Immunosuppressive receptor LILRB1 acts as a potential regulator in hepatocellular carcinoma by integrating with SHP1

2.1 Tissue microarray and Immunohistochemistry

The tissue microarray containing 75 pairs of liver carcinoma tissue and the paired para-carcinoma tissue was purchased from OUTDO Biotechnology co., LTD (Shanghai, China). For immunohistochemistry (IHC) assay, slides were heated in an oven for 30 minutes at 60 ∘ C. After deparaffinization in xylene and rehydration in gradient ethanol, LILRB1 antigen was retrieved by 0.01 M citrate salt solution (PH 6.0) pressure antigen retrieval. Then, the slides were cleaned by PBST and blocked with 5% BSA for 1 h at room temperature (RT). Primary antibody against LILRB1 (ab185325, abcam, Cambridge, UK) was employed for the incubation with the slides at 4 ∘ C overnight. The next day, the endogenous peroxidase was neutralized by 30% H 2 O 2 for 15 min. Then sections were incubated with secondary antibody and streptavidin horseradish peroxidase (HRP) for 15 min at RT (PV-9000, ZSGB-BIO, China). Antigen-antibody complexes were visualized after staining with DAB (ZL1-9081, ZSGB-BIO, China). Photograph observation was performed under a Biological inverted microscope (IX51, Olympus, Japan). Comprehensive analysis included measuring staining intensity and the number of positive cells (negative: - ; weakly positive: + , 20%; middling positive: + ⁣ + , 20%–50%; strong positive: + ⁣ + ⁣ + , $>$ 50%).

2.2 Cell culture and treatment

The human immortalized normal hepatocyte cell line-LO2 and three HCC cell lines including HepG2, Huh7, and MHCC97L were purchased from Shanghai cell bank, Chinese Academy of Sciences. These cells were cultured with complete Dulbecco’s modified Eagle’s medium (DMEM; 11320-033, GIBCO, USA) supplemented with 10% fetal bovine serum (10091-148, GIBCO, USA), 100 U/ml penicillin and 100 μ g/ml streptomycin (Sigma, USA) in a 37 ∘ C humidified incubator with 5% CO 2 .

2.3 Data collection, correlation and survival analysis

The expressions of LILRB1 mRNA and Shp1 mRNA in normal liver tissues or precancerous tissue were collected from Oncomine database (Wurmbach Liver Statistics) and analyzed by GEPIA version 2. In Wurmbach Liver Statistics database, 10 normal, 30 liver cancer precursor and 35 HCC patients were enrolled in our correlation analysis between LILRB1 and pathological types. Expression value of LILRB1 which was indicated as the value of Log2 median-centered intensity in these specimens were used to assess the Pearson correlation of LILRB1 and SHP1 or SHP2 expression by SPSS software in different liver tissues. GEPIA was used to analyze the Overall Survival and Disease Free Survival.

2.4 Western blotting

Total protein extracts were prepared using RIPA protein lysate (P0013B, Beyotime, China). 40 μ g protein was separated by 10% SDS-PAGE and transferred to an activated PVDF membrane (IPVH00010, Millipore, Thermo Scientific, USA). After blocking with 5% fat-free milk at room temperature for 1 h, the membrane was incubated with primary antibodies against LILRB1 (ab185325, abcam, Cambridge, UK), SHP1 (ab227503, abcam, Cambridge, UK) and SHP2 (ab110194, abcam, Cambridge, UK) at 4 ∘ C overnight. Actin (BM0627, Boster, China) served as the internal control. Next day, the membrane was incubated with Goat anti-Rabbit Secondary Antibody, HRP (BA1054, BOSTER, China) or Goat anti-Mouse Secondary Antibody, HRP (BA1051, BOSTER, China) for an additional 1 h at RT. After washing with TBST for 3 times, the proteins were visualized using the Imagequant LAS 4000 mini machine (GE Healthcare Life Sciences, USA) after incubating with ECL (NCI5079, Thermo Scientific, USA). All samples were subjected to at least 3 independent experiments.

2.5 Co-immunoprecipitation

OPEN IN VIEWER

Total protein lysates extracted from LO2 cells were incubated with antibodies against SHP1 (ab227503, abcam, Cambridge, UK) and isotype control IgG (ab172730, abcam, Cambridge, UK) overnight. And the complexes were incubated with protein A/G beads (88802, Thermo Scientific, USA) for another 3 h at room temperature. Then the immunoprecipitate was resuspended in 2x SDS loading buffer were boiled for 5 minutes and subjected to western blotting assay using LILRB1 (ab185325, abcam, Cambridge, UK) and SHP1 antibodies according to the above method.

Differences between clinical information and LILRB1 expression was analyzed by GraphPad 6.0 using chi-square and Fisher’s exact tests. Pearson correlation coefficient was used for analyzing the relation between LILRB1 and SHP1 or SHP2 by SPSS 25.0 software. Different degrees of staining of histological staining were analyzed by image pro plus. Quantitative analysis of protein and RNA was conducted by Quantity One and Graph Pad Prism using one-way ANOVA, followed by Student’s t -tests, respectively. 0.05 was considered statistically significant.

3.1 The expression of LILRB1 is depressed in liver cancer cells and liver cancer tissue array

LILRB1, as an immunoregulatory receptor, functions mainly as a key regulator in immune cells and hematologic malignant, but is seldom involved in the progress of solid cancer types [21]. To investigate the role of LILRB1 on HCC, we firstly determined its expression pattern in normal liver LO2 cells, and three liver cancer cell lines including MHCC97L, Huh7 and HepG2. As shown in Fig. 1A, the protein level of LILRB1 was drastically reduced in three liver cancer cell lines compared to that in the LO2 cells (Fig. 1A). Next, qRT-PCR was employed to further evaluate the transcription level of the LILRB1 gene in these cell lines. The results showed that relative expression of LILRB1 mRNA was significantly lower in the three cancer cell lines than that in normal liver cells (Fig. 1B).

To further ascertain the expression pattern of LILRB1 in hepatocellular carcinoma, a tissue microarray containing 75 pairs of HCC samples and paired adjacent liver tissue were employed to evaluate the level of LILRB1 in human specimens by immumohistochemical staining. The IHC results showed that the expression of LILRB1 was notably downregulated in most of the HCC samples compared to the paired adjacent liver tissue (Fig. 1C). Through IHC staining, we discovered that a low expression of LILRB1 ( + and - ) was found in 48 of 75 HCC tissues (64%), but we observed only 25 of 75 (33%) in the paracarcinoma tissue. Conversely, only 27 of 75 (36%) HCC specimens presented a high level of LILRB1 ( + ⁣ + and + ⁣ + ⁣ + ), compared to 50 of 75 (67%) in adjacent normal tissue (Table 1). Based on Multiple-factor analysis, we observed that there was a significant difference between LILRB1 expression and liver histologic subtype ( p = 0.0003) (Table 1). On the basis of IHC results, we also compared the LILRB1 expression in HCC with that in the paired adjacent liver tissue. The results showed that LILRB1 level was significantly higher in 49 of 75 (65%) paired paracarcinoma tissues than that in paired HCC samples. 15 of 75 subjects had no expression difference in LILRB1 both in HCC samples and paired paracarcinoma tissues and only 11 of 75 samples showed more expression of LILRB1 in HCC samples than that in their paired paracarcinoma tissues (Table S1). These data implies that LILRB1 is decreased in liver cancer cells and is possibly participates in the progression of liver cancer.

Next, we conducted the association between LILRB1 expression and clinicopathological characteristics of HCC patients in the tissue microarray. Among these 75 patients, we observed a significant decrease of LILRB1 in male subjects ( p = 0.0002), which was not found in female patients ( p = 0.6776) (Table 2). Though both different age ( ⩽ 50 and $>$ 50) and different tumor diameter ( ⩽ 5 cm and $>$ 5 cm) groups presented a significant correlation for LILRB1 in HCC tissues compared to normal control tissues, a more marked association was observed in aged subjects ( p = 0.0197 vs 0.0099) and large tumors ( p = 0.0422 vs 0.0018) (Table 2). However, only patients with higher pathologic grades (II - III + III, p = 0.0003) and clinical stages (T3 + T4, p = 0.0003) presented a significant association between relative LILRB1 expression and histologic subtype, and this relationship did not appear in patients with low pathologic (I - II + II) and clinical stage (T1 + T2) (Table 2). Besides, combined with Gene Expression Profiling Interactive Analysis (GEPIA version 2), we did not observe the correlation between LILRB1 expression and different cancer stages (Fig. 1D). Possibly, there is only significant correlation between LILRB1 level and high malignant HCC patients. These results indicate that frustrated LILRB1 is associated with part of the patients’ age, and pathologic grades, tumor diameter, and clinical stages.

OPEN IN VIEWER

OPEN IN VIEWER

Combined with the online Oncomine database, we further explored the expression profile of LILRB1 mRNA in HCC. The results showed that the expression of LILRB1 mRNA was much lower in HCC tissues ( n = 35) than that in liver cancer precursor tissues ( n = 30) and normal liver ( n = 10) (Fig. 2A). The decline of LILRB1 was also verified utilizing GEPIA. Once matched the TCGA normal data with Genotype-Tissue Expression (GTEx) data, we found that the expression of LILRB1 showed significantly lower in liver cancer subjects ( n = 369) than that in normal subjects ( n = 160) (Fig. 2B).

3.4 HCC patients with low LILRB1 have a short period of Disease Free Survival

To further prove the crucial determinant of LILRB1 in the development of HCC, we analyzed the impact of LILRB1 on survival rates. As shown in Fig. 3, among the 182 HCC patients including low LILRB1 group ( n = 91) and high LILRB1 group ( n = 91), median overall survival was similar and had no difference among both group (Fig. 3A). However, in the same population group, the median DFS was much higher in patients with low LILRB1 than that in HCC patients with high LILRB1 (Fig. 3B). These findings demonstrate that LILRB1 level is associated with Disease Free Survival of HCC patients, implying the key role of LILRB1 in the treatment of HCC.

OPEN IN VIEWER

The SH2 domain of SHP1/2 phosphatases constitutively bind to the cytoplasmic ITIMs of LILRB1, which play a pivotal role in cellular homeostasis and tumor development [26]. We therefore extensively searched the expression association of LILRB1, SHP1 and SHP2 in different liver subtypes. First, we determined the protein expression of SHP1, SHP2 and LILRB1 in different liver cell lines. We observed a moderate expression of SHP1, SHP2 and LILRB1 in normal liver cells. However, both the protein levels of SHP1 and LILRB1 were robustly declined in liver cancer cell lines such as MHCC97L, Huh7 and HepG2 cells (Fig. 4A and B). Though a moderate increase in the expression of SHP2 was observed in MHCC97L cells compared to LO2, no alteration appeared in Huh7 and HepG2 cells compared with LO2 cells (Fig. 4A and B). Additionally, based on search results from the online Oncomine database, we observed that transcription levels of LILRB1 had a significantly positive correlation with SHP1 in normal liver tissues ( n = 10, p = 0.028 and R = 0.686) (Fig. S1A), liver cancer precursor ( n = 30, 0.001 and R = 0.704) (Fig. S1B), and HCC specimens ( n = 35, p = 0.002 and R = 0.512) (Fig. S1C). However, there was no association between the expression of LILRB1 mRNA and SHP2 among normal liver tissues ( n = 10, p = 0.30 and R = 0.365) (Fig. S1D), liver cancer precursor ( n = 30, p = 0.30 and R = - 0.396) (Fig. S1E), and HCC tissues ( n = 35, p = 0.634 and R = 0.083) (Fig. S1F). Similarly, utilizing GEPIA, there was significant correlation between LILRB1 and SHP1 (PTPN6) in liver ( R = 0.58), normal ( R = 0.77), tumor ( R = 0.66) and the three combination ( R = 0.42) (Fig. 4C). Though the moderate correlation between LILRB1 and SHP2 (PTPN11) was discovered in liver ( R = 0.21) and normal ( R = 0.41), there was a weak link between them in tumor ( R = 0.18) and this three combination ( R = 0.10) (Fig. 4D). Thus, it may be SHP1 but not SHP2 that transduces the LILRB1-evoked downstream signal which possibly takes part in the progression of HCC. To further prove the direct interaction between LILRB1 and SHP1 in liver cells, the antibody against SHP1 was used to immunoprecipitate specific complexes. After subjecting SDS-polyacrylamide gel electrophoresis using the immunoprecipitate, as shown in Fig. 4E, LILRB1 was immunoprecipitated using specific antibody against SHP1 but not the IgG antibody (Fig. 4E). The data proved that there was fairly strong interactions between LILRB1 and SHP1 in liver cells. And the correlation between LILRB1 and SHP1 might contribute to the progression of HCC.