A Novel Anti-Cadherin 8 Monoclonal Antibody, Ca₈Mab-4, for Versatile Applications

1. Introduction

The cadherin superfamily includes over 100 cell-surface glycoproteins, which are characterized by conserved extracellular cadherin repeats [1,2]. The first identified cadherins (E-Cadherin/CDH1, N-Cadherin/CDH2, P-Cadherin/CDH3) and their closest relatives are classified as classical cadherins. These are divided into type I (CDH1–CDH4 and CDH15) and type II (CDH5–CDH12, CDH18–CDH20, CDH22, and CDH24) cadherins [3]. Classical cadherins are crucial for tissue development and maintenance in vertebrates [4]. In the nervous system, they participate in a wide range of developmental processes, including neurulation, neuronal migration, neurite outgrowth, axonal fasciculation, synaptic differentiation, and synaptic plasticity [5].

Each cadherin in the brain is expressed in specific groups of functionally connected nuclei and laminae [6]. A type II classical cadherin, cadherin-8 (CDH8), plays a crucial role in cold sensation, with its neural circuitry formed by sensory neurons projecting into the spinal cord [7]. The CDH8-expressing sensory neurons were found to connect to CDH8-expressing dorsal horn neurons in the spinal cord, and CDH8 was located near the synaptic junctions formed between these neuronal groups [7].

The neuron-specific transcription factor T-box brain 1 (TBR1) is crucial for brain development [8]. TBR1 haploinsufficiency changed the expression of CDH8, resulting in decreased inter- and intra-amygdalar connectivity and cognitive problems in a mouse model [9]. These developmental abnormalities are likely to impair neuronal activation in response to behavioral stimuli, as evidenced by a reduced number of c-FOS–positive neurons in the TBR1 (+/−) amygdalae [9].

Autism spectrum disorder is characterized by impairments in social communication and learning disability and is implicated to arise from aberrant synaptic connectivity [10]. For instance, rare variants in the neuroligin and neurexin genes, which encode synaptic adhesion molecules that interact across the synaptic cleft, have been linked to increased susceptibility to autism [11,12]. Furthermore, rare familial microdeletions on chromosome 16q21 that disrupt CDH8 were identified in families with autism spectrum disorder and learning disabilities [13]. In a family, three of the four boys with autism and learning disabilities inherited the deletion, but it was not present in their four unaffected siblings or their unaffected mother [13]. Therefore, CDH8 is proposed as a susceptibility factor for autism and learning disabilities.

Monoclonal antibodies (mAbs) that detect CDH8 by Western blotting or immunohistochemistry (IHC) have been developed for various applications; however, suitable mAbs for flow cytometry are not currently available. Using the Cell-Based Immunization and Screening (CBIS) method, our laboratory has previously developed anti-CDH1 [14] and anti-CDH15 [15] mAbs for use in flow cytometry, Western blotting, and IHC. The CBIS method involves high-throughput flow cytometry–based screening, and mAbs produced using this approach usually recognize conformational epitopes, which allows their use in flow cytometry. Notably, some of these mAbs are also compatible with Western blotting and IHC. In this study, we used the CBIS method to develop highly versatile anti-CDH8 mAbs.

2. Materials and Methods

2.1. Cell Lines

Mouse myeloma P3X63Ag8U.1 (P3U1), human glioblastoma (GBM) LN229, and Chinese hamster ovary (CHO)-K1 were obtained from the American Type Culture Collection (ATCC, Manassas, VA, USA). Human esophageal squamous cell carcinoma (SCC) TE5 was obtained from the Cell Resource Center for Biomedical Research Institute of Development, Aging and Cancer at Tohoku University (Miyagi, Japan).

2.2. Stable Transfectants

Genes encoding human CDH8 (NM_001796.5) were obtained from the RIKEN BioResource Research Center (Ibaraki, Japan). The CDH8 cDNA was subcloned into the pCAG-Ble vector with an N-terminal MAP16 tag [16]. Additionally, the CDH8 cDNA with an N-terminal PA16 tag [17] was constructed. These plasmids were transfected into LN229 or CHO-K1cells, and stable transfectants were sorted using an anti-MAP16 tag mAb (clone PMab-1) [16] or an anti-PA16 tag mAb (clone NZ-1) [17] using the Neon transfection system (Thermo Fisher Scientific, Inc.). Finally, MAP16-CDH8-overexpressed LN229 (LN229/CDH8) and PA16-CDH8-overexpressed CHO-K1 (CHO/CDH8) were established.

Type I cadherin-overexpressed CHO-K1 cell lines were previously established in [14]. Type II cadherin-overexpressed CHO-K1 cell lines were established previously [18]. Truncated, seven-domain (7D), and atypical cadherin overexpressed CHO-K1 cell lines were previously established in [18]. Each cadherin expression was confirmed using an anti-CDH1 mAb (clone Ca₁Mab-3 [14],), an anti-CDH3 mAb (clone MM0508-9V11, Abcam, Cambridge, UK), an anti-CDH6 mAb (clone 427909, R&D Systems Inc., Minneapolis, MN, USA), an anti-CDH15 mAb (clone Ca₁₅Mab-1 [15]), an anti-CDH17 mAb (clone 2618, Thermo Fisher Scientific, Inc.), and another anti-PA16-tag mAb (clone NZ-33 [19]) to detect other cadherins.

2.3. Production of Hybridomas

Female BALB/cAJcl mice (CLEA Japan, Tokyo, Japan) were intraperitoneally immunized with LN229/CDH8 cells (1 × 10⁸ cells/injection) mixed with 2% Alhydrogel adjuvant (InvivoGen, San Diego, CA, USA). Following three additional weekly immunizations (1.0 × 10⁸ cells/injection), a booster dose (1 × 10⁸ cells/injection) was administered two days before spleen excision. Hybridomas were produced as previously described [15].

2.4. Flow Cytometry Analysis and Determination of Dissociation Constant Values

CHO/CDH8 and TE5 cells were harvested with 1 mM EDTA were washed with phosphate-buffered saline (PBS) containing 0.1% bovine serum albumin (BSA; blocking buffer). The cells and incubated with Ca₈Mab-4 and flow cytometric data were acquired and the dissociation constant (K_D) values were calculated as described previously [15].

2.5. Western Blotting

Western blotting was performed using 1 μg/mL Ca₈Mab-4, 1 μg/mL of NZ-1, or 1 μg/mL of an anti-isocitrate dehydrogenase 1 (IDH1) mAb (clone RcMab-1) as described previously [15].

2.6. IHC Using Cell Blocks

The formalin-fixed paraffin-embedded (FFPE) cell sections were stained with Ca₈Mab-4 (0.1 or 10 μg/mL), MpMab-2 (10 μg/mL, IgG₁ isotype control, http://www.med-tohoku-antibody.com/topics/001_paper_antibody_PDIS.htm), or NZ-33 (0.01 μg/mL) using the BenchMark ULTRA PLUS with OptiView DAB IHC Detection Kit or ultraView Universal DAB Detection Kit (Roche Diagnostics, Indianapolis, IN, USA).

3. Results

3.1. Development of Anti-CDH8 mAbs by the CBIS Method

An immunogen, LN229/CDH8, was prepared as described in the Materials and Methods. LN229/CDH8 (1 × 10⁸ cells/mouse) was intraperitoneally injected five times into two BALB/cAJcl mice (Figure 1A). Hybridomas were produced by fusing splenocytes with myeloma P3U1 (Figure 1B). The supernatants of the hybridoma were screened to identify those positive for CHO/CDH8 and negative for CHO-K1 (Figure 1C). As a result, 54 positive wells out of 956 (5.6%) were found. Limiting dilution was then performed to clone hybridomas producing anti-CDH8 mAb (Figure 1D). Finally, 4 clones were established, and the purified mAbs were prepared.

3.2. Flow Cytometry Analyses of Ca₈Mab-4 Against CHO-K1, CHO/CDH8, and TE5

Among four clones, we selected Ca₈Mab-4 (IgG₁, κ) based on its reactivity in flow cytometry and suitability for Western blotting (http://www.med-tohoku-antibody.com/topics/001_paper_antibody_PDIS.htm). Figure 2 shows the flow cytometry analysis using Ca₈Mab-4 against CHO/CDH8 and CHO-K1. Ca₈Mab-4 reacted in a dose-dependent manner with CHO/CDH8 from 10 to 0.01 μg/mL (Figure 2A). In contrast, Ca₈Mab-4 did not recognize CHO-K1 even at 10 μg/mL (Figure 2A). Additionally, Ca₈Mab-4 reacted with human esophageal SCC TE5 in a dose-dependent way (Figure 2B), indicating that TE5 expresses endogenous CDH8. The binding affinity of Ca₈Mab-4 was assessed through flow cytometry. The fitted binding isotherms of Ca₈Mab-4 binding to CHO/CDH8 and TE5 are shown in Figure 2C. The K_D values were 3.8 × 10⁻⁹ M for CHO/CDH8 and 4.9 × 10⁻¹⁰ M for TE5. These findings demonstrate that Ca₈Mab-4 has a high binding affinity for CDH8-positive cell lines.

3.3. Determination of the Specificity of Ca₈Mab-4 Using CDHs-Overexpressed CHO-K1

We previously established CHO-K1 cells, which overexpressed type I cadherins (CDH1–CDH4 and CDH15) [14,15], type II cadherins (CDH5–CDH12, CDH18–CDH20, CDH22, and CDH24), a truncated cadherin (CDH13), 7D cadherins (CDH16 and CDH17), and an atypical cadherin (CDH26) [18]. Therefore, the specificity of Ca₈Mab-4 to those cadherins was determined. As shown in Figure 3A, Ca₈Mab-4 recognized CHO/CDH8 but did not react with other cadherins-overexpressed CHO-K1 cells. The cell surface expression of each cadherin was confirmed in Figure 3B. These results indicate that Ca₈Mab-4 is a specific mAb to CDH8 among those CDHs.

3.4. Western Blotting Using Ca₈Mab-4

We next tested whether Ca₈Mab-4 is suitable for Western blotting. Whole-cell lysates from CHO-K1 and CHO/CDH8 were analyzed. Ca₈Mab-4 detected bands around 63–100 kDa in CHO/CDH8, but not in CHO-K1 (Figure 4A). An anti-PA16 mAb (NZ-1) primarily detected 100 kDa in CHO/CDH8 (Figure 4B). An internal control, IDH1, was detected by RcMab-1 (Figure 4C). These results demonstrate that Ca₈Mab-4 can detect CDH8 in Western blotting.

3.5. IHC Using Ca₈Mab-4 in FFPE Cell Blocks

We next tested whether Ca₈Mab-4 is suitable for IHC in FFPE sections from CHO-K1 and CHO/CDH8. Ca₈Mab-4 showed intense membranous and cytoplasmic staining in CHO/CDH8 but not in CHO-K1 (Figure 5A). Additionally, an anti-PA16 tag mAb (NZ-33) exhibited a similar staining (Figure 5B). Ca₈Mab-4 also showed a membranous staining in TE5, but the isotype control mAb (MpMab-2) did not. These results indicate that Ca₈Mab-4 can detect exogenous and endogenous CDH8 in IHC of FFPE sections of cultured cells.

4. Discussion

CDH8 has five extracellular cadherin repeats, one of which mediates calcium-dependent homophilic and heterophilic interactions [20]. In this study, we developed a novel anti-CDH8 mAb using the CBIS method (Figure 1). A clone Ca₈Mab-4 showed strong recognition of both exogenous and endogenous CDH8 in flow cytometry and IHC (Figure 3 and Figure 5). Importantly, Ca₈Mab-4 exhibited high affinity (Figure 2C) and specificity for CDH8 without detectable cross-reactivity to other 21 cadherins, including type II, type I, 7D, truncated, and atypical CDH (Figure 3). Therefore, Ca₈Mab-4 could be helpful for isolating CDH8-positive cells using fluorescence-activated cell sorting. Since cross-reactivity data are unavailable for commercially supplied mAbs, careful validation is necessary, and caution should be exercised when using these mAbs. Additionally, identifying the epitope recognized by Ca₈Mab-4 will be crucial for developing highly specific anti-CDH8 mAbs. Furthermore, Ca₈Mab-4 is suitable for IHC of cell block specimens (Figure 5). Notably, IHC was performed on an automated slide-staining system, ensuring standardized and reproducible staining conditions. Overall, Ca₈Mab-4 is a versatile antibody with broad applications in basic research and potential clinical use.

We found that Ca₈Mab-4 recognized the human esophageal SCC TE5 cell line in flow cytometry and IHC (Figure 3 and Figure 5). Although we examined the reactivity of Ca₈Mab-4 in other esophageal SCC and glioblastoma cell lines, TE5 is the only cell line recognized by Ca₈Mab-4. No studies have examined the role of CDH8 in tumors. Further studies will be essential to clarify the roles of CDH8 in tumor proliferation and metastasis, as well as its expression in various human tumors. Although the extracellular domain of cadherins mediates calcium-dependent homophilic binding [2], CDH8 was reported to make not only homophilic binding, but also heterophilic one with another type II cadherin, CDH11 [21]. Since CDH11/OB-cadherin is predominantly expressed in mesenchymal cells and involved in fibrosis [22], the interaction between CDH8-positive tumor cells and CDH11-positive mesenchymal cells in the tumor microenvironment should be investigated in future studies [23].

We previously cloned cDNAs from hybridomas and produced recombinant mouse IgG_2a-type mAbs to enhance antibody-dependent cellular cytotoxicity (ADCC). Using human tumor xenograft models, antitumor activities have been evaluated [24,25]. We have cloned the cDNA of Ca₈Mab-4, and the IgG_2a-type Ca₈Mab-4 will be produced and evaluated for in vitro ADCC and antitumor efficacies in mouse tumor xenograft models.

Using in situ hybridization, CDH8 was detected in the developing cortex of a 9-week-old human embryo [13]. As shown in Figure 5, Ca₈Mab-4 is suitable for IHC. Therefore, Ca₈Mab-4 will contribute to the analysis of the distribution and subcellular localization of CDH8 in the human central nervous system. CDH8 is a TBR1 target in the cortex [26], and these have been implicated as risk factors in behavioral disorders such as autism [27,28]. This pathway is consistent with the hypothesis that dendritic defects contribute to the pathogenesis of disorders arising from aberrant neuronal wiring. Ca₈Mab-4 will also help clarify the hypothesis.

The CDH8-mediated adhesive code that determines neuronal connectivity has been clarified in mouse models [7,29,30]. In the mouse retina, the dendrites of over 40 different retinal ganglion cells (RGCs) arborize within the inner plexiform layer [31]. The dendrites are limited to one or several distinct sublaminae. Within these sublaminae, RGC dendrites receive synaptic inputs from at least 70 types of interneurons, including amacrine and bipolar cells [32,33]. In TBR1-expressing RGCs, the TBR1-CDH8 axis is required for their laminar specification [34]. Therefore, Ca₈Mab-4 would be an essential tool to distinguish or isolate CDH8-positive RGCs in human retina or in vitro differentiated RGCs from induced pluripotent stem cells [35].