ADrosophilametallophosphoesterase Mediates Deglycosylation of Rhodopsin
Jinguo Cao,Yi Li,Wenjing Xia,Keith Reddig,Wen Hu,Wei Xie,Hongsheng Li,Junhai Han
DOI: https://doi.org/10.1038/emboj.2011.254
2011-01-01
Abstract:Article29 July 2011free access A Drosophila metallophosphoesterase mediates deglycosylation of rhodopsin Jinguo Cao Jinguo Cao Institute of Life Science, The Key Laboratory of Developmental Genes and Human Disease, Southeast University, Nanjing, China Search for more papers by this author Yi Li Yi Li Institute of Life Science, The Key Laboratory of Developmental Genes and Human Disease, Southeast University, Nanjing, China Search for more papers by this author Wenjing Xia Wenjing Xia Institute of Life Science, The Key Laboratory of Developmental Genes and Human Disease, Southeast University, Nanjing, China Search for more papers by this author Keith Reddig Keith Reddig Department of Neurobiology, University of Massachusetts Medical School, Worcester, MA, USA Search for more papers by this author Wen Hu Wen Hu Institute of Life Science, The Key Laboratory of Developmental Genes and Human Disease, Southeast University, Nanjing, China Search for more papers by this author Wei Xie Wei Xie Institute of Life Science, The Key Laboratory of Developmental Genes and Human Disease, Southeast University, Nanjing, China Search for more papers by this author Hong-Sheng Li Corresponding Author Hong-Sheng Li Department of Neurobiology, University of Massachusetts Medical School, Worcester, MA, USA Search for more papers by this author Junhai Han Corresponding Author Junhai Han Institute of Life Science, The Key Laboratory of Developmental Genes and Human Disease, Southeast University, Nanjing, China Search for more papers by this author Jinguo Cao Jinguo Cao Institute of Life Science, The Key Laboratory of Developmental Genes and Human Disease, Southeast University, Nanjing, China Search for more papers by this author Yi Li Yi Li Institute of Life Science, The Key Laboratory of Developmental Genes and Human Disease, Southeast University, Nanjing, China Search for more papers by this author Wenjing Xia Wenjing Xia Institute of Life Science, The Key Laboratory of Developmental Genes and Human Disease, Southeast University, Nanjing, China Search for more papers by this author Keith Reddig Keith Reddig Department of Neurobiology, University of Massachusetts Medical School, Worcester, MA, USA Search for more papers by this author Wen Hu Wen Hu Institute of Life Science, The Key Laboratory of Developmental Genes and Human Disease, Southeast University, Nanjing, China Search for more papers by this author Wei Xie Wei Xie Institute of Life Science, The Key Laboratory of Developmental Genes and Human Disease, Southeast University, Nanjing, China Search for more papers by this author Hong-Sheng Li Corresponding Author Hong-Sheng Li Department of Neurobiology, University of Massachusetts Medical School, Worcester, MA, USA Search for more papers by this author Junhai Han Corresponding Author Junhai Han Institute of Life Science, The Key Laboratory of Developmental Genes and Human Disease, Southeast University, Nanjing, China Search for more papers by this author Author Information Jinguo Cao1,‡, Yi Li1,‡, Wenjing Xia1, Keith Reddig2, Wen Hu1, Wei Xie1, Hong-Sheng Li 2 and Junhai Han 1 1Institute of Life Science, The Key Laboratory of Developmental Genes and Human Disease, Southeast University, Nanjing, China 2Department of Neurobiology, University of Massachusetts Medical School, Worcester, MA, USA ‡These authors contributed equally to this work *Corresponding authors: Department of Neurobiology, University of Massachusetts Medical School, 364 Plantation Street, Worcester, MA 01605, USA. Tel.: +1 508 856 6702; Fax: +1 508 856 6070; E-mail: [email protected] of Life Science, The Key Laboratory of Developmental Genes and Human Disease, Southeast University, 2 Sipailou Road, Nanjing, Jiangsu 210096, China. Tel.: +86 25 8379 0962; Fax: +86 25 8379 0962; E-mail: [email protected] The EMBO Journal (2011)30:3701-3713https://doi.org/10.1038/emboj.2011.254 PDFDownload PDF of article text and main figures. Peer ReviewDownload a summary of the editorial decision process including editorial decision letters, reviewer comments and author responses to feedback. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Oligosaccharide chains of newly synthesized membrane receptors are trimmed and modified to optimize their trafficking and/or signalling before delivery to the cell surface. For most membrane receptors, the functional significance of oligosaccharide chain modification is unknown. During the maturation of Rh1 rhodopsin, a Drosophila light receptor, the oligosaccharide chain is trimmed extensively. Neither the functional significance of this modification nor the enzymes mediating this process are known. Here, we identify a dmppe (Drosophila metallophosphoesterase) mutant with incomplete deglycosylation of Rh1, and show that the retained oligosaccharide chain does not affect Rh1 localization or signalling. The incomplete deglycosylation, however, renders Rh1 more sensitive to endocytic degradation, and causes morphological and functional defects in photoreceptors of aged dmppe flies. We further demonstrate that the dMPPE protein functions as an Mn2+/Zn2+-dependent phosphoesterase and mediates in vivo dephosphorylation of α-Man-II. Most importantly, the dephosphorylated α-Man-II is required for the removal of the Rh1 oligosaccharide chain. These observations suggest that the glycosylation status of membrane proteins is controlled through phosphorylation/dephosphorylation, and that MPPE acts as the phosphoesterase in this regulation. Introduction N-linked glycosylation within the endoplasmic reticulum (ER) (Kornfeld and Kornfeld, 1985; Helenius, 1994) is important for the folding, assembly, oligomerization, quality control, and subcellular localization of membrane receptors and secretory proteins (Helenius, 1994; Fiedler and Simons, 1995; Helenius and Aebi, 2001). Elimination of glycoprotein N-glycosylation sites frequently results in significantly reduced levels of protein secretion (Fiedler and Simons, 1995). During transportation from ER to the cell surface, the oligosaccharide chains of glycoproteins are trimmed and in many cases are added with new sugar residues. The glycosylation status of membrane proteins is a key determinant for the specificity and efficacy of cell–protein and cell–cell interactions (Varki, 1993; Dwek, 1996). Appropriate N-glycosylation also increases the stability of glycoprotein by increasing plasma residence times (Hoffmeister et al, 2003) and providing steric protection from proteases and non-specific interactions (Rudd et al, 2001). In photoreceptor neurons, N-glycosylation of the light receptor rhodopsin is critical for its exit from the ER and for its targeting to light sensory organelles (Hargrave, 1977). Pharmacologically blocking rhodopsin glycosylation has prevented newly synthesized rhodopsin from maturation and inhibited the membrane disc morphogenesis (Fliesler et al, 1984, 1986). Importantly, mutations at the sites of N-glycosylation in rhodopsin have been identified in patients of autosomal dominant retinitis pigmentosa (Dryja and Li, 1995; Papermaster, 1995), a retinal degeneration disorder that causes visual defects including night blindness and tunnel vision (Dryja and Li, 1995). Rh1, the major rhodopsin in Drosophila, undergoes N-glycosylation during biosynthesis (O'Tousa, 1992; Brown et al, 1994). Mutation of residue asn20, the single site of N-glycosylation in Rh1, retains Rh1 in the secretary pathway and drastically reduces the level of mature Rh1 (Webel et al, 2000). Interestingly, the oligosaccharide chain is either extensively trimmed or completely removed in mature Rh1 (Huber et al, 1990; Katanosaka et al, 1998), indicating that Rh1 undergoes deglycosylation during its transport from ER to the light sensory organelle rhabdomere (Satoh et al, 1997). However, the enzymes mediating this process remain unknown. More importantly, the physiological significance of the Rh1 deglycosylation is unclear. Metallophosphoesterases (MPPEs) are a superfamily of enzymes that contain two well-characterized groups, phosphomonoesterases and phosphodiesterases. The activity of MPPEs depends on the presence of metal ions such as Mn2+, Ni2+, Mg2+, Zn2+, Fe2+ and Co2+ (Cohen, 1989). A wide variety of proteins and nucleotides are MPPE substrates, including protein phosphoserine phosphatases, sphingomyelin phosphodiesterases, nucleotidases, 2′,3′-cAMP phosphodiesterases and nucleases (Aravind and Koonin, 1998; Keppetipola and Shuman, 2008; Fujita et al, 2009). A recent study suggested that PGAP5, a human MPPE, mediated the remodelling of the glycan moiety on glycosylphosphatidylinositol-anchored proteins by removal of a side-chain ethanolamine-phosphate from the second mannose of glycosylphosphatidylinositol (Fujita et al, 2009). Although several MPPEs have been characterized based on their biochemical activity and structure, their native substrates are largely unknown. More importantly, in vivo functions of these enzymes have yet to be well characterized (Chen et al, 2004). CG8889, the Drosophila orthologue of human metallophosphoesterase 1, is highly expressed in the eye (Xu et al, 2004). In this work, we demonstrate that CG8889, hereafter referred to as dMPPE, functions as a phosphoesterase, and is required for Rh1 deglycosylation in photoreceptor neurons. Our study presents the first evidence that the glycosylation status of membrane receptors can be regulated through phosphorylation/dephosphorylation. Results Rh1 has a higher molecular weight in a dmppe mutant We obtained a dmppe mutant, CG8889e02905, from the Harvard Exelixis Collection. In newly eclosed, dark-reared mutants, we detected a small reduction in light sensitivity of the fly eye based on electroretinogram (ERG) recordings (Figure 1A and B). As light sensitivity is largely determined by the density of rhodopsin in the rhabdomeral membrane (Johnson and Pak, 1986), we stained cross-sections of eye with an Rh1 antibody to examine the Rh1 level in rhabdomeres. In this experiment, flies were reared in the dark from early pupal stage to prevent light-induced endocytosis. Four hours after eclosion, the majority of Rh1 protein was localized in the rhabdomere in wild-type flies. In contrast, in the dmppe mutant, many Rh1-positive vesicles (RPVs) were detected in the cell body of photoreceptors (Figure 1C). It is previously reported that Rh1 transports to rhabdomere through multivesicular bodies (MVBs; Satoh et al, 2005). To investigate whether the RPVs in the mutant were MVBs, we stained eye sections with the Rh1 and an immunogold antibody, and examined the result via electron microscopy (EM). The EM images revealed that RPVs appeared to be collections of small vesicles (Figure 1D), which are reminiscent of the reported MVBs. Thus, Rh1 could be retained in the secretion pathway in the mutant. Figure 1.Newly eclosed dmppe mutants show reduced light sensitivity, abnormal distribution and high molecular weight of Rh1. (A) ERG recordings revealed the reduction of light sensitivity in newly eclosed dmppe mutants. Flies were raised in the dark and examined within 4 h after eclosion. Fly eyes were stimulated with a series of 1 s light pulses of increasing intensities as labelled on the top. The first response appearing is marked with an asterisk. The scale bar next to the top trace is 5 mV. (B) Quantification of light sensitivities in newly eclosed flies. The mean relative sensitivities shown were calculated as described in Materials and methods. The error bar represents standard error of the mean (s.e.m.). *Indicates that the sample is significantly different from others in the group. (C) Rh1 distribution in the photoreceptors of newly eclosed flies. Cross-sections were prepared as described in Materials and methods. The sections were stained with a monoclonal Rh1 antibody (4C5). The number of RPVs per ommatidium was calculated for each genotype. Scale bar, 5 μm. (D) Immunogold electron microscopy reveals that RPV appeared to be the aggregation of small vesicles. Sections were prepared as described in Materials and methods. One RPV (boxed) is enlarged in the upper panel. Scale bar, 2 μm. (E) Reduction of Rh1 level in newly eclosed dmppe mutants. Flies were raised in the dark and heads were collected at the indicated time after eclosion. The scaffold protein INAD was probed in parallel. (F) Western blots show the increase in Rh1 MW in the mutant. The MWs of other visual molecules are normal. WT: wild type. Download figure Download PowerPoint We further compared the Rh1 protein level between wild type and the mutant using western blots. The results showed a slight reduction of Rh1 level in the newly eclosed mutant and normal Rh1 protein levels in 1-day-old mutants (Figure 1E). Intriguingly, in the mutant, we observed an increase in the molecular weight (MW) of Rh1 (Figure 1E and F), while all other visual molecules examined were similar to wild type (Figure 1F). The mRNA sequence of ninaE, the gene encoding the Rh1 protein, was not changed in the mutant (data not shown); therefore, the increase in Rh1 MW is due to abnormal post-translational modification. The increase in Rh1 MW is due to a failure of deglycosylation As dMPPE is predicted to be a phosphoesterase, the increase in Rh1 MW could be due to hyperphosphorylation. However, the following observations argue against this explanation. First, treatments with non-specific phosphatases, calf intestinal phosphatase or potato acid phosphatase, did not change the MW of Rh1 purified from the mutant (Figure 2A). Second, a truncated Rh1 protein encoded by the mutant allele ninaEΔ365, which lacks the phosphorylation sites (Vinos et al, 1997), still had an increased MW in the dmppe mutant background (Figure 2B). Figure 2.The increase in Rh1 MW is due to a failure of deglycosylation. (A) The increase in Rh1 MW is not caused by phosphorylation modification. After digestion with calf intestinal phosphatase (CIP) or potato acid phosphatase (PAP) at 37°C for 16 h, purified Rh1 was subjected to SDS–PAGE and then immunoblotting. (B) Deletion of phosphorylation sites in Rh1 did not prevent the increase in Rh1 MW in the mutant background. Each lane was loaded with one fly head and probed with a polyclonal antibody against the N terminus of Rh1. Note: in the ninaEΔ356 allele, only truncated Rh1 can be detected, as this mutant allele was crossed into a ninaE null mutant background. (C) Rh1 bands shifted after the treatment with PNGase F. After digestion with PNGase F in 37°C for 16 h, purified Rh1 was subjected to SDS–PAGE and immunoblotting. RS: rescuing flies, dmppe; p[trp∷dmppe]. +: Incubation with PNGase F. −: Incubation without PNGase F. (D) The MW of Rh1 was reduced after the treatment with Endoglycosidase H. Purified Rh1 was incubated with Endoglycosidase H at 37°C for 16 h and examined by western blots. Download figure Download PowerPoint As the MW of Rh1 from the mutant is close to that of immature and glycosylated Rh1, we purified Rh1 proteins from both wild type and mutant using Rh1 antibody-linked affinity beads and treated them with different glycosidases. After treatment with peptide: N-glycosidase F (PNGase F), the MW of Rh1 from the mutant was reduced (Figure 2C), indicating that Rh1 is glycosylated in the mutant. As a control, the mature Rh1 protein from wild type showed undetectable change in the MW (Figure 2C). To characterize the oligosaccharide chain present on glycosylated Rh1, we digested Rh1 from the mutant with endoglycosidase H (Endo H). Endo H only cleaves unmodified core oligosaccharides but not complex N-linked glycoproteins (Twining, 1984). The result showed that Rh1 from the mutant was deglycosylated by Endo H (Figure 2D), suggesting that Rh1 in the mutant contains a core oligosaccharide. Rh1 transport is slow in the dmppe mutant To understand why some Rh1 protein mislocalized in the newly eclosed mutant, we examined the dynamics of Rh1 glycosylation and distribution in both wild type and the mutant. Western blots of wild type showed that Rh1 was undetectable at 63% of pupal development (% pd), while immature 40 kDa Rh1 was detected at 68% pd, and mature 36 kDa Rh1 was detected at 71% pd (Figure 3A). Thus, the extensively trimmed Rh1 begins to form between 68% and 71% pd. In immunostaining experiments, most Rh1 protein resides in the cell body but not in the rhabdomere at 68% pd, ∼25% of Rh1 resides in the rhabdomere at 71% pd and ∼86% of Rh1 resides in the rhabdomere at 79% pd (Figure 3C and E). Taken together, these observations suggest that the immature high-MW Rh1 mostly exists in the cell body and the mature low-MW form localizes in the rhabdomere. In other words, Rh1 undergoes oligosaccharide trimming during its trafficking to the rhabdomere. Figure 3.Deglycosylation of Rh1 is essential for transport, but not required for membrane localization. (A) Time course of Rh1 processing in wild-type flies. Pupal heads were collected at the indicated time. The amount loaded is labelled on the top. (B) The Rh1 deglycosylation process is disrupted in dmppe mutant. Pupae were collected at different pupal development time points. (C) Rh1 distribution in the developing photoreceptors of wild-type flies. Sections were prepared as described in Materials and methods. Cross-sections were stained with anti-Rh1 antibody (green) and rhodamine-phalloidin (red). % pd: % of pupal development. Scale bar, 5 μm. (D) Rh1 distribution in the developing photoreceptors of the dmppe mutant. Scale bar, 5 μm. (E) Quantification of the percentage of Rh1 in rhabdomere during pupal development. Quantification was performed as described in Materials and methods. (F) Most Rh1 localized normally in matured mutant photoreceptors. Flies were raised in the dark and examined at 2 days after eclosion. Collection and fixation were performed under dim red light. Scale bar, 5 μm. (G) Constant Rh1 levels in adult mutants. Flies were raised in the dark and examined at the indicated time after eclosion. The scaffold protein INAD was probed in parallel as a loading control. Download figure Download PowerPoint In contrast, in the dmppe mutant, at ∼68% pd, the Rh1 MW was 40 kDa. It changed gradually to 39 kDa during development and remained the same size through eclosion (Figure 3B). These observations suggest that the process of Rh1 deglycosylation is disrupted in the mutant. Unlike the wild type, the mutant showed almost no Rh1 in the rhabdomere before 71% pd, and only ∼49% of Rh1 was localized in the rhabdomere at 79% pd (Figure 3D and E). These data indicate that oligosaccharide trimming accelerates the transport of Rh1 to the rhabdomere. Although we detected some RPVs in the cytoplasm, most of the glycosylated Rh1 was still loaded into rhabdomere in the newly eclosed mutant. To investigate the fate of RPVs, we examined the distribution and protein level of Rh1 in more matured adults. In 2-day-old, dark-reared mutants, Rh1 was completely localized in the rhabdomere (Figure 3F), and the overall Rh1 level was comparable to wild type (Figure 3G), suggesting that Rh1 in the RPVs ultimately transfers to the rhabdomere. Deglycosylation is not required for the normal signalling of Rh1 Glycosylated Rh1 localized normally in the rhabdomere in matured mutants and the rhabdomeral density of Rh1 was comparable to wild type; therefore, we next examined whether the retained oligosaccharide chain interfered with the signalling of Rh1. Compared with the wild-type flies, 2-day-old, dark-reared mutants displayed normal ERG recording traces (Supplementary Figure S1A), ERG amplitude (Supplementary Figure S1B), ON and OFF transients (Supplementary Figure S1C) and sensitivity to light (Supplementary Figure S1D). These data indicate that the deglycosylation of Rh1 is not required for the normal activation of visual signalling. In addition, the deglycosylation of Rh1 was also not required for the normal deactivation of visual signalling. In 2-day-old, dark-reared mutants, the termination speed of light response and the minimum light intensity required to produce prolonged depolarization after potential were comparable to that of wild type (Supplementary Figure S1E and F). Deglycosylation is essential for the stability of endocytosed Rh1 Although the deglycosylation of Rh1 is not required for normal visual signalling, old mutants reared in regular light conditions displayed obvious reduced sensitivity to light. ERG recording revealed that the overall light sensitivity of photoreceptor cells strikingly decreased in 20-day-old mutants raised under regular light cycles (12 h light/12 h dark; Figure 4A). Western blotting showed that most Rh1 and much of INAD had been degraded in mutant, but not in wild-type flies (Figure 4B), suggesting that old mutants undergo retinal degeneration. Subsequent EM examinations revealed many small splits and tiny holes in the rhabdomere of 18-day-old mutants raised in the same condition (Figure 4C). However, 16-day-old, dark-reared mutant displayed normal rhabodomeral structure, Rh1 level, and sensitivity to light (Supplementary Figure S2). Thus, the visual functional defects and retinal degeneration could be due to loss of Rh1 protein. Indeed, retinal degeneration also occurs in ninaE mutant flies that have extremely low levels of Rh1 (Leonard et al, 1992). Figure 4.Glycosylated Rh1 is not stable once it is endocytosed, and leads to retinal degeneration. (A) Quantification of light sensitivities in wild-type and dmppe mutant flies. Flies were raised in 12 h light/12 h dark conditions. Newly eclosed adults were collected and reared for the indicated time. For 1-day-old flies, collected adults were reared for 24 h. Relative light sensitivity was measured and calculated as described in Materials and methods. (B) Rh1 and INAD levels decreased in the older mutant. Tubulin was probed as a loading control. The reduction of INAD level indicates the older mutant undergoes retinal degeneration. (C) EM analyses revealed the older mutant underwent retinal degeneration. Flies were raised in 12 h light/12 h dark conditions for the indicated time. Each picture shows a single ommatidium. One degenerated rhabdomere from the mutant is enlarged in the right panel. Scale bar, 2 μm. (D) Glycosylation does not affect light-induced endocytosis of Rh1. Two-day-old dark-reared flies were stimulated with white light (700 lux) for 8 h and stained with the monoclonal Rh1 antibody. Scale bar, 5 μm. (E) Rh1 distribution after blue light stimulation. Two-day-old, dark-reared flies were stimulated with pure blue light (700 lux) for 2 or 3 h. Cross-sections were prepared and stained as described in Materials and methods. Three large endocytic Rh1 particles in the cell bodies are marked with arrows. Scale bar, 5 μm. The right panel shows the ratio of Rh1 signal intensity in the cell body for each genotype and treatment. Quantification was performed as described in Materials and methods. (F) Rh1 level decreased in the mutant after 3 h under blue light stimulation. Two-day-old, dark-reared flies were stimulated with pure blue light (700 lux) for the indicated time. After stimulation, fly heads were collected and total Rh1 levels were compared by western blots. INAD was probed as a loading control. *indicates that the sample is significantly different from the control (P<0.01; Student's t-test). Download figure Download PowerPoint Given that the glycosylation status affects the endocytosis of several membrane receptors (Partridge et al, 2004), the reduction of Rh1 in old mutants could be due to excessive endocytosis and subsequent degradation. To test this hypothesis, we first investigated the light-induced Rh1 endocytosis in the mutant. After exposure to regular light (700 lux) for 8 h, the number of endocytic Rh1 particles per ommatidium in the mutant was comparable to that in wild type (2.3±0.3 versus 2.6±0.4; Figure 4D). This observation suggests that the oligosaccharide chain does not significantly interfere with Rh1 endocytosis. Since glycosylation of Rh1 does not affect its endocytosis, we next examined whether glycosylation of Rh1 changed the stability of endocytosed Rh1. For this purpose, we prevented the endocytosed Rh1 from being recycled back to the rhabdomere by stimulating the fly with pure blue light, which converts the inactive Rh1 rhodopsin to the active metarhodopsin, but does not change it back to rhodopsin for recycling (Xu et al, 2004; Han et al, 2007). After 2 h of blue light stimulation (700 lux), equal ratios of endocytosed Rh1 signal were detected in wild type and the mutant (4.8±0.5 versus 5.1±0.6%; Figure 4E). At this time point, the degradation of endocytosed Rh1 had not begun in either fly (Figure 4F). After 3 h of stimulation, however, significant Rh1 degradation was observed in the dmppe mutant (Figure 4F). In addition, the intensity of visible endocytic Rh1 signal increased further only in the wild type (arrows in Figure 4E) but not in the mutant (Figure 4E). Thus, the insufficiently deglycosylated Rh1, after endocytosis, is not as stable as the extensively processed Rh1. dMPPE mediates the process of Rh1 deglycosylation In the dmppe mutant, a p element is inserted into the 5′ untranslated region of the dmppe gene. Homozygotes of this allele are viable. To determine if this insertion actually disrupts the dmppe gene, we examined dmppe mRNA levels with northern blot analysis and real-time RT–PCR. Using a probe that specifically recognizes a long transcriptional form of dmppe mRNA, we confirmed the loss of dmppe mRNA in the mutant (Figure 5A). Using real-time RT–PCR, we showed that the amount of dmppe mRNA decreased ∼50-fold in the mutant (Figure 5B). Furthermore, western blots also demonstrated the absence of dMPPE protein in the mutant (Figure 5C). Thus, the dmppe gene is disrupted by this p element insertion. Figure 5.dMPPE is required for Rh1 deglycosylation. (A) Northern blot shows the mRNA of the dmppe gene is absent in dmppe mutants. Total RNA was isolated from the heads of <3-day-old cn,bw and cn,pBacCG8889e02905,bw flies using Trizol. Total RNA (10 μg) was subjected to 1% denaturing agarose gels and hydrated with a DIG-labelled dmppe cDNA probe. The membrane was stripped and re-probed with ninaE probe for the loading control. (B) Real-time RT–PCR reveals that the dmppe mRNA level decreases ∼50-fold in dmppe mutants. Three data sets were averaged. (C) Western blots showing loss of dMPPE protein in the mutant. Each lane was loaded with three fly heads. (D) Overexpression of dMPPE restored the MW of Rh1. cn, mppe,bw;p[hs∷mppe] flies were heat shocked every day for 1 h from early pupal stage. Newly eclosed adults were collected for western blots. dMPPE protein levels are showed on the lower panel. (E) Photoreceptor-specific expression of dMPPE restored the MW of Rh1. dMPPE was specifically expressed in the photoreceptor through a trp gene promoter. dMPPE protein levels are shown on the lower panel. (F) Rh1 distribution in newly eclosed rescue (RS) flies. Quantifications of RPV number per ommatidium are presented in the right panel. The RS represents cn, mppe,bw; p[trp∷mppe]/TM6c flies. Scale bar, 5 μm. (G) Light sensitivity was restored in the RS fly. The relative sensitivities were calculated as described in Materials and methods. *indicates that the sample is significantly different from the control (P<0.01; Student's t-test). Download figure Download PowerPoint To confirm that loss of dMPPE is responsible for the defect in Rh1 deglycosylation, we attempted to rescue the phenotype by generating transgenic flies that express a dMPPE cDNA. In a cn,dmppe,bw; p[hs-dmppe] fly, expression of dMPPE from the pupal stage converted Rh1 to the extensively processed form (Figure 5D). To show that dMPPE functions in the photoreceptor, we successfully rescued the defect of Rh1 deglycosylation in another transgenic fly (cn, dmppe,bw; p[trp∷dmppe]) that expresses dMPPE through a photoreceptor-specific trp gene promoter (Han et al, 2006; Figure 5E). In this latter fly, we also observed the rescue of two other phenotypes of the dmppe mutant: the incomplete Rh1 delivery (Figure 5F) and the reduction of light sensitivity (Figure 5G) in newly eclosed flies. Taken together, the above observations indicate that dMPPE is required for the deglycosylation of Rh1. dMPPE predominately localizes in Golgi and colocalizes with trafficking Rh1 To investigate why dMPPE is required for Rh1 deglycosylation, we first examined the expression pattern of dMPPE in the adult fly head, and determined that dMPPE was highly expressed in the retina (Figure 6A). We next stained the dissected ommatidium with both anti-Rh1 and anti-dMPPE antibodies to display the subcellular localization of dMPPE. The images showed that dMPPE resides in the cell body of photoreceptors, with some surrounding the nucleus (Figure 6B). To further display the organellar distribution of endogenous dMPPE, we stained dissected ommatidium, whose ER or Golgi had been labelled with GFP. The results showed that dMPPE predominately exists in the Golgi (Figure 6C and D). Figure 6.dMPPE localizes in Golgi and colocalizes with trafficking Rh1. (A) dMPPE is predominately expressed in the retina. Dissected fly heads were costained with dMPPE antibody (green), a 24b10 antibody (red, showing retina) and DAPI (blue, for nuclei). R=retina, L=lamina, M=medulla, B=brain. Scale bar, 200 μm. (B) dMPPE distribution in photoreceptors. Dissected ommatidia were costained with dMPPE antibody (green), an Rh1 antibody (red, showing rhabdomere bundles) and DAPI (blue, for nuclei). Scale bar, 10 μm. (C) In GMR-Gal4/p[UAS∷GFP-K