LMNB2-mediated high PD-L1 transcription triggers the immune escape of hepatocellular carcinoma

Abstract

While immune checkpoint inhibitors targeting programmed cell death-ligand 1 (PD-L1) demonstrate clinical efficacy in hepatocellular carcinoma (HCC), tumor cells frequently evade immune surveillance through PD-L1 overexpression, a phenomenon whose regulatory mechanisms remain poorly understood. Through integrated analysis of single-cell transcription sequence data, we identified aberrant upregulation of Lamin B2 (LMNB2) specifically in immunotherapy-sensitive HCC patients. Functional characterization revealed that LMNB2 acts as a transcriptional regulator of PD-L1, potentiating immune escape mechanisms in HCC cells during co-culture with Jurkat cells. Notably, we discovered that speckle-type POZ protein (SPOP) directly interacts with LMNB2 to mediate its ubiquitination and proteasomal degradation, thereby maintaining physiological PD-L1 expression levels. Clinically relevant SPOP mutations or reduced SPOP expression impaired this regulatory mechanism, leading to LMNB2 accumulation and subsequent PD-L1 hyperactivation. Importantly, combinatorial targeting of LMNB2 with Atezolizumab (PD-L1 inhibitor) displayed a synergistic effect on suppressing tumor progression both in vitro and in vivo, particularly in HCC models with SPOP mutations or LMNB2 overexpression. These findings unveil a novel ubiquitination-dependent regulatory axis in HCC immune evasion and propose targeted co-inhibition strategies to overcome HCC immunotherapy resistance.

Fig. 1. LMNB2 is significantly elevated in HCC, negatively related to immune infiltration.

A Intersection of upregulated DEGs in drug-resistant HCC samples (GSE211850, GSE182593, GSE151412, and GSE192771). B Annotation and clustering for ST profiling. C Analysis expression of LMNB2 in HCC cells between patient responders and patient non-responders. D–F Debatching effect, clustering, and annotation for sc-RNA seq of 10 HCC samples (T) and 10 normal liver samples (P) (GSE235057). G Expression of LMNB2 in hepatocyte cells in sc-RNA seq of total10 HCC samples. H Expression of LMNB2 in hepatocyte cells in sc-RNA seq of separate 10 HCC samples. I Proportion of cell types in scRNA-seq of 10 HCC samples. J Proportion of CD4+ T cells in the 10 HCC samples grouped with high/low LMNB2 expression of hepatocyte cells. K Proportion of CD8+ T cells in the 10 HCC samples grouped with high/low LMNB2 expression of hepatocyte cells. L Proportion of NKT cells in the 10 HCC samples grouped with high/low LMNB2 expression of hepatocyte cells. M Proportion of B cells in the 10 HCC samples grouped with high/low LMNB2 expression of hepatocyte cells. Data are shown as mean ± SD (n ≥ 3). *p < 0.05, ns p ≥ 0.05.

Fig. 2. LMNB2 can promote tumor growth by inducing T-cell apoptosis.

A WB verified the overexpression and knockdown efficiency of LMNB2 in Hepa1-6 cells. B Schematic representation of xenograft tumors with sh-NC + EV, sh-NC + LMNB2, and sh-LMNB2 + EV. C Weight of xenograft tumors of sh-NC + EV, sh-NC + LMNB2, and sh-LMNB2 + EV. D Volume of xenograft tumors of sh-NC + EV, sh-NC + LMNB2, and sh-LMNB2 + EV. E IHC staining of LMNB2, CD3, and CD8 in sh-NC + EV, sh-NC + LMNB2, and sh-LMNB2 + EV in xenograft tumors. Scale bar, 200 μm. F IHC staining scores for LMNB2, CD3, and CD8 in sh-NC + EV, sh-NC + LMNB2, and sh-LMNB2 + EV in xenograft tumors. G WB verified the overexpression and knockdown efficiency of LMNB2 in Huh7 and HepG2 cell lines. H Apoptosis of Jurkat cells co-cultured with Huh7/HepG2 cells treated with sh-NC + EV, sh-NC + LMNB2, or sh-LMNB2 + EV was detected by flow cytometry. I Statistical analysis of apoptotic levels in Jurkat cells (H). J The cell cycle of Jurkat cells co-cultured with Huh7/HepG2 cells treated with sh-NC + EV, sh-NC + LMNB2, and sh-LMNB2 + EV was analyzed by flow cytometry. K Cell cycle statistics of Jurkat cells (J). L–O The levels of IL-2, IL-4, IL-10, and IFN-γ produced by Jurkat cells were detected using ELISA. Data are shown as mean ± SD (n ≥ 3). *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Fig. 3. LMNB2 is a bona fide transcription factor of PD-L1.

A RNA-seq analysis of control and LMNB2 knockdown Huh7 cells. B Volcano map of DEGs in control and LMNB2 knockdown Huh7 cells. C GSEA analysis of PD-L1 signaling in control and LMNB2 knockdown Huh7 cells. D Correlation analysis between LMNB2 and PD-L1 expression in the TIMER database. E Correlation analysis between LMNB2 and PD-L1 expression in the GEPIA database. F Correlation analysis between LMNB2 and immune cell infiltration in HCC using the TIMER database. G LMNB2 promotes the protein expression of exogenous PD-L1 in HepG2 cells. H LMNB2 promotes the protein expression of endogenous PD-L1 in HepG2 cells. I Model diagram of LMNB2 binding to the PD-L1 promoter to promote its transcription. J LMNB2 promotes the mRNA expression of PD-L1. K LMNB2 promotes the activity of PD-L1. L Model diagram of the six truncated PD-L1 promoters. M WB verified the overexpression of LMNB2 in 293T cells for dual-luciferase reporter analysis. N Statistical of the LMNB2 effect on the activity of six truncated PD-L1 promoters (O). Distribution of LMNB2 in ST-RNA-seq of HCC samples. P Correlation analysis between LMNB2 and PD-L1 in ST-RNA-seq in HCC samples. Q IHC staining of LMNB2 and PD-L1 in HCC samples. Scale bar, 200 μm. R Correlation analysis between LMNB2 and SPOP IHC scores. Data are shown as mean ± SD (n ≥ 3). ***p < 0.001, ****p < 0.0001.

Fig. 4. SPOP promotes LMNB2 lysosome-mediated degradation through K29-linkage ubiquitination.

A Coomassie blue staining for the enrichment of LMNB2. B Mass spectrometry of related proteins in LMNB2 enrichment. C and D Interaction between LMNB2 and SPOP. E Interaction between LMNB2 and SPOP in vitro. F Interaction between LMNB2 and SPOP-WT, SPOP-ΔMATH, SPOP-ΔBTB, and SPOP-ΔNLS. G Mode diagram of binding ability between LMNB2 and SPOP-WT, SPOP-ΔMATH, SPOP-ΔBTB, or SPOP-ΔNLS. H Co-localization of LMNB2, SPOP-WT, SPOP-ΔMATH, SPOP-ΔBTB, and SPOP-ΔNLS. Scale bar, 50 μm. I–K SPOP promotes the protein expression of LMNB2. L and M Effect of SPOP on the half-life of LMNB2. N Identifying degradation pathway of LMNB2 induced by SPOP. O Effects of SPOP-WT, SPOP-ΔMATH, SPOP-ΔBTB, and SPOP-ΔNLS on LMNB2 degradation. P Effects of SPOP-WT, SPOP-ΔMATH, SPOP-ΔBTB, and SPOP-ΔNLS on LMNB2 ubiquitination. Q and R Identification of ubiquitination linkage of LMNB2 induced by SPOP. Data are shown as mean ± SD (n ≥ 3). ***p < 0.001.

Fig. 5. SPOP-induced immune surveillance depends on LMNB2‑PD‑L1 axis in Huh7 cells.

A Huh7 cells achieving sh-NC + EV, sh-SPOP + EV, sh-SPOP+sh-LMNB2 + EV, SPOP + LMNB2 + sh-NC, SPOP + LMNB2-△SBC+sh-NC and SPOP-M35L + LMNB2 + sh-NC were co-cultured with Jurkat cells for 24 h after treatment with DMSO or Atezolizumab (10 ng/mL). WB of Huh7 cells in the HCC cell-Jurkat cell co-culture system for the detection of PD-L1 expression levels. All quantitation was normalized to the protein level of GAPDH in the sh-NC + EV group. B Huh7 cells achieving the above treatment were co-cultured with Jurkat cells for 24 h. Flow cytometry analysis of PD-1 binding on Huh7 cell surface. C Statistics of mean fluorescence intensity (MFI) for PD-1 in (B). D Huh7 cells achieving the above treatment were co-cultured with Jurkat cells for 24 h after treatment with DMSO or Atezolizumab (10 ng/ml). Apoptosis levels in Jurkat cells were detected by flow cytometry. E Statistical analysis of apoptotic levels in Jurkat cells (D). F Huh7 cells achieving the above treatment were co-cultured with Jurkat cells for 24 h after treatment with DMSO or Atezolizumab (10 ng/ml). The cell cycles of Jurkat cells were detected by flow cytometry. G Cell cycle statistics of Jurkat cells (F). H–K The levels of IL-2, IL-4, IL-10, and IFN-γ produced by Jurkat cells were detected using ELISA. Data are shown as mean ± SD (n ≥ 3). *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Fig. 6. Atezolizumab counteracts defective SPOP-LMNB2-PD-L1 axis-induced immune escape of HCC in vivo.

A Mode diagram of Atezolizumab injection in C57BL/6 J mice with a subcutaneous tumor. B Schematic representation of xenograft tumors of sh-NC + EV, sh-SPOP + EV, sh-SPOP + sh-LMNB2 + EV, SPOP + LMNB2 + sh-NC, SPOP + LMNB2-△SBC + sh-NC, and SPOP-M35L + LMNB2 + sh-NC groups in DMSO or Atezolizumab. C Weight of xenograft tumors in the above groups. D Volume of xenograft tumor in the above group. E IHC staining of SPOP, LMNB2, PD-L1, CD3, and CD8 in xenograft tumors. Scale bar, 200 μm. F Statistical analysis of the IHC staining scores of SPOP, LMNB2, PD-L1, CD3, and CD8 in xenograft tumors. Data are shown as mean ± SD (n ≥ 3). *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, ns p ≥ 0.05.

Fig. 7

Schematic of the proposed mechanism through which PD-L1 transcription activation induced by SPOP dysfunction induces immune escape of HCC cells.