Biochemical Journal

Research article

Activation of PLC by an endogenous cytokine (GBP) in Drosophila S3 cells and its application as a model for studying inositol phosphate signalling through ITPK1

Yixing Zhou , Shilan Wu , Huanchen Wang , Yoichi Hayakawa , Gary S. Bird , Stephen B. Shears

Abstract

Using immortalized [3H]inositol-labelled S3 cells, we demonstrated in the present study that various elements of the inositol phosphate signalling cascade are recruited by a Drosophila homologue from a cytokine family of so-called GBPs (growth-blocking peptides). HPLC analysis revealed that dGBP (Drosophila GBP) elevated Ins(1,4,5)P3 levels 9-fold. By using fluorescent Ca2+ probes, we determined that dGBP initially mobilized Ca2+ from intracellular pools; the ensuing depletion of intracellular Ca2+ stores by dGBP subsequently activated a Ca2+ entry pathway. The addition of dsRNA (double-stranded RNA) to knock down expression of the Drosophila Ins(1,4,5)P3 receptor almost completely eliminated mobilization of intracellular Ca2+ stores by dGBP. Taken together, the results of the present study describe a classical activation of PLC (phospholipase C) by dGBP. The peptide also promoted increases in the levels of other inositol phosphates with signalling credentials: Ins(1,3,4,5)P4, Ins(1,4,5,6)P4 and Ins(1,3,4,5,6)P5. These results greatly expand the regulatory repertoire of the dGBP family, and also characterize S3 cells as a model for studying the regulation of inositol phosphate metabolism and signalling by endogenous cell-surface receptors. We therefore created a cell-line (S3ITPK1) in which heterologous expression of human ITPK (inositol tetrakisphosphate kinase) was controlled by an inducible metallothionein promoter. We found that dGBP-stimulated S3ITPK1 cells did not synthesize Ins(3,4,5,6)P4, contradicting a hypothesis that the PLC-coupled phosphotransferase activity of ITPK1 [Ins(1,3,4,5,6)P5+Ins(1,3,4)P3→Ins(3,4,5,6)P4+Ins(1,3,4,6)P4] is driven solely by the laws of mass action [Chamberlain, Qian, Stiles, Cho, Jones, Lesley, Grabau, Shears and Spraggon (2007) J. Biol. Chem. 282, 28117–28125]. This conclusion represents a fundamental breach in our understanding of ITPK1 signalling.

  • calcium
  • Drosophila
  • inositol phosphate
  • insect physiology
  • phospholipase C (PLC)

INTRODUCTION

Stimulus-dependent activation of PLC (phospholipase C) hydrolyses PtdIns(4,5)P2 to generate two intracellular messengers: diacylglycerol which activates protein kinase C, and Ins(1,4,5)P3 which binds to specific receptors {IP3R [Ins(1,4,5)P3 receptor]; itpr in Drosophila} that gate intracellular Ca2+ stores [1]. This is a ubiquitous signalling response that regulates diverse aspects of cellular biology in almost all animal cell types [1]. As such, it is an intensively studied area of cell biology. Additionally, research into the metabolism of Ins(1,4,5)P3 has spurred the development of a largely separate signalling industry [2]: the study of an elaborate network of interconnected metabolites, many of which have multiple biological functions (Figure 1 and [2]). Thus the discovery that PLC is activated by an extracellular agonist endows that agent with a large number of new signalling activities.

Figure 1 The pathways of inositol phosphate metabolism

(A) A depiction of the pathways of inositol phosphate metabolism in animal cells, with a particular emphasis on the multiple contributions from the inositol polyphosphate multikinase, IPMK (reactions involving InsP1, InsP2 and the inositol pyrophosphates [2] are omitted for clarity). The reactions depicted by broken arrows are missing from Drosophila, which does not express a homologue of ITPK1. IP3K, inositol trisphosphate 3-kinase; 5-PASE, inositol polyphosphate 5-phosphatase; PTEN, phosphatase and tensin homologue deleted on chromosome 10. (B) The phosphotransferase activity of ITPK1 (see [25] for details).

Drosophila melanogaster is a genetically-tractable eukaryotic model that has proved useful in unraveling the complexities of many aspects of Ca2+ signalling [3]. For example, experiments with Drosophila mutants have demonstrated that the fly's single itpr gene is important for a range of physiological responses [46]. Cultured Drosophila cells have also been used to study the receptor-dependent activation of Ins(1,4,5)P3 production and Ca2+ mobilization [79]. However, the range of opportunities offered by this model organism have not yet been fully exploited for studies into the function and agonist-regulated metabolism of the other inositol phosphate signals (Figure 1). Such studies, which typically rely on analysing cells that have been radiolabelled with [3H]inositol during several days of proliferation [10], are best suited to immortalized cell lines. Such cells have been derived from Drosophila [11]. However, there is limited insight into the nature of peptides that act through endogenous receptors to activate PLC in these immortalized cells. The application of these cells to inositol phosphate research has therefore been restricted. To bypass this problem, some groups [7,8] have heterologously expressed exogenous receptors into immortalized Drosophila cells. We now describe a more physiologically relevant model that utilizes endogenous receptors: the S3 imaginal-disc cell line [11], in which we show that PLC is activated by a recently discovered insect cytokine, dGBP [Drosophila GBP (growth-blocking peptide)] [12].

dGBP is a member of a large family of insect cytokines [13]. These peptides range from 19–30 amino acid residues in length, and are produced upon proteolytic cleavage of longer prepropeptides [13]. The first GBP to be identified was the factor responsible for the reduction in the rate of growth of lepidopteran larvae following their colonization by the larvae of a parasitic wasp, Costesia kariyai [14,15]. The endocrinological perturbation that up-regulates GBP synthesis in the host stops it from forming the sclerotized pupal cuticle that would otherwise prevent the parasitic larvae from emerging. The GBP family also regulate morphogenesis, cell proliferation and innate immune responses, for example by stimulating plasmatocyte adhesion and spreading [12,13,16]. However, to date little progress has been made in understanding the molecular mechanisms of action of these cytokines [12]. Recently, the discovery of dGBP has revealed that this peptide acts through c-Jun N-terminal kinase to regulate gene expression [12]. However, in view of the multi-functionality of the GBP family (see above), it has remained probable that these peptides recruit additional signalling pathways. In the present study, we add substantially to dGBP's signalling repertoire by demonstrating its ability to activate multiple facets of the PLC-dependent inositol phosphate/Ca2+ cascade.

An increased insight into the molecular actions of insect cytokines such as dGBP is of interest in itself, but the importance of this field of research goes beyond the goal of expanding our understanding of insect physiology [16]. Knowledge of the roles of GBPs in immune responses in pest insects that impair human health or reduce crop yield may lead to the development of improved control programs. There is also considerable evolutionary conservation of innate immunity genes, pathways and effector mechanisms. Research into these defense mechanisms in Drosophila may ultimately improve our understanding of human immune responses. Indeed, the identification of the human homologue of the Toll receptor was initially prompted by the discovery that this protein mediates antifungal immune responses in Drosophila [17].

Another goal of the present study was to determine whether our discovery that dGBP activates PLC in S3 cells could be exploited to explore the signalling activities of inositol phosphate metabolizing enzymes from higher animals. In pursuit of this idea, we noted that the Drosophila genome does not encode a homologue of the mammalian ITPK1 (inositol tetrakisphosphate 1-kinase) [18]. Therefore S3 cells offer a rare example of a model for characterizing any gain-of-function that might arise from the heterologous expression of human ITPK1. For example, the phosphorylation of Ins(1,3,4)P3 by ITPK1 in mammals (Figure 1A) has been proposed to be the rate-limiting step in the synthesis of Ins(1,3,4,5,6)P5 and InsP6 [19]. However, there continue to be differences of opinion [1821] as to the relative importance of this pathway compared with the alternative ITPK1-independent route to Ins(1,3,4,5,6)P5 and InsP6 (Figure 1A). We aimed to address this debate by studying the impact of ITPK1 upon Ins(1,3,4,5,6)P5 and InsP6 synthesis in basal and PLC-activated S3 cells. Furthermore, ITPK1 has additional significance to mammalian cell signalling because this enzyme controls the synthesis of Ins(3,4,5,6)P4 (see Figure 1B), which regulates the conductance of the Cl channel/transporter ClC-3 (chloride channel, voltage-sensitive 3) [22]. In this way, ITPK1 helps regulate a number of physiological processes, such as salt and fluid secretion, insulin secretion, and neurotransmission [2224]. However, it has proved difficult to characterize the mechanism by which Ins(3,4,5,6)P4 synthesis is accelerated following PLC activation. To explain this phenomenon, we have previously put forward a hypothesis that is based on the laws of mass action [25]. We proposed that an elevated rate of phosphorylation of Ins(1,3,4)P3 to Ins(1,3,4,6)P4 by ITPK1 is coupled through an unusual phosphotransferase reaction to an increased rate of dephosphorylation of Ins(1,3,4,5,6)P5 to Ins(3,4,5,6)P4 (Figure 1B) [25]. In the present study, we have tested this hypothesis by determining whether PLC-dependent Ins(3,4,5,6)P4 synthesis could be heterologously introduced into S3 cells by the expression of ITPK1.

MATERIALS AND METHODS

Materials

3H-labelled myo-inositol and InsP6 were purchased from PerkinElmer. [3H]Ins(1,4,5)P3 and [3H]Ins(1,3,4,5)P4 were purchased from American Radiolabeled Chemicals. [3H]Ins(3,4,5,6)P4 was prepared as described previously [27]. 14C-labelled inositol phosphates were HPLC-purified from carbachol-stimulated, [14C]inositol-labelled parotid acinar cells. Ins(1,4,6)P3 was purchased from Cayman Chemicals. Sorbitol dehydrogenase, probenecid and NAD were obtained from Sigma–Aldrich. dGBP was synthesized as described previously [12]. TG (thapsigargin) was purchased from Calbiochem.

Cell culture and stable transfection of S3 cells

Using the EcoRI and XhoI restriction sites, human ITPK1 cDNA was cloned into a Drosophila expression vector, pMT/V5-His (Invitrogen), that includes the metal-inducible metallothionein promoter [26]. Colonies were selected and plasmid DNAs were verified by sequencing. S3 cells (3 ml; 1.5×106 cells/well) were seeded in 6-well dishes in Schneider's Drosophila medium (Invitrogen) supplemented with 10% heat-inactivated fetal bovine serum (Invitrogen), 100 units/ml penicillin and 100 μg/ml streptomycin. After 1 h, cells were transfected (as described in the Effectene Transfection Reagent Handbook; Qiagen) with 1 μg of ITPK1-pMT/V5-His expression vector and 0.05 μg of the resistance plasmid, pCoBlast (Invitrogen). Control cells were transfected with empty vector and the resistance plasmid. Fresh medium was added the following day. At 3 days following transfection, 25 μg/ml blasticidin (Invitrogen) was added to select for stable cell lines, designated S3 and S3ITPK1. Colonies were visible in 2–3 weeks, whereupon cells were then transferred to T-25 flasks. Cultures were maintained at 25°C. Where indicated, the expression of ITPK1 was induced for 2 days with 0.5 mM CuSO4 and the degree of expression was verified by Western blotting analysis [27] of S3 cell lysates, which were prepared in insect cell lysis buffer supplemented with protease inhibitor cocktail (product number 554779; BD Biosciences).

RNA interference

To obtain cDNA fragments of itpr, total RNA was isolated from S3 cells using RNeasy Mini and QIAshredder Kits (Qiagen). Total RNA (1 μg) was reverse transcribed with the following primers: itpr exon 5, forward primer 5′-CCTCAAGCGTTTGCATCATGC-3′ and reverse primer 5′-CTGTTTTCCCTTGGGTTTGTCATTTATG-3′; itpr exons 13–18, forward primer 5′-CTGTTCCTGAATCCGGGCATTC-3′ and reverse primer 5′-GCCACAAACAAGCCCTCCTTGC-3′. RT (reverse transcription)–PCR was performed under the following conditions: 55°C for 1 h and 94°C for 2 min; then 40 cycles of the following: 94°C for 15 s, 55°C for 30 s and 68°C for 2 min. A final incubation at 72°C for 10 min was performed.

dsRNA (double-stranded RNA) was prepared by using a MEGAscript RNAi kit (Ambion). For preparing template DNAs, two primers containing a T7 RNA polymerase promoter sequence at their 5′ ends were used. For cDNA fragment of itpr exon 5 (dsRNA-1; 792 bp), the forward primer was 5′-TAATACGACTCACTATAGGGGGGCACCTCAATCCAATATG-3′ and the reverse primer was 5′-TAATACGACTCACTATAGGGTATGGTGGAGTTCATGGTCG-3′. For cDNA fragment of itpr exon 13–18 (dsRNA-2; 744 bp), the forward primer was 5′-TAATACGACTCACTATAGGGCGGAACGCATTCTGTACAAC-3′ and the reverse primer was 5′-TAATACGACTCACTATAGGGCCGAGGTAGCGCTAACACAAT-3′. For EGFP (enhanced green fluorescent protein) dsRNA (as a control), a PCR product (698 bp) was amplified from a pEGFP vector. The forward primer was 5′-TAATACGACTCACTATAGGGAGAGTGAGCAAGGGCGAGGAG-3′, and the reverse primer was 5′-TAATACGACTCACTATAGGGAGAGATCGCGCTTCTCGTTGG-3′.

S3 cells were washed three times with serum-free medium and then treated with 20 μg/ml dsRNA for 30–60 min. Fresh medium with 10% fetal bovine serum was then added and the cells were further incubated at 25°C for 3 days prior to use. The levels of itpr transcript were determined by RT–PCR (see above) using 2.5 ng of total RNA as a template.

Measurements of inositol phosphates in intact cells

All experiments were performed at 25°C. Both S3 and S3ITPK1 cells (6-well dish; each well contained 5×105 cells in 2 ml) were labelled for 4 days with [3H]myo-inositol (PerkinElmer; 35 μCi per ml). After 2 days, fresh medium and [3H]myo-inositol were added. On day 4, the medium was replaced with a serum-free HBSS (Hepes-buffered salt solution; 120 mM NaCl, 20 mM Hepes, pH 7.4, 5.4 mM KCl, 0.8 mM MgCl2, 10 mM glucose and, unless otherwise stated, 1 mM CaCl2). After a further 1–3 h, LiCl was added to a final concentration of 10 mM for 20 min and then dGBP was added for various times. Assays were quenched by aspiration of medium followed immediately by the addition of 1 ml of ice-cold 0.6 M perchloric acid/0.02 mM InsP6. After 20 min on ice, the soluble material was removed and neutralized with 0.4 ml of 1 M potassium bicarbonate/0.04 M EDTA. Samples were stored at 4°C overnight, centrifuged (10000 g for 5 min at 4°C) to remove insoluble potassium perchlorate, diluted 1:1 with 1 mM EDTA and then analysed using either HPLC or gravity-fed anion-exchange columns [10].

For most HPLC experiments, samples were separated on a 250×4.6 mm Q-100 HPLC column (Thomson Instruments). This is equivalent to the Adsorbosphere SAX columns used in previous studies [28]. The flow-rate was 1 ml/min and the gradient was generated from Buffer A (1 mM EDTA) and Buffer B (1 mM EDTA plus 2 M NH4H2PO4, pH 3.35) in the following manner: 0–10 min, 0% B; 10–110 min, B increased linearly to 63%. Radioactivity was assessed either with an in-line detector (Radiomatic 500TR) or in 1 ml fractions.

In some experiments, samples were separated on a 125×4.6 mm Partisphere SAX column (Krackeler Scientific) at a flow-rate of 1 ml/min on one of two gradients generated from Buffer A (1 mM EDTA) and Buffer C (1.2 M NaH2PO4, pH 4.2). For InsP3 separation: 0–5 min=0% C, 5–76 min C increased linearly to 49% (InsP3 eluted at approximately 65 min). For InsP4 separation: 0–5 min=0% C, 5–125 min C increased linearly to 100% (InsP4 eluted at approximately 90 min). Fractions of 0.8 ml were collected.

Where necessary, inositol phosphates in HPLC fractions were desalted using Sep-Pak Accell Plus QMA columns (Waters) which were eluted with triethylammonium bicarbonate, and then the samples were freeze-dried [29].

Calcium assays

All procedures were performed at 25°C. Changes in intracellular Ca2+ in populations of Fluo-4-loaded S3 cells were monitored with a fluorometric imaging plate reader (FLIPRTETRA Molecular Devices). Approximately 1×105 cells/well were plated in uncoated 96-well plates and incubated for 24 h in 100 μl Schneider's Drosophila medium. Then the medium in each well was exchanged for 100 μl HBSS supplemented with 4 μM Fluo-4/AM (Fluo-4 acetoxymethyl ester; Invitrogen) plus 2.5 mM probenecid. After 60 min, cells were washed twice and then incubated in fresh HBSS either in the absence or presence of extracellular Ca2+ (1 mM) as indicated. Other reagent additions, indicated in the Figure legends, were made using a remotely controlled robotic 96-tip head. The excitation wavelength for Fluo-4 was 488 nm. Fluorescence emission was selected with a 510–570 nm bandpass filter and monitored simultaneously in all wells with a cooled charge-coupled device camera.

Unless otherwise stated, changes in intracellular Ca2+ in individual Fluo-4-loaded S3 cells were monitored in uncoated 48-well plates that were mounted on the stage of a Zeiss LSM 710 confocal microscope equipped with a 10× [0.35 NA (numerical aperture)] objective (approximately 6×105 cells/well; 200 μl). The loading of Fluo-4 and analysis of its fluorescence was as described above, except that the incubation volumes/well were 200 μl. Typically, we calculated the average fluorescence intensity of 30–40 single S3 cells for each condition. The data were then expressed as the ratio F/F0, where F0 is the average basal fluorescence intensity before cell treatment, and F is the measured fluorescence intensity.

In vitro ITPK1 assays

The Ins(3,4,5,6)P4 kinase activity of ITPK1 was assayed in 120 μl of buffer containing 2 M Hepes, pH 8.6, 7 mM ATP, 8.7 mM MgSO4 and 0.7 mM EDTA, 5 μg of recombinant ITPK1 ([27]), and 150 μl HPLC fractions containing putative [3H]Ins(3,4,5,6)P4/[3H]Ins(1,4,5,6)P4. Reactions (4 h at 37°C) were quenched with 0.2 vol. of ice-cold 2 M HClO4 plus 0.1 mg/ml phytate. After 10 min, samples were centrifuged (10000 g for 5 min at 4°C), and the supernatants were removed and neutralized with 0.35 vol. of 1 M K2CO3 plus 40 mM EDTA. Samples were left at 4°C overnight before centrifugation and analysis of the supernatant by HPLC. The assay buffer for the phosphotransferase activity of ITPK1 contained 50 mM KCl, 20 mM Hepes (pH 7.0), 6 mM MgSO4, 5 mM ADP, 0.5 mg/ml of BSA, 5 μM [3H]Ins(1,3,4,5,6)P5, 1 μg of recombinant ITPK1 and, where indicated, 5 μM Ins(1,3,4)P3. Reactions (30 min) were quenched and neutralized as described above. The Ins(1,3,4)P3-stimulated dephosphorylation of Ins(1,3,4,5,6)P5 to Ins(3,4,5,6)P4 (i.e. the phosphotransferase activity, see Figure 1B and [25]) was recorded using gravity-fed anion-exchange columns [10].

Structural analysis of InsP3

The treatment of [3H]InsP3 by periodate oxidation, reduction with sodium borohydride and dephosphorylation with alkaline phosphatase was as described in [10]. The recovery of 3H through each of these procedures was not significantly different from that observed when standards of [3H]Ins(1,4,5)P3 were processed in parallel. Thus we conclude that none of our samples contained significant quantities of [3H]Ins(4,5,6)P3 (which would lose the 2-[3H] label during the process [10]). Polyols were resolved by chromatography on a CarboSep CHO-682 HPLC column (Pb form, 300×7.8 mm coupled to a 50×4.6 mm guard column; Transgenomics). Water was used as the eluate, the flow rate was 0.3 ml/min, and the column temperature was 60°C. The polyol standards that were used to calibrate the column were all obtained from Sigma–Aldrich, except altritol (Tokyo Chemistry Industry) and D-[3H]iditol, which was synthesized by periodate oxidation, reduction and dephosphorylation of [3H]Ins(1,4,5)P3 [10].

RESULTS

A Drosophila cytokine, dGBP, activates PLC in S3 cells

Immortalized Drosophila cells have been a relatively untapped resource for studying the regulation of inositol phosphate metabolism by agonist-dependent activation. In the present study we found that the S3 imaginal-disc cell line [11] incorporated radiolabelled [3H]inositol into a variety of inositol phosphates, which were resolved by HPLC (Figure 2A; the elution positions of isomers of InsP1, InsP2, InsP3, InsP4, InsP5 and InsP6 are labelled 1–6). The levels of Ins(1,3,4,5,6)P5 relative to InsP6 in S3 cells are relatively low compared with those normally observed in mammalian cells (e.g. Supplementary Figure S1 at http://www.BiochemJ.org/bj/448/bj4480273add.htm). The significance of that comparison is discussed below.

Figure 2 The effect of dGBP upon the inositol phosphate profile of S3 cells

(A and B) [3H]inositol-labelled S3 cells were incubated with either vehicle (A) or 0.5 μM dGBP (B) for 2 min and then the samples were quenched, neutralized, separated on a Q100 HPLC column and radioactivity was measured in-line. The numbers that annotate each peak depict the various inositol phosphate species (1, InsP1; 2, InsP2 etc) as determined with appropriate standards (the present study and [28,49]). Results are from representative experiments (one of four). (C) A dose–response relationship between the degree of PLC activation and dGBP concentration (means±S.E.M. from three experiments). The time of stimulation was 0.5 min, and the results were obtained using gravity-fed anion-exchange columns.

We challenged S3 cells with a newly discovered Drosophila cytokine, dGBP [12]. The peptide promoted increases in the levels of InsP1, InsP2, InsP3 and InsP4 (Figures 2A and 2B) that epitomize PLC activation. These results (Figure 2) provide the first demonstration that dGBP activates various aspects of the PLC-dependent inositol phosphate signalling cascade. The levels of two peaks of InsP3 were elevated by dGBP (Figures 2B, 3A and 3B). These were identified as Ins(1,3,4)P3 and Ins(1,4,5)P3 respectively, on the basis of their co-elution with the standards of these compounds (Supplementary Figures S2A–S2C at http://www.BiochemJ.org/bj/448/bj4480273add.htm), and also by direct structural analysis (Supplementary Figure S2D and Supplementary Table S1 at http://www.BiochemJ.org/bj/448/bj4480273add.htm). Ins(1,3,4,5)P4 is the metabolic intermediate between Ins(1,4,5)P3 and Ins(1,3,4)P3 (Figure 1A). Peak 4a (Figure 2B) was identified as Ins(1,3,4,5)P4 on the basis of its elution characteristics on two different HPLC systems (Supplementary Figure S3 at http://www.BiochemJ.org/bj/448/bj4480273add.htm). Peak 4b, which we identified as Ins(1,4,5,6)P4 (Supplementary Figure S3B), was also elevated following addition of dGBP (Figures 2A and 2B). There is current interest in understanding the mechanisms of the receptor-dependent metabolism of Ins(1,4,5,6)P4, because of evidence that this particular inositol phosphate is a transcriptional regulator [30,31]. S3 cells may be a useful model for that field of research. dGBP also caused InsP5 levels to increase by approximately 70% (control=1.7±0.18; 2 min stimulation with dGBP=3.0±0.33; P<0.02).

Note that the height of the dGBP-stimulated Ins(1,4,5)P3 peak in S3 cells is small compared with the quantities of the Ins(1,4,5)P3 metabolites, particularly InsP2 and InsP1 (Figure 2B). These results suggest a considerable metabolic flux through Ins(1,4,5)P3, as is typically observed in mammalian cells following agonist-dependent activation of PLC. The degree of the stimulation of PLC by dGBP was also typical of that seen following the activation of endogenous receptors in mammalian cells (Supplementary Figure S1 and [32]). From a dose–response curve for 30 s activation of PLC by dGBP, an EC50 value of 0.2 μM was obtained (Figure 2C).

Mobilization of intracellular Ca2+ by dGBP

Ins(1,4,5)P3-dependent Ca2+ mobilization is a major consequence of PLC activation [1]. When S3 cells were incubated in medium containing 1 mM Ca2+, the addition of dGBP produced a relatively sustained dose-dependent increase in intracellular Ca2+; the EC50 value was 0.05 μM (Figures 3A and 3B). The EC50 value for PLC activation was 4-fold higher (Figure 2C), but differences of this magnitude are not uncommon when comparing dose/response curves data for these two assays (e.g. [33]). Through a phenomenon termed ‘receptor-reserve’ [34], a full Ca2+ response may be achieved with less than maximal PLC activation [35].

Figure 3 Effects of dGBP peptide and TG on calcium mobilization in S3 cells

Changes in intracellular Ca2+ concentration were monitored using Fluo-4 on a fluorometric imaging plate reader (FLIPRTETRA). (A) Wild-type S3 cells were incubated in medium containing 1 mM Ca2+ and were stimulated with the following concentrations of dGBP: 0.003 μM (a), 0.01 μM (b), 0.03 μM (c), 0.1 μM (d), 0.3 μM (e), 0.5 μM (f) or 1 μM (g). Peak [Ca2+] signals were used to generate the dose–response relationship in (B). (C) Cells were initially incubated in nominally Ca2+-free medium and (where indicated by the arrow) either vehicle (dotted line), 0.5 μM dGBP (continuous line) or 2 μM TG (broken line) were added. Where indicated, 1 mM Ca2+ was added to the medium. (D) The increase in intracellular [Ca2+] in S3 cells incubated in medium containing 1 mM [Ca2+]; either 0.5 μM dGBP (continuous line) or 2 μM TG (broken line) were added (as indicated by the arrow). (E) Cells were incubated in nominally Ca2+-free medium and 2 μM TG and 0.5 μM dGBP were added where indicated. (F) S3 cells were incubated in nominally Ca2+-free medium and the peak [Ca2+] signal in response to dGBP was recorded in S3 cells that were pre-treated with either vehicle (left-hand bar) or 2 μM TG (right-hand bar) as in (E). The results are either representative experiments (one of three) or, in (B) and (F), means±S.E.M. from three experiments.

When S3 cells were incubated in nominally Ca2+-free medium, a rapid increase in intracellular [Ca2+] was still observed after the addition of 0.5 μM dGBP (Figure 3C), but now the Ca2+ response was transient (Figure 3C) instead of being sustained (Figure 3A). These results provide a classical demonstration that dGBP initially mobilizes Ca2+ from intracellular stores. Nevertheless, the activation of a Ca2+ entry pathway following dGBP-dependent store depletion was observed when extracellular Ca2+ was restored, resulting in a sustained increase in intracellular [Ca2+] (Figure 3C). Thus dGBP promotes a biphasic Ca2+ response that is typically observed following the activation of PLC-coupled receptors: depletion of an intracellular Ca2+ store by Ins(1,4,5)P3, which then drives enhanced Ca2+ entry [1,36].

In cells derived from mammals and Drosophila, Ca2+ signals can also be generated by the addition of TG to inhibit ATP-driven Ca2+ uptake by the endoplasmic reticulum [9]; we reproduced these observations in S3 cells (Figures 3C and 3D). In the absence of extracellular Ca2+, the depletion of intracellular Ca2+ stores by TG was sufficient to near-completely inhibit the ability of dGBP to mobilize Ca2+ (Figures 3E and 3F). It is typical for the Ins(1,4,5)P3-regulated Ca2+ store to overlap with the TG-releasable Ca2+ pool [36].

Comparison of the effects of itpr knockdown by dsRNA on dGBP- and TG-induced Ca2+ mobilization in S3 cells

To explore further the participation of PLC-dependent Ins(1,4,5)P3-mediated Ca2+ mobilization in the dGBP response, we knocked down expression of the Drosophila Ins(1,4,5)P3 receptor itpr in S3 cells by the addition of dsRNA. Two different dsRNAs gave similar results (results not shown); the effects that were obtained with dsRNA-1 (see the Materials and methods section) are shown in Figure 4.

Figure 4 Effects of itpr knockdown by dsRNA on dGBP- and TG-induced calcium mobilization in S3 cells

(A) RT–PCR analysis of control or dsRNA-treated cells using gene-specific primers for itpr. (B) dGBP-stimulated PLC activity of control and dsRNA-treated cells. (CE) Changes in intracellular Ca2+ were monitored in single Fluo-4-loaded S3 cells seeded in uncoated 48-well plates mounted on the stage of a Zeiss LSM 710 confocal microscope. Cells were incubated either in either nominally Ca2+-free medium (D) or medium containing 1 mM Ca2+ (C and E). Either 2 μM TG or 0.5 μM dGBP was added as indicated. The results in (BE) are the means of three experiments.

The degree of expression of itpr was substantially reduced by 3 days of treatment with dsRNA (Figure 4A). There was no effect of the dsRNA upon the degree of dGBP-stimulated PLC activity (Figure 4B). The dsRNA also did not alter the size of the cell's Ca2+ stores, since there was no effect upon the degree of Ca2+ mobilization that was induced by either TG (Figure 4C) or ionomycin (results not shown). In contrast, itpr knockdown almost completely reduced the ability of dGBP to mobilize intracellular Ca2+, both in the absence (Figure 4D) or presence (Figure 4E) of extracellular Ca2+. These results confirm the involvement of itpr, and hence PLC-dependent Ins(1,4,5)P3 release, in the dGBP-dependent Ca2+ responses described above (and see Figure 3).

The effects of exogenous ITPK1 upon the inositol phosphate profile of unstimulated S3 cells

The heterologous expression of mammalian inositol phosphate kinases in lower eukaryotes has previously proven to be a productive experimental strategy for uncovering new aspects of inositol phosphate metabolism and signalling [37]. The use of an itpk1-null cell line as a host for human ITPK1 offers a gain-of-function strategy for studying the roles of this particular kinase. For example, we wished to address some of the uncertainties surrounding the relative importance of ITPK1 in the pathway of Ins(1,3,4,5,6)P5 and InsP6 synthesis [1821].

We generated a stable S3 cell line (S3ITPK1) carrying a plasmid encoding the human ITPK1 gene under the control of the inducible metallothionein promoter. Optimal expression was achieved 48 h post-induction (Figure 5A). The expression of ITPK1 increased InsP6 levels by approximately 70%, but Ins(1,3,4,5,6)P5 levels increased approximately 7-fold (Figures 5B and 5C). There was a corresponding increase in the levels of Ins(1,4,5,6)P4 (Figure 5C), another metabolite of Ins(1,3,4,5,6)P5 (Figure 1A). These results demonstrate that ITPK1 can augment the synthesis of these particular inositol phosphates in vivo. Nevertheless, the ratio of Ins(1,3,4,5,6)P5/InsP6 in S3ITPK1 cells was still substantially below that usually observed in mammalian cells (e.g. Supplementary Figure S1 and [21,32,38]).

Figure 5 The inositol phosphate profile in S3ITPK1 cells

(A) Western blotting analysis of ITPK1 in wild-type S3 and S3ITPK1 cells. Molecular mass is given on the left-hand side in kDa. (B) [3H]inositol-labelled S3ITPK1 cells were incubated in serum-free medium for 2 h, and then quenched, neutralized, separated on a Q100 HPLC column, and radioactivity was measured in-line. The numbers that annotate each peak depict the various inositol phosphate species (1, InsP1; 2, InsP2 etc). (C) A comparison of mean levels of Ins(1,4,5,6)P4, Ins(1,3,4,5,6)P5 and InsP6 in wild-type S3 with S3ITPK1 cells (*P<0.001). Data are either representative experiments (one of four) or means±S.E.M. from four experiments.

Does ITPK1 expression in PLC-activated S3 cells introduce a gain-of-function of Ins(3,4,5,6)P4 synthesis?

Ins(1,4,5)P3 is but one of several functionally significant inositol phosphates [2]. Among these, Ins(3,4,5,6)P4 is a regulator of certain aspects of cell physiology through its control of a Cl channel conductance [2224]. We have proposed that, during PLC activation, an increased rate of phosphorylation of Ins(1,3,4)P3 to Ins(1,3,4,6)P4 by ITPK1 is coupled through a phosphotransferase reaction to an increased rate of dephosphorylation of Ins(1,3,4,5,6)P5 to Ins(3,4,5,6)P4 (Figure 1B) [25]. Thus one of the goals of the present study was to determine whether PLC-dependent Ins(3,4,5,6)P4 synthesis would be a gain-of-function following heterologous expression of human ITPK1 in the itpk1-null S3 cells.

We first checked whether the expression of ITPK1 might result in an any unexpected modification of PLC activity, which is normally recorded by measuring agonist-dependent accumulation of InsP(1–4) [10]. That total includes the relatively large quantities of an InsP3 isomer that accumulated in dGBP-stimulated S3ITPK1 cells (peak 3x in Figure 6), but not in wild-type S3 cells (Figures 2A and 2B). Structural analysis (Supplementary Figures S4 and S5 at http://www.BiochemJ.org/bj/448/bj4480273add.htm) identified peak 3x as Ins(1,4,6)P3. The latter is not a novel isomer; Ins(1,4,6)P3 has previously been identified in mammalian cells, and its levels also increase somewhat upon receptor activation [32]. However, the high levels of Ins(1,4,6)P3 that we observed following PLC activation in S3ITPK1 cells are unprecedented. There are no known kinases that can synthesize Ins(1,4,6)P3 from an InsP2, but either Ins(1,3,4,6)P4 or Ins(1,4,5,6)P4 can be precursors [39,40]. Thus Ins(1,4,6)P3 was included in our assessment of PLC activity. As explained below, we also deemed that our estimate of dGBP-dependent PLC activity (InsPn in Figure 7A) should include the synthesis of Ins(1,2,3,4,6)P5.

Figure 6 The inositol phosphate profile in dGBP-stimulated S3ITPK1 cells

(A) [3H]inositol-labelled S3 cells were incubated with 0.5 μM dGBP for 2 min and then the samples were quenched, neutralized and separated on a Q100 HPLC column, and radioactivity was measured in-line. The numbers that annotate each peak depict the various inositol phosphate species (1, InsP1; 2, InsP2 etc). (B) The InsP3 region of the chromatograph (continuous line) was compared with the elution of a [3H]Ins(1,4,5)P3 standard from a parallel run (broken line). Peak 3x was identified as Ins(1,4,6)P3 (see the text for details).

Figure 7 Time course of inositol phosphate metabolism in dGBP-stimulated S3 and S3ITPK1 cells

[3H]inositol-labelled wild-type S3 cells (●) or S3ITPK1 cells (▼) were incubated with 0.5 μM dGBP for the indicated times and then the samples were quenched, neutralized and separated on a Q100 HPLC column and radioactivity was measured in-line. (A) ‘InsPn’ is used as an index of PLC activity and refers to InsP1–4 [41,50] plus Ins(1,2,3,4,6)P5 (see the text for details). (BE) Levels of individual [3H]inositol phosphates. Time-course data for Ins(1,3,4,5)P4 in wild-type S3 cells are presented in Supplementary Figure S3(D) (at http://www.BiochemJ.org/bj/448/bj4480273add.htm). The results are the means±S.E.M. from four independent experiments. Note that the dGBP-mediated increase in Ins(1,3,4,5,6)P5 levels in wild-type cells is statistically significant (see the text for details).

In both S3 and S3ITPK1 cells, the addition of dGBP stimulated an initial rapid phase of PLC activity that was largely completed within 2 min, after which the levels of inositol phosphates first plateaued and then decreased somewhat (Figure 7A). That transition of the rapid initial phase of PLC activity to a slower secondary rate is a typical manifestation of receptor desensitization, which is important in regulating the duration of an agonist's actions [41]. For example, such desensitization contributes to a largely transient Ins(1,4,5)P3 response [41]. That was the case in our experiments: a rapid initial 9-fold increase in Ins(1,4,5)P3 concentration immediately following agonist addition, followed by a slower drop in levels (Figure 7B).

The value of dGBP-stimulated InsPn was somewhat higher in S3ITPK1 cells than in wild-type S3 cells (Figure 7A). However, the presence of Li+ in these assays does not completely prevent the loss of inositol phosphates by their dephosphorylation to inositol [10]. That loss is probably smaller in S3ITPK1 cells, in which a larger proportion of total Ins(1,4,5)P3 metabolites were ‘preserved’ for the PLC assay by their conversion into Ins(1,4,5,6)P4 (Figure 7C), Ins(1,3,4,6)P4 (Figure 7D and Supplementary Figure S6 at http://www.BiochemJ.org/bj/448/bj4480273add.htm), Ins(1,2,3,4,6)P5 {presumably produced by off-target IP5K (inositol pentakisphosphate 2-kinase)-mediated phosphorylation of Ins(1,3,4,6)P4 [42]; Figure 7E} and Ins(1,4,6)P3 (Figure 7F). Indeed, the levels of those particular metabolites, when added together, near-quantitatively account for the difference in the estimated PLC activity between the two cell types (Figure 7A). Thus we conclude that PLC activity was not greatly modified by ITPK1 expression. There were also no significant differences in the degree of dGBP-mediated Ca2+ mobilization between wild-type S3 and S3ITPK1 cells (Supplementary Figure S7 at http://www.BiochemJ.org/bj/448/bj4480273add.htm).

We now return to our goal of determining whether, during PLC activation, phosphorylation of Ins(1,3,4)P3 to Ins(1,3,4,6)P4 by ITPK1 is coupled through a phosphotransferase reaction to an increased rate of dephosphorylation of Ins(1,3,4,5,6)P5 to Ins(3,4,5,6)P4 (Figure 1B and [25]). Ins(1,3,4,6)P4 was not detected in wild-type S3 cells, but the addition of dGBP to S3ITPK1 cells led to a substantial accumulation of Ins(1,3,4,6)P4 (Figure 7D and Supplementary Figure S6B). These results confirm that the ITPK1 was catalytically active and had access to a substantial supply of Ins(1,3,4)P3. Furthermore, these results indicate that the flux through the Ins(1,4,5)P3 kinase/Ins(1,3,4,5)P4 5-phosphatase pathway (Figure 1A) in dGBP-stimulated S3 cells was much greater than was evident from the low levels of Ins(1,3,4)P3 (Figure 7G).

If Ins(3,4,5,6)P4 were to have been produced, it would have co-eluted with its enantiomer, Ins(1,4,5,6)P4, in peak 4b (Figure 6A). To distinguish between these enantiomers, we incubated aliquots of peak 4b with human ITPK1, which stereoselectively phosphorylates Ins(3,4,5,6)P4, but not Ins(1,4,5,6)P4 [43]. We found that there was no phosphorylation of peak 4b by ITPK1 [Figure 8A, closed circles; we estimated that the limits of detection of Ins(3,4,5,6)P4 in this assay were 1% of the total peak]. For a positive control, we added [3H]Ins(3,4,5,6)P4, and found that 80% of that was phosphorylated by ITPK1 to Ins(1,3,4,5,6)P5 (Figure 8A, open circles). That is, peak 4b in S3ITPK1 was determined to be Ins(1,4,5,6)P4. It seems unlikely that Ins(3,4,5,6)P4 was formed in S3ITPK1 cells but then rapidly metabolized; when [3H]Ins(3,4,5,6)P4 was incubated at 25°C with S3ITPK cell lysates, its rate of dephosphorylation was <0.03%/μg/h (results not shown). Indeed, in mammalian cells Ins(3,4,5,6)P4 is one of the more metabolically resistant inositol phosphates [23,38].

Figure 8 Enzymatic analysis of the proportions of Ins(3,4,5,6)P4 and Ins(1,4,5,6)P4 in dGBP-stimulated S3ITPK1 cells and UTP-stimulated BHKITPK1 cells

(A) Aliquots of the 3H-labelled peak 4b from S3 cells stimulated with dGBP for 5 min (Figure 2B) (●) or an [3H]Ins(3,4,5,6)P4 standard (○) were incubated with ITPK1 as described in the Materials and methods section. Samples were quenched, neutralized and separated on a Partisphere SAX column. Note that none of peak 4b was phosphorylated. Similar data were obtained with aliquots of peak 4b isolated from S3 cells stimulated with dGBP for 15 min (results not shown). (B) Aliquots of the 3H-labelled peak 4b from BHKITPK1 cells stimulated for 5 min with UTP (Supplementary Figure S1 at http://www.BiochemJ.org/bj/448/bj4480273add.htm) were incubated in either the presence (●) or absence (○) of ITPK1 as described in the Materials and methods section.

In separate control experiments, we incubated ITPK1 with aliquots of the corresponding peak 4b from UTP-stimulated BHK (baby hamster kidney) cells (Supplementary Figure S1); virtually all of that peak [equivalent to approximately 10% of the Ins(1,3,4,5,6)P5 peak] was found to be Ins(3,4,5,6)P4 (Figure 7B). This comparison suggests that the mechanism by which ITPK1 synthesizes Ins(3,4,5,6)P4 in mammalian cells is not reconstituted in S3 cells simply by the expression of ITPK1. That conclusion is inconsistent with our hypothesis that mass-action effects govern the phosphotransferase activity of ITPK1 [25].

One possible consequence of an elevated level of Ins(1,4,6)P3 in S3ITPK1 cells (Figure 7F) could be to inhibit the Ins(1,4,5)P3 5-phosphatase [44]. Indeed, Ins(1,4,5)P3 levels were elevated in dGBP-stimulated cells when ITPK1 was expressed (Figure 7B). We therefore next investigated whether Ins(1,4,6)P3 might also inhibit the phosphotransferase activity of ITPK1, as such an effect was a potential explanation for the absence of Ins(3,4,5,6)P4 in these cells. We assayed the phosphotransferase activity in vitro as previously described [25], that is, from the ability of Ins(1,3,4)P3 to stimulate ITPK1-catalysed dephosphorylation of Ins(1,3,4,5,6)P5 into Ins(3,4,5,6)P4 in vitro (Supplementary Figure S8 at http://www.BiochemJ.org/bj/448/bj4480273add.htm). We found that Ins(1,4,6)P3 did not significantly affect phosphotransferase activity until its concentration was 10-fold greater than that of the Ins(1,3,4,5,6)P5 substrate (their maximum concentration ratio in intact cells; see below), and then only a 10% inhibition was observed (Supplementary Figure S8). Thus the presence of Ins(1,4,6)P3 does not obscure our conclusion that the expression of ITPK1 in S3 cells was insufficient to introduce PLC-coupled Ins(3,4,5,6)P4 synthesis.

DISCUSSION

A central conclusion in the present study is that PLC is activated by dGBP, a member of a multi-functional family of insect cytokines [13]. The Drosophila homologue of the GBP family was identified only quite recently, and it has been shown to stimulate transcription of antimicrobial peptides by activation of c-Jun NH2-terminal kinase [12]. However, the GBP family is known to regulate other diverse aspects of insect cell physiology, including larval growth, morphogenesis and cell proliferation [1216]. Therefore it was our goal to search for additional signalling pathways that dGBP might recruit, in order to gain a more complete understanding of the mechanisms that underlie the various actions of these cytokines. Our demonstration that dGBP activates PLC in S3 cells opens up a number of directions for further research into insect physiology. PLC-mediated hydrolysis of PtdIns(4,5)P2 yields diacylglycerol, which activates protein kinase C, and Ins(1,4,5)P3, which binds itpr and gates intracellular Ca2+ stores [1]. Furthermore, as shown in Figure 7, dGBP also promotes increases in a number of other inositol phosphates, several of which have their own distinct cellular functions [2].

We used several experimental approaches to ascertain that, in S3 cells, dGBP mobilizes Ca2+ from intracellular Ins(1,4,5)P3-releasable stores as a consequence of PLC activation. First, we used HPLC (Figures 2 and 7) and chemical analyses (Supplementary Figure S2) to determine that dGBP promotes the increases in Ins(1,4,5)P3 and related metabolites that are typical of agonist-dependent stimulation of PLC activity in mammalian cells (e.g. Supplementary Figure S1). Secondly, we used fluorescent Ca2+ probes to demonstrate that dGBP promotes the biphasic Ca2+ response that is typically observed following the activation of PLC-coupled receptors [1,36]: depletion of intracellular Ca2+ stores, which then drives enhanced Ca2+ entry (Figure 3). Thirdly, we used dsRNA to knock down itpr expression, whereupon the ability of dGBP to mobilize intracellular Ca2+ stores was almost completely eliminated (Figure 4).

The various elements of the inositol phosphate cascade that are activated by dGBP (Figure 7) have independent functions [2], and so may contribute to the peptide's multi-functionality. The Ins(1,4,5)P3-mediated mobilization of Ca2+ (Figure 3) probably mediates several physiological responses [1]. The dGBP-mediated increases in Ins(1,3,4,5)P4 levels (Supplementary Figure S3D) may also have some signalling significance [45]. Changes in Ins(1,4,5,6)P4 levels (Figure 7C) also promise to be important, in the light of evidence that Ins(1,4,5,6)P4 regulates gene transcription [30,31]. Ins(1,3,4,5,6)P5 levels in wild-type S3 cells were also elevated by 70% within 2 min of dGBP addition (Figure 7H and see the Results section). Such a sizable Ins(1,3,4,5,6)P5 response to PLC activation is not usually observed in mammalian cells (e.g. Supplementary Figure S1H). However, in the murine F9 teratocarcinoma cell line, PLC activation has been associated with large increases in Ins(1,3,4,5,6)P5 levels, which regulate key aspects of the β-catenin signalling pathway [20]. In F9 cells, Ins(1,3,4,5,6)P5 synthesis occurs primarily through direct phosphorylation of Ins(1,4,5)P3 by IPMK (inositol polyphosphate multikinase) [20]. This is also the case in Drosophila (Figure 1A and [18]). Thus S3 cells may be of value for further exploring these signalling functions for Ins(1,3,4,5,6)P5 and their contribution to the physiology of dGBP.

The genetic tractability of Drosophila S3 cells is also an advantage when using them to explore the mechanisms of agonist-dependent regulation of the metabolism and functions of the inositol phosphate family. In the present study, we exploited the observation [18] that the Drosophila genome is itpk1-null, and so we heterologously expressed human ITPK1 in S3 cells in order to gain insight into the signalling functions of that particular kinase. These experiments yielded three significant new conclusions with regards to the actions of ITPK1.

First, we obtained data that are relevant to the ongoing debate concerning the degree to which ITPK1 contributes to Ins(1,3,4,5,6)P5 synthesis in mammalian cells [1821,46]. Specifically, we found that the expression of ITPK1 in S3 cells yielded several-fold increases in the steady-state levels of Ins(1,3,4,5,6)P5 and one of its metabolites, Ins(1,4,5,6)P4 (Figure 5C). Nevertheless, in both wild-type S3 and S3ITPK1 cells, the ratio of Ins(1,3,4,5,6)P5 to InsP6 (Figures 5C, 7H and 7I) was relatively low compared with that in mammalian cells (e.g. Supplementary Figure S1 and [21,32,38]). These results suggest that factors other than ITPK1 expression itself contribute to the synthesis of the relatively large quantities of Ins(1,3,4,5,6)P5 that are typically present in mammalian cells. For example, there is evidence that much of the mammalian cell's Ins(1,3,4,5,6)P5 resides in a metabolically-resistant pool [47]. The mechanism by which Ins(1,3,4,5,6)P5 transfers into that pool in mammalian cells may be key to the accumulation of that inositol phosphate at relatively high levels.

Secondly, the sizable accumulation of Ins(1,2,3,4,6)P5 in PLC-activated S3ITPK1 cells (Figure 7E) is also of interest. We attributed that result to the relatively high levels of Ins(1,3,4,6)P4 (Figure 7D) successfully out-competing the lower levels of Ins(1,3,4,5,6)P5 (Figure 7H) for phosphorylation by IP5K; such competition has previously been shown in vitro [42]. Such a phenomenon could help explain a previously puzzling observation [21] that the InsP6 pool in rat-1 cells was largely unaffected when Ins(1,3,4,5,6)P5 synthesis was compromised by knockdown of IPMK expression. An ITPK1-dependent alternative pathway of InsP6 synthesis in rat-1 in cells could solve that problem: phosphorylation of Ins(1,3,4)P3 into Ins(1,3,4,6)P4 (Figure 7D), then into Ins(1,2,3,4,6)P5 (Figure 7E), which then can be converted into InsP6 [48].

The third, and arguably the most important, observation to emerge from our studies with ITPK1, was the discovery that the phosphotransferase activity of ITPK1 is not sufficient by itself to recapitulate PLC-dependent synthesis of Ins(3,4,5,6)P4, an intracellular signal that serves multiple biological roles through its regulation of Cl channel activity [22]. Thus our results of the present study counter a previous hypothesis [25] that Ins(3,4,5,6)P4 levels are solely regulated by the mass-action effects of substrate supply that are exerted upon ITPK1 (see Figure 1B). This outcome was unexpected, because previous structural and metabolic data [25] had indicated that the phosphotransferase activity of human ITPK1 is inherently well-adapted to regulation by mass-action effects in vitro. Perhaps the absence of Ins(3,4,5,6)P4 in PLC-activated S3ITPK1 cells (Figure 8A) is because metabolic compartmentalization prevents ITPK1 from accessing Ins(1,3,4,5,6)P5. It is also possible that there is a yet to be discovered mechanism, not active in S3 cells, by which the phosphotransferase activity of ITPK1 might be activated relative to its kinase activity. A potential mechanism might involve covalent modification of ITPK1, such as by its phosphorylation or acetylation [46]. In any case, the present study indicates that the regulation of the Ins(3,4,5,6)P4 signalling cascade is more complex than hitherto was appreciated. As the number of physiological responses that are regulated by Ins(3,4,5,6)P4 increases [22], so the need to unravel these mechanisms becomes more urgent.

In previous work, the heterologous expression in yeast of inositol phosphate kinases of animal and plant origin uncovered new aspects of inositol phosphate metabolism and signalling [37]. However, the use of immortalized Drosophila cells as a host offers several advantages over the yeast model, including their more direct relevance to mammalian systems: unlike yeasts, flies have an Ins(1,4,5)P3-releasable Ca2+ pool [46]. Genetic manipulation by dsRNA is also cheap and effective in cultured fly cells (e.g. Figure 4). Furthermore, our demonstration in the present study that dGBP stimulates PLC adds value to S3 cells as a resource for further studies into receptor-dependent regulation of inositol phosphate function and metabolism. Our work with ITPK1 offers an example of how that model can usefully be exploited. Finally, we add substantially to the repertoire of the GBP family by demonstrating its ability to activate multiple facets of the PLC-dependent inositol phosphate signalling cascade.

AUTHOR CONTRIBUTION

Yoichi Hayakawa, Gary Bird and Stephen Shears designed the study. Yoichi Hayakawa synthesized the dGBP peptide. Yixing Zhou, Shilan Wu, Huanchen Wang, Gary Bird and Stephen Shears performed experiments. All authors contributed to writing the paper.

FUNDING

This work was supported by the Intramural Research Program of the National Institutes of Health/National Institute of Environmental Health Sciences.

Acknowledgments

We thank Dr X. Qian and Dr J. Mitchell for assistance, and Dr J. Putney and Dr S. Hicks for helpful comments during the writing of the paper.

Abbreviations: BHK, baby hamster kidney; dGBP, Drosophila growth-blocking peptide; dsRNA, double-stranded RNA; GBP, growth-blocking peptide; HBSS, Hepes-buffered salt solution; IPMK, inositol polyphosphate multikinase; IP5K, inositol pentakisphosphate 2-kinase; ITPK1, inositol tetrakisphosphate 1-kinase; itpr, Drosophila Ins(1,4,5)P3 receptor; PLC, phospholipase C; RT, reverse transcription; TG, thapsigargin

References

View Abstract