Maintenance of Rhodopsin levels in Drosophila photoreceptor and phototransduction requires Protein Kinase D

ABSTRACT

During Drosophila phototransduction, the G protein coupled receptor (GPCR) Rhodopsin (Rh1) transduces photon absorption into electrical signal via G-protein coupled activation of phospholipase C (PLC). Rh1 levels in the plasma membrane are critical for normal sensitivity to light. In this study, we report that Protein Kinase D (dPKD) regulates Rh1 homeostasis in adult photoreceptors. Although eye development and retinal structure are unaffected in the dPKD hypomorph (dPKD^H), it exhibited elevated levels of Rh1. Surprisingly, despite having elevated levels of Rh1, no defect was observed in the electrical response to light in these flies. By contrast the levels of another transmembrane protein of the photoreceptor plasma membrane, Transient receptor potential (TRP) was not altered in dPKD^H. Our results indicate that dPKD is dispensable for eye development but is required for maintaining Rh1 levels in adult photoreceptors.

Introduction

Drosophila phototransduction represents a proto-typical G protein coupled receptor (GPCR) signaling pathway whereby specialized sensory cells called photoreceptors transduce light energy into an electrical response that is transmitted to the brain.^1,2 The signaling pathway underlying phototransduction involves photoisomerization induced activation of GPCR molecule Rhodopsin (Rh1) to form meta-rhodopsin which in turn activates the enzyme phospholipase C (PLC). PLC breaks down the plasma membrane lipid phosphatidylinositol 4,5 bisphosphate (PIP₂) to produce diacylglycerol (DAG). The signaling downstream of DAG somehow results in opening of ion channels TRP and TRPL. Besides ion channel opening, DAG is converted into phosphatidic acid which is subsequently used to synthesize phosphatidylinositol (PI) on endoplasmic reticulum. The entire signaling pathways occurs at the interface of plasma membrane and endoplasmic reticulum, at specialized sites called PM-ER membrane contact site.³ The phototransduction output is regulated by multiple kinases and phosphatases which act by regulating the levels of the various phosphoinositide intermediates. The unequivocally known kinases include phosphatidylinositol-4-kinase (PI4K), phosphatidylinositol-4-phosphate-5-kinase (PIP5K), DAG kinase (DGK), Ca²⁺/Calmodulin Kinase,^4,5 Although several regulators of the phototransduction cascade have been identified, the list is still growing and novel regulators are being characterized.

One of the important signaling intermediates in phototransduction pathway is DAG. The precise signaling event by which DAG results in opening of ion channels is not clear. One of the potential effectors of DAG is Protein Kinase C (PKC). In mammals, one of the PKC isoforms is known to activate another effector kinase namely Protein Kinase D isoform1 (PKD1).⁶ This raises the possibility that PKD might have a role in photoreceptors. Unlike mammals which have three isoforms of PKD, Drosophila genome encodes for a single PKD gene which is ubiquitously expressed in all tissues during development and is required for regulating signal transduction, cell proliferation, membrane trafficking and secretion (reviewed in^7,8). Studies elucidating the regulatory mechanisms of PKD activation, its subcellular localization and the molecular details of its role in signaling pathways have been worked out.^7–9

In this study, we addressed the potential role of Drosophila PKD in phototransduction in adult photoreceptors. Our results show that 70% reduction in dPKD transcript (in dPKD hypomorph) in the photoreceptor cells leads to an increase in the total Rh1 levels. This alteration in Rh1 level is not underpinned by alteration in its transcript levels. This defect was specific to Rh1 and was not observed in case of another plasma membrane protein, TRP. Interestingly, despite having altered Rh1 levels the dPKD hypomorph exhibits normal light sensitivity and did not exhibit any defect in photo response. However, further reduction in transcript levels combined with continuous light exposure revealed significant reduction in photo response. Overall this study reflects that dPKD contributes to the maintenance of Rh1 levels in adult photoreceptors.

Materials and methods

Fly culture

All fly lines were reared at 25°C, 50% relative humidity either under normal or constant illumination wherever indicated. Illumination intensity was maintained at ~2000 lux throughout experimental conditions. Fly lines were reared on medium containing corn flour, sugar, yeast powder and agar along with antibacterial and antifungal agents. White-eyed (w¹¹¹⁸) and Red Oregon-R (ROR) were used as wild type controls according to the eye color of the experimental flies used. Following fly lines were used in this study – Rh1GAL4 (C Desplan, Rockefeller University), GMRGAL4 (BL1104),¹⁰ dPKD^H (BL 37604), dPKD^RNAi

Optical neutralization

Flies were anesthetized on ice, decapitated using a sharp blade, and fixed on a glass slide using a drop of colorless nail paint. Imaging was done using 40X oil objective of Olympus BX43 microscope. At least 5 flies of each genotype and condition were imaged.

Immunoblotting

Heads from flies hatched within 24hrs (unless otherwise specified) were decapitated in 2X SDS-PAGE sample buffer followed by boiling at 95°C for 5 min. For detection of rhodopsin, samples were incubated at 37°C for 30 min and then subjected to SDS-PAGE and western blotting. The following antibodies were used: anti-rhodopsin (1:250–4C5, Developmental Studies Hybridoma Bank, University of Iowa,USA), anti-α-tubulin (1:4000,E7c, DSHB) and anti-TRP (1:4000, Raghu Padinjat, NCBS). All secondary antibodies (Jackson Immunochemicals) were used at 1:10,000 dilution. Quantification of the blot was done using Image J software from NIH (Bethesda, MD, USA).

QPCR

qPCR was performed with the ABI 7500 Fast Real-Time PCR instrument (ABI). RNA was extracted from fly heads and subjected to cDNA conversion. A reaction mixture was setup with 2X SyBr Master mix (ABI), primer pairs specific for the transcripts (1:40 dilution of 100mM, primer standardization with ~100% efficiency) and template. Transcript levels of RP49 were used for normalization across samples. Minimum of 3 biological replicates were used for each genotype and duplicate measurement of each sample was performed.

Primers used-

1. RP49- CGGATCCGATATGCTAAGCTGT (F)/GCGCTTGTTCGATCCGTA (R)

2. dPKD- GCAGCTCACTGAATATCCCCGG (F)/TGTGCGTATTGTACTGGCAGTCCC (R)

3. Rh1- CATCATTGCTGCTGTCTCCGCC (F)/AGCGTGATGGTGACCACAGCCA (R)

Electrophysiology

External electrical recordings were done from the eye of the flies. Flies were anesthetized on ice and immobilized at the end of a disposable pipette tip using a drop of nail paint. The recording electrode (GC 100 F-10 borosilicate glass capillaries, 1 mm O.D and 0.58 mm I.D from Harvard) was filled with 0.8% w/v NaCl solution and was placed on the surface of eye and the reference electrode was placed on the neck region/thorax. Flies were dark adapted for 5min followed by ten repeated green light flashes of 2s duration, each after 10s interval. A fiber optic guide was used to deliver the light stimulus from a LED light source placed within a distance of 5mm from the fly eye. Calibrated neutral density filters were used to vary the intensity of the light over 5 log units. Voltage changes were amplified using a DAM50 amplifier (WPI) and recorded using pCLAMP 10.2. Analysis of traces was performed using Clampfit (Axon Laboratories). All recordings were done with eye color matched controls.

IR curve analysis was done as described in.³ Briefly, the flies were exposed to five different light intensities increasing by 1 log unit.

Statistical analysis

Raw data was initially processed using Microsoft Excel. Plotting and statistics of the processed data sets were performed using GraphPad Prism v5.04 for Windows OS. For all experiments, error bars represent SEM, and p values were calculated by using the Student’s unpaired t-test. * P < 0.05, ** P < 0.01, *** P < 0.001.

Results

Protein Kinase D is required for maintaining normal levels of Rhodopsin in adult photoreceptors

Regulation of transport of phototransduction protein is essential for photoreceptor survival and function. In order to figure out the potential role of dPKD in adult eye, we used a hypomorphic allele which shows ~70% reduction in dPKD transcript level (dPKD^H).⁸ dPKD^H did not exhibit any rough eye phenotype or gross structural eye defects. To access the ommatidial organization, we performed optical neutralization ¹¹ on flies that were reared in incubators having constant light illumination. As shown in (Figure 1(a)) newly eclosed (0 Day) dPKD^H did not show any sign of retinal degeneration. The rhabdomeral integrity was maintained even after 10 days of rearing in constant light (Figure 1(a)). These results reflect that depletion of dPKD does not impact photoreceptor development and overall rhabdomere structure. Next, we asked whether dPKD function is essential for photoresponse. Photoreceptors use Rhodopsin, a GPCR, present on the rhabdomeral membrane as receptors to detect light. We checked the protein levels of Rhodopsin 1 (Rh1), the major opsin of photoreceptors in dPKD^H by Western Blotting. As shown in (Figure 1(b) (i), (ii)), compared to wild type flies of matched eye color, dPKD^H showed higher levels of Rh1. Since Rh1 is known to undergo massive trafficking during late pupal stage, ¹² we reared dPKD^H under constant light throughout development and checked Rh1 levels in newly eclosed flies. It is known that upon constant light rearing, the Rh1 levels in wild type retinae gets reduced but as shown in (Figure 1(c) (i), (ii)), even under strong light stimulation, dPKD^H maintained elevated levels of Rh1 compared to wild type under similar conditions. We next checked if the elevated Rh1 level in dPKD^H was due to enhanced transcription of this gene. As shown in (Figure 1(d), there was no change in the transcript levels of Rh1 in dPKD^H flies.

Figure 1.

(a) Optical neutralization images of dPKD^H compared to Wild Type flies grown at Normal light-0 day (newly eclosed) and Constant Light- 10 days. N = 7 flies per genotype and condition. (b) Western Blot and quantitation for Rhodopsin in dPKD^H compared to Wild Type at Normal light conditions (Incubators without internal illumination). Rh1 levels are normalized to Tubulin. N = 3 with 5 flies. Protein equivalent of 2 heads loaded per lane. (c) Western Blot and quantitation for Rhodopsin in dPKD^H compared to Wild Type at Constant light conditions. Rh1 levels are normalized to Tubulin. N = 3 with 5 flies. Protein equivalent of 2 heads loaded per lane. (d) Relative mRNA levels of Rhodopsin in dPKD^H compared to Wild Type. NinaE¹¹⁷ was used as a negative control for Rh1. N = 3, 30 heads per genotype. (e) Western Blot and quantitation for TRP in dPKD^H compared to Wild Type at Constant light conditions. TRP levels are normalized to Tubulin. N = 3 with 5 flies. Protein equivalent of 2 heads loaded per lane.

Next, we asked whether such perturbation in the rhabdomeral membrane protein was specific to Rh1 or was applicable to other plasma membrane proteins involved in phototransduction. We checked the level of another membrane protein transient receptor potential (TRP) and found it to be unchanged in dPKD^H compared to wild type (Figure 1(e). This result reflects that loss of dPKD activity specifically affects levels of Rh1 and not other photoreceptor membrane proteins in adult photoreceptors.

Consequences of elevated Rhodopsin on phototransduction in dPKD^H

Since Rhodopsin acts as the receptor for light, we went ahead to check the impact of elevated Rh1 on the functional output of photoreceptors. We did extracellular recordings from the eye (electroretinogram-ERG) to measure the electrical response to light. The ERG response of white eyed wild type flies shows a pattern where the response to first flash of light is usually higher and shows gradual reduction in value in response to subsequent flashes of light. Compared to wild type of same age, dPKD^H showed significantly higher response to first flash of light stimulation, although the response to subsequent 9 flashes were similar (Figure 2(a). Although the response to first flash of light in dPKD^H was higher than wildtype, the percent reduction in their response to subsequent nine flashes was not statistically different (Figure 2(b). Thus, the higher response to first flash of light recorded in dPKD^H seems to represent a subtle phenotype. Lack of a robust ERG defect on the face of elevated levels of total rhodopsin was intriguing and led us to perform another electrophysiological assay which directly correlates with the levels of receptor on plasma membrane. In this assay, the flies were exposed to logarithmically increasing light intensities. An Intensity-Response (IR) curve was plotted (for details see materials and methods) for both wild type and dPKD^H. As shown in (Figure 2(c) the IR curve of wild type and dPKD^H showed no significant difference. These results reflect that despite having higher rhodopsin levels, dPKD^H shows no defect in light sensitivity.

Figure 2.

(a) ERG response to 1st flash of light(1st response) and average of 9 subsequent responses to light in dPKD^H compared to Wild Type. N = 6–8 flies per genotype. (b) Percentage change in response across 10 flashes of light stimulations in dPKD^H compared to Wild Type. N = 4 flies per genotype. (c) Intensity-Response (IR) curve for both Wild Type and dPKD^H. N = 5 flies per genotype.

Photoreceptor specific reduction of dPKD recapitulates dPKD^H phenotypes

dPKD^H was generated by Minos transposon insertion, to rule out any potential strain associated impact in our observation we used other allelic combinations to reduce dPKD levels. To have a better understanding we switched from the pan-tissue hypomorphic allele (dPKD^H) to eye specific depletion of dPKD. We knocked down dPKD in outer photoreceptor cells using Rh1Gal4 as well as GMRGal4. GMRGal4 expression driver expresses throughout development, including both photoreceptors and supporting, non-neuronal cell whereas Rh1Gal4 expression driver is activated shortly prior to eclosion of adult flies within mature, fully differentiated photoreceptors only and following patterning and morphogenesis of the eye. As shown in (Figure 3(a,b)), RNAi mediated knockdown of dPKD resulted in higher Rh1 levels. We confirmed the level of dPKD knockdown in these flies by Q-PCR (Figure 3(c). To exclude any RNAi associated off-target effects, we used additional dominant negative constructs of dPKD (dPKD^KD) where Lysine-572 is replaced with Tryptophan thus, destroying the presumptive ATP binding site and making the protein kinase-dead.⁸ Similar effects were observed in GMR>dPKD^KD i.e. Rh1 levels were elevated without having any effect on TRP levels (Figure 3(d,e)). Altogether, these results testified increased rhodopsin levels in photoreceptors having reduced dPKD activity. This intrigued us to revisit the potential role of PKD in supporting phototransduction. It was possible that the remaining 30% dPKD expressed in dPKD^H was sufficient to support its function hence no ERG defect was observed. To further reduce the levels of dPKD we combined the hypomorph with the RNAi line and tested the ERG response. These flies showed slightly reduced response compared to controls, when reared in incubators with no internal illumination, although the difference was statistically not significant (Figure 3(f). To test if this subtle difference observed held any significance, we exposed the flies to enhanced stimulation by rearing them in incubators having constant light illumination for 6 days post eclosion. Under these conditions GMR>dPKD^RNAi;dPKD^H showed significantly reduced response (Figure 3(g). These flies also showed elevated levels of Rh1 compared to GAL4 control (Figure 3(h). Thus, the reduced ERG response phenotype observed was accompanied by elevated Rh1 levels.

Figure 3.

(a) Western Blot for Rhodopsin in GMRGal4>dPKD^RNAi compared to Wild Type controls at Normal light conditions. N = 3 with 5 flies. Protein equivalent of 2 heads loaded per lane. (b) Western Blot Rhodopsin in Rh1Gal4>dPKD^RNAi compared to Wild Type controls at Normal light conditions. N = 3 with 5 flies. Protein equivalent of 2 heads loaded per lane. (c) Relative mRNA levels of dPKD in GMRGal4>dPKD^RNAi compared to Wild Type controls. N = 3, 30 heads per genotype. (d) Western Blot for Rhodopsin in (i) GMRGal4>dPKD^KD and (ii) Rh1Gal4>dPKD^KD compared to Wild Type controls at Normal light conditions. N = 2 with 5 flies. Protein equivalent of 2 heads loaded per lane. (e) Western Blot for TRP in (i) GMRGal4>dPKD^KD and (ii) Rh1Gal4>dPKD^KD compared to Wild Type controls at Normal light conditions. N = 2 with 5 flies. Protein equivalent of 2 heads loaded per lane. (f) Average peak amplitude (in mV) in GMRGal4>dPKD^RNAi,dPKD^H/dPKD^H compared to control (GMRGal4>dPKD^H/+) in Normal light conditions. N = 5 flies. (g) Average. peak amplitude (in mV) in GMR>dPKD^RNAi,dPKD^H/dPKD^H compared to control (GMRGal4>dPKD^H/+) in Constant light conditions. N = 5 flies. (h) Western Blot for Rhodopsin in GMRGal4>dPKD^RNAi,dPKD^H/dPKD^H compared to controls. N = 2 with 5 flies. Protein equivalent of 2 heads loaded per lane.

Conclusion and outlook

This study focused on unraveling the potential function of a Serine/Threonine kinase, PKD in Drosophila photoreceptors. Most of the studies done on understanding the role of PKD have been performed ex-vivo in cells or cell-based systems.^7,13 In another study, we have characterized the role of dPKD in Drosophila growth using a hypomorphic allele (dPKD^H).⁸ We used dPKD^H to address the role of dPKD in adult photoreceptors. Our results show that reducing dPKD transcript levels by 70% did not result in any defect in photoreceptor development and ommatidial organization. However, higher level of total Rh1 was observed without any change in its transcript levels. The alteration in the protein level was specific for Rh1 and was not observed in case of another plasma membrane protein, TRP. Any perturbation in the levels of Rh1 is likely to affect the phototransduction output, ² however, we did not notice any robust defect in the light response and light sensitivity in dPKD^H. The fact that the light sensitivity (measured by IR curve) of these flies was not altered seems to indicate that the increased amount of Rh1 present within the photoreceptor cell was not participating in the signaling cascade. Since phototransduction machinery is housed in the apical plasma membrane, it is possible that the elevated Rh1 protein level in dPKD^H seen on Western blots was not present at the plasma membrane else it would have contributed to altered light sensitivity. Cellular imaging of rhodopsin along with other phototransduction molecules in photoreceptors will help in revealing the localization of the pool(s) of Rh1 which is contributing to higher protein level.

Although we did not notice any ERG defect in dPKD^H, there is a possibility that the remaining 30% transcript available in the dPKD^H was sufficient to support photo response. This is further supported by the fact that we observed a reduced ERG response only when we combined dPKD^RNAi and dPKD^H and exposed the flies to enhanced stimulation (rearing in constant light exposure for 6 days post eclosion). To further analyze dPKD function in photo response, an eye specific dPKD null allele could be helpful.

Altogether, this study implicates a role of dPKD in regulating Rh1 levels in photoreceptors. However, the underlying mechanism is not clear and following lines of questions need to be addressed (Model, Figure 4) (1) Whether dPKD is required for the transport of newly synthesized Rh1 from Golgi to plasma membrane? In mammalian cells, dPKD function is required in the Golgi apparatus for transport of cargoes to the plasma membrane,¹⁴ which raises a possibility that Rh1 transport might be affected upon PKD inhibition,^12,15,16 To answer this, Rh1 transport during early pupal developmental stages in dPKD inhibition needs to be revisited. Rh1 not only has signaling roles but structural roles as well.^17–19 Interestingly, photoreceptor degeneration was not obvious in dPKD mutants as determined by rhabdomere morphology, similar to NinaA mutant.²⁰ Lack of retinal degeneration might also indicate that Rh levels at plasma membrane might not be compromised severely. (2) Where does the elevated pool of Rh1 reside in dPKD^H? Localization of rhodopsin loaded vesicles (RLVs) population within photoreceptors would provide further insight. We expect higher RLV count in dPKD^H which are likely outside the rhabdomeral membrane. Molecular understanding of this defect would also require characterization of the nature of RLVs, whether they undergo degradation post accumulation through the autophagosomal pathways or (3) they escape degradation and do recycle back through the recycling endosomes to plasma membrane with delayed kinetics. The fact that the defect is observed only under conditions of prolonged enhanced stimulation indicates that either dPKD is required to support Rh cycle only during fast signaling or the amount of active dPKD in the strains used in this study was sufficient to support function. In light of this, the genetic interaction between dPKD and other members of Rh1 transporting machinery like Crumbs,²¹ Retromer complex²² and highroad²³ needs to be studied.

Figure 4.

Concluding Model. Dotted arrows represent steps in Rh1 signaling which might require dPKD activity.

Funding Statement

This work was supported by the National Centre for Biological Sciences-TIFR.

Acknoweldgments

This work was supported by the National Centre for Biological Sciences-TIFR. Stocks were obtained from the Bloomington Drosophila Stock Center (NIH P40OD018537). We would also like to thank Prof. Raghu Padinjat, NCBS-TIFR, for providing lab space and ERG setup used in this study. We acknowledge the generous support of numerous colleagues who provided us with fly stocks and antibodies. We thank the NCBS Fly facility and NCBS-CIFF for support. SA is supported by fellowship from CSIR. SY is supported by fellowship from NCBS-TIFR.

The, monoclonal antibody was obtained from the Developmental Studies Hybridoma Bank, created by the NICHD of the NIH and maintained at The University of Iowa, Department of Biology, Iowa City, IA 52242.

Author contribution

SA-Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting and revising the article.

SY- Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting and revising the article.

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.