Cell lines and cultures

In this study, a variety of prostate, gastric, and colorectal cancer cells were used. The cell lines (LNCAP, C4-2, 22RV1, PC3, DU145, VCAP, NCI-H660, HT29, MKN45) were obtained from Yanshun Biotechnology Company (Shenzhen, China), Meisen CTCC Company (Zhejiang, China), the American Type Culture Collection (ATCC, Manassas, VA), Cell Bank of Chinese Academy of Sciences (Shanghai, China), and Fuheng Biology (Shanghai, China). LNCAP, C4-2, 22RV1, and HT29 cells were cultured in RPMI 1640 medium (Gibco, USA) supplemented with 10% fetal bovine serum (FBS), 2 mM L-glutamine, and 1% penicillin/streptomycin. DU145, VCAP, and MKN45 cells were maintained in high-glucose Dulbecco's Modified Eagle's Medium (DMEM) supplemented with 10% FBS, 2 mM L-glutamine, and 1% penicillin/streptomycin. The PC3 cell line was cultured in Ham's F-12 K medium (Procell) with 10% FBS and 1% penicillin/streptomycin. The NCI-H660 cell line was cultured in RPMI 1640 medium (Gibco, USA) with 10 nM beta-estradiol, 10 nM hydrocortisone, 1 × ITS (Insulin-Transferrin-Selenium), 5%FBS, and 1% penicillin/streptomycin. All cell lines were incubated at 37 °C in a humidified atmosphere containing 5% CO2.

Sample sources

In the study, the samples included: 4 clinical samples of NEPC collected from The Fifth Affiliated Hospital, Southern Medical University. Additionally, patient-derived xenograft (PDX) samples were obtained from Prof. Yuanqiao He. All patients provided informed consent to participate in the research project. The collection and research use of patient tissues were conducted with the consent and under the supervision of the ethics committee of The Fifth Affiliated Hospital, Southern Medical University. Tissue microarrays of normal human visceral organs, benign prostatic hyperplasia, and prostatic adenocarcinoma were all from a biobank Zhongke Guanghua Company (China).

Immunohistochemistry staining

To analyze and evaluate the expression level of CEACAM5 in prostate cancer tissue, immunohistochemical staining for CEACAM5 was performed on clinical samples of NEPC, PDX samples, CDX samples, and tissue microarrays of normal human visceral organs, benign prostatic hyperplasia, and prostatic adenocarcinoma. The sections were stained with a rabbit monoclonal Anti-CEACAM-5/CD66e Antibody (11077-R327, Sinobiological Inc. China). CEACAM5 detection was carried out using a biotin-conjugated goat anti-rabbit IgG secondary antibody and an ABC kit (Vector Laboratories, USA), followed by colorimetric detection with diaminobenzidine (DAB; Vector Laboratories, USA). The images were captured using a 3DHISTECH™ scanner (Sysmex, UK). In the pathological analysis, five visual fields were selected from each pathological section to calculate the mean positive rate per sample, thereby determining the positivity presented graphically.

Flow cytometry assay

To analyze the cell surface expression of CEACAM5 protein, flow cytometry was employed. The cells were first gently washed twice with precooled PBS, followed by enzymatic digestion using 0.25% EDTA. Subsequently, the cells were fixed with 0.25% freshly prepared PFA for 5 min at room temperature (RT) and then blocked with 3% BSA. After blocking, the Alexa Fluor® 647 conjugated anti-human CEACAM5 antibody (392806, Biolegend, USA) was added to the cells at a dilution of 1:100 and incubated on ice for 1 h. Finally, the cells were washed three times with PBS before being analyzed using a Beckman Coulter flow cytometer and FlowJo software.

Phage library screening for CEACAM5-targeted nanobody

The nanobody screening was conducted as previously described [29]. Briefly, the isolation of CEACAM5-targeted targeted nanobodies was performed with a naive phage nanobody library screening (1 × 109 diversity) (KTSM-CND002, Shenzhen KangTi Life Technology Co., Ltd. China,). The Naïve nanobody phage library was established on PBMC RNA from over 100 alpacas Alpaca (Lama pacos) and three rounds of bio-panning were performed in the immune tubes coated with 40 μg CEACAM5-Fc protein (11077-H02H, Sinobiological Inc. China). This was followed by an incubation with 3% BSA-PBS solution for 1 h at RT. The nanobody phage library was then incubated in the tube for 1 h at RT. Unbound phage clones were removed by washing with PBST (PBS + 0.1% Tween20). Bound phages were eluted using a trypsin solution (0.25 mg/ml) and neutralized with AEBSF (4 mg/ml). The eluted phage clones were amplified, rescued with M13 helper phages in E. coli TG1 cells, and precipitated using PEG–NaCl (a solution of 20% PEG 800 and 2.5 M NaCl), after which they were resuspended in PBS. The phage library was collected for titration and subsequent screening rounds. One round subtractive screening with Fc fragment prior to third round selection was conducted to remove the potential Fc binders. After three rounds of biopanning, 96 phage clones were randomly selected and amplified for phage ELISA analysis.

Briefly, microtiter plates were coated with 2 μg/ml of purified CEACAM5-Fc or Fc fragment and incubated overnight at 4 °C. After blocking with a 3% BSA solution for 1 h at RT, phage clones were added and the plates were incubated for an additional hour at RT. The plates were then washed three times with PBST to remove unbound phages. Subsequently, an HRP-conjugated anti-M13 monoclonal antibody (Sino Biological, Beijing, China) was added and the plates were incubated for 1 h at RT. After another round of washing with PBST, the plates were developed with the TMB peroxidase substrate. The reaction was halted with 1 M HCl, and the absorbance was measured at 450 nm using an automated microplate reader (LabServ K3 TOUCH, Thermo Fisher Scientific, USA). Clones exhibiting an absorbance value more than threefold higher with CEACAM5-Fc compared to Fc fragment were considered positive. Based on the ELISA results, 8 positive clones were identified, and after sequencing, two unique sequences were obtained.

Protein purification

The various nanobodies were purified as described previously [29, 30]. Briefly, the recombinant plasmids pET-22B-B12 were transformed into BL21 (DE3) and then the bacterial clones were incubated at 37 °C and 200 rpm until reaching 0.6 of OD600 value, followed by induction with 1.0 mM IPTG at 30 °C and 200 rpm overnight. The cultures were pelleted with 8000 g for 15 min at 4 °C. Cell pellets were resuspended into PBS containing polymyxin B (10000unit/ml) to release B12 nanobodies from bacterium periplasm through incubation at 37 °C and 220 rpm shake for 2 h.

The lysate was spun down for 45 min at 12,000 × g, and the supernatants were loaded on a gravity column with 1 mL Ni–NTA agarose resin (Qiagen, Germany). The protein-bound resin was washed with 50 mL Wash Buffer I (300 mM NaCl, 50 mM NaH2PO4, 20 mM imidazole, pH 8.0, 1 mM PMSF) and 50 mL Wash Buffer II (300 mM NaCl, 50 mM NaH2PO4, 40 mM imidazole, pH 8.0, 1 mM PMSF) and then eluted with 25 ml Elution Buffer (300 mM NaCl, 50 mM NaH2PO4, 250 mM imidazole, pH 8.0, 1 mM PMSF). Finally, the eluate was then fractionated using a Superdex-150 gel filtration column with an AKTA Pure System (GE Healthcare Life Sciences, USA) in 1 × PBS. The purified protein was identified by SDS-PAGE, quickly frozen in liquid nitrogen, and stored at −80 °C until use. Control Nb, B12-PE38, Con-PE38, B12-PE38 mut and ConNb-PE38 mut were induced using the same procedures as B12 nanobody, but all these proteins were purified as cytoplasmic proteins for which cell pellets were dissolved in lysis buffer (300 mM NaCl, 50 mM NaH2PO4, 10 mM imidazole, pH 8.0, 1 mM PMSF) and crushed through three cycles of low-temperature, high-pressure homogenization. The subsequent protein purification procedures are the exact same as B12 nanobodies. The purified proteins were analyzed and identified by western blot with 6 × His tag, HA tag or anti-VHH antibodies.

ELISA assay

To verify the binding activity of the B12 nanobody to CEACAM5, ELISA was performed using the following procedure: 5 μg/ml of human CEACAM5 (CE5-H5226, ACRO biosystems, China) was coated onto 96-well microplates and incubated at 4 °C overnight. Unbound protein was removed by washing with PBS before incubation with a 3% BSA-PBS solution for 1 h at RT. The nanobodies were diluted from 8 to 1 nM with 1 × PBST and incubated in the plates at RT for 1 h. The plates were then washed three times with 1 × PBST, and the nanobodies were detected using an HRP-conjugated anti-HA tag antibody (SinoBiological, China) and TMB peroxidase substrate (BioLegend, San Diego, CA, USA). The absorbance was measured at 450 nm.

SPR assay

The experiment was used to validate the direct interaction between B12 nanobody and in antigen CEACAM5 to calculate the equilibrium constants of the two proteins. CEACAM5-his protein was purchased from ACRO biosystems company (CE5-H5226). CEACAM5 was immobilized on a chip, and different concentrations of nanobody B12 were sequentially added to analyze the binding affinity. The reaction signals were recorded within 360 s, to draw kinetic curves, and calculate each corresponding parameter.

Protein labeling with IR800

IR800 dye-labeled nanobodies were used for cell ELISA and in vivo imaging. Briefly, B12 or control nanobodies were diluted to a concentration of 1 mg/mL in PBS (pH 6.5). IR800-maleimide was added and the mixture was incubated for 2 h at RT in the dark. Unconjugated dye was removed using a 10 kDa molecular weight cutoff spin desalting column (Millipore). The concentrations of the nanobodies were measured using a BCA protein assay kit (Pierce). The proteins (B12-PE38, ConNb-PE38, B12-PE38 mut and ConNb-PE38 mu) were also labeled using the same procedure.

Cell ELISA

To analyze the binding activity of B12 nanobodies to cancer cell lines expressing CEACAM5, Cell ELISA was performed as follows: Cancer cell lines (PC3, HT29) were cultured in 96-well plates at a density of 5 × 104 cells/well overnight. The cells were fixed with 4% paraformaldehyde for 10 min and then washed once with PBST, followed by incubation with a 3% BSA for 1 h at RT. The nanobodies, previously labeled with Dye IR800 as described above, were diluted from 250 to 0 nM with 1 × PBST and incubated with the plates at RT for 1 h. After washing the plates three times with PBST, fluorescence intensity was measured using a 784 nm laser channel on a Sapphire Capture system (Sapphire, USA).

In vivo imaging

Once tumors had grown to approximately 500 mm3, the PC3 tumor-bearing mice were randomly assigned to two groups (n = 3) and injected intravenously 80 μg B12-IR800 and ConNb-IR800 respectively. At various time points post-injection, fluorescence imaging was conducted using the IVIS Spectrum system (PerkinElmer, USA). Upon completion of the whole-body image, the animals were euthanized, and key organs (heart, liver, spleen, lungs, kidneys) along with the tumors were extracted for ex vivo fluorescence analysis. In vivo imaging with various proteins, including B12-PE38 mut-IR800 and ConNb-PE38 mut-IR800, was conducted following identical procedures.

In vitro internalization and uptake assay of immunotoxins

B12-PE38 was labeled with Cy5 Dye at a 2:1 ratio as previously described. For uptake assays, 1 × 105 PC3 cells were plated in poly-lysine-coated 24-well plates and exposed to varying concentrations of B12-PE38 over different durations to assess drug uptake efficiency. After washing, cells were fixed with 4% paraformaldehyde for 10 min, and fluorescence was quantified using an inverted microscope and Image J software with a minimum of three replicates. Flow cytometry was also used to evaluate B12-PE38 uptake in PC3 cells, with cells collected, washed, fixed, and resuspended in PBS for APC channel fluorescence detection.

In vitro cell viability assay

To verify the targeted cytotoxicity of B12-PE38 on cancer cells expressing CEACAM5 protein, CCK8 (Cell Counting Kit-8) assay was conducted to determine the cytotoxicity and the half-maximal inhibitory concentration (IC50) values. PC3, MKN45, and HT29 cells were seeded into 96-well plates at a density of 5 × 103 cells per 100 μL of culture medium per well. Following overnight incubation in a humidified atmosphere at 37 °C with 5% CO2, the cells were briefly washed once with PBS. Subsequently, different concentrations (0, 15.625, 31.25, 62.5, 125, 250, 500, and 1000 nM in 100 μL of culture medium, n = 3) of the purified B12, B12-PE38, Con-PE38, or B12-PE38 mut were introduced into the respective wells. The plates were then incubated for 3 days at 37 °C in a humidified incubator. Afterward, 10 μL of CCK-8 solution (Abcam, China) was carefully added to each well, taking care to avoid bubble formation. The plates were incubated for an additional hour at 37 °C. The absorbance was measured at 450 nm using an automated microplate reader (LabServ K3 TOUCH, Thermo Fisher Scientific, USA) following a brief shaking step. Data analysis for cell viability and IC50 values was performed using GraphPad Prism software.

EDU assay

The EDU assay kit was purchased from Biosharp Company (China). A total of 1 × 105 cells were seeded in a 96-well plate and cultured overnight. The cells were then treated with complete medium containing a gradient of concentrations of B12-PE38, or with PBS as a baseline control, as well as with B12 and ConNb-PE38 as negative controls, for approximately 12 h. Following this treatment, the EDU solution was diluted in complete medium at a ratio of 1:1000 and incubated with the cells for 2 h to label DNA synthesis. After aspirating the medium and washing the cells twice with PBS, they were fixed with a 4% paraformaldehyde solution for 10 min to preserve cellular morphology. Subsequently, EDU-positive cells were detected using the protocol recommended by the kit, and the nuclei were counterstained with Hoechst 33342 to visualize cell nuclei. Finally, at least three replicates of fluorescence images were captured using an inverted microscope, and the data were analyzed to calculate the differences in cell proliferation rates.

Mouse xenograft models and treatments

All experiments on animals in the present study were done following the approved protocol by Institutional Animal Care and Use Committee (IACUC) of the Shenzhen People’s hospital and were carried out in accordance with relevant institutional and national guidelines and regulations. Nude mice and NCG mice at 6–8 weeks old were purchased from Gempharmatech (Guangzhou, China) and were maintained under pathogen-free conditions in the animal center of the Shenzhen People’s hospital. Mice were euthanized when showed obvious signs of discomfort or when maximal tumor size reached 2000 mm3.

To establish the subcutaneous tumor models, 100 μl of PBS containing approximately 5 × 106 PC3 cells, 2 × 106 HT29 cells was injected into the right lower back of the mice. Tumor size was measured with vernier calipers and calculated using the following formula: (length × width × width)/2.

To determine the antitumor efficacy of B12-PE38, PC3 tumor-bearing mice were first selected for antitumor study. In brief, tumor-bearing mice were randomly divided into control group (PBS), control group (B12), low-dose group (0.4 mg/kg B12-PE38), and high-dose group (0.6 mg/kg B12-PE38) when the tumor size reached approximately 150 mm3 (n = 5). Drugs were administered intravenously as indicated schedule in Fig. 5e every other day for six injections. During treatment, tumor size and body weight in mice were monitored. For survival study, mice were sacrificed when tumor size reached 2000 mm3. To evaluate the therapeutic effect of B12-PE38 in colorectal cancers, HT29 tumor-bearing mice were randomly divided into 2 groups and treated with PBS, and B12-PE38 (0.6 mg/kg) (n = 5). The mice were treated as scheduled above for six injections. During treatment, tumor size and body weight were recorded. After treatment, the major organs including heart, liver, spleen, lung, kidney, and tumor were collected, and frozen sections were prepared and analyzed by H&E staining, Ki67 and TUNEL staining.

For combination therapy, tumor-bearing mice (PC3) received totally 8 does of immunotoxin B12-PE38 formulated at 0.4 mg/kg and 4 does of docetaxel (DTX) at 2.5 mg/kg, as demonstrated in administration schedule presented in Fig. 6a. Meanwhile, mice receiving B12-PE38 or docetaxel alone were used positive control groups, but those receiving PBS were used a negative control group. Tumor growth and volume were monitored every 2 days until tumor size reached 2000 mm3.

Patient derived CRPC xenograft model and treatment

A piece of fresh PDX tissue (2nd passage) was generously provided by Nanchang University and confirmed to express CEACAM5 through IHC. The sample was originated from a patient diagnosed with PCa. Informed consent was obtained from the patient, and all procedures involving human samples were approved by the Medical Ethics Committee of the Shenzhen People’s Hospital, Nanfang University and Nanchang University.

Upon receipt, the PDX tissues was swiftly dissected into 3 × 3x3 mm fragments on ice and implanted subcutaneously into the right forelimb of NCG mice (NOD/ShiLtJGpt-Prkdcem26Cd52Il2rgem26Cd22/Gpt) to propagate the tumors. Once the tumor volume reached 800 to 1000 mm3, it was harvested, dissected into smaller fragments, and subcutaneously inoculated into the right forelimb of new NCG mice. When these tumors reached a volume of 100 mm3, the mice were randomly assigned to four treatment groups: PBS, docetaxel (DTX), B12-PE38, and a combination of B12-PE38 and DTX. The combined treatment group received a total of 8 doses of immunotoxin B12-PE38 at a dosage of 0.4 mg/kg and 4 doses of docetaxel at a dosage of 2.5 mg/kg, as detailed in the administration schedule depicted in Fig. 6d. Tumor growth and volume were monitored every 2 days until the tumor size reached 2000 mm3 for the purpose of anti-tumor efficacy evaluation and survival assessment, with 5 mice per group.

Mouse bone metastasis xenograft models and treatments

To establish intra-tibial bone xenograft models, PC3 cells (5 × 106) were injected directly into the tibia of mice, and the models were monitored by X-ray every 7 days until osteolytic changes in the tibial bone were detected, thereby confirming the successful establishment of the model. Once osteolytic lesions in the tibia were observed based on X-ray indications, the mice were randomly assigned to three treatment groups: PBS, B12-PE38 at 0.3 mg/kg, and B12-PE38 at 0.6 mg/kg. A total of 6 doses of the drugs were administered (Fig. 7a). During the treatment period, the intra-tibial xenografts were monitored by X-ray every 7 days. In the survival analysis, the visible signs of bone fractures, as monitored by X-ray and a weight decrease of more than 20%, were considered an endpoint for the experiment.

Statistics

All statistical analyses were performed using GraphPad 9.0 (GraphPad Software Inc., United States). Numerical data were presented as means ± standard error of mean (SEM). For more than two groups of continuous variables, one-way Anova or two-way Anova were performed to determine the difference significance among groups. Differences between two groups of variables were compared and confirmed by two-tailed Student’s t-test. Survival curves were depicted using Kaplan–Meier’s method and compared by the log-rank test. A p < 0.05 was considered as a statistically significance.