Background: Neutrophils extracellular traps (NETs) are extracellular strands of decondensed DNA in complex with citrullinated histones (H3cit) and granule proteins including myeloperoxidase (MPO) (Sørensen O E, et al. 2016). Although this structure was initially described as an antimicrobial defense mechanism (Brinkmann V, et al. 2007), recent evidence has linked NETs to tumor metastases, and a previous study has demonstrated that the interaction of NETs and CD5+ B cells favored the transition from autoimmunity to lymphoma (Sangaletti S, et al. 2014). But it remains unclear whether NETs promote diffuse large B cell lymphoma (DLBCL) progression and can be a therapeutic candidate. Thus, our objectives were to detect NETs in the DLBCL samples, define the role of NETs in progression of DLBCL and reveal the underlying mechanism.

Methods: We collected 123 tissue samples from DLBCL patients. NETs in tissues were detected by immunofluroscent staining. In cocultured model, we tested the effect of NETs on cell proliferation and migration of DLBCL cell lines. In vivo, murine A20 cells were inoculated in BALB/C mice (n = 5 mice/group), tumor growth and axillary lymph nodes metastasis were monitored by computer tomography. Tumor tissues and plasma were collected and NETs were detected by immunofluroscence and ELISA.

Results: We detected more NETs structures in advanced stage samples than early stage samples (P= 0.035). Higher NETs expression was significantly associated with more B symptoms, extranodal involvement, higher IPI, elevated LDH level and lower overall response rate. In a univariate analysis, we found high NETs expression was associated with significant inferior survival (OS, HR: 3.250, 95%CI: 1.451-7.278, P= 0.002; progression free survival, HR: 2.678, 95%CI: 1.271-5.642, P= 0.007). In a multivariate analysis, high NETs expression was also found to be an independent predictor of OS and PFS (OS, HR: 2.824, 95%CI: 1.242-6.423, P= 0.013; PFS, HR: 2.398, 95%CI: 1.125-5.113, P= 0.024).

Next, we wanted to determine whether DLBCL could induce NETs formation and how tumor-induced NETs were formed. As previous study proved that IL-8 could prime neutrophils for NETosis (Alfaro C, et al.2016), we detected IL-8 expression in the same tissue samples by immunostaining, and found advanced stage presented with high IL-8 expression and IL-8 also predicted inferior OS, suggesting that IL-8 may interact with NETs. Using a transwell chamber assay, we found only the media of human Su-dhl2 and murine A20, not human Su-dhl4 or Su-dhl6, potentiated species-matched neutrophils to generate extensive NETs. The level of IL-8 was highest in the supernatant of Su-dhl2 (3181.00 ± 386.60 pg/ml), nearly 10.63 and 8.57 folds higher than Su-dhl4 and Su-dhl6 (P= 0.002); and blocking IL-8 significantly reduced the ability of Su-dhl2 to induce NETs. CXCL1 and CXCL2, IL-8 murine homologues, were also high in the media of A20 cells, and NETosis induced by A20 cells was attenuated by murine CXCL1 or/and CXCL2 neutralizing antibody. We also found IL-8 mainly bound to CXCR2, not CXCR1, to induce NETs via the activation of MAPK and Src pathway.

Then we sought to investigate whether NETs formation affected the growth and dissemination of DLBCL. In vitro, we treated Su-dhl2 cells and A20 cells with PMA-stimulating neutrophils media, and we observed that the treatment of NETs increased tumor proliferation at 72 hours and promoted cell migration, which were attenuated by administration of DNase I, a known inhibitor of NETs formation. In vivo, we found that NETs formed in the tumor tissues, which was associated with significantly increased serum MPO-DNA complexes in untreated mice. However, the systemic administration of DNase I or CXCR2 neutralizing antibody significantly decreased NETs formation, and retarded heterotopic tumor growth and axillary lymph nodes metastases compared with untreated group.

Conclusion: In the present study, we firstly demonstrated the value of NETs to predict DLBCL inferior survival, and provided clear evidence that tumor-derived IL-8 bound to its receptor CXCR2, not CXCR1, to induce NETs formation via the activation of Src and MAPK pathway. And we also proved that this structure could promote DLBCL progression. Most importantly, our findings provided preliminary evidence that the use of DNase I or CXCR2 neutralizing antibody could reduce DLBCL progression and will hopefully pave the way for future clinical trials.

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution