BRD7 as key factor in PBAF complex assembly and CD8⁺ T cell differentiation

Upon infection, naive CD8⁺ T cells differentiate into cytotoxic effector cells to eliminate the pathogen-infected cells. Although many mechanisms underlying this process have been demonstrated, the regulatory role of chromatin remodeling system in this process remains largely unknown. Here we show that BRD7, a component of the polybromo-associated BAF complex (PBAF), was required for naive CD8⁺ T cells to differentiate into functional short-lived effector cells (SLECs) in response to acute infections caused by influenza virus or lymphocytic choriomeningitis virus (LCMV). BRD7 deficiency in CD8⁺ T cells resulted in profound defects in effector population and functions, thereby impairing viral clearance and host recovery. Further mechanical studies indicate that the expression of BRD7 significantly turned to high from naive CD8⁺ T cells to effector cells, which bridged BRG1 and PBRM1 to the core module of PBAF complex, consequently facilitating the assembly of PBAF complex rather than BAF complex in the effector cells. The PBAF complex changed the chromatin accessibility at the loci of Tbx21 gene and upregulated its expression, leading to the maturation of effector T cells. Our research demonstrates that BRD7 and the PBAF complex are key in CD8⁺ T cell development and present a significant target for advancing immune therapies.

Increased expression of BRD7 in the effector CD8⁺ T cells. The SWI/SNF complexes are involved in genome-wide transcriptional regulation (16, 17). However, their roles in antiviral immune response remain unknown. To elucidate the roles of SWI/SNF chromatin remodeling complexes in regulating CD8⁺ T cell responses, we analyzed the expression of components of BAF, PBAF, and ncBAF complexes after influenza virus infection. Among these complex components, we found that the expression of BRD7 was the highest in effector CD8⁺ T cells at day 10 postinfection (p.i.), and the expression maintained high in memory CD8⁺ T cells (at ~42 days p.i.) from OT-I TCR-transgenic mice when compared with their mRNA expression in naive CD8⁺ T cells (Figure 1A). In addition, when we analyzed the kinetics of mRNA expression of several PBAF components during in vivo responses to infection, we found that Brd7 mRNA expression was upregulated at day 6 p.i. and reached the maximum at day 8 (Figure 1B), suggesting that the expression of BRD7 is concurrent with the development of effector CD8⁺ T cells. Of note, the CD8⁺ T cells of these mice recognize a peptide fragment of chicken ovalbumin (OVAp) bound to H-2K^b when infected with PR8-OVA virus, a recombinant strain of A/PR/8/34 (H1N1) virus expressing OVA. These data therefore indicate that BRD7 in CD8⁺ T cells may play a critical role in regulating immune response.

Figure 1

BRD7 is upregulated in effector CD8⁺ T cells. (A) Differential expression of relative mRNAs in naive, effector, and memory OT-I CD8⁺ T cells was analyzed. Naive OT-I cells purified from spleen of OT-I mice were transferred into recipient CD45.1 mice and then infected with PR8-OVA at day 1 after transfer. Effector (10 days [d] p.i.) and memory (~42 d p.i.) OT-I cells were purified with FACS for mRNA quantification relative to that of naive OT-I cells. (B) Time course of mRNA expression of several PBAF components in OT-I T cells purified from spleen of OT-I mice, which were transferred into recipient CD45.1 mice and then infected with PR8-OVA at day 1 after transfer. **P < 0.01, ***P < 0.001, and ****P < 0.0001 (2-way ANOVA). Mean ± SEM of 3 mice per group. Data are representative of 2 independent experiments.

In order to examine the role of BRD7 in CD8⁺ T cell response, we generated mice that conditionally deleted Brd7 in T cell–specific conditional BRD7-deleted mice (Brd7^ΔT) by crossing mice homozygous for loxP-flanked alleles of Brd7 (Brd7^fl/fl) to mice expressing a transgenic encoding Cre recombinase from T cell–specific Cd4 promoter (Cd4-Cre) (Supplemental Figure 1; supplemental material available online with this article; https://doi.org/10.1172/jci.insight.171605DS1); this resulted in lacking expression of BRD7 in mature CD8⁺ T cells (Supplemental Figure 2A). The thymic T cell development and T cell circulation were not affected in Brd7^ΔT mice since similar percentages and absolute numbers of CD4⁺ and CD8⁺ T cells were found in thymus, mediastinal lymph nodes (mLN), and spleen from both BRD7 WT mice (Brd7^fl/fl) and BRD7-deleted littermate pairs (Brd7^ΔT) (Supplemental Figure 2, B and C). To explore the potential for any off-target effects on related Brd family proteins, we carried out Western blotting to scrutinize their expression. The findings revealed that, with the exception of Brd7, the expression of Brd2, Brd4, Brg1, and Trim-28 remained unaltered in the BRD7-deficient mice (Supplemental Figure 3). Thus, this T cell–specific Brd7-deleted mice allowed us to specifically analyze the role of BRD7 in CD8⁺ T cell responses during infection.

BRD7 deficiency impairs the differentiation of effector CD8⁺ T cells. To examine the response of CD8⁺ T cell subsets of BRD7-sufficient and BRD7-deficient mice during virus infection, we infected Brd7^fl/fl mice and Brd7^ΔT mice with the influenza HKx31 strain and analyzed the CD8⁺ T cells in the spleen and lungs at the peak of the response (day 10 p.i.). We analyzed influenza antigen–specific CD8⁺ T cell responses by staining the H-2D^b-NP (NP_366–374) and H-2D^b-PA (PA_224–233) tetramers. There were no obvious differences in the percentages and numbers of the influenza-specific H-2D^b-NP⁺CD8⁺ and H-2D^b-PA⁺CD8⁺ T cells in the spleen and lungs from Brd7^fl/fl mice and Brd7^ΔT mice (Figure 2, A–C). However, the frequency of antigen-specific KLRG1⁺IL-7R^– SLECs was significantly lower in BRD7-deficient CD8⁺ T cells than in BRD7 WT CD8⁺ T cells (Figure 2, D and E). This finding was not due to tissue distribution, since similar results were found in both spleen (Supplemental Figure 4A) and the lung (Figure 2, D and E). Antigen-specific BRD7-deficient CD8⁺ T cells also showed increased expression of CD62L and CD27 (Supplemental Figure 4B), a defective SLEC differentiation phenotype (7, 39). This phenotype was not due to abnormal activation of CD8⁺ T cells, since the expression of CD69 and CD44 were not altered in BRD7-deficient CD8⁺ T cells (Supplemental Figure 4B). Likewise, to investigate the role of BRD7 in a secondary effector CD8⁺ T cell response, we used a 2-infection model in our study. In the primary infection, mice were infected i.n. with the influenza PR8 virus (H1N1 virus). In the secondary infection, mice were rechallenged with the HKx31 virus i.n. 6 weeks after the first infection. We found that the frequency of NP⁺CD8⁺ T cells in the lung and spleen was similar between Brd7^fl/fl mice and Brd7^ΔT mice (Figure 2F), while a reduction of antigen-specific KLRG1⁺IL-7R^– SLECs was found in the Brd7^ΔT mice after infection with HKx31 viruses of previously PR8-primed mice (Figure 2G). These results indicate a defect in the primary and secondary effector differentiation of CD8⁺ T cells without BRD7.

Figure 2

BRD7 controls SLEC differentiation. (A) Flow cytometry of influenza H-2D^b-NP⁺CD8⁺ and H-2D^b-PA⁺CD8⁺ T cells from spleen and lungs of BRD7 WT (Brd7^fl/fl) (n = 4) and BRD7-deficient (Brd7^ΔT) (n = 3) mice. Mice were infected with influenza HKx31 virus, and NP366 and PA244 tetramer were stained in lung or spleen CD8⁺ T cells at 10 d p.i. (B and C) Frequency and cell number of H-2D^b-NP⁺ or H-2D^b-PA⁺ cells among CD8⁺ T cells in A. (D and E) Flow cytometry of KLRG1 and CD127 on H-2D^b-NP⁺CD8⁺ T cells in the lungs obtained from Brd7^fl/fl (n = 4) or Brd7^ΔT (n = 3) mice infected with HKx31 at 10 d p.i. (F) Flow cytometry of H-2D^b-NP⁺CD8⁺ T cells from spleen and lungs of PR8-primed Brd7^fl/fl (n = 4) and Brd7^ΔT (n = 4) mice rechallenged with HKx31 at 6 weeks after PR8 infection. (G) Flow cytometry of KLRG1 and CD127 on H-2D^b-NP⁺CD8⁺ T cells in the spleen obtained of PR8-primed Brd7^fl/fl (n = 4) and Brd7^ΔT (n = 4) mice rechallenged with HKx31 at 6 weeks after PR8 infection. (H and I) Flow cytometry of KLRG1 and CD45.1 on NP⁺CD8⁺ T cells from chimeras. Mixed bone marrow chimeras were generated by reconstitution of lethally irradiated CD45.1 mice with bone marrow from Brd7^fl/fl (CD45.2⁺) plus Brd7^fl/fl (CD45.1⁺CD45.2⁺) (n = 3), or Brd7^ΔT (CD45.2⁺) plus Brd7^fl/fl (CD45.1⁺CD45.2⁺) (n = 3) at a ratio of 1:1. Chimeras were infected with HKx31 at 6 weeks after reconstitution and were sacrificed at 10 d p.i. for analysis. (J–M) Flow cytometry of OT-I CD8⁺ T cells (J and K) or KLRG1⁺CD127^– SLECs and KLRG1^–CD127⁺ MPECs (L and M) in CD45.1 host mice (n = 4) 9 days after transfer of Brd7^fl/fl (CD45.1⁺CD45.2⁺) plus Brd7^ΔT (CD45.2⁺) OT-I CD8⁺ T cells at a ratio of 1:1 and infection with PR8-OVA virus 1 day after transfer. Data are shown as mean ± SEM. **P < 0.01 (2-tailed Student’s t test). Data are representative of 3 independent experiments.

To explore whether the defective differentiation of SLECs in BRD7-deficient mice depended on the direct role of BRD7 in CD8⁺ T cells, we generated chimeras by reconstituting lethally irradiated CD45.1 mice with a various 1:1 mixture of bone marrow cells of Brd7^fl/fl mice and Brd7^ΔT mice, followed by infection with HKx31 at 6 week after reconstitution. The frequency of antigen-specific effector CD8⁺ T cells was analyzed in the spleen of chimeras. We found that the BRD7-deficient CD8⁺ T cells produced much lower antigen-specific KLRG1⁺ effector cells than BRD7 WT CD8⁺ T cells, even in the presence of WT CD8⁺ T cells (Figure 2, H and I), and this suggested that BRD7 influenced the differentiation of SLECs depending on the intrinsic role of BRD7 in CD8⁺ T cells. Furthermore, to rule out the possibility that the deletion of BRD7 in CD4⁺ T cells may affect the phenotype, we crossed BRD7-deficient mice with mice of the OT-I TCR-transgenic strain to generate OT-I BRD7-deficient mice. The mixture of OT-I BRD7-deficient naive CD8⁺ T cells (CD45.2⁺) and OT-I BRD7 WT naive CD8⁺ T cells (CD45.1⁺CD45.2⁺) at a ratio of 1:1 was adoptively transferred into CD45.1 mice, followed by infection with PR8-OVA viruses after 24 hours. We found that OT-I BRD7-deficient CD8⁺ T cells failed to generate as many SLECs as OT-I BRD7 WT CD8⁺ T cells did (Figure 2, J–M). Altogether, these results show that BRD7 regulated the differentiation in a CD8⁺ T cell–intrinsic manner.

BRD7 is required for the effector function of CD8⁺ T cells. After exposure to antigen, activated CD8⁺ T cells undergo an effector differentiation process and gain the functional ability to produce cytotoxic and effector cytokines to eliminate intracellular pathogens. To examine whether BRD7 deficiency affects the effector functions, we measured cytokine production from CD8⁺ T cells after NP-peptide stimulation by intracellular staining and flow cytometric analysis. The proportion of BRD7-deficient CD8⁺ T cells producing IFN-γ (Figure 3A) and TNF-α (Figure 3B) was significantly decreased. Likewise, the production of the cytolytic effector molecule GzmB (Figure 3C) and perforin (Figure 3D) was hardly detected in antigen-specific BRD7-deficient CD8⁺ T cells. Low production of these cytolytic effector cytokines will result in a defect in the ability to kill targets. Indeed, BRD7-deficient CD8⁺ T cells showed a diminished cytolytic effector function in an in vivo cytolysis assay (Figure 3, E and F). Of note, in the in vivo cytolysis experiment, we stained splenocytes stimulated with NP peptide by low-dose carboxyfluorescein diacetate succinimidyl ester (CFSE^lo cells) and splenocytes without NP peptide stimulation by high-dose CFSE (CFSE^hi cells). Then, we mixed CFSE^lo cells and CFSE^hi cells at a ratio of 1:1 and injected the mixture into WT mice that were not infected with HKx31, WT mice infected with HKx31, or BRD7-deficient mice infected with HKx31 separately. Almost complete clearance of NP-pulsed cells was observed in WT mice infected with HKx31, whereas the killing of these cells was less effective in BRD7-deficient mice infected with HKx31. Similarly, Brd7^ΔT mice showed a higher degree of weight loss (Figure 3G) and more severe pathology in the lung than Brd7^fl/fl mice did after infection with the aggressive influenza PR8 viruses (Figure 3H). Taken together, these data suggest that BRD7 was required for the production of functional CD8⁺ effector T cells and the clearance of viruses.

Figure 3

BRD7 is required for effector function. (A and B) Intracellular cytokine staining of IFN-γ (A) or TNF-α (B) produced by splenic CD8⁺ T cells from Brd7^fl/fl (n = 4) and Brd7^ΔT (n = 4) mice infected with HKx31 virus stimulated with NP peptide at 10 d p.i. Number beside outlined areas indicates percent of IFN-γ⁺CD8⁺ (A) or TNF-α⁺CD8⁺ (B) T cells. Frequency of IFN-γ⁺ (A) or TNF-α⁺ (B) cells among CD8⁺ T cells. (C and D) Flow cytometry of granzyme B (GzmB) (C) or perforin (Prf1) (D) produced by splenic NP⁺CD8 ⁺ T cells from Brd7^fl/fl (n = 4) and Brd7^ΔT (n = 3) mice infected with HKx31 virus at 10 d p.i. Numbers beside outlined areas indicate mean fluorescence intensity (MFI) of GzmB (C) or Prf1 (D) among NP⁺CD8⁺ T cells. MFI of GzmB (C) or Prf1 (D) among NP⁺CD8⁺ T cells. (E) In vivo cytolysis assay. Noninfected WT host mice or infected BRD7 WT (n = 3) or BRD7-deficient (n = 3) mice received equal numbers of low CFSE–labeled B6 splenocytes loaded with NP peptide plus high CFSE–labeled B6 splenocytes without NP peptide stimulation at 10 d p.i. Cytotoxic T lymphocyte activity was assessed 4 hours after transfer. Numbers above bracketed lines represent percentages of cells per CFSE peak. (F) Frequency of peptide-pulsed cells in E was shown. (G) Body weight of Brd7^fl/fl (n = 8) and Brd7^ΔT (n = 7) mice infected with 0.5 LD₅₀ of A/PR/8/34 (H1N1) virus at various times p.i. Body weight at day 0 was set as 100%. (H) Histological examination of lung from Brd7^fl/fl or Brd7^ΔT mice noninfected with PR8 virus (top panel) or infected at day 8 p.i. (bottom panel). Lung from Brd7^ΔT mice infected with PR8 virus showed severe tissue damage and lymphocytic infiltration. Data are shown as mean ± SEM. *P < 0.05, **P < 0.01, and ***P < 0.001 (2-tailed Student’s t test). Data are representative of 3 independent experiments.

CD8⁺ T cells will proliferate and differentiate into effector cells to eradicate pathogen during virus infection. This process should be independent of virus strain. To further demonstrate the role of BRD7 in CD8⁺ T cell responses to other viral infection, we further utilized the LCMV infection model in our study, which initiates an acute infection. We infected Brd7^fl/fl mice or Brd7^ΔT mice with the Arm strain of LCMV, and we analyzed genotypes of mice at the peak of the response (day 8 p.i.). Similar percentages and numbers of the LCMV gp33-H-2D^b–specific CD8⁺ T cells were detected in Brd7^fl/fl and Brd7^ΔT mice (Figure 4, A and B), whereas the proportion of antigen-specific KLRG1⁺IL-7R^– SLECs was fewer in Brd7^ΔT mice when compared with littermate controls (Figure 4, C and D). Also, a defect in the production of effector cytokines was detected in CD8⁺ T cells from Brd7^ΔT mice infected with LCMV (Figure 4, E and F). Moreover, a reduced clearance of LCMV virus was observed in Brd7^ΔT mice compared with littermate controls (Figure 4G). These results indicate that BRD7 mediated the SLEC development independently of virus strain or TCR specificity.

Figure 4

BRD7 controls SLEC differentiation during LCMV infection. (A) Flow cytometry of Arm gp33⁺CD8⁺ T cells from spleen of Brd7^fl/fl (n = 3) and Brd7^ΔT (n = 4) mice. Brd7^fl/fl and Brd7^ΔT mice were infected with LCMV-Arm virus, and antigen-specific gp33 tetramer was stained in spleen CD8⁺ T cells at day 8 p.i. Numbers beside outlined areas indicate percent of gp33⁺CD8⁺ T cells. (B) Frequency (left) and cell number (right) of gp33-specific cells among CD8⁺ T cells in A. (C) Flow cytometry of KLRG1 and CD127 on gp33⁺CD8⁺ T cells in the spleen obtained from Brd7^fl/fl (n = 3) or Brd7^ΔT (n = 4) mice infected with Arm at 8 d p.i. Numbers in quadrants indicate percent of KLRG1⁺CD127^– SLECs (top left) or KLRG1^–CD127⁺ MPECs (bottom right). (D) Frequency of KLRG1⁺CD127^– SLECs or KLRG1^–CD127⁺ MPECs among gp33⁺CD8⁺ T cells in C. (E and F) Intracellular cytokine staining of IFN-γ (E) or TNF-α (F) produced by splenic CD8⁺ T cells from Brd7^fl/fl (n = 3) and Brd7^ΔT (n = 4) mice infected with Arm virus stimulated with gp33 peptide at day 8 p.i. Number beside outlined areas indicates percentage of IFN-γ⁺CD8⁺ (E) or TNF-α⁺CD8⁺ (F) T cells. Frequency of IFN-γ⁺ (E) or TNF-α⁺ (F) cells among CD8⁺ T cells in left. (G) LCMV Arm virus titers in the spleen of Brd7^fl/fl (n = 3) or Brd7^ΔT (n = 4) mice infected with LCMV Arm strain were determined at day 8 p.i. Data are shown as mean ± SEM. *P < 0.05, **P < 0.01 (2-tailed Student’s t test). Data are representative of 3 independent experiments.

BRD7 regulates the expression of genes critical for effector differentiation. Cell proliferation and death control the expansion of CD8⁺ T cells. To explore the proliferation of CD8⁺ T cells in vivo after infection, we evaluated the 5-bromodeoxyuridine (BrdU) incorporation in antigen-specific cells at day 10 after HKx31 infection. We found that the BrdU incorporation of NP⁺CD8⁺ T cells was similar in BRD7-deficient and BRD7 WT mice (Supplemental Figure 5A), indicating that BRD7 deficiency did not alter the proliferation of antigen-specific CD8⁺ T cells. In addition, when we quantify the apoptotic CD8⁺ T cells from infected mice using annexin V and propidium iodide staining, we did not find notable differences of the apoptotic antigen-specific CD8⁺ T cells between BRD7-deficient and WT mice (Supplemental Figure 5B). These data suggest that BRD7 did not mediate cell proliferation or apoptosis of CD8⁺ T cells, which is consistent with the above data that BRD7 deficiency did not change the proportion of antigen-specific CD8⁺ T cells.

To better elucidate the underlying molecular mechanism by which BRD7 controls the differentiation of SLECs, we analyzed the global gene expression profiles of BRD7 WT and BRD7-deficient antigen-specific CD8⁺ T cells using RNA-Seq. We sorted NP⁺CD8⁺ T cells from Brd7^fl/fl mice and Brd7^ΔT mice infected with HKx31 by flow cytometry at day 10 p.i. BRD7 deficiency led to upregulation of 856 genes and downregulation of 647 genes compared with their WT counterparts (Figure 5A). The differential expressed genes were highly enriched among transcripts induced in effector CD8⁺ T cells and activated CD8⁺ T cells (Figure 5B). We also observed a notably different expression in SLEC signature genes, including Zeb2, Tbx21, Sell (Cd62l), and Klrg1 (Figure 5C). These genes were further categorized as transcription factors, chemokine receptors, adhesion molecules, and killer cell lectin-like receptors (Figure 5C), indicating that BRD7 exerted broad regulatory effects upon SLEC signature genes.

Figure 5

BRD7 orchestrates the expression of genes critical for SLEC differentiation. (A) Scatter plot analysis of differentially expressed genes of NP⁺CD8⁺ T cells from Brd7^fl/fl and Brd7^ΔT mice infected with HKx31 at 10 d p.i. FDR ≤ 0.05 and log₂ fold change ≥ 2 were used as the threshold to evaluate the significance of differences in gene expression. (B) Gene set enrichment analysis (GSEA) in A among transcriptional differences between naive and activated CD8⁺ T cells (left) or memory and effector cells (right). NES, normalized enrichment score. (C) Differentially expressed genes of transcription factor (red), chemokine receptors (black), adhesion molecules (blue), and killer cell lectin-like receptors (pink) between WT and BRD7-deficient cells, represented as the log₂ fold change. The y axis displays the log₂ fold change of each DEG, and the x axis lists the gene name. (D) Genome browser tracks displaying RNA-Seq and ATAC-Seq data at a selected locus comparing WT and BRD7-deficient cells. Tag density from different groups was calculated by normalizing to the total mapped reads. (E) qPCR analysis of mRNA in BRD7-deficient naive CD8⁺ T cells (T_N) or NP⁺CD8⁺ T cells (T_eff) (n = 4) presented relative to expression in BRD7 WT cells (n = 4). Data are shown as mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001 (2-tailed Student’s t test). Data are representative of 2 independent experiments.

As an essential component of the PBAF complex, BRD7 deficiency may result in the change of chromatin structure. To test this possibility with a higher-resolution method, we performed assay of transposase assessable chromatin-sequencing (ATAC-Seq), which detects the insert of Tn5 transposase in open chromatin regions (40). ATAC-Seq analysis of NP⁺CD8⁺ T cells showed a less accessible chromatin configuration at SLEC signature genes, including Zeb2, Gzmb, and Ifng, in cells from Brd7^ΔT mice than that from Brd7^fl/fl mice (Figure 5D), and this was consistent with RNA-Seq data. The different expression of some important molecules involved in the differentiation of SLECs was further validated by quantitative PCR (qPCR) (Figure 5E). Consistent with RNA-Seq and ATAC-Seq data, all these genes were decreased in BRD7-deficient cells compared with their WT counterparts in the effector CD8⁺ T cells but not in naive CD8⁺ T cells. Together, these results reveal that BRD7 was a major regulator of genes involved in SLEC differentiation and function.

The PBAF complex enriches at Tbx21 promoter and affects T-bet expression in effector CD8⁺ T cells. Based on the fact that BRD7 is the specific component of the PBAF complex, a chromatin remodeling complex, we further examined the chromatin state of the BRD7-bound regions by ChIP-Seq with antibodies to BRD7 in BRD7 WT OT-I CD8⁺ T cells from OT-I mice infected with PR8-OVA at day 8 p.i. Notably, we found that BRD7 was enriched at the Tbx21 loci (Figure 6A), and this was further validated with ChIP-PCR (Figure 6B). T-bet is a major driver for CD8⁺ T cell SLEC lineage commitment and controls the expression of a large number of molecules critical for the effector differentiation of CD8⁺ T cells, including Zeb2, Runx3, Ifng, Gzma, and Prf1 (5, 41–44). The loss of T-bet leads to abrogated cytotoxic function and influences nearly 50% of the SLEC-specific genes. We hypothesized that BRD7 bound to the Tbx21 loci and regulated T-bet expression. Indeed, RNA-Seq and ATAC-Seq revealed a defect in transcriptional activity of the Tbx21 gene after BRD7 deficiency (Figure 5 and Figure 6A). A significant decrease of T-bet expression in BRD7-deficient NP⁺CD8⁺ T cells was also observed by qPCR (Figure 6C) and flow cytometry analysis (Figure 6D). Taken together, our results suggest that T-bet acts as the downstream factor for BRD7 to drive during the differentiation of effector cells.

Figure 6

BRD7 is enriched at Tbx21 loci and regulates T-bet expression. (A) Representative alignments of ChIP-, ATAC- and RNA-Seq measurements at Tbx21 loci. ATAC- and RNA-Seq measurements from cells in Figure 5. ChIP-Seq with antibody to BRD7 was assessed with OT-I CD8⁺ T cells from OT-I mice infected with PR8-OVA at day 8 p.i. ChIP-Seq with antibody to H3K9ac was assessed with OT-I CD8⁺ T cells from Brd7^fl/fl and Brd7^ΔT OT-I mice infected with PR8-OVA at day 8 p.i. (B) ChIP analysis (n = 3) shows the deposition of BRD7 at the promoter regions of Tbx21 loci. (C) qPCR analysis of T-bet mRNA in BRD7-deficient NP⁺CD8⁺ T cells (n = 3) presented relative to expression in BRD7 WT cells (n = 3). (D) Flow cytometry of T-bet in BRD7 WT (n = 4) or BRD7-deficient (n = 4) NP⁺CD8⁺ T cells. Frequency of T-bet expressing NP⁺CD8⁺ T cells. (E) ChIP-PCR assay shows the deposition of H3K9me3, H3K27me3, and H3K14ac at the promoter regions of Tbx21 loci. Brd7^fl/fl mice (n = 3) and Brd7^ΔT mice (n = 3) were used in the experiments. Data are shown as mean ± SEM. *P < 0.05 and **P < 0.01 (2-tailed Student’s t test). Data are representative of 2 independent experiments.

BRD7 was previously reported to bind to H3K9ac (26). To test the possibility that BRD7 binds to H3K9ac and increases the H3K9ac deposition at the Tbx21 locus, we performed ChIP-Seq with antibodies to H3K9ac in BRD7 WT or BRD7-deficient OT-I CD8⁺ T cells from OT-I mice infected with PR8-OVA at day 8 p.i. No significant difference of H3K9ac enrichment was observed at Tbx21 loci in BRD7-deficient cells compared with BRD7 WT cells (Figure 6A), implying that BRD7 regulated the T-bet expression independently of its binding to H3K9ac. To examine the possibility that the PBAF complex co-opts epigenetic mechanisms for target gene regulation, we further performed a ChIP assay with antibodies to H3K9me3, H3K27me3, and H3K14ac on WT and BRD7-deficient OT-I CD8⁺ T cells from OT-I mice infected with PR8-OVA at day 8 p.i. The H3K9me3, H3K27me3, and H3K14ac modifications at the Tbx21 locus were not obviously altered after BRD7 deficiency (Figure 6E). Thus, BRD7 regulated the T-bet expression independently of H3K9me3, H3K27me3, and H3K14ac modification at the Tbx21 locus.

BRD7 functions as a bridge for the PBAF complex to efficiently assemble in effector CD8⁺ T cells. Based on the result that BRD7 regulated the chromatin accessibility of Tbx21 locus, we assumed that BRD7 within the PBAF complex plays an important role in effector CD8⁺ T cell differentiation. To better explore the role of BAF and PBAF complexes in CD8⁺ T cell differentiation, we performed immunoprecipitation with a BRG1-specific antibody with lysates from naive or OT-I CD8⁺ T cells that were from OT-I mice infected with PR8-OVA at day 8 p.i. and were denoted as effector cells. The pull-down substrates were subsequently analyzed with mass spectrometry (MS) (Figure 7A). Being the ATPase subunit of BAF and PBAF complexes, BRG1 is shared by both BAF and PBAF complexes. Therefore, we could observe the component changes of both BAF and PBAF complexes. Interestingly, we found that BAF/PBAF-shared components appeared in the BRG1-associated proteins from both naive and effector CD8⁺ T cells (Figure 7, B–E), while PBAF-specific components, including BRD7, PBRM1, PHF10, and ARID2, existed only in those from effector cells (Figure 7, D and E). We further confirmed this result with an IP experiment and found that BRG1 interacted with BRD7 only in effector CD8⁺ T cells and not in naive CD8⁺ T cells (Figure 7F). Thus, we proposed that the SWI/SNF complexes appear merely as BAF complex in naive CD8⁺ T cells and that only the PBAF complex is assembled in the effector CD8⁺ T cells.

Figure 7

BRD7 functions as a bridge for PBAF complex to efficiently assemble in effector CD8⁺ T cells. (A) MS analysis of BRG1-associated proteins in naive OT-I CD8⁺ T (n = 2) or effector OT-I CD8⁺ T cells (n = 2) from OT-I mice infected with PR8-OVA at day 8 p.i. (B) Venn diagram showing the overlap of components. List 1: Reported components of BAF complex. List 2: Identified components of BAF complex from naive cells in A. List 3: Reported BAF-specific components. List 4: Identified BAF-specific from naive cells in A. (C) SWI/SNF components from naive T cells in A were clustered with STRING analysis. (D) Venn diagram showing the overlap of components. List 1: Reported components of PBAF complex. List 2: Identified components of PBAF complex from effector cells in A. List 3: Reported PBAF-specific components. List 4: Identified PBAF-specific from effector cells in A. (E) SWI/SNF components from effector cells in A were clustered with STRING analysis. (F) Co-IP of BRG1-associated proteins in naive CD8⁺ T (n = 2) or effector CD8⁺ T cells (n = 2) from OT-I mice infected with PR8-OVA. (G) ChIP (n = 3) shows the deposition of BRG1 at the promoter regions of Tbx21 loci. (H and I) The mRNA expression (n = 3) of Brg1 and Tbx21 of OT-I T cells of shRNA was analyzed 8 days after transfer into recipient mice infected with PR8-OVA virus. (J) Co-IP of BRG1 in BRD7 WT and BRD7-deficient CD8⁺ T cells from OT-I mice infected with PR8-OVA at day 8 p.i. (K and L) ChIP (n = 3) shows the deposition of BRG1 and SMARCC1 at the promoter of Tbx21 in BRD7 WT and BRD7-deficient CD8⁺ T cells from OT-I mice infected with PR8-OVA at day 8 p.i. Data are shown as mean ± SEM. *P < 0.05 and ***P < 0.001 (2-tailed Student’s t test). Data are representative of 2 independent experiments.

Because of the chromatin remodeling function, it is reasonable to assume that the ATPase unit of the PBAF complex would also be enriched at the loci of target genes. To test this hypothesis, we used BRG1-specific antibody in the ChIP assay and found that BRG1 was specifically enriched at the loci of Tbx21 in the OT-I CD8⁺ T cells from OT-I mice infected with PR8-OVA at day 8 p.i. (Figure 7G). Based on the fact that BRG1 deficiency resulted in the loss of double-negative T cells in the thymus during T cell maturation (21), we utilized BRG1-specific shRNA (shBRG1) to specifically deplete BRG1 in mature T cells. Accordingly, the depletion of BRG1 resulted in a downregulation of Tbx21 mRNA expression, indicating that BRG1 plays a role in regulating T-bet (Figure 7, H and I). Thus, we suggested that the PBAF complex bound to Tbx21 loci and regulated the expression of T-bet in the effector CD8⁺ T cells.

To explore the role of BRD7 within the PBAF complex in the effector CD8⁺ T cells, we further analyzed the interactions between SMARCC1, BRG1, and PBRM1 based on the assembly process of PBAF complex in human 293T cells (17). According to this modular assembly model of the PBAF complex, the form of BAF core module and ATPase module are 2 parallel steps. ATPase module is recruited after the core module incorporation with PBAF-specific components ARID2, BRD7, and PHF10. After incorporation with the ATPase module, the PBAF complex intermediate finalizes its formation by binding with PBRM1 (17). However, this model did not propose any role played by BRD7 in assembling the PBAF complex. Since BRD7 can individually interact with SMARCC1, PBRM1, or BRG1 (17), we supposed that BRD7 may function as a bridge between PBAF core module and ATPase module. To test this hypothesis, we analyzed the interaction between SMARCC1, PBRM1, or BRG1 after BRD7 deficiency in the effector CD8⁺ T cells. We found that the binding of BRG1 to either SMARCC1 or to PBRM1 became weakened after BRD7 deficiency in the OT-I CD8⁺ T cells (Figure 7J), indicating that the binding of the core module of BAF- or PBAF-specific component PBRM1 to the ATPase module depended upon BRD7. In addition, the enrichment of BRG1 at Tbx21 locus was significantly less after BRD7 deficiency in effector CD8⁺ T cells (Figure 7K), while the enrichment of SMARCC1 at the Tbx21 locus was unaffected after BRD7 deficiency in effector CD8⁺ T cells (Figure 7L), implying that the recruitment of the ATPase module to the target gene depended on the assembly of the core module of PBAF with BRD7. Altogether, these results indicate that BRD7 was the bridge between the BAF core module and the ATPase module during the assembly of PBAF complex in effector CD8⁺ T cells.

FH, YL, YQ, and YY designed the experiments, performed most of these experiments, analyzed the data, and wrote the manuscript; ZZ, B Luo, YW, JL, JC, and WZ performed some of the experiments; HZ provided scientific expertise and the interpretation of data for the work; and B Liu contributed to the idea generation, experimental design, and manuscript writing and conceived the project.

View Supplemental data

View Supporting data values

We gratefully acknowledge Lilin Ye (Institute of Immunology, Third Military Medical University, Chongqing, China) for providing the LCMV Arm virus.

We acknowledge National key research and development program (2021YFC2301904) and Basic and Applied Basic Research Fund Committee of Guangdong Province (SL2022A04J01923) to B Liu, and we acknowledge National Key R&D Program of Department of Science and Technology of China (2022YFC0870700), Important Key Program of Natural Science Foundation of China (92169201, 92369205), Emergency Key Program of Guangzhou National Laboratory (EKPG21-24), and Key R&D Program of Department of Science and Technology of Guangdong (2022B1111020004) to HZ.

Address correspondence to: Bingfeng Liu, Yat-Sen University, No. 74, 2nd Zhongshan Road, Guangzhou, Guangdong Province, 510080 PRC. Phone: 86.20.87332588; Email: liubf5@mail.sysu.edu.cn.

Conflict of interest: The authors have declared that no conflict of interest exists.

Copyright: © 2024, Huang et al. This is an open access article published under the terms of the Creative Commons Attribution 4.0 International License.

Reference information: JCI Insight. 2024;9(15):e171605.https://doi.org/10.1172/jci.insight.171605.

1. Gray SM, Kaech SM, Staron MM. The interface between transcriptional and epigenetic control of effector and memory CD8(+) T-cell differentiation. Immunol Rev. 2014;261(1):157–168.
   View this article via: CrossRef PubMed Google Scholar
2. Kaech SM,, and Cui WTranscriptional control of effector and memory CD8+ T cell differentiation. Nat Rev Immunol. 2012;12(11):749–761.
   View this article via: CrossRef PubMed Google Scholar
3. Chang JT, et al. Molecular regulation of effector and memory T cell differentiation. Nat Immunol. 2014;15(12):1104–1115.
   View this article via: CrossRef PubMed Google Scholar
4. Rutishauser RL, Kaech SM. Generating diversity: transcriptional regulation of effector and memory CD8 T-cell differentiation. Immunol Rev. 2010;235(1):219–233.
   View this article via: CrossRef PubMed Google Scholar
5. Joshi NS, et al. Inflammation directs memory precursor and short-lived effector CD8(+) T cell fates via the graded expression of T-bet transcription factor. Immunity. 2007;27(2):281–295.
   View this article via: CrossRef PubMed Google Scholar
6. Cannarile MA, et al. Transcriptional regulator Id2 mediates CD8+ T cell immunity. Nat Immunol. 2006;7(12):1317–1325.
   View this article via: CrossRef PubMed Google Scholar
7. Kallies A, et al. Blimp-1 transcription factor is required for the differentiation of effector CD8(+) T cells and memory responses. Immunity. 2009;31(2):283–295.
   View this article via: CrossRef PubMed Google Scholar
8. Yang CY, et al. The transcriptional regulators Id2 and Id3 control the formation of distinct memory CD8⁺ T cell subsets. Nat Immunol. 2011;12(12):1221–1229.
   View this article via: CrossRef PubMed Google Scholar
9. Yao S, et al. Interferon regulatory factor 4 sustains CD8(+) T cell expansion and effector differentiation. Immunity. 2013;39(5):833–845.
   View this article via: CrossRef PubMed Google Scholar
10. Hess Michelini R, et al. Differentiation of CD8 memory T cells depends on Foxo1. J Exp Med. 2013;210(6):1189–1200.
    View this article via: CrossRef PubMed Google Scholar
11. Kim MV, et al. The transcription factor Foxo1 controls central-memory CD8⁺ T cell responses to infection. Immunity. 2013;39(2):286–297.
    View this article via: CrossRef PubMed Google Scholar
12. Pearce EL, et al. Control of effector CD8⁺ T cell function by the transcription factor Eomesodermin. Science. 2003;302(5647):1041–1043.
    View this article via: CrossRef PubMed Google Scholar
13. Zhou X, et al. Differentiation and persistence of memory CD8(+) T cells depend on T cell factor 1. Immunity. 2010;33(2):229–240.
    View this article via: CrossRef PubMed Google Scholar
14. Gray SM, et al. Polycomb repressive complex 2-mediated chromatin repression guides effector CD8⁺ T cell terminal differentiation and loss of multipotency. Immunity. 2017;46(4):596–608.
    View this article via: CrossRef PubMed Google Scholar
15. Shin HM, et al. Epigenetic modifications induced by Blimp-1 Regulate CD8⁺ T cell memory progression during acute virus infection. Immunity. 2013;39(4):661–675.
    View this article via: CrossRef PubMed Google Scholar
16. Clapier CR, et al. Mechanisms of action and regulation of ATP-dependent chromatin-remodelling complexes. Nat Rev Mol Cell Biol. 2017;18(7):407–422.
    View this article via: CrossRef PubMed Google Scholar
17. Mashtalir N, et al. Modular organization and assembly of SWI/SNF family chromatin remodeling complexes. Cell. 2018;175(5):1272–1288.
    View this article via: CrossRef PubMed Google Scholar
18. Sokpor G, et al. Context-specific chromatin remodeling activity of mSWI/SNF complexes depends on the epigenetic landscape. Signal Transduct Target Ther. 2021;6(1):360.
    View this article via: CrossRef PubMed Google Scholar
19. Mashtalir N, et al. Chromatin landscape signals differentially dictate the activities of mSWI/SNF family complexes. Science. 2021;373(6552):306–315.
    View this article via: CrossRef PubMed Google Scholar
20. Yuan J, et al. Structure of human chromatin-remodelling PBAF complex bound to a nucleosome. Nature. 2022;605(7908):166–171.
    View this article via: CrossRef PubMed Google Scholar
21. Gebuhr TC, et al. The role of Brg1, a catalytic subunit of mammalian chromatin-remodeling complexes, in T cell development. J Exp Med. 2003;198(12):1937–1949.
    View this article via: CrossRef PubMed Google Scholar
22. Zhang F, Boothby M. T helper type 1-specific Brg1 recruitment and remodeling of nucleosomes positioned at the IFN-γ promoter are Stat4 dependent. J Exp Med. 2006;203(6):1493–1505.
    View this article via: CrossRef PubMed Google Scholar
23. Hodges C, et al. The many roles of BAF (mSWI/SNF) and PBAF complexes in cancer. Cold Spring Harb Perspect Med. 2016;6(8):a026930.
    View this article via: CrossRef PubMed Google Scholar
24. Osman S. Reshuffling chromatin: how the human chromatin remodeler PBAF recognizes nucleosomes. Nat Struct Mol Biol. 2022;29(5):419.
    View this article via: CrossRef PubMed Google Scholar
25. Kaeser MD, et al. BRD7, a novel PBAF-specific SWI/SNF subunit, is required for target gene activation and repression in embryonic stem cells. J Biol Chem. 2008;283(47):32254–32263.
    View this article via: CrossRef PubMed Google Scholar
26. Sun H, et al. Solution structure of BRD7 bromodomain and its interaction with acetylated peptides from histone H3 and H4. Biochem Biophys Res Commun. 2007;358(2):435–441.
    View this article via: CrossRef PubMed Google Scholar
27. Peng C, et al. The transcriptional regulation role of BRD7 by binding to acetylated histone through bromodomain. J Cell Biochem. 2006;97(4):882–892.
    View this article via: CrossRef PubMed Google Scholar
28. Burrows AE, et al. Polybromo-associated BRG1-associated factor components BRD7 and BAF180 are critical regulators of p53 required for induction of replicative senescence. Proc Natl Acad Sci U S A. 2010;107(32):14280–14285.
    View this article via: CrossRef PubMed Google Scholar
29. Chiu YH, et al. BRD7, a tumor suppressor, interacts with p85α and regulates PI3K activity. Mol Cell. 2014;54(1):193–202.
    View this article via: CrossRef PubMed Google Scholar
30. Drost J, et al. BRD7 is a candidate tumour suppressor gene required for p53 function. Nat Cell Biol. 2010;12(4):380–389.
    View this article via: CrossRef PubMed Google Scholar
31. Hu K, et al. Poly(ADP-ribosyl)ation of BRD7 by PARP1 confers resistance to DNA-damaging chemotherapeutic agents. EMBO Rep. 2019;20(5):e46166.
    View this article via: CrossRef PubMed Google Scholar
32. Wang XM, et al. BRD7 mediates hyperglycaemia-induced myocardial apoptosis via endoplasmic reticulum stress signalling pathway. J Cell Mol Med. 2017;21(6):1094–1105.
    View this article via: CrossRef PubMed Google Scholar
33. Zhao R, et al. BRD7 plays an anti-inflammatory role during early acute inflammation by inhibiting activation of the NF-кB signaling pathway. Cell Mol Immunol. 2017;14(10):830–841.
    View this article via: CrossRef PubMed Google Scholar
34. Kikuchi M, et al. TRIM24 mediates ligand-dependent activation of androgen receptor and is repressed by a bromodomain-containing protein, BRD7, in prostate cancer cells. Biochim Biophys Acta. 2009;1793(12):1828–1836.
    View this article via: CrossRef PubMed Google Scholar
35. Chirkova T, et al. CX3CR1 is an important surface molecule for respiratory syncytial virus infection in human airway epithelial cells. J Gen Virol. 2015;96(9):2543–2556.
    View this article via: CrossRef PubMed Google Scholar
36. Smith JJ, et al. The SWI/SNF chromatin remodeling assemblies BAF and PBAF differentially regulate cell cycle exit and cellular invasion in vivo. PLoS Genet. 2022;18(1):e1009981.
    View this article via: CrossRef PubMed Google Scholar
37. Wei Z, et al. Vitamin D switches BAF complexes to protect β cells. Cell. 2018;173(5):1135–1149.
    View this article via: CrossRef PubMed Google Scholar
38. Pan D, et al. A major chromatin regulator determines resistance of tumor cells to T cell–mediated killing. Science. 2018;359(6377):770–775.
    View this article via: CrossRef PubMed Google Scholar
39. Hikono H, et al. Activation phenotype, rather than central- or effector-memory phenotype, predicts the recall efficacy of memory CD8+ T cells. J Exp Med. 2007;204(7):1625–1636.
    View this article via: CrossRef PubMed Google Scholar
40. Buenrostro JD, et al. Transposition of native chromatin for fast and sensitive epigenomic profiling of open chromatin, DNA-binding proteins and nucleosome position. Nat Methods. 2013;10(12):1213–1218.
    View this article via: CrossRef PubMed Google Scholar
41. Kallies A, Good-Jacobson KL. Transcription factor T-bet orchestrates lineage development and function in the immune system. Trends Immunol. 2017;38(4):287–297.
    View this article via: CrossRef PubMed Google Scholar
42. Xin A, et al. A molecular threshold for effector CD8(+) T cell differentiation controlled by transcription factors Blimp-1 and T-bet. Nat Immunol. 2016;17(4):422–432.
    View this article via: CrossRef PubMed Google Scholar
43. Dominguez CX, et al. The transcription factors ZEB2 and T-bet cooperate to program cytotoxic T cell terminal differentiation in response to LCMV viral infection. J Exp Med. 2015;212(12):2041–2056.
    View this article via: CrossRef PubMed Google Scholar
44. Omilusik KD, et al. Transcriptional repressor ZEB2 promotes terminal differentiation of CD8⁺ effector and memory T cell populations during infection. J Exp Med. 2015;212(12):2027–2039.
    View this article via: CrossRef PubMed Google Scholar
45. Xiao L, et al. Targeting SWI/SNF ATPases in enhancer-addicted prostate cancer. Nature. 2022;601(7893):434–439.
    View this article via: CrossRef PubMed Google Scholar
46. Carcamo S, et al. Altered BAF occupancy and transcription factor dynamics in PBAF-deficient melanoma. Cell Rep. 2022;39(1):110637.
    View this article via: CrossRef PubMed Google Scholar
47. Shu S, et al. Synthetic lethal and resistance interactions with BET bromodomain inhibitors in triple-negative breast cancer. Mol Cell. 2020;78(6):1096–1113.
    View this article via: CrossRef PubMed Google Scholar
48. Guo A, et al. cBAF complex components and MYC cooperate early in CD8⁺ T cell fate. Nature. 2022;607(7917):135–141.
    View this article via: CrossRef PubMed Google Scholar
49. Barisic D, et al. Mammalian ISWI and SWI/SNF selectively mediate binding of distinct transcription factors. Nature. 2019;569(7754):136–140.
    View this article via: CrossRef PubMed Google Scholar
50. Zhang J, et al. Host protein moloney leukemia virus 10 (MOV10) acts as a restriction factor of influenza A virus by inhibiting the nuclear import of the viral nucleoprotein. J Virol. 2016;90(8):3966–3980.
    View this article via: CrossRef PubMed Google Scholar
51. Huang F, et al. Identification of a novel compound targeting the nuclear export of influenza A virus nucleoprotein. J Cell Mol Med. 2018;22(3):1826–1839.
    View this article via: CrossRef PubMed Google Scholar
52. Ma X, et al. TRIM28 promotes HIV-1 latency by SUMOylating CDK9 and inhibiting P-TEFb. Elife. 2019;8:e42426.
    View this article via: CrossRef PubMed Google Scholar