Association of brefeldin A-inhibited guanine nucleotide-exchange protein 2 (BIG2) with recycling endosomes during transferrin uptake

Endocytosis of Tfn and Localization of BIG1 and BIG2 in COS7 Cells.

To evaluate the relationship of BIG1 and BIG2 to endosomal structures, the proteins were observed microscopically during the uptake of Alexa Fluor 568-tagged Tfn. Before addition of Tfn and after 2 min of uptake, when most of the Tfn would still be in EEs, it did not overlap with BIG1 or BIG2, which remained largely concentrated near cell nuclei, perhaps with Golgi structures (Fig. 1). After 10 min however, when some of the Tfn was expected to be in REs, BIG2 partially coincided with Tfn in small punctate collections around the initial BIG2 concentration, whereas no change in BIG1 was seen. After 30 and 60 min of uptake, when Tfn was expected to have reached Rab4/Rab11-positive REs (16), its colocalization in punctate concentrations with BIG2, but not BIG1, was widespread (Fig. 1); colocalization of BIG2 and EEA1 was not detected (data not shown). With increasing time of Tfn uptake, BIG2-stained structures increased in size, number, and distribution, consistent with BIG2 accumulation in REs.

Separation by Density Gradient Centrifugation.

To evaluate the effects of Tfn uptake on the distribution of BIG1 and BIG2 and several proteins associated with specific intracellular structures, postnuclear supernatant fractions from cells before and after 2 or 60 min of incubation with Tfn were subjected to centrifugation in iodixanol density gradients. Western blotting of total proteins before gradient fractionation revealed that amounts of BIG2 and Rab11(but not BIG1, Rab4, Exo70, or TfnR) were significantly greater (19% and 17%, respectively) after 60 min of Tfn uptake than they were at zero time or 2 min (Fig. 2A).

In cells fractionated after 2 min of Tfn uptake, Tfn was confined to fractions 15–21, which also contained all of Rab4 and EEA1, plus a large fraction of Rab11, as they did before Tfn addition (Fig. 2B). Most of the BIG1 and BIG2 were also present in those fractions, but their distributions quantitatively did not parallel those of the other proteins. After 60 min of uptake, Tfn was seen additionally in fractions 7–13, as was Rab11. At no time did distribution of Rab11 in the gradient parallel that of Tfn or TfnR (Fig. 2B). Relative amounts of both BIG2 and Exo70 in fractions 9–11, which may include a population of REs, were clearly greater at 60 min than at zero or 2 min (Fig. 2B). Distributions of BIG1, Rab4, and EEA1 were apparently unchanged with time, whereas TfnR was at all times most abundant in fractions 7–15, consistent with the presence of REs in these fractions (14).

Efficacy and Specificity of BIG1 and BIG2 Small Interfering RNA (siRNA).

To assess in a different way the involvement of BIG2 (and BIG1) in Tfn/TfnR recycling, we used siRNA to decrease amounts of endogenous BIG1 and/or BIG2 in COS7 cells. Amounts of endogenous BIG1 and BIG2 were clearly decreased after 24 h of incubation with BIG1 or BIG2 siRNA, respectively, and were even lower after 48 h (Fig. 3A). Quantification of immunoreactive proteins by densitometry 72 h after addition of BIG1 or BIG2 siRNA (Fig. 3 B and C) showed that endogenous BIG1 and BIG2 were decreased by 93% (7 ± 0.04% vs. mock) and 92% (8 ± 0.05% vs. mock), respectively.

When incubation was continued after removal of siRNA at 72 h, increased BIG1 or BIG2 was detectable after 24 h and more so after 48 h (Fig. 3A). Treatment of cells with BIG1 or BIG2 siRNA for 72 h did not affect amounts of the other BIG protein or of ARF or actin (Fig. 3 B and C). Likewise amounts of these proteins were unaffected by the vehicle alone (mock) or siRNA negative control (nontargeting) (Fig. 3 B and C).

Effects of BIG1 and BIG2 siRNA on TfnR.

In polarized and nonpolarized cells, TfnR and Tfn are believed to remain together and, after release of iron from Tfn, to return together to the cell surface from the peripheral endosomes, either directly or indirectly via pericentriolar REs (13, 14). BIG1 siRNA did not alter the amount or distribution of TfnR seen on confocal microscopy (Fig. 4A), but BIG2 siRNA alone or with BIG1 siRNA clearly increased TfnR in pericentriolar REs very close to structures containing giantin, a marker for cis- and median-Golgi, which were not discernably affected by the siRNAs (Fig. 4A), or by Tfn uptake (data not shown). Quantification of TfnR on Western blots by densitometry confirmed that it was 30% and 40% higher in cells incubated with BIG2 siRNA or BIG2 plus BIG1 siRNA, respectively, and the smaller increment with BIG1 siRNA alone was not statistically significant (Fig. 4 B and C).

Effect of BIG1 and BIG2 siRNA on Tfn Uptake and Intracellular Trafficking.

Tfn bound to the TfnR is believed to move sequentially through Rab4-positive EEs and, at least in part, through REs that contain Rab11 before being returned to the cell surface and released. To try to quantify initial rates of Tfn uptake, we measured Tfn content of cells in the first 7 min of uptake (Fig. 5A). Accumulation of Tfn in cells was not constant during this time, presumably because TfnR is cycling even without external ligand, and Tfn accumulation is always the difference between uptake and release (Fig. 5A). Although TfnR was significantly increased in cells incubated with BIG2 siRNA (Fig. 4C), confocal microscopy failed to reveal the effects of BIG1 or BIG2 siRNA alone or together on the amounts or localization of TfnR. These cells did accumulate significantly more Tfn after 3, 5, and 7 min of uptake than did cells incubated with vehicle or BIG2 siRNA alone, both of which had no effect (Fig. 5A).

To assess Tfn release from REs, cells that had taken up Alexa Fluor 488-tagged Tfn for 60 min were washed and incubated in medium containing unlabeled Tfn; amounts of 488-tagged Tfn in medium and cells were quantified at intervals thereafter, as shown in Fig. 5B. There were no significant differences in release among mock-transfected (vehicle alone) and siRNA-transfected cells before 10 min. At later times, however, Tfn release was significantly slower from cells transfected with BIG2 or BIG2 plus BIG1 siRNA than from cells transfected with vehicle alone or BIG1 siRNA (Fig. 5B).

Microscopically (Fig. 5C), Alexa Fluor 568-tagged Tfn appeared to be largely absent from peripheral sites in most cells within 10 min, albeit a bit more persistent, perhaps, in cells transfected with BIG2 siRNA, with or without BIG1 siRNA. This finding would be consistent with the initial, faster release from all cells of ≈50% of intracellular Tfn representing largely the clearance of Tfn from EEs. After 30 and 60 min, cells transfected with BIG2, with or without BIG1, siRNA appeared to contain more Tfn than mock- or BIG1 siRNA-transfected cells (Fig. 5C).

Antibodies and Reagents.

Rabbit anti-BIG1 and anti-BIG2 antibodies against synthetic peptides CSQPPEQELGINKQ (italic C added to facilitate coupling to keyhole limpet hemocyanin) and RLKHSQAQSK, respectively (9), were affinity-purified with the peptides used for immunization coupled to a SulfoLink kit (Pierce) according to the manufacturer’s instructions. Mouse monoclonal antibody against ARF (catalog no. MA3-060) was purchased from Affinity BioReagents (Neshanic Station, NJ); mouse monoclonal antibody against TfnR (catalog no. 13-6800) and rabbit anti-Rab11 antibodies (catalog no. 71-5300) were purchased from Zymed Laboratories; mouse anti-Tfn monoclonal antibody (catalog no. ab10208) was from Abcam (Cambridge, MA); rabbit anti-giantin antibody (catalog no. PRB-114C) was from Covance (Richmond, CA); Alexa Fluor 488- or 594-conjugated secondary antibodies, Alexa Fluor 568- or 488 -tagged Tfn, anti-Alexa Fluor 488 rabbit IgG fraction (catalog no. A-11094), and goat anti-mouse IgG–horseradish peroxidase were from Molecular Probes; o-phenylenediamine was from Sigma; EDTA-free protease inhibitor was from Roche; and all of the reagents for siRNA were from Dharmacon Research (Layfayette, CO).

Immunofluorescence Microscopy.

COS7 cells (used because they are more favorable than HepG2 cells for immunofluorescence microscopy) were grown at 37°C in 5% CO₂/95% and DMEM with 10% FBS (GIBCO), 100 units/ml penicillin G, and 100 μg/ml streptomycin. For experiments, 2 × 10⁴ cells were incubated in four-well culture slides for 24 h and then for 3 h in serum-free medium containing 0.2% BSA to deplete endogenous Tfn. After incubation with Alexa Fluor 568-tagged Tfn (25 μg/ml) for the time indicated in the figures, cells were fixed with 3% paraformaldehyde in PBS for 12 min at room temperature, washed with PBS, and incubated in blocking buffer (10% normal goat serum and 0.1% saponin in PBS) for 30 min. Cells were incubated with anti-BIG1 or -BIG2 antibodies (diluted with blocking buffer, respectively, 1:1,000 and 1:500) for 1 h at room temperature. After being washed with PBS, cells were incubated for 1 h with Alexa Fluor 488-labeled anti-rabbit IgG (diluted 1:1,000 in blocking buffer) and washed with PBS. Coverslips were mounted in Prolong Gold anti-fade reagent with DAPI (Molecular Probes) and inspected with a confocal microscope (Leica model TCS4D/DMIRBE).

Separation of EEs and REs on Optiprep Iodixanol Gradient.

COS7 cells grown to 70% confluence on 100-mm culture plates were incubated without serum for 3 h at 37°C, then without or with Alexa Fluor 488-tagged Tfn for the time indicated in the figures before washing and scraping in 50 mM Hepes (pH 7.2) containing 0.25 M sucrose, 78 mM KCl, 8.37 mM CaCl₂, 10 mM EGTA, and 4 mM MgCl₂ (1 ml per plate), followed by homogenization (20 strokes in a Dounce homogenizer), and centrifugation for 5 min at 1,000 × g. Samples (1 ml) of supernatants were applied to 5–20% iodixanol gradients (10 ml), which were centrifuged for 16 h at 27,000 rpm on a SW41 rotor before collection from the bottom of 500-μl fractions. Trichloroacetic acid-precipitated proteins from each fraction were dispersed in 50 μl of NuPAGE LDS sample buffer, proteins in samples (25 μl) were separated in NuPAGE gels, and blots were incubated with antibodies against the indicated proteins. Signals were detected by using LAS-3000 (Fujifilm), and the intensities of the blots were quantified by densitometry using multi gauge software (Fujifilm).

siRNA Treatment and Tfn Uptake and Release.

siRNAs against human BIG1 (5′-TGAATCACCTCAACTTAGA-3′) and BIG2 (5′-CAACTACGACTGTGAT TTA-3′) were designed and synthesized as Option A4 by Dharmacon Research. Dharmacon siCONTROL nontargeting siRNA was used as a negative control. COS7 cells were incubated for 24 h in four-well culture slides for immunofluorescence microscopy or in 100-mm plates for Western blotting, followed by incubation for 72 h in DharmaFECT2 siRNA transfection reagent containing 100 pmol of siRNA duplexes or other additions as indicated in the figures according to the manufacturer’s instructions. The cells were then washed and incubated for 3 h in serum-free medium containing 1% BSA.

For Tfn uptake, cells were incubated with Alexa Fluor 488-tagged Tfn (25 μg/ml) for the time indicated in the figures, placed on ice, quickly washed four times with ice-cold PBS, and lysed in 0.5 ml of lysis buffer (1% Triton X-100/0.1% SDS/0.2% BSA/50 mM NaCl/1 mM EDTA/10 mM Tris, pH 7.4/1:1,000 protease inhibitors). For Tfn recycling, cells were incubated with Alexa Fluor 488-tagged Tfn at 37°C for 60 min, placed on ice, washed four times with ice-cold PBS containing 0.1% BSA, stripped of surface-bound Tfn with stripping buffer (25 mM citric acid/24.5 mM sodium citrate/280 mM sucrose, pH 4.6), and incubated with serum-free DMEM containing 1% BSA and 2.5 mg/ml unlabeled Tfn for the times indicated in the figures. Medium was collected, and cells were lysed in 0.5 ml of lysis buffer. Alexa Fluor 488-tagged Tfn was quantified by ELISA.

Quantification of Tfn in Cells and Medium by ELISA.

ELISA plates were prepared by using Immulon 2HB microplates (Fisher) incubated overnight at 4°C with 10 μg/ml rabbit anti-488 IgG in 0.1 M sodium bicarbonate, pH 9.3. After being washed four times in PBST (PBS containing 0.05% Tween 20), plates were incubated overnight at 4°C with PBS containing 1% BSA and washed three times with PBST. Sample (50 μl) and PBS (50 μl) were then added to each well, followed by incubation for 2 h at room temperature. After five washes with PBST, 100 μl of 50 ng/ml mouse anti-Tfn in PBS with 0.1% BSA were added to each well, and 60 min later, the plates were washed five times with PBST before adding 100 μl of 50 ng/ml goat anti-mouse IgG–horseradish peroxidase in PBS with 0.1% BSA to each well. After 60 min and five washes with PBST, 100 μl of substrate solution (o-phenylenedianine dihydrochloride, 30 mg tablet dissolved in 15 ml of 20 mM sodium citrate buffer, pH 5.0, plus 100 μl of 3% H₂O₂). Addition of 100 μl of 1 M H₂SO₄ to each well 20 min later terminated the reactions. Absorbance at 490 nm was quantified by using an ELISA plate reader (Bio-Rad).