Expression and prognostic impact of NTF3 and TrkC in hepatocellular carcinoma

Abstract Background Treatment of patients with NTRK fusion-positive cancers using first-generation tropomyosin-related kinase (Trk) inhibitors is associated with high response rates, regardless of tumor histology. However, there have been few studies on neurotrophin-3 (NTF3) and TrkC ligands in hepatocellular carcinoma (HCC). Methods We used immunohistochemistry to evaluate NTF3 and TrkC expression levels in tissue samples. Gene expression profiling interactive analysis was used to determine TrkC and NTF3 expression in HCC. Western blotting, quantitative reverse transcription polymerase chain reaction, and enzyme-linked immunosorbent assays were utilized to analyze TrkC and NTF3 levels in HCC cell lines. Proliferation tests and cell migration were also explored. Results NTF3 and TrkC levels were lower in HCC tissue (median H- scores 149.09 and 54.60, respectively) than those in para-cancerous tissue (192.69 and 71.70, respectively); no statistical difference was found in the survival rate. Positive correlations were observed between NTF3 and TrkC levels in both HCC and para-cancerous tissues. Alpha-fetoprotein was the only clinical characteristic associated with TrkC levels. The transcription of NTF3 was lower in HCC samples compared to normal samples. NTF3 overexpression inhibited the proliferation of MHCC97-L and HepG2 cells but did not significantly affect cell migration. Conclusions The transcription of NTF3 was lower in HCC samples compared to normal samples, indicating a potential association with disease-free survival and overall survival in HCC. NTF3 and TrkC expression levels were lower in HCC tissues than those in para-cancerous tissues. Our results indicate that NTF3 may be a prognostic factor for HCC.


Introduction
Primary liver cancer is the sixth most commonly diagnosed cancer and the third leading cause of cancer-related deaths worldwide, with approximately 906,000 new cases and 830,000 deaths in 2020 [1].The prognosis of liver cancer is poor, and only 5-15% of patients are eligible for surgical resection [2].Therefore, investigating the mechanisms of liver cancer and identifying novel therapeutic targets is urgently needed to improve prognosis.
Neurotrophins (NTFs), including nerve growth factor (NGF), brain-derived neurotrophic factor (BDNF), NTF3, NTF4, and NTF5, were originally identified as promoters of neuronal survival [3].NTFs are ligands for receptor tyrosine kinases of the tropomyosin-related kinase (Trk) family.The three most common types of Trk receptors are TrkA, TrkB, and TrkC.TrkA preferentially binds NGF, TrkB binds BDNF and NTF4 or NTF5, and TrkC binds NTF3 [4].Pan-cancer analysis revealed that NTF signaling positively correlates with anti-tumor immunity, clinical outcomes, and the response to targeted therapies and immunotherapies in cancer [5].
Trk activation stimulates cancer cell proliferation, aggressiveness, and metastasis [6].NTF3, a member of the NTF family, promotes neuronal growth and development, maintains neuronal morphology, regulates neuronal functions, and repairs damaged neurons [7].TrkC activates neuronal survival pathways, including the Ras/MEK/MAPK and PI3K/AKT pathways, which promote cellular functions such as proliferation, growth, and survival in cancer [8].
In this study, we evaluated NTF3 and TrkC expression levels in liver cancer and normal tissues.Furthermore, we constructed an NTF3 overexpression plasmid and investigated the effects of NTF3 on hepatocellular carcinoma (HCC) cell proliferation and migration to better understand the functions of NTF3 and TrkC in HCC.
The microarray included 197 tissue cores with 97 pairs of HCC and para-cancerous samples and three unpaired HCC samples.Patient clinical data provided by the company were used to obtain information on clinical diagnoses and tumor-node-metastasis (TNM) stages.The patients had undergone surgical resection between July 2010 and March 2012, and complete clinical and follow-up data until November 2017 was available.All patients provided written informed consent.

Immunohistochemical staining of NTF3 and TrkC in the hepatocellular tissue microarray
NTF3 and TrkC levels in tissue samples were evaluated using immunohistochemistry (IHC) according to the standard streptavidin-peroxidase method.Briefly, histological slides of tissue samples were deparaffinized using xylene, followed by an alcohol gradient.The antigen was retrieved using citric acid antigen repair buffer (pH 6.0; ZSGB-BIO, Beijing, China).Endogenous peroxidase activity was blocked by incubating the slides in a 3% H 2 O 2 solution (Kit I; ZSGB-BIO) prepared in methanol at approximately 25 °C for 10 min.Next, the slides were washed with phosphate-buffered saline (PBS, pH 7.4) and blocked using a blocking buffer [10% fetal bovine serum (FBS; ZSGB-BIO, Beijing, China) in PBS].The slides were subsequently incubated overnight at 4 °C with rabbit anti-human polyclonal antibody against NTF3 (ab53685, dilution 1:500; Abcam, Cambridge, UK) and rabbit anti-human monoclonal antibody against TrkC (C44H5, dilution 1:500; Cell Signaling Technology, Danvers, MA, USA).The slides were then washed and incubated with polyclonal anti-rabbit immunoglobulin G secondary antibody (Kit II; ZSGB-BIO).Immunocomplexes were detected using 3,3′-diaminobenzidine (ZSGB-BIO).Staining was monitored using a microscope (Leica, Wetzlar, Germany) and terminated when sufficient staining was achieved, after which the slides were dehydrated and stored.Tissue images were acquired and analyzed using CaseViewer (Danjier, Shandong, China).The H-score for each sample, defined as a continuous variable, was calculated by a pathologist without clinical or molecular data using the following formula: [9].All stained tumor samples yielded H-scores ranging from 0 to 300.The median H-score for HCC tissues were used as the threshold to designate high and low NTF3 and TrkC expression levels.

Analysis of NTF3 and TRKC expression and clinical characteristics using gene expression profiling interactive analysis (GEPIA)
GEPIA (http://gepia.cancer-pku.cn/index.html),a commonly used interactive tool comprising 9,736 tumors and 8,587 normal tissue samples from TCGA and the Genotype-Tissue Expression (GTEx) databases, was used to analyze TrkC and NTF3 expression and determine any association with HCC [10].The log-rank test was performed, and the resulting hazard ratio (HR) and p or Cox p-values were plotted.

Cell culture
Huh7, HepG2, and Hep3B cells were obtained from our laboratory at the Second Affiliated Hospital, Zhejiang University School of Medicine.HCCLM3 and MHCC97-L cells were obtained from Xuanwu Hospital, Capital Medical University.The cells were cultured in Dulbecco's Modified Eagle Medium (DMEM; Gibco, Grand Island, NY, USA) supplemented with 10% FBS (Gibco), 100 units/mL penicillin, and 100 µg/mL streptomycin (Gibco) in a humidified atmosphere at 37 °C and 5% CO 2 .

Enzyme-linked immunosorbent assay (ELISA) analysis of NTF3 levels in cell serum
An ELISA kit was purchased from Jianglaibio (JL11950; Shanghai, China).Cells from each cell line were inoculated into 24-well plates for 24 h.Cell supernatants were centrifuged at 1000 × g for 20 min, and 50 μL of supernatant was added to the plates.Thereafter, different concentrations of standard substances (800, 400, 200, 100, 50, and 25 pg/mL) and diluent (blank control) from the ELISA kit were added.Next, horseradish peroxidase (100 µL) was added and the plates were incubated for 60 min at 37 °C.Thereafter, the liquid in the plate was discarded, and the wells were washed five times.Solvents A and B were added and incubated for 15 min, and the reaction was terminated by the addition of an ending solution.The absorbance was immediately read at 450 and 630 nm using a microplate reader (SpectraMax M3; Molecular Devices, Silicon Valley, CA, USA).

Cell transfection
Primers were designed for two NTF3 subtypes to construct an NTF3 overexpression plasmid.The primer sequences were as fo l l o w s : NTF3-201, fo r w a r d : 5 ′ -G CC ACC ATG T CCATCTTGTTTTATGTGA-3′ and reverse: 5′-TGTTCT TCCGATTTTTCTCGAC-3′ and NTF3-202, forward: 5′-GCCAC CATGGT TAC T T T TGCCACGATC T-3′ and reverse: 5′-TGTTCTTCCGATTTTTCTCGAC-3′.PCR amplification showed that Huh7 cells expressed the NTF3-201 subtype.The PCR products were ligated into the pCDH plasmid vector (MiaoLing Plasmid Platform, Wuhan, China) containing the green fluorescent protein copGFP and then transformed into Escherichia coli DH5α (MiaoLing Plasmid Platform).After PCR and sequencing, the plasmids were purified for transfection.Lipofectamine ® LTX (15338100; Thermo Fisher Scientific) was used to transfect the cells according to the manufacturer's protocol.Briefly, 4 µg of pCDH vector (pCDH-control) or pCDH-NTF3 plasmid was mixed with 5 µL of Lipofectamine and incubated for 25 min at approximately 25 °C.The mixture was transfected into 90% confluent MHCC97-L and HepG2 cells for 1 h.The transfected cells were cultured at 37 °C in DMEM for 24 h, and green fluorescence was observed using a fluorescence microscope (Leica).

Cell counting kit-8 (CCK-8) assay
We used a CCK-8 kit (Dojindo, Shanghai, China) to measure cell proliferation capacity.The cells were seeded in 96-well plates at a density of 5000 cells/well.After incubation, 10 μL of CCK-8 solution was added to each well at different times.Absorbance was measured at 450 nm using a microplate reader (SpectraMax M3; Molecular Devices).

Colony formation assay
A colony formation assay was performed to analyze cell proliferation.Briefly, the cells were seeded in 6-well plates at a density of 1000 cells/well and incubated for 14 days.Thereafter, the colonies were fixed with 10% formaldehyde for 30 min and stained with 0.1% crystal violet (C0121; Beyotime) for 30 min at approximately 25 °C.Subsequently, colonies containing > 50 cells were counted, and the mean colony size was calculated.Each clone was plated in triplicate for each experiment.Cloning formation rate (%) = (number of clones/number of inoculated cells) × 100%.

Wound healing assays
For the wound healing assay, cultured cells in DMEM containing 10% FBS were grown to 100% confluence on plastic dishes with culture inserts (81176; Ibidi, Gräfelfing, Germany).The culture inserts were removed after 12 h.The cells were viewed and photographed at 0, 24, and 48 h after wounding using a microscope (Leica).Wound closure was quantitatively analyzed using ImageJ software.Each test was performed in triplicate.Cell mobility (%) = (initial scratch width -x scratch width after h)/initial scratch width value × 100%.

Transwell migration assay
For the migration assay, 5 × 10 5 cells suspended in 100 μL serum-free medium were added to the upper transwell chamber (8 mm, Corning Costar; Corning Life Sciences, Corning, NY, USA), and 600 μL of medium supplemented with 20% serum was placed in the lower chamber.After incubation, the cells on the upper surface of the membrane were gently removed with a cotton swab.Cells invading the lower surface were fixed with methanol and stained with 0.1% crystal violet (C0121; Beyotime) for 30 min at approximately 25 °C.Stained cells were counted in five randomly selected fields using a light microscope (Leica, Wetzlar, Germany).Each clone was plated in triplicate for each experiment.

Statistical analysis
All statistical analyses were performed using SPSS software (version 22.0; SPSS Inc., Chicago, IL, USA).For descriptive statistics, data distributions were analyzed using the Shapiro-Wilk test.Normally distributed variables are expressed as mean ± standard deviation and were analyzed using Student's t-test.Variables with a skewed distribution are expressed as medians (interquartile ranges).H-scores are expressed as medians (interquartile ranges) and were analyzed using the Mann-Whitney U test.The Spearman correlation coefficient was used for the correlation analysis between NTF3 and TrkC.TrkC and NTF3 expression levels and clinicopathological features of patients with HCC were analyzed using the chi-squared test and multivariate logistic regression analysis.The roles of TrkC and NTF3 in cell survival were assessed using the Kaplan-Meier method.Regarding RT-qPCR, Trkc data were analyzed using single sample parameter analysis, and NTF3 data were analyzed using the Kruskal-Wallis H-test.Statistical significance was set at p < 0.05.

NTF3 and TrkC expression levels in HCC and paracancerous tissues
Of the 97 pairs of HCC and para-cancerous samples and three unpaired HCC samples, two HCC and one para-cancerous sample were defoliated during IHC.As a result, the final analysis included 98 HCC and 96 para-cancerous samples, consisting of 94 pairs, four unpaired HCC samples, and two unpaired para-cancerous tissues.Microarray samples are described in detail in Table S1.IHC staining images of NTF3 and TrkC expression levels in HCC and para-cancerous tissues are presented in Figure 1(A).Positive staining was observed in the cytoplasmic and membranous areas.For HCC tissue, the median H-score of NTF3 was 149.09 [interquartile range (IQR), 174.39-114.16],and that of TrkC was 54.60 (IQR, 71.48-37.48).These scores were significantly lower than those for para-carcinoma tissues at 192.69 (IQR, 203.91-176.27)and 71.70 (IQR, 83.82-60.26),respectively (p < 0.001, Figure 1(B)).

Correlation analysis between NTF3 and TrkC levels in HCC and para-cancerous tissues
The correlation between NTF3 and TrkC expression levels in HCC and para-cancerous tissues was investigated using Spearman's correlation coefficient.A positive correlation was observed between NTF3 and TrkC expression in both HCC (Spearman, r 2 = 0.426, p < 0.001) and para-cancerous (Spearman, r 2 = 0.487, p < 0.001) samples (Figure 1(C)).

Comparison of NTF3 and TrkC expression levels in HCC tissues of patients with different clinical characteristics
In chi-squared test analysis, statistical differences were not observed between the high and low expression of NTF3 in the following categories: sex, age, hepatitis B virus (HBV) infection, liver cirrhosis, cancer embolus status, alpha-fetoprotein (AFP) levels, and TNM stage of patients with HCC (p > 0.05).In addition, there were no statistical differences between the high and low TrkC expression groups regarding patient sex, age, HBV infection, liver cirrhosis and cancer embolus status, or TNM stage.AFP was the only clinical characteristic associated with TrkC expression levels (p < 0.001; Table 1).

Transcriptional analysis of NTF3 and TrkC expression in HCC and normal samples
For GEPIA analysis, tissue samples were divided into two groups as follows: 369 HCC and 160 normal samples.Notably, TrkC expression was lower in HCC samples than that in normal samples; however, this difference was not statistically significant (Figure 2(A)).Furthermore, we observed a discrepancy in NTF3 expression between the HCC and normal samples (p < 0.05) (Figure 2(B)); in particular, NTF3 expression was lower in HCC samples than that in normal samples.In both HCC and normal samples, the expression of TrkC and NTF3 was correlated (p = 7.2e-06 and p = 0.023, respectively, Figure 2(C-D)).These findings indicated that TrkC mRNA expression was independent of the HCC stage (p = 0.931) and prognosis (disease-free survival, DFS, p = 0.420; overall survival, OS, p = 0.870) (Figure 2(E-G)).We did not find a significant correlation between NTF3 mRNA expression and HCC prognosis for different HCC stages (p = 0.579); however, we observed a correlation between NTF3 mRNA expression and DFS and OS (p = 0.009 and p = 0.003, respectively) (Figure 2(H-J).

NTF3 and TrkC expression levels in HCC cell lines
Western blot analysis revealed that TrkC was only expressed in MHCC97-L cells (Figure 3(A)).Furthermore, RT-qPCR analysis revealed that TrkC mRNA expression was higher in HepG2 cells than that in Huh7 and HCCLM3 cells (p < 0.05, Figure 3(A).We also observed a statistically significant difference in NTF3 mRNA expression between HepG2 and Hep3B cells (p < 0.05, Figure3(A)).ELISA was used to assess NTF3 secretion; NTF3 concentrations in MHCC97-L cells were significantly lower than those in Huh7, Hep3B, and HepG2 cells (p = 0.044, p = 0.017, p = 0.008, respectively), as shown in Table 2.

NTF3 overexpression inhibited MHCC97-L and HepG2 cell proliferation
As shown in Figure 3(B), the absorbance of MHCC97-L and HepG2 cells in the control, pCDH-vector, and pCDH-NTF3 groups was not significantly different at 0 h.However, the absorbance of the pCDH-NTF3 group was significantly lower than that of the control and pCDH-vector groups at 24, 36, and 48 h (p < 0.05).These results suggest that NTF3 overexpression inhibited the proliferation of MHCC97-L and HepG2 cells.After Giemsa staining, the cells were counted and analyzed.The clonogenicity of HepG2 cells was significantly lower in the pCDH-NTF3 group than that in the control and pCDH-vector groups (Figure 3(B)).Regarding MHCC97-L cells, the clones in the pCDH-NTF3 group differed from those in the other groups (p < 0.05).

NTF3 overexpression had no significant effect on MHCC97-L and HepG2 cell migration
Cell migration was detected using wound healing and transwell assays.Regarding MHCC97-L cells, no significant difference was observed in the migration rate of the control and pCDH-NTF3 groups at 12 h (p = 0.530) and 24 h (p = 0.796) (Figure 3(C)).HepG2 cell mobility was similar in the control, pCDH-vector, and pCDH-NTF3 groups.At 12 h, the p-value of 0.144 illustrated the difference between the control and pCDH-NTF3 groups and the p-value 0.151 indicated the difference between the pCDH-vector and pCDH-NTF3 groups.At 24 h, no discrepancy was observed between the control and pCDH-NTF3 groups (p = 0.456).After 36 h, no discrepancy was observed in HepG2 cells between the pCDH-NTF3 and control (p = 0.359) and pCDH-vector (p = 0.603) groups.As shown in Figure3(D), after 48 h, no discrepancy was observed in MHCC97-L cells between the pCDH-NTF3 and control (p = 0.139) or pCDH-vector (p = 0.291) groups.Therefore, NTF3 overexpression had no significant effect on the migration ability of MHCC97-L and HepG2 cells.

Discussion
According to the GLOBOCAN 2020 database, there were an estimated 431,383 new cases and 412,216 deaths from liver cancer in 2022 [1].The prognosis of liver cancer is poor, and > 90% of all liver cancers are HCC, for which chemotherapy and immunotherapy are the best options.Trans-arterial chemoembolization improves 2-year survival in patients with intermediate-stage HCC by 23% when compared to conservative therapy [2].Therefore, it is particularly important to research new therapeutic measures for improving the prognosis HCC.NTRK1, NTRK2, and NTRK3 encode TrkA, TrkB, and TrkC, respectively.NTRK fusions are exceedingly rare, occurring in 0.31% of adult and 0.34% of pediatric tumors.Although NTRK fusions are rare, other NTRK abnormalities affect 14% of patients with cancer [11].The Trk expression levels were not associated with the copy number of each different NTRK, and NTRK fusion was not detected in Trk-overexpressing HCC.The Trk overexpression group tended toward shorter recurrence-free survival (p = 0.092) and OS (p = 0.079) than the low expression group, with a frequent multicentric occurrence, which was statistically significant in specific subpopulations [12].One study reported that TrkC mRNA expression was increased in all five tumors (100%) compared to their matched normal tissue samples [13].In our study, we analyzed TrkC mRNA expression using GEPIA with 369 HCC and 160 normal samples.We observed that TrkC expression was lower in HCC samples than in normal samples, but this difference was not statistically significant.Studies with large sample can help to minimize errors and provide more accurate information on mRNA expression levels.Using GEPIA, we also found that NTF3 mRNA expression was lower in HCC samples than in normal samples, and this was associated with decreased DFS and OS (p = 0.009 and p = 0.003, respectively) in patients with HCC.Another study reported that NTF3 expression was significantly reduced in patients with HCC based on the Gene Set Enrichment Analysis database.They also found that low NTF3 expression was associated with poorer outcomes, including shorter OS, recurrence-free survival, progression-free survival, and disease-specific survival, in 74 patients with HCC [14].Yang et al. discovered that NTF3 was significantly downregulated in patients with HCC based on analysis of both the TCGA and GEO databases.Low NTF3 expression was also found to be correlated with shorter OS and DFS in patients with HCC.In addition, they observed decreased expression of NTF3 in the majority of HCC samples (21/32, 65.63%), but no significant relationship was found between NTF3 protein levels and clinicopathologic features [15].These results are consistent with our own research findings regarding NTF3 mRNA expression.
Liu et al. discovered that the majority of samples in the HCC cohort [59 cases (73.75%)] exhibited low expression of NTF3, whereas a small number of samples [21 cases (26.25%)] demonstrated high expression [16].We conducted a follow-up study on NTF3 expression in tissue samples and found that low NTF3 expression was associated with a low median OS, although the difference was not statistically significant.In both HCC and normal samples, there was a correlation between NTF3 and TrkC expressions.To further investigate this correlation, we analyzed NTF3 and TrkC expression levels in HCC cell lines using western blotting, qRT-PCR, and ELISA.We observed that overexpression of NTF3 inhibited the proliferation of MHCC97-L and HepG2 cells, but did not have a significant effect on cell migration.The binding between TrkC and NTF3 results in a preferential activation of the PI3K/AKT pathway, which prevents apoptosis and increases cell survival [17].Liu et al. through CIBERSORT analysis, showed that NTF3 expression is positively correlated with CD4+ cells, mast cells, NK cells, macrophages, and B cells in the tumor microenvironment.Additionally, NTF3 expression is negatively correlated with immune checkpoints PD-L1, TIGIT, and TIM-3 [17].Another study, conducted through in vivo and in vitro experiments, revealed that NTF3 significantly inhibits the progression of HCC cells.This inhibition is regulated by c-Jun, which binds to the p75 neurotrophin receptor (p75NTR) and activates the JNK and P38 MAPK pathways to induce apoptosis [14].
This study had some limitations in terms of interpreting the results.First, we only used GEPIA to analyze TrkC and NTF3 mRNA expression without conducting clinically relevant trials.GEPIA is an interactive web application for gene expression analysis, comprising 9736 tumors and 8587 normal samples from the TCGA and GTEx databases, and uses the output of a standard processing pipeline to obtain RNA sequencing data  [10].We believe that this database contains a large number of samples to analyze for gene expression.Second, we only explored the effect of NTF3 on HCC cell proliferation and migration and not that of TrkC.TrkC expression did not significantly differ with different clinical characteristics, with the exception of AFP levels, and also did not affect the survival rate of patients with HCC.Furthermore, TrkC expression was observed only in MHCC97-L cells.In future studies, we will explore TrkC expression in HCC by expanding the cell lines.

Conclusion
In summary, the transcription of NTF3 was lower in HCC samples compared to that of normal samples.The low NTF3 mRNA expression observed in HCC samples may indicate poor DFS and OS.NTF3 and TrkC expression levels were lower in HCC tissues than those in para-cancerous tissues, and a positive correlation was observed between NTF3 and TrkC mRNA and protein expression levels.However, NTF3 and TrkC expression levels had no significant correlation with the survival rate.Of all the clinical characteristics studied only AFP levels were related to TrkC expression levels.Lastly, NTF3 overexpression inhibited MHCC97-L and HepG2 cell proliferation but had no significant effect on cell migration.Our results indicate that NTF3 may be a prognostic factor for HCC.

Figure 1 .
Figure 1.immunohistochemical analysis of ntf3 and trkc in hepatocellular carcinoma (Hcc) and para-cancerous tissues.(a) immunohistochemical staining of ntf3 and trkc.(B) the H-score of ntf3 and trkc in Hcc and para-cancerous tissues.(c) the correlation analysis between ntf3 and trkc expression levels in Hcc and para-cancerous tissues.(d) Survival analysis of ntf3 and trkc in Hcc.

Figure 2 .
Figure 2. GePia was used to analyze NTF3 and TrkC gene expression and prognosis in hepatocellular carcinoma (Hcc) and normal samples.(a-B) NTF3 and TrkC gene expression.(c-d) the correlation analysis between NTF3 and TrkC gene expression in Hcc and normal samples.(e-G) the analysis of TrkC on tumor stage, disease-free survival (dfS), and overall survival (oS) in Hcc.(H-J) the analysis of NTF3 on tumor stage, dfS, and oS in Hcc.

Figure 3 .
Figure 3. ntf3 and trkc expression levels in Hcc cell lines were tested using western blotting and rt-qPcr (a).ntf3 overexpression inhibited MHcc97-l and HepG2 cell proliferation in the cell counting kit-8 (ccK-8) and colony formation assays (B). the wound healing assays of ntf3 on MHcc97-l and HepG2 cells (c). the transwell migration assay of ntf3 on MHcc97-l and HepG2 cells (d).

Table 1 .
association the expression of ntf3 and trkc with clinicopathological features in hepatocellular carcinoma patients.