Abstract
The Mammalian phosphatase of regenerating liver (PRL) family is primarily recognized for its oncogenic properties. Here we found that in Drosophila, loss of prl-1 resulted in CO2-induced brain disorder presented as irreversible wing hold up with enhancement of Ca2+ responses at the neuron synaptic terminals. Overexpression of Prl-1 in the nervous system could rescue the mutant phenotype. We show that Prl-1 is particularly expressed in CO2-responsive neural circuit and the higher brain centers. Ablation of the CO2 olfactory receptor, Gr21a, suppressed the mutant phenotype, suggesting that CO2 acts as a neuropathological substrate in absence of Prl-1. Further studies found that the wing hold up is an obvious consequence upon knockdown of Uex, a magnesium transporter, which directly interacts with Prl-1. Conditional expression of Uex in the nervous system could rescue the phenotype of prl-1 mutants. We demonstrate that Uex acts genetically downstream of Prl-1. Our findings provide important insights into mechanisms of Prl-1 protection against olfactory CO2 stimulation induced brain disorder at the level of detailed neural circuits and functional molecular connections.
Background
Drosophila phosphatase of regenerating liver-1(prl-1) is the only homologous gene of the mammalian PRL family (including PRL-1, PRL-2 and PRL-3) that belongs to the smallest class (molecular masses of 20-22kDa) of protein tyrosine phosphatases (PTPs) (Diamond, Cressman et al., 1994, Zeng, Hong et al., 1998). In recent years, much attention has been paid to PRLs due to their implication in various cancers (Bessette, Qiu et al., 2008, Rios, Li et al., 2013, Saha, Bardelli et al., 2001, Zeng, Dong et al., 2003). The highly expressing PRL-3 has almost been exclusively and particularly correlated to disease aggressiveness and clinical outcome for such multiple tumor types as colorectal(Molleví, Aytes et al., 2008, Xing, Peng et al., 2009), ovarian(Polato, Codegoni et al., 2005, Ren, Jiang et al., 2009), breast(Wang, Peng et al., 2006) and gastric(Wang, Cai et al., 2009) cancers. Analysis of PRL-1, PRL-2 mRNA in a large number of variant human tissue specimens revealed its significant overexpression in hepatocellular and gastric carcinomas, but significant underexpression in ovarian, breast, kidney carcinomas (Dumaual, Sandusky et al., 2012). Variant expression was also noted in other non-cancer tissue types(Dumaual, Sandusky et al., 2006). These contradictory results suggested high tissue-specific and a pleiotropic role for PRL in the diseases process. Nevertheless, neither the mechanism nor the regulation of PRL have been clarified, the exact biological function of these PRL molecules remains currently unknown and a clear mutant background of PRL has yet to be established in any genetic model.
Characterization of the Prl family members has been conducted for early embryos of Drosophila, amphioxus and zebrafish, where Prls were seen to be uniformly expressed in the central nervous system (CNS) and suggested to be likely involved in early neural development (Lin, Lee et al., 2013). In a mouse model PRL-2 was noted as ubiquitously expressed, particularly in the hippocampal pyramidal neurons, ependymal cells, cone and rod photoreceptor cells (Gungabeesoon, Tremblay et al., 2016). We initially investigated the function of Drosophila Prl-1 by generation of prl-1 mutant using CRISPR/Cas9 method (Bassett & Liu, 2014, Bassett, Tibbit et al., 2013). Surprisingly, we found that in the absence of Prl-1, a CO2-induced irreversible wing hold up phenotype resembling spasticity develops. This led us to discover a possible role for Prl-1 in the nervous system.
CO2-evoked behavioral responses in winged insects are important for food foraging, reproduction and survival (Guerenstein, Christensen et al., 2004, McMeniman, Corfas et al., 2014, Stange & Stowe, 2015). Drosophila, in particular, is highly sensitive to CO2 where the detection of a CO2 threshold is usually accompanied by clear physiological and behavioral responses. These have been previously explained in terms of anesthetic and toxic effects related to high concentrations of CO2 (Badre, Martin et al., 2005, Dijken, Sambeek et al., 1977), or as a stress odorant eliciting avoidance behavior upon the detection of CO2 levels as low as 0.1% (Suh, Wong et al., 2004). The recent discovery that Drosophila is equipped with a single population of CO2-responsive neurons harbored in the third segment of the antenna and that the avoidance behavior is mediated by two chemosensory receptors, Gr21a and Gr63a, represents a significant breakthrough in understanding how Drosophila senses and processes CO2 stimulus (Jones, Cayirlioglu et al., 2007, Scott, Jr et al., 2001, Suh et al., 2004). Upon CO2 stimulation, avoidance behavior can be assessed using a T-maze assay in laboratory conditions. Correspondingly, the neural activity at sensory neuron synaptic terminals in the CO2-responsive V- glomeruli of the antennal Lob (AL) can be detected by Ca2+ imaging (Jones et al., 2007, Suh et al., 2004). However, to what extent CO2 stimuli has influence from the level of functional molecular connections at that of detailed neural circuits to the physiological function at bodily levels remains obscure. Here we reveal the localization of Prl-1 in CO2-responsive neural circuit and the higher brain centers and uncover a Prl-1 involved protective mechanism, based on its interaction with the Uex in the nervous system.
Uex, is the homologous gene of human ancient domain proteins (ACDPs), also known as CNNMs including CNNM1-4, of which CNNM4 acts as a Na+/Mg2+ exchanger and regulates Mg2+ efflux (Yamazaki, Funato et al., 2013). CNNMs were identified to interact with hPRL-1 or hPRL-2 in the mediation of cancer metastasis (Funato, Yamazaki et al., 2014, Hardy, Uetani et al., 2015).A recent study on human genetic diseases has reported that mutations in CNNM2 are causative for seizures and mental retardation in patients with hypomagnesemia(Arjona, de Baaij et al., 2014). In this study we initially confirmed that their homologues, Prl-1 and Uex, also physically interacted in Drosophila. We show that the wing hold up is an obvious consequence upon knockdown of Uex and the expression of Uex is dramatically decreased in prl-1 mutants. From the previous focus being almost exclusively oncogenic, it was surprising to uncover a highly significant Prl-1-Uex complex based neuroprotective role in Drosophila. Ectopic expression of either Drosophila Prl-1 or hPRL in the nervous system could rescue the mutant phenotype. Our study may open up a potential new focus upon the functional pathway of the hPRL beyond studies of oncogenic properties.
Results
Loss of Prl-1 results in irreversible wing hold up in Drosophila
An often overlooked behavioral phenomenon when using standard CO2 anesthesia in Drosophila is that the flies respond with a temporary holding up of wings which ceases upon recovery. Surprisingly, we observed an irreversible wing hold up phenotype developing in prl-1 mutants beyond CO2 stimulation. To characterize the occurrence of this wing hold up phenotype, we began by dividing the prl-1 mutant or wild type flies into several age groups (day 1, day 2, or day 3), keeping them separately (n=180 for each group, 20 flies per vial). All experimental groups were subjected to acute CO2 under a flow of 5L/m for 20 sec as a minimum manipulation. Non CO2-treated groups served as controls. After recovery, all age groups of WT flies showed normal wing posture. However, about 85% of mutants that had received the CO2 administration on day 3 showed irreversible wing hold up (Fig.1A-D and Fig. S1B). The mutants that had either received, or had not received the specific CO2 administration on day 1, began to show latent and progressive wing hold up, gradually reaching a percentage of 60% around day-21. This was both a considerably lower level and a slower response than that of the flies that had received the CO2 stimulation on day 2 or day 3 (Fig. S1B). These observations revealed that the 3-day old mutant flies were particularly sensitive to CO2 stimulation and present a rapid prevalence of irreversible wing hold up. As male prl-1 mutants displayed the more prominent phenotype upon the CO2 stimulation (Fig. S1A), we exclusively analyzed male responses in the following experiments.
We then considered whether this wing hold up in prl-1 mutants had resulted from a side-effect of general anesthesia. Gaseous nitrogen (N2) or volatile ether (Dijken et al., 1977, Van Voorhies, 2009) was used to anesthetize 3 day old prl-1 mutant flies, with the CO2 treatment used as the control. After recovery, neither N2 nor ether anesthesia groups displayed the wing hold up as seen in the CO2 treated groups (Fig. 1E). These observations indicated that irreversible wing hold up, as induced by CO2 exposure in prl-1 mutants, does not simply occur as a direct result of general anesthesia.
Expressing Prl-1 or hPRL in the nervous system can rescue the mutant phenotype
We next conducted a series of assays to screen the specific tissues or cells of prl-1 mutants with overexpression of Prl-1, in which the wing hold up could be rescued. The uas-prl-1 transgene line was expressed in the mutant flies under the control of several tissue or cell type-specific Gal4 drivers, including Tublin-gal4, pan-neuronal Elav-Gal4, Repo-Gal4, TH-Gal4, motor neuronal D42-Gal4, and muscle Mhc-Gal4. Surprisingly, among all drivers genetically manipulated in the prl-1 mutants, only the pan-neuronal expressed Elav-Gal4 could completely rescue the wing hold up in the 3-day old groups that had either received, or not received the acute CO2 treatment (Fig.1F and Fig. S1C). While other drivers, such as D42-Gal4 exhibited a partial rescue, the Tublin-gal4, TH-Gal4, Mhc-gal4 drivers failed to provide any rescue of the mutant phenotype (Table S1). These observations indicated a potent role for Prl-1 in the Drosophila nervous system, specifically relating to CO2 stimulation. Given this neurologic impairment in prl-1 mutant flies, we then tested the function of hPRL in the context of the Drosophila brain. We generated transgenic flies containing hPRL-1 or hPRL-2. Ectopic expression of either UAS-hPRL-1 or UAS-hPRL-2 driven by Elav-Gal4 was able to rescue the wing hold up (Fig. 1G). Our results suggest that hPRL may have retained this conserved neuroprotective function.
Prl-1 is enriched in the CO2 neural circuit and higher brain centers
To evaluate the expression pattern of Prl-1 in the brain, we first performed an immunoblot assay with the whole head tissues of adult flies. A positive signal was detected in the control animals by using a rabbit polyclonal antibody against the full-length peptide of Prl-1, which was abolished in prl-1 mutants (Fig. S1D). To accurately target the Prl-1 expressing neurons, we also generated a transgenic Gal4 line which contained a 6.1Kbp of genomic DNA immediately upstream to the open reading frame of the prl-1 gene. We flanked the Gal4 sequence with both the 5’ and 3’ flanking regions of the prl-1 gene and constructed a fused green fluorescent protein (EGFP) with Prl-1 sequence in the N-terminal to make a high-fidelity reporter. By directly viewing GFP fluorescence in the transgenic flies expressing EGFP-Prl-1 under the control of the Prl-Gal4 driver, the whole head showed green with particularly robust GFP signals in the third segment of antennae (Fig.2A-A’). Confocal scanning revealed that the Prl-1 protein was apparently expressed in the basiconic sensillum (Fig.2B-B’), which houses CO2-responsive neurons (Scott et al., 2001, Suh et al., 2004). Prl-1 expression had also expanded along the axons of the olfactory CO2 neurons to its projected V-glomeruli in the antennal lobe (AL), and further to the higher processing center, the mushroom body (MB) (Kwon, Dahanukar et al., 2007, Suh et al., 2004). The distribution of EGFP-Prl-1 in the AL and the MB was also assessed using immunofluorescence staining of the adult brain with a GFP antibody (Fig.2C-D’). It is notable that the enrichment of Prl-1-Gal4 expression in the stereotyped V-glomeruli corresponds very well to the identical dendritic structure of CO2 sensory neurons expressing Gr21a-Gal4 and Gr63a-Gal4 (Jones et al., 2007) and their projection neurons expressing PNv-Gal4s (v201089, v200516) (Lin, Chu et al., 2013) (Fig.3A, and Fig. S2).
Ablation of Gr21a, an olfactory CO2 receptor, suppresses the prl-1 mutant phenotype
Analysis of CO2-evoked avoidance responses using T-maze assays (Kwon et al., 2007, Suh et al., 2004) revealed no significant differences between prl-1 mutants and the WT (Fig. S3A). This indicates that the olfactory sensing of CO2 remains active in prl-1 mutants. However, irreversible wing hold up in the prl-1 mutants proceeded with age when maintained in the ambient environment, and at a highly accelerated rates for 3-day old flies treated with the acute CO2 stimulation. We then considered whether the irreversible wing hold up could be blocked by suppressing CO2 perception via removing ligands from the receptors. We employed the RNA interference system to genetically knock down Gr21a in the CO2 sensory neurons. The bigeneric progeny of prl-1 mutants, each group bearing Elav-Gal4 or Gr63a-Gal4 along with Gr21a-RNAi, no longer displayed any rapid prevalence of irreversible wing hold up under acute CO2 exposure for 3-day old flies. Progressive irreversible wing hold up had also dramatically decreased from 60% to less than 5% in the non CO2 treated groups (Fig.3B and Fig.S3B). However, driven by Gr63a-Gal4, overexpression of Prl-1 only in the CO2 sensory neurons could not rescue the phenotype (data not shown). These observations suggest that the mutant flies are highly susceptible to brain disorder when the Prl-1 linked defense against CO2 insult is compromised. Therefore, the irreversible wing hold up is a deficit in olfactory information processing in the central brain, where CO2 might act as a neuropathological trigger in absence of Prl-1.
Enhancement of CO2-evoked Ca2+ activity and elevated ROS in the prl-1 mutants
Upon activation of CO2-responsive neural circuit driven aversion responses, Ca2+ transients could be visualized using two-photon microscopy with the expression of calcium-sensitive fluorescent protein (GCaMP) in the AL (Jones et al., 2007, Suh et al., 2004). We analyzed the V-glomerular activation pattern by Ca2+ imaging in WT and prl-1 mutants upon CO2 -evoked aversion responses. The GCaMP indicator (UAS-GCaMP6.0) is driven by the Elav-GAL4 activator in all neurons (Jones et al., 2007, Wang, Wong et al., 2003). Upon 20 sec CO2 exposure, the V glomeruli were activated (Fig.3C). The intensity of Ca2+ activity viewed in the prl-1 mutants was 2-fold higher than that in the controls (average peak ΔF/F of prl-1−/− is 1.21±0.51; average peak ΔF/F of WT is 0.55±0.46) (Fig. 3D-E). The overexpression of a uas-prl-1 transgene driven by Elav-Gal4 in the mutants could restore CO2-evoked Ca2+ activity to their original levels (Fig.S3D-F).
Oxidative stress represents a number of sequential and integrated processes that lead to cell vulnerability in the brain (Coyle & Puttfarcken, 1993).We took advantage of the GSTD1-ARE-GFP flies to evaluate the reactive oxygen species (ROS) levels in the prl-1 mutants, where GSTD1 is regulated via the Keap1/cnc signaling pathway in response to oxidative stress (Nguyen, Nioi et al., 2009, Sykiotis & Bohmann, 2008). By viewing GFP fluorescence in different age groups of flies, higher levels of ROS, as indicated by stronger GFP signals, were observed in prl-1 mutants as compared to those of the control animals (Fig. S3C). We also performed knockdown of the cnc levels in PRL-1 mutants, resulting in exacerbated wing hold up phenotype in the mutants (data not shown).The pan-neuronal Elav-gal4 driven Prl-1 overexpression was able to rescue the elevated ROS phenotype (Fig. S3C). This data may serve as an indicator that PRL-1 may function as a neurological antioxidant via the Keap1/cnc pathway.
Irreversible wing hold up results from a decrease of Uex in the prl-1 mutants
We next asked whether prl-1 cases the underlying genes responsible for the irreversible wing hold up. We focused on Drosophila Uex, the only known homologous gene of the CNNM family members that have been recognized as Prl-binding partners in mammalian cells (Funato et al., 2014, Hardy et al., 2015). To examine whether Prl-1 also interacts with Uex in Drosophila, WT Prl-1 and catalytically inactive mutant Prl-1-D77A/C109S constructs were generated with HA-tagged at their N-terminus. These were transiently expressed in S2 cells and the lysates were subjected to IP with an anti-HA antibody. Co-immunoprecipitation of endogenous Uex could be detected when exogenous HA-Prl-1 was precipitated from S2 cells (Fig. 4A) revealing that Uex can also interact with Prl-1 in Drosophila. A GST-pulldown assay validated the direct interaction between Prl-1 and Uex, whereas this interaction was decreased with the mutation of the catalytic sites Prl-1-C109S or Prl-1-D77A/C109S (Fig. 4B). The Co-IP results showed that these amino acid substitution mutants of Prl-1 largely decreased the Prl-1-Uex binding ability (Fig. 4A). This suggested that Uex does not function as a phosphate substrate of Prl-1. The co-localization of Prl-1 and Uex on the plasma membrane of S2 cells was also detected by double-immunofluorescence staining with the antibodies against Prl-1 and Uex (Fig. 4C).
Given the abnormal wing posture in prl-1 mutant flies and a direct interaction existing between Prl-1 and Uex, we then searched for a similar phenotype of abnormal expression of Uex. However, the loss of Uex function resulting from CRISPR/Cas9 led to embryonic lethality. RNA interference was then used to knock down Uex expression in adult flies and a panel of Gal4 lines was employed to detect wing posture. Strikingly, all flies expressing Uex-RNAi (line 36116) driven by pan-neuronal Elav-Gal4 displayed a similar irreversible wing hold up, 1 day after eclosion (Fig. 4E and Fig.S4). This was associated with short life span of no more than 10 days. Several Gal4 lines were employed to map abnormal wing posture resulting from the knockdown of Uex. Among all these differing cell-type-or region-specific Gal4 lines, expressed within CNS or peripheral nervous system (PNS) or muscles, only Elav-Gal4 could induce the irreversible wing hold up (Fig. S4). By analyzing the pan-neuronal RNAi-induced knockdown of Uex expression, our Western blot data showed that a single protein with a mass of ~92 kDa had been obviously reduced compared to the wild type (Fig. 4G). We also generated a UAS-Uex line using the complete protein coding sequence to exclude any off-target effect. As figure 4F shows, these flies displayed normal wings with co-expression of UAS-Uex and Uex-RNAi lines. These results indicate that irreversible wing hold up is directly correlated to the function of Uex, the knockdown of Uex in the nervous system resulting in the same phenotype.
We hypothesize that the irreversible wing hold up might be accounted for by the decreased expression of Uex in the prl-1 mutants, considering the direct interaction between Prl-1 and Uex in the brain. We probed the rate of Uex expression within the different age groups of flies after eclosion. Compared with the control animals, the protein level of Uex had decreased in prl-1 mutant flies (Fig. 4G). More strikingly, there was a gradual degradation over time of Uex protein, from the newly eclosed mutant flies to the 3 day old mutant flies (Fig. 4G). This may explain why a particularly rapid prevalence of wing hold up occurs for 3-day old mutant flies treated with acute CO2. Alternatively, we performed ectopic expression of Uex in the prl-1 mutant background to test whether Uex functions downstream of Prl-1 signaling. A conditional RU486-dependent Gal4 (GeneSwitch) was used to induce tissue-specific transgene expression (Osterwalder, Yoon et al., 2001). As RU486 was fed to the flies at the adult stage, prl-1 mutant flies that were exposed with 20s CO2 on day 3, showed a reversible wing posture within one week (Fig. 4H). The RU486-induced Uex overexpression in prl-1 mutants was verified by Western blotting (Fig. 4I). The Uex protein was recovered to a high level using the Elav-GeneSwitch driver. These results confirmed that Prl-1 positively regulates Uex at the genetic and molecular level in Drosophila.
Discussion
Here, we have identified in a Drosophila model that, in the absence of Prl-1, olfactory CO2 stimulation can cause irreversible wing hold up, negating any possibility of flight. The prl-1 mutant flies retain normal responses to anesthesia including CO2, N2 or volatile ether and also remain responsive to CO2 with no significant change in avoidance behavior. It seems that the sensing of CO2 remains relatively un-affected in the mutants but the processing of olfactory information in the central brain is rendered defective, resulting in an olfactory CO2-induced irreversible wing hold up phenotype. Overexpression of either Drosophila Prl-1 or hPRL in the nervous system could rescue the mutant phenotype. The ablation of the olfactory CO2 sensory neurons blocks phenotype development. Therefor CO2 might act as a neuropathological substrate for brain disorders in absence of Prl-1.
CO2 has also had a long history as a human anesthetic where increased CO2 levels decreases hippocampal neuronal excitability and may cause sedation (Capps, 1968, Jones, 1984). At the cellular level, CO2-induces changes in pH and may alter neuronal excitability by altering extracellular adenosine and ATP concentrations (Xu, Uh et al., 2011). CO2 insufflation-induces oxidative stress and has a toxic effect on neuroblastoma cells leading to DNA damage (Montalto, Currò et al., 2013). In E. coli, CO2 was seen to exacerbate the toxicity of ROS in a dose-dependent manner (Ezraty, Ducret et al., 2011). Here, our observation that prl-1 mutants display both enhancement of Ca2+ transients in the CO2-responsive neural circuit and elevated ROS production and that both could be rescued by overexpression of Prl-1 in the nervous system, suggests a potent neuropretective role for Prl-1. We propose that CO2 could act as an olfactory pathway linked neuropathological substrate for brain disorders in absence of Prl-1. This finding might open up a new focus upon the functional pathway of the hPrl beyond studies of its oncogenic properties.
It is notable that, in prl-1 mutants, the expression of Uex is clearly decreased. Knockdown of Uex resulted in the same wing hold up as observed in the prl-1 mutants, while abnormal wing posture in prl-1 mutants could be rescued by the expression of Uex in the nervous system. Thus irreversible wing hold up could be regarded as a downstream consequence of a deficiency in Uex due to the absence of Prl-1. The mammalian counterparts of Uex have been predicted as magnesium transporters involved in tumor progression (Gulerez, Funato et al., 2016, Kostantin, Hardy et al., 2016). Biochemical analyses of cultured cells revealed that the PRL acts through the inhibition of CNNM to regulate intracellular magnesium homeostasis (Funato et al., 2014, Hardy et al., 2015, Hirata, Funato et al., 2014). In vertebrate neurons, Mg2+ acts as a physiological Ca2+ antagonist for blocking the excitatory N-methyl-D-aspartate receptors in the CNS (Iseri & French, 1984, Zito & Scheuss, 2009) and has therefore been suggested as a possible means of resolving muscle rigidity and spasm in cases of tetanus(Ceneviva, Thomas et al., 2003). In our experiment, although it was inaccessible to measure the Mg2+ status in the prl-1 mutants, enhanced Ca2+ activities were observed at CO2-responsive neuron synaptic terminals. In human, CNNM2 mutations cause impaired brain development and seizures in patients with hypomagnesemia (Arjona et al., 2014). We hypothesize that the decreased Uex in Prl-1 mutants could cause disturbed Mg2+ homeostasis and consequently lead to a brain disorder presented as irreversible wing hold up.
The irreversible wing hold up in prl-1 mutants seems to be a prominent form of hyperreflexia resembling spasticity. Spasticity in mammals is a common complication beyond brain injury, presenting as intermittent or sustained involuntary activation of muscles (Pandyan, Gregoric et al., 2005). In an early study Takano et al had noted that mPRL is expressed in neurons and oligodendrocytes in the brain and is enhanced in the cerebral cortex following transient forebrain ischemia in rats (Takano, Fukuyama et al.,1996). The possible role of PRL-1 in an ischemia stroke situation where neurons undergo restrictions in both oxygenated blood inflow and CO2 blood-export, has not yet to be followed up. Analysis of the cerebral cortex and hippocampus tissue sections from a small sample set of Alzheimer’s disease patients revealed that there seemed a trend toward increased expression of PRL-1 mRNA expression and also a considerable correlations with patient age in the brain (Dumaual et al., 2012). The prl-1 mutant flies also exhibited age-dependent phenotypic manifestation. Intriguingly, in Parkinson’s and Alzheimer’s diseases there is profound olfactory disorder in odor threshold detection, odor memory, and/or odor identification occurring prior to disease onset (Doty, 2017, Kalia & Lang, 2015, Rudzinski, Fletcher et al., 2008), often associated with aspects of limb spasticity (Erro & Stamelou, 2017, Karlstrom, Brooks William et al., 2007). Drosophila models of human neurodegenerative diseases have already included observations of the wing hold up or hold out phenotypes (Clark, Dodson et al., 2006, Freibaum, Lu et al., 2015, Pandey & Nichols, 2011).The cause of these disorders is unclear. It is worth considering that olfactory CO2 stimulation, even of atmospheric levels, might be directly involved in these neurologic disease-related detrimental processes, for which Prl-1 provides defense. Our findings might be applicable for the potential involvement of hPRL in neuroprotection.
Materials and methods
Fly strains and genetics
The following transgenic flies were used: (1) Elav-Gal4, (2) Tubulin-Gal4, (3) Repo-Gal4, (4) TH-Gal4, (5) Orco-Gal4, (6) Or476-Gal4, (7) Gr21a-Gal4, (8) Gr63a-Gal4, (9) V201089-Gal4, (10) V200516-Gal4, (11) MB247-Gal4, (12) OK107-Gal4, (13) GF-Gal4, (14) D42-Gal4, (15) 24B-Gal4, (16) Vglut-Gal4, (17) MHC-Gal4, (18) Mef2-Gal4, (19) Elav-GeneSwitch-Gal4, (20) Uas-mCD8::GFP, (21) Uas-GCaMP6.0,(22) Uas-uex IR, (23) Uas-Gr21a IR, (24) GSTD-ARE-GFP.
Plasmid construction and infection: prl-1 was cloned in a pAHW vector (Carnegie Institution of Washington), and pGEX-4T-1 vector. Mutations were introduced by site-directed mutagenesis using a GBclonart mutagenesis kit.
Generation of transgenic flies: prl-1, EGFP-prl-1, uex, hPrl-1, and hPrl-2 were cloned in a pUAST-attb plasmid using specific primers (see below) and amplified by PCR. The prl-Gal4 was cloned in a pW25-Gal4 plasmid, which flanked the Gal4 sequence with both the 5’ and 3’ flanking regions of the prl-1 gene. All constructs were integrated into a single attP docking site, VK33 on chromosome 3L, using common phiC31 site-specific integration, as Matthew P Fish described(Venken & Bellen, 2007).
Antibody generation: Full-length cDNA of prl-1, and N-terminal 300bp cDNA of uex were cloned into a pGEX-4T-1 expression vector. The GST-fusion protein was affinity purified using Sepharose-4B beads (GE Healthcare). The polyclonal antibody was raised in rabbits. Animal experiments were conducted in accordance with the Guidelines for the Care and Use of Laboratory Animals of Zhejiang University.
Immunofluorescence staining
Immunofluorescence staining of the adult brains and S2 cells were conducted as previously described (Riemensperger, Strauss et al., 2011, Rogers, Wiedemann et al., 2004). The following primary antibodies were used: rabbit anti-Prl-1,1:500 (this study); rabbit anti-Uex,1:500; mouse anti-Brpnc82, 1:50 (DSHB);DAPI (1 g/ml; Sigma-Aldrich).
IP and GST pull-down
Cells and fly tissues were lysed with TAP buffer (1% Triton, 50mM Tris pH 8.0, 125mM NaCl, 5% Glycerol, 0.4% NP-40, 1.5mM MgCl2, 1mM EDTA, 25mM NaF, and 1mM Na3VO4), supplemented with protease inhibitor (Roche, Laval, QC, Canada). For IP, 1-2 mg of proteins was incubated with 1 μg of HA antibody (Abcam, ab9110) and Protein A-agarose beads (Roche Applied Science) according to the manufacturer’s protocol. The supernatants eluted from immunoprecipitated beads were loaded for Western blotting following standard protocols. For the GST pull-down assays, 500 μg of proteins were incubated with glutathione sepharose (GE Healthcare, Canada) for 3 h.
Calcium imaging
Sample preparation and calcium imaging were as described in. CO2 was delivered at a flow rate of 5 ml min−1. Sample preparation and calcium imaging were as described in(Jones et al., 2007, Suh et al., 2004). CO2 was delivered at a flow rate of 5 ml min−1. The adult Drosophila was fixed to the scotch tape by its dorsal parts with its wings and the maxillary pulp was also immobilized using a scotch tape slice. The dorsal parts of the adult brain were dissected within a drop of AHL solution(Jones et al., 2007, Suh et al., 2004) and then covered with a coverslip for Ca2+ imaging. Imaging of Ca2+ was performed on Olympus fluoresce microscope with a x20 objective lens. Images were acquired at 1.42 frames per second. For quantitative analysis of Ca2+ imaging data, images were batch processed with Image J to determine fluorescence intensity. The initial 120 seconds sequential images, occurring prior to the 20 second CO2 stimulus, were subjectively selected and the average fluorescence intensity (F) was set as the basal level. Changes in fluorescence intensity (△F) in the images were calculated and △F/F was used to denote Ca2+ responses. Heat map images were generated with Matlab (Mathworks Inc., Natick, MA, USA) by setting the basal fluorescence level at zero.
Behavior assays
About 50 flies were placed into the T-maze for each CO2 avoidance assay. At the T-maze choice point, flies sensed the converging currents of fresh air and CO2 from each arm. After 1 min of choice, the avoidance index (AI) was calculated.
Statistics
All the raw data were analyzed parametrically with excel and GraphpadPrism 5 software. The data were evaluated via Two-tailed Student’s t test. All data are presented as mean ±SEM. *P < 0.05; **P < 0.01; ***P < 0.001.
Conflict of interest
The authors declare that they have no conflict of interest.
Acknowledgments
We thank Jun Ma, Hao Wang, Lijun Kang for helpful suggestions on data analysis; Qi Zeng, Chris Wood and Jingwei Zhao for discussions and comments on the manuscript. We also thank Yu Cai (GSTD1 - ARE-GFP), Jiangqu Liu (GCaMP), the Bloomington Drosophila Stock Center and Qinghua Drosophila Stock Center for providing the fly stocks. This work was supported by the National Basic Research Program of China (2013CB945600 (X.Y.), 2012CB966800 (W.G.)).