Exploring Excitotoxicity and Regulation of a Constitutively Active TRP Ca²⁺ Channel in Drosophila

ABSTRACT

Unregulated Ca²⁺ influx affects intracellular Ca²⁺ homoeostasis, which may lead to neuronal death. In Drosophila, following the activation of rhodopsin the TRP Ca²⁺ channel is open to mediate the light-dependent depolarization. A constitutively active TRP channel triggers the degeneration of Trp^P365/+ photoreceptors. To explore retinal degeneration, we employed a multidisciplinary approach including live imaging using GFP tagged actin and arrestin 2. Importantly, we demonstrate that the major rhodopsin (Rh1) was greatly reduced before the onset of rhabdomere degeneration; a great reduction of Rh1 affects the maintenance of rhabdomere leading to degeneration of photoreceptors. Trp^P365/+ also led to the up-regulation of CaMKII, which is beneficial as suppression of CaMKII accelerated retinal degeneration. We explored the regulation of TRP by investigating the genetic interaction between Trp^P365/+ and mutants affecting the turnover of diacylglycerol (DAG). We show a loss of phospholipase C in norpA^P24 exhibited a great reduction of the DAG content delayed degeneration of Trp^P365/+ photoreceptors. In contrast, knockdown or mutations in DAG lipase (InaE) that is accompanied by slightly reduced levels of most DAG but an increased level of DAG 34:1, exacerbated retinal degeneration of Trp^P365/+. Together, our findings support the notion that DAG plays a role in regulating TRP. Interestingly, DAG lipase is likely required during photoreceptor development as Trp^P365/+; inaE^N125 double mutants contained severely degenerated rhabdomeres.

Introduction

Excitotoxicity describes the neuronal injury often observed in pathological conditions such as stroke, which is caused by the excessive release of glutamate in the central nervous system [1]. Glutamate activates glutamate receptors, particularly the ionotropic NMDA receptors, resulting in excessive Ca²⁺ influx [2] that perturbs intracellular Ca²⁺ homoeostasis critical for the neuronal structure and function. Excessive Ca²⁺ influx often leads to Ca²⁺ overload that may affect the endoplasmic reticulum and the Golgi, thereby disrupting protein synthesis [3] and trafficking [4]. Moreover, Ca²⁺ overload impacts the integrity of mitochondria leading to the release of cytochrome C to trigger apoptosis [5]. Elevated Ca²⁺ may globally activate the Ca²⁺-dependent proteases, endonucleases, and phospholipases to alter the macromolecular composition of cells. Importantly, excitotoxicity appears to be a shared mechanism of neurodegeneration among some disorders including Parkinson’s and Alzheimer’s diseases [6].

Here we explored how a constitutively active TRP Ca²⁺ channel orchestrates excitotoxicity in Drosophila. In photoreceptors, TRP and TRP-like (TRPL) channels are responsible for the light-dependent depolarization following the activation of rhodopsin [7,8]. In the visual cascade, activated rhodopsin interacts with the heterotrimeric Gq protein resulting in activation of phospholipase Cβ4 (PLCβ4) [9]. Subsequently, PLCβ4 breaks down phosphatidylinositol-4, 5-bisphosphate (PIP₂) to generate inositol triphosphate (IP₃) and diacylglycerol (DAG). DAG can be further metabolized by DAG lipase to release fatty acids. Both DAG [10,11] and polyunsaturated fatty acids (PUFAs) [12,13] have been implicated in the gating of the TRP and TRPL channels in Drosophila.

Drosophila Trp^P365 mutants express a missense mutation in the 5^th transmembrane domain leading to a constitutively active Ca²⁺ channel possibly by increasing the probability of spontaneous Ca²⁺ entry [14]. Trp^P365 displays severe light-dependent retinal degeneration [15], which can be reduced by overexpression of CalX, a Na⁺/Ca²⁺ exchanger protein involved in the regulation of intracellular Ca²⁺ [16]. We explored mechanisms of degeneration in Trp^P365/+ photoreceptors for insights into excitotoxicity, as retinal degeneration in Drosophila can be explored via cell biological, biochemical, and genetics strategies. The major visual system of Drosophila is the compound eye; each consists of 800 unit-eyes (ommatidia). Within each unit-eye (ommatidium), there are eight photoreceptors, R1-8, where visual transduction takes place. Within each photoreceptor, TRP is localized in the rhabdomere that consists of densely packed membrane in which rhodopsin is also localized. Rhabdomeres are arranged in a stereotypical pattern in each ommatidium that can be easily identified microscopically. A loss of rhabdomeres is a hallmark of retinal degeneration, which can be investigated in live flies monitoring the epifluorescence of rhodopsin or GFP tagged reporters.

In this report, we employed transgenes that express actin-GFP [17] or Arrestin2-GFP [18] to uncover defects leading to degeneration of Trp^P365/+ photoreceptors. We show a great reduction of the major rhodopsin Rh1 occurs prior to the loss of rhabdomere, whereas TRPL, a cation channel, displays a more complex regulation. We also demonstrate an up-regulation of CaMKII, however, suppression of CaMKII enhances the degeneration of Trp^P365/+ photoreceptors.

We explored the regulation of TRP by investigating the genetic interaction between Trp^P365/+ and mutants regulating DAG turnover. Importantly, we show that the degeneration of Trp^P365/+ photoreceptors is delayed in the norpA^P24 background, which contains reduced levels of DAG as quantified by LC-MS (liquid chromatography-mass spectrometry). In contrast, degeneration of Trp^P365/+ is greatly enhanced in the knockdown of inaE or inaE^N125 with reduced DAG lipase activity [13]. Interestingly, inaE^N125 contains an increased level of an unsaturated DAG. Together, these results support the notion that activation of TRP is regulated by DAG. We also observed a severe loss of rhabdomeres in pupal photoreceptors of Trp^P365/+; inaE^N125 double mutants, implicating DAG lipase in the formation of rhabdomeres during development.

Materials and methods

Fluorescence microscopy

Adult flies were anesthetized by CO₂, immobilized in clay placed in a 50 mm Petri dish for imaging the compound eye. Similarly, pupa (>p13 or >80% pupal development) [19] with puparia removed were positioned in clay placed in a Petri dish. The compound eye was examined using an upright Olympus AX70 microscope equipped with a 10X lens for detecting dpp (deep pseudopupil) or a 40X water immersion lens (LUMPLFL 40X) for examining multiple ommatidia. Image acquisition was performed at 100X magnification for dpp and 400X for rhabdomeres/ommatidia using IPLab image acquisition software (BioVision Technologies, Exton, PA, USA) and the Retiga camera from QImaging (Surrey, BC, Canada). Exposure time was made constant throughout the experiment based on the brightest signal in the control group. Multiple flies (n ≥ 5) were analysed.

Fly handling for microscopy

Adult flies or pupae were sorted and manipulated under a dissecting microscope with a light source of 600 lux for less than one minute for adults or three minutes for pupae. During imaging, compound eyes were subjected to the blue light (1300 lux) from the fluorescent microscope, and images were taken immediately for Act88F-GFP or Arr2-GFP. In general, the fluorescent intensity of Act88F-GFP or Arr2-GFP remains constant during the image acquisition. Heat shock treatment was performed by incubating flies at 37°C for one hour every 12 hours following eclosion.

Fluorescent image analysis

All image manipulation was performed under the guideline of Rossner and Yamada [20]. Fluorescent images included in the Figures are similar in appearance to the raw images. Experimentally, we collected newly eclosed flies of the desirable genotype, placed them in a vial, and aged at 25°C in a 12 h light/dark cycle (ambient light 300 lux) for various amounts of time. Flies were analysed for GFP-marked rhabdomeres by water-immersion fluorescence microscopy. Retinal degeneration was scored based on the number of rhabdomeres present in a given ommatidium (unit eye) similar to that described by Cerny et al. (2015). Scores from five flies from multiple crosses were calculated, and from each fly, 8 ommatidia were counted to obtain an average of the rhabdomere number. Control flies with similar eye colours (red-eyed or white-eyed) were used for comparison.

Quantitative Western blotting

One fly head was dissected, placed in an Eppendorf tube, and extracted with 15 μl of 2X Laemmli sample buffer via sonication. Retinal extracts were prepared similarly after first removing retinas from fly heads. Protein extracts were size-fractionated by SDS/PAGE (10–12%) and transferred onto a nitrocellulose filter. Filters were incubated with desired primary antibodies, followed by the fluorophore-conjugated secondary antibodies (e.g. Alexa Fluor 680 Goat Anti-Rabbit IgG, Invitrogen, Carlsbad, CA). The fluorophore signal was analysed by the Odyssey Infrared Imaging System (LI-COR, Lincoln, NE, USA). Individual protein content was normalized by comparing it to that in 1-day old wild-type. Inactivation-no-afterpotential D (INAD) [21], a photoreceptor protein whose level is relatively constant, was used as the loading control. We carried out three to five analyses (n = 3–5) with each analysis using one fly head or four retinas.

Antibodies for arrestin 2 (Arr2) [18], Ca²⁺/Calmodulin-dependent protein kinase II (CaMKII) [22], eye-PKC [23], and no-receptor potential A (NORPA) [24] were used. Antibodies against TRP-like (TRPL) were generated in rabbits using a bacterial fusion protein corresponding to 757–1124 aa of TRPL. Polyclonal antibodies for activated CaMKII (phosphorylated CaMKII at Thr²⁸⁷) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Monoclonal antibody for Rh1 (4C5) was obtained from Developmental Studies Hybridoma Bank [created by the NICHD and maintained at The University of Iowa].

Enrichment of rhabdomere membranes for lipid extraction

Membrane fractions enriched with rhabdomere membranes were obtained by differential centrifugation using a protocol modified according to [10]. Briefly, 100 freshly dissected heads were collected in an Eppendorf tube and sonicated following the addition of 0.4 ml of 1X PBS. The homogenates were centrifuged at 1,000 X g for 1 min to obtain P1, and the supernatant was re-centrifuged at 3,000 X g for 6 min to obtain P2. The supernatant was subsequently centrifuged at 12,000 X g for 15 min to obtain P3. The P3 fraction enriched with TRP was used for lipid extraction. The P3 membrane fraction was sonicated and resuspended in one ml of chloroform: methanol (2:1). Lipid extraction was performed with constant agitation for one hour (Eppendorf vortex mixer). All operations were carried out at 4° C.

DAG quantitation by ultra high-performance liquid chromatography-Mass spectrometry (UHPLC-MS)

The chloroform/methanol extract was collected following centrifugation (12,000 X g for 15 min), dried down, and re-dissolved in 100 μl of the buffer A [acetonitrile: isopropanol (7:3) containing 0.1% acetic acid and 10 mM NH₄Ac]. Fractionation of various DAG species by UHPLC was modified according to [10]. Ten μl of the sample were injected into a C8 reverse-phase column (100 mm X 2.1 mm X 2.6 μm particles, Kinetex) (Phenomenex, Torrance, CA) in the Vanquish UHPLC system (ThermoFisher, Waltham MA). Separation of DAG was optimized using the following programme which included 1 min at 90% buffer A/10% buffer B, which was followed by a linear gradient from 90% to 100% buffer A in 10 min. The flow rate was set at 0.3 ml/min. Buffer B contains 0.1% acetic acid and 10 mM NH₄Ac. The column was re-equilibrated with 90% buffer A for 10 min before another sample application. The retention times of DAG were within the 2.8–6.1 min interval.

The mass spectra were acquired using a Q Exactive HF Hybrid Quadrupole-Orbitrap Mass Spectrometer (ThermoFisher, Waltham, MA) in positive ion mode. The spectra were recorded using a full scan covering the ranges from 400–800 m/z. The resolution was set to 45,000 restricting the Orbitrap loading time to a maximum of 100 ms with a target value of 1E6 ions. The capillary voltage was set to 3kV with a sheath gas (N₂) flow value of 60 and an auxiliary gas flow (N₂) of 35. The capillary temperature was set to 300° C, while the drying gas in the heated electrospray source was at 150° C. The skimmer voltage was held at 25 V while the tube lens was set to a value of 130 V. The spectra were recorded from min 0 to min 10 of the UHPLC gradients.

Chromatograms from UHPLC-MS were analysed using Xcalibur (ThermoFisher, Waltham MA) based on the mass of the Na⁺ adducts of various DAG. DAG standards used included 1,2-dipalmitoyl-sn-glycerol, 1,2-dioleoyl-rac-glycerol and 1,2-distearoyl-sn-glycerol (Cayman Chemicals, Ann Arbour, MI). Chloroform, methanol, isopropanol, and acetonitrile (HPLC grade) were from Fisher Scientific (Waltham MA).

Real-time reverse transcription-polymerase chain reaction (RT-PCR)

Total RNA was extracted from 20 fly heads by a modified method of Chirqwin et al [25] and dissolved in 20 μl water. We used 5 μl of RNA for the first-strand cDNA synthesis employing Superscript III (Invitrogen, Carlsbad, CA) primed with random hexamers. Quantitative PCR in triplicate was performed via CFX96 Real-Time System (BIO-RAD, Hercules, CA) using iQTM SYBR Green Supermix (BIO-RAD). All expression values were normalized to RpL32 (rp49). Some of the primer sequences used for PCR were selected from the FlyPrimerBank [26] and listed below: RpL32, AGCATACAGGCCCAAGATCG, TGTTGTCGATACCCTTGGGC; inaE, TTGTTAGCGTCTCGTTGGTTATC, CGGCATCCAGAATACTGCCA; inaE-A, CACCAGCCCATCAACACTCA, GCGTGTGCTCGGTTGTATCT; arr1, CATGAACAGGCGTGATTTTGTAG, TTCTGGCGCACGTACTCATC.

Generation of Rh1-GAL4 lines

Full-length GAL4 cDNA (about 2.6 Kb) was obtained from pGaTB plasmid by HindIII digestion and subcloned into pGem7zf. The recombinants with flanking SacI (5ʹ) and SmaI (3ʹ) were digested with SacI and SmaI and ligated into YC4, which contains Drosophila Rh1 rhodopsin promoter [27]. The Rh1-GAL4 DNA construct in YC4 was microinjected into yw embryos (The Rainbow Transgenics Inc, Camarillo, CA). Multiple transgenic lines were selected based on the rescue of the yellow phenotype in the wings.

Drosophila stocks

Both Trp^P365 and inaE^N125 were obtained from Dr. W. Pak (Purdue University). Transgenic flies for GMR-GAL4 (stock #1104), inaE RNAi (#64885) and the GFP-tagged actin transgenes including Act5C (#9257), Act57B (#9256), Act87E (#9249), and Act88F (#9253) were obtained from Bloomington Drosophila Stock Centre (BDSC) (NIH P40OD018537). Fly cultures were maintained in the BDSC standard cornmeal medium at 25°C in a 12 h dark/light cycle. Standard cross was used to introduce suitable genetic background into Trp^P365/+.

Statistical analysis

One-way ANOVA and two-tailed Student’s t-test were employed for statistical analysis.

Results

Use of Actin-GFP and Arr2-GFP to investigate retinal morphology of live flies

Trp^P365 undergoes rapid retinal degeneration characterized by the loss of deep pseudopupil (dpp) [15]. Dpp is the optical superposition of rhodopsin epifluorescence in the rhabdomere from neighbouring ommatidia [28]. Dpp can be readily observed in live white-eyed flies. Here we employed a more sensitive GFP-based assay, which can be easily applied to both white- and red-eyed flies at various stages of retinal degeneration.

First, we investigated the use of GFP tagged actin as a reporter for the rhabdomere that is supported by the actin-based cytoskeleton. There are six distinct but highly homologous actin isoforms in Drosophila including Act5C, Act42A, Act57B, Act79B, Act87E, and Act88F. Based on the microarray data, the actin isoforms expressed in the eye include Act5C, Act57B, Act87E, and Act88F. We targeted the expression of each of the four isoforms as GFP tagged protein using the GMR driver [29]. We observed that, when ectopically expressed in R1-R8 photoreceptors, each GFP tagged actin isoform [17] could be incorporated into the cytoskeleton of rhabdomeres (Figure 1a). The green fluorescence associated with actin-GFP was readily detected in the eye, like dpp (Figure 1c). Accordingly, GFP-labelled rhabdomeres in each ommatidium also could be easily quantified and compared (Figure 1a).

Figure 1.

Live retina imaging using Act88F-GFP or Arr2-GFP as the reporter. Act88F-GFP was ectopically expressed in all photoreceptors of the compound eye via the GMR-GAL4. The GFP tagged actin was incorporated into the cytoskeleton of rhabdomeres, which was detected by the use of water-immersion (a) or regular objectives as fluorescent dpp (c). Water immersion microscopy allows corneal neutralization for visualization of the photoreceptor in live flies without any tissue manipulation. The expression of Arr2-GFP was engineered under the control of the Rh1 rhodopsin promoter. Arr2-GFP binds to activated Rh1 and becomes enriched in rhabdomeres of R1-6 photoreceptors. Arr2-GFP was detected via water-immersion (b) or regular objectives (d). Transgenic flies expressing Arr2-GFP were white-eyed while those expressing Act88F-GFP, red-eyed. One-day-old flies were used. Scale bar, 5 μm (a,b) or 20 μm (c,d)

We also employed GFP-tagged Arrestin 2, Arr2-GFP, to detect the rhabdomere [18] as Arr2 associates with light-activated rhodopsin and thus is recruited to the rhabdomere (Figure 1b). Arr2-GFP was engineered for expressing in R1-6 photoreceptors, and by the blue light stimulation, it gave rise to a fluorescent dpp (Figure 1d), similar to that by Act88F-GFP.

Exploring the degeneration of Trp^P365/+ photoreceptors via Act88F-GFP or Arr2-GFP

It is well established that Ca²⁺ dynamically controls the cytoskeleton by regulating actin polymerization and depolymerization [30]. We speculate that constitutively active Ca²⁺ channels would impact the rhabdomere by destabilizing the cytoskeleton. Indeed, Trp^P365/+ contained aberrant rhabdomeres [15], based on phalloidin staining. To further investigate how the cytoskeleton is affected, we employed Act88F-GFP [17]. We show that the degeneration of Trp^P365/+ photoreceptors was characterized by irregular ommatidia clusters with small rhabdomeres (Figure 2b), which was followed by a reduction of Act88F-GFP fluorescence in rhabdomeres as observed in 3-day old flies (Figure 2c). Subsequently, a great reduction of GFP was evident in 5-day old Trp^P365/+ (Figure 2d), which indicated a loss of rhabdomeres. Indeed, as the fly aged the average number of rhabdomere per ommatidium was decreased (Figure 2e). Based on the findings, we conclude that the constitutively active TRP activity may affect the intracellular Ca²⁺ level that directly or indirectly impacts the stability of the rhabdomere cytoskeleton.

Figure 2.

The age-dependent deterioration of rhabdomeres in Trp^P365/+ via Act88F-GFP. Trp^P365/+ flies of various ages were examined for Act88F-GFP that is incorporated into the cytoskeleton of rhabdomeres (b-d). One-day-old wild-type was used as a positive control (a). Degeneration of Trp^P365/+ photoreceptors was characterized by a reduction of Act88F and a loss of rhabdomeres. The stereotypical arrangement of rhabdomeres within each ommatidium was also affected. The time-dependent decline of the rhabdomere number in Trp^P365/+ (Mean ± SEM, n = 5) (e). Scale bar in a-d, 5 μm

Using Arr2-GFP we show that degeneration resulted in a distortion of the fluorescent dpp (Figure 3b, c). Consistently, we show a misalignment of rhabdomere clusters (Figure 3e, f), which was also observed using Act88F-GFP (Figure 2b, c). The Arr2-GFP intensity in the rhabdomere was initially reduced (Figure 3e, in 3-day old), but appeared more enriched in 5-day old Trp^P365/+. Consistently, rhabdomere numbers were reduced as flies aged (Figure 3g), which was also observed via Act88F-GFP (Figure 2e).

Figure 3.

The age-dependent progression of retinal degeneration in Trp^P365/+ via Arr2-GFP. Fluorescent dpp from Arr2-GFP in Trp^P365/+ was reduced in intensity (a-c), and the outline of dpp became distorted in older flies (b, c). Trp^P365/+ photoreceptors exhibited defects in the alignment of ommatidium clusters, which was observed in 3-day old flies (e). Moreover, 5-day old Trp^P365/+ flies showed a loss of rhabdomeres (f). The time course of rhabdomere number reduction (Mean ± SEM, n = 5) (g). Scale bar in d-f, 5 μm

A great reduction of Rh1 precedes the loss of rhabdomeres in Trp³⁶⁵/+ photoreceptors

It has been shown that Rh1 is required for the maintenance of rhabdomere [31] as it promotes morphogenesis of the rhabdomere terminal web [32,33]. We investigated the relationship between rhabdomere morphology and the Rh1 content following retinal degeneration. Experimentally, Trp^P365/+ flies of various ages were collected and analysed microscopically for rhabdomere morphology via Arr2-GFP (Figure 4a–d). Subsequently, each fly head was analysed by Western blotting to quantify the total Rh1 content (Figure 4e, f).

Figure 4.

A drastic reduction of Rh1 precedes deterioration of rhabdomeres in Trp^P365/+. Trp^P365/+ flies of various ages were first characterized via Arr2-GFP followed by Western blotting. Shown is one representative set of analyses (a-e). Retinal morphology was visualized via Arr2-GFP (a-d). The rhabdomere size in 1-day old Trp^P365/+ was comparable to that of wild-type (w¹¹¹⁸), but greatly reduced in 3-day old (c) with a drastic loss of rhabdomeres in 8-day old (d). Western blotting probed with Rh1, TRPL, or INAD antibodies (e). The Rh1 level was greatly reduced in 1-day old Trp^P365/+ flies while the TRPL level was initially reduced, but further elevated in 3-day old Trp^P365/+. The remaining Rh1, either in the monomer or dimer form, displayed slower mobility compare to that of 1-day old wild-type. Shown in the histogram are the time-dependent changes of the total Rh1, TRPL, or INAD level when compared to that in the newly eclosed (0-day old) Trp^P365/+ (f). Molecular weight standards of the Western blot (in Kd) are shown on the right. Scale bar in a-d, 5 μm

We observed a drastic reduction of Rh1 in 1-day old Trp^P365/+ (11.3 ± 3.2%, n = 3), which is further decreased to about 4.5 ± 2.6% (n = 3) in 3-day old Trp^P365/+ (Figure 4e, f). Significantly, the reduction of Rh1 was also accompanied by a slower mobility, suggesting the remaining Rh1 was either hyperphosphorylated or glycosylated (Figure 4e). Remarkably, despite a loss of Rh1, 1-day old Trp^P365/+ retained retinal morphology similar to that of wild-type (Figure 4a, b). Taken together, we show a great loss of Rh1 occurred before the morphological changes in rhabdomeres, consistent with the notion that Rh1 plays a critical role in promoting the stability of rhabdomeres.

We also explored whether TRPL is similarly affected in degenerating retinas. TRPL is a cation channel involved in the light-dependent depolarization of photoreceptors. Interestingly, we observed the level of TRPL was reduced in 1-day old (42.5 ± 3.7%, n = 3), but became greatly elevated by almost 10fold in 3-day old Trp^P365/+ (Figure 4 e, f). TRPL was drastically reduced in 8-day old, which coincided with a great loss of rhabdomeres (Figure 4d). Based on our findings, we conclude that TRPL is regulated by the constitutively active TRP Ca²⁺ channel via distinct mechanisms, different from those of Rh1.

Upregulation of the CaMKII activity in Trp^P365/+ photoreceptors

Unregulated TRP channels are likely to disrupt Ca²⁺ homoeostasis to impact diverse physiological processes. An increase of cytosolic Ca²⁺ promotes its binding to calmodulin (CaM); Ca²⁺ bound CaM regulates diverse molecular targets including transcription factors, protein phosphatases, and protein kinases such as Ca²⁺/calmodulin-dependent protein kinase II (CaMKII). Ca²⁺ is also required for the activation of the classical protein kinase C (PKC) family, which is dependent on DAG [34]. Aberrant activation of PKC or CaMKII may promote retinal degeneration by turning on apoptotic pathways [35].

We investigated whether CaMKII was affected in Trp^P365/+. In wild-type retinal extracts we observed several CaMKII isoforms including three major polypeptides of 490 amino acids (aa), 509 aa, and 530 aa [36](Figure 5a). Significantly, the total CaMKII content in Trp^P365/+ photoreceptors was increased (27.3 ± 4.5%, n = 3), when compared to wild-type (Figure 5c).

Figure 5.

Up-regulation of CaMKII and the functional consequence of the ala-mediated suppression of CaMKII in Trp^P365/+ photoreceptors. Total and activated CaMKII (with phosphorylation at Thr²⁸⁷) levels in retinas of 1-day old Trp^P365/+ and wildtype (w¹¹¹⁸) were analysed by Western blotting using INAD as loading controls (a). Quantitation of the total and activated CaMKII level (Mean ± S.E.M, n = 3) is shown in the histogram (c). The level of eye-PKC in Trp^P365/+ was not significantly different from that of wild-type control (b). Modulation of retinal degeneration by ala in Trp^P365/+ photoreceptors (d-i). Suppression of CaMKII led to irregular rhabdomere arrangement (e) with reduced rhabdomere number in Trp^P365/+; ala (f). The age-dependent reduction of the rhabdomere number is shown in j(Mean ± SEM, n = 5). All flies examined were red-eyed, and Arr2-GFP was used as a reporter

We also measured activated CaMKII using polyclonal antibodies that recognize phospho-Thr²⁸⁷, an autophosphorylation site of CaMKII. We show phosphorylation of the three major isoforms of CaMKII (490, 509, and 530 aa) from both wild-type and Trp^P365/+ (Figure 5a). Importantly, the total phosphorylated CaMKII isoforms were increased by approximately 42 ± 5.3% (n = 3) in 1-day old Trp^P365/+ (Figure 5c).

In contrast, the eye-specific PKC, a classical PKC involved in the visual signalling [23], was not significantly affected in 1-day old Trp^P365/+ (Figure 5b). Similarly, levels of INAD, NORPA, and Arr2 were not altered (Figure 5a, b). Based on the findings, we conclude that abnormal TRP activity results in the up-regulation of the CaMKII activity with an increase in both activated and total CaMKII content. CaMKII participates in diverse signalling processes; it may modulate apoptosis to impact the survival of photoreceptors.

Suppression of CaMKII exacerbates retinal degeneration of Trp^P365/+ photoreceptors

To explore whether the elevated CaMKII activity plays a role in initiating the degeneration of Trp^P365/+ photoreceptors, we investigated whether retinal degeneration could be modified when CaMKII was suppressed. To suppress CaMKII, we employed the heat-shock driven overexpression of an inhibitory peptide (ala) of CaMKII [37]. As the expression of ala is controlled by the hsp70 promoter, all fly strains were treated with heat shock accordingly.

We compared the age-dependent degeneration of Trp^P365/+; ala with that of Trp^P365/+ or ala. As shown in Figure 5, 3-day-old Trp^P365/+; ala flies contained irregular rhabdomere clusters with missing rhabdomeres (Figure 5e), while Trp^P365/+ alone appeared less degenerated with reduced rhabdomere size and number (Figure 5h). In contrast, overexpression of ala alone does not affect retinal morphology (Figure 5d) [18]. Significantly, 5-day-old Trp^P365/+; ala contained significantly reduced rhabdomere numbers (Figure 5f, j), when compared to Trp^P365/+ (Figure 5i, j). Taken together, retinal degeneration of Trp^P365/+ photoreceptors appears enhanced in the ala background, supporting that inhibition of CaMKII exacerbates the degeneration of photoreceptors expressing the constitutively active TRP channel. It is likely that activation of CaMKII exerts beneficial effects that promote the proper maintenance of rhabdomeres.

Degeneration of Trp^P365/+ photoreceptors is partially suppressed when PLCβ4 is missing

TRP is the founding member of the mammalian TRP superfamily whose members are involved in diverse physiological processes. How is TRP in Drosophila photoreceptors gated remains controversial; second messengers including DAG and PUFAs, or mechanical force [38] caused by hydrolysis of PIP₂ and protons [39], have been implicated. As DAG is involved in the gating of the canonical TRP family (TRPC) that Drosophila TRP belongs to, we explored whether DAG regulates it. Therefore, we investigated how mutations regulating the DAG synthesis modified retinal degeneration of Trp^P365/+.

In the visual cascade, DAG is generated by PLCβ4 encoded by the norpA gene. In norpA^P24, a null allele, we anticipate that DAG would be greatly reduced. We investigated the genetic interaction between norpA^P24 and Trp^P365/+ and compared the retinal morphology of single mutants to that of the double mutant. As shown in Figure 6, at 5-day old both single mutants exhibited overt degeneration with disorganized and smaller rhabdomeres (Figure 6e, h). Significantly, the degeneration of Trp^P365/+ (Figure 6e) is morphologically distinct from that of norpA^P24 (Figure 6h), which was characterized by a more uniform reduction of Arr2-GFP in rhabdomeres. In contrast, double mutants of the same age retained retinal morphology with discrete R1-6 rhabdomeres (Figure 6a), like wild-type (not shown). As flies aged, single mutants displayed a severe loss of rhabdomeres, which was evident in 8-day old flies (Figure 6f, i), while double mutants still contained mostly discrete rhabdomeres (Figure 6b). At 10-day old, double mutants began to show disorganized rhabdomeres (not shown), and at 16-day old, a severe loss of rhabdomeres (Figure 6c), similar to that of 8-day old single mutants. A comparison of retinal degeneration is summarized based on the number of the remaining rhabdomeres (Figure 6j). Taken together, we conclude that the degeneration of Trp^P365/+ is reduced in the norpA^P24 background.

Figure 6.

Degeneration of Trp^p365/+ photoreceptors is delayed in the norpA^P24 genetic background. Retinal morphology of Trp^P365/+; norpA^P24 double mutants (a-c) was compared to that of single mutants Trp^P365/+ (d-f) or norpA^P24 (g-i), using Arr2-GFP as the reporter. single mutants displayed overt degeneration with a severe loss of rhabdomeres, which was evident in 8-day old (f, i), while double mutants of the same age maintained retinal morphology (b) comparable to that of wild-type. degeneration in the double mutant was evident in 16-day old, which displayed a greatly reduced number of rhabdomeres (c). the time-dependent reduction of the rhabdomere number in the single and the double mutants (Mean ± SEM, n = 5) (j). ***, p < 0.0001; **, p < 0.001 by ANOVA. all flies tested were white-eyed. Scale bar in a-i, 5 μm

Both adult and pupal photoreceptors of Trp^P365/+; inaE^N125 display a severe loss of rhabdomeres

To further support the role of DAG to regulate TRP, we investigated whether reduced DAG lipase would affect the steady-state DAG level to impact retinal degeneration in Trp^P365/+. DAG is metabolized by DAG lipase (InaE) to generate 2-acylglycerol and fatty acid [13]. DAG also can be converted to phosphatidic acid (PA) by DAG kinase (RdgA) [40]. We investigated how a reduction of DAG lipase modified retinal degeneration via genetic interactions between Trp^P365/+ and inaE^N125, a hypomorphic allele.

Unexpectedly, we observed that Trp^P365/+; inaE were born with severely degenerated rhabdomeres (Figure 7b) when compared to single mutants (Trp^P365/+ or inaE) (Figure 7e). This finding strongly suggests that DAG lipase is critical during the development of photoreceptors. Moreover, rhabdomeres in the double mutants appeared further degenerated as the fly aged (Figure 7c).

Figure 7.

Genetic interaction between Trp^P365/+ and inaE ^N125 led to a severe loss of rhabdomeres. Retinal morphology of the Trp^P365/+; inaE ^N125 double mutants was compared to that of single mutants Trp^P365/+ via Arr2-GFP. Retinas from pupae (>80% pupal development; a, d), newly emerged (b, e), and 2-day old flies (c, f) were examined. double mutants displayed drastically reduced rhabdomere clusters in both adult and pupal photoreceptors. all flies analysed were red-eyed

To investigate the critical period leading to the degeneration of the double mutant, we examined pupal photoreceptors. Indeed, double mutants contained aberrant and degenerated rhabdomeres (Figure 7a), while single mutants, Trp^P365/+ (Figure 7d) or inaE^N125 (not shown), had the full complement of the R1-6 rhabdomeres, similar to wild-type (not shown). Our results support a strong genetic interaction between Trp^P365/+ and inaE^N125, suggesting a reduction of DAG lipase accelerates the degeneration of Trp^P365/+ photoreceptors. Moreover, a decrease of DAG lipase leads to abnormal rhabdomere biogenesis in pupal photoreceptors of Trp^P365/+.

Knockdown of inaE potentiates the degeneration of Trp^P365/+ photoreceptors

We investigated how retinal degeneration of Trp^P365/+ could be modified by reduced expression of DAG lipase in adult photoreceptors. We employed RNAi-mediated knockdown by expressing double-strand interference RNA for inaE using the Rh1-GAL4. To evaluate the efficiency of RNAi-mediated knockdown, we performed real-time RT-PCR to compare inaE expression in fly heads with and without the RNAi transgene. It is important to note that the expression of inaE is not restricted to photoreceptors. Genome annotation predicts six InaE protein isoforms, however, the critical isoform in photoreceptors is not known. We analysed the mRNA expression of all InaE isoforms by quantitative PCR and observed a reduction of 49.7 ± 8.2% (n = 3) in heads of the knockdown flies. We also investigated the mRNA for InaE-A, which appears reduced in the inaE^N125 mutant, and observed a reduction of 29.4 ± 5.7% (n = 3), relative to control.

Importantly, we show that knockdown of inaE led to a more severe degeneration characterized by a reduction of both rhabdomere number and Arr2-GFP intensity, both of which were evident in 3-day (Figure 8b, e) and 5-day old flies (Figure 8c, f). In contrast, knockdown of inaE alone did not affect the rhabdomere size and intensity (not shown). We quantified the average number of rhabdomeres with or without RNAi, and the results are summarized in Figure 8g. Taken together, knockdown of DAG lipase appeared to potentiate retinal degeneration of Trp^P365/+, which may be caused by increased DAG. An elevated DAG content is also likely to promote the activation of protein kinase C (PKC) to regulate TRP. However, TRP is negatively regulated by an eye-specific PKC in photoreceptors [41]. We conclude that suppression of DAG lipase enhances retinal degeneration of Trp^P365/+ possibly by increasing the DAG content.

Figure 8.

The RNAi-mediated knockdown of inaE potentiates the degeneration of Trp^P365/+ in differentiated photoreceptors. Degeneration of Trp^P365/+ (a-c) was examined with or without inaE knockdown (a-f) using Arr2-GFP as the reporter. the knockdown of inaE by RNAi was controlled by the Rh1 promoter using the GAL4/UAS strategy. retinal morphology of 1-day to 5-day old flies were compared, and all flies tested were red-eyed. Scale bar, 5 μm. (g), quantitation of retinal degeneration by rhabdomere numbers. The remaining rhabdomeres per ommatidium were compared. shown are mean ± SEM (n = 5)

The MS analysis shows a great reduction of DAG in norpA^P24 and an increase of DAG-34:1 in inaE^N125

To obtain evidence supporting the notion that changes in DAG levels in the mutants are responsible for modifying Trp^P365/+ degeneration, we quantified the DAG content. We prepared lipid extracts from rhabdomere enriched membranes (Figure 9a) and performed LC-MS analysis. It was reported that following light stimulation, DAG species that contain saturated fatty acids are not affected whereas several species of DAG containing unsaturated fatty acids are increased [10]. If these unsaturated DAG species are critical for activating TRP, a reduction would lessen the TRP activity impacting the progression of retinal degeneration in Trp^P365/+ mutants.

Figure 9.

Quantitation of the DAG content in rhabdomere membranes of wild-type, norpA^P24 and inaE^N125. (a) Western blotting of various membrane fractions obtained by differential centrifugation. shown is a representative preparation from wild-type fly heads. TRP and Rh1 were used as markers for rhabdomeres. each lane contained the equivalent of one fly head or the corresponding membrane fractions (P1-3), and the remaining supernatant (S3). the P3 fractions were used for lipid extraction. (b) Reverse-phase LC-MS analysis of wild-type (w¹¹¹⁸), norpA^P24, and inaE^N125 lipids. The chromatograms show the total ion current profile (m/z 400–800) of lipids from wild-type (top panel), norpA^P24 (middle), and inaE^N125 (bottom) with the retention time from 1.8–6.4 min. the retention time of various DAG ranges from 2.8 to 6.1 min (solid lines above retention time). (c) Changes of eight DAG species in the mutants, compared to those of wild-type. with one exception (DAG-32:1), most DAG including both saturated and unsaturated ones were greatly reduced in norpA^P24. the lipid in inaE^N125 contained moderately reduced levels of DAG with the exception of DAG-34:1, which is elevated to about 120 ± 1% compared to wild-type. shown are mean ± SEM (n = 3)

We quantified the steady-state level of the most abundant DAG in the lipid extracts, which included seven unsaturated as well one saturated DAG from wild-type and mutants (Figure 9b, c). We show that norpA^P24 contained reduced levels of seven DAG including DAG-36:0. The reduction ranged from 87 ± 1% to 34 ± 4% when compared to the wild-type level. The most drastic reduction was observed in DAG-34:2 (34 ± 4%) and DAG-36:4 (36 ± 11%) (Figure 9c). These findings strongly suggest that PLCβ4 encoded by norpA contributes to the majority of DAG synthesis in the eye. Similarly, we also quantified DAG levels in inaE^N125 that contains reduced activity of DAG lipase. We speculate that the DAG content would be increased when the lipase is decreased, although DAG also can be converted to PA by DAG kinase. Interestingly we show that DAG-34:1 was significantly elevated to 120 ± 1% in inaE^N125 while the rest of DAG is either moderately reduced (87 ± 3% to 70 ± 3%) or not significantly altered (Figure 9c). Our results suggest a possible link between DAG-34:1 and TRP activation. Taken together, we conclude that an increase of the critical DAG contributes to the acceleration whereas a reduction of DAG delays the degeneration of Trp^P365/+ photoreceptors.

Discussion

Drosophila TRP is the founding member of the mammalian TRP channels that participate in diverse physiology. Drosophila TRP functions downstream of PLCβ4 to initiate the light-dependent depolarization of photoreceptors. Importantly, a gain-of-function mutation at the fifth transmembrane domain (S5) gives rise to a constitutive TRP Ca²⁺ channel leading to the degeneration of photoreceptors [15]. A constitutively active TRP is likely to perturb intracellular Ca²⁺ homoeostasis to impact diverse biological processes. However, it appears that elevated Ca²⁺ alone is not sufficient to cause degeneration of photoreceptors as a moderate rise of intracellular Ca²⁺ fails to elicit retinal degeneration in calX mutants missing Ca²⁺/Na⁺ exchanger [16]. Specifically, retinal degeneration is dependent on light in both calX and Trp^P365/+ mutants [15]. Light activates rhodopsin leading to a rise of the intracellular Ca²⁺ that critically regulates recycling and turnover of rhodopsin. Mis-regulation of Ca²⁺ during visual signalling may result in the accumulation of hyperphosphorylated metarhodopsin, which promotes Arr1-mediated endocytosis [42,43] and Arr2-metarhodopsin complex formation [44,45] leading to retinal degeneration. Here we explored how elevated Ca²⁺ influx triggers retinal degeneration. We show the time-dependent loss of the rhabdomere in Trp^P365/+ using either Arr2-GFP or Act88F-GFP as the reporter. We also observed abnormal spacing of ommatidia, possibly resulting from disruption of the rhabdomere structure or alignment. Ca²⁺ is known to regulate actin polymerization and depolymerization, and abnormal Ca²⁺ entry in Trp^P365/+ may impact the rhabdomere cytoskeleton and rhodopsin turnover.

Dynamic regulation of Rh1 and TRPL in Trp^P365/+ photoreceptors

We measured the Rh1 content by Western blotting and show a drastic reduction in 1-day old Trp^P365/+. Moreover, the residual Rh1 displays slower mobility, suggesting defects in posttranslational modifications. It remains to be determined whether defective processing is responsible for the enhanced degradation and turnover of Rh1. Interestingly, a great reduction of Rh1 precedes deterioration of rhabdomeres, when the Rh1 content and the morphology of rhabdomere were analysed in the same fly. Rh1 plays a critical role in the maintenance of rhabdomeres; Rh1 is required for rhabdomere biogenesis by orchestrating the morphogenesis of the rhabdomere terminal web, the F-actin meshwork that supports the microvilli. Here we show a drastic reduction of Rh1 rhodopsin in adult photoreceptors also affects the maintenance of the rhabdomere leading to degeneration of Trp^P365/+ photoreceptors.

TRPL is affected by dysregulation of Ca²⁺ in a more complex manner. In wild-type, we observed that the steady-state TRPL level was much lower in white-eyed compared to red-eyed flies (Shieh et al., unpublished), suggesting that in white-eyed flies more light input down-regulates the TRPL content. This repression could be achieved at the transcriptional, translational, or posttranslational level. To maintain a lower steady-state level in white-eyed flies, TRPL may be constantly synthesized and degraded. Moreover, TRPL undergoes light-regulated endocytosis by trafficking to the intracellular compartments leading to light adaptation [46–48]. We show that TRPL was reduced in 1-day old, but greatly elevated in 3-day old Trp^P365/+, suggesting that TRPL is dynamically upregulated in response to the loss of Rh1. How light orchestrates the interplay of these regulatory mechanisms to regulate the TRPL level remains to be explored.

Role of CaMKII in promoting the survival of photoreceptors

A rise of cytoplasmic Ca²⁺ activates CaMKII, a master regulatory protein mediating the activity-dependent neuronal inputs to modulate diverse intracellular pathways [49]. Activated CaMKII triggers neuronal remodelling, and modulates synaptic plasticity [50]. In Drosophila photoreceptors, CaMKII catalyzes the phosphorylation of Arr2 [22,51] that regulates deactivation of activated rhodopsin. We show an up-regulation of CaMKII in Trp^P365/+ with increases in both total and activated CaMKII levels. Interestingly, suppression of CaMKII exacerbates retinal degeneration of Trp^P365/+ photoreceptors. CaMKII is likely to have multiple substrates in photoreceptors; CaMKII phosphorylates Arr2 to promote the dissociation of phosphorylated Arr2 from metarhodopsin [45]. Inhibition of CaMKII would result in the accumulation of Arr2-metarhodopsin complex accelerating retinal degeneration [44]. CaMKII also regulates apoptosis by phosphorylating the major initiator caspase, DRONC, at Ser¹³⁰, which diminishes its death-inducing activity [35]. Moreover, CaMKII may activate the inhibitor of apoptosis that negatively regulates DRONC leading to the Ca²⁺-dependent inactivation of the apoptotic pathways. Suppression of apoptosis by CaMKII is likely to delay retinal degeneration.

Role of DAG in regulating drosophila TRP

Drosophila TRP belongs to the TRPC subfamily, which also includes seven members in the human genome, TRPC1-7. Some of the TRPC channels are expressed in the brain and the heart and have been implicated in pathophysiology including neurodegenerative diseases and cardiac myopathy [52]. Most of mammalian TRPC channels are activated upon PLCβ stimulation, and in some cases, by DAG directly [53]. The role of DAG to open TRPC is also supported by recent Cryo-EM studies that revealed two lipid-binding pockets in the human TRPC3 [54]. While the role of DAG to activate mammalian TRPC has been firmly supported, how is Drosophila TRP gated remains controversial [55]? Potential second messengers include DAG [10], and fatty acids such as PUFAs [13,56]. Recently, it was suggested that TRP could be activated by a combination of the PIP₂ hydrolysis and proton, both of which are regulated by PLCβ4 [39]. It is possible that the gating of TRP may involve a combination of factors including DAG, protons, and the hydrolysis of PIP₂. It is important to note that TRP is tethering to INAD that orchestrates the formation of multimolecular protein complexes [57,58], which might facilitate its interaction with multiple second messenger(s) for regulating its gating. It is possible that hydrolysis of PIP₂ in the membrane may change the lipid microenvironment to sensitize TRP to DAG and protons.

Genetic evidence indicates that the gating of TRP requires PLCβ4 (NORPA), which generates DAG. We investigated whether DAG is critical in the opening of TRP via genetic interactions with either norpA^P24 or inaE^N125. Indeed, the degeneration of Trp^P365/+ photoreceptors is greatly delayed in the norpA^P24 background that fails to generate DAG. This finding supports the notion that DAG or its metabolites opens the TRP channel, which is also supported by studies in the rdgA mutant, that exhibits constitutive TRP activity [11]. The rdgA mutant is likely to have elevated DAG content as it lacks DAG kinase [59].

We examined how reduced DAG lipase activity (InaE) [13], either in the inaE^N125 genetic background or by RNAi-mediated knockdown, modifies the degeneration of Trp^P365/+ photoreceptors. A reduced DAG lipase activity may elevate the DAG level, which could prolong the activation of TRP accelerating retinal degeneration. Indeed, Trp^P365/+; inaE^N125 double mutants contain severely degenerated photoreceptors, which further deteriorated following eclosion. Moreover, knockdown of inaE in adult photoreceptors enhances the degeneration of Trp^P365/+ photoreceptors. Taken together, our findings support the notion that perturbation of DAG lipase may affect the DAG content to modify retinal degeneration. Genetic interactions between Trp^P365/+ and inaE^N125 were also reported in [13], they show a drastic reduced visual response in the double mutant. These authors proposed that TRP and DAG lipase may interact or subserve closely related functions.

Changes of DAG levels in norpA^P24 and inaE^N125 may modulate degeneration of Trp^P365/+ photoreceptors

To provide evidence supporting changes of the steady-state DAG level modulate retinal degeneration, we performed LC-MS to quantify various DAG species in rhabdomere membranes. Previously, it was reported that light treatment increases several unsaturated DAG including DAG-32:1, DAG-32:2, DAG-34:1, DAG-34:2, DAG-36:3 and DAG-36:4 in the rhabdomere, implicating these DAG species in the gating of TRP. However, it is not known which is the critical DAG species and the relative potency of the various unsaturated DAG.

Interestingly, we show a great reduction of both saturated (DAG-36:0) and unsaturated DAG (DAG-32:2, DAG-34:1, DAG-34:2, DAG-36:2, DAG-36:3, and DAG-36:4) in norpA^P24, compared to wild-type. This reduction would impact the activation of TRP channels delaying the progression of retinal degeneration. Similarly, we also quantified DAG in inaE^N125 and observed a moderate reduction of several DAG (DAG-32:2, DAG-34:1, DAG-34:2, DAG-36:3, DAG-36:0), but an increase of DAG-34:1. These findings are consistent with the notion that unsaturated DAG is critical for activating TRP.

Role of DAG lipase in photoreceptor development

We show that pupal photoreceptors of Trp^P365/+; inaE^N125 are severely degenerated with barely recognizable rhabdomere clusters, while those of inaE^N125 retain wild-type morphology. In contrast, the more severe alleles of inaE (xl29, xl15, and xl18) also contain smaller rhabdomeres [13]. These findings suggest that DAG lipase is important for rhabdomere biogenesis during photoreceptor development. inaE^N125 is a hypomorphic allele that retains about 70% of the InaE protein [13]. It is possible that the residual DAG lipase activity is negatively impacted in the Trp^P365/+ background leading to a drastic reduction of the critical fatty acid metabolites. A reduction of DAG lipase also transiently increases the content of DAG, which could be further converted to PA by RdgA [59]. It has been shown that elevated PA levels may interfere with membrane transport during the critical period of rhabdomere biogenesis [60]. Thus, DAG lipase/InaE appears to have dual functionalities in the eye; it fine-tunes the PA level to regulate rhabdomere formation and modulates the DAG content to regulate TRP in the visual signalling.

In summary, we demonstrate that the abnormal TRP channel activity in Trp^P365/+ likely affects Ca²⁺ homoeostasis impacting the maintenance of photoreceptors. Specifically, altered intracellular Ca²⁺ probably accelerates the turnover of Rh1 triggering degeneration of rhabdomeres. We propose that Ca²⁺ regulates the steady-state Rh1 level as activation of Rh1 increases the cytoplasmic Ca²⁺ to fine-tune the Rh1 content and activity (Figure 10). The Ca²⁺-dependent feedback regulation of Rh1 may involve CaMKII and RdgC, the rhodopsin phosphatase [45,61,62]. It might also involve CAMTA, a Ca²⁺/CaM regulated transcription activator, which turns on the expression of an E3 ubiquitin ligase to decrease Rh1 activity [63]. Lastly, we show that retinal degeneration of Trp^P365/+ is delayed in norpA^P24 flies, which appears accompanied by a reduced DAG content in the rhabdomere. In contrast, degeneration of Trp^P365/+ is exacerbated in inaE^N125 flies that contain elevation of an unsaturated DAG. Our findings support the notion that changes in DAG level affect the regulation of the TRP channel in Drosophila.

Figure 10.

Regulation of the Rh1 turnover by Ca²⁺ in Drosophila photoreceptors. In photoreceptors, activated Rh1 (Rh1*) couples to Gq to activate NORPA (PLCβ4), which generates DAG to open the TRP channel. the increase of intracellular Ca²⁺ activates CaMKII, RdgC, and possibly CAMTA, which participate in the modulation of Rh1 deactivation and turnover. shown is a schematic representation of the critical proteins and pathways

The abbreviations used are: aa, amino acids; Arr2, arrestin 2; CaMKII, calcium/calmodulin-dependent protein kinase II; DAG, diacylglycerol; dpp, deep pseudopupil; GFP, green fluorescent protein; inaC, inactivation-no-afterpotential C; INAD, inactivation-no-afterpotential D; inaE, inactivation-no-afterpotential E; IP₃, inositol trisphosphate; LC-MS, liquid chromatography-mass spectrometry; norpA, no-receptor-potential A; PA, phosphatidic acid; PIP₂, phosphatidylinositol-4, 5-bisphosphate; PLC, phospholipase C; PKC, protein kinase C; PUFAs, polyunsaturated fatty acids; rdgA, retinal degeneration A; Rh1, rhodopsin 1; Trp, transient receptor potential; TRPC, canonical or classical TRP; TRPL, TRP-like.

Funding Statement

This work was supported in part by the National Eye Institute [EY08126].

Disclosure statement

No potential conflict of interest was reported by the authors.