Noordermeer et al. (1998)
position of the gene product of interest in the gradient can then be compared to these markers.
Method C: Coimmunoprecipitation for Identifying Interacting Proteins
The entire genome of flies is known, and several projects overseen by the BDGP have the goal of generating mutants in virtually all of the genes. The next step is to identify the function of all these genes and to identify physically interacting proteins. The latter question can be approached by immunoprecipitations using antibodies recognizing native proteins or epitope-tagged fusion proteins. The benefits of each of these approaches are discussed further in the section on protein expression.
An example of the utility of this approach is a recent study demonstrating the suprising result that certain GABA and glutamate gated-channels may heterodi-merize in vivo. A GABA-gated channel had been identified previously, but when expressed in a heterologous system, the pharmacology was distinct from the native channel (Zhang et al., 1995) and glutamate was known to regulate ligand binding (Smith et al., 2000). Coimmunoprecipitation from membranes prepared from head extracts demonstrated that a GABA and glutamate gated-channel can coassemble using two radiolabeled ligands and antibodies to both subunits.
For immunoprecipitations, antibodies should be bound to protein A, G, or A/G beads depending on the host species of the antibody (see manufacturers instructions) to as high a stoichiometry as possible. This allows for fewer beads to be used, thereby lowering nonspecific background binding. Antibodies should be affinity purified (see Harlow and Lane, 1999) and incubated overnight with protein A- or g-linked beads at 4° C. Nonspecific rabbit IgG, preimmune IgG, and/or beads alone should be used as controls for the affinity-purified antibodies. For the best results, IgGs are linked covalently to the protein A beads using 5-20 mM dimethylpimelimidate or 0.5-4 mM disuccinimidyl suberate (DSS) as described (Harlow and Lane, 1999). Covalent coupling prevents the IgG subunits from masking interacting proteins during analysis and allows the reuse of the beads. The beads are then washed with high salt (1M NaCl or 4 M MgCl2) and detergent-containing buffers (up to 1% TX-100) to release unbound antibodies followed by a wash in phosphate-buffered saline (PBS) with 0.1% bovine serum albumen (BSA). The lysate can then be incubated with control beads to preclear the extract (Harlow and Lane, 1999). The specific antibody beads are mixed with the precleared lysate (SI) or subcellular S2-P3 fractions in equivalent volumes for 2h (or overnight) at 4°C, washed with 0.1% BSA in PBS, and finally in PBS. Bound material is eluted by 0.1 M glycine, pH 2.5, or by boiling in reducing SDS sample buffer. The immunoprecipitate is electrophoresed on SDS-PAGE and is Western blotted for the antigen and suspected interactors. Alternatively, analysis of the immuoprecipitate by two-dimensional gels followed by matrix-assisted laser desorption/ionization time-of-flight mass spectroscopy (MALDI-TOF MS) can be used to identify interacting proteins (for review, see Nordhoff et al., 2001).
Two-dimensional gels allow one to separate proteins based on molecular weight and isoelectric point on large gels. This makes it more likely to pick a spot containing a single protein, and mass spectrometric analyses (MALD-TOF) can allow the identification of the protein. A good control, if it is available, is an immunopreci-pitation reaction for a true protein null or one in which the epitope is truncated.
Some biochemical studies of Drosophila proteins have utilized S2 cells to study neuronal proteins. These cells tend to express a large set of proteins, including neuronal ones. RNAi (for review, see Worby et al., 2001) and biochemical experiments have been very successful in these cells, including looking for or demonstrating protein/protein interactions (Hwang et al., 2000; Worby et al., 2001b, 2002). These studies can also serve as a starting point for the choice of generating mutants in newly identified interacting proteins and subsequent determination of their in vivo function in flies (Schmucker et al., 2000). A nice example of this latter approach is a series of studies of DOCK, an adaptor protein that functions in axonal guidance. Immunopreciptation of DOCK from S2 cells and embryos led to the observation that DOCK interacts with several phosphotyro-sine-containing proteins, including p270 (Clemen et al., 1996). Further coimmu-noprecipitations of a tagged version of DOCK (SH2) expressed in S2 cells demonstrated several interacting proteins, including the phosphotyrosine containing Down syndrome cell adhesion molecule (Dscam-p270) (Schmucker et al., 2000; Worby et al., 2001a). Mutants of DSCAM were isolated and shown to also have a role in axon guidance (Schmucker et al., 2000) consistent with coimmuno-precipitation data of Dock and Dscam. Dock and Dscam also interact genetically (Schmucker et al., 2000).
Method D: Lipid Analyses
The effect of diet on lipid metabolism has been studied in flies, and mutants have been isolated that putatively should disrupt lipid metabolism in the nervous system (Yoshioka et al., 1983; Fujita et al., 1987; Inoue et al., 1989; Stark et al., 1993; Pavlidis et al., 1994). To address these issues, methods of extracting and analyzing lipids in fly heads have been developed. Lipids have been found to modify many parameters of a neural cell, but still the role of lipid composition in membranes is not well understood. Genetic approaches in yeast have been helpful at addressing this issue, and an example of addressing the role of lipids in neuronal function is the study of the easily shocked gene, which encodes an ethonolamine kinase. Mutants in this gene are paralyzyed easily by brief, intense electrical stimulation (Pavilidis et al., 1994). Phospholipid analysis indicated that phospha-tidylethanolamine levels were depressed. Synthesis of phosphatidylethanolamine requires ethanolamine kinase and highlights the importance of phospholipid metabolism for neuronal function. PIP2 has been shown to play critical roles in the nervous system. PIP and PIP2 levels can be analyzed readily in the Drosophila head by thin-layer chromatography coupled to immunological detection and using labled inorganic phosphate as described later. A complementary approach to the direct measurement of PIP2 has been demonstrated by a biosensor of PIP2 (Hardie et al., 2001). With this technique, one can determine relative levels of PIP2 in vivo.
Briefly, 20 fly heads are homogenized in 400 yl of 0.32 M sucrose, 50 mM Tris, pH 7.4, and 1mM EDTA for 60s in 1.5-ml Eppendorf tubes. Then 0.8ml of chloroform/methanol (2:1 v/v) is added, and the tissue is further homogenized for 60s. Another 0.8ml is added and vortexed for 2min. This mixture is centri-fuged at 2500 x g for 5 min and the lower organic phase is removed and saved. Two more extractions are done with 2 volumes of acidic chloroform/methanol/ 12N HCL (2:1:0.0125, v/v/v) and the lipid containing lower phases are pooled and neutralized with one drop of 4 N NH4OH. The organic phase is dried under nitrogen gas and the sides of the container are rinsed with chloroform/methanol (2:1). Samples are resuspended in chloroform/methanol (2:1) and either stored at —80° C or applied to a thin-layer chromatography (TLC) plate. The TLC plates (Whatman, aluminum backed) are prepared by treating with 1% potassium oxalate in methanol, removing the excess, and heating at 110° C for 1 h. Samples are applied to the plates based on the number of heads extracted or by quantitating total phospholipids by measuring phosphorus content (Ames, 1966). The TLC plates are then developed by solvents depending on the desired separation (Inoue et al., 1989; Stark et al., 1993). For example, PIP and PIP2 can be analyzed by developing in 65% isopropanol and 35% 2 M acetic acid. Once the solvent front reaches the top of the plate, the TLC plate is allowed to dry in a hood. After drying, the TLC plate can be analyzed by several methods. For example, one can analyze both PIP and PIP2 levels by antibody binding and detection. When the TLC plate is completely dry (no acetic acid odor), the plates are blocked in PBS-B-PVP (PBS with 1% BSA and 1% polyvinylpyrrolidone, an average molecular weight of 10,000). The PIP (1:1000 Assay Designs, Ann Arbor, MI) and PIP2 (1:1000 Assay Designs or Echelon Labs, Salt Lake City, UT) antibodies are applied and incubated for 2 h. The plates are washed four times in PBS-B-PVP, incubated with either HRP or alkaline phosphatase-coupled secondary antibodies for 1 h, and then rinsed again as for primary antibodies. The plates are developed in the appropriate reagents (Lane and Harlow).
Few phospholipid-specific antibodies exist, but methods based on the incorporation of radioactive precursors can be used (Inoue et al., 1989; Stark et al., 1993). For example, flies can be fed 1-12% sucrose containing 32P-labeled inorganic phosphate for up to 24 h. Labeled flies are frozen in liquid nitrogen, vortexed, and their heads collected. Lipids are extracted as described earlier and are then analyzed by various methods on TLC plates (Inoue et al., 1989; Stark et al., 1993). Identification of specific molecules is made by using standards and detected using iodine or ninhydrin vapors (Skipski and Barclay, 1969). Quantitation can be achieved by scraping the radioactive spots identified by autoradiography, and the radioactivity measured using a liquid scintillation counter. Alternatively, one can use a phosphoimager such as the one made by Molecular Dynamics and its associated software.
Was this article helpful?