Frizzled‐7‐targeting antibody (SHH002‐hu1) potently suppresses non–small‐cell lung cancer via Wnt/β‐catenin signaling

Abstract

Non–small‐cell lung cancer (NSCLC) is one of the deadliest cancers worldwide, and metastasis is considered one of the leading causes of treatment failure in NSCLC. Wnt/β‐catenin signaling is crucially involved in epithelial–mesenchymal transition (EMT), a crucial factor in promoting metastasis, and also contributes to resistance developed by NSCLC to targeted agents. Frizzled‐7 (Fzd7), a critical receptor of Wnt/β‐catenin signaling, is aberrantly expressed in NSCLC and has been confirmed to be positively correlated with poor clinical outcomes. SHH002‐hu1, a humanized antibody targeting Fzd7, was previously successfully generated by our group. Here, we studied the anti‐tumor effects of SHH002‐hu1 against NSCLC and revealed the underlying mechanism. First, immunofluorescence (IF) and near‐infrared (NIR) imaging assays showed that SHH002‐hu1 specifically binds Fzd7⁺ NSCLC cells and targets NSCLC tissues. Wound healing and transwell invasion assays indicated that SHH002‐hu1 significantly inhibits the migration and invasion of NSCLC cells. Subsequently, TOP‐FLASH/FOP‐FLASH luciferase reporter, IF, and western blot assays validated that SHH002‐hu1 effectively suppresses the activation of Wnt/β‐catenin signaling, and further attenuates the EMT of NSCLC cells. Finally, the subcutaneous xenotransplanted tumor model of A549/H1975, as well as the popliteal lymph node (LN) metastasis model, was established, and SHH002‐hu1 was demonstrated to inhibit the growth of NSCLC xenografts and suppress LN metastasis of NSCLC. Above all, SHH002‐hu1 with selectivity toward Fzd7⁺ NSCLC and the potential of inhibiting invasion and metastasis of NSCLC via disrupting Wnt/β‐catenin signaling, is indicated as a good candidate for the targeted therapy of NSCLC.

The novel humanized Fzd7‐targeting antibody (SHH002‐hu1) generated by our group was demonstrated to selectively bind NSCLC cells overexpressing Fzd7 and specifically target Fzd7+ NSCLC tumor tissues, enabling the specific Wnt/β‐catenin signaling inhibition. SHH002‐hu1 shows the potential for inhibiting migration and invasion of NSCLC cells in vitro, inhibiting the growth of NSCLC xenografts, and preventing the popliteal lymph node metastasis of NSCLC, resulting from attenuating EMT of NSCLC cells. SHH002‐hu1 is indicated as a good candidate for the targeted therapy of NSCLC.

Abbreviations

ALK

aplastic lymphoma kinase

CRD

cysteine‐rich domains

EMT

epithelial–mesenchymal transition

Fzd7

Frizzled‐7

ICIs

immune checkpoint inhibitors

IF

immunofluorescence

IHC

immunohistochemistry

LEF

lymphoid enhancing factor

LN

lymph node

NIR

near infrared

NSCLC

non–small‐cell lung cancer

PD‐1

programmed cell death protein‐1

PD‐L1

programmed cell death ligand‐1

TCF

T‐cell factor

TKIs

tyrosine kinase inhibitors

VEGF

vascular endothelial growth factor

1. INTRODUCTION

Lung cancer is the most frequent cause of cancer‐related death worldwide. According to histological classification, ~85% of lung cancer patients belong to the NSCLC group. With the support of next‐generation sequencing and other genomic profiling tools, the treatment of NSCLC has changed from the empirical use of chemotherapy and radiotherapy to a hallmark of targeted therapy and personalized medicine. ¹ , ² Although epidermal growth factor receptor–tyrosine kinase inhibitors (epidermal growth factor receptor [EGFR]–TKIs) and ALK–TKIs have a favorable and durable treatment response, most patients will eventually develop progressive disease within ~1 year of treatment. Furthermore, acquired resistance develops and limits the long‐term efficacy of these EGFR/ALK–TKIs. ³ , ⁴ , ⁵ Currently, ICIs that target the PD‐1/PD‐L1 axis have significantly shifted the treatment paradigm in advanced NSCLC. However, the population of patients who benefit from the treatment remains modest, some of whom would relapse and progress eventually. ⁶ , ⁷ Moreover, NSCLCs harboring EGFR mutations or ALK rearrangements are associated with low objective response rates to PD‐1/PD‐L1 inhibitors. ⁸ , ⁹ , ¹⁰ , ¹¹ Therefore, it is imperative to explore novel therapeutic options to improve the treatment outcome of NSCLC.

The Wnt/β‐catenin pathway has been broadly implicated in human cancers, including NSCLC. The aberrant activation of Wnt/β‐catenin signaling is tightly linked to the increment of prevalence, advancement of malignant progression, and the development of poor prognostics. ¹² , ¹³ , ¹⁴ , ¹⁵ , ¹⁶ Metastasis is considered one of the leading causes of treatment failure and death in NSCLC patients, and the Wnt/β‐catenin pathway is demonstrated to be correlated strongly with EMT. Furthermore, Wnt/β‐catenin signaling is involved in the resistance to EGFR‐TKIs and ICIs. ¹⁷ , ¹⁸ Wnt/β‐catenin signaling has been confirmed to play a role in the survival of EGFR mutant NSCLC during EGFR TKI treatment and in drug resistance. ¹⁹ , ²⁰ In addition, β‐catenin activation results in the binding of the β‐catenin/TCF/LEF complex to the promoter region of the CD274 gene to induce PD‐L1 expression, further promoting tumor immune evasion and impairing the effects of PD‐1/PD‐L1 inhibitors. ¹⁷ , ²¹ In conclusion, Wnt/β‐catenin signaling is the potential therapeutic target of NSCLC.

Fzd7, one of the major receptors, is upregulated along the Wnt/β‐catenin signaling pathway in NSCLC. ¹⁴ , ²² Given the important role that Fzd7 plays in tumorigenesis and progression, to date, several methods have been designed to antagonize Wnt/β‐catenin signaling by targeting Fzd7. ²³ , ²⁴ Pharmacological inhibition of Fzd7 by small interfering peptides, small‐molecule inhibitors, anti‐Fzd7 antibody, or extracellular peptide of Fzd7 has displayed anti‐cancer activities via disrupting the canonical Wnt signaling pathway. ²⁵ , ²⁶ Previously, a humanized antibody targeting Fzd7 (SHH002‐hu1) was generated by our group using hybridoma and antibody humanization techniques. SHH002‐hu1 exhibits an extremely high affinity for Fzd7 (Kd < 1.0 × 10⁻¹² M). ²⁷

Here, we focused our study on the anti‐tumor effects of SHH002‐hu1 against NSCLC and the underlying mechanism. First, IF and NIR imaging assays confirmed the specificity and targeting of SHH002‐hu1. Wound healing and transwell invasion assays demonstrated that SHH002‐hu1 could inhibit the migration and invasion of NSCLC cells. Subsequently, TOP‐FLASH/FOP‐FLASH luciferase reporter, IF staining, and western blot assays verified that SHH002‐hu1 attenuates significantly the activities of Wnt/β‐catenin signaling and inhibits EMT in NSCLC cells. Finally, an in vivo study exhibited that SHH002‐hu1 could suppress the growth and popliteal LN metastasis of NSCLC xenografts. The above‐indicated SHH002‐hu1 is a candidate antibody drug for targeted therapy of NSCLC.

2. MATERIALS AND METHODS

2.1. Cell culture and transfection of siRNA

The human bronchial epithelial cell line BEAS‐2B, human embryonic kidney cell line HEK293T, and human NSCLC cell line A549 were obtained from the ATCC (New York, USA) and cultured in DMEM (Gibco, Grand Island, USA), supplemented with 10% (v/v) FBS (Gibco, Auckland, NZ). The human NSCLC cell lines H1299 and H1975 were obtained from the ATCC and cultured in RPMI 1640 medium, supplemented with 10% (v/v) FBS.

Fzd7‐specific small interfering RNA (siRNA) was synthesized by Genepharma (Shanghai, China). The nucleotide target sequence for Fzd7 (siFzd7) was: 5′‐CCGUCUUCAUGAUCAAGUATT‐3′. Then, transfection was performed using Lipofectamine RNA iMAX reagent (Invitrogen, CA, USA).

2.2. IF assay for the binding of SHH002‐hu1

A549/H1975/BEAS‐2B cells were inoculated into a 6‐well plate. When cell confluency reached 70%–80%, the cells were incubated with SHH002‐hu1 and then goat anti‐human IgG (H&L) (Alexa Fluor 647, Abcam, USA). Subsequently, images were taken using an Olympus fluorescence microscope (Olympus, Tokyo, Japan).

2.3. Dynamics and NSCLC targeting capability using NIR imaging in vivo

Here, 1 × 10⁷ A549/H1975 cells were inoculated subcutaneously into 5‐week‐old female BALB/c nude mice (Shanghai Laboratory, Animal Research Center, China) to develop an implant tumor. An IRB‐NHS fluorescence probe (Keyuandi Biotechnology, Shanghai, China) was incubated with SHH002‐hu1. The product was collected and named NIRB‐SHH002‐hu1. NIRB‐SHH002‐hu1 (50 nmol/kg) was then injected intravenously into NSCLC tumor‐bearing mice. Additionally, free SHH002‐hu1 (2.5 μmol/kg) was mixed with NIRB‐SHH002‐hu1 (50 nmol/kg) to evaluate competitive blocking. Then, the NIR fluorescence imaging system (IVIS Spectrum CT, PerkinElmer, USA) was used to conduct fluorescence imaging. The analysis of the region of interest function was utilized for analyzing tumor/normal tissue ratios (T/N ratio)

2.4. Wound healing assay

Here, 3 × 10⁴ A549/H1975 cells or A549/H1975 cells transfected with Fzd7‐targeted siRNA (Fzd7‐silencing cells) were inoculated into each well of a Culture‐Insert (Ibidi, Martinsried, Germany) fixed in a 24‐well plate. After adherence, the Culture‐Insert was removed. After the incubation with SHH002‐hu1, the drug‐dissolved medium was replaced. Images were taken using an Olympus inverted microscope (Olympus, Tokyo, Japan). The wound migrated distances were measured using the Image‐Pro‐Plus program and calculated as follows: L _n  = (L ₀ – L _time)/2.

2.5. Transwell invasion assay

Here, 4 × 10⁴ of A549/H1975 cells or Fzd7‐silencing cells suspended in serum‐free medium were plated into the upper wells of the 24‐well transwell chamber (Millipore, Billerica, USA), coated with Matrigel (Corning, Bedford, USA), and then treated with SHH002‐hu1. At 12 h later, non‐invasive cells on the upper layer were removed, whereas the invaded cells were fixed and stained. Images were taken using an Olympus inverted microscope; the invaded cells were counted using the Image‐Pro‐Plus program and invasion percentages were quantified based on the untreated control.

2.6. TOP‐FLASH/FOP‐FLASH luciferase reporter assay

The TCF/LEF transcriptional activity was assessed using the TOP‐FLASH/FOP‐FLASH luciferase reporter assay. A549/H1975 cells or Fzd7‐silencing cells were seeded into 48‐well plates and transiently transfected with the TOP‐Flash/FOP‐Flash luciferase reporter gene (GeneChem, Shanghai, China) and Renilla luciferase transfection control reporter gene plasmids using Lipofectamine™ 2000 (Invitrogen, CA, USA). Then, the cells were treated with 100 nmol/L SHH002‐hu1, at 2 h later, Wnt3a (200 ng/mL) was added. The luciferase activity of TOP‐Flash or FOP‐Flash was monitored at 24 h using the dual‐luciferase reporter assay system (Promega, WI, USA). The TOP/FOP ratio was then calculated to assess the activity of the Wnt/β‐catenin pathway.

2.7. IF assay for the expression and location of β‐catenin

Here, 5 × 10⁵ A549/H1975 cells or Fzd7‐silencing cells were seeded onto coverslips in 6‐well plates and allowed to adhere. When reaching 70%–80% confluency, the cells were treated with 100 nmol/L of SHH002‐hu1, and at 2 h later, Wnt3a (200 ng/mL) was added. After another 22 h, cells were fixed and then incubated with anti‐β‐catenin (Cell Signaling Technology, USA) followed by a FITC‐conjugated secondary antibody.

2.8. Western blot assay

Here, 5 × 10⁵ A549/H1975 cells or Fzd7‐silencing cells were inoculated into 6‐well plates and allowed to adhere. When reaching 70%–80% confluency, the cells were treated with 25/50/100 nmol/L SHH002‐hu1, and at 2 h later, Wnt3a (200 ng/mL) was added. At 48 h later, the whole cell proteins were extracted from cells using RIPA buffer (Beyotime, Shanghai, China), the nuclear extracts were prepared using a NE‐PER Nuclear and Cytoplasmic Extraction Kit (Pierce Biotechnology, Rockford, USA), and the membrane proteins were extracted using a Membrane and Cytosol Protein Extraction Kit (Beyotime). Western blots were probed with anti‐phospho LRP6 (Ser1490), anti‐LRP6, anti‐active β‐catenin (non‐phospho β‐catenin, Ser45), anti‐β‐catenin, anti‐E‐cadherin, anti‐N‐cadherin, anti‐Vimentin, anti‐Snail, anti‐Histone H3, anti‐c‐Myc, anti‐CD44, anti‐β‐actin (Cell Signaling Technology, MA, USA), anti‐VEGFA, and anti‐Frizzled‐7 (Abcam, Cambridge, UK) antibodies.

2.9. Xenograft model and administration

NSCLC xenograft models were established by subcutaneously injecting A549/H1975 cells into BALB/c nude mice as described previously. When the average tumor volume reached 150 mm³ (A549)/200 mm³ (H1975), mice were randomized into five groups (n = 5 for each group), and the experiment started: (1) PBS control; (2) 1.25 mg/kg SHH002‐hu1; (3) 2.5 mg/kg SHH002‐hu1; (4) 5 mg/kg SHH002‐hu1; (5) 10 mg/kg XAV‐939 (Absin, Shanghai, China). SHH002‐hu1 was administered intravenously every 3 days, and XAV‐939 was intraperitoneally administered every 3 days. Tumor development was measured periodically and the tumor volume was determined using the formula V = (length × width²)/2. When the average H1975 transplanted tumor volume of the PBS control reached 2000 mm³, administration was ended, and the H1975 tumors of each group were isolated for further study. For the A549 transplanted tumor, the treatment ended on day 28, and the mice were humanely euthanized.

2.10. IHC and IF analysis for NSCLC tumor tissues

Paraffin sections were cut into 5‐μm sections and fixed in 4% paraformaldehyde. For IHC staining, the sections were incubated with anti‐Ki‐67 (Cell Signaling Technology, MA, USA), anti‐E‐cadherin and anti‐N‐cadherin, and HRP‐labeled secondary antibody and analyzed using the Vectastain ABC Kit (Dako, Copenhagen, Denmark). For IF staining, the sections were incubated with anti‐CD31 (Cell Signaling Technology, MA, USA) and anti‐β‐catenin, and followed by FITC‐conjugated secondary antibody.

2.11. Popliteal LN metastasis study

Here, 1 × 10⁷ A549‐luc/H1975‐luc cells that stably expressed luciferase were inoculated subcutaneously into the footpads of 5‐week‐old female BALB/c nude mice to develop an implant tumor. When the average tumor volume reached 50 mm³, mice were randomized into two groups (n = 5 for each group), and the administration began: (1) PBS control; (2) 5 mg/kg SHH002‐hu1 (intravenous injection, every 3 days). The mice were imaged using an IVIS spectrum imaging system 15 min after intraperitoneal injection of d‐luciferin potassium salt (Beyotime, Shanghai, China) to monitor lymphatic metastasis. When the average footpad tumor volume of PBS control reached 250 mm³, the mice were humanely euthanized, and the popliteal LNs were enucleated.

2.12. Statistical analysis

All the quantitative data were expressed as the mean ± SD, and all experiments were repeated independently at least three times. Student's t‐test and ANOVA were used to determine significant differences. GraphPad Prism 8 software was used to analyze statistical differences in data, a p‐value < 0.05 represented a statistically significant difference.

3. RESULTS

3.1. SHH002‐hu1 selectively binds NSCLC cells overexpressing Fzd7 and specifically targets Fzd7⁺ NSCLC tumor tissues

First, the IF assay was executed to confirm the Fzd7⁺ cell‐binding capacity of SHH002‐hu1. The western blot assay (Figure 1A) showed that there was little Fzd7 expressed on BEAS‐2B and HEK293T cells. We then found that, among the three detected NSCLC cell lines, A549 and H1975 cells expressed high levels of Fzd7 relative to BEAS‐2B cells. Subsequently, the IF assay (Figure 1B) showed that SHH002‐hu1 selectively bound Fzd7 overexpressing A549 and H1975 cells, rather than noncancerous bronchial epithelial cells BEAS‐2B. In addition, A549/H1975 cells showed little binding with SHH002‐hu1 when Fzd7 expression was suppressed by siFzd7, which confirmed the selectively of SHH002‐hu1.

FIGURE 1.

SHH002‐hu1 specifically targets Fzd7⁺ NSCLC cells and Fzd7⁺ NSCLC tumor tissue. (A) Western blot assay to detect the expression of Fzd7 in HEK293T/BEAS‐2B/NSCLC cells. The detection antibody was rabbit anti‐human Fzd7 purchased from Abcam. (B) IF assay to detect the binding of SHH002‐hu1 with NSCLC cells. Bars = 100 μm. (C, D) NIRB imaging assay to evaluated the bio‐distribution of NIRB‐SHH002‐hu1 in A549/H1975‐bearing nude mice. In blocking experiments, free SHH002‐hu1 inhibited the probes from binding to the tumor sites. Tumor/normal tissue ratios calculated at 12 h post‐injection of probe groups into A549/H1975‐bearing nude mice from the region of interest were shown. Data are given as the mean ± SD (n = 3), **p < 0.01, ***p < 0.001, Student's t‐test was used to determine significant differences

Next, NIR imaging was used to assess the dynamics and Fzd7⁺ NSCLC targeting capability of NIRB‐SHH002‐hu1 in vivo (Figure 1C,D). After the injection of NIRB‐SHH002‐hu1, the fluorescent signals spread throughout the body immediately. At 2 h later, some fluorescent antibody was excreted through the kidneys. At 4 h later, A549/H1975 NSCLC xenografts were distinguished by fluorescence. The fluorescence signal was maintained for 24 h. At 48 h later, there were no fluorescent signals detected. Meanwhile, the blocking group showed no intense fluorescent signals at the tumor site, indicating that tumor targeting was mediated by SHH002‐hu1. The fluorescent signal was significantly different, with a maximal tumor/normal tissue ratio at 12 h of 8.62 ± 0.46 (A549)/2.23 ± 0.07 (H1975) and 0.41 ± 0.03 (A549)/0.3 ± 0.05 (H1975) for the NIRB‐SHH002‐hu1‐treated group and the corresponding blocking group, respectively. We also conducted the NIR imaging assay with the Fzd7‐negative H1299 cells; the results are shown in Figure S1. Unlike in A549/H1975 xenografts models, NIRB‐SHH002‐hu1 could not target H1299 tumor tissue effectively with shorter maintenance time in vivo. Consequently, SHH002‐hu1 can selectively bind NSCLC cells that overexpressed Fzd7 and specifically target Fzd7⁺ NSCLC tumor tissues.

3.2. SHH002‐hu1 inhibits the migration and invasion of A549 and H1975 cells in vitro

Wound healing and transwell invasion assays were executed to investigate whether SHH002‐hu1 could inhibit the migration and invasion of A549/H1975 cells. As shown in Figure 2A–D, 100 nmol/L SHH002‐hu1 significantly inhibited the migration of A549/H1975 cells, and cells transfected with Fzd7‐targeted siRNA had a marked decrease in cell migration compared with the control groups; however, SHH002‐hu1 could not inhibit the migration of Fzd7‐silencing A549/H1975 cells. Figure 2E–H shows that the invasion of A549/H1975 cells through Matrigel was obviously inhibited by SHH002‐hu1 and siFzd7; likewise, SHH002‐hu1 exhibited little effect on the invasion of Fzd7‐silencing A549/H1975 cells. Microscopic views of 25/50/100 nM SHH002‐hu1 groups are shown in Figure S2A–D. In conclusion, SHH002‐hu1 inhibits the migration and invasion of NSCLC cells in an Fzd7‐dependent manner.

FIGURE 2.

SHH002‐hu1 inhibits the migration and invasion of A549/H1975 cells. (A, B) Photomicrographs of A549/H1975 cell migration, bars = 100 μm. (C, D) Quantitative analysis of A549/H1975 cell migration using ImageJ software. (E, F) Microscope views from a transwell invasion assay to estimate A549/H1975 cell invasion, bars = 100 μm. (G, H) Quantitative analysis of A549/H1975 cell invasion using ImageJ software. Data were presented as the mean ± SD, n = 3, *p < 0.05, **p < 0.01, ***p < 0.001, NS, no significance, Student's t‐test was used to determine significant differences[Corrections made on 14 March 2023, after first online publication: Figure 2 has been corrected in this version.]

To exclude the impact of cell proliferation on cell migration and invasion, a CCK‐8 assay was utilized to detect the effect of SHH002‐hu1 on A549/H1975 cell proliferation under the same culture conditions (serum starvation) with wound healing and transwell invasion assays. As shown in Figure S3A, SHH002‐hu1 exhibited little effect on A549/H1975 cell proliferation in the above concentration range. Additionally, SHH002‐hu1 also showed no significant cytotoxicity toward noncancerous bronchial epithelial cells BEAS‐2B, indicating the relatively fewer side effects of SHH002‐hu1 in the clinical application. Above all, SHH002‐hu1 exhibited the potential to impair the migration and invasion of NSCLC cells.

As we know, c‐Myc, a downstream protein of Wnt/β‐catenin signaling is involved in the proliferation of tumor cells. Hence, we executed the cell colony formation assay to verify the efficacy of SHH002‐hu1 in inhibiting cell proliferation, ²⁵ and the colony formation assay was conducted in medium containing 10% serum. The results (Figure S3B,C) revealed that 100 nM SHH002‐hu1 or Fzd7 siRNA markedly decreased colony formation and growth.

3.3. SHH002‐hu1 strongly attenuates Wnt/ β‐catenin signaling pathway activated by Wnt3a

TOP‐FLASH/FOP‐FLASH luciferase reporter assay was used to measure Wnt/β‐catenin downstream TCF/LEF transcription activity. As shown in Figure 3A,B, SHH002‐hu1, as well as siFzd7, markedly decreased the transcriptional activity of TCF/LEF stimulated by Wnt3a; and SHH002‐hu1 showed no additive effect when treating Fzd7‐silenced A549/H1975 cells. Since the transcriptional activity of β‐catenin requires its translocation from the cytoplasm to the nucleus, an IF assay was performed to evaluate the distribution pattern of β‐catenin. Wnt3a promoted the translocation of β‐catenin from the cytoplasm to the nucleus in NSCLC cells compared with the control. Furthermore, SHH002‐hu1 or siFzd7 strongly suppressed the nuclear β‐catenin accumulation induced by Wnt3a (Figure 3C–F); likewise, the additive effect was not found in the siFzd7 ⁺ SHH002‐hu1 groups. Notably, SHH002‐hu1 or siFzd7 not only impaired the nuclear translocation and accumulation of β‐catenin, but also downregulated the expression of the total β‐catenin protein.

FIGURE 3.

SHH002‐hu1 inhibits Wnt3a‐induced transcriptional activity of TCF/LEF and nuclear translocation of β‐catenin in A549/H1975 cells. (A, B) TOP/FOP ratio in A549/H1975 cells. The relative luciferase signal normalized to the control Renilla luciferase and relative to the signal in the absence of Wnt3a is shown. (C, D) SHH002‐hu1 attenuated the Wnt3a‐induced accumulation of β‐catenin in the nucleus of A549/H1975 cells. IF staining of β‐catenin (green) is shown, nuclei are counterstained with DAPI (blue), bars = 50 μm. (E, F) Quantitative analysis of (C) and (D). The co‐localization of blue and green fluorescence staining indicated the nuclear β‐catenin⁺ cells. Nuclear β‐catenin⁺ cells were counted using ImageJ software. Data are presented as the mean ± SD, n = 3, *p < 0.05, **p < 0.01, ***p < 0.001, NS, no significance, Student's t‐test was used to determine significant differences

As shown in Figure 4A, Wnt3a or SHH002‐hu1 showed little effect on the expression of Fzd7, indicating that the function blocking of Wnt/β‐catenin signaling resulted from SHH002‐hu1 competition with Wnt3a for Fzd7 binding. Figure 4B–D demonstrates that SHH002‐hu1 or siFzd7 significantly downregulated the expression of nuclear β‐catenin and active β‐catenin induced by Wnt3a; while, SHH002‐hu1 could not enhance the effects of siFzd7, further verifying that SHH002‐hu1 abolished the functional activity of Wnt/β‐catenin signaling in an Fzd7‐dependent manner. Figure 4E and Figure S4A,B showed that SHH002‐hu1 significantly reduced both Wnt3a‐induced nuclear accumulation of β‐catenin and phosphorylation of LRP6 in a concentration‐dependent manner. Intriguingly, total β‐catenin expression was also reduced by SHH002‐hu1 (Figure S4C), which was consistent with the results of the IF assay. Figure S4C also revealed that SHH002‐hu1 inhibited the expression of active β‐catenin, c‐Myc, VEGFA, and CD44 in a concentration‐dependent manner. Above all, we demonstrated that SHH002‐hu1 strongly attenuated the Wnt/β‐catenin signaling pathway activated by Wnt3a via targeting Fzd7.

FIGURE 4.

SHH002‐hu1 blocks the phosphorylation of LRP6, decreases the expression of downstream oncoproteins of Wnt/β‐catenin signaling, and inhibits the EMT of NSCLC cells. (A) Western blot analysis for the expression of Fzd7, β‐actin was used as the loading control. (B) Western blot analysis for the expression of nuclear β‐catenin and active β‐catenin. Histone H3 was used as the loading control for nuclear proteins. (C) Relative expression level of nuclear β‐catenin indicated in (B). (D) Relative expression level of active β‐catenin in (B). Data are presented as the mean ± SD, n = 3, *p < 0.05, **p < 0.01, ***p < 0.001. Student's t‐test was used to determine significant differences. (E) Western blot analysis for the expression of key proteins involved in Wnt/β‐catenin signaling. (F) Western blot analysis for the expression of EMT marker proteins in A549/H1975 cells. siFzd7 was set as a positive control

3.4. SHH002‐hu1 reverses EMT of NSCLC induced by Wnt3a through abating Wnt/β‐catenin signaling

One hallmark of EMT is the increased level of active β‐catenin. Active β‐catenin directly binds to transcription factors associated with the promoters of the key EMT inducers Slug, ZEB1, and Twist and induces the expression of these inducers. ²⁸ , ²⁹ Here we detected the expression of various classical EMT markers in NSCLC cells to investigate the impact of SHH002‐hu1 on EMT. Figure 4F and Figure S4D indicated that Wnt3a increased the expression of N‐cadherin and Vimentin, and decreased the expression of E‐cadherin. The expression of EMT‐related transcription factor Snail was also increased by Wnt3a. Whereas, the stimulated EMT of A549/H1975 cells induced by Wnt3a was significantly weakened, along with the attenuated Wnt/β‐catenin pathway caused by SHH002‐hu1, which was probably why SHH002‐hu1 strongly suppressed the migration and invasion of NSCLC cells.

3.5. SHH002‐hu1 inhibits the growth of human NSCLC xenografts by blocking the Wnt/ β‐catenin pathway

The impact of inhibiting Wnt/β‐catenin signaling on lung tumor growth was assessed using human NSCLC xenografts in nude mice. The pharmacological β‐catenin inhibitor XAV‐939, a tankyrase inhibition that stabilizes axin and antagonizes Wnt signaling, was set as a positive control. ³⁰ , ³¹ The treatment of 5 mg/kg SHH002‐hu1 and 10 mg/kg XAV‐939 resulted in significant inhibition of growth in human NSCLC xenografts, compared with the PBS control; notably, 5 mg/kg SHH002‐hu1 exhibited stronger anti‐NSCLC effects than 10 mg/kg XAV‐939 in A549 xenografts (Figure 5A,B). Figure S5 indicated that SHH002‐hu1 inhibited the growth of NSCLC xenografts in a dose‐dependent manner. To confirm that SHH002‐hu1 functioned by blocking Fzd7 specifically, we verified whether SHH002‐hu1 could inhibit the tumor growth of Fzd7 knockout A549/H1975 xenografts. As shown in Figure S6A,B, knocking out of Fzd7 significantly inhibited NSCLC tumorigenesis and tumor growth; while SHH002‐hu1 showed no additive effect on the tumor growth of Fzd7 knockout A549/H1975 xenografts, confirming that SHH002‐hu1 functions by blocking Fzd7 in vivo.

FIGURE 5.

SHH002‐hu1 inhibits the growth of human NSCLC xenografts. (A) Representative images of isolated tumors from A549/H1975 tumor‐bearing nude mice. (B) A549/H1975 tumor growth curves. Data are given as the mean ± SD (n = 5), *p < 0.05, **p < 0.01, NS, no significance. (C) IHC staining of Ki‐67 (brown staining) on paraffin sections of A549/H1975 xenografted tumors. Bars = 25 μm. (D) Ki‐67‐positive cells from tumor tissues were counted using ImageJ software, and the percentage of Ki‐67‐positive cells of each group is shown. Data are given as the mean ± SD (n = 3), *p < 0.05, **p < 0.01, NS, no significance, Student's t‐test was used to determine significant differences

Tumor xenografts were harvested for IHC and IF at the end of the treatment period. IHC demonstrated that there was a significant decrease in the numbers and intensity of cell proliferation marker Ki‐67 in 2.5/5 mg/kg SHH002‐hu1 and XAV‐939 groups compared with the PBS control (Figure 5C,D). Then, the staining of E‐cadherin and N‐cadherin was executed to evaluate the EMT of NSCLC. Figure 6A shows that SHH002‐hu1, as well as XAV‐939, obviously upregulated the expression of epithelial marker E‐cadherin in A549 and H1975 tumor tissues; meanwhile, Figure 6B demonstrated SHH002‐hu1 and XAV‐939 downregulated the expression of mesenchymal marker N‐cadherin effectively, suggesting the potential of SHH002‐hu1 to impair EMT of NSCLC in vivo. In addition, CD31 staining was executed by IF assay to evaluate tumor angiogenesis in vivo. The markedly reduced intensity of CD31 staining in the SHH002‐hu1/XAV‐939 treated group, especially the high‐dose SHH002‐hu1 group (Figure 6C), indicated that SHH002‐hu1 significantly inhibited NSCLC angiogenesis.

FIGURE 6.

SHH002‐hu1 inhibits angiogenesis and EMT in A549/H1975‐tumor xenografts, and represses the accumulation of β‐catenin in the nucleus. (A, B) IHC staining of E‐cadherin and N‐cadherin (brown staining) on paraffin sections of A549/H1975 xenografted tumors. (C) IF staining of CD31 (green fluorescence) on paraffin sections of A549/H1975 implanted tumors. (D) IF staining of β‐catenin (green fluorescence) on paraffin sections of A549/H1975 xenografted tumors. Bars = 25 μm

The IF assay was performed to demonstrate whether SHH002‐hu1 could impact the distribution pattern of β‐catenin. Figure 6D shows that 2.5 mg/kg SHH002‐hu1 and 10 mg/kg XAV‐939 effectively inhibited the nuclear import and accumulation of β‐catenin; 5 mg/kg SHH002‐hu1 not only impaired the nuclear translocation of β‐catenin, but also significantly downregulated the expression of total β‐catenin protein. To sum up, SHH002‐hu1 exhibits superior anti‐tumor activities against NSCLC through impairing the EMT of NSCLC cells and blocking NSCLC angiogenesis by disrupting the Wnt/β‐catenin pathway, and achieved a better therapeutic effect than the small‐molecule inhibitor, XAV‐939.

3.6. SHH002‐hu1 suppresses LN metastasis of NSCLC

To examine the role of SHH002‐hu1 in the NSCLC metastasis in vivo, a popliteal LN metastasis model was used. Figure 7A,B shows the representative images of the footpad tumors and popliteal LNs from nude mice bearing A549 footpad xenografts and the images of bioluminescence. Figure 7A,B and Figure S7A indicated that almost all the nude mice in the control group developed severe lymphadenopathy (swollen lymph nodes), whereas there were fewer enlarged lymph nodes observed in nude mice from the SHH002‐hu1 group. The volume of the popliteal LN was smaller in the SHH002‐hu1 group compared with the control (Figure 7C,E). Moreover, the SHH002‐hu1 group exhibited a lower LN metastatic rate than the control (Figure 7D). We further performed the IF assay to detect the expression and localization of β‐catenin in popliteal LNs. The results (Figure S7B) indicated that SHH002‐hu1 could markedly downregulate the expression of nuclear β‐catenin, as well as total β‐catenin. Collectively, SHH002‐hu1 inhibited the popliteal LN metastasis of A549/H1975 cells Strikingly, additionally, we found that the primary tumor size in the SHH002‐hu1 group was lower than in the control group (Figure S7A), suggesting that SHH002‐hu1 suppressed the tumorigenesis of NSCLC.

FIGURE 7.

SHH002‐hu1 suppresses LN metastasis of NSCLC in vivo. (A) Representative images of the footpad tumors and popliteal LNs from nude mice bearing A549 footpad xenografts. (B) Representative images of bioluminescence from nude mice bearing A549 footpad xenografts. (C) Representative images of enucleated popliteal LNs. (D) The ratio of popliteal LN metastasis was calculated for all groups. (E) The statistical data for popliteal LN volume. Data are given as the mean ± SD (n = 5), *p < 0.05, **p < 0.01, Student's t‐test was used to determine significant differences

4. DISCUSSION

Wnt signaling via Fzd7 is reported to have oncogenic potential, and Fzd7 is overexpressed in multiple tumors. ²² Selective targeting of Fzd7 suppresses oncogenic Wnt signaling without interfering with tumor‐suppressive Wnt7a signaling. ³² , ³³ OMP‐18R5, an anti‐Fzd7 antagonistic antibody developed by OncoMed Pharmaceuticals, Inc., decreases the tumor‐initiating cell frequency and suppressed the growth of tumor in the lung, pancreatic, and metastatic breast cancer patient‐derived xenografts. ¹⁴ Further, clinical data demonstrated the efficacy of OMP‐18R5, indicating that the targeting of Fzd7 is a promising therapeutic strategy for cancer treatment. ²⁵ , ³⁴ Given the important role of Wnt ligand binding, the sequences in CRD were highly conserved between Fzd receptors. The CRD of Fzd7 shares over 40% sequence identity with other members of the Fzd family, especially Fzd1/2/5/8. ³⁵ , ³⁶ , ³⁷ OMP‐18R5, initially identified by binding to Fzd7, interacts with five Fzd receptors through a conserved epitope. The lack of specificity toward its intended target poses concerns for its off‐target effects and probably becomes one of the main obstructions to the clinical application of OMP‐18R5.

SHH002‐hu1, a high‐affinity antibody specifically targeting the CRD domain of Fzd7 generated by our group, was demonstrated to show no cross‐reactivity with Fzd1/2/5/8. SHH002‐hu1 has been shown to be a potential therapeutic agent that may be used alone or in combination with the antiangiogenesis drugs in the treatment of triple‐negative breast cancer. ²⁷ Considering that striking benefit was observed when OMP‐18R5 was used in the treatment of NSCLC, ³³ here, we studied the anti‐tumor effects of SHH002‐hu1 against NSCLC, especially around the capacity of inhibiting NSCLC invasion and metastasis. Fzd7 is involved in both canonical and non‐canonical Wnt signaling activation. The canonical Wnt/β‐catenin pathway is triggered by a Wnt ligand (Wnt1, Wnt3a, and Wnt8) binding to Fzd7 and low‐density lipoprotein LRP5/6. Fzd7 can also associate with Ror2 or Ryk to transmit non‐canonical Wnt signals from ligands including Wnt5 or Wnt 11, either via phencyclidine (PCP) or Ca²⁺. ³⁸ , ³⁹ What is more, Fzd7 regulates metastasis via canonical or non‐canonical Wnt signaling in various types of tumors. ⁴⁰ This study showed that the phosphorylation of LRP6, the upregulated active β‐catenin, and downstream targets of β‐catenin induced by Wnt3a were downregulated by SHH002‐hu1, indicating that SHH002‐hu1 suppresses the canonical Wnt signaling pathway. Hence, we performed experiments to study the effect of SHH002‐hu1 on the canonical, rather than the non‐canonical, Wnt signaling pathway.

Primarily, SHH002‐hu1 shows excellent targeting toward Fzd7⁺ NSCLC cells and NSCLC tumor tissue, enabling specific Wnt/β‐catenin signaling inhibition. SHH002‐hu1 significantly inhibits the migration and invasion of A549/H1975 cells in vitro. The activation of the Wnt/β‐catenin pathway facilitates EMT to promote the invasion and metastasis of various tumors. ⁴¹ , ⁴² , ⁴³ To dig deep into the expression of the key factors involved in EMT, we found that SHH002‐hu1 significantly attenuated the EMT of A549/H1975 cells. In addition, SHH002‐hu1 exhibited superior anti‐tumor activities by inhibiting the growth of NSCLC xenografts and suppressing the LN metastasis of NSCLC. The Wnt/β‐catenin pathway activity positively correlated with elevated angiogenesis in various cancers. Here, we also demonstrated that SHH002‐hu1 exhibited antiangiogenic activities against NSCLC in vivo. Figure 8 shows the schematic illustration delineating the possible role of SHH002‐hu1 in NSCLC.

FIGURE 8.

Schematic illustration delineating the role of SHH002‐hu1 in NSCLC. SHH002‐hu1 abolishes the functional activity of Fzd7 in NSCLC by competitive binding Fzd7 extracellular CRD with Wnt ligands

For now, the acquired resistance greatly restricted the clinical application of EGFR/ALK–TKIs, and the low ORRs to anti‐PD‐1/PD‐L1 antibodies narrowed the clinical option of ICIs. It is meaningful to study the correlations of Fzd7 expression with the occurrence of EGFR‐TKIs resistance or no response to PD‐1/PD‐L1 axis inhibitors, and further verify the effectiveness of SHH002‐hu1 toward NSCLC resistance to multiple types of EGFR‐TKIs or failing to respond to anti‐PD‐1/PD‐L1 antibodies. Nevertheless, here, SHH002‐hu1 was demonstrated to exhibit NSCLC targeting behavior and strong anti‐tumor activities when used as a single agent, benefiting from the high affinity to Fzd7 and attenuating the EMT of NSCLC by abating the Wnt/β‐catenin signaling pathway. To sum up, SHH002‐hu1 was indicated as a good candidate for the targeted therapy of NSCLC.

FUNDING INFORMATION

This study was supported by the National Natural Science Foundation of China (NSFC81703401); the Key Clinical Program of Shanghai Municipal Health Commission (20214Y0516); the “Chen Guang” project supported by the Shanghai Municipal Education Commission and Shanghai Education Development Foundation (18CG72); and the Construction Project of Shanghai Key Laboratory of Molecular Imaging (18DZ2260400).

CONFLICT OF INTEREST

The authors declare no conflict of interest.

ETHICS STATEMENT

Approval of the research protocol by an Institutional Reviewer Board: N/A.

Informed Consent: N/A.

Registry and the Registration No. of the study/trial: N/A.

Animal Studies: All animal experiments were performed in accordance with protocols approved by the Animal Ethics Committee of Shanghai University of Medicine and Health Sciences.

Supporting information

Figure S1.

Li K, Mao S, Li X, et al. Frizzled‐7‐targeting antibody (SHH002‐hu1) potently suppresses non–small‐cell lung cancer via Wnt/β‐catenin signaling. Cancer Sci. 2023;114:2109‐2122. doi: 10.1111/cas.15721