PU.1-dependent regulation of UCH L1 expression in B-lymphoma cells

Abstract

Elevated levels of ubiquitin C-terminal hydrolase L1 (UCH L1) have been detected in a variety of malignancies, and recent studies show the oncogenic capacity of overexpressed UCH L1 in vivo in animal models. Here we demonstrate that expression of endogenous UCH L1 is significantly higher in B-lymphoma cells than in transformed cells of epithelial and fibroblastic origin. The specific hematopoietic transcription factor PU.1 induces UCH L1 expression through direct activation of the uch l1 promoter. Using chromatin immunoprecipitation (ChIP) assays and direct mutagenesis we identified PU.1 binding sites on the uch l1 promoter, at least three of which are involved in this activation. We also show that the viral transcriptional co-activator EBNA2 dramatically increases PU.1-dependent up-regulation of endogenous UCH L1 expression. Finally, inhibition of PU.1 expression with specific shRNA resulted in reduction of UCH L1 mRNA and protein levels in Epstein-Barr virus (EBV)-transformed B-cells. We propose that the ubiquitin-editing enzyme UCH L1 is a multifunctional pro-oncogenic factor involved in development and progression of certain lymphoid malignancies, including EBV-associated lymphomas.

Figure 1

High levels of UCH L1 protein detected in B-lymphoma cells. Total cell lysates from four epithelial (U2OS, 293, HeLa, and C33A, lanes 1–4), four fibroblastic (NIH3T3, GM0010A, Cos7, and GM00637F, lanes 5–8), and four B-lymphoblastoid (BL30, X-50/7, KR4, and Raji, lanes 9–12) cell lines were separated in SDS-PAGE and probed with UCH L1 and GAPDH antibodies.

Figure 2

PU.1 directly binds to the uch l1 promoter at each of five PU.1 sites in vitro. (A) Schematic illustration of uch l1 promoter with five putative PU.1-binding sites. (B) Purified recombinant PU.1 separated on SDS-PAGE and detected with Coomassie blue stain; 4 µg of BSA was used as loading control. (C) High molecular weight DNA-protein complexes of uch l1 promoter and PU.1 stained with SYBR green EMSA to visualize DNA. (D) Western blot analysis was performed on the same gel for visualization of PU.1 in complex with uch l1 promoter with PU.1 antibody.

Figure 3

Mutations in PU.1-binding sites abolish PU.1-mediated activation of the uch l1 promoter. (A) ChIP/PCR analysis was performed to determine binding of PU.1 factor to the putative binding sites on uch l1 promoter with the use of specific PU.1 antibody in KR4 LCLs. Normal IgG was used as negative control in immunoprecipitations. PCR reactions were performed with primers targeting PU.1-binding sites (see ‘Materials and methods’), and amplified DNA products were resolved in 2% agarose gel. (B) NIH 3T3 cells were co-transfected with control or PU.1 (250 ng/well) expression vectors, along with UCHL1p–LUC wild-type and mutant reporter plasmids (500 ng/well). Luciferase assays were performed 48 h post-transfection. The data represent three independent experiments prepared in triplicate and normalized to β-galactosidase activity.

Figure 4

UCH L1 expression is induced in primary B-cells after EBV immortalization. Total RNA and protein were extracted from uninfected PMBCs from two donors (I and II) and from the same PBMCs after EBV immortalization. (A) Reverse-transcription PCR analysis was performed with UCH L1-specific primers and GAPDH primers used as internal control. (B) Western blot analysis was performed with UCH L1 antibody. β-Actin served as loading control.

Figure 5

Type III EBV latency products induce UCH L1 expression in a Burkitt lymphoma cell line latently infected with EBV. uch l1 promoter activity was determined in paired uninfected and EBV-infected BL30 cell lines. (A) Cells (2 × 10⁵) were nucleofected with UCHL1p–LUC reporter construct. Luciferase assays were performed 48 h post-transfection. The data represent three independent experiments prepared in triplicate and normalized to β-galactosidase activity. (B) Total RNA was extracted from BL30 and BL30-EBV cells, and real-time PCR analysis with specific primers for UCH L1 was performed. GAPDH was used as internal control. (C) Western blot analysis for UCH L1 protein levels in lysates from BL30 and BL30-EBV cells was performed with UCH L1 antibody. GAPDH was used as loading control. Quantification is shown for UCH L1 protein expression. (D) Total RNA was extracted from BL30 and BL30-EBV cells, and real-time PCR analysis with specific primers for PU.1 was performed. GAPDH was used as internal control. (E) Western blot analysis for UCH L1 protein levels in lysates from BL30 and BL30-EBV cells was performed with PU.1 antibody. GAPDH was used as loading control.

Figure 6

PU.1 and EBNA2 form endogenous complex and synergistically induce endogenous UCH L1 expression. (A) PU.1 was immunoprecipitated from total lysates of KR4 LCLs, IPs were resolved in 12% PAGE and probed with EBNA2 antibody. Rabbit normal immunoglobulins were used as control for IPs. (B) NIH 3T3 cells (low endogenous level of UCH L1) were used for all experiments. Cells were co-transfected with control (1000 ng/well) or PU.1 (250 ng/well) or EBNA2-HA (750 ng/well) or EBNA2-HA (750 ng/well) and PU.1 (250 ng/well) together, along with UCHL1p–LUC wild-type or with five PU.1 binding sites mutated (500 ng/well). Control DNA was used as filler where necessary. Luciferase assays were performed 48 h post-transfection and are normalized to β-galactosidase activity. (C) Next, cells were co-transfected with EBNA2-HA (750 ng/well) and PU.1 (250 ng/well) expression vector or control. Total RNA and protein were extracted 48 h post-transfection, RT-PCR was performed to detect expression of UCH L1 RNA levels with UCH L1-specific primers. (D) Western blot analysis was used to detect endogenous UCH L1 protein levels with UCH L1 antibody. GAPDH served as loading control.

Figure 7

Inhibition of PU.1 expression results in reduction of UCH L1 levels in transformed B-cells. PU.1 shRNA or GFP shRNA was stably nucleofected in the EBV-transformed B-cell line KR4. Cells were selected and maintained in medium with puromycin (1 µg/mL). (A) Total RNA was extracted from cells expressing GFP shRNA and PU.1 shRNA, and real-time PCR analysis with specific primers for PU.1 and UCH L1 was performed. GAPDH was used as internal control. (B) Western blot analysis for UCH L1 protein levels was performed in cell lysates from cells expressing GFP shRNA and PU.1 shRNA with PU.1 and UCH L1 antibody. GAPDH was used as loading control.

Figure 8

Working model: PU.1 up-regulates uch l1 expression in B-cell lymphomas. In B-lymphomas, the hematopoietic specific factor PU.1 induces UCH L1 expression. Induced expression of UCH L1 is then used to activate oncogenic pathways [14].