SEARCH

SEARCH BY CITATION

Keywords:

  • Drosophila development;
  • metamorphosis;
  • midgut;
  • PTP52F;
  • receptor protein tyrosine phosphatase

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. References
  8. Supporting Information

To date our understanding of Drosophila receptor protein tyrosine phosphatases (R-PTPs) in the regulation of signal transduction is limited. Of the seven R-PTPs identified in flies, six are involved in the axon guidance that occurs during embryogenesis. However, whether and how R-PTPs may control key steps of Drosophila development is not clear. In this study we investigated the potential role of Drosophila R-PTPs in developmental processes outside the neuronal system and beyond the embryogenesis stage. Through systematic data mining of available microarray databases, we found the mRNA level of PTP52F to be highly enriched in the midgut of flies at the larva–pupa transition. This finding was confirmed by gut tissue staining with a specific antibody. The unique spatiotemporal expression of PTP52F suggests that it is possibly involved in regulating metamorphosis during the transformation from larva to pupa. To test this hypothesis, we employed RNA interference to examine the defects of transgenic flies. We found that ablation of endogenous PTP52F led to high lethality characterized by the pharate adult phenotype, occurring due to post pupal eclosion failure. These results show that PTP52F plays an indispensable role during the larva–pupa transition. We also found that PTP52F could be reclassified as a member of the subtype R3 PTPs instead of as an unclassified R-PTP without a human ortholog, as suggested previously. Together, these findings suggest that Drosophila R-PTPs may control metamorphosis and other biological processes beyond our current knowledge.


Abbreviations
ConA

concanavalin A

CS

C1290S mutant

EcR

ecdysone receptor

HA

hemagglutinin

PTK

protein tyrosine kinase

PTP

protein tyrosine phosphatase

PTP-d

PTP-δelta

PTP-σ

PTP-sigma

RNAi

RNA interference

R-PTP

receptor protein tyrosine phosphatase

WT

wild-type

Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. References
  8. Supporting Information

Reversible tyrosine phosphorylation is a very important post-translational modification. A diverse array of biological processes controlled by protein tyrosine phosphorylation and dephosphorylation in various metazoans are mediated by the coord inated action between protein tyrosine kinases (PTKs) and protein tyrosine phosphatases (PTPs) [1,2]. The characterization of PTKs, PTPs and their numerous substrates is of critical importance to our understanding of how the homeostasis of signal transduction is achieved. Over the last two decades, because of advances in biochemical, molecular and genetic approaches, the functions of many of the key players involved in this context, in particular PTKs, have been clarified [3,4]. In contrast, our understanding of PTPs in the regulation of protein tyrosine phosphorylation dependent signal transduction has lagged behind. Issues including substrate specificity as well as spatial and temporal control of cell signaling determined by PTPs remain elusive.

Since its development more than 15 years ago, the substrate trapping technique [5] has been used to identify numerous bona fide and potential substrates of PTPs, significantly accelerating the biochemical characterization of these enzymes. Recently, studies have used specific PTP knockout mice to delineate their roles in developmental control and disease conditions. A few studies have clearly defined the critical role of some non-receptor PTPs, such as PTP1B in regulating insulin responsiveness [6,7], leptin signaling [8–10] and breast tumorigenesis [11], T-cell PTP in regulating T-cell signaling [12], as well as SHP-2, which participates in Helicobacter pylori induced gastric ulcer [13], Noonan syndrome [14] and LEOPARD syndrome [15,16]. Genetic studies of other PTPs in mice, however, have encountered some difficulties, mostly due to the redundancy of multiple regulatory components in a given signaling pathway. This problem was particularly present in the investigation of receptor PTPs (R-PTPs). One such problem has been encountered in the genetic characterization for the functional role of R-PTPs belonging to the subtype R2A phosphatases [17]. Since the first knockout mouse model was generated, it has been suggested that PTP-sigma (PTP-σ) might participate in determining axon regeneration and extension [18]. However, detailed investigations found that the deletion of PTP-σper se failed to produce an obvious phenotype during neuronal development [19], largely due to the functional compensation contributed by other R-PTPs in the same subtype. To date, it is clear that PTP-σ and PTP-delta (PTP-δ), both subtype R2A PTPs, compensate for each other in the control of neuronal development. Quantitative analyses have demonstrated that PTP-σ+/−/PTP-δ−/− and PTP-σ−/−/PTP-δ+/− mutants exhibit intermediate phenotypes in motoneuron survival and phrenic nerve branching, whereas no significant defect has been detected in either PTP-σ or PTP-δ single mutants [17]. These data suggest that more studies involving multiple knockouts in mice are needed to clarify the functional role of individual PTPs. Considering the complexity of functional compensation among mammalian PTPs, we proposed that other simpler model organisms might provide alternatives for genetically characterizing the PTPome.

Drosophila is an excellent model organism for investigating diverse tyrosine phosphorylation dependent signaling pathways, which are regulated by many modulators exhibiting similar functions across species throughout evolution [20–22]. Its relatively simple genome allows the precise dissection of signaling networks without the genetic redundancy related complications often seen in mammalian genetic screenings [23–25]. Moreover, while there are a large number of classical PTPs in mammals (38 in humans), there are only 15 putative ptp genes in the Drosophila genome (eight non-receptor PTPs and seven R-PTPs). Several of the Drosophila PTPs have been classified as orthologs of human PTPs based on the similar sequence alignment between the two, suggesting evolutionarily conserved functions [26]. In fact, a number of early breakthroughs in functional characterizations of R-PTPs have been made in genetic studies of Drosophila, and all R-PTPs have been found to be involved in axon guidance during embryogenesis [27–30].

To date, the findings used to define the functional role of R-PTPs in axon guidance have been limited to embryogenesis, however. Other than this stage, it is not clear whether and how R-PTPs participate in the regulation of Drosophila development. Therefore, in this study we investigated potential R-PTP involvement in development beyond the embryonic stage. Systematic data mining of available microarray databases found PTP52F to be highly enriched in the midgut tissue of prepupal flies. Our genetic studies demonstrated that PTP52F plays a critical role in the control of Drosophila development during the larva–pupa transition.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. References
  8. Supporting Information

Profiling of PTPs during Drosophila development by in-gel phosphatase activity assay

While individual PTPs have been implicated in the regulation of Drosophila development during embryogenesis, the expression and activity profile of PTPs at other developmental stages remain uncharacterized. Such data might provide more insight into the role of PTPs in the control of the developmental process. We used an in-gel phosphatase activity assay to visualize the possible participation of PTPs at each developmental stage. This assay displays a diverse array of active PTPs in total extracts of cells and tissues according to the molecular weights of these phosphatases resolved by SDS/PAGE. As shown in Fig. 1, the overall PTP activity in the embryonic stage was significantly higher than in other developmental stages, suggesting that rapid protein tyrosine dephosphorylation plays a critical role in signal transduction during this stage. We found the activity of many PTPs to be diminished during the larva–pupa transition, and increased slightly in adult flies (Fig. 1). Although some PTPs were visible in adult flies, their activity was far weaker than the activity of those in embryos (Fig. 1). The biological implication of such interesting observations requires further investigation. Nonetheless, it should be addressed that all phosphatases detected by this assay format are likely to be non-R-PTPs due to the inherent limitation of the technique [31,32]. Indeed none of the Drosophila R-PTPs, which run greater than 100 kDa in SDS gels according to theoretical calculation (http://www.uniprot.org), was unraveled by the in-gel phosphatase activity assay (Fig. 1). Obviously, other methods are needed to collect information for profiling R-PTPs at different stages of Drosophila development.

image

Figure 1.  In-gel assay reveals the dynamic change of PTP activity in multiple stages during Drosophila development. The total protein extracts (35 μg each) were collected from the whole fly at various developmental stages as indicated, and then applied to an SDS gel cast in the presence of radioisotope-labeled PTP substrate. Left panel: After the complete process with the denaturation and renaturation buffers, the gel was exposed to an X-ray film for visualization of the PTPs. The clear zones shown in each lane represent the activity of phosphatases according to their molecular weights resolved by SDS/PAGE. Right panel: The Coomassie blue staining of the same gel indicates an equal amount of protein extract loaded in each lane.

Download figure to PowerPoint

Data mining of microarray databases to depict the mRNA expression profiles of R-PTPs during Drosophila development

Since the protein expression profiles of R-PTPs were not available, we switched our focus to other information such as the mRNA level of these phosphatases over various developmental stages. Flybase provides modENCODE temporal expression data for each gene [33,34] (http://www.modencode.org). Since this database shows the expression pattern of every gene throughout the development of flies and is easily accessed, we decided to perform data mining to profile R-PTPs using already existing information embedded in Flybase. We analyzed all R-PTPs in the Drosophila genome, including dLAR, PTP4E, PTP10D, PTP52F, PTP99A and PTP69D but excepting dIA2 which is a naturally inactive phosphatase. We examined the mRNA levels of these R-PTPs at various developmental stages (the embryo, early and late third instar larva, white prepupa, pupa and adult). As seen in Fig. 2, all R-PTPs except PTP52F were highly expressed during embryogenesis. PTP4E, PTP10D and PTP69D levels were particularly pronounced in embryos (Fig. 2A,B), suggesting that they may play essential roles in the control of developmental events at this time. These findings are in agreement with those of two recent studies reporting that PTP4E, PTP10D and PTP69D act in coordination of axon guidance during embryogenesis and that they have redundant and compensating functions [35]. Perhaps the most interesting observation in the profiling was the gradually increased expression of PTP52F from the embryonic stage to the larva–pupa transition (Fig. 2A,B), at which time metamorphosis begins and most larval tissues are readily remodeled for the development to adult flies [36]. The discovery that PTP52F is specifically expressed at this particular stage of development suggested that R-PTPs may be involved in the control of metamorphosis, despite a very low activity of non-R-PTPs being observed at this time of pupal formation (Fig. 1).

image

Figure 2.  Data mining reveals the mRNA expression profile of R-PTPs in multiple stages during Drosophila development. The mRNA levels of six Drosophila R-PTPs at various developmental stages were obtained from the modENCODE Temporal Expression Data of the Flybase. (A) Heat map showing relative changes in R-PTP expressions following the onset of embryogenesis until adult. The heat map scale indicates the relative fold expression of genes within this group as log2 of the actual value. (B) A simple graphical representation of the heat map for easy visualization. Note that PTP52F and PTP99A were expressed in relatively high levels during the larva–pupa transition.

Download figure to PowerPoint

PTP52F is highly enriched in midgut during the larva–pupa transition

We wanted to examine the expression of R-PTPs in various tissues of Drosophila, particularly focusing on the tissue distribution of PTP52F, which is highly expressed during the larva–pupa transition (Fig. 2). We used data mining to profile the expression of R-PTPs in the third instar larval tissues including salivary gland, central nervous system, trachea, tubule, hindgut and midgut using the FlyAtlas database, which provides the most comprehensive collection of the mRNA level data on each gene in each tissue during the third instar larva and adult stages of development [37] (http://www.flyatlas.org). Data were mined using three criteria: (a) mRNA SIGNAL, the abundance of mRNA; (b) mRNA ENRICHMENT, mRNA of the specific gene compared with the total mRNA of the whole flies; and (c) the Affymetrix PRESENT CALL, the number of times a specific gene was detectably expressed out of four arrays being analyzed. To summarize our findings, we present a simplified anatomy of the third instar larvae showing major tissues with relative expression levels of R-PTPs (Fig. 3). Four out of six R-PTPs (dLAR, PTP4E, PTP99A and PTP69D) were detected in the central nervous system, suggesting that they play important roles in the neuronal formation not only during embryogenesis but also at the beginning of metamorphosis. To our surprise, we found PTP52F to be exclusively expressed in gut tissues, and to be particularly enriched in the midgut (Fig. 3). Since this kind of tissue distribution of R-PTPs had never been recorded in flies, we hypothesized that PTP52F may have a specific role in the regulation of gut tissue during the larva–pupa transition. Thus, the remainder of the study was devoted to the characterization of PTP52F and the study of its potential involvement in Drosophila development.

image

Figure 3.  Tissue distribution of R-PTPs at the third instar larval stage. The diagram on the left shows the major tissues that are undergoing metamorphosis during the transformation from larva to pupa. The relative mRNA level of each R-PTP in these tissues obtained from the FlyAtlas is shown in the right panel. For easy visualization, the numerical value of mRNA expression has been converted into a scale of enrichment indicated by the plus sign. Note that PTP52F was specifically enriched in the midgut.

Download figure to PowerPoint

Identification of PTP52F as an active plasma-membrane-localized phosphatase

We performed sequence analysis on the full-length cDNA of Ptp52F gene that we cloned from flies and found it to be identical to the one published previously [38]. As shown in Fig. 4(A), the PTP52F protein contains a putative signal sequence, six fibronectin type III repeats, a single transmembrane segment and a single phosphatase domain in the intracellular region. We generated an antibody specifically targeting the C-terminal tail outside the phosphatase domain (Fig. 4A). This antibody recognized the wild-type (WT) phosphatase ectopically expressed in Drosophila S2 cells (Fig. 4B). We noticed that the full-length PTP52F ran as a single band at a higher position (∼ 200 kDa) than the predicted size (∼ 160 kDa) in SDS gels (Fig. 4B), suggesting the occurrence of post-translational modifications. The phosphatase activity of the WT form of PTP52F, but not the C1290S mutant (CS) form, was confirmed by the typical pNPP assay (Fig. 4C). Furthermore, an in vitro assay demonstrated that partially purified PTP52F could dephosphorylate pTyr proteins in S2 cell lysates (Fig. S1), suggesting the tyrosine phosphatase activity of this R-PTP. To examine the subcellular localization of PTP52F, we performed immunofluorescence staining with PTP52F antibody together with F-actin co-staining in S2 cells attached to the lectin concanavalin A (ConA) substrate. When spreading on the ConA-coated surface, cortical actin structure in S2 cells would be concentrated at the cell periphery, thus allowing us to determine the leading edge of the plasma membrane as demonstrated in our previous study [39]. Using F-actin staining as the guidance, we have observed that the ectopically expressed WT form of PTP52F is highly enriched near the plasma membrane (Fig. 4D). Considered together, these findings suggest that PTP52F acts as an active form of receptor tyrosine phosphatase.

image

Figure 4.  Characterization of PTP52F as an active phosphatase located in the cell periphery. (A) The layout of fibronectin III repeats (FN3) and a phosphatase domain in the basic architecture of full-length PTP52F. The motif used as the epitope for antibody generation on the C-terminal region is marked. (B) Various amounts of plasmid were used for expressing the WT form of HA-tagged PTP52F in S2 cells. Aliquots of total lysates were subjected to immunoblotting with anti-PTP52F antibody and anti-HA antibody. (C) The HA-tagged WT form and CS form of full-length PTP52F were expressed in S2 cells and then purified by immunoprecipitation with anti-HA antibody. Equal amounts of both WT and CS forms of immunoprecipitated PTP52F were subjected to the phosphatase activity assay using pNPP as a substrate. The inset summarizes the results of immunoblotting with anti-PTP52F antibody. Similar levels of immunoprecipitated PTP52F were used for the activity assay. (D) The WT form of PTP52F was expressed in S2 cells. The subcellular localization of PTP52F was then examined in ConA-coated coverslips by immunofluorescence staining with anti-PTP52F antibody. Co-staining with cortical F-actin showed the clear localization of PTP52F in the cell periphery. Bar, 10 μm.

Download figure to PowerPoint

Endogenous PTP52F protein is localized specifically in the larval midgut

We used the antibody that we generated to examine the expression of endogenous PTP52F in total protein extracts isolated from both WT and PTP52F knockdown (RNA interference, RNAi) flies at the third instar larval stage, when mRNA of PTP52F was robustly enriched (Figs 2 and 3). As shown in Fig. 5A, the specific band at ∼ 200 kDa appeared only in the WT flies but not in the RNAi line, suggesting that endogenous PTP52F protein was indeed expressed during the larva–pupa transition. The excellent performance of this antibody indicated that we could use it to further characterize PTP52F in developing flies with various genetic backgrounds. The results of our data mining of the FlyAtlas (Fig. 3) suggested that there would be a robust level of PTP52F protein in the prepupal midgut tissue. To find out, we performed immunofluorescence staining with anti-PTP52F antibody. As shown in Fig. 5B, there was a strong signal of PTP52F in the midgut of WT flies, suggesting that a high level of mRNA at this developmental stage (Figs 2 and 3) might lead to enhanced expression of its protein in larval epidermal cells or adult epidermal progenitor cells, both of which are major types of cells in the larval midgut [40]. In contrast, immunofluorescence staining of the salivary gland with anti-PTP52F antibody resulted in negative signal (Fig. S2), consistent with an undetectable level of PTP52F mRNA in this part of the prepupal flies (Fig. 3). We further performed immunofluorescence staining using PTP52F knockdown flies driven by Tubulin-Gal4 to ensure the specificity of the antibody in vivo. Focusing on the midgut tissues, we observed that the PTP52F signal was completely ablated in the RNAi transgenic flies, whereas the staining of PTP52F in the WT flies was robust (Fig. 5C). Using this antibody, we further tested the distribution of endogenous PTP52F in the whole gut tissue of the WT flies at the pupal stage. Interestingly, the staining of PTP52F protein was highly enriched in the midgut but much weaker in the hindgut region (Fig. 5D), consistent with the information of its mRNA distribution provided by the FlyAtlas (Fig. 3).

image

Figure 5.  Expression of endogenous PTP52F in the larval midgut. (A) Aliquots of total protein extracts isolated from the WT and PTP52F knockdown (RNAi) flies at pupal formation (PF) and 2 h after pupal formation (2 h-APF) were subjected to immunoblotting with anti-PTP52F antibody. (B) The midgut tissues isolated from the PF stage of WT flies were subjected to immunofluorescence staining with anti-PTP52F antibody. Bar, 20 μm. (C) The midgut tissues isolated from both WT and PTP52F knockdown (KD) flies were stained with anti-PTP52F antibody. Note that the fuzzy green color shown in the PTP52F KD tissue was contributed by the background auto-fluorescence. Bar, 200 μm. (D) The diagram shown at the top indicates the layout of gut tissue at the prepupal stage. Abbreviations: GC, gastric ceaca; MT, malphigian tubule; HG, hindgut. Images in the lower panel show that the expression of PTP52F was specifically restricted to cells located in the midgut but not in the hindgut, as revealed by immunofluorescence staining with anti-PTP52F antibody. Bar, 20 μm.

Download figure to PowerPoint

The essential role of PTP52F in Drosophila development

To test whether enriched expression of PTP52F in the midgut of prepupal flies plays a key role in development, we ablated endogenous PTP52F from the whole body of flies using the Tub-Gal4 driver and then examined the phenotype. The RNAi transgenic flies developed throughout embryogenesis up to the larval stage, although the speed was significantly slower than WT flies (data not shown). Interestingly, knockdown of PTP52F led to death of organisms during the post pupal stage (Fig. 6A). We observed a high percentage of pupal death, which was shown as the typical pharate adult phenotype (53%) due to eclosion failure at the pupal cage (Fig. 6A). A small fraction of flies died as early pupa (17%), presumably also caused by eclosion failure (Fig. 6B). Some flies (30%) survived for a little longer beyond the pupal stage, although they eventually died either during eclosion or soon after eclosion (Fig. 6B). Since PTP52F is highly enriched in the midgut during the larva–pupa transition (Fig. 5) and the midgut undergoes extensive metamorphic changes during this stage, we also checked whether PTP52F knockdown affects the developmental process of the midgut. As shown in Fig. 6C we noticed a significant delay of midgut degradation in PTP52F RNAi transgenic flies, whereas the midgut of WT flies became much shorter at 4 h after the pupal formation stage. Taken together, these results demonstrate that PTP52F plays an essential role in the control of key steps that guide metamorphosis and eclosion and it must be expressed properly during the larva–pupa transition.

image

Figure 6.  Ablation of endogenous PTP52F via RNAi led to pupal lethality. (A) The pharate adult phenotype was observed when endogenous PTP52F expression was knocked down by the Tubulin-Gal4 driver. (B) The table summarizes the lethal phase analysis of PTP52F knockdown flies. Note that over 50% of flies died as pharate adults. Most flies developed until the pupal stage but then died due to eclosion failure. Only a small fraction of flies (13%) died upon completion of eclosion. (C) The images of midgut tissues isolated from both WT and PTP52F knockdown (KD) flies at 4 h after pupal formation (4 h-APF) were captured for a morphological comparison. Note that the metamorphosis of PTP52F-KD flies was significantly delayed as demonstrated by the much longer gut tissue compared with the WT flies. Bar, 1 mm.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. References
  8. Supporting Information

Data accumulated so far have suggested that the involvement of Drosophila R-PTPs is mostly in neural-specific functions [27,38,41]. Essentially nothing was known about the functions for Drosophila R-PTPs outside the nervous system until a very recent finding that demonstrated the role of PTP4E and PTP10D in the determination of tracheal tube geometries in embryos [42]. However, it must be noticed that, no matter what the origin of the cell type or what signaling pathway has been discussed so far, the functional role of R-PTP was only shown in Drosophila embryos due to the lethality of R-PTP mutant flies during embryonic development. In the current study, PTP52F was found by data mining to be specifically enriched in the midgut during the larva–pupa transition, a finding we confirmed through genetic manipulation combined with immunoblotting and immunofluorescence approaches. Thus, we were able to show for the first time that an endogenous R-PTP protein is highly expressed outside the nervous and tracheal systems at a developmental stage other than embryos.

To date, our understanding of PTP52F in the regulation of signal transduction is limited, largely due to the lack of knowledge about substrates and signaling pathways regulated by this phosphatase. A recent study showed that Tartan, a transmembrane protein, is a candidate substrate of PTP52F [41]. Genetic experiments further illustrate that PTP52F and Tartan act synergistically in pathways of motor axon guidance in embryos [41]. Tartan was also detected in tracheal tissue [43]. However, nothing is known about the distribution or potential functions of Tartan in the midgut of larva and pupa. Therefore, it is difficult to predict whether the phenotype of pharate adult lethality shown in the PTP52F knockdown flies (Fig. 6) was influenced by the deregulation of Tartan or other yet-to-be-identified substrates of PTP52F. Further investigations are required to delineate the molecular basis for PTP52F-mediated regulation of signaling pathways in the midgut during the larva–pupa transition. Nonetheless, it is interesting to point out that the pharate adult phenotype revealed by the ablation of PTP52F has also been reported in some mutants of ecdysone response genes [40,44], including ecdysone receptor (EcR) [45], a key mediator of metamorphosis during the transformation from larva to pupa. The similarity of the phenotype suggests a possibility that PTP52F may participate in ecdysone-mediated signal transduction. Indeed our preliminary data support a genetic interaction between EcR and PTP52F. Using the eye phenotype as a readout, we found that overexpression of EcR B1 form driven by GMR-Gal4 led to the severe loss of hexagonal ommatidia (Fig. S3), whereas only mild disruption of the ommatidia arrangement was observed in PTP52F RNAi lines (Fig. S3). Interestingly, the eye phenotype revealed in EcR overexpressed flies could be rescued through the knockdown of PTP52F (Fig. S3), suggesting a possible synergistic interaction between these gene products in a signaling pathway. Based on these observations, we hypothesized that Ptp52F might be a downstream gene transcriptionally regulated by EcR. As a typical steroid hormone receptor, EcR recognizes specific consensus motifs on the promoter region of effector genes [46,47]. We therefore analyzed the 2 kb region upstream of the first exon of the Ptp52F gene to search for potential ecdysone response elements using the software nubiscan-2.0 (http://www.nubiscan.unibas.ch/), which is an in silico tool that helps predict nuclear receptor binding sites. Based on the analysis, two possible ecdysone response elements with high scores were identified (Fig. S4A). Importantly, the sequences of these elements are similar to the consensus motif shown in the known EcR-regulated genes (Fig. S4B). Together these results suggest that Ptp52F may be a downstream gene of ecdysone action.

The latest version of computational and bioinformatics analysis defined PTP52F as an unclassified member in the PTP super family without a clear ortholog in humans [26]. However, based on the sequence of full-length PTP52F clones obtained by us (Fig. 4A) and in an earlier report by Zinn’s group [26,38], we propose that the classification of Drosophila R-PTPs be revised. Apparently PTP52F contains only one catalytic domain in the intracellular region instead of two tandem putative phosphatase domains, as suggested previously [38]. The overall layout of PTP52F architecture composed of multiple fibronectin repeats, a single transmembrane segment plus a single phosphatase domain (Fig. 4A) reclassifies this phosphatase as a member belonging to the subtype R3, together with PTP4E and PTP10D in the Drosophila PTP family (Fig. 7A). Thus, the earlier study, which defined PTP52F as an unclassified phosphatase, should be modified. As shown in Fig. 7A, there are three Drosophila R-PTPs and six human R-PTPs in the subtype R3. Obviously, it is difficult to classify the ortholog pair merely based on the sequence alignment. Other important criteria such as the regulatory role in evolutionarily conserved signaling pathways or the tissue-specific expression profile of R-PTPs across species must be taken into consideration. Recently, PTP10D and PTP4E were regarded as the functional orthologs of human density-enriched PTP-1 due to their similar expression in epithelial cells and their shared ability to downregulate receptor tyrosine kinase mediated signaling [42]. Following the same principle of consideration, we propose that PTP52F might be the functional ortholog of human and mouse stomach-associated PTP-1 (SAP-1) (Fig. 7B). Accumulated data clearly show that mammalian SAP-1 is exclusively expressed in gastrointestinal epithelial cells [48], similar to the midgut-enriched expression of PTP52F in flies. Such a specific expression profile of both PTP52F and SAP-1 suggests that they may regulate evolutionarily conserved signaling pathways in the gut tissue.

image

Figure 7.  Reclassification of PTP52F as a member of subtype R3 PTPs. (A) Sequence analysis suggests that PTP52F, together with PTP10D and PTP4E, belongs to the subtype R3 in the PTP superfamily. There are six members of human R-PTPs in this subtype. (B) A proposed model showing that PTP52F might be the functional ortholog of human and mouse SAP-1.

Download figure to PowerPoint

In conclusion, data shown in the present study suggest that endogenous Drosophila R-PTPs act in developmental control outside the nervous and tracheal systems and also beyond the early embryonic stage and thus potentially play an indispensable role in the regulation of metamorphosis. Although additional experiments are needed to support this hypothesis, our study has opened a new avenue for understanding the role of Drosophila R-PTPs that may mediate signal transduction during development and other biological processes in areas beyond our current knowledge.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. References
  8. Supporting Information

In-gel phosphatase activity assay

An in-gel phosphatase activity assay was performed as described in previous studies [31,32]. The SDS gel was prepared with the 32P-labeled substrate to obtain a gel solution of 1.5 × 106 cpm (20 mL)−1, equivalent to approximately 2 μm pTyr. The fly lysates of specific stages were collected with lysis buffer containing 1% NP-40 and stored at −80 °C. For each lane, 35 μg of total protein was loaded. After electrophoresis, the gel was processed with various buffers sequentially for fixation, protein denaturation and renaturation. In the final step of renaturation, dithiothreitol (3 mm) was included in the buffer for activation of PTPs in the gel. The dephosphorylation reaction was terminated by staining the gel with Coomassie blue. After destaining, the gel was dried and exposed to X-ray film.

Microarray data mining

For the expression profile of R-PTPome, we utilized the modENCODE Temporal Expression Data from Flybase. The National Human Genome Research Institute’s model organism encyclopedia of DNA elements (modENCODE) project provides the biological research community with a comprehensive encyclopedia of genomic functional elements in the model organisms Caenorhabditis elegans and D. melanogaster [33,34]. The expression levels were analyzed and different levels of expression were recorded. The tissue-specific expressions of R-PTPs were found in FlyAtlas, which is the Drosophila gene expression atlas available online [37] (http:http://www.flyatlas.org).

Generation of PTP52F antibody

A fragment of synthetic peptide from the intracellular domain of PTP52F (amino acids 1400–1417) was used as the peptide for antibody production. The epitope was selected based on the scores from an antigenicity prediction program. The rabbit antisera showing the specificity was further purified by affinity chromatography. The purified antiserum was used at a final dilution of 1 : 10 000 for immunoblots and 1 : 1000 for immunofluorescence staining.

Cloning and expression of PTP52F

The cDNA sequence corresponding to full-length PTP52F was obtained by reverse transcription of total mRNA from the pupa formation stage of the fly. The full-length WT and CS mutant were constructed with a hemagglutinin (HA) tag at the C terminus and cloned into pAc5.1A (for protein expression in S2 cells) vector. All cDNAs were examined by sequencing.

Cell culture, transfection, and immunoprecipitation and immunoblotting of PTP52F

Drosophila S2 cells were routinely maintained in Schneider’s medium supplemented with 10% fetal bovine serum. For transient transfection with the PTP52F expression vector, S2 cells (5 × 105 cells per 6 cm plate) were incubated with a mixture of plasmid DNA (5 μg per 6 cm plate) and Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA), according to the manufacturer’s instructions. For immunoprecipitation, the cells were lysed in buffer containing 50 mm Tris/HCl (pH 8.0), 1% NP-40, 150 mm NaCl, 2 mm Na3VO4 and protease inhibitors. An aliquot of total lysate (1 mg) was precleaned with protein G-Sepharose (GE Healthcare, Uppsala, Sweden) for 30 min at room temperature followed by immunoprecipitation with anti-HA antibody (clone HA-7, Sigma) at 4 °C for 3 h and then by the elution of the beads with 2× sample buffer. For immunoblotting, 35 μg of total protein was loaded with 2× sample buffer for each sample followed by gel running and blocking with primary antibody and secondary antibody. Rabbit anti-PTP52F antiserum and mouse anti-HA IgG (Sigma, St. Louis, MO, USA) was used in this study.

Phosphatase activity assay

Assays were performed as previously described [49] with a few modifications. Briefly, S2 cells overexpressing HA-tagged PTP52F-WT or PTP52F-CS were harvested in lysis buffer. HA-PTP52F was immunoprecipitated from an aliquot of total lysate (1.5 mg) by HA antibody. The immunocomplex was incubated in phosphatase activity assay buffer (50 mm Hepes, pH 7.5, 1% NP-40, 10 mm dithiothreitol and 20 mmp-nitrophenyl phosphate). The reaction was carried out at 37 °C for 1 h. After the reaction was terminated by 2 m NaOH, the phosphatase activity in the immunocomplex was measured by spectrometric analysis at 405 nm.

Fly stocks

All flies were maintained at 25 °C unless otherwise indicated. The following strains were obtained from various sources: OreR flies (Bloomington), PTP52F knockdown line (3116, VDRC), Tubulin-gal4, GMR-gal4 (Bloomington). The GAL4-UAS system was used to generate progeny expressing the target gene in a tissue-specific pattern [50].

Immunofluorescence staining

S2 cells were plated and transfected as described above. After 48 h of incubation, the cells were suspended and replated on ConA (0.5 mg·mL−1; C2010; Sigma) coated glass coverslips and stained with anti-PTP52F antibody followed by Cy2-conjugated goat anti-rabbit IgG (Jackson ImmunoResearch, West Grove, PA, USA). For F-actin staining, cells were stained with tetramethyl rhodamine isocyanate conjugated phallodin (Jackson ImmunoResearch, West Grove, PA, USA). The samples were visualized using a Zeiss LSM 510 confocal microscope. Late third instar larval midguts were dissected in NaCl/Pi, and were fixed, permeabilized and stained with PTP52F antibody followed by Cy3-conjugated goat anti-mouse IgG (Jackson ImmunoResearch, West Grove, PA, USA) according to the protocol described in a previous study [51]. The samples were visualized using a Zeiss LSM 510 confocal microscope.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. References
  8. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. References
  8. Supporting Information

Fig. S1. Tyrosine phosphatase activity and plasma membrane localization of PTP52F.

Fig. S2. Immunofluorescence staining of salivary gland with anti-PTP52F antibody.

Fig. S3. Genetic interaction between EcR and PTP52F.

Fig. S4. Identification of potential ecdysoneresponse elements in the promoter region of Ptp52F gene.

Please note: As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer-reviewed and may be re-organized for online delivery, but are not copy-edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.

FilenameFormatSizeDescription
febs8696_sm_FigS1-S4.zip476KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.