Revisiting the specificity of the MHC class II transactivator CIITA in classical murine dendritic cells in vivo

Abstract

Ciita was discovered for its role in regulating transcription of major histocompatibility complex class II (MHCII) genes. Subsequently, CIITA was predicted to control many other genes based on reporter and ChIP-seq analysis but few such predictions have been verified in vivo using Ciita^-/- mice. Testing these predictions for classical dendritic cells (cDCs) has been particularly difficult, since Ciita^-/- mice lack MHCII expression required to identify cDCs. However, recent identification of the cDC-specific transcription factor Zbtb46 allows the identification of cDCs independently of MHCII expression. We crossed Zbtb46^gfp mice onto the Ciita^-/- background and found that all cDC lineages developed in vivo in the absence of Ciita. We then compared the complete transcriptional profile of wild-type and Ciita^-/- cDCs to define the physiological footprint of CIITA for both immature and activated cDCs. We find that CIITA exerts a highly restricted control over only the MHCII, H2-DO and H2-DM genes, in DC1 and DC2 cDC subsets, but not over other proposed targets, including Ii. These findings emphasize the caveats needed in interpreting transcription factor binding sites identified by in-vitro reporter analysis, or by ChIP-seq, which may not necessarily indicate their functional activity in vivo.

Keywords: Ciita; Dendritic cell; Transcription factor; Zbtb46.

Fig. 1. BM cDC progenitors can be identified without MHCII in Zbtb46^+/GFP mice and their development is CIITA-independent

(A) Flow cytometry analysis of BM cDC progenitors from Ciita^+/+Zbtb46^+/GFP and Ciita^−/−Zbtb46^+/GFP mice defined as Lin⁻SiglechH⁻ CD11c⁺CD135⁺CD115⁺CD117⁻Zbtb46^gfp+ and Lin⁻SiglechH⁻CD11c⁺CD135⁺CD115⁻CD117^int Zbtb46^gfp+ for pre-CD4 and pre-CD8 DCs, respectively (n = 5). Contour plots are from a single experiment representative of 2 experiments with 2–3 mice per experiment. (B) Frequency of pre-CD4 and pre-CD8 DCs in vivo as defined in. Percent for pre-CD4 DC was based on a pre-gate defined as Lin⁻SiglechH⁻CD11c⁺CD135⁺CD115⁺cells. Percent for the pre-CD8 DC was based on a pre-gate defined as SiglechH⁻CD11c⁺CD135⁺CD115⁻cells. Data are shown as a dot plot with mean bar for two pooled experiments with 2–3 mice per experiment (n = 5). (C) Output of sorted pre-CD4 and pre-CD8 DC progenitors as defined in (A) after 5 days of Flt3L-treated cultures. Shown are contour plots from a single experiment pre-gated as B220⁻SiglecH⁻CD11c⁺Zbtb46^gfp+ representative of 2 experiments with 2–3 mice per experiment (n = 5).

Fig. 2. Splenic CD172a⁺ and CD24⁺ DCs can be identified without MHCII as a linage marker in Zbtb46^+/GFP mice and develop in Ciita-deficient mice

(A) Flow cytometry analysis of splenic CD172a⁺ and CD24⁺ DCs from Zbtb46^+/gfp mice using either B220⁻SiglecH⁻CD11c⁺MHCII⁺ or B220⁻SiglecH⁻CD11c⁺Zbtb46^gfp+ pre-gates. Contour plots on the left are from a single, representative experiment, and dot plots on the right pool 5 experiments with 1–3 mice per experiment with mean bars (n = 8–10). (B) Flow cytometry analysis of splenic cDC populations in Ciita^+/+Zbtb46^+/GFP and Ciita^−/− Zbtb46^+/gfp mice with populations using B220⁻SiglecH⁻ CD11c⁺Zbtb46^gfp+ pre-gates. Contour plots on the left are from a single, representative experiment and dot plots on the right pool 5 experiments with 1–3 mice per experiment (n = 7–10).

Fig. 3. CIITA has a restricted transcriptional footprint in cDCs

(A) Volcano plot showing differentially expressed genes in homeostatic splenic CD24⁺ (left) and CD172a⁺ (right) DCs sorted from Ciita^+/+Zbtb46^+/gfp and Ciita^−/−Zbtb46^+/gfp mice. (B) Volcano plot showing differentially expressed genes in CD24⁺ (left) and CD172a⁺ (right) DCs from whole BM Flt3L cultures derived from Ciita^+/+Zbtb46^+/gfp and Ciita^−/−Zbtb46^+/gfp mice and activated with IFN-γ and LPS. In (A,B) fold-change is derived from the mean difference in Log₂ expression levels and P-values are derived from Welch’s t-test executed with Log₂ expression levels of two biological replicates from one experiment (n = 2). (C) Heat map relative expression levels of select differentially expressed genes from (A) and (B). (D) Heat map of Log₂ expression levels of selected previously reported CIITA target genes from (A) and (B).