Key words

1 Introduction

Twenty years after the identification of nuclear export sequence s (NESs) in the HIV-1 Rev protein [1] and in PKI, the heat stable inhibitor of the catalytic subunit of cAMP-dependent protein kinase [2], the number of proteins carrying characteristic hydrophobic NESs is still growing. CRM1, a member of the importin β-family of nuclear transport receptor s, has been recognized as the major factor that mediates transport of proteins out of the nucleus, although other family members are involved in export of selected cargoes as well (for reviews see refs. 3–5). More recently, CRM1 was identified as a potential target in anti-cancer therapy (for review see ref. 6), based on the observation that it is upregulated in many tumor cells. This probably relates to the fact that many tumor suppressors like p53 [7] or p27 [8] are bona fide CRM1 cargoes.

We initially designed an assay to analyze nuclear export in digitonin -permeabilized cells that was based on similar principles as the classic nuclear import assay developed by Adam and Gerace [9]. The assay used GFP -NFAT (nuclear factor of activated T cells) as a CRM1-dependent reporter protein [10]. In vivo, nuclear trafficking of GFP-NFAT can easily be controlled by adding the calcium ionophore ionomycin to cells, a treatment that leads to rapid nuclear import of the protein [11, 12]. Nuclear export , on the other hand, occurs upon medium exchange and can be recapitulated in digitonin-permeabilized cells in vitro. After such an export reaction in permeabilized suspension cells, residual nuclear fluorescence can be quantified by flow cytometry , allowing the analysis of a large number of samples in a short period of time. Again, the method for quantification was based on a previous approach [13]. Although this assay proved very powerful for the identification of CRM1 [10], as well as for the characterization of co-factors like Ran [14], RanBP1 [15], RanGAP [16], and Nup214 [17], it has not widely been used by other researchers, because it requires a cell line stably transfected with the GFP-NFAT pla smid. We therefore set out to develop a similar assay that now uses transiently transfected cells expressing GFP-NFAT or GFP-snurportin1 (GFP-SPN1) as CRM1-dependent export cargoes and mCherry-emerin as a marker that allows gating for transfected cells in the subsequent flow cytometry analysis. Other combinations of cargoes and markers are possible. Nuclear accumulation of the export cargoes is achieved by treating the cells with the selective CRM1-inhibitor leptomycin B (LMB). For GFP-NFAT, ionomycin is used as an additional import-inducing agent (see Note 1 ). The marker protein should remain associated with the cells upon digitonin permeabilization . Emerin is a protein that localizes to the inner nuclear membrane . In our transfection assays, it is also found in stable association with other membranes, e.g., the endoplasmic reticulum . Alternatively, other membrane proteins can be used that should, of course, not interfere with nucleocytoplasmic transport (see Note 2 ).

Using the same principles described here, it should be possible to monitor nuclear export in vitro of virtually any CRM1-dependent export cargo (see Note 3 ). Several aspects of the methods are similar to those described in a previous chapter in Methods in Molecular Biology [18], where we presented the method using cells that stably express GFP -NFAT .

2 Materials

2.1 Plasmids and Cells

  1. 1.

    NFAT reporter: for the specific construct we use, the coding sequence of human NFAT was amplified from pSH107c (kindly provided by G.R. Crabtree, Stanford, CA) and cloned via HindIII and BamHI into a pS65T-C1 (Clontech)-based eukaryotic expression vector, behind the coding sequence of a modified version of GFP (S65T; [10]).

  2. 2.

    SPN1 reporter: for the specific construct we use, the coding sequence of human SPN1 was amplified from GST-SPN1 [19] and cloned via XhoI and EcoRI into the pEGFP-C1 vector [20].

  3. 3.

    Emerin reporter: for the specific construct we use, the coding sequence of human emerin was amplified from emerin-GFP (kindly provided by Eric Schirmer, Edinburgh) and cloned via XhoI and BamHI into the pmCherry-C1 vector (Clontech), yielding mCherry-emerin.

  4. 4.

    Ran expression pla smid: we use human wild-type Ran in pET11d.

  5. 5.

    CRM1 expression pla smid: we use a His-tagged human CRM1 in the pQE60 vector (kindly provided by I. Mattaj, EMBL, Heidelberg, Germany).

  6. 6.

    Cells for transport assay: HeLa p4 cells (P4.R5 MAGI) [21] were obtained through the NIH AIDS Reagent Program (Division of AIDS, NIAID, NIH, Dr. Nathaniel Landau; see Note 4 ).

  7. 7.

    Growth medium for assay cells: Dulbecco’s Modified Eagle’s medium (DMEM), containing 10 % fetal bovine serum (FBS), penicillin (100 U/mL), and streptomycin (100 μg/mL).

  8. 8.

    Cells for cytosol preparation: HeLa suspension cells (CSH HeLa strain, obtained from Dr. Frauke Melchior, Heidelberg).

  9. 9.

    Growth medium for cytosol preparation: Joklik’s modified S-MEM containing 10 % fetal bovine serum (FBS), 100 U/mL and 100 μg/mL penicillin streptomycin.

2.2 Buffers for Assay and Preparation of Cytosol

  1. 1.

    Lysis buffer: 5 mM HEPES-KOH, pH 7.3, 10 mM KOAc, 2 mM Mg(OAc)2, 2 mM dithiothreitol (DTT), 0.1 mM phenylmethyl sulfonyl fluoride (PMSF), 1 μg/mL of each leupeptin, pepstatin, and aprotinin.

  2. 2.

    Transport buffer: 20 mM HEPES-KOH, pH 7.3, 110 mM KOAc, 2 mM Mg(OAc)2, 1 mM EGTA, DTT, and protease inhibitors as above (see Note 5 ).

2.3 Buffers and Reagents for Preparation of Recombinant Transport Factors

  1. 1.

    Transport buffer: as above.

  2. 2.

    Ran buffer: 50 mM HEPES-NaOH, pH 8.0, 500 mM NaCl, 5 mM MgCl2, 0.1 mM PMSF, and 1 μg/mL of each leupeptin, pepstatin, and aprotinin.

  3. 3.

    CRM1 buffer: 50 mM HEPES-NaOH, pH 7.4, 500 mM NaCl, 2 mM MgCl2, 20 mM β-mercaptoethanol, 0.1 mM PMSF, and 1 μg/mL of each leupeptin, pepstatin, and aprotinin.

    CRM1 low salt buffer: 50 mM HEPES-NaOH, pH 7.4, 50 mM NaCl, 2 mM Mg(OAc)2, 2 mM DTT.

  4. 4.

    CRM1 high salt buffer: 50 mM HEPES-NaOH, pH 7.4, 500 mM NaCl, 2 mM Mg(OAc)2, 2 mM DTT.

  5. 5.

    PBS: 137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, 1.8 mM KH2PO4.

  6. 6.

    Imidazole 5 M stock.

  7. 7.

    Standard bacterial culture reagents.

  8. 8.

    DEAE Sepharose.

  9. 9.

    Preparative S75 column (see GE Healthcare for specifications to match if another is used).

  10. 10.

    PD-10 Desalting Columns or a HiPrep 26/10 Desalting (Sephadex G-25) type column (e.g., GE Healthcare, though any similar desalting column will do).

  11. 11.

    MonoQ anion exchange column (see GE Healthcare for specifications to match if another is used).

  12. 12.

    Isopropylthio-β-∆-galactoside (IPTG) 1 M stock.

  13. 13.

    Ammonium sulfate.

  14. 14.

    GDP 100 mM stock.

2.4 Additional Reagents for Transport Assays

  1. 1.

    ATP-regenerating system: (A) 100 mM ATP in 20 mM HEPES, 100 mM Mg(OAc)2 (see Note 6 ). Adjust pH to 7.4 with NaOH. (B) 80 mg/mL creatine phosphate in H2O. (C) 2000 U/mL creatine phosphokinase in 50 % glycerol, 20 mM HEPES, pH 7.4. Store all components at −20 °C and mix 2A:2B:1C (ATP:creatine phosphate:creatine phosphokinase) before use.

  2. 2.

    Leptomycin B (LMB; Enzo Life Sciences): prepare 370 μM stock (for preincubation of permeabilized cells) and 10 μM working dilution in ethanol (for treatment of intact cells); store at −20 °C.

  3. 3.

    Ionomycin: prepare stock (1 mM in DMSO) and store at −80 °C.

  4. 4.

    Digitonin : prepare stock (1 % in DMSO) and store at −20 °C.

  5. 5.

    Trypan blue: use 1:1 with cell suspension.

  6. 6.

    Oligonucleotide (see Note 7 ): dissolve desalted oligonucleotides (5′AGAG GAAAATTTGTTTCATA and 5′TATGAAACAAATTTTCCTCT), each at 200 μM, in 40 mM Tris–HCl, pH 7.4, 20 mM MgCl2, 50 mM NaCl, combine equal volumes and anneal by heating to 65 °C for 5 min and slow cooling to room temperature. Freeze in aliquots and store at −20 °C.

  7. 7.

    Wheat germ agglutinin (WGA): WGA, which inhibits most nucleocytoplasmic transport pathways, is dissolved in transport buffer at 2 mg/mL and stored at −80 °C.

3 Methods

3.1 Preparation of Cytosol

  1. 1.

    Grow HeLa suspension cells in spinner flasks to mid-log phase and collect them by centrifugation at 300 × g for 15 min (3–10 L can easily be handled). All of the following steps are performed on ice or at 4 °C.

  2. 2.

    Wash twice with PBS and once with transport buffer.

  3. 3.

    Resuspend in 1 volume of lysis buffer and swell for 10 min on ice.

  4. 4.

    Lyse the cells in a dounce homogenizer (see Note 8 ). After centrifugation at 1500 × g for 15 min, the supernatant is cleared by centrifugation at 120,000 × g for 1 h.

  5. 5.

    Dialyze overnight against transport buffer (10,000 K d cutoff), freeze the resulting cytosol (~10 mg/mL) in aliquots in liquid nitrogen, and store at −80 °C.

3.2 Preparation of Recombinant Transport Factors

3.2.1 Ran

Various methods for the preparation of Ran have been described in the literature. We typically use a modified version of the one described by Melchior et al. [22]. Human wild-type Ran is expressed from pET11d in BL21-(DE3) cells.

  1. 1.

    Grow 2 L of a bacterial culture to an OD600 of 0.6 and induce expression with 0.5 mM isopropylthio-β-∆-galactoside (IPTG) for 3 h at 37 °C. All subsequent steps are performed on ice or at 4 °C.

  2. 2.

    Harvest cells and resuspend them in 100 mL of Ran buffer, and lyse them. We find that the use of an EmulsiFlex-C3 (Avestin) for lysing produces reproducibly good protein, but other lysing methods should be able to be applied.

  3. 3.

    Clear lysate by centrifugation at 100,000 × g for 30 min.

  4. 4.

    Incubate the supernatant in batch with 70 mL DEAE Sepharose, equilibrated in Ran buffer, for 1 h with gentle agitation. Pour the slurry into a column and collect the flowthrough and approximately 100 mL of the subsequent wash (Ran buffer; wash until no protein can be detected).

  5. 5.

    Precipitate Ran from the DEAE flowthrough with ammonium sulfate: add salt gradually to 55 % saturation, stir for 2 h, and centrifuge for 20 min at 100,000 × g. Resuspend precipitate in 6–8 mL transport buffer containing 250 μM GDP, and clarify by centrifugation at 14,000 × g for 15 min.

  6. 6.

    Load onto a preparative S75 column, equilibrated in transport buffer. Collect 5-mL fractions and analyze them by SDS-PAGE .

  7. 7.

    Pool Ran -containing fractions, freeze in aliquots in liquid nitrogen, and store at −80 °C.

The final concentration of Ran (ca. 95 % pure) should be 2–3 mg/mL.

3.2.2 CRM1

The preparation of His-tagged CRM1 from pQE60-CRM1 (kindly provided by I. Mattaj, EMBL, Heidelberg, Germany) was originally described in [23].

  1. 1.

    Express the protein at 37 °C overnight in TG1 cells without induction by IPTG.

  2. 2.

    Harvest bacteria, wash them in PBS, and store cell pellets at −80 °C until purification.

  3. 3.

    Thaw and resuspend bacteria in cold CRM1 buffer, and lyse them. As for Ran we recommend an EmulsiFlex-C3 (Avestin). All following steps are performed on ice or at 4 °C.

  4. 4.

    Clear the lysate by centrifugation at 100,000 × g for 30 min.

  5. 5.

    Incubate the supernatant with Ni-NTA sepharose equilibrated in CRM1 buffer at 4 °C for 1 h.

  6. 6.

    Wash beads four times with 50 mL CRM1 buffer, each supplemented with 30 mM imidazole.

  7. 7.

    Transfer beads to an empty column and elute with CRM1 buffer containing 300 mM imidazole.

  8. 8.

    Change to CRM1 low salt buffer using a desalting column attached to an FPLC system.

  9. 9.

    Load the protein onto a MonoQ anion exchanger column equilibrated in CRM1 low salt buffer and elute with an increasing concentration of CRM1 high salt buffer.

  10. 10.

    Pool the peak fractions containing CRM1, change into transport buffer with desalting columns, and freeze in aliquots in liquid nitrogen. Store at −80 °C.

3.3 Transfections and Treatment of Cells

  1. 1.

    Grow HeLa p4 cells in DMEM medium in 24-well dishes.

  2. 2.

    Transfect cells with Lipofectamine (Life Technologies) or the calcium phosphate method [24], using 0.8 μg pla smid per ~105 cells per 24-well, with or without coverslips (see Note 9 ).

  3. 3.

    Incubate overnight and induce nuclear import of the reporter protein with 5 nM LMB for 90 min. For GFP -NFAT , also add 1 μM ionomycin for the last 30 min. The localization of GFP-NFAT and mCherry-emerin before and after treatment of cells with ionomycin/LMB is shown in Fig. 1. Continue as described in Subheading 3.4 or 3.5.

    Fig. 1
    figure 1

    Subcellular localization of GFP -NFAT and mCherry-emerin . HeLa p4 cells were transfected with plasmids coding for GFP-NFAT (green) and mCherry-emerin (red) and analyzed by confocal microscopy . GFP-NFAT localized to the cytoplasm or was homogenously distributed in the cells. Nuclear import of GFP-NFAT was induced by LMB and ionomycin. The nuclear localization of GFP-NFAT did not change upon permeabilization of the cells with digitonin . Note that the membrane association of mCherry-emerin did not vary in any of the tested conditions, confirming its suitability as a marker protein

    Full size image

3.4 Analysis of Transport by Fluorescence Microscopy

Although the major advantage of our approach is the quantitative analysis by flow cytometry (see Subheading 3.5), we briefly outline the method for microscopic analysis as well.

  1. 1.

    Wash cells grown on coverslips with cold transport buffer and permeabilize with digitonin (final concentration ~70 μg/mL in transport buffer) on ice (see Fig. 1). The required amount of the detergent should be tested on a separate slide for each cell type, using trypan blue (see Note 8 ).

  2. 2.

    Subject cells to preincubation and transport reaction, using reagents as outlined in Subheading 3.5, steps 3 and 4. Use ~40 μL solution per coverslip and incubate at 30 °C in a moisturized chamber.

  3. 3.

    Wash with cold transport buffer. Fix cells with 3.7 % formaldehyde in PBS.

  4. 4.

    Analyze cells by fluorescence microscopy and quantify images using local software.

3.5 Analysis of Transport by Flow Cytometry

  1. 1.

    Trypsinize cells: wash with PBS, add trypsin. Remove most of the trypsin solution immediately to minimize carryover of the protease. Incubate at 37 °C until cells detach. Collect cells in cold transport buffer containing 10 % fetal bovine serum (FBS) to inactivate the trypsin. The following steps are performed on ice or at 4 °C unless otherwise indicated. Centrifuge for 5 min at 300 × g and wash once in transport buffer.

  2. 2.

    Resuspend cells in transport buffer at 107/mL. Add digitonin to 70 μg/mL (0.7 μL of a 1 % stock per 106 cells). Leave on ice for 3 min and check permeabilization with trypan blue, using 5 μL of cells. Add more digitonin if less than 90–95 % of the nuclei appear blue. Wash with transport buffer and centrifuge as in step 1.

  3. 3.

    Preincubation: Resuspend cells at 107/mL in transport buffer and add 100 nM LMB and 25 μL ATP-regenerating system per mL of cell suspension. Incubate for 15 min in a 30 °C water bath. Wash cells twice with transport buffer. The preincubation results in depletion of transport factors like CRM1 that initially remain associated with the permeabilized cells. Furthermore, endogenous CRM1 will be modified by LMB. Without the preincubation, Ran will be the only rate-limiting factor for CRM1-dependent export [10].

  4. 4.

    Resuspend cells in transport buffer at 2 × 107/mL. The transport reaction (40 μL in a 5-mL flow cytometry tube) should contain: 10 μL of cells (200,000), 1 μL ATP-regenerating system, 0.5 μL annealed oligonucleotide for cells transfected with GFP -NFAT (final concentration: 1.25 μM). Add cytosol (~1 mg/mL usually yields optimal transport) and/or recombinant transport factors.

  5. 5.

    Incubate on ice or in a 30 °C water bath for 30 min.

  6. 6.

    Add 4 mL cold transport buffer. Centrifuge for 10 min at 300 × g and 4 °C. Remove most of supernatant by aspiration, leaving ~200 μL behind.

  7. 7.

    Analyze export of the GFP -cargo by flow cytometry . We use a FACSCanto II Cell Analyzer (Becton Dickinson), and typically count 5000–10,000 cells. For measuring GFP-fluorescence, gate the cells such that only mCherry-positive ones are analyzed. In most cases, the median fluorescence is best suited for statistical analysis of transport. Export reactions may be standardized by assigning a GFP-cargo fluorescence value of 100 to a 4 °C control (see Note 10 ).

Typical results for nuclear transport reactions as analyzed by flow cytometry are shown in Figs. 2 and 3. With respect to the efficiency of nuclear export of GFP -NFAT , they are very similar to those obtained using stably transfected cells [10, 18]. Cytosol, which should contain all relevant transport factors, strongly promoted nuclear export (i.e., resulted in low levels of fluorescence). CRM1 and Ran were only active when added together, indicating that both factors are rate-limiting under our experimental conditions. The physiological significance of the results was validated using several control conditions. First, we included wheat germ agglutinin (WGA) in our reaction. WGA is a lectin that binds to various O-glycosylated nucleoporins [25] and is known to inhibit the majority of nuclear transport pathways [26, 27]. Here, it clearly inhibited CRM1-dependent export of GFP-NFAT and GFP-SPN1. Next, reactions were performed in the absence of an ATP-regenerating system (−ATP) or at 4 °C. Under both conditions, the cells retained a high level of fluorescence. Taken together, these controls validate our in vitro assay as a transport system that faithfully reconstitutes nuclear export in transiently transfected, permeabilized cells.

Fig. 2
figure 2

Analysis of nuclear export by flow cytometry . HeLa p4 cells were transfected with plasmids coding for GFP -NFAT and/or mCherry-emerin . (a) Cell populations negative or positive for mCherry-emerin could be clearly distinguished by analyzing non-transfected cells and cells transfected with mCherry-emerin, alone or in combination with GFP-NFAT. A gate (red area ) was set in a way that only mCherry-emerin-positive cells were included in further analysis. FSC forward scatter, AU arbitrary units. (b) Comparison of GFP-NFAT fluorescence in the total and the gated population . A prominent peak for non-transfected cells can be observed on the left side of the fluorescence intensity profile in the total population (left panel ). Gating for the transfection marker mCherry-emerin (as described above) led to a simplified fluorescence intensity profile of GFP-NFAT, as most non-transfected cells were excluded from the analysis (right panel ). The fluorescence intensity of nuclei was normalized to a 4 °C control sample (indicated by the dashed line ) and was clearly reduced upon export of GFP-NFAT. Note the small remaining peak with a high GFP-NFAT fluorescence in the export reaction, which could result from cells that had not been sufficiently permeabilized and were therefore not accessible for export factors

Full size image
Fig. 3
figure 3

CRM1 and Ran promote nuclear export of GFP -NFAT and GFP-SPN1. HeLa p4 cells were transfected with plasmids coding for GFP-NFAT or GFP-SPN1 and mCherry-emerin and subjected to nuclear export reactions in vitro. Nuclear fluorescence of GFP-NFAT (a) and GFP-SPN1 (b) was analyzed by flow cytometry . Total populations were gated for mCherry-emerin-positive cells and the median fluorescence intensities of samples incubated at 4 °C were set as 100 %. Reactions contained 0.7 mg/mL cytosol, 2 μM Ran, 300 nM CRM1, and/or 150 μg/mL WGA, as indicated. The average and standard deviation of two independent experiments are shown

Full size image

4 Notes

  1. 1.

    For GFP -NFAT , ionomycin alone is sufficient to induce nuclear import . If LMB is used, cells must be washed thoroughly to prevent the drug from modifying CRM1 that is added to the reaction.

  2. 2.

    In our experiments, we did not observe an obvious influence of mCherry-emerin on CRM1-dependent export. New combinations of cargoes and transfection markers should be analyzed by fluorescence microscopy to verify efficient nuclear import of the cargo and unaltered localization of the marker upon LMB treatment as well as their unchanged localization upon permeabilization of the plasma membrane by digitonin .

  3. 3.

    Depending on the size of the CRM1-cargo, the protein might leak out of the nucleus in digitonin -permeabilized cells. This is not a problem for GFP -NFAT , which is >100 kDa in size. Smaller proteins can be fused to larger tags (e.g., GST or double-GFP) to minimize passive diffusion.

  4. 4.

    We typically use HeLa p4, because the transfection rate in these cells is very high.

  5. 5.

    EGTA inhibits the cellular phosphatase calcineurin, which dephosphorylates NFAT , resulting in nuclear import of the protein [12]. Thus, re-import of the reporter protein, which would complicate the analysis, is prevented under our conditions.

  6. 6.

    Use sodium-ATP. Lithium-ATP may interfere with export of GFP -NFAT [10].

  7. 7.

    This double-stranded oligonucleotide mimics the DNA-binding site of NFAT . It can stimulate export of GFP -NFAT about twofold, probably by promoting the release of GFP-NFAT from chromatin.

  8. 8.

    Check permeabilization with trypan blue, which stains the nuclei. As an alternative to dounce homogenization, the plasma membrane can be permeabilized with digitonin . The nuclear membrane remains intact, because of its lower level of cholesterol, as compared to the plasma membrane. Add 0.5–1 μL of a 10 % solution of digitonin in DMSO for 107 cells.

  9. 9.

    Transfection conditions are chosen such that essentially all mCherry-positive cells are also positive for our GFP -export cargo. In the final analysis by flow cytometry , this allows gating on mCherry-positive cells and measuring the residual GFP-fluorescence, assuming that even GFP-negative cells did contain the export cargo at the beginning of the reaction. Depending on the efficiency of the transfection , we obtain 15–30 % of total cells that express both, the red and the green fluorescent protein (see Fig. 1). The calcium phosphate method should not be used for proteins like NFAT , whose intracellular localization is controlled by the cellular calcium level.

  10. 10.

    It is also possible to simultaneously monitor nuclear import of fluorescently labeled import substrates (e.g., Cy5-BSA-NLS ; [10]). Furthermore, one may use propidium iodide to determine the DNA content of the cells. This can be done for nuclear import or export, allowing analysis of nuclear transport with respect to the cell cycle [28].