SFPQ subcellular mislocalisation entails altered cancer cell function

Dear Editor, This study demonstrates that splicing factor proline and glutamine rich (SFPQ) is present on the cell surface in cancer cells, where it interacts with S100A4 and so forth proteins to alter cell function when SFPQ is a nuclear protein normally. Protein subcellular mislocalisation can alter cell function in cancer development, metastasis or drug resistance and may serve as a therapy target.1 Overexpression of SFPQ is essential for the malignancy of cancer cells.2-4 While SFPQ normally resides in the nucleus, it can be found in other subcellular locations in lung cancer cells.5 We further explore if SFPQ appears on the cell surface and its influence on cancer cells. SFPQ is more highly stained in the cancer nucleus and cytoplasmic region than the adjacent normal area in immunohistochemistry (IHC) stain in multiple cancer tissues (Figure 1A‒J). The cytoplasmic staining was 3.8‒8.6 times greater than that of the control region and nuclear staining was also more prominent (3.6‒5.6 times more than the control regions in stain density quantification analysis; Figure 1B,D,F,H,I). SFPQ is present outside of the cell nucleus in solid cancers in most cases but not in control tissues. To observe the SFPQ distribution in subcellular organelles, the cell membrane, cytoplasm and nucleus were isolated and used in a Western blot to observe the SFPQ level. With subcellular fractions from normal and cancer cell lines, we found that SFPQ was present in all of the cell membranes, cytoplasm and nuclear fractions from cancer cells but only in nuclear fractions from non-cancer cell lines (Figure 1K‒M). We then used immunofluorescence staining to determine whether SFPQ is present on the cell surface of solid cancer cell lines. We fixed the cells with paraformaldehyde (which fixes the cells but does not penetrate the cell membrane) and stained them with an SFPQ antibody. Most of the cancer cells were positive for SFPQ, which suggested that SFPQ was present on the cell surface (Figure 1N). Internalisation assay with pHrodo-labelled SFPQ antibody and lung cancer cell H661 was positive, suggesting SFPQ is a cell surface protein6 (Figure 1O). Additionally, we found SFPQ and the cell surface marker cadherin E overlapped well in immunofluorescence (IF) stain with lung cancer cells and IHC stain in lung cancer patient tissue (Figure S1) and flow cytometry result suggests that SFPQ is on the cell surface (Figure S1C). This shows SFPQ presence on the cancer cell surface. We then examine if interacting cell surface SFPQ affects cancer cell function. High SFPQ expressing lung cancer cells H661 (Figure 2A) were used to identify cell surface SFPQ binding antibodies, and 9/19 antibodies tested positive in IF stain (Figure 2B shows examples of them). Binding SFPQ could inhibit cell proliferation in lung cancer lines (Figure 2C) and other cancer cell lines from other cell types (Figure 2D). The cell proliferation marker ki67 expression levels are low in those proliferation-inhibited cells (Figure 2E). The inhibition levels are consistent with the SFPQ membrane level (Figure 2A,C,D). Some antibodies also altered apoptosis (Figure 2F) and inhibited cell invasion (Figure 2G,H). When we applied some of those antibodies to other cancer cell lines, SFPQ antibody 1A6 inhibited cell proliferation (Figure S2) and invasion in those cancer cells (Figure S3) while knockdown the SFPQ in those cells reduced the antibody inhibition. These data indicate that different SFPQ antibodies affect cancer cells differently. In vivo study, lung cancer cells A549, H648 and H661 were intravenous; (IV) injected into NSG mice, and anti-SFPQ 1A6 or control IgG was applied to the mice six times in intervals of 3 days. There are more cancer cell clusters in the normal, IgG-treated mouse than the SFPQ antibody-treated mouse in haematoxylin and eosin stain, SFPQ IHC analysis and human cell quantification (Figure 2I‒L). Together, these data indicate that SFPQ antibody treatment attenuates the ability of lung cancer cells to grow in vivo. These findings suggest that cell surface SFPQ could be a therapeutic target. We performed an immunoprecipitation proteomics analysis of SFPQ with the cell membrane fraction to identify SFPQ interacting proteins. Proteomic analysis revealed many proteins bound to SFPQ (Table S1). By analysing their possible functions and signal transduction pathways via Ingenuity pathway analysis (https://digitalinsights.qiagen.com/products/qiagen-ipa), these proteins were found to potentially include pathways of signal transduction that impact cell proliferation, apoptosis, metabolism, senescence and migration (Figure 3A,B and Tables S2 and S3). To confirm that these proteins interact with SFPQ, SFPQ antibody immunoprecipitation and Western blotting were performed with antibodies against PPIA, H2AX, S100A4 and VIM in the lung cancer cell membrane fraction (Figure 3C). PPIA, H2AX, S100A4 and VIM bands were present in the SFPQ pull-down material, demonstrating that SFPQ binds to PPIA, H2AX, S100A4 and VIM in the cell membrane. IF double stain also shows SFPQ and S100A4 colocalised in cell membrane (Figure S5). The involvement of S100A4 in cancer cell migration has been reported previously7-9; we then examined the effect of the expression of SFPQ on S100A4 expression and cell invasion in solid cancer cells. SFPQ and S100A4 are higher in lung cancer cell lines (Figure 3D). SFPQ or S100A4 was knocked down with shRNA in lung cancer cells (with scrambled shRNA used as a control), and SFPQ and S100A4 expression levels and SFPQ/S100A4 complex were inhibited (Figure 3E,F). The cell invasion and migration marker MMP2 was reduced in those lung cancer cells with SFPQ or S100A4 knockdown (Figure 3G,H). We also observed if the SFPQ/S100A4 complex exists in other solid cancer cell lines. We found that SFPQ/S100A4 formed a complex in those cancer cells but not in normal bronchial cells (Figure S4). Knockdown SFPQ or S100A4 inhibited cell invasion in breast cancer cells (Figure S5). These indicate that membrane SFPQ/S100A4 affects cell invasion in cancer cells. Taken together, SFPQ is present on the cell surface and affects cell function via interaction with S100A4, PPIA, etc. in lung cancer and other solid cancer cells. Anti-SFPQ antibodies could inhibit cancer cell proliferation and invasion by binding to SFPQ cells. Therefore, cell surface SFPQ may serve as a prognostic marker and therapeutic target for those cancers. Libang Yang and Robert Kratzke conceived, designed and directed the studies with input from Craig A. Henke, Sabine Karam and Scott M. Dehm. Libang Yang and Robert Kratzke wrote the manuscript with assistance from all the authors. Libang Yang, Adam Gilbertsen, Bo Sun, Yingming Li and Blake Jacobson established primary human cell lines, performed Q-PCR, Western blot analysis, performed gain and loss of function experiments, mouse studies and immunohistochemistry. Proteomic sample preparation, mass spectrometry analysis and database searching were performed at the Center for Mass Spectrometry and Proteomics in the Department of Biochemistry, Molecular Biology and Biophysics at the University of Minnesota. Supporting agencies are listed here: https://cbs.umn.edu/cmsp/about. This work was supported by the National Institutes of Health (grant R01 HL125227 to C.A.H.). The authors declare they have no conflicts of interest. This manuscript represents original work, and all sources have been properly cited. Ethical approval was obtained and informed consent was secured where necessary. The data that support the findings of this study are available upon request from the corresponding author. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.