Methods for studying membrane-proximal GAP activity on prenylated Rab GTPase substrates

Summary/abstract:

Rab GTPases are key regulators of membrane trafficking. When GTP-bound, or ‘active,’ Rabs are anchored to membranes and recruit effector proteins that mediate vesicle formation, transport, and fusion. Rabs are inactivated by GTPase activating proteins (Rab-GAPs), which catalyze GTP hydrolysis, rendering Rabs cytosolic. In vivo, C-terminal prenylation modifications link activated Rabs to organelle and vesicle membranes, yet historically, in vitro Rab-GAP activity assays have been performed in the absence of membranes. We have developed a method for assaying Rab-GAP activity in a physiological context, with dissociation of the Rab from the membrane serving as a readout for Rab-GAP activity. Given that membrane binding status is a key consequence of Rab activation state, this assay will be useful for the study of a wide range of Rab/Rab-GAP pairs.

1. Introduction

Rab GTPases are key drivers of eukaryotic vesicle trafficking [1,2]. Rab function is dictated by nucleotide binding: when GTP-bound, or ‘active,’ Rabs are anchored to membranes via C-terminal prenylation modifications and recruit effector proteins that control the formation, transport, and receipt of vesicles. When a membrane transport pathway is complete, Rabs hydrolyze GTP, rendering them GDP-bound and ‘inactive.’ They are then extracted from the membrane by a GDP dissociation inhibitor (GDI), which chaperones the Rab until it is activated again. This nucleotide exchange-hydrolysis cycle is facilitated by guanine nucleotide exchange factors (GEFs), which exchange GDP for GTP, and GTPase activating proteins (GAPs), which catalyze GTP-to-GDP hydrolysis [3,4]. Because nucleotide binding controls Rab effector recruitment— and, thus, the initiation and termination of vesicle trafficking events—identifying and characterizing the specific GEFs and GAPs that control individual Rabs is key to fully understanding the trafficking pathways in which Rabs function.

There are several established biochemical assays for interrogating Rab-GEF activity in vitro, most of which employ fluorescent GDP analogs that can be detected spectroscopically as they dissociate from the activated Rab [5]. Measuring Rab-GAP activity, however, is less straightforward, as it is difficult to directly measure GTP hydrolysis without using radiolabeled GTP (e.g., [γ³²P]-GTP). Free phosphate reporters (e.g., malachite green, commercial phosphate sensors) that detect phosphate released by a Rab following GTP hydrolysis are a feasible alternative, but we have found that this approach is often not practical, as a) extensive steps must be taken to remove contaminating free phosphate from buffers, protein solutions, and reaction vessels, and b) high concentrations of Rab are required for reliable phosphate detection above background. Perhaps most importantly, most published Rab-GAP assay methods do not include key physiological considerations such as membranes and Rab prenylation, both of which are important contributors to Rab activation state [1,6-8].

We have developed a Rab-GAP activity assay that takes advantage of the relationship between Rab activation and membrane binding: if a Rab is inactivated by a Rab-GAP, it will no longer bind membranes if stoichiometric GDI is also present. The assay is rooted in standard liposome flotation techniques, in which liposomes and bound proteins are separated from free proteins via a discontinuous sucrose gradient. In this protocol, we describe how we use our assay to study Saccharomyces cerevisiae Gyp1, which serves as a Rab-GAP for Ypt1 (homolog of human Rab1) at the late Golgi/trans-Golgi Network (TGN) [9-11]. The assay is sensitive enough to allow researchers to not only reconstitute Rab-GAP reactions, but also to investigate how other factors, such as a lipids and additional proteins, influence Rab-GAP function.

Briefly, our method consists of 5 steps: 1) preparation of liposomes, 2) activation of Rab of interest (Ypt1) on liposomes via EDTA-mediated nucleotide exchange, 3) addition of Rab-GAP of interest (Gyp1) to catalyze GTP hydrolysis, 4) collection of liposomes and any associated proteins, and 5) visualization of reaction outcome via SDS-PAGE.

2. Materials

2.1. Liposome preparation

1. Equipment and consumables

   1. Glass syringes, 10-500 uL
   2. Screw-cap glass vials
   3. 25-mL glass pear-shaped flasks
   4. Rotary evaporator
   5. Parafilm or solid flask stopper
   6. 37°C incubator
   7. Liposome extrusion apparatus (Avanti Polar lipids cat. 610000) with 2 1-mL glass syringes (Avanti Polar Lipids cat. 610017)
   8. 19 mm 100 nm pore filters (Whatman cat. 800309)
   9. 10 mm filter supports (Whatman cat. 230300)
2. Reagents

   1. Lipid stocks (in chloroform)
   2. Fluorescent lipophilic dye (e.g., DiR, Invitrogen cat. D12731) (see Note 1)
   3. Chloroform
   4. Methanol
   5. HK buffer: 20 mM HEPES pH 7.4, 150 mM potassium acetate

2.2. Rab-GAP activity assay

1. Equipment

   1. 30°C incubator
   2. Optima TLX ultracentrifuge (Beckman Coulter)
   3. TLA100 fixed-angle rotor (Beckman Coulter)
   4. 7 x 20 mm open-top thickwall polycarbonate centrifuge tubes (Beckman Coulter cat. 343775)
   5. 55°C heating block
   6. BioRad ChemiDoc MP Imaging System (or similar imaging instrument)
2. Reagents

   1. Purified prenylated Ypt1-GDI complex (see Note 2) or other prenylated Rab-GDI complex of interest
   2. Purified Gyp1 (see Note 3) or other Rab-GAP of interest
   3. Purified myristoylated Arf1 (see Note 4)
   4. 100 nm TGN-mimetic liposomes, prepared in part 3.1 (see Note 5)
   5. 100 mM EDTA, pH 8.0
   6. 200 mM magnesium chloride
   7. 10 mM GTP
   8. 10 mM GMP-PNP, GTPγS, or GDP (see Note 6)
   9. HK buffer with added DTT: 20 mM HEPES pH 7.4, 150 mM potassium acetate, 1 mM DTT
   10. HKM buffer with added DTT: 20 mM HEPES pH 7.4, 150 mM potassium acetate, 2 mM magnesium chloride, 1 mM DTT
   11. 2.5 M sucrose (in HKM buffer)
   12. 0.75 M sucrose (in HKM buffer)
   13. 2x and 5x SDS-PAGE sample buffers with β-mercaptoethanol added fresh to 5% v/v (for 5x buffer: 300 mM Tris-Cl pH 6.8, 10% w/v SDS, 50% v/v glycerol, 0.05% w/v bromophenol blue; dilute in water to make 2x buffer)
   14. SDS-PAGE gels (see Note 7)
   15. Lipid-safe Coomassie stain (e.g., Bio-Rad Bio-Safe Coomassie, cat. 1610786 or Bio-Rad QC Colloidal Coomassie, cat. 161-0803)

3. Methods

3.1. Liposome preparation

This section details our protocol for liposome preparation. Briefly, a lipid film is first deposited onto the walls of a glass flask under rotary evaporation, then rehydrated in an aqueous buffer. The resulting multilamellar liposomes are then extruded through a porous polycarbonate filter, which yields a suspension of uniformly-sized unilamellar liposomes [13,14]. Table 1 lists the lipid composition we use when making TGN-mimetic [14] liposomes (see Note 5).

Table 1.

Composition of TGN-mimetic liposomes used in Gyp1/Ypt1 Rab-GAP assay. DOPC, 1,2-dioleoyl-sn-glycero-3-phosphocholine; POPC, 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine; DOPE, 1,2-dioleoyl-sn-glycero-3-phosphoethanolamine; POPE, 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoethanolamine; DOPS, 1,2-dioleoyl-sn-glycero-3-phospho-L-serine; POPS, 1-palmitoyl-2-oleoyl-sn-glycero-3-phospho-L-serine; DOPA; 1,2-dioleoyl-sn-glycero-3-phosphate; POPA, 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphate; PI, phosphatidylinositol; PI(4)P, phosphatidylinositol-4-phosphate; CDP-DAG, 1,2-dipalmitoyl-sn-glycero-3-cytidine diphosphate; PO-DAG, 1-palmitoyl-2-oleoyl-sn-glycerol; DO-DAG, 1,2-dioleoyl-sn-glycerol; DiR, 1,1′-dioctadecyl-3,3,3′,3′-tetramethylindotricarbocyanine iodide.

TGN liposome composition
Lipid	Mol. percent
DOPC	24
POPC	6
DOPE	7
POPE	3
DOPS	1
POPS	2
DOPA	1
POPA	2
PI	29
PI(4)P	1
CDP-DAG	2
PO-DAG	4
DO-DAG	2
Ceramide	5
Cholesterol	10
DiR	1

3.1.1. Day 1

1. Using glass syringes, combine lipids in a screw-cap glass vial at a 2 mM final concentration in 20:1 chloroform:methanol.

2. Using a glass syringe, dispense 500 μL of the lipid solution into a 25-mL pear-shaped flask and dry completely via rotary evaporation (slow-to-medium speed rotation) under vacuum. The lipids should dry in a thin, uniform film at the bottom of the flask (see Note 8).

3. Gently dispense 1 mL HK buffer into the flask, taking care to not disturb the lipid film. Cover the flask opening with a stopper or Parafilm and incubate, stationary, for 14-16 h at 37°C.

3.1.2. Day 2

• 4.Gently swirl the flask by hand to suspend the hydrated lipids in the HK buffer.
• 5.To form liposomes, extrude the lipids through a 100 nm pore size filter (see Note 9) in a liposome extrusion apparatus 25 times (see Note 10). Store liposomes (protected from light if using DiR) at 4°C for up to one month.

3.2. Rab-GAP activity assay

This section details our Rab-GAP assay protocol. First, a prenylated Rab is activated on liposomes via EDTA-mediated nucleotide exchange. Next, the Rab-GAP of interest is added and the reaction is incubated to allow for GAP-catalyzed GTP hydrolysis. Following this incubation, liposomes and bound proteins are collected via flotation through a sucrose density gradient. Inactivated, GDP-bound Rabs will dissociate from the liposomes and will not be present in the recovered lipid fraction, while active, GTP-bound Rabs remain liposome-bound. (Figure 1)

Figure 1. Schematic depicting the workflow for the Rab-GAP activity assay.

3.2.1, Activation of prenylated Rab on liposomes by EDTA-mediated nucleotide exchange. Upon activation, the GTP-bound Rab becomes anchored to the liposome membranes. 3.2.2, Addition of Rab-GAP followed by a 5 minute incubation to allow for GAP-catalyzed GTP hydrolysis. Inactive, GDP-bound Rab is extracted from the liposomes by the GDI. 3.2.3, Preparation of sucrose gradient in which the reaction input is loaded into the densest sucrose layer (bottom) and then collected from the least dense (top) layer following ultracentrifugation. 3.2.4, Visualization of reaction outcome by SDS-PAGE (see Note 19).

In cells, Gyp1 is recruited to the late Golgi/TGN by the GTPase Arf1 [11], so we have included Arf1 in this protocol. GAPs may be recruited to membranes by other proteins (e.g., S. cerevisiae Msb3, a Rab-GAP for the Rab5 homolog Vps21, is recruited by the BLOC-1 complex [16,17]) or by membrane features such as curvature or lipid packing (e.g., human ArfGAP1 [18,19]). As such, this protocol should be adapted to the specific conditions the Rab-GAP of interest requires.

3.2.1. Nucleotide exchange

1. Combine the following reagents in the order listed:

   • 20 uL 1 mM liposomes
   • 1 ug prenylated Ypt1-GDI (or Rab of interest) (see Note 11) (see Note 12)
   • 5 ug myristoylated Arf1 (for Gyp1-containing reactions) (see Note 2) (see Note 11)
   • 2 uL 10 mM GTP, GDP, or non-hydrolyzable GTP analog (see Note 6)
   • 3 uL 100 mM EDTA
   • HK buffer to 77 uL, minus volume of Rab-GAP to be added in step 3.2.2.

2. Incubate at 30°C for 30 minutes to allow for EDTA-mediated nucleotide exchange.

3. Add 3 uL 200 mM magnesium chloride to stop nucleotide exchange. Gently pipette or flick the tube to mix.

4. Incubate the reaction at room temperature for 1 minute.

3.2.2. Rab-GAP recruitment

1. Add 3 ng Gyp1 (or Rab-GAP of interest) (see Note 11) (see Note 12). Gently pipette or flick the tube to mix.

2. Incubate the reaction at room temperature for 5 minutes.

3.2.3. Liposome flotation and recovery

1. Using a cutoff pipet tip, slowly add 50 uL 2.5 M sucrose to each reaction and gently mix. Avoid creating bubbles (see Note 13).

2. Slowly transfer 100 uL of the reaction-sucrose mixture to a 7 x 20 mm open-top thickwall polycarbonate tube. Avoid creating bubbles (see Note 13). Save 15-20 uL of the remaining reaction-sucrose mixture as a reaction input for SDS-PAGE (Figure 1, Figure 2).

3. Slowly, gently dispense a 100-uL layer of 0.75 M sucrose atop the previous layer, taking care to not mix the layers or create bubbles (see Note 13) (see Note 14). There should be a faint demarcation between the two layers.

4. Slowly, gently dispense a 20-uL layer of HKM buffer atop the 0.75 M sucrose layer, taking care to not mix the layers or create bubbles (see Note 14). There should be a clear demarcation between the two layers.

5. Load tubes into a TLA100 rotor and centrifuge at 100,000 x g at 20°C for 20 minutes. Set the instrument to accelerate and decelerate as slowly as the instrument permits.

6. Carefully collect 30 uL from the top of the sucrose gradient (see Note 15) (see Note 16).

Figure 2. Representative result of a Gyp1/Ypt1 Rab-GAP assay.

“Input” reaction samples were taken before liposome flotation, and “Membrane-bound” samples were recovered following ultracentrifugation/liposome flotation. The addition of Gyp1/Arf1 catalyzes Ypt1 GTP hydrolysis, displacing Ypt1 from membranes. Membrane extraction is dependent on GTP hydrolysis, as Ypt1 activated with the non-hydrolyzable analog GMP-PNP remains membrane associated under the same reaction conditions. Gyp1 with a C-terminal TAP tag was detected via anti-TAP western blot; Ypt1 and Arf1 were detected via Coomassie stain. Lipids were visualized by DiR dye. * = Gyp1 degradation product. Figure adapted from ⁹.

3.2.4. Visualization of reaction outcome

1. Add 2x SDS sample buffer to reaction inputs. Add 5x SDS sample buffer to recovered lipids. Briefly heat samples in a 55°C heat block prior to gel loading (see Note 17).

2. Run input and recovered samples on an SDS-PAGE gel (see Note 7). Do not run the dye front off the gel, as the lipids run roughly with the dye front (see Note 18).

3. Visualize lipids in the gel by lipid tracer dye (recommended) or by Coomassie stain (see Note 1) (see Note 18) (see Note 19). Visualize proteins by Coomassie stain and/or western blot. (Figure 2) If quantification of technical replicates is desired, digital densitometry (e.g., ImageJ intensity measurement tools) can be used to quantify protein and lipid band intensities (see Note 20).

4. NOTES

1. We use DiR, a near-infrared fluorescent lipophilic dye, for liposome visualization. In addition to being visible by eye (and making liposomes easier to visually locate when working with them), it strongly fluoresces at λ = 700-800 nm, which is useful for visualizing lipids in SDS-PAGE gels. (Similar dyes that fluoresce at shorter wavelengths are commercially available.) Single-step low-methanol/-ethanol Coomassie stains can also detect lipids in SDS-PAGE gels, but they are not nearly as sensitive.

2. See [8] for Ypt1-GDI complex purification protocol.

3. See [11] for Gyp1 purification protocol.

4. At concentrations < 2 nM, Gyp1 must bind Arf1 to be robustly recruited to membranes, so we include it in our Gyp1 Rab-GAP assays. Arf1 is activated alongside Ypt1 via EDTA-mediated nucleotide exchange (step 3.2.1). For Arf1 purification protocol, see [19]. Adjustments to this protocol should be made to accommodate or test different Rab-GAP recruitment mechanisms.

5. Gyp1 functions at the late Golgi/TGN, and a TGN-like lipid composition [14] is critical for recruitment of Gyp1 to liposomes [11]. We therefore use TGN-mimetic liposomes [14] (Table 1) when assessing Gyp1 function in vitro. In cells, membrane lipid composition varies significantly among organelles [20, 21] and often contributes to protein localization and function [22], so we recommend that researchers tailor this liposome recipe according to the specific intracellular localization of the Rab and Rab-GAP of interest.

6. It is important to include a negative control for GTP hydrolysis, as Rabs can hydrolyze GTP independently of Rab-GAP assistance (albeit extremely slowly [23]). One option is to use a catalytically inactive Rab-GAP mutant, but this requires purification of an additional protein. Instead, we prefer to use a non-hydrolyzable GTP analog (e.g., GMP-PNP or GTPγS) for hydrolysis controls. We also recommend including a control for background membrane binding in which the Rab substrate is exchanged with GDP. The ‘inactive,’ GDP-bound Rab should not associate with liposomes.

7. We recommend using SDS-PAGE gels with ≥12% acrylamide to ensure good retention of lipids within gels. Lipids generally run within or just above the dye front.

8. A 30°C water bath can be used to hasten solvent evaporation but is not essential.

9. We usually use a 100-nm pore-size when extruding liposomes for flotation-based assays because extrusion through these small pores reduces the likelihood of persistent multilamellar structures [13].

10. It is important to collect liposomes after an odd number of passes through the extrusion apparatus, as the starting side of the filter will collect any unwanted debris. We typically pass the lipids through the extrusion apparatus 19-25 times.

11. We typically work from purified protein stocks of the following concentrations: 2-5 uM prenylated Rab-GDI complex, 50-200 nM Rab-GAP, and, when appropriate, 50-100 uM myristoylated Arf1. Following purification, all proteins are aliquoted, snap frozen in liquid nitrogen, and stored at −80°C. Protein aliquots can be frozen and thawed at least one time without significant loss of activity.

12. These reagent ratios may not be appropriate for all Rabs/Rab-GAPs and should be optimized as needed. It is best to use only the minimum amount of Rab-GAP necessary for Rab inactivation, especially when aiming to discern mechanisms of Rab-GAP recruitment or activity regulation. We recommend performing an initial titration of purified Rab-GAP to determine the minimum concentration required for robust Rab activation within the 5-minute reaction time.

13. The sucrose solutions are viscous. Work slowly with a cutoff pipet tip (~2 mm snipped off the working end) to avoid bubbles and inaccurate volume transfer. The thickwall polycarbonate tubes may be held in empty 1.7 mL Eppendorf tubes or an appropriately sized PCR tube rack.

14. Core engagement and steady breathing can help stabilize the hand. Researchers should take their time when preparing and collecting from the sucrose gradient.

15. When recovering lipids, we collect 2x 15-uL fractions from the top layer of the sucrose gradient. We find that the brief respite between the two fractions helps us maintain a steady hand.

16. Sometimes, floated lipids can form a tight or stringy-looking pellet on the wall of the tube. This pellet often forms on the inside wall of the tube, so it may be helpful to use a marker to mark the orientation of each tube prior to centrifugation in order to locate and collect lipid pellets. If floated lipids are not visible, they are dispersed in the top layer of the sucrose gradient. Therefore, simply collect 2x 15-uL samples from the top of the sucrose gradient (see Note 18).

17. Addition of SDS sample buffer can cause reactions containing HK buffer to precipitate out of solution. Brief incubation in a 55°C heat block can be used to re-solubilize samples just prior to gel loading.

18. In-gel lipid visualization provides a quantitative measure of liposome recovery. Liposome recovery is important to consider when interpreting the results of this assay, as inconsistent lipid recovery can confound comparisons between experimental and control reactions.

19. If using a lipophilic dye to visualize lipids, take an image of the lipids before Coomassie staining to ensure strong lipid signal.

20. If a desired, densitometry can be used to quantify protein band and lipid intensities for technical replicates. We typically normalize protein band intensity to lipid signal, i.e., $\frac{proteinbandintensity}{lipidbandintensity}$. However, if there is significant variation in lipid band intensity between samples, it is not appropriate to normalize or draw any conclusions from the experiment, and instead the experiment should be repeated. Comparisons should only be made when roughly equal amounts of lipids are recovered from each sample.