MT4‐MMP in tumor‐associated macrophages is linked to hepatocellular carcinoma aggressiveness and recurrence

Dear Editor, Hepatocellular carcinoma (HCC) is among the deadliest human cancers, ranking as the fifth most common primary cancer and the thirdmost commoncancer-associated cause of death.1 The tumor microenvironment has been proposed in recent years to be closely related to resistance and tumor recurrence.2 It includes inflammatory cells, cancer cells, and extracellular matrix (ECM).3 Polarization and phenotypic transition of macrophages are the primary features of tumor-associated chronic inflammation.4 Tumor-associated macrophages (TAMs) show an M2-like phenotype and are closely involved in tumorigenesis, metastasis, ECM remodeling, and poor prognosis.5 MT4-MMP (also named MMP17), as a member of the MMP family of ECM remodelers, is membrane anchored by a glycosylphosphatidylinositol anchor (GPI)6 and directly resolves a specific set of substrates, such as fibrinogen, proTNF, ADAMTS4, and αM integrin.7 MT4-MMP has been reported to be expressed by macrophages and to be associated with inflammation and angiogenesis.8 In particular, the MT4-MMP transcript was recently observed in gastrointestinal adenocarcinomas, prostate cancer, osteosarcomas, leukemias, lung carcinomas, glioblastomas, cervical carcinomas, melanomas, and breast cancer.9 However, the role of MT4-MMP in HCC is still poorly understood. We explored the MT4-MMP expression profile in human HCC. Forty-three and 12 pairs peritumoral and tumor tissues, respectively, were analyzed by qPCR and western blot. Overall survival (OS) was the time from surgery to death or final follow-up, while recurrencefree survival (RFS) was defined as the interval between surgery and recurrence. Data revealed that the MT4-MMP mRNA and protein level was obviously upregulated in peritumoral tissues over tumor tissues (Figure 1A and B). Next, 316 paired peritumoral and tumor tissues were further analyzed for MT4-MMP expression by immunohistochemistry (Figure 1C). Patient clinical and demographic characteristics are compiled in Table S1.

Survival analyses showed that patients in the peritumoral MT4-MMP-high group had markedly decreased median OS and RFS (median OS, 58.0 months; RFS, 24.0 months) relative to patients in the peritumoral MT4-MMP-low group (median OS, undefined, P = .008; RFS, 41.0 months, P = .014). In contrast, there was no significant difference between the OS and RFS of the tumoral MT4-MMP-high and -low groups (median OS, 71 vs 77 months, P = .644; RFS, 27 vs 31 months, P = .795) ( Figure 1D). Multivariate analyses also showed that peritumoral MT4-MMP was independently associated with OS (HR = 1.957, P = .006) and RFS (HR = 1.640, P = .015) in HCC patients (Table  S2). To assess MT4-MMP location, we performed colocalization immunofluorescence staining in human HCC tissues, including peritumoral and tumor tissues. MT4-MMP was expressed in TAMs of peritumoral and tumor tissues, as shown by the colocalization of MT4-MMP with CD68+CD206+ macrophages ( Figure 1E). Notably, MT4-MMP was not expressed in tumor cells (AFP+), fibroblasts (α-SMA+), or M1 macrophages (iNOS+) ( Figure 1F-H). Therefore, we clarified the clinical value of peritumoral MT4-MMP by establishing a prognostic nomogram based upon multivariate Cox analyses. The C-index was 0.737 (95% CI 0.697-0.777) for OS prediction and the calibration plot for the probability of OS at 3 and 5 years exhibited optimized agreement between the nomogram-derived predictions and actual observations ( Figure 1I).
In vitro assay, we found that the expression levels of MT4-MMP in M2 and TAMs were significantly higher than in the M0 and M1 groups ( Figure S1). Similarly, MT4-MMP expression was higher in TAMs than in HCC cells ( Figure S2). To assess the role of MT4-MMP in TAMs for HCC progression, we generated stable MT4-MMP overexpression and knockdown clones of TAMs ( Figure  S3). MHCC97H cells were selected and harvested after 48 hours of coculture with TAMs for use in proliferation, apoptosis, and metastasis assays. Right panel: Quantification of EMT-related marker staining in MHCC97H cells. All histogram bars represent mean ± SEM. *P < .05, **P < .01, ***P < .001 versus the control group (Ctrl). Three experimental replicates in all panels of HCC cells, whereas TAMs with knockdown of MT4-MMP expression lowered cell viability and increased cell apoptosis (Figure 2A and B). The stable overexpression of MT4-MMP in TAMs contributed to the accumulation of cell motility, migration, and invasion in HCC cells, while the suppression of MT4-MMP expression produced the opposite effect ( Figure 2C and D). In view of the critical role of epithelial-mesenchymal transition (EMT) in tumorigenesis, metastasis, and recurrence, we explored EMT marker expression in control and hepatoma cells cocultured with TAMs with MT4-MMP overexpression or knockdown. IF staining demonstrated that TAMs overexpressing MT4-MMP had reduced epithelial E-cadherin expression and increased mesenchymal N-cadherin and vimentin expression, whereas knockdown of MT4-MMP yielded the opposite phenotype ( Figure 2E).

F I G U R E 2 Continued
In conclusion, we have identified that high MT4-MMP expression in peritumoral tissues has a strong relationship with HCC aggressiveness and recurrence and can be seen in TAMs. Additionally, through in vitro experiments, we verified that MT4-MMP in TAMs promoted HCC proliferation and metastasis. Hence, MT4-MMP in TAMs may become a potential biomarker for predicting prognosis and may be a promising therapeutic target for HCC.

C O N F L I C T O F I N T E R E S T
The authors declare that there is no conflict of interest.

A U T H O R C O N T R I B U T I O N S
Feng Qi and Jinglin Xia designed the study and analyzed and wrote the manuscript. Jia Li and Mengzhou Guo conducted the in vitro experiment and data analysis. Feng Qi and Biwei Yang collected the patient samples and followup information and performed the clinical data analysis. Feng Qi, Jia Li, and Biwei Yang performed the bioinformatics analysis. All the authors read and approved the final manuscript.

E T H I C S A P P R O VA L A N D C O N S E N T T O PA R T I C I PAT E
All protocols with human specimens were applied under the examination and approval of the Ethical Committee of Zhongshan Hospital, Fudan University. Written informed consent was provided by each patient (B2020-051).