Affinity Purification of Protein Complexes from Drosophila Embryos in Cell Cycle Studies

Abstract

The ability to identify protein interactions is key to elucidating the molecular mechanisms of cellular processes, including mitosis and cell cycle regulation. Drosophila melanogaster, as a model system, provides powerful tools to study cell division using genetics, microscopy, and RNAi. Drosophila early embryos are highly enriched in mitotic protein complexes as their nuclei undergo 13 rounds of rapid, synchronous mitotic nuclear divisions in a syncytium during the first 2 h of development. Here, we describe simple methods for the affinity purification of protein complexes from transgenic fly embryos via protein A- and green fluorescent protein-tags fused to bait proteins of interest. This in vivo proteomics approach has allowed the identification of several known and novel mitotic protein interactions using mass spectrometry, and it expands the use of the Drosophila model in modern molecular biology.

1 Introduction

Mitotic progression involves a complex and highly dynamic network of protein–protein interactions. The ability to dissect these associations is important to understand how specific proteins function relative to each other. The affinity purification of individual bait proteins, followed by the identification of associated factors by mass spectrometry, has allowed the detection of such networks involved in many cellular processes [1, 2]. However, in mitosis, many crucial interactions are restricted to a small time window or to discrete subcellular structures and often require specific posttranslational modifications. For this reason, mapping the mitotic protein interactome has proved challenging [3].

Proteins controlling cell division tend to be well conserved between species and have been studied in many model organisms. Yeast and vertebrate cells can be arrested and even synchronized in mitosis, in order to enrich for complexes of interest prior to their purification. This strategy has been successful, although it remains possible that an observed interaction does not reflect the normal physiology of the system, due to the artificial cellular state induced by the drug treatment. The fruit fly Drosophila melanogaster has contributed greatly to increasing our molecular understanding of mitosis. The genetic tools available in flies are extremely well developed for in vivo studies, and Drosophila cells in culture also offer several advantages. For example, we have previously developed protocols for the purification of protein complexes from Drosophila cultured cells [4], which have led to the identification of a number of new interactions and to the dissection of several molecular mechanisms of cell cycle regulation, mitosis, and cytokinesis [5–7]. Yet, these cells, such as Schneider (S2) or D.Mel-2 cells, are notoriously difficult to arrest and synchronize in mitosis, which limits the isolation and detection of transient, low-abundance complexes acting specifically during cell division. Moreover, cultured cell lines, adapted to grow in vitro by the process of immortalization, do not always accurately represent the in vivo physiology of the biological system. Therefore, the development of an in vivo proteomics methodology would add a new approach to the Drosophila toolbox and expand its value as a leading model organism in the field of mitosis.

We decided to exploit Drosophila early embryos as starting material for the identification of mitotic protein interactions. After fertilized eggs are laid, 13 rounds of rapid and synchronous mitotic divisions of nuclei in a syncytium take place before cellularization of the blastoderm occurs. This developmental stage relies on proteins and mRNAs that are maternally deposited in the egg. Divisions occur approximately every 10 min, where M phase and S phase alternate without intervening gap phases [8]. Therefore, embryos collected before 2 h of age constitute a starting material that is highly enriched in mitotic protein complexes.

We express proteins of interest (bait proteins) in fusion with a tag to facilitate their isolation by affinity purification. A single-step purification, which takes advantage of a high affinity between a tag and a resin, enormously simplifies the process of protein isolation, making it easy and straightforward even for an inexperienced biochemist. Moreover, a fast procedure increases the chances of preserving unstable protein complexes.

Here, we describe protocols and reagents that we have developed or adapted for the expression and purification of proteins fused to either protein A (PrA)- or green fluorescent protein (GFP)-tags. A transgene coding for the fusion protein expressed maternally in eggs results in the presence of the tagged protein in the early embryos. This bait protein is incorporated into complexes with endogenous partner proteins. Following affinity purification of the bait, associated proteins can be identified by mass spectrometry. This method has been validated by the purification of several known and novel mitotic protein–protein interactions in our laboratories (unpublished). As expected, several of the associations identified in embryos were missed when Drosophila cultured cells were used, reflecting the importance of using embryos as a highly enriched mitotic source tissue. This approach also bears potential to probe interactions involved in other processes that occur in Drosophila early embryos.

2 Materials

2.1 Embryo Collection

• 1. Vials and bottles with standard Drosophila cornmeal food.

• 2. Transgenic flies expressing the bait protein: Few to tens of thousands of healthy and newly enclosed fruit flies.

• 3. From theses flies, 1 g of syncytial embryos per purification (see Note 1 ).

• 4. Embryo collection cages (Genesee Scientific) that can be fitted onto 100 mm-diameter plastic Petri dishes or large fly population cages (Genesee Scientific, cylindrical cages with 30 cm diameter and 60 cm length) suitable for large-scale embryo collections. Similar (homemade) cages made of plexiglass or durable acrylic can also be used. More information at: http://www.flystuff.com/general.php.

• 5. Grape juice agar (see Note 2 ): Dissolve 12.5 g sucrose and 11.25 g agar in 375 mL water. Boil until completely dissolved. Add 125 mL grape juice, allow to cool down, and distribute into plastic Petri dishes, which can be fitted onto embryo collection cages (100 mm-diameter), or into larger trays (e.g., 15 × 25 cm polystyrene trays) when using large fly population cages.

• 6. Yeast paste (see Note 2 ): Mix dry active yeast with distilled water until a homogenous paste with a consistency similar to peanut butter is obtained. Add water or yeast again as needed.

• 7. Sieves or mesh basket (Genesee Scientific) for small-scale embryo collections. For large-scale embryo collections, use large diameter (e.g., 10–20 cm) sieves (Endecotts Ltd, 70 μm pore size for embryo collection and 425 μm pore size for the collection of flies stuck to the yeast paste).

• 8. Paintbrushes: For embryo collection, normal soft paintbrushes can be used (size 3–6).

• 9. 50 % household bleach.

• 10. Distilled water.

• 11. Embryo wash solution (EWS): 0.7 % (w/v) NaCl and 0.05 % (w/v) Triton X-100.

• 12. Paper towels.

• 13. Flat-ended spatula.

• 14. Stereo microscope.

• 15. Benchtop micro centrifuge.

• 16. 1.5-mL microfuge tubes.

• 17. Analytical laboratory scale.

• 18. Liquid nitrogen.

2.2 Affinity Purification

We use rabbit IgG-coupled paramagnetic beads for PrA affinity purification. The detailed method for covalently coupling rabbit IgG to magnetic beads has been described previously [4]. For GFP affinity purification, we use the GFP-Trap system (ChromoTek GmbH).

2.2.1 Affinity Resins

• 1. Rabbit IgG-coupled paramagnetic beads for PrA affinity purifications: Home-made by covalently coupling rabbit IgG (MP Biomedicals) to Dynabeads^® M-270 Epoxy (Life Technologies).

• 2. GFP-Trap system (see Note 3 ) (ChromoTek GmbH).

2.2.2 Solutions

• 1. Extraction Buffer (EB) for PrA or GFP-Trap purification: 20 mM Tris–HCl pH 7.5, 150 mM NaCl, 2 mM MgCl₂, 0.5 mM Na-EGTA pH 8.0, 1 mM dithiothreitol (DTT), 0.1 % NP-40, 5 % glycerol, 1 mM phenylmethanesulfonyl fluoride (PMSF), EDTA-free complete protease inhibitor cocktail (Roche) (see Notes 4 and 5 ).

• 2. Wash buffer (WB): Same as EB.

• 3. Final wash buffer (FWB): 20 mM Tris–HCl pH 7.5, 150 mM NaCl, 2 mM MgCl₂, 0.5 mM Na-EGTA pH 8.0, 1 mM DTT (see Note 6 ).

• 4. Elution solution: Elution of the PrA-baits requires a freshly made and filtered solution of 0.5 M NH₄OH and 0.5 mM EDTA.

• 5. 1× Laemmli sample buffer: 60 mM Tris–HCl pH 6.8, 2 % (w/v) SDS, 10 % glycerol, 5 % 2-mercaptoethanol, 0.01 % (w/v) bromophenol blue.

2.2.3 Tools

• 1. Prechilled plastic sample micropestles or 7/15-mL glass Dounce tissue grinder (see Note 5 ).

• 2. Prechilled 10 mL plastic syringes and 0.8 mm × 40 mm needles.

• 3. Prechilled plastic/glass funnel-filter device: Place a double-layer Miracloth filtering cloth (EMD Millipore) into the funnel.

• 4. Prechilled 1.5-mL microfuge tubes and 15-mL conical tubes.

• 5. Cell Culture microscope.

• 6. Refrigerated benchtop micro centrifuge.

• 7. Rotating wheel.

• 8. Magnetic stands (Promega) suitable for microfuge or larger tubes.

• 9. Aspirator.

• 10. Vacuum concentrator.

• 11. Heat block.

• 12. Apparatus suitable for conventional (denaturing) sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and Western blotting. Use 4–12 % gradient SDS-polyacrylamide gels for optimal separation of small to large size proteins.

• 13. Silver-nitrate staining kit (Life Technologies).

3 Methods

3.1 Production of Embryos

3.1.1 Generation of Transgenic Flies with Embryonic Expression of the Bait Protein

Transgenic flies expressing the PrA- or GFP-tagged proteins can be generated by various methods including standard P-element random insertion in the embryonic germline or can be obtained from different laboratories or stock centers. We routinely use the Gateway system (Life Technologies) to generate recombinant plasmids suitable for the N- or C-terminal tagging of the bait protein (Table 1) [4]. When the endogenous promoter is used to drive the bait’s expression, we use standard cloning [9]; the authentic promoter sequence is amplified by PCR and cloned upstream to the bait’s CDS (previously fused in frame (5′ or 3′) with the affinity tag). Then the whole promoter-tag-CDS cassette is subcloned into the pCaSpeR 4 P-element vector lacking a general promoter followed by embryo injection using standard procedures [10].

Table 1 Vectors available for early embryonic expression of bait proteins of choice in fusion with GFP or PrA

We usually generate a stable population of transgenic flies (a stock) that can be expanded for the production of large amounts of embryos expressing the fusion protein of interest. We have experimented with the poly-ubiquitin promoter (pUB, ubiquitous expression), the Maternal α-Tubulin promoter (MAT, female germline expression), and the bait protein’s own promoter for the immediate and constitutive expression of the bait protein. This strategy is often successful, but the (over)expression of the bait and/or the position of the tag can sometimes be toxic by interfering with oogenesis or early embryogenesis. In extreme cases, it is impossible to obtain a transgenic stock. In other cases, the stock is difficult to amplify due to few eggs laid and/or a low percentage of embryos hatching. This problem can be overcome by using an inducible promoter. We often make transgene encoding the tagged protein under the control of the inducible Gal4-UAS system [11]. Its expression can be driven in the female germline and in early embryos using a Maternal α-Tubulin-Gal4-VP16 driver (we use an insertion of this transgene on chromosome II, which is homozygous viable and fertile; Bloomington stock 7062 or 7063). It is important to use the UASp variant because the more widely used UAS _T does not (or poorly) allow expression in the female germline [12, 13]. Other Gal4 drivers enable expression of the same bait protein in the same transgenic line in a variety of other tissues (providing tools also for genetic or biochemical experiments).

When using the Gal4-UASp system, a Maternal α-Tubulin-Gal4-VP16; MKRS/TM6B stock can be crossed to a w; If/CyO; UASp-transgene/TM6B stock to generate a w; Maternal α-Tubulin-Gal4-VP16/CyO; UASp-transgene/TM6B expression stock. The markers and balancer chromosomes as well as the localization of the driver and transgenic constructs may differ. Once generated, the stock can be easily amplified. The toxicity of the transgenes expressed will vary. In some cases, overexpression of the transgene greatly reduces fertility, and the stock tends to retain the balancers. In other cases, the expression stock tends to lose the balancers and becomes homozygous with two copies of the driver and/or two copies of the other transgene. This will not happen if a transgene insertion is homozygous lethal. The natural adaptation of the expression stock helps its rapid amplification and maximizes its fertility and the yields of embryo collections. In any case, several insertions of the transgene should be screened for health of the expression stock and levels of the bait protein. Expression can be tested by Western blot on protein extracts prepared from embryos or abdomens from well-fed females. We usually aim for the highest possible expression if the stock is healthy because low expression has been a limiting factor in complex purification, particularly for large bait proteins. However, if the amount of the protein is important, we choose the stock expressing the tagged protein at a similar level to the endogenous protein. Finally the expression level of UASp transgenes can also be tuned by varying the temperature between 18 and 29 °C because the GAL4 driver is more active at higher temperatures.

3.1.2 Embryo Collection

• 1. Amplify the expression fly lines in vials or bottles containing standard Drosophila cornmeal food with yeast.

• 2. A few thousands (small cage) to tens of thousands of flies (large cage) should be added to an embryo collection cage, until they distribute individually on its walls, floor, and roof (see Note 7 ). The female to male ratio should be between 1 to 1 and 3 to 1. Flies should be no more than a few days old and need to be well fed for 1–2 days before maximum embryo production is reached. Maintain cages at 25 °C and at least 70 % humidity with controlled 12/12-h reverse light/dark cycles. Replace the food plates every 12 h, covering approximately one-third of the grape juice agar Petri dish with yeast paste.

• 3. When embryo production is maximal, begin collections in the morning. Replace the grape juice agar Petri dish with a fresh one, pre-warmed to room temperature for at least 30 min, containing yeast paste on roughly one-tenth of its surface, in the middle. Remove the Petri dish after 2 h and replace with a fresh food plate (see Note 8 ). After another 2 h, replace the food plates again, and collect embryos by washing the surface of the food plate with small amounts of distilled water and a paintbrush (see Note 9 ). In this process, resuspend the yeast paste on the surface of the food plate as it contains several embryos. Filter the suspension using a sieve and wash embryos thoroughly with distilled water to remove any remaining yeast. A sieve with larger pore size can be placed on top to collect the flies that were stuck to the plate, whilst letting the embryos through. Alternatively, flies can be removed with tweezers before filtration through the sieve.

• 4. Remove the chorion from embryos by soaking them in 50 % bleach for 2–5 min, with gentle agitation using a paint brush. Completion of dechorionation can be verified under a stereo microscope. Do not leave in bleach longer than necessary.

• 5. Thoroughly wash the embryos in distilled water followed by an EWS wash and place the sieve on a paper towel to remove the remaining liquid from the embryos.

• 6. Transfer the embryos to a 1.5-mL microfuge tube using a small, flat-ended spatula or a paintbrush with the tip dipped into distilled water.

• 7. Centrifuge at low speed for a few seconds to bring embryos to the bottom of the tube. Note the weight of the embryos and snap-freeze in liquid nitrogen. Dechorionated embryos can be stored at −80 °C for several months or years. Embryo collections can be repeated several times per day, for several days, until the desired amount of embryos is obtained or until embryo production has waned. Make sure to change the food plate at least once a day (ideally more) and change the cage itself when it becomes dirty.

3.2 Affinity Purification

3.2.1 Protein A Affinity Purification

This method has been designed with the ultimate goal of identifying co-purified proteins by mass spectrometry. The tag consists of two repeats of the immunoglobulin-binding domain of Protein A (PrA) from Staphylococcus aureus, and its purification takes advantage of the high binding affinity of PrA to rabbit IgGs. The principle is illustrated in Fig. 1. Because some proteins can stick nonspecifically to the affinity beads used or to the complex being purified, there is a need to compare the set of proteins detected in any given purification with that obtained with an unrelated protein. For this control, we have been using a strain of flies that express PrA-GFP in embryos (Maternal α-Tubulin-Gal4-VP16/CyO; UASp-PrA-GFP/TM6B). The bait protein of interest and the control bait should be purified in parallel, under identical conditions. Proceed as follows:

Fig. 1

In vivo proteomics and the rationale of affinity purification of protein complexes from Drosophila transgenic embryos. The flowchart demonstrates the major steps of the purification of tagged proteins along with their interactors, which are described in detail in Subheading 3. The inset illustrates how the bait protein, tagged either with PrA or GFP, binds to an affinity bead via direct interaction with IgG or GFP-binding domain, respectively. Proteins interacting with the bait protein co-purify on the affinity beads and are subsequently identified using mass spectrometry

• 1. Work at 4 °C throughout embryo lysis and purification (unless otherwise stated). Add 0.5 volume (relative to the weight of embryos) of EB to embryos while they are still frozen in 1.5-mL microfuge tubes. For example, for 1 g of embryos, add 0.5 mL of EB. The starting mass of embryos used may be adjusted depending on the bait protein and particular fly line (see Note 1 ).

• 2. Crush embryos thoroughly and for several seconds using prechilled plastic micropestles that fit a 1.5-mL microfuge tube. Alternatively, a Dounce tissue grinder can be used (see Note 10 ). Pool lysates at this point if they were initially contained in multiple tubes.

• 3. Add 4 volumes of EB (relative to the weight of embryos) after transferring to a larger tube and vortex for 10 s. If the protein is suspected to be associated with chromatin, add nuclease (see Note 11 ).

• 4. Pass lysates through a needle using a prechilled syringe. Repeat four times. However, this step may not be necessary if a Dounce homogenizer was used in step 2. The lysate can be checked under a cell culture microscope to ensure destruction of nuclei.

• 5. Centrifuge lysates at 4,000 × g for 20 min at 4 °C (see Note 12 ).

• 6. Transfer supernatant containing soluble proteins to a new tube, taking care to avoid the white, fatty layer, which is mainly composed of yolk, the nutritive droplets of the embryos. As it is sticky, it is important to remove as much yolk as possible to minimize the presence of nonspecific proteins in purification products (see Note 13 ).

• 7. A small fraction of the clarified supernatant can be taken for subsequent Western blot analysis (“soluble input”). The pellet can be resuspended in a volume of 1× Laemmli sample buffer equal to the volume of supernatant and a small fraction can be kept for Western Blot analysis (“pellet”).

• 8. Before mixing with the extract, the required amount of IgG-conjugated magnetic beads (approximately 200 μL of bead suspension (20 mg of dry beads) per 1 g of embryos) is washed briefly in EB in separate tubes prepared for each sample (see Note 14 ).

• 9. Vortex the tubes gently and place them on a magnetic rack, wait for beads to settle and aspirate off or carefully remove the EB.

• 10. Resuspend beads with the protein extract’s clarified supernatant and transfer to a new tube. Incubate on a rotating wheel at 4 °C from 30 min to 4 h (see Note 15 ).

• 11. Place tubes on the magnetic rack and wait until all beads have adhered to the side of the tube facing the magnet. A small fraction of the unbound supernatant can be taken for subsequent Western blot analysis (“unbound”). Discard the supernatant using an aspirator or pipette.

• 12. Add 1 volume of WB (relative to the original extract) and resuspend the beads by vortexing briefly. Place tubes on the rotating wheel and wash at 4 °C for 5 min. Place tubes on the magnetic rack and discard the WB. Repeat washes four times.

• 13. Perform two more washes with 1 volume of FWB (relative to the original extract) (see Note 6 ).

• 14. Transfer the suspension to new microfuge tubes attached on the magnetic rack, wait for beads to settle, and carefully remove all supernatant (see Note 16 ).

• 15. Add 0.5 mL freshly made Elution solution to each tube. Resuspend the beads by briefly vortexing and place the tubes on a rotating wheel for 5 min incubation at room temperature.

• 16. Insert the tubes in the magnetic rack and transfer supernatants (eluates) to new microfuge tubes.

• 17. Perform a second round of elution, pool the two eluates (total volume of 1 ml), place the tubes on the magnetic rack, and transfer eluates again to new tubes to remove any remaining magnetic particles.

• 18. Divide the combined eluates into two fractions of 100 μL (10 %) and 900 μL (90 %) and lyophilize both samples in a vacuum concentrator (see Note 17 ).

• 19. Keep the lyophilized 90 % fractions at −20 °C until sample preparation for mass spectrometric analysis.

• 20. Resuspend the precipitates from the 10 % fractions in 1× Laemmli sample buffer and boil for 5 min at 95 °C.

• 21. Visualize the eluted proteins by SDS-PAGE followed by silver-nitrate staining (Fig. 2), saving a small amount of the sample (“eluted from beads”) for Western blot analysis (see Note 18 ).

  Fig. 2

  

  Examples of protein complexes obtained by PrA affinity purifications from early embryos. A small fraction of the purification products were analyzed on a 4–12 % SDS polyacrylamide gel stained with silver nitrate. The bait proteins are indicated at the top. Right: specifically associated proteins identified by mass spectrometry are indicated at the level of their predicted molecular weight and under their respective bait protein. The main fraction of the purification products was analyzed without gel separation, and therefore we have no experimental positional information for the proteins labeled on the right. Purified bait proteins are marked by asterisks and indicated in bold font names in the panel on the right. IgG HC immunoglobulin G heavy chain, IgG LC immunoglobulin G light chain, MW molecular weight protein markers

• 22. If a purification product looks clean and reveals specific bands including a clear band at the expected molecular weight for the bait protein (as shown in Fig. 2), we generally proceed to gel-free mass spectrometric identification of purified proteins from the lyophilized precipitate obtained in step 18. This fraction can be processed and analyzed directly by most proteomics facilities.

3.2.2 GFP-Trap Affinity Purification

Over the past years, thousands of transgenic flies expressing proteins fused to a fluorophore, such as GFP and its variants, were established by different laboratories or consortia, in order to follow the localization and dynamics of the protein of interest by microscopy (fixed preparations or live imaging). Conveniently, the same fly lines can also be used for one-step affinity purifications against the GFP tag of the chimeric protein and its interacting factors from different ontogenetic stages or tissues for proteomic studies. Combined knowledge of localization and interactions is highly informative on a protein’s function. Here we describe a general protocol for the GFP-Trap-based purification of GFP-fusions from Drosophila syncytial embryos. The general principle is similar to the PrA affinity purification and is illustrated in Fig. 1. We recommend purifying the bait protein of interest and the control GFP bait (we made and used the yw; Bc Gla/CyO, P[w+; Ub-GFP] balancer stock that constitutively expresses GFP) in parallel, under identical conditions. The embryo lysis and soluble protein extract preparation is identical to the above procedure for PrA affinity purifications (see Subheading 3.2.1, steps 1–7). Follow these subsequent steps for the GFP-Trap purifications:

• 1. Pre-equilibrate GFP-Trap agarose beads: use 50–100 μL of the original GFP-Trap agarose suspension per sample (see Note 19 ) in 5 mL EB in 15-mL conical tubes.

• 2. Mix gently four times and sediment the beads by brief and slow centrifugation (e.g., for 3 min at 500 × g) with a slow deceleration speed in order to avoid turbulence.

• 3. Add the clarified supernatant of the embryo extract to the beads followed by incubation on a rotating wheel at 4 °C from 30 min to 2 h (see Note 15 ).

• 4. Sediment the beads and take a small fraction of the unbound supernatant for subsequent Western blot analysis. Carefully discard residual supernatant using an aspirator or pipette.

• 5. Wash the beads five times in 10 mL WB by gentle rotation at 4 °C for 5 min each time.

• 6. Sediment the beads and discard the supernatant.

• 7. Transfer the beads to a new 15-mL conical tube and wash two times in 10 mL FWB (see Note 6 ) by gentle rotation at 4 °C for 10 min.

• 8. Sediment the beads and discard the supernatant.

• 9. Transfer the beads to a microfuge tube and further wash two times for 2 min with FWB.

• 10. Sediment the beads and discard the supernatant.

• 11. Remove as much buffer as possible from the agarose beads; take 10 % of the beads for SDS-PAGE analysis using silver staining (load 9 % of the beads) and Western blotting (load 1 % of the beads). An example of a silver-stained gel of purification products is shown in Fig. 3.

  Fig. 3

  

  Examples of affinity purification of three bait proteins using GFP-Trap. GFP alone, GFP-Polo [16], and GFP-BubR1 [17] fusions were purified from transgenic Drosophila syncytial embryos using GFP-Trap agarose beads. Small fractions of the preparations (9 %) were run on a 4–12 % SDS polyacrylamide gel and stained with silver nitrate (left panel ), the remaining isolated material (90 %) was analyzed using mass spectrometry and examples of identified proteins are listed in the panel on the right. The main fraction of the purification products was analyzed without gel separation and therefore we have no experimental positional information for the proteins labeled on the right. Baits are marked by asterisks and indicated in bold font names in the panel on the right

• 12. Keep the rest of the beads (90 %) at 4 °C for subsequent on-beads tryptic digestion and mass spectrometric analysis (see Note 20 ).

4 Notes

• 1. The amount of embryos needed for affinity purification may be adjusted depending on the expression level of the bait protein. A method for very large-scale embryo production has been published previously in this series [14]. Overexpression of tagged proteins in transgenic embryos may be toxic or decrease fitness, leading to poor fertility and embryo production. In other cases, the expression of the bait protein may be extremely low, due to rapid degradation or transcriptional downregulation. Therefore, we recommend researchers to perform pilot experiments to observe the egg deposition rate in comparison to wild-type (e.g., Oregon-R) flies and to determine the overexpression level of the bait protein via Western blot. This approach should help evaluate the size of the population required to generate a sufficient amount of embryos. For protein identification by mass spectrometry, we use approximately 1 g of embryos per purification. It is recommended to collect embryos in a separate population cage room, fly incubator, or chamber. To obtain a maximum yield of embryos, it is critical to maintain constant environmental conditions of 25 °C, at least 70 % relative humidity and controlled reverse light/dark cycles.

• 2. Flies are attracted to the acidic and aromatic smell of fruit juice, and they like to feed on yeast paste. Grape juice agar food plates/trays can be stored at 4 °C for several days. Prior to use, pre-warm the plates/trays to room temperature for 30 min. After initial mixing of the yeast paste, gas bubbles will form within a few hours. It is important to mix again to eliminate these bubbles as they can prevent firm adhesion of the yeast paste to the agar medium. Yeast paste can be stored at 4 °C for several days. Heat the yeast paste for a few seconds in microwave oven (do not let it melt) prior to spreading on agar, as lysed yeast cells are preferred by the flies.

• 3. The most common method for the affinity purification of GFP (and its variants)-tagged proteins is co-immunoprecipitation using antibodies recognizing the tag. However, the large immunoglobulin complexes, as major contaminants, cause “peptide masking” during mass spectrometric analysis, which may lead to the loss of important information about potential interactors or post-synthetic modifications. The GFP-Trap system instead utilizes the small 13 kDa intact GFP-binding domain of the special heavy-chain antibody of Camelidae to enable the rapid and efficient one-step isolation of GFP-tagged baits expressed in Drosophila embryos for subsequent analysis by mass spectrometry [15]. GFP-Trap is extremely stable with a high binding affinity to GFP. We recommend the agarose beads version of GFP-Trap as we found that the use of the magnetic beads version—although more convenient during washes—increases the background of nonspecific interactions with bait proteins. For the complete list of recognized fluorophores (RFP-Trap is also available) visit the FAQ section at www.chromotek.com.

• 4. The composition of the extraction buffer (EB) depends on a particular protein or complex to be purified. Proteins that are part of large organelles or subcellular structures may be extracted with higher salt and/or detergent; however the concentration of mild detergents should not exceed 0.5 % during the binding step. If the detergent concentration of the EB is higher than 0.5 %, it should be diluted prior to mixing with the beads (binding). Similarly, too much salt can slow down or prevent efficient binding of the bait protein to the beads. Chemical inhibitors of specific enzymatic activities can also be added to the EB at the extraction step, depending on the proteins to be purified and on the subsequent analysis. If the protein of interest is rapidly degraded by the 26S proteasome, use 25 μM MG132 proteasome inhibitor in the EB and throughout the purification. If the bait protein is purified for mass spectrometry-based identification of potential posttranslational modifications (PTM) instead of for the identification of interactors, the stringency of the buffers can be increased (with more detergent or salt) and it is recommended to use specific phosphatase inhibitors, such as 50 nM okadaic acid (for PP2A-related phosphatases) or broader specificity phosphatase inhibitor cocktails (e.g., from Sigma or Roche) to preserve the phosphorylation state of proteins in the extract. For the inhibition of deubiquitylating enzymes (DUBs), use 10 mM of the sulfhydryl alkylating agent N-ethylmaleimide. Such inhibitors may also be used if the interaction between the bait protein and its partners is likely to be PTM-dependent.

• 5. Always use autoclaved or filter-sterilized ultra-pure water and stock solutions during the extraction and purification steps. Use properly cleaned and autoclaved microfuge and conical tubes, funnels, Dounce tissue grinder and micropestles, needles and syringes.

• 6. Final wash buffer (FWB) must not contain glycerol, protease inhibitors, or detergents, which may interfere with the trypsin digestion or other steps of sample preparation and analysis by liquid chromatography and mass spectrometry. If this problem persists, a protein precipitation step can be applied after elution (see Note 17 ).

• 7. Overcrowding of flies in the population cage will result in injury and stressing of the flies, leading to a reduced embryo yield.

• 8. The first collections of the day tend to yield fewer embryos than subsequent collections, and many of these embryos can be defective if they have been retained by the mothers for too long. Consequently, we recommend discarding the first plates.

• 9. During embryo collection, care should be taken not to stab the agar with the paintbrush, as agar flakes can clog the sieve used in later steps and are difficult to separate from the embryos. Around 100 mg of embryos is typically obtained from one good collection on a 100 mm-diameter Petri dish (significantly more in large cages from trays).

• 10. When using Dounce homogenizer, the initial lysis of the embryos can be done with a loose-fit pestle, and a finer homogenization step can be done with the tighter-fit pestle. Due to high backpressure, the initial strokes are usually difficult. Be careful to avoid “squirting up” of the homogenate; when there is resistance, release it by twisting the pestle left-and-right and continue pushing down gently.

• 11. If the bait and/or its interactors are likely to be chromatin associated, add 200 μg/mL (or 2,000 Kunitz units) final concentration of DNase I (Sigma DNEP dissolved in 150 mM NaCl, 4 mM MgCl₂ and pre-activated at 37 °C for 5 min, although other nucleases can also be used) to the lysate and incubate at 37 °C for 5 min, mix gently, and incubate for a further 10 min at room temperature prior to initial centrifugation. Alternatively the DNase I treatment may be performed at 4 °C for 1 h on a rotating wheel. DNA digestion can be monitored on an agarose gel after extraction of nucleic acids from lysates. For this, we found that DNazol (Life Technologies) gives a very good yield of DNA purification from extracts.

• 12. The duration and speed of centrifugations has to be optimized for different purifications. Large molecular complexes may be lost in the pellet after high-speed centrifugation.

• 13. To help remove the yolk, the homogenized lysate can be passed through a double layer of Miracloth (pre-equilibrated with EB and placed in a funnel). The supernatant can also be centrifuged once more to remove remaining yolk.

• 14. The ideal amount of beads used for purification depends mainly on the expression level of the bait protein to be purified.

• 15. A long incubation will favor a high yield of the bait protein while a short incubation will minimize the risk of disassembly of low affinity complexes.

• 16. It is important to transfer the samples to new tubes before the elution because some proteins from the extract stick on the walls of the tube and could be redissolved in presence of the elution solution, contaminating the purification product.

• 17. Alternatively, a protein precipitation step using methanol or acetone can be added to get rid of nonvolatile impurities that may remain in the eluted samples (even after lyophilization) and interfere with mass spectrometry. However, we find that it is usually not necessary.

• 18. The various small fractions kept during the procedure can be used for Western blot analysis to follow the path of the bait proteins purified during the preparation. We recommend loading equivalent fractions of the insoluble pellet, soluble input, unbound, eluates, and the bead fraction kept after elution (obtained by boiling the beads in 1× Laemmli sample buffer; load without bead particles). Use appropriate dilutions to ensure the loading of equal percentage of the total sample for every fraction. This analysis reveals if a protein has been efficiently expressed, extracted, bound, and eluted, and if it has been degraded or stayed intact (Fig. 4).

  Fig. 4

  

  Following the bait protein during the major steps of the affinity purification procedure. Polo-protein A fusion was purified from syncytial embryos using IgG-coupled magnetic Dynabeads. Small fractions were collected at different steps during the purification procedure. For each fraction, the same proportion of the total sample (0.1 %) was loaded on the gel, which was analyzed by Western blotting with HRP-conjugated rabbit IgG. This analysis reveals that Polo-protein A was incompletely extracted, efficiently bound to the beads, and efficiently eluted. No breakdown of the protein was detected

• 19. The binding efficiency of the GFP-Trap agarose beads is high. We found that 100 μL of original bead suspension is sufficient to bind 90–95 % of GFP alone prepared from 1.5 g syncytial embryos; however the binding to GFP-tagged proteins is less efficient.

• 20. Quantitative elution of the bait and its interactors from GFP-Trap is difficult due to the high affinity of the GFP-binding domain to GFP. However, we found that the elution of the purified proteins from beads is not necessary. The best results were obtained by direct tryptic digest of the material on beads followed by mass spectrometry. However, as an alternative, 0.3 M Glycine-HCl pH 2.5, 6 M Guanidine-HCl, or 1 M NH₄OH can be used for elution.