Development of a Novel Anti−CD44 Monoclonal Antibody for Multiple Applications against Esophageal Squamous Cell Carcinomas

CD44 is a cell surface glycoprotein, which is expressed on normal cells, and overexpressed on cancer cells. CD44 is involved in cell adhesion, migration, proliferation, survival, stemness, and chemo−resistance. Therefore, CD44 is thought to be a promising target for cancer diagnosis and therapy. In this study, we established anti−CD44 monoclonal antibodies (mAbs) by immunizing mice with a CD44 variant (CD44v3−10) ectodomain and screening using enzyme−linked immunosorbent assay. We then characterized them using flow cytometry, Western blotting, and immunohistochemistry. One of the established clones (C44Mab−46; IgG1, kappa) reacted with CD44 standard isoform (CD44s)−overexpressed Chinese hamster ovary−K1 cells (CHO/CD44s) or esophageal squamous cell carcinoma (ESCC) cell lines (KYSE70 and KYSE770). The apparent KD of C44Mab−46 for CHO/CD44s, KYSE70, and KYSE770 was 1.1 × 10−8 M, 4.9 × 10−8 M, and 4.1 × 10−8 M, respectively. C44Mab−46 detected CD44s of CHO/CD44s and KYSE70, and CD44 variants of KYSE770 in Western blot analysis. Furthermore, C44Mab−46 strongly stained the formalin−fixed paraffin−embedded ESCC tissues in immunohistochemistry. Collectively, C44Mab−46 is very useful for detecting CD44 in various applications.


Introduction
CD44 is a cell surface glycoprotein, which is expressed on normal cells, and overexpressed on cancer cells. [1]. The CD44 gene is comprised of 19 exons in human [2]. The first five and the last five exons are constant and encode the shortest isoform of CD44 (85-95 kDa), called the CD44 standard isoform (CD44s). CD44s is widely distributed and plays critical roles in normal homeostasis, including hematopoiesis and immune system, and organogenesis [3]. The middle nine exons can be alternatively spliced and assembled with the ten exons contained in CD44s. They are referred to as CD44 variant isoforms (CD44v) [4]. For instance, CD44v3−10 is expressed in epithelial tissues and overexpressed in carcinomas [3]. Furthermore, CD44v3−10 functions as a co−receptor of receptor tyrosine kinases by binding to heparin−binding epidermal growth factor-like growth factor and fibroblast growth factors (via v3−encoded region) [5] and hepatocyte growth factor (via v6−encoded region) [6], and confers oxidative stress resistance via v8−10−encoded region [7]. Moreover, both CD44s and CD44v are major receptors for hyaluronic acid (HA), which is involved in cell adhesion, migration, and proliferation [8]. They also play important roles in tumor progression, metastasis, stemness, and resistance to chemo− and radiotherapy [9,10]. Therefore, CD44 is a promising target for cancer diagnosis and therapy.
Several CD44−targeting monoclonal antibodies (mAbs) have been developed for preclinical research applications. IM7 mAb was shown to inhibit HA−induced vascular endothelial growth factor production in human vascular endothelial cells [11] and significantly decrease cell migration and invasion in breast cancer cells [12]. Another mAb, H4C4 mAb, was found to reduce tumor growth, metastasis, and post−radiation recurrence [13]. A humanized mAb specific for CD44, RG7356, has been found to be directly cytotoxic for leukemia B cells, but with no effect on normal B cells. Administration of RG7356 to immune−deficient mice engrafted with human chronic lymphocytic leukemia cells was found to result in complete clearance of engrafted leukemia cells [14]. Finally, CD44v6−specific humanized mAbs (BIWA−4 and BIWA−8) labeled with 186 Re showed the therapeutic efficacy in head and neck SCC xenograft bearing nude mice in another study [15].
Sulfasalazine is an inhibitor of xCT, the subunit of cystine-glutamate transporter. CD44v has been found to interact with the xCT and promotes oxidative stress resistant in cancer cells [7,16]. The clinical trials of sulfasalazine have been evaluated in combination with cisplatin in patients with CD44v−expressing advanced gastric cancer refractory to cisplatin [17], and the reduction of the CD44v−positive gastric cancer cells was observed in sulfasalazine treated patients [18]. These results suggest that CD44v is an important marker to predict and evaluate the sensitivity of sulfasalazine.
Previously, we established the novel mAb, C 44 Mab−5 (IgG 1 , kappa) against CD44 using the Cell−Based Immunization and Screening method [19]. C 44 Mab−5 showed high sensitivity for flow cytometry and immunohistochemical analysis in oral cancers. In this study, we developed a novel anti−CD44 mAb, C 44 Mab−46 (IgG 1 , kappa), and evaluated it for various applications, including flow cytometry, Western blotting, and immunohistochemical analyses.

Splenocytes (B cells) Myelomas
Fusion (to 96 well plates)   Then, cells were treated with Alexa Fluor 488−conjugated anti−mouse IgG. Fluorescence data were subsequently collected, followed by the calculation of the apparent dissociation constant (K D ) by GraphPad PRISM 8.

Western Blot Analysis
Western blotting was performed to further assess the sensitivity of C 44 Mab−46. Lysates of CHO−K1 and CHO/CD44s cells were probed. As shown in Figure 4A, C 44 Mab−46 detected CD44s as a~85 kDa band. However, C 44 Mab−46 did not detect any band from lysates of CHO−K1 cells. These results indicated that C 44 Mab−46 specifically detect exogenous CD44s. Next, we examined the detection of endogenous CD44 using lysates from KYSE70 and KYSE770 cells. As shown in Figure 4B, C 44 Mab−46 could detect CD44 as 85 and 48kDa double bands from lysates of KYSE70 cells. In contrast, C 44 Mab−46 could detect CD44 as more than 100 kDa triplet bands, probably CD44v isoforms, from lysates of KYSE770 cells. We also performed the peptide blocking experiment and found that 85 kDa band in CHO/CD44s ( Figure 4C), 85 and 48 kDa bands in KYSE70, and CD44v isoforms in KYSE770 ( Figure 4D) were blocked in the presence of the CD44 peptide (amino acids 161−181, WT). These results suggest that C 44 Mab−46 recognizes endogenous CD44s and CD44v(s) in the esophageal cancer cell lines. (1 µg/mL) plus the CD44 peptide (10 µg/mL, WT), and subsequently with peroxidase−conjugated anti−mouse immunoglobulins. Black arrows indicate the predicted size of CD44s (~85 kDa). The red arrow indicates the CD44 variants.

Immunohistochemical Analysis Using C 44 Mab−46 against ESCC Tissues
To investigate whether C 44 Mab−46 can be used for immunohistochemical analyses using paraffin−embedded tumor sections, we used tissue microarray of ESCC. In a well differentiated ESCC section ( Figure Figure S1). We also performed the peptide blocking experiment and found that the recognition of ESCC tissues by C 44 Mab−46 was completely blocked by CD44 peptide (amino acids 161−180, WT) (Supplemental Figure S2). In immunohistochemical analyses using C 44

Discussion
CD44 have functions in many processes in normal cells (hematopoietic, immune system, and organogenesis), and in pathological situations (inflammation and cancer) [3]. CD44 exhibits passive ligand binding, including adhesion to hyaluronan and other components of the extracellular matrix. Furthermore, CD44 acts as a co−receptor for growth factors and as a scaffold for enzymes. Therefore, establishment and characterization of anti−CD44 mAbs are thought to be important for the development of CD44 targeting therapy and diagnosis.
In this study, we developed C 44 Mab−46, which can recognize all isoforms of CD44, and shows utility in flow cytometry, Western blotting, and immunohistochemistry applications. As shown in Figure 2, C 44 Mab−46 similarly recognized endogenous CD44 of KYSE70 and KYSE770 cells by flow cytometry; however, the expression pattern of CD44s and CD44v was different in Western blot analysis (Figure 4). Furthermore, we detected a 48 kDa band in KYSE70 cells, which is also blocked by the CD44 peptide ( Figure 4D). By the radio labeled CD44s pulse chasing assay, a CD44s precursor form (about 48 kDa) was first translated, after which the precursor received N− and O−glycosylation and reached 85 kDa mature form [22]. This result suggests that the 48 kDa band is a non−glycosylated precursor form of CD44 in KYSE70 cells. We previously determined the epitope of C 44 Mab−46 as 174 −TDDDV− 178 by enzyme−linked immunosorbent assay using synthetic peptides [21]. Although the Thr174 of CD44 have been confirmed as an O−glycan site [23], C 44 Mab−46 could recognize the epitope in the absence of glycosylation. In the future, it should be determined whether the 48 kDa CD44 is exposed on cell surface, and O−glycosylation at Thr174 affect the recognition by C 44  CD44s and CD44v interact with various proliferation−, migration− and invasion− promoting membrane proteins. CD44 binds to platelet−derived growth factor β−receptor and transforming growth factor−β type I receptor [24], and modulates their functions. CD44s also interacts with podoplanin (PDPN) and colocalizes at cell−surface protrusions. PDPN−induced migration requires CD44 in MDCK cells, and knockdown of CD44 and PDPN in oral SCC cells affect cell spreading, suggesting that CD44 directly interacts with PDPN and modulates the intracellular signaling and cell migration [25]. It is interesting that C 44 Mab−46 inhibits these protein-protein interactions and influences their functions.
CD44v6 is highly expressed in many cancers, and acts as co−receptor for at least three receptor tyrosine kinases (c−Met [26,27], EGFR [28], and VEGFR−2 [29]), which contribute to the oncogenic functions of CD44v6. Therefore, humanized anti−CD44v6 mAb BIWA4 (bivatuzumab)−mertansine drug conjugate, was evaluated in clinical trials [30]. However, the clinical trials were discontinued because of the severe skin toxicities [31,32]. A first−in−human phase I clinical trial with RG7356, targeting to the constant region of CD44, showed an acceptable safety profile in patients with advanced CD44−expressing solid tumors. However, the study was terminated due to no evidence of a clinical and/or pharmacodynamic dose-response relationship with RG7356, but not due to safety concerns [33]. Therefore, the development of anti−CD44 targeting mAbs with more potent and fewer side effects is desired.
We previously established cancer−specific mAbs (CasMabs) targeting to PDPN [34][35][36][37] and podocalyxin [38], which are expressed on many cancers, including ESCC [39][40][41][42]. These CasMabs recognize cancer−specific aberrant glycosylation of the target proteins [43]. It is worthwhile to develop a cancer−specific anti−CD44 antibody using the CasMab method. We used the glioblastoma LN229−secreted CD44v3−10 ectodomain as an antigen and generated many mAbs other than C 44 Mab−46. We selected the mAbs, which react with cancer cells, but not with normal cells by flow cytometry and immunohistochemistry. Furthermore, we previously converted the IgG 1 subclass of C 44 Mab−5 into a mouse IgG 2a , named 5−mG 2a , and produced a defucosylated version, 5−mG 2a −f. The 5−mG 2a −f ex-hibited ADCC/CDC activities in vitro and significantly reduced tumor growth in oral cancer cells [44]. Therefore, the production of a class switched and defucosylated version of C 44 Mab−46 is warranted to evaluate the in vivo anti−tumor activity. Since KYSE70 was used in this study, it was reported to be tumorigenic in athymic nude mice [45]; the cell line can be used to evaluate the anti−tumor activity as in the in vivo model.

Purification of CD44ec
After LN229/CD44ec was cultured using DMEM complete medium without G418, CD44ec was purified from the supernatants using RAP tag system, comprised of an anti−RAP tag mAb (clone PMab−2) and a RAP peptide (GDDMVNPGLEDRIE) [55,56]. The filtered culture supernatant (5 L) was passed through PMab−2−Sepharose (2 mL bed volume), and the same process was repeated three times. The beads were then washed with 100 mL of phosphate−buffered saline (PBS, Nacalai Tesque, Inc.), and eluted with 0.1 mg/mL of RAP peptide in a step−wise manner (2 mL × 10).

Hybridoma Production
Female BALB/c mice (6−weeks old) were purchased from CLEA Japan (Tokyo, Japan). The animals were housed under specific pathogen−free conditions. The Animal Care and Use Committee of Tohoku University approved all animal experiments.
Subsequently, splenic cells were fused with P3U1 cells using polyethylene glycol 1500 (PEG1500; Roche Diagnostics, Indianapolis, IN, USA). The hybridomas were then grown in RPMI media supplemented with hypoxanthine, aminopterin, and thymidine (HAT) for selection (Thermo Fisher Scientific Inc.). The culture supernatants were screened for the anti−CD44ec antibody production using enzyme−linked immunosorbent assay (ELISA).

Immunohistochemical Analysis
Paraffin−embedded ESCC tissue microarray (Product Code: BC02011, US Biomax Inc., Rockville, MD, USA) were deparaffinized in xylene and rehydrated. Then, they were autoclaved in citrate buffer (pH 6.0; Agilent Technologies Inc.) for 20 min. After blocking with SuperBlock T20 (Thermo Fisher Scientific, Inc.), sections were incubated with C 44 Mab−46 (5 µg/mL) for 1h at room temperature and then treated with the EnVision+ Kit for mouse (Agilent Technologies Inc.) for 30 min. Color was developed using 3,3 −diaminobenzidine tetrahydrochloride (DAB; Agilent Technologies Inc.) for 2 min. Counterstaining was performed with hematoxylin (FUJIFILM Wako Pure Chemical Corporation). Hematoxylin and eosin (HE) staining (FUJIFILM Wako Pure Chemical Corporation) was performed using consecutive tissue sections. Leica DMD108 (Leica Microsystems GmbH, Wetzlar, Germany) was used to examine the sections and obtain images.