Protein Kinase Bα Is Required for Vesicle Trafficking and Class II Presentation of IgA Fc Receptor (CD89)-Targeted Antigen

Abs and fluorochromes

All anti-Ig Abs (including FITC-, indocarbocyanine 3 (Cy3)-, indocarbocyanine 5 (Cy5)-, and HRP-conjugated Abs) and fluorochrome or HRP-conjugated streptavidin were purchased from Jackson ImmunoResearch Laboratory, unless indicated. FcαR was ligated with the anti-FcαR mAb My43 (mouse IgM produced in our laboratory) (20) and cross-linked with either a FITC-conjugated rabbit anti-mouse IgM or FITC-conjugated donkey anti-mouse (DAM) IgM. Anti PKBα and anti-HA (hemagglutinin peptide sequence) mAbs were purchased from Cell Signaling Technology. Biotin-conjugated-anti-I-A^d and anti-lamp1 Abs were purchased from BD Pharmingen. Alexa 488-conjugated dextran was purchased from Molecular Probes.

Cell culture

The FcαR/γγ transfectants used in these studies were a gift from Dr. J. G. van de Winkel (University Medical Center, Utrecht, Netherlands) and have been described previously (9, 14). FcR- and γ-chain-negative IIA1.6 cells were transfected with pCAV plasmids encoding FcαR and the wild-type γ-chain (CD89/wt), or with FcαR and a mutant γ-chain (N-terminal cytoplasmic domains of the wild-type γ-chain and the C-terminal sequence of FcγRIIA, CD89/IIA). Transfectants were cultured in RPMI 1640 (Cellgro; Mediatech) medium supplemented with 10% fetal clone III (HyClone), 40 μg/ml gentamicin, 2 mM l-glutamine, 1 mM sodium pyruvate, and 2 μM methotrexate. Where indicated, CD89/IIA cells were transfected with the pcDNAI plasmid (conferring ampicillin and neomycin resistance) encoding a constitutively active PKBα (CA-PKBα; a gift from Dr. P. Stahl, Washington University, St. Louis, MN). CA-PKBα transfectants are referred to as IIA/CA-PKBα hereafter. IIA/CA-PKBα cells were cultured in selection medium containing 0.2 mg/ml G418. The OVA-specific, T cell hybridoma DO.11.10 cells (a gift from Dr. P. Marrack, National Jewish Medical and Research Center, Denver, CO) was cultured in DMEM (Cellgro; Mediatech) supplemented with nonessential amino acids (Invitrogen Life Technologies), 1 mM sodium pyruvate, 2 mM l-glutamine, 0.75 mg/ml dextrose, 0.85 mg/ml sodium bicarbonate, 50 μg/ml gentamicin, and 10% FBS (Sigma-Aldrich).

Chemical inhibitor and small interfering RNA (siRNA) treatment

The PKB inhibitor 1L-6-hydroxymethyl-chiro-inositol 2-(R)-2-O-methyl-3-O-octadecylcarbonate was purchased from EMD Biosciences. The PKB inhibitor was dissolved in DMSO to 1000 times the final required concentration. Inhibitors were further diluted in cell culture medium immediately before use and added to cell cultures as required. Nonspecific RNA duplexes, PKBα-specific siRNA duplexes, and transfection reagents were purchased from Qiagen. Transfection was performed according to the manufacturer’s instructions. CD89/wt cells were seeded at 1.5 × 10⁵ cells per well of a 24-well culture plate overnight. Media were removed, and cells were gently washed in serum-free medium twice before addition of 100 μl of serum-free medium. Transfection reagent (6 μl) was added to 300 μl of serum-free culture medium containing a 1 μg/ml final concentration of RNA duplexes. Samples were vortexed briefly and incubated for 25 min at room temperature to allow formation of transfection complexes. Three hundred microliters of medium containing transfection complexes was then added to cells and followed by incubation for 2 h at 37°C. Five hundred microliters of serum-containing medium was added and followed by a further 4-h culture before harvesting cells and culturing in 75 cm² flasks. Media were changed daily, and cells were harvested for experiments at 96 h posttransfection.

Ag presentation assays

The Ag presentation assay was similar to that described previously (9, 14). CD89-transfected A20IIA1.6 cells (2 × 10⁵ in 100 μl) were cultured in triplicate with 100 μl of anti-nitroiodophenol (NIP)-IgA/NIP-OVA complexes (IgA-OVA, 1 μg/ml) and DO.11.10 T cells (5 × 10⁴ in 100 μl) for 20 h. After incubation, supernatants were removed and frozen. For the PKB inhibition assay, cells were pretreated with the indicated concentration of PKC inhibitor for 30 min and then incubated with anti-NIP-IgA/NIP-OVA complexes for 4 h in the presence of PKB inhibitors. Inhibitors and noninternalized IgA-OVA complexes were removed by washing four times with PBS followed by culture with DO.11.10 T cells. For fluid phase Ag uptake, IIA1.6 cells were incubated with OVA (0.5, 1, or 2 mg/ml) for indicated times, and incubated with DO.11.10 for 20 h. Ag presentation was assessed by measuring the IL-2 concentration in the supernatant by ELISA.

IL-2 ELISA

ELISA plates were coated with 50 μl of 2.5 μg/ml rat anti-mouse IL-2 mAb in 0.1 M NaHPO₄ (pH 9.0) overnight at 4°C. Plates were washed three times with wash buffer (PBS, 0.05% Tween 20) and blocked with 200 μl of PBS, 1% BSA, 0.05% Tween 20, and 0.05% NaN₃ for 1 h at room temperature. Plates were washed three times with wash buffer before adding 100 μl of murine IL-2 standard (BD Pharmingen) or culture supernatants in binding buffer (PBS, 1% BSA, and 0.05% Tween 20) before incubating overnight at 4°C. Plates were washed four times with wash buffer before adding 100 μl of 2.5 μg/ml biotin-conjugated, rat anti-mouse IL-2 mAb for 1 h at room temperature, then washed four times with wash buffer before the addition of 100 μl of 0.25 μg/ml HRP-conjugated streptavidin, and incubated for 45 min at room temperature. Plates were then washed six times, after which 90 μl of ABTS peroxidase substrate (Kirkegaard & Perry Laboratories) was added to each well and allowed to develop for 2–5 min at room temperature before stopping the reaction with the addition of 110 μl of 10% SDS in ddH₂O. Absorbance at 450 nm was measured using a Dynex MRX Revelation Microplate Absorbance Reader (Dynex Technologies). Statistical differences in IL-2 concentration between DMSO- and inhibitor-treated cells were determined by Student’s t test.

SDS-PAGE and immunoblotting

Cell lysates corresponding to 10⁶ cell equivalents were resolved by SDS-PAGE. Proteins were transferred to nitrocellulose membranes and incubated for 1 h at 4°C with 5% nonfat dry milk and 0.5% Tween 20 in PBS. Membranes were incubated with primary Abs (0.2–2.0 μg/ml) overnight at 4°C, and washed six times for 5 min in PBS then incubated with 3% nonfat dry milk/0.05% Tween 20 in PBS. HRP-conjugated secondary Ab was then added to a final concentration of 0.2 μg/ml and incubated for a further 2 h at 22°C before washing six times for 5 min in PBS. Membrane-bound HRP-conjugated Abs were detected by ECL (Amersham Pharmacia Biotech).

Flow cytometry

Cells were incubated at 4°C in 0.1% w/v BSA in PBS and surface stained with FITC-conjugated or PE-conjugated Abs for 1 h before washing three times with ice-cold PBS. Cells were then fixed in 1.0% w/v paraformaldehyde in PBS and analyzed using a BD Biosciences FACScan in combination with CellQuest software (BD Biosciences).

Receptor internalization and confocal microscopy

CD89-transfected cells (10⁶ cells per sample in a 100-μl volume) were incubated with 1 μg of anti-FcαR mAb (My43) for 1 h at 4°C before washing three times in ice-cold PBS. Cells were incubated with 1 μg of FITC-conjugated DAM for 1 h at 4°C followed by three washes in PBS and a 30-min incubation at 37°C. For PKB inhibitor experiments, cells were pretreated with desired final concentrations of inhibitor at 4°C for 30 min before FcαR endocytosis. Endocytosis was stopped by addition of 10 ml of ice-cold 1% paraformaldehyde in PBS to each 1-ml sample. Cells were permeabilized in 0.5% w/v saponin, 0.1% w/v BSA, 0.1% w/v NaN₃ in 1× PBS for 30 min, before intracellular staining for vesicle markers lamp1, class II, Rab4, or Rab5a. Cy3-conjugated secondary Abs were then used to detect bound primary Abs. Cells were mounted on slides and analyzed using a Zeiss LSM Meta confocal microscope.

Colocalization of CD89 with vesicle markers was assessed using NIH Image J software (〈http://rsb.info.nih.gov/ij〉). Briefly, colocalized CD89 (green) and vesicle (red) pixels were determined and subtracted from the total CD89-associated signal (integral pixel intensity). Noncolocalized integral pixel intensities were then taken as a percentage of the total CD89-associated integral pixel intensity and used to determine the percentage colocalization of the total CD89-associated signal with the vesicle marker. Data were then expressed using GraphPad Prism software. Statistical significance between different experimental conditions was determined by ANOVA.

Aggregation of class II⁺ vesicles was assessed by measuring the area over which vesicles were distributed relative to the total midsection area of the cell. A more aggregated phenotype results in a smaller area being occupied. The percentage of aggregation was expressed such that all values were shown relative to 100% (infinite) aggregation. Statistical significance between different experimental conditions was determined by ANOVA.

Fluid phase pinocytosis

CD89/wt cells were untreated or treated with DMSO or PKB inhibitor in DMSO (25 μM final concentration) for 30 min. Cells were then fixed, or treated with FITC-conjugated OVA, at a final concentration of 1.0 mg/ml in the presence of PKB inhibitor for 1, 2, or 4 h at 37°C. Cells were washed in ice-cold PBS three times before fixation and analysis by confocal microscopy. Alternatively, 96 h after PKBα “knockdown” with siRNA, cells were pulsed with FITC-OVA then fixed and counterstained for class II before analysis.

CD89-mediated class II Ag presentation requires PKBα activation

CD89/γ-chain-induced signaling is necessary for CD89-mediated Ag presentation (9, 14, 15). We tested the effect of a PKB chemical inhibitor on class II presentation of CD89-targeted NIP-OVA (Fig. 1⇓). This inhibitor functions as a lipid analog that competes for binding of the pleckstrin homology domain of PKB with PI(3,4,5)P₃, thus preventing membrane recruitment and activation (21, 22). The PKB inhibitor caused a dose-dependent inhibition of CD89 Ag presentation with an IC₅₀ of 14.5 μM (Fig. 1⇓a). DMSO, the carrier for the inhibitor had no effect on Ag presentation. Washing out the inhibitor before addition of Ag allowed a complete recovery of Ag presentation, indicating that DO.11.10 T cell hybridomas were not adversely affected by inhibitor-treated APCs (Fig. 1⇓b). The PKB inhibitor had no effect on expression of CD89 (Fig. 1⇓c) or class II (Fig. 1⇓d) as assessed by flow cytometry. Thus inhibition was not due to adverse effects on the ability to capture or present Ag. The PKB inhibitor used in this study blocks PKB activity in vitro with an IC₅₀ of 5 μM and growth of cancer cell lines with variable IC₅₀ values (1–10 μM) (23). The IC₅₀ for inhibition of PI3-kinase by PKB inhibitor I is 83 μM. Therefore, our data are consistent with PKB being a target of the inhibitor in our experiments.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1.

CD89-mediated Ag presentation is blocked by chemical inhibition of PKB. a, Culture-plate-adherent CD89/wt cells were pretreated for 30 min at 37°C with DMSO or PKB inhibitor at the final concentrations indicated. NIP-OVA and IgA (anti-NIP) were added at final concentrations of 2 and 6 μg/ml, respectively. Cells were then incubated for 4 h at 37°C. Cells were washed three times in culture medium before addition of DO.11.10 T cells and incubation for 20 h at 37°C. Supernatants were harvested and assayed for IL-2 by ELISA. b, The PKB inhibitor was removed by washing before Ag pulse (50/rec). Data in a and b show mean ± SD for triplicate samples and are representative of four independent experiments. Asterisks denote IL-2 concentrations significantly different from DMSO-treated cells (p < 0.05). c and d, APCs were untreated or DMSO- or inhibitor-treated (50 μM) for the time periods shown, before fixation and staining for CD89 (c) and I-A^d (d) and before analysis by flow cytometry. In the upper left panel, background staining is shown by faint dotted line, specific staining is shown by solid and heavier dotted lines. In the upper right and lower panels, untreated cells are shown in gray fill, DMSO is shown by solid lines, and inhibitor is shown by dotted lines.

We tested the effect of the PKB inhibitor on the presentation of Ag acquired by fluid phase pinocytosis (Fig. 2⇓). The PKB inhibitor caused a dose-dependent inhibition of Ag presentation with an IC₅₀ value of 19.14 μM (Fig. 2⇓a). DMSO showed a small inhibitory effect (<10%) on fluid phase Ag presentation that was not statistically significant. To demonstrate that PKB inhibition did not prevent fluid phase pinocytosis, we treated cells with inhibitor before incubation with FITC-conjugated OVA (Fig. 2⇓b). PKB inhibition did not inhibit fluid phase pinocytosis as shown by flow cytometry. By confocal microscopy analysis we observed a punctate granular distribution of fluorescent FITC-OVA consistent with pinocytosis (Fig. 2⇓b). We also observed similar results using the pinocytic marker Alexa-conjugated dextran (data not shown).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2.

Chemical inhibition blocks presentation of fluid phase Ag, but not pinocytosis. a, Culture-plate-adherent CD89/wt cells were pretreated for 30 min at 37°C with DMSO or PKB inhibitor at the final concentrations indicated. OVA was added at a final concentration of 1 mg/ml. Cells were then incubated for 4 h at 37°C. Cells were washed three times in culture medium before the addition of DO.11.10 T cells and incubation for 20 h at 37°C. Supernatants were harvested and assayed for IL-2 by ELISA. Data show mean ± SD for triplicate samples and are representative of four independent experiments. Asterisks denote IL-2 concentrations significantly different from DMSO-treated cells (p < 0.05) b, Cells were untreated or pretreated for 30 min at 37°C with DMSO or PKB inhibitor (25 μM) before incubation with FITC-conjugated OVA at 1 mg/ml for times indicated. Cells were then washed to remove excess OVA. Graph shows mean fluorescence intensity as assessed by flow cytometry. Image shows distribution of OVA at 2 h in individual cells. White lines indicate cells from different fields of view.

Most chemical inhibitors are selective rather than specific, but the PKB inhibitor is reported to have good specificity (21, 22, 23). However, we wished to confirm that PKB was indeed required for CD89-mediated Ag presentation using an independent approach. Therefore, we adopted a molecular strategy to confirm the requirement for PKB for Ag presentation. CD89/wt cells were transfected with nonspecific or PKBα-specific short interfering RNA duplexes. This strategy permits specific knockdown of PKBα expression (24). At 96 h posttransfection, we observed a 20–45% inhibition of presentation of CD89-targeted OVA depending on Ag concentration (Fig. 3⇓a). With lower IgA/OVA concentrations, and presumably a lower degree of CD89 cross-linking, the dependency on PKBα increased. In the same experiments, we compared class II presentation of fluid phase OVA (Fig. 3⇓b). In contrast to results obtained for the chemical inhibitor, presentation of Ag acquired by pinocytosis was not inhibited, suggesting that other PKB isoforms (PKBβ), and not PKBα, regulate presentation of fluid phase Ag. Transfection of PKBα-specific siRNA resulted in an ∼40% specific knockdown of PKBα expression (relative to nonspecific RNA) at 96 h posttransfection (Fig. 3⇓c).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 3.

CD89-mediated Ag presentation is blocked by specific siRNAⁱ knockdown of PKBα expression. Cells were treated with nonspecific siRNA (NS) or PKBα-specific siRNA (PKBα) as described in Materials and Methods. Nonspecific and PKBα cells were incubated with IgA anti-NIP and NIP OVA (a) or OVA (b) as described. DO.11.10 T cells were then added and cocultured before assay of IL-2 in supernatants. Data show the mean ± SD IL-2 production from triplicate samples. Asterisks denote IL-2 concentrations significantly different from nonspecific siRNA-treated cells (p < 0.05). Results are representative of four independent experiments. c, Whole cell lysates were prepared and immunoblotted for PKBα before reprobing membranes for β-actin. Relative expression of PKBα to β-actin is shown. d, Cells were fixed and stained for CD89 or I-A^d and analyzed by flow cytometry. Isotype control-associated fluorescence is shown as a faint dotted line. Untreated cells are shown as a gray filled area, nonspecific-RNA-treated cells are shown by a solid line, and PKBα-RNA is shown by a heavy dotted line.

PKBα knockdown had no measurable effect on expression of CD89 or class II (Fig. 3⇑d). We also examined the subcellular distribution of class II relative to trans-Golgi network (Rab6), late endosome (LE) (Rab7), and recycling/early endocytic vesicles (Rab4). We did not observe any difference in class II subcellular localization, indicating that synthesis, assembly, and transport was intact in PKBα-inhibited cells (data not shown).

PKBα is required for CD89 trafficking to MIIC

CD89/IIA cells exhibit a reduction in trafficking of internalized CD89 to MIIC and aggregation of MIIC vesicles (9). Aggregated CD89-positive vesicles have all the characteristics of MIIC in that they contain lamp1, cathepsin B, and class II in CD89/wt cells (15). We tested the effect of PKBα siRNA knockdown, on MIIC induction (Fig. 4⇓a). After treatment with nonspecific RNA, we observed the formation of large CD89⁺ aggregates in CD89/wt cells (Fig. 4⇓a). Transfection with siRNA specific for PKBα inhibited formation of large aggregates, resulting in formation of several smaller CD89⁺ vesicles with a more punctuate distribution. Importantly, this phenotype is typical of CD89/IIA cells (9, 15). We also tested the effect of the PKB inhibitor on aggregation of CD89⁺ vesicles and obtained similar results to that observed for siRNA (data not shown).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 4.

PKBα knockdown inhibits delivery of cross-linked CD89 to class II-containing vesicles. Cells were treated with nonspecific siRNA (NS) or PKBα-specific siRNA (PKBα) as described in Materials and Methods. a, CD89 was ligated with the anti-CD89 mAb My43 and cross-linked with FITC-conjugated DAM IgM (green) at 4°C before washing and incubation for 30 min at 37°C. Cell lysates were immunoblotted for PKBα and PKCδ. The bar graph shows percent knockdown of PKBα expression relative to PKCδ. b, CD89 (green) was cross-linked and internalized as in a. Cells were then fixed, permeabilized, and counterstained for class II (red). Colocalization of CD89 and class II is indicated by yellow (merged image) or white (colocalized pixels). Images shown are representative of at least 100 cells for each condition. White lines separate cells from different fields of view. Graphs show percent colocalization of internalized CD89 with I-A^d. Each data point represents an individual cell.

To confirm that PKBα knockdown inhibits transport of CD89 to MIIC, we allowed cells to internalize cross-linked CD89 and counterstained them for the MIIC marker class II (Fig. 4⇑b). In nonspecific siRNA-treated cells, large characteristic CD89⁺ vesicles were observed that colocalized extensively with the intracellular pool of I-A^d class II that presents OVA peptide to DO.11.10 cells. In PKBα siRNA-treated cells, numerous smaller vesicles were observed with poor class II colocalization. This suggests that PKBα is required for transport from earlier class II-negative or class II-low endocytic organelles to class II-rich MIIC.

PKBα is not required for delivery of fluid phase Ag to class II-containing vesicles

Because PKBα-specific siRNA did not inhibit presentation of fluid phase Ag, one would expect that delivery of Ag to class II-containing vesicles would not require PKBα. Therefore, we performed siRNA knockdown of PKBα and tested the effect on delivery to class II-containing vesicles. We observed no difference between delivery of FITC-OVA to class II-containing vesicles in nonspecific RNA- and PKBα siRNA-treated cells (Fig. 5⇓). Large MIIC aggregates as formed by signaling were not generated in cells pulsed with fluid phase Ag. Similarly, delivery of FITC-OVA to lamp1-containing vesicles after PKBα knockdown was not inhibited, and lamp1 vesicles did not aggregate substantially (data not shown).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 5.

PKBα knockdown does not inhibit delivery of fluid phase Ag to class II-containing vesicles. CD89/wt cells were transfected with nonspecific (NS) or PKBα-specific siRNA as described in Materials and Methods. After 96 h, cells were harvested and incubated with 1 mg/ml FITC-OVA (green) for 4 h at 37°C (a). Cells were washed and fixed before permeabilization and counterstaining for class II (red). Images show representative cells with colocalization indicated in yellow and colocalized pixels in white. White lines separate cells from different fields of view. b, Graph shows percentage colocalization of CD89 and class II. Each data point represents an individual cell. c, Immunoblot shows PKBα expression after treatment with nonspecific or PKBα siRNA relative to loading control (actin).

Reconstitution of PKBα activity “rescues” delivery of CD89 to late endosomes, but not delivery to MIIC or Ag presentation

CD89/IIA cells are 8- to 16-fold less efficient at CD89-mediated class II Ag presentation than CD89/wt cells (9, 14). CD89/IIA cells are a good system in which to determine whether reconstitution of PKBα is sufficient to rescue Ag presentation. We used a molecular approach in which CD89/IIA cells were transfected with the pcDNAI plasmid encoding CA-PKBα (Fig. 6⇓a). Addition of an N-terminal Src myristoylation sequence to PKBα results in membrane targeting and constitutive activation (25). CD89/IIA cells were deficient in Ag presentation as reported previously (9, 14). We observed that IIA/CA-PKBα cells had a similar ability to present CD89-targeted OVA as CD89/IIA cells (Fig. 6⇓a). We also tested the effect of CA-PKBα expression on presentation of fluid phase OVA (which is intact in CD89/IIA cells) (9). Consistent with our siRNA experiments, CA-PKBα had no significant impact on fluid phase presentation by the CD89/IIA cells (data not shown). The inability of CA-PKBα to rescue Ag presentation by CD89/IIA cells was not due to lack of expression of the mutant, which was expressed at similar levels to endogenous PKBα (Fig 6⇓b). Furthermore, CD89/wt, CD89/IIA, and IIA/CA-PKBα cells all had similar levels of expression of CD89 (Fig. 6⇓c) and class II (Fig. 6⇓d). These data show that CA-PKBα is insufficient to rescue Ag presentation by CD89/IIA cells.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 6.

CA-PKBα is insufficient to rescue class II presentation by CD89/IIA cells. a, CD89/wt, CD89/IIA, and IIA/CA-PKBα cells were incubated with IgA anti-NIP and NIP OVA as described before coculture with DO.11.10 T cells. Data shows mean IL-2 concentration in supernatants for triplicate samples. Near identical results were obtained in two independent experiments. b, Expression of CA-PKBα (HA peptide-tagged) was detected by immunoblotting whole cell lysates with an anti-HA Ab. Endogenous PKBα was detected with an anti-PKBα Ab. Anti-PKBα does not detect the HA-tagged CA-PKBα. The epitope recognized by anti-PKBα lies within the region of CA-PKBα that was truncated (aa 4–129). Cells were fixed and stained for CD89 (c) or class II (d) expression before analysis by flow cytometry. The gray filled area shows CD89/wt cells, the faint dotted line shows CD89/IIA cells, and the faint dotted line shows IIA/CA-PKBα cells. Isotype control-associated fluorescence is also shown (faint dotted line on left of histogram).

We then tested the effect of CA-PKBα expression on MIIC formation in CD89/IIA cells (Fig. 7⇓). As reported previously, CD89/wt cells formed large MIIC-like structures containing CD89 colocalized with lamp1 and class II, whereas CD89/IIA cells showed reduced MIIC formation and lamp1/class II colocalization (9, 15) (Fig. 7⇓, a and b). When IIA/CA-PKBα cells were analyzed similarly, the phenotype observed was that of CD89 colocalization with lamp1 in large aggregated vesicles (Fig. 7⇓a). To confirm our observations, the percentage of internalized CD89 that colocalized with lamp1 was measured using NIH Image J software. The mean degree of CD89/lamp1 colocalization was similar in CD89/wt and IIA/CA-PKBα cells, although there was greater heterogeneity within the CD89/wt cells.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 7.

CA-PKBα“rescues” CD89 transport to lamp1, but not class II compartment CD89/IIA cells. CD89 (green) was cross-linked and internalized in CD89/wt, CD89/IIA, and CD89/IIA, CA-PKBα cells as described in Materials and Methods. Cells were fixed, permeabilized, and counterstained for Lamp1 (a) or class II (b). Images shown are representative of at least 100 cells. White lines separate cells from different fields of view. Graphs show percentage colocalization of internalized CD89 with lamp1 or class II. c, Class II vesicle aggregation after CD89 cross-linking and internalization was measured for each cell line. Each data point in graphs a–c represents an individual cell. Data are representative of two independent experiments.

In contrast, there was only a small (but statistically significant) increase in CD89/class II colocalization in IIA/CA-PKBα cells as compared with CD89/IIA cells (Fig. 7⇑b). Consistent with this, aggregation of class II⁺ vesicles was lower in IIA/ and IIA/CA-PKBα cells than wt cells (Fig. 7⇑c). Our data show that expression of CA-PKBα is sufficient to rescue trafficking of CD89 to aggregated lamp1 vesicles, but not entirely rescue the appearance of class II in these vesicles in CD89/IIA cells. This suggests that PKBα is sufficient for CD89 transport to late endocytic compartments but not for formation of fully functional class II-containing MIIC.