The deubiquitinating enzyme UCHL1 is a favorable prognostic marker in neuroblastoma as it promotes neuronal differentiation

In our study, there were 64 pediatric patients with neuroblastic tumors containing 19 GNB and 38 NB, who were histologically diagnosed in Xinhua hospital affiliated to Shanghai Jiaotong University School of Medicine during October 2012 and February 2015. The recorded data of each patient were analyzed and performed follow-up through phone calls. Each tumor specimen was stored in liquid nitrogen for TMA analysis. 77 pediatric patients with neuroblastic tumors (40 GNB/GN and 37 NB) were histologically diagnosed in Xinhua hospital affiliated to Shanghai Jiaotong University School of Medicine during September 2012 and December 2016. Each tumor specimen was further for high performance liquid chromatography/mass spectra / mass spectra analysis. Protein of 12 of 77 patients was extracted for immunoblotting analysis. All experimental protocols were approved by the Ethics Committee of the Xinhua hospital affiliated to Shanghai Jiaotong University School of Medicine.

Tissue specimens were separated out a small part and shaped in the special mold for microarray preparation. After fixed in the 4% paraformaldehyde over night, they were trimmed and embedded in paraffin as a planned array. Then, samples were sectioned (5 μm) and attached to poly-L-lysine coated slides. IHC staining was performed using a standard immunoperoxidase staining procedure (primary antibody to UCHL1, 1:400, Cell Signaling Technology, Danvers, Massachusetts, USA). Hematoxylin was used as a counterstain. The tissue sections were viewed independently by two pathologists in a blind fashion. IHC staining was scored on a specialized scale from 0 to 4: 0 represented negative expression, 1 represented weakly positive expression (0–10% positive cells), 2 represented mildly positive expression (10–30% positive cells), 3 represented moderately positive expression (30–50% positive cells), 4 represented strongly positive expression (50–100% positive cells). The scale was determined according to the average number of positive cells in 10 random fields of one slide.

Tissue array analysis results of NB patient tumor samples were obtained from the R2 Genomics Analysis and Visualization Platform (http://​r2.​amc.​nl) using the following publicly available datasets: Oberthuer (ArrayExpress: E-TABM-38) [27], Versteeg (GEO: GSE16476) [28], and Seeger (GEO: GSE3446) [29], which included comprehensive information on the relevant clinical and prognostic factors selected for analysis. Four patients without survival information was not included in the Oberthuer dataset, and the information of age, MYCN status and stage were not contained in the Seeger dataset. For Kaplan-Meier analysis, the best p value and corresponding cutoff value was selected according to the R2 Genomics Analysis and Visualization Platform. The Asgharzadeh dataset, which was also obtained from the R2 Genomics Analysis and Visualization Platform, contained 209 neuroblastic tumors with included differentiating histology, and was used for differentiation analysis.

The human neuroblastoma cell lines, SH-SY5Y and SK-N-BE (2) [30], were obtained from ATCC (Manassas, USA). SH-SY5Y and SK-N-BE (2) cells were maintained in a 1:1 mixture of Eagle’s Minimum Essential Medium and F12 Medium with 10% fetal bovine serum, 1% Penicillin-Streptomycin, 2 mM L-glutamine, 1% non-essential amino acids and 1% Sodium Pyruvate (All from Life Technologies GmbH, Darmstadt, Germany). For neural differentiation assay, all trans-RA (Sigma-Aldrich, St Louis, MO) was dissolved in dimethyl sulphoxide (DMSO, Sigma-Aldrich) and 10 mM stock solutions were prepared. SH-SY5Y and SK-N-BE (2) cells were treated with 10 μM RA for 3,5 or 7 days, DMSO was used as negative control. Pictures were taken to monitor neurite outgrowth. The reversible, competitive, active-site directed inhibitor, LDN57444 (Selleckchem, Houston, USA), was used for UCHL1 inhibition [31].

Oligonucleotides with the following nucleotide sequences were used for the cloning of shRNA-encoding sequences into a lentiviral vector plvx-shRNA2 (clontech, Tokyo, Japan): human UCHL1 (shUCHL1), 5’-GATCCCGGGTAGATGACAAGGTGAATCTCGAGATTCACCTTGTCATCTACCCGTTTTTG-3′; Scrambled control (shNC), 5’-GATCCCCTAAGGTTAAGTCGCCCTCGCTCGAGCGAGGGCGACTTAACCTTAGGTTTTTG-3′ (Synthetized by Sangon Biotech). High titer lentiviral stocks were produced, and SH-SY5Y cells were infected with scrambled control lentivirus (shNC) or lentivirus expressing shRNA inhibiting UCHL1 (shUCHL1) according to the manufacturer’s protocol (http://​www.​clontech.​com/). GFP positive cells were selected by flow cytometry sorting and passaged for further study.

Cells were harvested and lysed in the RIPA buffer (Beyotime, Haimen, China) containing PMSF (Beyotime) for 30 min on ice. Lysates were clarified by centrifugation at 15000 g for 30 min. Protein concentration of the supernatant fraction was determined by the Bradford assay. Protein samples were diluted in 4× SDS loading buffer (TaKaRa) and heated to 95 °C for 5 min and fractionated in a 10% or 8% SDS-polyacrylamide gel. Proteins were electroblotted onto a polyvinylidene fluoride and incubated for 1 h in 5% bovine serum albumin in phosphate buffer solution (PBS) or nonfat dry milk dissolved in PBS containing 0.1% Tween-20 (PBST) at room temperature. The blotting membranes were incubated with primary antibodies overnight at 4 °C, extensively washed in PBST, incubated with HRP-conjugated secondary antibody (Cell Signaling Technology) for 1 h at room temperature, and washed again with PBST. The blotting membranes were developed with chemiluminescent reagents (Millipore, Billerica, MA, USA) according to the instructions provide by the manufacturer. Primary antibody to UCHL1, AKT, phospho-AKT (Ser473), ERK1/2, phospho-ERK1/2 (Thr202/Tyr204), p21 and Cyclin D1 were purchased from Cell Signaling Technology, and primary antibody to tyrosine hydroxylase (TH) and growth-associated protein 43 (GAP43) were purchased from Abcam (Cambridge, MA, USA). The blots were quantitative analyzed through the software of Image J (X64, v. 2.1.4).

Total RNA was extracted with TRIzol (Invitrogen) and reverse-transcribed into cDNA with the reverse transcription kit from TakaRa (Tokyo, Japan). The levels of mRNAs were measured by real-time PCR with SYBR Green reagent from Roche (Natley, NJ, USA) and normalized to the level of GAPDH mRNA. Primer sequences were as follows: TH forward 5′-TCATCACCTGGTCACCAAGTT-3′, reverse 5′-GGTCGCCGTGCCTGTACT-3′; GAP43 forward 5′-GAGGATGCTGCTGCCAAG-3′, reverse 5′-GGCACTTTCCTTAGGTTTGGT-3′; TUBB3 forward 5′-CCTGGAACCCGGAACCAT-3′, reverse 5′-AGGCCTGAAGAGATGTCCAAAG-3′; NPY forward 5′-TACCCCTCCAAGCCGGACAA-3′, reverse 5′-CATTTTCTGTGCTTTCTCTCAT-3′. All primers were synthesized by Sangon Biotech (Shanghai, China).

Cells with different treatments were incubated with BrdU (BD Biosciences, San Jose, CA, USA) at a final concentration of 10 μM in the cell culture medium for 4 h. The cells were harvested and washed with PBS. After fixation and permeabilization, the cells were treated with 300 μg/ml DNase (Roche). The incorporated BrdU was stained with anti-BrdU-FITC antibody (BD Biosciences) and then analyzed by flow cytometry.

Cell proliferation was monitored using CCK8 (Dojindo, Kumamoto, Japan) according to the manufacturer’s instructions. The cells were seeded onto 96-well plates, and cell proliferation was assessed at the indicated time points by measurement of the absorbance at 450 nm.

Statistical analyses were performed using SPSS version 18.0 software for Windows. The association of UCHL1 expression with clinical pathologic characteristics was analyzed by the chi-square criterion test. Survival analysis was assessed by Kaplan-Meier analysis together with single variable or multivariate Cox analysis. All measurement data are presented as mean ± S.E.M. Statistical significance was evaluated using unpaired nonparametric test. Significance was expressed as: * P <  0.05, ** P <  0.01, and *** P <  0.001.

To determine the potential clinic implication of UCHL1 in NB, TMA of neuroblastic tumors obtained from Xinhua Hospital was analyzed by IHC for UCHL1 protein expression. As shown in Fig. 1a-c, UCHL1 was differentially expressed in different samples of neuroblastic tumor patients, according to which patients were classified into UCHL1 high or UCHL1 low based on IHC score. 46.9% of patients demonstrated low UCHL1 expression (IHC score: 0–2), while 53.1% demonstrated high UCHL1 expression (IHC score: 3–4). Then the clinical pathologic characteristics and overall survival (OS) rates were evaluated with respect to UHCL1 expression. There was no significant association of UCHL1 expression with clinicopathologic parameters, such as age at diagnosis, stage, primary site and bone marrow metastasis (Table 1), while UCHL1 expression was higher in tumors of stage 1 than tumors of stage 4 (Fig. 1d). Kaplan-Meier analysis demonstrated that patients with high UCHL1 expression in the TMA cohort showed significantly good OS (3 years OS; UCHL1 high vs. low: 80.6% vs. 51.5%; P = 0.0222; Fig. 1e) as compared to the ones with low UCHL1 expression. Further analysis suggested UCHL1 could be scored as a continuous variable retained prognostic significance for OS in single variable analysis (hazard ratio [HR]: 0.698, 95% confidence interval [CI]: 0.506–0.964, P = 0.029) and multivariable analysis (HR: 0.738, 95% CI: 0.535–1.018, P = 0.064; Tables 2 and 3). In addition, high expression of UCHL1 also showed significantly good OS in NB patients among the TMA cohort (3 years OS; UCHL1 high vs. low: 83.3% vs. 45.0%; P = 0.0317; Fig. 1f). Together, these indicates that higher UCHL1 expression in NB patients is associated with better clinical outcome.

To further evaluate UCHL1 as a potential prognostic factor in NB, we reanalyzed three independent patient datasets (Oberthuer, Versteeg and Seeger) to examine the correlation between the mRNA expression of UCHL1 and the survival rate in NB. In the Oberthuer dataset, patients with high UCHL1 mRNA levels showed better OS (15-year OS; UCHL1 high vs. low: 91.4% vs. 77.6%; P = 0.004) and event-free survival (EFS; 15-year EFS; UCHL1 high vs. low: 79.3% vs. 62.6%; P = 0.004) (Fig. 2a). In multivariable analysis, increment in every unit of UCHL1 mRNA correlated with a decrease in HR for OS (HR: 0.338, 95% CI: 0.116–0.980, P = 0.046) and EFS (HR: 0.476, 95% CI: 0.228–0.994, P = 0.048) independent of established risk markers (MYCN status, tumor stage and patient age at diagnosis; Table 4). Moreover, decreased UCHL1 expression correlated significantly with advanced tumor stages (Fig. 2b). By Kaplan-Meier analysis and multivariable analysis, we also confirmed that high UCHL1 expression is prognostic for favorable outcome in the Versteeg and Seeger dataset (Fig. 2c and d; Tables 5 and 6). Taken together, our analyses of three independent microarray datasets also indicate that UCHL1 is a potential prognostic marker in NB.

A critical role of UCHL1 in neurogenesis in the embryonic brain has been reported [25], we then analyzed the association of UCHL1 with NB tumor differentiation in the TMA cohort. In TMA samples, 10 of 19 GNB samples and 12 of 18 well-differentiated NB samples demonstrated high UCHL1 expression, compared to 5 of 20 poorly differentiated NB samples (Table 7). Compared with poorly differentiated NB, well-differentiated NB and GNB had higher expression of UCHL1 (Fig. 3a). We further explored UCHL1 expression from another set of samples analyzed by mass spectrometry quantitative proteomics, which comprised of 37 NB and 40 GNB/GN (detail data were not shown). UCHL1 expression was significantly higher in GNB/GN and well-differentiated NB than poorly differentiated NB (Fig. 3b). Immunoblotting analysis also showed that the protein expression of UCHL1 was higher in GNB/GN than NB. Furthermore, UCHL1 expression was positively correlated with the expression of TH, a known differentiation marker in NB (Fig. 3c and d). Similar results were obtained in an independent dataset (Asgharzadeh dataset) of 209 neuroblastic tumors when these data were reanalyzed for UCHL1 mRNA (Fig. 3e). To further confirm the association between UCHL1 and NB differentiation, we investigated the correlation of UCHL1 expression levels with differentiation states in primary NB from the Seeger microarray dataset. When the tumors were separated into UCHL1 high and low expression groups, tumors of the high UCHL1 group expressed significantly higher mRNA levels of differentiation markers such as GAP43, TH, neuronal enolase 2 (ENO2) and dopamine-b hydroxylase (DBH) (Fig. 3f). Similar results could be found in the Oberthuer and Versteeg datasets (Additional file 1: Figure S1). Taken together, our findings illustrate an association of high UCHL1 expression with NB differentiation. We went on to examine how UHCL1 affects NB cell behavior and to explore its signaling functions.

To gain insight into the role of UCHL1 in neuronal differentiation of NB cells, we used RA, a differentiation agent for clinical treatment of high-risk NB after terminating the chemotherapy, to induce neural differentiation of the NB cell lines, SH-SY5Y and SK-N-BE (2). Results from Real-time PCR and immunoblotting analysis confirmed that UCHL1 was upregulated in SH-SY5Y and SK-N-BE (2) cells upon RA administration (Fig. 4a and b). To evaluate UCHL1 regulating neuronal differentiation, UCHL1 expression was reduced in SH-SY5Y and SK-N-BE (2) cells using a lentivirus-expressing shRNA specific to UCHL1 (named shUCHL1), and SH-SY5Y and SK-N-BE (2) cells infected with a lentivirus-expressing scrambled shRNA (named shNC) were used as controls (Fig. 4c and Additional file 2: Figure S2). As compared with shNC cells, RA-induced neurite outgrowth was inhibited in shUCHL1 SH-SY5Y (Fig. 4d). The reversible, competitive, active-site directed inhibitor of UCHL1, LDN57444, was next used to examine the regulation of UCHL1 on neuronal differentiation. Compared with DMSO-pretreated SH-SY5Y cells, LDN57444-pretreated SH-SY5Y cells had poor capacity of RA-induced neuronal differentiation (Fig. 4e). What’s more, similar phenomenon was also exhibited in SK-N-BE (2) cells (Fig. 4f). To further confirm the function of UCHL1 on RA-induced neuronal differentiation, we detected several established neural differentiation relevant proteins or genes. As shown in Fig. 5a and b, the inhibited neurite outgrowth in shUCHL1 SH-SY5Y and SK-N-BE (2) cells was accompanied by downregulation of expression of neuron-specific protein, TH and GAP43. In addition, the established neural differentiation relevant genes, including TH, GAP43, TUBB3, and NPY were down-regulated in RA-treated SH-SY5Y and SK-N-BE (2) cells after UCHL1 knockdown in comparison to the controls (Fig. 5c and d). Similar changes of neural differentiation relevant proteins or genes were also detected in LDN57444-pretreated SH-SY5Y cells (Fig. 5e and f). These results together suggest that UCHL1 promotes neuronal differentiation of NB cells.

To explore the potential molecular mechanisms by which UCHL1 promoted neuronal differentiation, we performed immunoblotting analysis to determine the activities of signaling pathways including AKT and ERK1/2, in cells. These signaling pathways are known as downstream of RA signaling and critical for RA-induced differentiation [32, 33]. As shown in Fig. 6a and b, UCHL1 knockdown dramatically inhibited the phosphorylation of AKT at Ser473 and Erk1/2 at Thr202/Tyr204 after RA administration in both SH-SY5Y and SK-N-BE (2) cells. These results suggest that the effect of UCHL1 inhibition on RA-induced neuronal differentiation is mediated by repression of AKT and ERK1/2 activities.

UCHL1 could either positively or negatively regulate cell proliferation of cancer cells [34]. To further characterize the potential function of UCHL1 in NB, we examined whether UCHL1-regulated cell differentiation was coupled with regulation of NB cell proliferation. BrdU and CCK8 assays showed that there was no significant difference of the proliferation between shNC and shUCHL1 SH-SY5Y cells (Fig. 7a and b). In addition to promoting NB neural differentiation, RA also could induce NB cell growth arrest [6, 7]. Here we found that UCHL1 knockdown reversed RA-induced cell growth arrest of SH-SY5Y cells, characterized by increased percentage of BrdU⁺ cells and cell viability in shUCHL1 SH-SY5Y cells in the presence of RA (Fig. 7a and b). Next, we tested the levels of the cell cycle regulators p21 and cyclin D1, which regulated cell proliferation. As shown in Fig. 7c, as compared with shNC SH-SY5Y cells, RA-upregulated cell cycle-dependent kinase inhibitor p21 level and RA-downregulated cyclin D1 level was inhibited in shUCHL1 SH-SY5Y cells, consistent with the effect of UCHL1 on RA-induced inhibition of cell proliferation. Besides, we also examined the effects of UCHL1 on RA-repressed proliferation of SK-N-BE (2) cells, and showed the similar phenomenon in Fig. 7d and e. These results indicate that UCHL1 also promotes RA-repressed proliferation of NB cells.