), spontaneous Ca²⁺ influx in pupal dopaminergic neurons from a mutant IP₃R (S224/G1891S) was significantly higher, whereas SOCE was lower as compared to wild-type dopaminergic neurons (Figure 4C). The IP₃R is required in Drosophila neurons during pupal development for flight (Agrawal et al., 2013). Spontaneous Ca²⁺ signals in Drosophila neurons also originate during late stage pupal development (Jiang et al., 2005). Our data support a model where loss of flight in itpr mutant flies may be an outcome of the up regulation of spontaneous Ca²⁺ influx during pupal development possibly resulting in an imbalance of excitatory and inhibitory neurotransmitters in the developing flight circuit.