A Cancer-Specific Monoclonal Antibody against Podocalyxin Exerted Antitumor Activities in Pancreatic Cancer Xenografts

Podocalyxin (PODXL) overexpression is associated with poor clinical outcomes in various tumors. PODXL is involved in tumor malignant progression through the promotion of invasiveness and metastasis. Therefore, PODXL is considered a promising target of monoclonal antibody (mAb)-based therapy. However, PODXL also plays an essential role in normal cells, such as vascular and lymphatic endothelial cells. Therefore, cancer specificity or selectivity is required to reduce adverse effects on normal cells. Here, we developed an anti-PODXL cancer-specific mAb (CasMab), PcMab-6 (IgG1, kappa), by immunizing mice with a soluble PODXL ectodomain derived from a glioblastoma LN229 cell. PcMab-6 reacted with the PODXL-positive LN229 cells but not with PODXL-knockout LN229 cells in flow cytometry. Importantly, PcMab-6 recognized pancreatic ductal adenocarcinoma (PDAC) cell lines (MIA PaCa-2, Capan-2, and PK-45H) but did not react with normal lymphatic endothelial cells (LECs). In contrast, one of the non-CasMabs, PcMab-47, showed high reactivity to both the PDAC cell lines and LECs. Next, we engineered PcMab-6 into a mouse IgG2a-type (PcMab-6-mG2a) and a humanized IgG1-type (humPcMab-6) mAb and further produced the core fucose-deficient types (PcMab-6-mG2a-f and humPcMab-6-f, respectively) to potentiate the antibody-dependent cellular cytotoxicity (ADCC). Both PcMab-6-mG2a-f and humPcMab-6-f exerted ADCC and complement-dependent cellular cytotoxicity in the presence of effector cells and complements, respectively. In the PDAC xenograft model, both PcMab-6-mG2a-f and humPcMab-6-f exhibited potent antitumor effects. These results indicated that humPcMab-6-f could apply to antibody-based therapy against PODXL-expressing pancreatic cancers.

PODXL regulates epithelial and tumor cell motility through interactions with the actin polymerization complexes, including the ERM (ezrin-radixin-moesin) family [13] and PDZ protein Na + /H + exchanger regulatory factors 1 and 2 (NHERF-1/2) [14].Ezrin and NHERF-1/2 are adaptor proteins that have diverse binding partners and facilitate the interaction of PODXL with the cytoskeletons of tumor cells [12].The interactions between PODXL, ezrin, and NHERF-1/2 have been reported to activate intracellular signaling including Rac1, RhoA, Cdc42, mitogen-activated protein kinase, and phosphatidylinositol-3 kinase pathways to promote motility [12].PODXL has been shown in gain-and loss-of-function studies to play an essential role in tumor progression by promoting migration, invasiveness, stemness, and metastasis in a variety of cancer cells [15].Therefore, PODXL has acquired increasing attention as a target of tumor immunotherapy.
Nevertheless, the further development of anti-PODXL mAbs targeting tumors has been hampered by concerns about possible toxicities to normal vascular/lymphatic endothelial cells [5] and kidney podocytes [6].For the development of antibody therapy against PODXL-positive cancers, cancer-specificity is necessary to reduce the risk of adverse effects on normal tissues.In this study, we developed a cancer-specific anti-PODXL mAb, PcMab-6 (IgG 1 , kappa), by screening more than one hundred hybridoma clones.We engineered PcMab-6 into a mouse IgG 2a -type (PcMab-6-mG 2a ) and a humanized IgG 1 -type (humPcMab-6) mAb and further produced the core fucose-deficient types (PcMab-6-mG 2a -f and humPcMab-6-f, respectively) to potentiate ADCC.This technique was clinically applied to mogamulizumab (Poteligeo ® ), a defucosylated antibody targeting CCR4 [20].We then examined the antitumor activity against mouse xenograft models of PDAC.
Body weight loss and skin disorders were not observed in the xenograft-bearing mice (Figure 3J-L).The mice on day 28 (MIA PaCa-2 and Capan-2) and day 25 (PK-45H) are shown in Supplementary Figure S1.
Body weight loss and skin disorders were not observed in the xenograft-bearing mice (Figure 6J-L).The mice on day 21 are shown in Supplementary Figure S2.

Discussion
Pancreatic cancer has become the third leading cause of death in men and women combined in the United States as of 2023 [21].PDAC is the most common type of pancreatic cancer and exhibits an extremely poor prognosis with a 5-year survival rate of approximately 10% [22].The four most common oncogenic events, including KRAS, CDKN2A, SMAD4, and TP53, play critical roles in cancer development [23,24].In contrast, PDAC is a heterogenous tumor with various histologies [25], heterogenous molecular landscapes [26], and clinical outcomes.Therefore, the identification of early diagnostic markers and therapeutic targets in each group is desired.In this study, we provided evidence that PODXL could be a promising target for antibody-based therapy, especially PcMab-6 (Figure 1) and its derivatives, including PcMab-6-mG 2a -f (Figures 2 and 3) and humPcMab-6-f (Figures 4-6).
High PODXL expression is significantly associated with worse overall survival and is predictive of shorter overall survival in multiple cancers, especially PDAC [27].Furthermore, PODXL is upregulated during epithelial-mesenchymal transition [28] and plays a key role in the extravasation of mesenchymal PDAC cells [29].PODXL enhances the extravasation through direct binding to ezrin, a cytoskeletal linker protein, which enables the transition of tumor cells from a non-polarized, rounded cell shape to an invasive extravasation-competent morphology [29].These results suggest that PODXL mediates the process of extravasation during tumor metastasis.Although the requirement of PODXL ligands such as E-selectin [30] in endothelial cells is unknown, it would be interesting to see whether PcMab-6 affects PODXL ligand interactions, tumor cell extravasation, and metastasis.
The minimum epitopes of one of the anti-PODXL non-CasMabs (clone PcMab-47) and another anti-PODXL CasMab (clone PcMab-60 [35]) were determined to be the peptide sequence, including Asp207, His208, Leu209, and Met210, of PODXL [36] and the peptide sequence, including Arg109, Gly110, Gly111, Gly112, Ser113, Gly114, Asn115, and Pro116, of PODXL [37], respectively.In contrast, the epitope of PcMab-6 has not been identified yet.The identification of the PcMab-6 epitope is essential for understanding the mechanism of cancer-specific recognition.Furthermore, recognition by PcMab-6 in flow cytometry was lower than that of PcMab-47 (Figure 1), which can be attributed to the possibility that the epitope sequence is partially exposed or modified in cancer cells but not in normal cells in living cells.The CasMab selection strategy can contribute to the development of novel mAbs against a variety of antigens.
Glycosylation in tumors is frequently altered [38].Anti-PODXL mAbs, which recognize tumor-specific glyco-epitopes on PODXL but do not react with PODXL expressed on normal cells, have been developed [39].One of these mAbs, PODO447, exhibits exquisite specificity for a tumor glycoform of PODXL but lacks reactivity with normal adult human tissue.Using an array of glycosylation defective cell lines, the epitope of PODO447 was revealed to be an O-linked core 1 glycan presented in the context of the PODXL peptide backbone [39].Furthermore, PODO447 coupled with monomethyl auristatin E (MMAE) showed efficacy in targeting human pancreatic and ovarian tumor xenografts in mouse models [40].
All therapeutic mAbs exhibit adverse effects, probably because of the recognition of antigens expressed in normal cells.The majority of patients treated with anti-EGFR therapeutic mAbs experience dermatological disorders, which affect their quality of life and adherence to the therapy [41].The major adverse effect associated with anti-HER2 therapeutic mAbs is cardiotoxicity [42].Since PODXL is expressed in normal adult tissues, including vascular/lymphatic endothelial cells [5] and kidney podocytes [6], the evaluation of the in vivo toxicity of PcMab-6 derivatives is essential.We investigated this toxicity against cynomolgus monkeys.The first protocol is one intravenous injection of 20 mg/kg of the mouse-human chimeric and core-fucose-deleted antibody of PcMab-6 (chPcMab-6-f) and observation for one week.The second protocol is four intravenous injections of 10 mg/kg of chPcMab-6-f (every week) and observation for one month.We confirmed that chPcMab-6-f did not show any toxicities against cynomolgus monkeys in either protocol (Supplementary Figure S3).
CARs are synthetic modular proteins that redirect immune cell reactivity toward a target of interest.This platform has shown substantial clinical effects against B cell and plasma cell malignancies and the potential to expand its application to solid tumors [43].The optimal epitope for the PODXL-targeting CAR has not yet been evaluated.However, it would be worthwhile to investigate the cancer specificity of the PcMab-6 single-chain variable fragment and the efficacy of CAR-T therapy against PODXL-positive tumors.

Animals
For the establishment of mAbs against PODXL, the animal experiment was approved by the Animal Care and Use Committee of Tohoku University.For the ADCC assay and antitumor activity in mouse xenograft models, the animal experiments were approved by the Institutional Committee for Experiments of the Institute of Microbial Chemistry.The loss of original body weight to a point > 25% and/or a maximum tumor size > 3000 mm 3  were identified as humane endpoints for euthanasia.

Antibodies
A mouse anti-PODXL mAb, PcMab-47 (IgG 1 , kappa), was developed as described previously [17].Normal mouse IgG (mIgG) was purchased from FUJIFILM Wako Pure Chemical Corporation (Osaka, Japan).Mouse IgG 2a (mIgG 2a ) and normal human IgG (hIgG) were purchased from Sigma-Aldrich Corp. (St.Louis, MO, USA).To generate PcMab-6-mG 2a , the V H cDNA of PcMab-6 and the C H of mouse IgG 2a were subcloned into a pCAG-Ble vector (FUJIFILM Wako Pure Chemical Corporation), and the V L and C L cDNAs of PcMab-6 were subcloned into a pCAG-Neo vector (FUJIFILM Wako Pure Chemical Corporation).To generate a humanized PcMab-6 (humPcMab-6), the CDR of PcMab-6 V H , the frame sequences of V H in human Ig, and the C H of human IgG 1 were cloned into the pCAG-Neo vector.The CDR of PcMab-6 V L , the frame sequences of V L in human Ig, and the C L of the human kappa chain were cloned into the pCAG-Ble vector.To generate PcMab-6-mG 2a -f and humPcMab-6-f, antibody expression vectors were also transfected into BINDS-09 (Fut8-knocked-out ExpiCHO-S cells) using the ExpiCHO Expression System.PcMab-6, PcMab-6-mG 2a -f, and humPcMab-6-f were purified using Protein G-Sepharose (GE Healthcare Bio-Sciences, Pittsburgh, PA, USA).
Cytolyticity (% lysis) was determined: % lysis is calculated as (E − S)/(M − S) × 100, where "E" denotes the fluorescence in the effector and target cell cultures, "S" denotes the spontaneous fluorescence of only target cells, and "M" denotes the maximum fluorescence after treatment with a lysis buffer (10 mM Tris-HCl (pH 7.4), 10 mM EDTA, and 0.5% Triton X-100).All data are shown as mean ± standard error of the mean (SEM).Welch's t-test was used for the statistical analyses.
To measure the antitumor activity of humPcMab-6-f, 100 µg of humPcMab-6-f or control hIgG in 100 µL of PBS was injected into the peritoneal cavity of each tumor xenograftbearing mouse on day 7.Additional antibodies were injected on day 14.Furthermore, human NK cells (8.0 × 10 5 cells) were injected around the tumors on days 7 and 14.The mice were euthanized on day 21 after the cell inoculation.
The tumor diameter was measured and the tumor volume was calculated using the following formula: volume = W 2 × L/2, where W is the short diameter, and L is the long diameter.All data are expressed as the mean ± SEM.Statistical analysis was performed using ANOVA with Sidak's post hoc test.p < 0.05 was considered statistically significant.

Figure 4 .
Figure 4. Establishment of the humanized and defucosylated antibody humPcMab-6-f.(A) The complementaritydetermining region (CDR) of PcMab-6 V H , the frame sequences of V H in human Ig,

Author
Contributions: H.S., T.O. and T.T. performed the experiments.M.K.K. and Y.K. designed the experiments.H.S. and Y.K. analyzed the data.H.S. and Y.K. wrote the manuscript.All authors have read and agreed to the published version of the manuscript.Funding: This research was supported in part by the Japan Agency for Medical Research and Development (AMED) under Grant Numbers JP22ama121008 (to Y.K.), JP22am0401013 (to Y.K.), 23bm1123027h0001 (to Y.K.), JP22ck0106730 (to Y.K.), and JP21am0101078 (to Y.K.) and by the Japan Society for the Promotion of Science (JSPS) Grants-in-Aid for Scientific Research (KAKENHI) grant nos.22K06995 (to H.S.), 21K07168 (to M.K.K.), and 22K07224 (to Y.K.).Institutional Review Board Statement: Animal experiments were approved by the Animal Care and Use Committee of Tohoku University (permit number: 2019NiA-001) and the Institutional Committee for Experiments of the Institute of Microbial Chemistry (approval nos.2022-010, 2022-017, 2022-021, and 2022-045).