Bik promotes proteasomal degradation to control low-grade inflammation

Abstract

Although chronic low-grade inflammation does not cause immediate clinical symptoms, over the longer term, it can enhance other insults or age-dependent damage to organ systems and thereby contribute to age-related disorders, such as respiratory disorders, heart disease, metabolic disorders, autoimmunity, and cancer. However, the molecular mechanisms governing low-level inflammation are largely unknown. We discovered that Bcl-2-interacting killer (Bik) deficiency causes low-level inflammation even at baseline and the development of spontaneous emphysema in female but not male mice. Similarly, a single nucleotide polymorphism that reduced Bik levels was associated with increased inflammation and enhanced decline in lung function in humans. Transgenic expression of Bik in the airways of Bik-deficient mice inhibited allergen- or LPS-induced lung inflammation and reversed emphysema in female mice. Bik deficiency increased nuclear but not cytosolic p65 levels because Bik, by modifying the BH4 domain of Bcl-2, interacted with regulatory particle non-ATPase 1 (RPN1) and RPN2 and enhanced proteasomal degradation of nuclear proteins. Bik deficiency increased inflammation primarily in females because Bcl-2 and Bik levels were reduced in lung tissues and airway cells of female compared with male mice. Therefore, controlling low-grade inflammation by modifying the unappreciated role of Bik and Bcl-2 in facilitating proteasomal degradation of nuclear proteins may be crucial in treating chronic age-related diseases.

Keywords: Apoptosis; Inflammation; NF-kappaB; Peptides.

Figure 1. Loss of Bik in mice causes the development of emphysema.

(A) Total RNA or protein was isolated from the cranial lobes of the lungs of 4-week-old mice, and changes in the basal expression levels of IL-6 and KC were analyzed by quantitative reverse-transcription PCR (qRT-PCR). n = 3/group. (B) The levels of IL-1α, MIP-1α, and (C) VEGF were analyzed in the lung homogenates of WT and bik^–/– mice at 80 weeks of age by Luminex. n = 9/group. (D) Baseline weighted mean alveolar volume in bik^+/+ and old bik^–/– mice at 13 to 44 weeks of age. n = 8–11; N = 2 (n, sample size in a single experiment; N, number of experimental repeats). (E) Mean alveolar chord length in 4-week-old female bik^+/+ and bik^/ mice. n = 3, N = 3. (F) Female CCSP-rtTA Bik transgenic mice and their littermates were kept with Dox water (400 mg/l) until 25 weeks of age. n = 10–13; N = 3. Mean alveolar chord length of lung tissues was measured in 10 randomly selected lung fields per mouse, using Visiopharm software. (G) MAECs isolated from female bik^+/+ and bik^–/– mice were plated on 6-well plates. Media were changed to rock inhibitor-/FBS-free media and then cells harvested 6 and 24 hours later for mRNA analysis. The mRNA expression levels of the indicated inflammatory cytokines were analyzed using qRT-PCR from n = 3 wells/group. (H) Media from bik^+/+ and bik^–/– MAECs plated on 6-well plates were analyzed for levels of IL-1α, IL-6, and KC using Luminex. n = 3 wells/group. The sample size in a single experiment is denoted as n; the number of experimental replicates is denoted as N. Two-tailed Student’s t test was used to compare between 2 groups, and grouped results were analyzed using 2-way ANOVA. Data are represented as mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001.

Figure 2. A SNP affects BIK promoter activity and expression

(A) AA of rs738276 demonstrates higher expression than GG (P = 2 × 10^–11) in lymphoblastoid cells. (B) Bik mRNA in 40 HAECs from current or former smokers (11 AA, 20 AG, 9 GG). (C) HAECs (n = 23) differentiated and treated with 50 ng/ml IFN-γ. Bik was increased in Bik AA compared with Bik GG HAECs (P = 0.002). Western blot analysis of Bik protein in Bik AA and Bik GG HAECs. (D and E) p750A-SA or p750G-SA constructs were transfected into (D) HAECs and (E) H1299 cells, and BIK gene promoter activity of A normalized to G values. n = 3; N = 3. (F) IRF-1 has a higher affinity to p750G compared with p750A. Nuclear extracts from H1299 cells were subjected to EMSA using biotin-labeled p750A or p750G oligonucleotide probes of Bik and IRF-1 genes. (G) Primary HAECs with Bik AA and Bik GG genotypes were processed for ChIP assays by using anti–IRF-1 or control IgG (negative control). Precipitated DNA was subjected to PCR to amplify the Bik gene fragment. (H) Protein lysates from WT and IRF-CRISPR knockout HAECs subjected to Western blotting. (I) Cytosolic and nuclear fractions of BikAA and BikGG HAECs analyzed for expression of indicated proteins by Western blotting. (J) Plasma samples from individuals with Bik AA and Bik GG genotypes analyzed using Luminex. In each group, 40 samples were analyzed, and in many samples, these cytokines were not detected. (K) Metaanalysis of increased decline in lung function associated with G allele in study participants of more than 60 years of age in ECLIPSE, LSC, COPDGene, and FHS. Additive model compared GG, GA, and AA genotypes of BIK SNP rs738276 using GG as a reference group. Two-tailed Student’s t test compared between 2 groups, and grouped results were analyzed using 2-way ANOVA. Data are represented as mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001.

Figure 3. Bik suppresses LPS and allergen-induced inflammation in mice.

(A) bik^+/+ and bik^–/– mice were instilled with 5 μg LPS intranasally, and neutrophil numbers were compared in the BAL fluid 4 hours later. n = 5/group. (B) TetoBik^– and tetoBik⁺ mice were put on Dox diet for 48 hours, and the expression of Bik was analyzed in the lung tissues by qRT-PCR or immunostaining. Bik mRNA and protein levels were shown by qRT-PCR and immunostaining, respectively. (C) Baseline macrophage levels in the BAL fluid of male and female CCSP-rtTA⁺TetOBik⁺ and CCSP-rtTA⁺TetOBik^– mice, 2 male and 3 female in TetOBik^– and 4 male and 3 female in CCSP-TetOBik^– per group. (D) TetoBik^– and TetoBik⁺ mice were placed on Dox diet for 48 hours and subsequently instilled with 50 μg LPS and lungs harvested; BAL fluid neutrophil numbers were quantified 4 hours later (n = 7/group 4 males and 3 female). (E) Female bik^–/– mice were instilled once daily over 2 days with 10⁸ virus particles of Ad-Bik or Ad-GFP, as control, diluted in final volume of 50 μL in PBS. Twenty-four hours later, mice were intranasally instilled with 5 μg LPS and euthanized and BAL fluid neutrophil numbers quantified 24 hours later. n = 4/group. (F) TetoBik^– and TetoBik⁺ mice were placed on Dox diet for 48 hours and subsequently instilled with 10 μg HDM for 5 days (n = 5/group). BAL cell numbers were analyzed 5 days later. Two-tailed Student’s t test was used to compare between 2 groups, and grouped results were analyzed using 2-way ANOVA. Data are represented as mean ± SEM. *P < 0.05; **P < 0.01.

Figure 4. The BH3 domain of Bik inhibits nuclear p65–induced transcriptional activity.

A549 NF-κB luciferase reporter cells were infected with either Ad GFP, Ad-Bik, or Ad-Bik^L61G (mutant Bik that does not kill cells) and subsequently treated with 10 ng/ml TNF-α for 6 hours. (A) Protein lysates were analyzed for the expression of Bik by Western blotting. (B) Percentages of viable cells were analyzed by trypan blue exclusion assay. n = 3. Cells were infected with the adenoviral vectors at MOIs indicated below the bar. (C) NF-κB transcriptional activity was analyzed in the cell lysates. n = 4. (D) Percentages of cells expressing nuclear p65 were analyzed by immunofluorescent staining. n = 6–8; N = 2. (E) A549 NF-κB luciferase reporter cells were treated with vehicle, TAT, or TAT-Bik^L61G peptides for 2 hours and subsequently treated with TNF-α for 6 hours. NF-κB transcriptional activity was analyzed in the cell lysates. n = 3. (F) A549 cells were treated with 10 μM control peptides or Bik^L61G peptides for 2 hours, followed by treatments with 10 ng/ml TNF-α for the indicated times in the presence of phosphormide. Cell lysates were analyzed for the level of IκBα protein by Western blotting. (G) Cytosolic-protein lysates from bik^+/+ and bik^–/– MAECs were analyzed for levels of phospho- and total IκBα by Western blotting. (H) Cytosolic lysates from bik^+/+ and bik^–/– MAECs were immunoprecipitated using anti-p65 antibody and analyzed for IκBα levels by Western blotting. (I) Cytosolic and nuclear fractions of lysates from bik^+/+ and bik^–/– MAECs were analyzed for levels of p65 and IκBα by Western blotting. (J) Nuclear fractions of lysates isolated from bik^+/+ and bik^–/– MAECs analyzed for levels of p65 and p50 by Western blotting. (K) HAECs transfected with siControl or IRF-1 siRNA. Western blot of nuclear lysates for Bik and p65 protein. (L) bik^–/– mice instilled with 50 μg HDM intranasally daily for 5 consecutive days and on days 6 and 7 intranasally treated with 10 μM of control TAT peptide, BH3 WT Bik peptide, or BH3 mutant Bik peptide. BAL fluids analyzed for inflammatory cell numbers. n = 4–6/group. Two-tailed Student’s t test was used to compare between 2 groups, and grouped results were analyzed using 2-way ANOVA. Data are represented as mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001.

Figure 5. Bik, localized to nuclear domains of the cell, reduces nuclear p65 by binding to Bcl-2.

(A) Primary HAECs and A549 cells were immunostained for Bik protein and analyzed by fluorescent microscopy for the localization of Bik protein. Original magnification, ×1,200. (B) Cytosolic and nuclear extracts from HEK293T and HAECs were analyzed for the localization of p65, Bik, and Bcl-2 by Western blotting. (C) HEK293T cells were transfected with EV or Bcl-2 plasmid, and protein lysates were analyzed for Bcl-2 and p65 levels by Western blotting. (D) WT and CRISPR/cas9 Bik-knockout HEK293T cells were transfected with plasmids expressing WT Bcl-2. Forty-eight hours later, nuclear lysates were analyzed for levels of p65 and Bcl-2 by Western blotting. The bar graph on the right-hand side shows densitometry (p65 fold change) of the Western blot from 3 independent experiments. (E) CRISPR/Cas9 Bcl-2–knockout cells were transfected with EV, WT, or mutant triple–flag-tagged Bcl-2. Forty-eight hours later, nuclear lysates were analyzed for the expression levels of p65 using Western blotting, nuclear lysates were immunoprecipitated using anti-p65 or Bcl-2 antibodies, and the Ips were probed for flag–Bcl-2 and p65. (F) Bik CRISPR/Cas9 knockout cells were transfected with EV or plasmids expressing WT or phosphorylation-mutant Bik constructs. Forty-eight hours later, cytosolic and nuclear lysates were analyzed for p65 levels by Western blotting. Two-tailed Student’s t test was used to compare between 2 groups, and grouped results were analyzed using 2-way ANOVA. Data are represented as mean ± SEM. *P < 0.05.

Figure 6. Bik through Bcl-2 interacts with the S19 regulatory particle components of the proteasome and regulates protein degradation.

(A) A549 NF-κB luciferase reporter cells treated with ABT-263 and assessed for NF-κB transcriptional activity 24 hours later (n = 3). (B) A549 cells expressing NF-κB luciferase reporter treated with vehicle or 10 nM ABT-263 for 24 hours, followed by treatments with sham or 10 ng/ml TNF-α for 6 hours, and analyzed by immunostaining. n = 6; N = 2. (C) bik^–/– MAECs treated with mock or 10 nM ABT-263 for 24 hours. Nuclear fractions of the cell lysates probed for p65 on Western blots. (D) Bcl-2 CRISPR/Cas9 knockout HEK293T cells transfected with 0.5 μg EV or Bcl-2 plasmids and 48 hours later treated with vehicle or 20 μM MG132 for 4 hours. Nuclear fractions probed for p65, β-catenin, and lamin. (E) Bik^–/– MAECs treated with mock or 10 nM ABT-263 for 24 hours followed by treatment with vehicle or 20 μM MG132 for 4 hours. Nuclear fraction probed for p65, β-catenin, and lamin A/C. (F) Bcl-2 CRISPR/Cas9 knockout HEK293T cells transfected with EV or flag–Bcl-2 and treated with vehicle or 20 μM MG132 for 4 hours. Nuclear lysates immunoprecipitated 4 hours later using anti-p65 antibody and the IPs analyzed for p65 ubiquitination using antiubiquitin. (G) Bcl-2 CRISPR/Cas9 knockout HEK293T cells transfected with Bcl-2 plasmid for 48 hours and treated with vehicle or 20 μM MG132, 10 μM MS-873, 10 μM bafilomycin, or 10 μM chloroquine for 4 hours. Nuclear lysates were analyzed for levels of p-65. (H) Volcano plot from Bcl-2 immunoprecipitates of nuclear lysates from bik^+/+ and bik^–/– MAECs. Blue dots indicate proteins enriched in bik^+/+ MAECs while red dots indicate proteins increased in bik^–/– MAECs; black dots indicate no change between bik^+/+ and bik^–/– MAECs. (I) Cytosolic and nuclear fractions of bik^+/+ and bik^–/– MAECs analyzed for p65, RPN1, RPN2, and Bik levels by Western blotting and immunoprecipitates using anti–Bcl-2 were analyzed for RPN1 and RPN2 levels by Western blotting. (J) bik^+/+ and bik^–/– MAECs grown on cell-culture chambers and analyzed for localization of RPN1 by confocal microscopy. Two-tailed Student’s t test was used to compare between 2 groups, and grouped results were analyzed using 2-way ANOVA. Data are represented as mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001. Original magnification, ×1,500 (B); ×500 (J).

Figure 7. Bik and Bcl-2 levels are reduced in female airway cells.

Lung tissues from male and female C57BL/6 mice were analyzed for (A) Bcl-2, (B) Bik mRNA, and (C) Bik and Bcl-2 protein levels by qRT-PCR and Western blotting. (D) Bcl-2 mRNA levels in male and female bik^–/– MAECs. Bcl-2 mRNA levels in (E) male and (F) female bik^–/– MAECs treated with vehicle or 10 nM estradiol and (G) in male or (H) female bik^–/– MAECs treated with vehicle or 10 nM 5αDHT. Bik mRNA levels in male and female WT MAECs (I). Bik mRNA levels in (J) male and (K) female bik^–/– MAECs treated with vehicle or 10 nM estradiol and in (L) male or (M) female bik^–/– MAECs treated with vehicle or 10 nM 5αDHT. mRNA levels were analyzed by qRT-PCR. n = 3–9/group; N = 2. (N) Bcl-2 and Bik protein levels in male and female bik^+/+ MAECs analyzed from protein extracts by Western blotting. Two-tailed Student’s t test was used to compare between 2 groups, and grouped results were analyzed using 2-way ANOVA. Data are represented as mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001.

Figure 8. Schematic showing that Bcl-2 anchored to the nuclear membrane by interacting with Bik.

Bik modifies the BH4 domain of Bcl-2 to interact with RPN1 and RPN2 and increases the activity of the nuclear proteasome to degrade nuclear p65.