C/EBPβ deficiency enhances the keratinocyte innate immune response to direct activators of cytosolic pattern recognition receptors

Abstract

The skin is the first line of defense to cutaneous microbes and viruses, and epidermal keratinocytes play a critical role in preventing infection by viruses and pathogens through activation of the type I interferon (IFN) response. Using RNAseq analysis, here we report that the conditional deletion of C/EBPβ transcription factor in mouse epidermis (CKOβ mice) resulted in the upregulation of IFNβ and numerous keratinocyte interferon-stimulated genes (ISGs). The expression of cytosolic pattern recognition receptors (cPRRs), that recognize viral RNA and DNA, were significantly increased, and enriched in the RNAseq data set. cPRRs stimulate a type I IFN response that can trigger cell death to eliminate infected cells. To determine if the observed increases in cPRRs had functional consequences, we transfected CKOβ primary keratinocytes with the pathogen and viral mimics poly(I:C) (dsRNA) or poly(dA:dT) (synthetic B-DNA) that directly activate PRRs. Transfected CKOβ primary keratinocytes displayed an amplified type I IFN response which was accompanied by increased activation of IRF3, enhanced ISG expression, enhanced activation of caspase-8, caspase-3 and increased apoptosis. Our results identify C/EBPβ as a critical repressor of the keratinocyte type I IFN response, and demonstrates that the loss of C/EBPβ primes keratinocytes to the activation of cytosolic PRRs by pathogen RNA and DNA to induce cell death mediated by caspase-8 and caspase-3.

Keywords: C/EBPβ; cell death; keratinocytes; type I IFN response.

Figure 1.

CKOβ epidermis displays significant alterations in gene expression and enrichment of type I interferon pathway. (A) Immunoblot analysis for C/EBPβ and β-actin in epidermal protein lysates from Cre and CKOβ mice. (B) Total RNA was isolated from the epidermis Cre and CKOβ mice and was subjected to RNA sequencing. Row-scaled heatmap showing total gene expression, with each column representing RNA isolated from a single mouse. Left DEG column indicates statistically up-regulated (top;green, 1202 genes) and down-regulated (bottom; purple, 1384 genes) and non-significant (middle; grey) genes in CKOβ mice compared to Cre control mice (FDR < 0.1). (C) Volcano Plot of RNAseq data in CKOβ mice (FDR < 0.1). (D) Results from gene set enrichment analysis with top 20 statistically significant (FDR < 0.1) positively enriched pathways displayed. (E) Reactome pathway enrichment results (top 10) in epidermis of CKOβ compared to Cre control.

Figure 2.

Deletion of C/EBPβ enhances keratinocyte type I IFN system. (A) Heatmap of genes differentially expressed in combined Interferon α/β signaling pathway ontologies (enrichment FDR = 8.3E-8). Left DEG column indicates statistically up-regulated (top;green) and down-regulated (bottom; purple) and non-significant (middle; grey) genes in CKOβ mice compared to Cre control mice (FDR < 0.1). Each column represents RNA isolated from a single mouse. (B) TaqMAN Real-Time PCR analysis for Ifnβ expression in Cre and CKOβ mouse epidermis. (C) TaqMan Real-Time PCR analysis for ISG expression in Cre and CKOβ mouse epidermis. Data are expressed as the mean ± SD N ≥ 3 mice. * denotes p-value <0.05.

Figure 3.

Deletion of C/EBPβ enhances the innate immune response to direct activators of cytosolic PRRs. (A) Cre and CKOβ primary mouse keratinocytes in culture were transfected with 800 ng/ml poly(I:C)-Lyovec. Cells were collected 8 h post transfection. Immunoblot analysis for C/EBPβ (the doublet is LAP* 37 kDa, and LAP 35 kDa), p-IRF3, IRF3, and β-actin in poly(I:C)-Lyovec treated cells. (B) Cre and CKOβ primary mouse keratinocytes were transfected with 800 ng/ml poly(I:C)-Lyovec. Cells were collected 18 h post transfection. TaqMAN Real-time PCR analysis for Ifnβ. (C) Cre and CKOβ primary mouse keratinocytes were transfected with 1600 ng/ml poly(I:C)-Lyovec. Growth media from cultured keratinocytes was collected 18 h post transfection. Secreted IFNβ protein levels were measured from cell supernatant by ELISA. (D) Cre and CKOβ primary mouse keratinocytes were transfected with 800 ng/ml poly(I:C)-Lyovec. Cells were collected 18 h post transfection. TaqMAN Real-Time PCR analysis for various ISGs. (E) Cre and CKOβ primary mouse keratinocytes were transfected with 2 μg/ml poly(dA:dT)-Lyovec. Cells were collected 18 h post transfection. TaqMAN Real-time PCR analysis for Ifnβ. (F) Cre and CKOβ primary mouse keratinocytes were transfected with 2 μg/ml poly(dA:dT)-Lyovec. Growth media from cultured keratinocytes was collected 18 h post transfection. Secreted IFNβ protein levels were measured from cell supernatant by ELISA. (G) Cre and CKOβ primary mouse keratinocytes were transfected with 2 μg/ml poly(dA:dT)-Lyovec. Cells were collected 18 h post transfection. TaqMAN Real-Time PCR analysis for various ISGs. Data are expressed as the mean ± SD N ≥ 3. * denotes p-value <0.05 compared to similarly treated control.

Figure 4.

Deletion of C/EBPβ enhances activation of caspase-8 and apoptosis in response to direct activators of cytosolic PRRs. (A) Cre and CKOβ primary mouse keratinocytes were transfected with 1600 ng/ml poly(I:C)-Lyovec or 2 μg/ml poly(dA:dT)-Lyovec. 18 h post treatment cells were stained with annexin-V and PI and analyzed by FACS. Data are expressed as the mean + SD N > 3. * denotes p-value <0.05 compared to similarly treated control. (B) Cre and CKOβ primary mouse keratinocytes were transfected with 800 ng/ml (low dose, labeled as L) or 1600 ng/ml poly(I:C)-Lyovec (high dose, labeled as H) and collected 18 h post transfection. Immunoblot analysis for C/EBPβ, cleaved caspase-8, cleaved caspase-3 and β-actin. (C) Cre and CKOβ primary mouse keratinocytes were transfected with 1600 ng/ml poly(I:C)-Lyovec and collected at 8 h and 18 h post transfection. Immunoblot analysis for C/EBPβ, cleaved caspase-8, cleaved caspase-3 and β-actin. (D) Cre and CKOβ primary mouse keratinocytes were transfected with 1 μg/ml (low dose, labeled as L) and 2 μg/ml poly(dA:dT)-Lyovec (high dose, labeled as H) and collected at 18 h post transfection. Immunoblot analysis for C/EBPβ, cleaved caspase-8, cleaved caspase-3 and β-actin. (E) Cre and CKOβ primary mouse keratinocytes were transfected with 2 μg/ml poly(dA:dT)-Lyovec and collected at 8 h and 18 h post transfection. Immunoblot analysis for C/EBPβ, cleaved caspase-8, cleaved caspase-3 and β-actin. (F) Cre and CKOβ primary mouse keratinocytes were treated with1600 ng/ml poly(I:C) (labeled “naked”) or transfected with 1600 ng/ml poly(I:C)-Lyovec (labeled as “Lyovec”). Cells were collected 18 h post treatment. Immunoblot analysis for C/EBPβ, cleaved caspase-8, cleaved caspase-3 and β-actin.

Figure 5.

C/EBPβ-deficient normal human epidermal keratinocytes (NHEKs) display enhanced activation of caspase-8 and apoptosis in response to direct activators of cytosolic PRRs. (A) Normal human epidermal keratinocytes (NHEKs) were transfected with control or human C/EBPβ targeting siRNA. 48 h hours post transfection NHEKs were transfected with 800 ng/ml (low dose, labeled as L) or 1600 ng/ml poly(I:C)-Lyovec (high dose, labeled as H). Cells were collected at 8 h and 18 h post treatment and immunoblot analysis for human C/EBPβ (the doublet is LAP* 48 kDa, and LAP 46 kDa), cleaved caspase-8, cleaved caspase-3 and β-actin was conducted.