Antitumor activity of chLpMab‐2, a human–mouse chimeric cancer‐specific antihuman podoplanin antibody, via antibody‐dependent cellular cytotoxicity

Abstract Human podoplanin (hPDPN), a platelet aggregation‐inducing transmembrane glycoprotein, is expressed in different types of tumors, and it binds to C‐type lectin‐like receptor 2 (CLEC‐2). The overexpression of hPDPN is involved in invasion and metastasis. Anti‐hPDPN monoclonal antibodies (mAbs) such as NZ‐1 have shown antitumor and antimetastatic activities by binding to the platelet aggregation‐stimulating (PLAG) domain of hPDPN. Recently, we developed a novel mouse anti‐hPDPN mAb, LpMab‐2, using the cancer‐specific mAb (CasMab) technology. In this study we developed chLpMab‐2, a human–mouse chimeric anti‐hPDPN antibody, derived from LpMab‐2. chLpMab‐2 was produced using fucosyltransferase 8‐knockout (KO) Chinese hamster ovary (CHO)‐S cell lines. By flow cytometry, chLpMab‐2 reacted with hPDPN‐expressing cancer cell lines including glioblastomas, mesotheliomas, and lung cancers. However, it showed low reaction with normal cell lines such as lymphatic endothelial and renal epithelial cells. Moreover, chLpMab‐2 exhibited high antibody‐dependent cellular cytotoxicity (ADCC) against PDPN‐expressing cells, despite its low complement‐dependent cytotoxicity. Furthermore, treatment with chLpMab‐2 abolished tumor growth in xenograft models of CHO/hPDPN, indicating that chLpMab‐2 suppressed tumor development via ADCC. In conclusion, chLpMab‐2 could be useful as a novel antibody‐based therapy against hPDPN‐expressing tumors.

Membrane proteins could be targeted by antibody-based therapy if (1) they possess cancer-specific mutations outside the membrane [25], (2) they are overexpressed in cancers rather than in normal tissues [26][27][28], or (3) they are posttranslationally modified by phosphorylation or glycosylation [29,30]. Although many anti-hPDPN monoclonal antibodies (mAbs) are commercially available, most of them react with the N-terminus of hPDPN, and they do not fulfill the above-mentioned criteria [6,[31][32][33][34][35]. hPDPN is highly expressed in normal lymphatic endothelial cells (LECs) and normal lung type-I alveolar cells at the same level as in cancer cells. We previously produced a rat anti-hPDPN mAb (NZ-1), which detects hPDPN with high specificity and sensitivity [6,10,31]. NZ-1 is efficiently internalized by glioma cell lines and accumulates in tumors in vivo; therefore, it has been suggested to be a suitable candidate for therapy against malignant gliomas [5,10]. Moreover, NZ-1 inhibits tumor cell-induced platelet aggregation and tumor metastasis [23]. NZ-1 mediates antibody-dependent cellular cytotoxicity (ADCC) and complement-dependent cytotoxicity (CDC) against tumor cells that express hPDPN [36]. Furthermore, human-rat chimeric antibodies, such as NZ-8 and NZ-12, exhibit high ADCC and CDC in vitro, and they show very high antitumor activities and neutralizing capabilities [36,37]. Nevertheless, these chimeric mAbs are not cancer-specific; therefore, cancer-specific anti-hPDPN chimeric or humanized antibodies should be developed to prevent unfavorable side effects.
In this study, we developed and characterized chLpMab-2, a human-mouse chimeric anti-hPDPN antibody derived from LpMab-2.

Flow cytometry
The cell lines were harvested after brief exposure to 0.25% trypsin/1 mmol/L EDTA (Nacalai Tesque, Inc.). After washing with 0.1% BSA in PBS, the cells were treated with primary mAbs for 30 min at 4°C, followed by treatment with FITC-labeled goat anti-human IgG (Thermo Fisher Scientific Inc.). Fluorescence data were acquired using Cell Analyzer EC800 (Sony Corp., Tokyo, Japan).

Preparation of effector cells
Effector cells were prepared as previously described [9]. Human peripheral blood mononuclear cells (MNCs) were obtained from leukocytes, which were separated from the peripheral blood of healthy donors. The study with human subjects was approved by the Ethics Committee of Tokushima University.
ADCC ADCC was determined with the 51 Cr release assay [9]. Target cells were incubated with 0.1 μCi of 51 Cr-sodium chromate at 37°C for 1 h. After washing three times with RPMI 1640 supplemented with 10% FBS, 51 Cr-labeled target cells were seeded in 96-well plates in triplicate. Human peripheral blood MNCs and chLpMab-2 or control human IgG were added to the cells. After 6 h of incubation, 51 Cr released from cells into the supernatant (100 μL) was measured using a gamma counter (PerkinElmer, Waltham, MA). The percentage of cytotoxicity was calculated using the following formula: % specific lysis = (E − S)/(M − S) × 100, where E is the release in the test sample, S is the spontaneous release, and M is the maximum release.

CDC
CDC was evaluated by the 51 Cr release assay as previously described [9]. The CHO/hPDPN cells (target cells) were incubated with 51 Cr-sodium chromate (0.1 μCi) for 1 h at 37°C. After incubation, the cells were washed with RPMI 1640 supplemented with 10% FBS. The 51 Cr-labeled cells were incubated with baby rabbit complement (Cedarlane, Ontario, Canada) at a dilution of 1:32 and chLpMab-2 (0.01-10 μg/mL) or control human IgG (0.01-10 μg/mL) for 3 h in 96-well plates. After incubation, 51 Cr in the supernatant was measured using a gamma counter. The percentage of cytotoxicity was calculated using the following formula: % specific lysis = (E − S)/(M − S) × 100, where E is the release in the test sample, S is the spontaneous release, and M is the maximum release.
Antitumor activity of anti-hPDPN antibodies CHO/hPDPN cells were trypsinized and washed with PBS. The cell density was adjusted with PBS to 5.0 × 10 7 cells/ mL, and 100 μL/animal of the cell suspension was subcutaneously inoculated into BALB/c nude mice. After 1 day, 100 μL of 1 mg/mL of chLpMab-2 and human IgG were injected into the peritoneal cavity of mice, once a week for 4 weeks (control group, n = 6; chLpMab-2 group, n = 6). Human NK cells (5.0 × 10 5 cells, Takara Bio Inc., Shiga, Japan) were injected around the tumors 4 and 11 days after cell inoculation. The tumor diameter was measured every 3-4 days and was calculated using the following formula: volume = W 2 × L/2, where W is the short diameter and L is the long diameter. The mice were euthanized 21 days after cell implantation.

Statistical analysis
All data were expressed as means ± SEMs. Student's t-test, Mann-Whitney U-test, one-way ANOVA followed by Tukey-Kramer multiple comparisons, and two-way ANOVA were performed as appropriate. P values less than 0.05 were considered to be statistically significant. All statistical tests were two-sided.

Antitumor activity of chLpMab-2
To study the antitumor activity of chLpMab-2 on primary tumor growth in vivo, the CHO/hPDPN cells were subcutaneously implanted into the flanks of nude mice. The LN319, PC-10, and NCI-H226 cells were not tested because they were not suitable for xenograft models. chLpMab-2 and control human IgG were injected into the peritoneal cavity of mice, once weekly for 4 weeks (n = 6 each), and human NK cells were injected twice around the tumors. Tumor formation was observed in mice from the control and treated groups. However, chLpMab-2 significantly reduced tumor development compared with control human IgG ( Fig. 6A and B). The tumor volume was significantly reduced by chLpMab-2 treatment on day 15, 18, and 21 (Fig. 6C). These results indicate that the administration of chLpMab-2 with the NK cells inhibited the primary tumor growth of the CHO/hPDPN cells in vivo.

Discussion
We previously produced LpMab-2 (mouse IgG 1 , kappa), one of the CasMabs against hPDPN [29]. LpMab-2 recognized not only the Thr55-Leu64 peptide of hPDPN but also an aberrant O-glycosylated hPDPN, which is attached to Thr55 or Ser56 of hPDPN [48]. LpMab-2 reacted with hPDPN-expressing cancer cells and not with normal cells, as revealed by flow cytometry and immunohistochemistry [29]; therefore, LpMab-2 is an anti-hPDPN CasMab that is potentially advantageous for antibody-based molecular-targeted therapy against hPDPNexpressing cancers. However, LpMab-2 is a mouse IgG 1 ; therefore, it cannot be used to study ADCC and CDC against hPDPN-expressing cancers.
To our knowledge, herein we developed the first cancerspecific human-mouse chimeric anti-hPDPN antibody (chLpMab-2) from LpMab-2. Previously, we had developed a human-mouse chimeric anti-hPDPN antibody  (chLpMab-7) [40] and human-rat chimeric anti-hPDPN antibodies such as NZ-8 [36] and NZ-12 [37]. Other groups have reported the development of human-mouse chimeric anti-hPDPN/hAggrus antibodies that were not cancer-specific [32,33]. Because hPDPN is expressed in many normal organs such as the lung and kidney, non-CasMabs against hPDPN are not suitable for antibodybased molecular targeting therapy.
In this study, we used FUT8-KO CHO-S cells (PDIS-5) to express afucosylated chLpMab-2. Afucosylated antibodies are known to exhibit high ADCC [53]. Consistent with the literature, the ADCC of afucosylated chLpMab-2 was 3.4 times higher than that of fucosylated chLpMab-2 (data not shown). In contrast, afucosylated chLpMab-2 showed a lower CDC (Fig. 5B) than fucosylated chLpMab-2 (data not shown). Gasdaska et al. have reported that the higher ADCC of afucosylated rituximab suggests an improvement in effectiveness and potency, however, its lower CDC may mitigate infusion toxicity [54]. They concluded that afucosylated rituximab was clinically better than fucosylated rituximab. In contrast, Niwa et al. have reported that fucose depletion can provide a panel of IgGs (IgG 1 , IgG 2 , IgG 3 , and IgG 4 ) with enhanced ADCC but that none of the IgGs affected CDC [53].
In our previous study, a human-mouse chimeric anti-hPDPN mAb showed CDC for antitumor activity in the absence of human NK cells in a mouse xenograft model [36]. In this study, chLpMab-2 exhibited ADCC for antitumor activity in the xenograft model with added NK cells [9]. Our results indicate that ADCC is important to induce the antitumor activity of anti-hPDPN CasMab in the xenograft model for two reasons: i) fucosylated chLpMab-2, which was expressed in CHO-K1 cells, showed high CDC but failed to induce enough ADCC and antitumor activity against hPDPN-expressing tumors (data not shown) and ii) afucosylated chLpMab-2 did not show enough CDC in vitro (Fig. 5B) but exhibited higher ADCC in vitro (Fig. 5A) and antitumor activity against hPDPNexpressing tumors (Fig. 6).
Taken together, chLpMab-2 may be useful as a novel antibody-based therapy against hPDPN-expressing tumors with no unexpected side effects.