Fig 1.
Domain structure of naturally occurring isoforms of CIITA and N-terminal deletion mutants used in this study.
A) Domain structure of CIITA; numbering refers to CIITA-FIII (K120 isoform). B) N-terminal ends of naturally occurring forms CIITA-FI, -FIII and–FIV generated through alternative promoter usage. C) Amino acid sequence of the acidic activation domain (AAD) of CIITA-FIII (single letter code). Acidic amino acids are shown in italics. The N-terminal ends of CIITA-FIV and of N-terminal deletion mutants are indicated by arrows, and replacement of K120 by IE in the I120E121 isoform is also indicated.
Fig 2.
Steady-state protein expression levels and relative MHC-II activation potential of naturally occurring and engineered N-terminal variants of CIITA.
A) Stable transfectants of indicated CIITA-variants were generated in HEK293-EBNA cells and bulk transfectants analyzed for cell surface HLA-DR expression by flow cytometry. NPL indicates cells transfected with empty expression vector (EBS-NPL). Mean fluorescence intensities (MFI) for the different transfectants are shown. B). HLA-DRA and HLA-DRB expression of mature (spliced) mRNA was determined by RT-QPCR. mRNA expression levels (scale on left) were determined with linearized plasmid DNA as standards. HLA-DR expression of the cells used for RNA isolation is also shown (scale on right). Expression levels were normalized and expression of CIITA-FIII was set at 100 (values and standard errors from 2 independent experiments). C) Western blot analysis for CIITA protein expression of cells shown in (A). CIITA was detected with serum K22 (top). The blot was stripped and re-probed with an Hsp90-specific antiserum as loading control (bottom). D) Determination of relative transactivation potential of different CIITA forms based on MFI values of cell surface HLA-DR expression compared to CIITA protein expression levels. The activity of CIITA-FIII was set at 100. Values are derived from duplicates of flow cytometry analysis, protein extraction and western blotting. E) HeLa cells were transiently transfected with the indicated CIITA-E163 constructs and analyzed three days after transfection by flow cytometry for EGFP (FL1) and HLA-DR (FL4) expression. For each construct three different concentrations of plasmid were transfected: 0.05, 0.1, and 0.2 μg/well in 12-well plates. Cells were co-transfected with 0.2 μg/well d2tomato as a transfection marker and total DNA amount was completed to 0.7 μg DNA/well with empty expression vector. EGFP and HLA-DR expression were quantified on tomato-positive cells. All transfections were carried out in duplicate. The symbols were connected by lines only to improve clarity.
Fig 3.
Determination of protein half-life of different CIITA isoforms and mutants by western blotting after cycloheximide treatment.
A, B) HEK293-EBNA cells stably transfected with the indicated isoforms and deletion mutants of CIITA were either left untreated (0 h), or cultured for 2 h in the presence of 200 μg/ml CHX before harvesting. CIITA protein expression was analyzed by western blotting with CIITA-specific antiserum K5. Protein expression levels were determined by densitometry analysis of x-ray films and are shown below each lane. Blots were stripped and re-probed with a Hsp90-specific antiserum as loading control.
Fig 4.
The first ten amino acids of CIITA-FIII act like a portable N-terminal degron and activation domain.
A) Western blot analysis for CIITA protein expression of the indicated stable transfectants in HEK293-EBNA cells. CIITA was detected with serum K22. A long and a short exposure of the same western blot are shown. B) Protein turnover of the different CIITA forms was determined as in Fig 3. For this blot, 40 μg of protein was loaded for the constructs containing the added FIII ends (lanes 3, 4, 7, 8, 11, 12) and 20 μg for those without (lanes 1, 2, 5, 6, 9, 10). C) The relative transactivation potential of different CIITA forms was determined as in Fig 2D. The activities of ∆36, ∆54, and ∆102 were set at 100. Values are derived from three experiments.
Fig 5.
Degradation and transactivation potential of the N-terminal end of CIITA-FIII.
A) The N-terminal end of CIITA-FIII reduces CIITA protein expression while increasing MHC-II activation. HeLa cells were transiently transfected with the indicated CIITA-E163 constructs and analyzed three days after transfection by flow cytometry for EGFP (FL1) and HLA-DR (FL4) expression as shown for Fig 2E. CIITA MRC is a construct expressing CIITA isoform III without EGFP tag. B) CIITA-FIII1-10 destabilizes a heterologous protein. pEYFP-N1 (lanes 1–3) or pEYFP-Nuc (3 x SV40-NLS; lanes 4–6) constructs containing N-terminal extensions of the first ten amino acids of CIITA-FIII (FIII-pEYFP, lanes 2, 5), or a CIITA-FIII N-terminal extension with the second and third amino acids (Arg2, Cys3) replaced by alanine and glycine (FIII-MAG-pEYFP) were transiently transfected into HeLa cells. Shown is the residual EYFP fluorescence after 4h of CHX (40 μg/ml) as determined by flow cytometry. The fluorescence intensity at time point 0h was set as 100%. Lanes 7 and 8 show pEGFP-N1 and pd2EGFP as controls. Experiments were carried out in duplicates. Statistical analysis; student t-test; * p < 0.05; ** p < 0.005.
Fig 6.
N-terminal ends of NTU proteins increase turnover and transactivation of ∆36.
A) N-terminal end sequences of the different constructs used. Arginine (R) and proline (P) residues in the added N-terminal ends are underlined. B) Western blot analysis for CIITA protein expression of stable transfectants in HEK293-EBNA of the indicated constructs. CIITA was detected with serum K5. C) Protein turnover of the different CIITA forms was determined as in Fig 3.
Fig 7.
Amino acid analysis of N-terminal degradation sequences.
A) N-terminal end sequences of mutated constructs based on full length CIITA-FIII (upper) or ∆36 with N-terminal extensions (lower). Mutated amino acid residues are underlined. B) Western blot analysis for CIITA protein expression of stable transfectants of the indicated constructs in HEK293-EBNA cells. CIITA was detected with serum K5. C, D) Protein turnover of the different CIITA forms was determined as in Fig 3.
Fig 8.
CIITA-FIII interacts more efficiently with protein partners.
CIITA was immunoprecipitated from protein extracts of HEK293-EBNA cells stably transfected with empty EBS-NPL vector (lanes 1, 4), CIITA-FIII (lanes, 2, 5), or CIITA-∆36 (lanes 3, 6) respectively. Input controls (lanes 1–3) or immunoprecipitated material (lanes 4–6) were separated by SDS-PAGE (8% gel), blotted and analyzed by western blotting. The membrane was cut in half and the upper part was probed with antibodies for CIITA (A), stripped, and reprobed consecutively with antibodies for p300/p400 (antibody RW144) (B), RFX (D), and Hsp90 as a control (F), the lower part was hybridized with antibodies against TBP (C), stripped and reprobed for S8 using Reliablot secondary reagents (E). For input controls longer exposures are shown, with the exception of Hsp90. Ratios of band intensities of bands in lane 5 versus lane 6 are shown on the right.
Fig 9.
Knock out of p300/CBP affects expression of CIITA independently from the FIII N-terminal end.
MEF-p300fl/fl-CBPfl/fl cells were infected either with a Cre-recombinase expressing lentivirus (Cre; lanes 1, 3, 5, 7, 7, 9) or with empty lentivirus vector (pLKO; lanes 2, 4, 6, 8, 10). One day after infection, the cells were transiently transfected with EBS-NPL empty vector (lanes 1, 2), CIITA-FIII-E163 (lanes 3, 4), CIITA-FIII-MRC (lanes 5, 6), CIITA-∆36-E163 (lanes 7, 8), and CIITA-FIII-∆36-E163 and analyzed three days after transfection. Steady state protein expression levels were analyzed by SDS-PAGE followed by western blotting. The blots are from a single membrane, which was cut into horizontal strips, and show immunoblotting with antibodies against p300 (A), CIITA (B), and Hsp90 (C).