Inhibitory Effects of Circulating Natural Autoantibodies to CD47-derived Peptides on Oral Squamous Cell Carcinoma Cells

Natural autoantibodies serve as an important anti-tumorigenic system due to its immune surveillance function. The present study aims to investigate whether circulating natural IgG autoantibodies against cluster of differentiation 47 (CD47) could suppress the proliferation of oral squamous cell carcinoma (OSCC) cells. Three OSCC cell lines were used for this study. The expression of fourteen tumor related genes including CD47 were tested. OSCC cells were grown, respectively, with 20% human plasma positive and negative for anti-CD47 IgG. Cell proliferation, apoptosis and invasion/metastasis were examined. The results showed that CD47 presented highest expression among all 14 genes detected in OSCC cells. Plasma anti-CD47 IgG signicantly inhibited the viability of all three OSCC cell lines. The proportions of apoptotic cells were remarkably higher in OSCC cells treated with anti-CD47 IgG-positive plasma than those treated with IgG-negative plasma. Furthermore, the cell invasion/metastasis was attenuated evidently with the attribution of plasma anti-CD47 IgG. In conclusion, natural autoantibodies against CD47 may be a potential target for OSCC immunotherapy.


Introduction
Oral squamous cell carcinoma (OSCC) is one of the most common cancers counting for more than 90% of all histological subtypes of oral cancer [1,2]. About 300,000 new cases are estimated to be diagnosed annually worldwide, and locoregional recurrences may occur in 30-35% of patients, 20% of whom die of OSCC eventually [3,4]. The high recurrence and metastasis rate of OSCC lead to the 5-year survival rate of less than 50% [5,6]. In recent years, treatment options, particularly for advanced (III and IV) malignancies, have been applied including surgery, radiation and/or chemotherapy as well as targeted therapy. [1,7]. Despite huge advances in diagnosis and treatment, the toxic side effects and drug resistance remain the main causes of high mortality and poor mortality in OSCC [1,3]. Therefore, there is an urgent need for new actionable molecular targets or new therapeutic agents with better e cacy to prevent the recurrence and metastasis of OSCC.
Cluster of differentiation 47 (CD47), also called integrin-related signal transducer or protein (IAST or IAP), is composed of an extracellular N-terminal V-set immunoglobulin superfamily (IgSF) domain attached to ve-time transmembrane domains and an intracellular C-terminus [8]. Being the principal ligand, CD47 expresses in several tumors and binds to the cytoplasmic tail of signal regulatory protein alpha (SIRPα) that is a surface glycoprotein and an immunoreceptor expressed on all myeloid cells. CD47-SIRPα interaction triggers a cascade of events and elicits a "do not eat me" signal to suppress macrophage phagocytosis, allowing cancer cells to escape the immune surveillance. Several lines of evidence have recently shown that overexpression of CD47 is associated with the pathogenesis of leukemia, non-Hodgkin's lymphoma (NHL), glioblastoma, bladder and breast cancer [9,10]. Therefore, blocking CD47-SIRPα checkpoint has been applied for immunotherapy. Inhibition of CD47 binding to SIRPα activates both innate and adaptive immunity by promoting presentation of tumor antigens to CD8 T lymphocytes, resulting in cancer cell destruction by macrophage and anti-tumor cytotoxicity due to antigenpresentation by dendritic cells, respectively [11,12]. There are several CD47 antagonists have shown strong e cacy against solid and hematological tumors undergoing clinical trials, such as Hu5F9 and TTI-621 [9,12]. Hu5F9 is a humanized monoclonal antibody binding to CD47 with high a nity and has been proposed to activate a pro-phagocytic signal pathway to eliminate the tumor cells [13]. TTI-621 is a   recombinant fusion protein composed of human SIRPα N-terminal domain fused to the Fc receptor of   IgG, reinforcing phagocytosis and antitumor activity by engagement of macrophage Fcγ receptors with  IgG1 Fc receptor [13,14]. However, the safety of these immune pharmaceuticals remains a key issue.
Natural autoantibodies are de ned as immunoglobulins produced by B1 lymphocyte cells without the stimulation of exogenous antigens [15]. They are present in normal individuals and physiologically maintain the tissue homeostasis such as elimination of invading pathogenic agents, clearance of apoptotic cells and destruc tion of cancer cells [16,17]. Recent studies suggested that the levels of natural anti-forkhead box P3(FOXP3) IgG autoantibodies are signi cantly higher in patients with esophageal cancer [18]; in addition, anti-CD25 and anti-baculoviral IAP repeat-containing protein 5(BIRC5) IgG levels are elevated obviously in patients with non-small cell lung cancer (NSCLC), who retained a better prognosis and longer overall survival [15,19]. Our previous study showed that anti-vascular endothelial growth factor receptor 1(VEGFR1) IgG positive plasma could signi cantly inhibit the proliferation of hepatocellular carcinoma (HCC) cells by induction of apoptosis and the progression-free survival of HCC patients received infusion with anti-VEGFR1 IgG positive plasma was prolonged [20]. However, studies of natural autoantibodies in OSCC cells are rarely documented. The present study was thus undertaken to analyze the inhibitory effects of anti-CD47 IgG-positive plasma on OSCC cell lines.  (Table 1).

Materials And Methods
RNA was isolated from OSCC cells lysates using the Trizol method (Invitrogen) after cells were incubated in a humidi ed atmosphere with 5% CO 2 at 37°C, and total RNA samples were quanti ed with the Nanodrop machine (Agilent Technologies, Santa Clara, CA). Next, the cDNA was generated with SuperRT cDNA Synthesis Kit (CWbiotech, Beijing, China), and UltraSYBR One Step qRT-PCR Kit (Low ROX, CWbiotech, Beijing, China) was then used to quantify the expression of a target gene. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as a housekeeping gene for normalization as previously described [21], and gene expression was determined by the 2 −△△CT method. qRT-PCR primer sequences are listed in Table 1. 2.3. Detection of plasma IgG levels. According to the results of qRT-PCR analysis, a linear peptide antigen of human CD47 was designed using a computational epitope prediction software (http://www.iedb.org) based on the features of the target proteins such as hydrophilicity, exibility, surface accessibility and antigenicity [19]; plasma IgG against the extracellular domain of the CD47 protein was detected with inhouse ELISA assay [22].
Human plasma samples were collected from healthy blood donors by the Blood Center of Dongguan, Guangdong Province, China. Pooled plasma from more than 100 randomly selected individuals was used as a quality control (QC) for relative quanti cation of plasma anti-CD47 IgG levels. This work was approved by the local Ethics Committee based in Dongguan and conformed to the requirements of the Declaration of Helsinki. Brie y, PBS containing 0.5% bovine serum albumin (BSA) was prepared as an analysis buffer and added to each negative control (NC) well, and a positive control (PC) sample was added to each PC well. The detection was performed according to the instructions of the ELISA kit, which was provided by Hailanshen Biotechnology Ltd, Qingdao, China. Optical density (OD) measurement was done on a microplate reader at 450 nm within 10 minutes at a reference wavelength of 620 nm and the speci c binding ratio (SBR) was used to represent plasma anti-CD47 IgG levels. Plasma with the highest SBR value from two healthy donors was mixed as anti-CD47 IgG-positive plasma; anti-CD47 IgG negative plasma was taken from two healthy donors with the lowest SBR value and mixed properly as described in our previous study [15,18,21,23,24]. SBR calculation is as follows: SBR = (OD sample -OD NC )/(OD PC -OD NC ).
2.4. Cell proliferation assay. 96-well plates were used to seed three OSCC cell lines in 100 µl/well with a density of 3 × 10 4 cells/ml, 5 × 10 4 cells/ml and 5 × 10 4 cells/ml, respectively, and then cultured in complete medium for 24 h under the same conditions as mentioned above; the medium containing 20% human plasma either negative or positive for anti-CD47 IgG antibodies was then used to culture OSCC cells for 48 h under the same conditions as indicated above [21]. Cell counting kit-8 (CCK-8) (Vazyme, Nanjing, China) was used to detect cell viability. Brie y, 10 µl CCK-8 solution was mixed with complete medium at a ratio of 1:10 and then added to each well. After incubation at 37°C for 2 h, the optical density (OD) was measured at a wavelength of 450 nm. Cell viability measurements were used to present data and calculated as follows: Cell viability = (OD positive -OD blank )/(OD negative -OD blank ).
2.5. Analysis of apoptosis. 6-well dishes were used to seed three OSCC cell lines seeded in 2 ml/dish with a density of 2 × 10 5 cells/ml, 2.5 × 10 5 cells/ml and 3 × 10 5 cells/ml, respectively, for initial culture for 24 h as mentioned above. Cells were then treated with the medium containing 20% human plasma either positive or negative for anti-CD47 IgG and harvested at 48 h; the apoptosis rate was evaluated using the Annexin V-FITC/PI Apoptosis Detection kit (BD Biosciences, Franklin Lakes, NJ, USA) according to the instructions from the manufacturer. In brief, 5 μl Annexin V-FITC and 5 μl PI were added to the buffer and incubated at room temperature for 15 min in the dark. Cells were analyzed by ow cytometry (BD FACSCanto) within 1 h. Annexin V-FITC +/PI-staining and Annexin V-FITC +/PI+ staining were used to de ne early apoptosis and late apoptosis.
2.6. Transwell assay. Cell invasion assays were performed in Transwell chambers (24-well format; 8 mm pore size; Corning, NY, USA) coated with Matrigel (1 mg/ml; Corning Incorporated, Corning, NY, USA) in triplicate. After treatment with the medium containing 20% human plasma either positive or negative for anti-CD47 IgG for 48 h, OSCC cells (5 × 10 5 cell/well) were planted into the upper chamber and 1000 μl complete culture was added to the lower chamber, and then incubated in a humidi ed atmosphere with 5% CO 2 at 37°C for additional 48 h. After that, polyoxymethylene and Giemsa (Salarbio, Beijing, China) were used for the xation and staining of invading cells on the underside of the membrane. Five visual elds were randomly selected and the average number of invading cells was calculated with an inverted microscope (Olympus Corporation, Tokyo, Japan, ×200).
2.7. Statistical analysis. All experimental data were expressed as mean ± standard deviation (SD). Student's t-test (two-tailed) and one-way analysis of variation were applied to examine the differences in cell viability, percentage of apoptotic cells and cell invasion/metastasis between OSCC cells treated with anti-CD47 IgG positive and negative plasma as well as in gene expression. P < 0.05 (*) and P < 0.01 (**) were considered to be statistically signi cant. All experiments were repeated at least three times.  (Figure 1(c)). Nevertheless, CD47 gene expression showed the highest level concertedly in all three OSCC cell lines. It is notable that CD47 expression was 1800-fold higher than ABCA1 in CAL27 cells. Based on the high expression of CD47 in three OSCC cells, we subsequently detected natural IgG autoantibodies against CD47 in human plasma and then analyzed the inhibitory effects of anti-CD47 IgGpositive plasma on OSCC cells.
3.3. Cell apoptosis induced by anti-CD47 IgG plasma. Because of the inhibitory effects of anti-CD47 IgGpositive plasma on the proliferation of OSCC cells, we investigated their apoptosis induced by anti-CD47 IgG-positive plasma in OSCC cells. The proportion of apoptotic cells was signi cantly higher in CAL27, SCC9 and SCC25 cells treated with anti-CD47 IgG-positive plasma than those treated with anti-CD47 IgGnegative plasma (P =0.00027 for CAL27 cells, P =0.028 for SCC9 cells, and P =0.0017 for SCC25 cells) (Figure 3). In comparison with SCC9, the apoptosis of CAL27 and SCC25 cells was induced more remarkably by anti-CD47 IgG-positive plasma.
3.5. Figure summary. This study described the screening processes of plasma enriched with natural anti-CD47 IgG autoantibodies and analyzed the effects on cell proliferation, apoptosis and invasion/metastasis of three OSCC cell lines in three stages. In order to depict the content of this study and highlight the strategy more clearly, a diagram was constructed and shown in Figure 5.

Discussion
Natural autoantibodies are considered to be important antitumor systems in vivo showing a decrease in early-stage cancer patients while a gradual increase during the tumor progresses [18,25]. A number of studies have been concerned about the role of natural autoantibodies in NSCLC, esophageal cancer, breast cancer and liver cancer, including those for p16, FOXP3, CD25, ABCC3, ANXA1, BIRC5, MYC and VEGFR1 [18,[25][26][27][28][29]. It has been supposed that natural autoantibodies may be useful anticancer agents for postoperative therapies to prevent the recurrence of human cancer.
The present study has demonstrated that some healthy individuals have remarkably high levels of natural IgG autoantibodies against CD47, which could signi cantly inhibit the proliferation of OSCC cell lines, such as SCC9 cells. Our previous study also showed that natural anti-ABCC3 IgG-positive plasma had similar suppressive effects on the OSCC cells previous [21]. The observation was interpreted with study of apoptosis and gene expression in cells. CD47 showed the highest expression levels in all three OSCC cell lines studied, consistent with other reports that CD47 was expressed more signi cantly in oral cancer tissue than normal mucosal tissues [30]. Plentiful CD47 was expressed on the surface of three OSCC cells and targeted by anti-CD47 IgG autoantibodies. Notably, SCC9 cell viability was suppressed most obviously whereas the apoptosis was not signi cantly altered, suggesting that distinct mechanisms are likely to be involved in inhibitory effects of anti-CD47 IgG autoantibodies on the proliferation of different OSCC cells. Cell migration and invasion are the key processes of tumor evolution contributing to poor prognosis of OSCC [31][32][33]. The natural anti-CD47 IgG-positive plasma could e ciently inhibit the invasion and migration of three OSCC cell lines in vitro, especially SCC9 cells. Therefore, we hypothesize natural IgG autoantibodies for CD47 may have potential to treat OSCC. To our knowledge, this is the rst report on inhibitory effects of natural anti-CD47 autoantibodies on the proliferation and invasion of OSCC cells in vitro.
In function, CD47 plays a critical role in the self-recognition process of tumors [34]. The expression of CD47 has been exploited in order to escape immunological eradication in various solid and hematologic cancers; its overexpression is clinically related to poor prognosis [34]. Although CD47 is involved in the regulation of cancer cell growth, its most important function related to tumor development is to prevent phagocytosis via ligating SIRPα with surrounding phagocytes [10]. Ligation of CD47 to SIRPα on the surface of macrophages promotes phosphorylation of the intracellular ITIM, subsequently recruit and activate inhibitory phosphatases SHP-1 and SHP-2 protein to produce a series of cascade reactions that inhibit phagocytosis of macrophages [9,35]. Based on the characteristics of CD47 and the mechanism behind natural autoantibodies against tumors, natural anti-CD47 IgG autoantibodies might play an antitumor role by mediating antibody-dependent cell-mediated cytotoxicity (ADCC) and complement dependent cytotoxicity (CDC). ADCC is a set of mechanisms that target cells coated with IgG antibodies of the proper subclasses to be the prey of cell-to-cell cytolysis executed by immune cells expressing FcRIIIA (CD16A) [36]. CDC is initiated when C1q, the initiating component of the classical complement pathway, xed to the Fc portion of target-bound antibodies [37], which triggers a cascade, leading to the assembly of the membrane attack complex (MAC) and cell lysis occurs as the MAC disrupts the plasma membrane of the target cell [38,39]. In addition, combination of autoantibodies to CD47 triggering the ADCC or CDC could compete the CD47-SIRPα interaction. Although our primary research has shown the effects of natural anti-CD47 IgG autoantibodies on OSCC cells, further research needs to be done to reveal the exact mechanism by which natural anti-CD47 IgG autoantibodies destroy OSCC cells, laying a solid foundation for future animal experiments and clinical trials.

Conclusions
Natural anti-CD47 IgG-positive plasma could inhibit cell proliferation, induce apoptosis of OSCC cells, and inhibit their invasion. However, restriction of human plasma utility clinically may compromise the antitumor effects of natural anticancer antibodies. Therefore, the intravenous immune globulin (IVIG) produced from plasma rich in natural anti-tumor antibodies may become a promising agent for immunotherapy of oral cancer and other types of cancer in the future.

Declarations Data Availability
The data used to support the ndings of this study are available from the corresponding author upon request.

Disclosure
A preprint version of this study has been submitted to https://www.researchsquare.com/article/rs-59440/private/draft. This manuscript is the result interpretation of our study.

Con icts of Interest
The authors declare that they have no con icts of interest.    Transwell assay in OSCC cells treated with anti-CD47 IgG-positive plasma or IgG-negative plasma. The data of invasion were expressed as mean ± standard deviation; *P 0.05, ** P 0.01.