Cotranslational endoplasmic reticulum assembly of FcepsilonRI controls the formation of functional IgE-binding receptors

Abstract

The human high affinity receptor for IgE (FcepsilonRI) is a cell surface structure critical for the pathology of allergic reactions. Human FcepsilonRI is expressed as a tetramer (alphabetagamma(2)) on basophils or mast cells and as trimeric (alphagamma(2)) complex on antigen-presenting cells. Expression of the human alpha subunit can be down-regulated by a splice variant of FcepsilonRIbeta (beta(var)). We demonstrate that FcepsilonRIalpha is the core subunit with which the other subunits assemble strictly cotranslationally. In addition to alphabetagamma(2) and alphagamma(2), we demonstrate the presence of alphabeta and alphabeta(var)gamma(2) complexes that are stable in the detergent Brij 96. The role of individual FcepsilonRI subunits for the formation of functional, immunoglobulin E-binding FcepsilonRI complexes during endoplasmic reticulum (ER) assembly can be defined as follows: beta and gamma support ER insertion, signal peptide cleavage and proper N-glycosylation of alpha, whereas beta(var) allows accumulation of alpha protein backbone. We show that assembly of FcepsilonRI in the ER is a key step for the regulation of surface expression of FcepsilonRI. The ER quality control system thus regulates the quantity of functional FcepsilonRI, which in turn controls onset and persistence of allergic reactions.

Figure 1.

In vitro translation of IgE-binding FcɛRIα. (A) Signal peptide cleavage and N-glycosylation of FcɛRIα depends on the source of microsomes used during in vitro translation. Transcripts of K^b-FcɛRIα were translated for 1 h at 30°C in the presence of microsomes derived from KU812, dog pancreas (dpm), or U373 (CC). After lysis of the microsomal pellets in 1% Brij96 lysis buffer, immunoprecipitates were obtained with polyclonal anti-α serum and either mock digested or digested with Endo H. Subsequently immunoprecipitates were analyzed by 12.5% SDS-PAGE under reducing conditions. Conversion of α_+sig to α_Nglyc occurs most efficiently in CC microsomes. (B) α_Nglyc generated with CC microsomes expresses properly folded IgE-binding epitopes. IgE selectively precipitates the fully N-glycosylated form of FcɛRIα. Representative experiments (n = 3).

Figure 2.

In vitro translation of FcɛRIβ and FcɛRIβ_var. (A) Transcripts of FcɛRIβ or FcɛRIβ_var were translated for 1 h at 30°C in the presence of CC microsomes. After lysis of the microsomal pellets in 1% Brij96 lysis buffer, immunoprecipitates were obtained with polyclonal anti-β serum and analyzed by 12.5% SDS-PAGE under reducing conditions. Representative experiments (n = 3).

Figure 3.

ER insertion and proper signal peptide cleavage of FcɛRIγ. (A) Proper insertion and signal peptide cleavage of K^b-γ in an in vitro translations with a [³⁵S]cysteine translation mix. (B) In vitro translations of K^b-γ with a [³⁵S]methionine translation mix. The single methionine of γ gets cleaved after insertion into the microsomes. [³⁵S]Cysteine present in the [³⁵S]methionine translation mix is sufficient for labeling of K^b-γ. In vitro translations were performed in the absence and presence of CC microsomes. Translation products were analyzed after immunoprecipitation from 1% Brij96 lysates of in vitro translations performed in the absence of microsomes (−) or lysates of microsomal pellets (+) and analyzed by 12.5% SDS-PAGE under reducing (A and B) or nonreducing (C and D) conditions. FcɛRIγ forms a dimer under nonreducing conditions. Representative experiments (n = 3).

Figure 4.

FcɛRI complexes form cotranslationally. (A) Cotranslational formation of αγ complexes. α and γ RNA were translated consecutively (lanes 1 and 3, 1^stα 2^ndγ) or at the same time (lane 2 and 4, α + γ) into CC microsomes. Direct loads of the microsomal pellets confirmed the presence of both translation products (lanes 1 and 2). αγ complexes were only retrieved by anti-α immunopreciptiation, if both proteins were translated at the same time (lane 4, α + γ). (B) α and γ were translated separately and microsomal pellets were mixed before lysis. Direct loads of microsomes showed the presence of both proteins. No αγ complexes were precipitated. (C) Immunoprecipitations of αγ complexes with serum directed against γ.

Figure 5.

Association studies of different FcɛRI subunits. (A) Cotranslation of β and γ chains results in efficient expression of both proteins in microsomes. Anti-β and -γ serum both failed to coprecipitate βγ complexes. (B) Demonstration of αβ complexes. Anti-α serum precipitates stable αβ complexes. The presence of β is in the complex is confirmed by anti-β reimmunoprecipitations. (C) Confirmation of αβ complexes in 293 cells transiently transfected with α_HA and GFP-β cDNA. (D and E) The β and γ subunits associate with α but not HLA-A2. FcɛRI complexes form only cotranslationally. Direct loads of microsomes (D) in which proteins were translated separately and mix before lysis (lanes 1–3) and microsomes from cotranslations (lanes 4–6) were compared with immunoprecipitations. Complexes were precipitated from 1% Brij96 lysates of in vitro translations from CC pellets and analyzed by 12.5% SDS-PAGE under reducing conditions. Representative experiments (n = 3).

Figure 6.

Cotranslational formation of tetrameric FcɛRI complexes. (A) α,β, γ, or α,β_var, γ mRNAs were simultaneously translated into CC microsomes. Microsomal pellets were solibilized in 1% Brij96 lysis buffer and immunoprecipitations with anti-γ or control serum were performed. Stable αβγ as well as αβ_varγ complexes were retrieved by anti-γ. (B) β_var induces the accumulation of α_+sig. αβ_varγ mRNAs were cotranslated and direct loads of in vitro translations were compared with anti-α immunoprecipitates. All translated proteins with a prominent band of ∼33 kD are present in direct loads of CC mircosomes translated with αβ_varγ . Anti-α immunoprecipitation shows an overrepresentation of α_+sig when compared with translations of α alone. Proteins were precipitated from 1% Brij96 lysates of in vitro translations from CC pellets and analyzed by 12.5% SDS-PAGE under reducing conditions. Representative experiments (n = 3).

Figure 7.

FcɛRIβ_var induces the accumulation of α_1sig in vivo. (A) β_var down-regulates surface IgE-binding epitopes. β and β_var were subcloned into a bicistronic vector expressing EGFP to control for equal expression levels (pIRES2-β-EGFP and pIRES2-β_var-EGFP). 293 cells were transiently transfected and treated with proteasome inhibitor for 2 h. After SDS lysis, immunoblots with anti-β serum were preformed to demonstrate the expression of both proteins. (B) mAb 15–1 recognizes the IgE-binding epitope of FcɛRIα. IgE binding capacity of CHOαγ was assessed with biotinylated IgE (black). Preincubation with mAb 15–1 (reactivity shown in blue) inhibits subsequent IgE binding (red). ΔMean fluorescence is shown on the abscissa. (C) CHOαβγ were transiently transfected with pIRES2-β-EGFP or pIRES2-β_var-EGFP (β (red) and β_var(black)). Reactivity of mAb 15–1, which is specific for the IgE-binding epitope, was monitored on the surface of a cell population gated for EGFP-expression as a marker for expression of β and β_var. pIRES2-β_var-EGFP specifically down-regulates surface 15–1 reactivity. (D) β_var induces accumulation of unglycosylated α protein backbone in vivo. A HA-tagged version of K^b-α (α_HA) was transiently transfected into 293 cells in the presence of the indicated FcɛRI subunits. Transfection was followed by anti-HA immunoprecipitation and anti-HA immunoblotting. α_HA folds properly and becomes N-glycosylated (α_Nglyc) in the absence of other complex subunits (lane 1). Cotransfection of βγ (lane 4) or β_varγ (lane 3) induces Golgi-associated glycosylation modification of α_HA (α_mod). Unglycosylated protein backbone accumulates specifically in the presence of β_var. (E) β_var induces the accumulation of α_+sig. Metabolic labeling of cells transiently transfected with α_HA, β and γ followed by anti-HA immunoprecipitations and EndoH digestion. Most α_HA is rapidly transformed into its fully N-glycosylated modification (α_Nglyc). Susceptibility to EndoH treatment defines α_+sig.

Figure 8.

FcɛRIβ_var induces the accumulation of improperly folded FcɛRIα in vivo. (A) NH₂-terminal EGFP-fusion proteins of β or β_var (GFP-β or GFP-β_var) were transiently transfected into 293 cells with lipofectamine. The fusion adds the expected 28kD to the molecular weight but otherwise does not interfere with the molecular characteristics of β or β_var. Pulse-chase analysis: cells were analyzed untreated or pretreated with proteasome inhibitor ZL₃VS (5 μm, 2 h). Anti-GFP immunoprecipitates were obtained from 1% NP-40 lysates and analyzed by 12.5% SDS-PAGE under reducing conditions. Inhibition of the proteasome stabilizes GFP-β_var. (B) CHOαγ were transiently transfected with GFP-β and GFP-β_var and analyzed by epifluorescence. Cells were treated with ZL₃VS for 2 h, fixed and stained with mAb 15–1 to visualize IgE-binding epitopes. Staining with the anti-α polyclonal serum was performed to visualize all forms of α. Transfection with the GFP-β does not influence mAb 15–1 reactivity. Cells transfected with GFP-β_var do not show mAb 15–1 reactivity but still remain positive with the anti-α serum indicative for the presence of unfolded α chain. Representative experiment (n = 4). (C, D) Pulse-chase experiments to confirm that FcɛRIα not targeted to proteasomal degradation by β_var. FcɛRIα_HA was transiently transfected into 293 cells in the presence of βγ or β_varγ cDNA. Transfection, immunoprecipitations and analysis were performed as described in (A). α_+sig is stabilized by the presence of β_var selectively (C). (D) Comparison of FcɛRIα protein levels in αβ_varγ transfectants in the presence and absence of proteasome inhibitors. No alterations in the amount or expression pattern of FcɛRIα in induced by the inhibitor.