Prostate cancer is a common cancer in men and the second leading cause of cancer deaths (1-3). Existing cancer treatment strategies such as surgery, chemotherapy, and radiation therapy (4-6) have been helpful in treating prostate cancer; however, new effective treatments are needed to overcome limitations such as side effects and recurrence rates (7).
Therapeutic Fc-fusion proteins have been successfully used to treat various diseases, including cancer, viral infections, and immune diseases (8-12). The Fc-fusion proteins are composed of an immunoglobulin (Ig) Fc domain linked to another peptide or antigen. Immunoglobulin IgG1 Fc-fusion proteins have emerged as promising therapeutic drugs in immunology (9). In addition, the Fc fusion can be applied to generate dimeric and hexameric structures, likely in a star shape (9, 13-15). Among diverse tissues, PAP highly exists in prostate tissues compared to other organs (16). PAP is a prostate tumor antigen, and the target of the FDA-approved anti-prostate cancer vaccine. In this study, thus, PAP Fc fusion proteins were produced as a prostate cancer vaccine candidate through the fusion of immunoglobulin A (IgA) and immunoglobulin M (IgM) Fc domains. IgA exists primarily as a monomer or dimer, while IgM primarily exists as a pentamer. The polymer structure is thought to increase antibody avidity for the antigen and is required for functions such as complement activation by IgM and epithelial secretion of IgA and IgM (13, 17, 18).
Plants have generally been used to obtain therapeutic recombinant proteins and monoclonal antibodies for diseases such as cancers, viral infections, and immune diseases (19-22). In previous studies, immunotherapeutic proteins have been produced using transgenic and transient expression systems including plant suspension cell cultures (23, 24). One advantage of using the plant expression system for immunotherapeutic proteins such as vaccines and antibodies is economical plant biomass production. In addition, there is no risk of contaminating therapeutic proteins with human pathogens, and proteins can be produced with similar glycosylation patterns as eukaryotic cells (25-27). In this study, plant was utilized to express PAP-IgA Fc, PAP-IgM Fc, PAP-IgA FcK, and PAP-IgM FcK. These proteins were then compared to determine the immunogenic effects of the IgA and IgM Fc domains.
Transgenic tobacco plant lines expressing PAP-IgA Fc, PAP-IgM Fc, PAP-IgA FcK, and PAP-IgM FcK were obtained through
Leaf samples of PAP-IgA Fc, PAP-IgM Fc, PAP-IgA FcK, and PAP-IgM FcK transgenic plant lines were obtained for immunoblot (Fig. 2). Transgenic lines with PAP-IgA Fc, PAP-IgM Fc, PAP-IgA FcK, and PAP-IgM FcK showed approximately 71 kDa (Fig. 2E, F) and 83 kDa (Fig. 2G, H) protein bands, respectively. The positive control PAP was observed at 45 kDa (Fig. 2E, G). In PAP-IgA Fc, PAP-IgM Fc, PAP-IgA FcK, and PAP-IgM FcK protein samples, 45 kDa and >240 kDa protein bands were observed in addition to the 71 and 83 kDa protein bands (Fig. 2E, G). In each of the four protein samples treated with anti-human Fcγ antibody, protein bands were detected at 71 kDa and >240 kDa. However, no 45 kDa protein band was observed in the PAP-IgA Fc, PAP-IgM Fc, PAP-IgA FcK, and PAP-IgM FcK transgenic lines (Fig. 2F, H, anti-IgG Fc). No protein band was observed in non-transgenic plant (Fig. 2E-H).
In SDS-PAGE, transgenic plants expressing PAP-IgA Fc (71 kDa), PAP-IgM Fc (82.8 kDa), PAP-IgA FcK (71.5 kDa), and PAP-IgM FcK (83.3 kDa) showed the expected recombinant fusion protein sizes (Supplementary Fig. 2). The purified bands of PAP-IgA Fc, PAP-IgM Fc, PAP-IgA FcK, and PAP-IgM FcK were clear, showing that the plant-derived recombinant proteins were properly purified. The 50-kDa protein band was observed in PAP-IgA Fc, but not in PAP-IgA FcK (Supplementary Fig. 2A, B). In PAP-IgM Fc, <50 kDa protein was detected, but not in PAP-IgM FcK (Supplementary Fig. 2C, D).
SEC-HPLC was performed to determine the size distribution of the fusion proteins PAP-IgA Fc, PAP-IgM Fc, PAP-IgA FcK, and PAP-IgM FcK. All proteins migrated on HPLC as double or triple peaks between 5 and 8 min (Fig. 3B-E). The three peaks corresponded to >150, 300, and 1,500 kDa similar to the molecular weight (MW) of a monomer, dimer, trimer, and polymer. When IgA and IgM Fc domains were fused to PAP, peaks were observed at 5 min, predicting the MW of more than 1,500 kDa (Fig. 3B-E). On HPLC with PAP-IgA Fc and PAP-IgA FcK, a peak was observed between 5.7 and 7 min. The polymer (P) on HPLC corresponded to a MW greater than 670 kDa. Both PAP-IgA Fc and PAP-IgA FcK formed dimers, while PAP-IgM Fc and PAP-IgM FcK formed polymers. The smaller complexes of PAP-IgA Fc and PAP-IgA FcK appeared as broad peaks in the calculated MW ranging at 250-670 kDa between 5.7 and 7 min.
The EM confirmed structures of PAP-IgA Fc, PAP-IgM Fc, PAP-IgA FcK, and PAP-IgM FcK (Fig. 3F-I). The recombinant proteins showed the complex forms with monomer assembly. The complexes were identified either as a dimer, in which two monomers were assembled, or a polymer, in which three or more monomers were assembled. This different shape may be due to the recombinant protein orientation bound to the chip.
After confirming the molecular size and shape of the PAP-IgA Fc, PAP-IgM Fc, PAP-IgA FcK, and PAP-IgM FcK fusion proteins, their immune responses were tested in BALB/c mice. The mice were injected three times in total, and blood samples were collected for serum IgG response analysis after each vaccination. As shown in Fig. 4A, immune response was not observed in the first bleeding samples in any of the experimental groups. However, in the second bleeding samples, IgG was observed in all the groups except for the mice injected with 1 × PBS. ELISA was performed on the immune serum, and the highest response observed was in the PAP-IgA FcK protein (Fig. 4B). The IgG1 subtype induction in the IgG response indicated that the immune responses were related to Th2 responses. As shown in Fig. 4C, the IgG1 absorbance values were lower than the total IgG response. The IgG1 response was the highest in the PAP-IgA FcK protein.
The spleens were harvested to determine the CD4+ and CD8+ T cellular responses from the mice injected with PAP-IgA Fc, PAP-IgM Fc, PAP-IgA FcK, and PAP-IgM FcK proteins (Fig. 4D-G). The CD4+ and CD8+ T cell number and activation state were measured in the splenocyte. The CD4+ and CD8+ T cells were assessed to determine the CD4+ and CD8+ T cell induction activity of PAP-IgA Fc, PAP-IgM Fc, PAP-IgA FcK, and PAP-IgM FcK proteins. This was conducted using the gating strategy shown in Supplementary Fig. 3. In the mean fluorescence intensity (MFI) values of both CD4+ and CD8+ T cells, PAP-IgA FcK (400 MFI) and PAP-IgM FcK (458 MFI) proteins had higher values compared to PAP-IgA Fc (342 MFI) and PAP-IgM Fc (306 MFI) proteins, respectively (Fig. 4D-G). Notably, for CD4+ and CD8+ T cells, the IgM Fc type with KDEL had the highest MFI values among the others (Fig. 4D-G).
This study demonstrated that the plant-derived PAP-Fc fusion vaccine protein had a potential of preventive and therapeutic agents for prostate cancer. PAP-IgA Fc, PAP-IgM Fc, PAP-IgA FcK, and PAP-IgM FcK were glycoproteins with PAP fused to the human immunoglobulin IgA Fc and IgM Fc domains. The ER signal peptide was fused to the N-terminus of PAP, and the ER retention signal motif was fused to the C-terminus of the IgA Fc and IgM Fc domains. The PAP-IgA Fc, PAP-IgM Fc, PAP-IgA FcK, and PAP-IgM FcK genes were expressed under the control of the enhanced CaMV 35S promoter (28, 29). This study hypothesized that the expression and immunogenicity of PAP-IgA Fc, PAP-IgM Fc, PAP-IgA FcK, and PAP-IgM FcK proteins would vary depending on the immunoglobulin (Ig) Fc domain type in transgenic plants. Thus, the goal of the current research was to determine the effect of the IgA and IgM Fc domain types on the expression of the recombinant protein as well as the efficacy of their immune responses. This would then determine their potential use in a recombinant Fc fusion vaccine expressed in plants.
Both PCR and RT-PCR analyses demonstrated that the PAP-IgA Fc, PAP-IgM Fc, PAP-IgA FcK, and PAP-IgM FcK transgenes were inserted and expressed in transgenic plant DNA genomes. Western blot revealed the PAP-Fc fusion protein expression in plants. Expression of both PAP-IgA FcK and PAP-IgM FcK with ER retention signal peptide was higher compared to PAP-IgA Fc and PAP-IgM Fc. PAP-IgA Fc (3 mg/kg), PAP-IgM Fc (5.6 mg/kg), PAP-IgA FcK (3.5 mg/kg), and PAP-IgM FcK (6.2 mg/kg) proteins were obtained from fresh leaves of transgenic plants. These results suggest that the ER retention signal motif enhanced production level of the KDEL tagged proteins. This high expression of PAP-IgA FcK and PAP-IgM FcK can be explained by the localization of ER (22, 29-31).
Typically, dimeric or trimeric structures are formed for IgA, and pentameric or hexameric structures are formed for IgM (32, 33). In addition, the Fc domain of Ig activates the effector function by binding to FcRs or mediating other immune responses such as the complement system (32, 34). The Fc fusion protein can significantly induce the immune response for cancer. Therefore, in this study, IgA and IgM Fc domains were fused to PAP to generate vaccine proteins with polymeric structures for prostate cancers in transgenic plants. In the SEC-HPLC analysis, polymer and monomer-sized peaks were identified in all of the plant-derived proteins, and dimer or trimer-sized peaks were identified in the PAP-IgA Fc and PAP-IgA FcK proteins. Likewise, Bio-TEM analysis identified dimer and polymer shapes of proteins. These results suggest that proteins fused with IgA Fc and IgM Fc domains can form their polymeric shapes in plant cells.
In the mice experiment, plant-based recombinant PAP-Fc fusion proteins induced PAP-specific antibody responses in mice. The PAP-IgA FcK recombinant protein showed a much higher response in terms of total IgG and IgG1 compared to the other recombinant proteins, suggesting a Th2 bias. Notably, the IgA protein form showed a higher IgG response in the PAP-IgA FcK protein, which had the ER retention signal peptide KDEL.
Recent studies have shown that T-cell responses are a crucial component of an effective therapeutic cancer vaccine (35, 36). The PAP-IgA Fc, PAP-IgM Fc, PAP-IgA FcK, and PAP-IgM FcK proteins induced CD4+ and CD8+ T cell responses in mice. The activation of CD4+ and CD8+ was measured based on the MFI value of CD69. PAP-IgA Fc, PAP-IgM Fc, PAP-IgA FcK, and PAP-IgM FcK vaccine proteins induced higher activity of CD4+ T cells. CD4+ T cells promote anti-cancer immunity through various mechanisms such as T cell activation, T cell homing, antigen presenting enhancement, effector function, and co-stimulation (37). These results were attributed to have immunogenic effects from dimeric and pentameric formation in PAP-IgA Fc, PAP-IgM Fc, PAP-IgA FcK, and PAP-IgM FcK. The PAP-IgA FcK and PAP-IgM FcK proteins were more efficient at inducing CD4+ T cell response compared to the PAP-IgA Fc and PAP-IgM Fc proteins. It is speculated that vaccines may be more effective with high mannose glycan structure modification caused by the ER retention signal peptide KDEL (29-31).
Unlike other microorganisms which do not have subcellular organisms such as endoplasmic reticulum and golgi complex, plants express and assemble PAP polymeric IgA and IgM Fc fusion antigenic proteins increasing immunogenicity with and without adjuvant (38). In addition, plant expression system has several advantages for recombinant protein production in cost, safety, and scalability compared to microorganism system (39).
In this study, the immunotherapeutic vaccine candidate proteins with IgA and IgM Fc fusions for prostate cancer were produced in transgenic plants using a stable expression system, and those PAP-Fc fusion proteins exhibited polymeric shapes to induce immune response for prostate cancer.
See supplementary information.
This research was funded by the National Research Foundation of Korea grant (2021R1F1A1063869) and Basic Science Research Program through the National Research Foundation of Korea grant (2020R1I1A1A01072021).
The authors have no conflicting interests.