Drosophila Fatty Acid Transport Protein Regulates Rhodopsin-1 Metabolism and Is Required for Photoreceptor Neuron Survival

Tight regulation of the visual response is essential for photoreceptor function and survival. Visual response dysregulation often leads to photoreceptor cell degeneration, but the causes of such cell death are not well understood. In this study, we investigated a fatty acid transport protein (fatp) null mutation that caused adult-onset and progressive photoreceptor cell death. Consistent with fatp having a role in the retina, we showed that fatp is expressed in adult photoreceptors and accessory cells and that its re-expression in photoreceptors rescued photoreceptor viability in fatp mutants. The visual response in young fatp-mutant flies was abnormal with elevated electroretinogram amplitudes associated with high levels of Rhodopsin-1 (Rh1). Reducing Rh1 levels in rh1 mutants or depriving flies of vitamin A rescued photoreceptor cell death in fatp mutant flies. Our results indicate that fatp promotes photoreceptor survival by regulating Rh1 abundance.


Introduction
Retinal degeneration is a major health concern that affects one in 2000 people worldwide (http://www.sph.uth.tmc.edu/Retnet/) [1]. Although human retinopathies are heterogeneous in physiopathology and severity, they all involve loss of photoreceptor (PR) neurons, which leads to blindness. The most frequent retinal disease is retinitis pigmentosa, which is caused by mutations in one or more of at least 54 distinct genes (Retnet). Among them, the most frequently mutated gene is rhodopsin (rho in mammals) which is mutated in 30-40% of all cases of autosomal dominant retinitis pigmentosa (ADRP) (Retnet) [2]. Rhodopsin is the light-sensitive protein of PRs that activates phototransduction. Approximately 100 rho mutations have been identified, and they affect folding, trafficking and activity of the rhodopsin protein. Despite extensive study, the mechanisms of retinal degeneration remain unclear.
Retinal degeneration has been studied extensively in Drosophila [3,4,5]. Many mutations in Drosophila phototransduction pathway genes induce PR degeneration. The mechanism of toxicity of these mutations is related either to a defect in folding, trafficking or activity of the Drosophila Rhodopsin-1 protein (Rh1), or to an accumulation of toxic Rh1-Arrestin2 (Arr2) complexes, or to a deregulation of the Ca 2+ homeostasis [4,5]. In addition, mutations can be introduced in Drosophila genes to model human diseases. For example, the rh1 P37H allele, which corresponds to rho P23H , the most frequent mutation of rhodopsin in ADRP, has been successfully introduced into Drosophila and induces PR degeneration [6]. We recently used Drosophila to identify genes involved in PR survival [7]. Using a new method of PR visualization, called Tomato/ GFP-FLP/FRT, we screened recessive lethal mutations and found fatp k10307 to be associated with PR cell death. The fatp k10307 mutation consists of the insertion of a P{lacW} element into the open reading frame of the fatty acid transport protein (fatp) gene, which has never been characterized in Drosophila. Its closest mammalian orthologs are fatp1 and fatp4 [8]. The mammalian members of the Fatp family of proteins have acyl-CoA synthetase enzymatic activity and facilitate cellular fatty acid uptake [9,10]. Each member of the family has a specific expression pattern and function. fatp1 is expressed in muscle, heart, brain, adipose tissue and retina [11,12], and is involved in thermogenesis and obesity. fatp4 is expressed most abundantly in the small intestine, brain, skeletal muscle, heart, skin, liver and kidney [10,13,14]. Loss of fatp4 in mice or in humans is associated with restrictive dermopathy related to the ichthyosis prematurity syndrome, a skin defect caused by altered lipid and fatty acid compositions [15,16,17].
In this work, we show that Drosophila fatp is required for PR survival. PR degeneration in a fatp null mutant is adult-onset and progressive. The onset of degeneration in the fatp mutant correlates with the time of expression of fatp in the normal adult retina. We then investigated the mechanisms of PR degeneration in the fatp mutant. We show that PRs in the fatp mutant exhibit an elevated photoresponse that is associated with high levels of Rh1. We also found that fatp mutant flies have a defect in Rh1 degradation and that reducing the level of Rh1 restored PR survival.

fatp is required for PR viability in adult Drosophila
To study the role of fatp in PR viability, we first used RNA interference to reduce fatp expression. fatp-interfering RNA was specifically produced in the retina under the control of eye-specific drivers (ey-Gal4 and GMR-Gal4) ( Figure 1A-1G). fatp knockdown led to a progressive loss of PRs, indicating that fatp expression is required for PR viability. To validate these findings, we used the fatp k10307 mutation, generated by the insertion of a 10.7 kb P{lacW} element into the first exon of the gene. This mutation is recessive and can be considered a null allele as it completely prevents fatp expression, abolishing the production of fatp mRNA and protein in homozygous fatp k10307 L1 larvae ( Figure S1). As a consequence, fatp k10307 mutant larvae died during the second instar. To study the role of fatp in adult PR and circumvent fatp k10307 larval lethality, we used the Tomato/GFP-FLP/FRT mosaic method [7]. This new method combines mitotic recombination and cornea neutralization techniques [18,19,20] to allow time-course analysis of homozygous mutant clones in living flies over several days. We evaluated PR presence, identified on the basis of GFP expression, for 14 days from hatching in fatp k10307 mutant retinas ( Figure 1H-1J). All fatp mutant PRs were present upon hatching but progressively disappeared starting on day four ( Figure 1H, 1H9). Losses of fatp mutant PRs were statistically significant in the retinas of 8-and 14-day-old adults ( Figure 1H0, 1H90, 1I). In contrast, little or no PR loss occurred in neighboring wild-type or heterozygous tissue, indicating that the fatp k10307 mutation was cell-autonomous ( Figure 1H, 1I, 1J).
The cell-autonomous nature of the PR loss in the fatp mutant ( Figure 1H-1I) suggested that fatp expression is required for PR viability. To confirm this requirement, we attempted to rescue PR viability in the fatp k10307 mutant with tissue-specific expression of wild-type fatp using the UAS/GAL4 system [21]. fatp re-expression from the rh1 promoter in the R1-R6 PRs fully rescued the fatp k10307 mutant PR phenotype ( Figure 1K-1N). These results demonstrate that fatp expression in PRs is required for their survival.
Next, we performed structural and ultra-structural analyses of fatp mutant PR in resin-embedded retina sections. We used classical brightfield microscopy to examine PR integrity in wholeeye fatp mutant clones generated by the EGUF/Hid method [22] ( Figure 2A-2E). PR loss started after approximately 15 days and increased progressively with age ( Figure 2A-2E). Using this approach, PR loss was detected later than with the Tomato/ GFP-FLP/FRT method. This may be because the Tomato/GFP-FLP/FRT method is based on the expression and targeting of GFP in the rhabdomere and thus detects early deficits whereas classical histological methods assess only the physical presence or absence of PRs. We also detected progressive and adult-onset PR loss by recording the number of nuclei between the apical and basal layers of fatp mutant retinas in horizontal cryosections ( Figure S2). We then studied fatp mosaic retinas by electron microscopy (EM) to characterize further PR degeneration in fatp mutant cells ( Figure 2F-2M). On EM images, we could distinguish homozygous fatp mutant PRs from the absence of pigment vesicles at the base of the rhabdomeres and in inter-ommatidial cells (IOCs) ( Figure 2F). We observed several levels of PR degeneration among fatp mutant cells ranging from normal PRs to PRs that were fully degenerated and engulfed by neighboring IOCs. Most normal PRs displayed no obvious sign that would predict future degeneration ( Figure 2G). Some PRs were electron-dense and their cytoplasm was contracted suggesting that they had initiated the degeneration process ( Figure 2H). In these PRs, some swelling mitochondria could be observed but the rhabdomeres were still intact. In more advanced stages of degeneration, the rhabdomeres were clearly affected with disorganized microvilli structures ( Figure 2I, 2J). In addition, the neighboring IOCs seemed to be activated with a typical spotted pattern in the cytoplasm. At the end of the degenerative process, PRs were phagocytosed by the neighboring IOCs and apparently digested ( Figure 2K-2M). Thus, the absence of fatp causes progressive adult-onset PR degeneration.

fatp is expressed in the adult retina
We determined the specificity of fatp expression in developing and adult Drosophila eyes. To do this, we generated an anti-Fatp antibody (C11-7) that was sensitive enough to detect ectopic Fatp and endogenous Fatp on western blots and in immunofluorescence experiments ( Figure 3A-3E). No fatp expression was observed in third instar eye imaginal discs ( Figure 3B), but we detected Fatp protein in PR cytoplasm juxtaposed to PR rhabdomeres and in IOCs of the adult retina ( Figure 3D, 3E). To capture the fatp expression profile, we exploited the fatp k10307 enhancer trap line, which carries the b-galactosidase reporter ( Figure S3). Using a bgalactosidase antibody, we detected b-galactosidase in the nuclei cell layer that includes the outer PRs and accessory cells in adult head cryosections ( Figure S3A-S3D). Immunolocalization experiments revealed that small amounts of Fatp consistently colocalized with the neuronal marker ELAV in the outer PRs and that Fatp was also present in apical IOCs ( Figure S3A-S3D). Thus, the fatp k10307 enhancer trap line faithfully reproduces the distribution of Fatp and can be used to follow fatp expression. Testing for b-galactosidase activity in this line revealed that fatp was also expressed in the adult optic lobe around the medulla and lamina, in the midgut and in the salivary glands but not in the eye imaginal disc of third instar larvae ( Figure S3E-S3L). In conclusion, the absence of fatp expression in the developing eye

Author Summary
Normal vision requires precise regulation of the visual response. The deregulation of the visual response can lead to retinal diseases and blindness. The most frequent retinal disease is retinitis pigmentosa, and in 30%-40% of such cases rhodopsin, the light-sensitive protein, is mutated. Approximately 100 rhodopsin mutations have been identified; they affect folding, trafficking, and activity of the rhodopsin protein and induce consequent photoreceptor neuron death. Despite extensive studies, many aspects of retinal degeneration remain unclear. Drosophila is a suitable model organism to study retinal diseases. To understand the mechanisms of retinal degeneration, we studied the previously uncharacterized fatty acid transport protein (fatp) gene in Drosophila. We found that flies that lack the fatp gene exhibit a deregulation of the visual response and an adult-onset and progressive retinal degeneration. In addition, we show that such retinal degeneration is due to the death of photoreceptor neurons in which rhodopsin proteins accumulate. In summary, we report a novel fly model of adult-onset retinal degeneration. We uncovered a novel and interesting mechanism by which Fatp, a potential regulator of lipid transport and metabolism, is responsible for the regulation of rhodopsin levels in the photoreceptor neurons.

fatp regulates the visual response
We investigated whether the requirement for fatp expression for PR viability is related to the visual response. First, we tested whether PR degeneration in the fatp mutant is light-dependent. fatp mutant flies reared in normal light conditions exhibited significantly greater PR losses than fatp mutant flies reared in complete darkness ( Figure 4A-4G). This indicates that light contributes to PRs loss in the fatp mutant. Then, to examine directly the visual response in the fatp mutant, we performed electoretinogram (ERG) recordings on white-eyed flies ( Figure 4H, 4I): we used 8-day-old flies, an age at which the integrity of fatp mutant PRs is still intact as observed in resin-embedded retinal sections (data not shown). We obtained ERG recordings that measure the summed responses of all retinal cells. Wild-type flies exhibited a corneal negative receptor potential in response to orange light, which returned to baseline when the light was switched off ( Figure 4H). The amplitude of the ERG was higher in fatp k10307 flies than controls, with no apparent difference in the kinetics of the visual response ( Figure 4H, 4I). These results suggest that Fatp is a negative regulator of the visual response.

The elevated levels of Rh1 cause PR loss in the fatp mutant
We wondered whether PR degeneration in the fatp mutant could be rescued by rh1 mutant alleles as previously described for light-dependent retinal degeneration mutants such as rdgC and norpA [23]. We therefore tested whether PR viability in the fatp mutant could be rescued by rh1 G69D and rh1 I17 (ninaE G69D and ninaE I17 ) alleles ( Figure 5 and Figure S5). rh1 G69D is a dominant negative allele whereas rh1 I17 is a null allele. Although the rh1 G69D allele causes retinal degeneration in old flies under constant light exposure, it could be used in our study because these mutants raised under a 12 h day/light cycle did not exhibit retinal degeneration ( Figure 5C, 5H, 5M; Figure S4; and [23,24]). Using the Tomato/GFP-FLP/FRT method, we compared PR loss in retinas in the single fatp k10307/k10307 mutant to that in fatp k10307/k10307 rh1 G69D/+ double mutant. PR loss was very much lower in fatp k10307/k10307 rh1 G69D/+ mutant retinas than in fatp k10307/k10307 mutant retinas ( Figure 5A-5E). In resin-embedded eye sections, PR viability was found to be fully rescued in the fatp k10307/k10307 rh1 G69D/+ double mutant ( Figure 5F-5J). EM showed that despite the smaller size of the rhabdomeres, PR ultra-structure was fully rescued in fatp k10307/k10307 rh1 G69D/+ double mutants ( Figure 5K-5N). Similarly, a rescue of PR degeneration was observed in fatp k10307/k10307 rh1 I17/+ double mutant ( Figure S5). Thus, rh1 mutant alleles rescued PR degeneration in the fatp mutant.
We tested whether rh1 is dysregulated in fatp mutant PRs. We assayed Rh1 protein in fatp mutant and control retinas by western blotting (Figure 6A, 6B). The Rh1 content of fatp k10307/k10307 mutant retinas was double that of control retinas. We confirmed that Rh1 was less abundant in fatp k10307/k10307 rh1 G69D/+ double mutant than fatp k10307/k10307 single mutant retinas ( Figure 6A, 6B). These results indicate that Rh1 metabolism is altered in the fatp mutant and that reducing Rh1 levels in fatp mutants protects the PRs. To confirm this conclusion, we examined whether rearing flies on a vitamin A-deficient medium could reduce PR loss in fatp mutant retinas ( Figure 6C-6F). Vitamin A is the precursor of the Rh1 chromophore and is required for Rh1 synthesis. A vitamin Adeficient diet rescues retinal degeneration in several mutants, including rdgB, crumbs, and arrestin mutants [25,26,27]. In retinas of flies reared on a vitamin A-deficient diet, occasional PR degeneration was visible but most rhabdomeres were present ( Figure S6). fatp mutant flies deprived of vitamin A produced no detectable Rh1 protein and the PR loss in these flies was very much lower than that in flies receiving a standard diet ( Figure 6C-6F). Thus, reducing the Rh1 level rescued PR degeneration in the fatp mutant. Therefore, PR degeneration in the fatp mutant appears to be due to an over abundance of Rh1.
To test whether high levels of Rh1 are toxic to the PR, we overexpressed rh1 in developing and adult PRs ( Figure S7). Ectopic expression of rh1 with the GMR driver led to the rough eye phenotype, indicating that rh1 overexpression is toxic in the developping eye ( Figure S7A, S7B). We also overexpressed rh1 specifically in differentiated R1-6 PRs using the rh1 promoter. rh1 overexpression led to rhabdomere degeneration ( Figure S7C, S7D). Thus, we conclude that elevated Rh1 levels in fatp mutant PRs can lead to PR pathogenesis.
Next we examined the possible involvement of Arrestin2 (Arr2) in Rh1 toxicity in fatp mutants. Stable Rh1-Arr2 complexes are toxic in adult PRs and are responsible for retinal degeneration in several mutants including norpA, rdgC, rdgB [26,28,29,30]. To examine this possibility, we tested for genetic interaction between fatp k10307 and arr2 3 mutations ( Figure 6G-6J). We found that PRs were rescued in fatp and arr2 double mutant retina, indicating that arr2 is required for the toxicity in fatp mutant PRs. Together with the excess of Rh1 detected in fatp mutant, these observations suggest that toxic Rh1-Arr2 complexes induce PR degeneration in the fatp mutant.
We tested whether the elevated Rh1 levels in fatp mutant PRs were due to deregulation of rh1 expression and/or degradation of Rh1. We first analyzed rh1 expression in fatp mutant retinas using a rh1-lacZ reporter to monitor rh1 transcription. We did not detect higher levels of lacZ in fatp mutant retinas than in control retinas ( Figure S8A, S8B). This shows that rh1 transcription is not upregulated by the fatp mutation. Presumably, therefore, fatp regulates Rh1 levels post-transcriptionally. To test for the involvement of fatp in Rh1 degradation, we forced Rh1 degradation by illuminating the retina with blue light as previously described [31]. Blue illumination maintains Rh1 in an active conformation and induces its degradation. In white-eyed control flies blue light illumination for 6 h induced a significant loss of Rh1 for the retinas whereas the same treatment of white-eyed fatp mutants did not result in decreased Rh1 abundance ( Figure S8C, S8D). This suggests that light induced-Rh1 degradation is impaired in fatp mutant PRs and this may explain why there is more Rh1 in these retinas than controls.

Discussion
Mutations resulting in inactive Rh1, impaired visual responses and PR degeneration have been studied extensively [4,5]. In this study, we describe the phenotype of fatp k10307 , the first mutation known to exhibit elevated Rh1 levels leading to loss of PRs.
We show that fatp expression is required for PR viability in adult Drosophila. In the absence of fatp, PRs degenerate progressively fatp k10307/k10307 +rh1.fatp mosaic retinas and (N) quantification (t-test, n. = 5). In the control (K), all PRs were present. In fatp mutant mosaic retina (L), 70% of the mutant PRs were lost. In these retinas, mutant PRs were rescued by re-expression of fatp under the control of the rh1 promoter (M). doi:10.1371/journal.pgen.1002833.g001 during adulthood. Moreover, we demonstrate that the requirement for fatp in adult PRs is cell-autonomous, which is in agreement with the presence of Fatp in adult PRs and with its absence from the developing eye imaginal disc. The agedependent PR degeneration in the fatp mutant is reminiscent of Drosophila models of ADRP [3,6]. We thus propose that fatpassociated degeneration is a new model of late-onset PR degeneration.
Our results indicate that PR death in fatp mutants is a consequence of elevated levels of Rh1. We demonstrate that reducing Rh1 levels, as a consequence of rh1 mutation or a vitamin A-deficient diet, efficiently restored PR viability in fatp mutants ( Figure 5, Figure 6, and Figure S5). Thus, the accumulation of Rh1 is toxic for the PRs in fatp mutants. One possibility is that Rh1 associates with Arr2, forming toxic Rh1-Arr2 complexes as in norpA, rdgC, rdgB mutants [26,28,29,30]. Indeed, we found that disrupting either rh1 or arr2 rescued PR loss in fatp mutants ( Figure 5, Figure 6, and Figure S5). This genetic evidence is consistent with Rh1-Arr2 complexes causing PR degeneration in fatp mutant. Definitive proof of this mechanism requires the direct assessment of Rh1/Arr2 complexes in fatp mutants. Also, we cannot exclude the existence of additional toxic mechanisms. For example, the elevated visual response in fatp mutants (Figure 4) may contribute to PR death because of a defect in Ca 2+ homeostasis.
We found that Rh1 protein levels were elevated in fatp mutant retinas and that this was probably due to decreased Rh1 degradation ( Figure S8C, S8D). fatp may regulate sphingolipid metabolism, which controls Rh1 trafficking and degradation [32,33]. In support of this hypothesis, the total ceramide content is higher in the skin of fatp4 2/2 than control mice [15]. Ceramidase facilitates the endocytic turnover of Rh1 and rescues retinal degeneration in arr2 and phospholipase C mutants [32,34]. Similarly, Drosophila Fatp may limit ceramide levels and inhibit endocytic turnover of Rh1. It is also possible that fatp regulates Rh1 synthesis, but we show that loss of fatp does not affect rh1 gene transcription ( Figure S8A, S8B). Alternatively, Fatp may regulate the synthesis or recycling of the retinal chromophore required for Rh1 synthesis. In mammals, Fatp1 inhibits two enzymes of the visual cycle in vitro, LRAT and RPE65, which respectively produce and consume retinyl-ester, a fatty acid-linked form of the chromophore [12]. A similar visual cycle was recently described in Drosophila [35,36,37]. Therefore, chromophore synthesis or recycling may be increased in the fatp mutants, resulting in upregulated Rh1 synthesis. In support of this hypothesis, we showed that inhibiting chromophore synthesis in vitamin Adeficient medium fully rescued PR viability in fatp mutant retinas ( Figure 6D-6F). Nevertheless, the mechanisms by which fatp may regulate the visual cycle remain to be elucidated.
The visual response was higher in the fatp mutant retina than control retina but the mechanisms involved are unclear. Previous work has shown that the ERG amplitude depends on the ratio between the level of Rh1 and the Rh1 kinase activity of gprk1 [38]: decreasing gprk1 activity resulted in higher ratios and elevated ERG amplitudes whereas increasing gprk1 activity resulted in lower ratios and lower ERG amplitudes [38]. Whether there is a disequilibrium between phosphorylated/unphosphorylated forms of Rh1 in fatp mutant retinas remains to be explored. Alternatively, fatp may be required for the production of lipid metabolites that regulate the phototransduction cascade. Indeed, it has been suggested that polyunsaturated fatty acids, which are potential diacyl-glycerol metabolites, act on TRP/TRPL channels [39,40,41]. Thus, the elevated visual response may be the consequence of a lipid metabolite dysregulation in fatp mutant retinas.
In conclusion, fatp mutation is a new model of retinal pathology in flies in which the up-regulation of Rh1 contributes to progressive PR degeneration. Whether a similar pathological mechanism exists in human retinal diseases remains to be determined.

Generation of UAS-fatp transgenic line
fatp cDNA (SD05207, Gold cDNAs Collection) was recovered from BDGP DGRC in pOT2 vector. fatp cDNA was cloned (XhoI/ EcoRI) into a pUAST-w+-attB transgenic fly vector. Best Gene, Inc (CA, USA) generated transgenic lines using PhiC31 integrasemediated transgenesis [44]. The vector DNA was injected in embryos carrying attP docking sites (strain 9750 at 65B2). w+ embryos were selected and stable transgenic fly stocks established.  Live fluorescent imaging of PRs CO2-anesthetized flies were placed in a 35 mm cell culture dish half-filled with 1% agarose, covered with water at 4uC and observed using an upright 510 Zeiss confocal fluorescent microscope as described [20]. For the time-course study of agedependent PR death in single flies, after visualization, each living fly was detached from the agarose, dried and transferred to a vial containing fly food medium.

Resin-embedded tangential sections
Tangential sections of adult eyes were performed as described [45]. PR viability was determined by counting the number of intact rhabdomeres on retina tangential plastic sections. At least 200 ommatidia from three different animals were scored per experimental condition.

Transmission electron microscopy
Drosophila eyes were dissected and fixed overnight at 4uC in 1.5% glutaraldehyde, 1% paraformaldehyde and 0.1M PIPES buffer (pH 7.4). After washing, eyes were post-fixed at room temperature in 1% OsO4, 0.1M PIPES (pH 7.4). They were then deshydrated with successive ethanol solutions followed by anhydrous propylen oxyde. Eyes were infiltrated with increasing concentrations of epoxy resin (EMbed 812 from EMS) in propylen oxyde for 1 day at room temperature and samples were mounted in pure resin into silicone embedding molds. Polymerization was performed at 60uC for 2 days. Ultrathin sections of 60 nm were stained with lead citrate and examined with a transmission electron microscope (Philips CM120) operating at 80 kV.

Generation of anti-Fatp C11-7 antibody
Two rabbits were immunized with two peptides of Fatp, Peptide 1: YQTSKGRYELLTPQ at the C-terminus of the protein and Peptide 2: NNNSETEKNIPQAK in the middle of the protein.
The serum from the two rabbits were pooled and affinity purified.

Electroretinogram
For ERG recordings, white-eyed flies were analyzed. Coldanesthetized flies were immobilized in clay. A tungsten electrode (0.5-1 MV, Intracell) was inserted in the back of the head and a glass electrode filled with 3 M KCl (2-6 MV) was poked through the cornea. Flies were dark-adapted for 2 min before recording.
An orange LED (591 nm, 2800 mcd, 40u beam, LY 5436-VBW-1, Osram, France) was placed at 1 cm from the head. The flash intensity reaching the eye was 650 mW/cm 2 , as measured with a PM100D power meter and S121C photodiode (Thorlabs, Maisons-Laffitte, France). Signals were filtered at 2 kHz and digitized at 10 kHz, using a MultiClamp 700A amplifier, a Digidata 1322A interface and pClamp-8 software (Molecular Devices, Sunnyvale, USA). Flash intensity and duration were controlled through pClamp and the Digidata analog output.

Histological detection of b-galactosidase activity
Horizontal adult eye sections were performed using a cryostat microtome (Microm) and deposited on superfrost Plus slides (Thermo). Third instar laval imaginal discs, midgut and salivary gland were dissected in 1X PBS. Samples were fixed 5 min in PBS 0.25% gluteraldehyde. They were stained in a solution of 7.
For blue light-induced degradation of Rh1, white-eyed heads were exposed to 10 mW blue light for 6 h and homogenized in Tris-buffered saline (20 mM Tris (pH 7.5), 150 mM NaCl) containing 0.5% Triton X-100 and protease inhibitors. After centrifugation, the supernatant was mixed with an equal volume of 2X Laemmli buffer and loaded onto a 12% acrylamide gel as described above. Figure S1 Recessive lethal fatp k10307 mutation is protein null. (A) The fatp k10307 mutation consists of the insertion of a P{lacW} element (10,7 kb) in the 59 end of the open reading frames. We assumed that it disrupts fatp expression. In addition, the P{lacW} element carries a lacZ sequence which captures fatp expression profile. We first checked that fatp k10307 -associated lethality is localized in the fatp locus. We determined that fatp k10307 was transheterozygous lethal over Df(2L)Exel7048 and Df(2L)BSC210, two deficiencies (101.5 and 199.6 kb respectively) covering the fatp locus and was not lethal over Df(2L)BSC342, a nearby deficiency that does not cover fatp locus. (B-E) Photographes of first and second wild-type Canton-S (Cs) and fatp k10307/k10307 mutant larvae (scale bar = 0.5 mm). First instar larvae (L1, B,C) were observed 42-47 h after egg laying and second instar larvae (L2, D, E) 67-71 h after egg laying. Whereas first instar homozygous mutant larvae are similar to willd type larvae in terms of size and mobility, second instar mutant larvae are dying as weakly mobile and small larvae. Their mouth hook is similar to second instar larvae (F, scale bar = 15 mm). This result indicates that fatp k10307 mutation is lethal in L2 instar. (G) Semi-quantitative duplex RT-PCR of fatp and rp49 mRNA in wild-type and fatp k10307/k10307 mutant retina and first instar larvae extract. In fatp k10307/k10307 , no amplification of fatp mRNA is detected suggesting that fatp k10307 is a null mutation. (C) Western blot analysis of Rh1 in unboiled head extracts from control and fatp k10307/k10307 flies exposed to blue light. Heads were either kept in the dark (light 2) or exposed to blue light for 6 hours (light +). Tubulin was used as a loading control. (D) Quantification of protein levels. Whereas Rh1 levels are twofold decreased after light illumination in control heads, Rh1 levels are not significantly decreased in fatp mutant heads. (TIF)