In cancer therapy, inhibition of the VEGF/VEGF receptors signal pathway has
been shown to suppress angiogenesis in lots of models, including genetic models of
cancer, leading to clinical development of a lot of VEGF inhibitors. Consequently, a
novel chimeric protein is confirmed by fusing the receptor binding domain of human
VEGFA165 to the Fc portion of the human IgG1 as an affibody, which is designed as
antagonist that blocks the physiological interaction between VEGF ligands and its
receptors, VEGFR-1 and VEGFR-2. It has been reported that Ig Fc portion could help
maintain the tertiary structure of their CH-2 domain after glycosylation of IgG
molecule [97]. Besides, glycosylation of the IgG molecule has been shown to
significantly affect the resulting the ability of the antibody to participate in ADCC and
complement system [102].
The stimulation of endothelial cells by growth factors, such as VEGF, is required
for the process of angiogenesis [103]. In previous lab results, it has been shown that
RBDV-Ig Fc could bind to human VEGFRs, VEGFR-1 and VEGFR-2, and inhibit
angiogenesis through VEGF and its receptors interaction in HUVE cells in vitro. This
can be strongly confirmed since the sequence alignment of receptor binding domain
of VEGF (RBDV) of human and mouse share a 92.7% similarity. As expected,
RBDV-Ig Fc fusion proteins can bind to immobilized mouse VEGF receptors, not
only VEGFR-1 but also VEGFR-2, in ELISA. We, therefore, demonstrated this
receptor binding domain containing fusion protein could also bind to mice cells which
express VEGF receptors on cell surface. According to the results, RBDV-Ig Fc could
bind both endothelial cells and the melanoma cell, B16/F10, of which was reported
VEGFR expression [101]. We also needed to confirm whether RBDV-Ig Fc could
inhibit signal transduction cascades in mouse endothelial cells, SVEC4-10. Clearly, in
the course of this study, we discovered that RBDV-Ig Fc consistently and significantly
inhibited the proliferation of SVEC4-10 cells, and has also suppressed the
proliferation when compared with their natural interactions with VEGF alone. This
can be supported by the research of the carboxyl terminal of VEGF. The heparin
binding region (amino acid 110-165) has been strongly indicated and that the heparin,
through simultaneous binding to VEGF and its receptors increases in signal amplitude
and duration [93]. Here our results have shown that the RBDV-Ig Fc inhibits VEGF
receptor activation and mitogenesis in mouse endothelial cell no matter whether
VEGF has been presented or not. Our results suggest that the mechanism of
endothelial cell growth inhibition mediated by the chimeric proteins may due to the
direct blocking of VEGF/VEGFR-2 interaction, resulting in inhibition of
VEGF-induced VEGFR-2 activation.
It was proposed that the disturbance of VEGFR-1 activity results in vascular
malformation [60], though the role of VEGFR-1 in angiogenesis is still the object of
debate. As we expected, VEGF-driven, capillary-like tube formation in SVEC4-10
cells was strikingly decreased by RBDV-Ig Fc. Morphological changes of SVEC4-10
cells, induced by VEGF, were inhibited by RBDV-Ig Fc. Moreover, the migration
ability of SVEC4-10 cells, which was induced by VEGF, was partially suppressed by
RBDV-Ig Fc. Base on the above results, we confirm that RBDV-Ig Fc could inhibit
the signal transduction through VEGF and VEGFR-1 in vitro. According to the results
above, RBDV-Ig Fc could block the angiogenesis signal transduction through
competing against VEGF.
Several strategies have previously been reported to function as VEGF binding
antagonists, including anti-VEGF antibody [81, 82], anti-VEGFR-2 receptor antibody
[86], RNA-based aptamers [104], and various peptides [90-92]. Compared with the
above molecules, RBDV-IgG1 Fc possesses some theoretical advantages, because it
can target both VEGF main receptors, VEGFR-1 and VEGFR-2. Furthermore,
coupling of the receptor binding domain of VEGF sequence to human IgG Fc region
sufficiently blocks VEGF receptors and antagonizes VEGF activity. There are several
theoretical advantages for targeting VEGFR receptors on endothelial cells as cancer
therapeutics. First, VEGFR-2 is expressed exclusively on proliferating endothelial
cells at tumor sites [105]; therefore, antagonists against the receptor may offer higher
specificity compared to other agents. Moreover, tumor vessel endothelial cells are in
direct contact with the blood, and they have greater accessibility compared to
antibodies against markers expressed on individual tumor cells.
With in situ injection of RBDV-Ig Fc, B16/F10 tumor growth was suppressed
and was significantly smaller than control groups, PBS and IgG1 Fc; however, the
proliferation of B16/F10 was not inhibited by RBDV-Ig Fc in vitro. We speculated
that the proliferation of B16/F10 was not only driven by VEGF signal transduction,
but also other growth factors in signal pathway; for example the hepatocyte growth
factor (HGF), scatter factor (SF), and thrombin [106, 107]. Besides, blockade of
angiogenesis which was activated by VEGFR-1 and VEGFR-2 signaling was
necessary to efficiently inhibit B16/F10 melanoma growth and metastasis [108].
Hence, RBDV-Ig Fc could inhibit tumor growth of B16/F10 through the suppressing
of the angiogenesis and proliferation of endothelial cells in vivo.
Tumor sections were obtained and stained with H&E, we discovered that tumor
associated blood vessels and the area surrounding the tumor were disrupted (Figure
18C) with RBDV-Ig Fc treatment. Moreover, the tumor cells were also damaged
randomly within the IgG1 Fc group (Figure 18B), which did not show any destructive
effects in negative group (Figure 18A). IgG-like fusion proteins may prove to be
better choices over smaller molecule substances or fragments such as peptides for
other in vivo applications by providing the Fc domain that not only confers long
pharmacokinetic half-life [97] but also supports secondary immune functions, such as
antibody dependent cellular cytotoxicity (ADCC) and complement-dependent
cytotoxicity (CDC). In a previous study, we have shown that RBDV-IgG1 Fc can
induce cell-mediated cell cytotoxicity by NK cells (Appendix2, Figure2). Tumor cell
killing by ADCC is triggered by the interaction between the Fc region of an antibody
bound to a tumor cell, and the Fcγ receptors on immune effector cells, such as
neutrophils, macrophages and natural killer (NK) cells. Among these receptors,
FcγRIIIa is a predominant FcγR on NK cells [109, 110]. CDC is initiated by
complement component C1q binding to the Fc region of IgG, which is bound to the
surface of a tumor cell. Of these two isotypes, the IgG1 isotype is widely accepted as
the most effective at recruiting the immune system. This is because the Fc regions of
IgG1 interact strongly with all types of FγR on lymphoid and myeloid effectors and
are strong complement activators [111, 112]. Because the damage of tumor cells was
in cytoplasm, and there were not a lot of immune cells in the tumor region, we believe
that the Ig Fc region of chimeric protein can damage tumor cells and endothelial cells
by CDC.
After confirming the suppressive ability of RBDV-Ig Fc to tumor growth with in
situ treatment, we treated mice bearing B16/F10 with i.v. injection of RBDV-Ig Fc
twice every five days. Tumor growth was effectively suppressed with RBDV-Ig Fc
treatment and showed significant difference from other groups in 4 days after
injection. The damage of the tumor region was similar to the RBDV-Ig Fc treatment
in situ (data not shown). It is strongly believed that RBDV-Ig Fc not only could bind
VEGF receptors in vitro, but also serve a function of targeting VEGF receptors in vivo.
Heart, lung, spleen, liver, and kidney were also separated from mice. We discovered
that RBDV-Ig Fc caused physiological damage in kidney (Figure 21C), spleen
(Figure 22C), and liver (Figure 23C). The organ sections of the mice with IgG1 Fc
treatment also showed a bit of break (Figure 21B, 22B, 23B). Although RBDV-Ig Fc
caused physiological damage to mice, it indeed extended the survival rate of the mice
with treatment of RBDV-Ig Fc (Figure20). According to the histological results, we
believe that RBDV, as a targeting domain of RBDV-Ig Fc, could bind cells which
expressed VEGF receptors in vivo; in addition, the systemic immune response was
enhanced by the IgG1 Fc portion of the chimeric protein resulted in organ damage.
In summary, an effective strategy was demonstrated in the engineering of
receptor binding domain of ligands against specific targeted cells which was fused
with an appropriate Fc portion of IgG for their therapeutic capacity. Also, this
chimeric protein, RBDV-Ig Fc, was effective in antiangiogenesis in vitro and
suppressing B16/F10 tumor growth in C57/BL6 mouse model. It is believed that
RBDV-Ig Fc can be used in the treatment of the growth of human cancer cells, which
was dependent on angiogenesis. The concept of using a receptor binding domain of
growth factor conjugated with Fc portion of IgG might be a useful strategy for cancer
therapy.
Figure 1. Role of the VEGF receptors. VEGFR-1 and VEGFR-2 are expressed in the cell surface of most blood ECs. Instead, VEGFR-3 is largely restricted to lymphatic EC. VEGF-A binds both VEGFR-1 and VEGFR-2. In contrast, PLGF and VEGFB interact only with VEGFR-1. VEGFC and VEGFD bind VEGFR-2 and VEGFR-3. There is much evidence that VEGFR-2 is the major mediator of EC mitogenesis and survival, as well as angiogenesis and microvascular permeability.
VEGFR-1 has an established signaling role in mediating monocyte chemotaxis. Also, in hematopoietic stem cells (HSC) or leukemic cells, both VEGFR-1 and VEGFR-2 may mediate a chemotactic and a survival signal. R1, VEGFR-1, R2, VEGFR-2; R3, VEGFR-3. (Olsson, A.K. et al. Nat. Rev. Molecular Cell Biology 2006)
Figure 2. Sequence alignment of human and mouse receptor binding domain of VEGF. The amino acid sequence of mouse VEGF was obtained from NCBI protein database (NCBI No: NP_001020421). Two sequences were compared to find out the mouse receptor binding domain of VEGF (mRBDV), and the highest similarity was obtained when RBDV compared with 179-288 amino acids of mouse VEGF. There was a 92.7% similarity between RBDV and mRBDV.
Ig G1 F c
Bam HI Xho I ApaI HindIII
Ig G1 F c
Bam HI Xho I ApaI HindIII
Ig G1 F c
Bam HI Xho I ApaI HindIII
Figure 3. Gene construction of the chimeric proteins. The receptor binding domain of the human vascular endothelial growth factor (VEGF) was fused to the N terminal of the Fc fragment of IgG1, followed by the 6X his tag for protein purification. LS means the leader sequence; RBDV, the receptor binding domain of vascular endothelial growth factor; IgG1 Fc is the constant region of the Fc domain of human immunoglobulin G1. Human IgG1 Fc molecule was also created as a control.
Figure 4. Restriction enzyme digestion of the pAAV-MCS/IgG1 Fc and pAAV-MCS/RBDV-IgG1 Fc. In lanes 1 and 4: uncut plasmids, lane 1 is pAAV-MCS/RBDV-IgG1 Fc, and lane4 is pAAV-MCS/IgG1 Fc. In lanes 2 and 5:
BamHI and HindIII digestion and the expected fragments of 1080 and 855 bp are
obtained, respectively. In lanes 3 and 6: BamHI and XhoI digestion and the expected fragments of 340 and 753 bp in pAAV-MCS/RBDV-IgG1 Fc plasmid, 120 and 753 bp in pAAV-MCS/IgG1 Fc plasmid (arrow indicated) can be observed. M means 100 bp ladder marker.(A)
(B)
(C)
Figure 5. The fluorescence expression in HEK-293 cell. HEK-293 cells were transfected with pAAV-MCS-hrGFP vector using calcium phosphate based transfection methods. (A) Light microscope picture of hrGFP-transfected HEK 293 cells, 100X magnification. (B) Fluorescence microscope photograph of the hrGFP-transfected HEK-293 cell, 100X magnification. (C) Histograms of the hrGFP-transfected cells from 10,000 events measured with flow cytometer. The transfection efficiency in over 70%.
Protein G Agarose
SephadexTM G-25 HisTrapTM
HP
Supernatants
High purity
Protein G Agarose
SephadexTM G-25 HisTrapTM
HP
Supernatants
High purity
Figure 6. Flow chart of procedure for the purification of chimeric proteins. The culture supernatants of HEK-293T cells were harvested after transfection. The resulting supernatants were first applied into protein G affinity chromatography.
Eluted fractions of Protein G were successively loaded into a nickel-charged HisTrap affinity chromatography and were desalted by using sephadex G-25.
Figure 7. SDS-PAGE analysis of recombinatant proteins. IgG1 Fc and RBDV-IgG1 Fc proteins were expressed and fisrt purified by protein G affinity chromatography, followed by HisTrap affinity chromatography and desalted by using sephadex G-25.Purified proteins were resolved by 10% SDS-PAGE under reducing condictions and stained with Coomassie blue. The molecular markers in kilodaltons (kDa) are shown on the left.
Figure 8. Characterization of purified chimeric proteins. Plasmid pAAV- MCS/IgG1 Fc or pAAV-MCS/RBDV-IgG1 Fc, driven by a human cytomegalovirus (CMV) immediate-early promoter, was transfected into human kidney epithelial (HEK) 293 cells. When the color of cultured medium changed to orange, cultured medium of transfectants were collected, run on reducing 10% SDS-polyacryamide gel electrophoresis, and then analyzed by Western blotting withAs for the purity of chimeric proteins, SDS-PAGE was further transferred onto NC membrane, analyzed by Wesern blotting with the antibody to human IgG (A) and to poly His (B).
Figure 9. Mouse VEGF Receptor 1 binding activities of purified RBDV IgG1 Fc.
The binding activity of the purified RBDV-IgG1 Fc to immobilized mouse VEGFR-1.
96-well plates was precoated with mouse VEGFR1- Fc (0.4μg/well in 100μl) at room temperature overnight. Different quantities of purified proteins with biotin labeled were added and incubated at room temperature for 1 hr, followed by incubation with HRP conjugate streptavidin for an additional 20 mins. The plates were then incubated with a peroxidase TMB substrate, followed by reading of the absorbance at 450 nm.
(*P<0.05). The data performed here are the mean ±SD of three independent experiments (n=6) corresponding to PBS or IgG1 Fc control.
Figure 10. Mouse VEGF Receptor 2 binding activities of purified RBDV IgG1 Fc.
The binding activity of the purified RBDV-IgG1 Fc to immobilized mouse VEGFR-2.
96-well plates was precoated with mouse VEGFR2- Fc (0.4μg/well in 100μl) at room temperature overnight. Different quantities of purified proteins with biotin labeled were added and incubated at room temperature for 1 hr, followed by incubation with HRP conjugate streptavidin for an additional 20 mins. The plates were then incubated with a peroxidase TMB substrate, followed by reading of the absorbance at 450 nm.
(*P<0.05). The data performed here are the mean ±SD of three independent experiments (n=6) corresponding to PBS or IgG1 Fc control.
Figure 11.
(A) MS1 cells
(B)B16/F10 cells
(C)SVEC4-10 cells
Figure 11. Cell surface binding ability. RBDV-IgG1 Fc chimeric protein was detected by flow cytometric analysis using MS1(A), B16/F10(B), and SVEC4-10(C).
The cells were incubated with chimeric proteins at 4 °C for 1 hour, followed by incubation with FITC-labeled goat anti-human IgG antibody for an additional hour (gray area). Cells stained with IgG1-Fc and followed staining with FITC-labeled goat anti-human IgG antibody were used as the negative control (solid line).
(A)
(B)
Figure 12. RBDV-IgG1 Fc inhibits the VEGF-induced proliferation of SVEC4-10 cells in a dose-dependent manner. SVEC4-10 cells were pre-cultured in the presence of 1, 2.5, 5, 10, and 20μg/ml of RBDV-IgG1 Fc(A) or IgG1 Fc(B) (In lanes 1, 2, 3, 4 and 5, respectively). In another condition, cells were pre-cultured in the presence of 2.5, 5, 10 and 20 μg/ml of RBDV-IgG1 Fc(A) or IgG1 Fc(B) before stimulation with 10ng/ml VEGF (In lanes 7, 8, 9, and 10, respectively) for 1 hour at room temperature. Cell proliferation was then measured with MTS proliferation assay as described in the “Materials and Methods”. The data shown as mean of proliferation inhibition percentages ±SD obtained from three independent experiments (n=9), * means p<0.05 when compared with the negative group. The VEGF (In lane 6, 10 ng/ml) stimulated a significant increase in endothelial cell proliferation as compared to the control, PBS (Lane 1; #P<0.01). compared with VEGF stimulated proliferation, the growth of SVEC4-10 cells were significantly inhibited by RBDV-IgG1 Fc in a dose-dependent way. (@, p<0.05)
(A)SVEC4-10
(B)B16/F10
Figure 13. RBDV-IgG1 Fc inhibits the VEGF-induced proliferation of SVEC4-10 cells, but not of B16/F10 cells. (A) SVEC4-10 cells and (B) B16/F10 cells were incubated with chimeric proteins for 72 hours and proliferation profile was determined by 1.9 mg/ml MTS. NC, PBS control. In lane 1, 5μg/ml of RBDV-Ig was added. In lane 2, 5μg/ml of IgG1-Fc was added. In lane 3, 10ng/ml VEGF was added.
In lane 4, RBDV-Ig and VEGF were added. In lane 5, IgG1 Fc and VEGF were added.
The data shown here are the mean of proliferation inhibition percentages ±SD obtained from three independent experiments (n=9). VEGF (In lane 3, 10ng/ml) stimulated a significant increase in endothelial cell proliferation as compared to the negative control, PBS (# P<0.001). The proliferation of SVEC4-10 cells were significantly inhibited by RBDV-IgG1 Fc when compared with VEGF or IgG1 Fc group (* P<0.05). There was no significant effect of VEGF or RBDV-Ig on B16/F10 cells.
Figure 14. Effect of RBDV-IgG1 Fc on in vitro tube formation, 100X magnification. To determine the angiogenic activity, the SVEC4-10 cells (3×104 cells/well/100μl) were pre-incubated with IgG1 Fc (10 μg/ml) or RBDV-IgG1 Fc (10 μg/ml) for 1 hour, and were added onto growth factor-reduced Matrigel with VEGF (10 ng/ml) in DMEM growth medium for 6 hours at 37°C. Under normal conditions, SVEC4-10 cells formed a network of tubes (A) that was enhanced when stimulated with VEGF (10 ng/ml). (B)SVEC4-10 cells were treated with IgG1 Fc. (C) Tube formation was apparently inhibited with RBDV-IgG1 Fc. (D) Three random fields were counted per well, and the total branches of the tube-like structures per field were calculated. The data represents a mean of the branch of tube-like structures ±SD obtained from three independent experiments (n=6). *P< 0.05 when compared with VEGF group. #P< 0.05 when compared with VEGF+IgG1 Fc group.
Figure 15. Effect of RBDV-IgG1 Fc on in vitro cell migration, 100X magnification.
To confirm the migration activity, the SVEC4-10 cells (3×104 cells/well/100μl) were pre-incubated with IgG1 Fc (10 μg/ml) or RBDV-IgG1 Fc (10 μg/ml) for 1 hour, and were added into the upper well of Transwell whose lower well contains 10 ng/ml of VEGF in DMEM growth medium for 6 hours at 37°C. After 6 hours, the cells in the upper well were removed by cotton buds, and the migrated cells on the other side of membrane were fixed and stained with PI. (A) Migration ability was enhanced when stimulated with VEGF (10 ng/ml). (B)SVEC4-10 cells were treated with IgG1 Fc. (C) migration ability of SVEC4-10 cells was apparently inhibited with RBDV-IgG1 Fc.
The small circular pores were 8μm pore of Transwell. (D) Three random fields were count per well, and the total number of cells were calculated. The data represents a mean of the branch of tube-like structures ±SD obtained from two independent experiments (n=4). *P< 0.05 when compared with VEGF group. #P< 0.05 when compared with IgG1 Fc group.
Figure 16. In vivo suppression of tumor growth with in situ RBDV-Ig Fc treatment. Female C57/BL6 (6-8 weeks of age) was inoculated with 1×106 B16/F10 cells subcutaneously in 200μl PBS. When the tumor average volume was up to 50 mm3, mice were injected with 150 μg proteins in situ. Tumor volume was measured every 2 days after injecting with proteins. The data represent a mean of tumor volume
±SD obtained from two independent experiments (n=6). For RBDV-Ig group, *means
P< 0.05 when compared with the negative group, # means P< 0.05 when compared
with IgG1 Fc group.Figure 17. Subcutaneous vascularizing of mouse dorsum. Mice were sacrificed when the tumor volume of negative group was up to 2,500 mm3. Dorsal skin was regularly cut from the mice. The vessels were obvious in (A) negative and (B) Ig-G1 Fc groups, but not in RBDV-Ig group (C).
Figure 18.
Figure 18. H&E staining of tumor sections, 200X magnification. Portions of the tumors were fixed in 4% paraformaldehyde overnight and subsequently paraffin embedded and cut in slides. Tumor sections were stained with H&E. (A) Negative group which was injected with 200μl PBS had lots of disorder blood vessels (BV) and the growth of the melanoma tumor B16/F10 was active. Tumors of the IgG1 Fc group (B) were randomly damaged, whereas tumors of the RBDV-Ig (C) group were disrupted around blood vessels.
Figure 19. In vivo tumor therapy with iv. injection of RBDV-Ig Fc. Female C57/BL6 (6-8 weeks of age) was inoculated with 1×106 cells subcutaneously in 200μl PBS. When the tumor average volume was up to 30 mm3, mice were intravenously (i.v.) injected with 150μg protein and injected again after five days. ▼ means the day with protein injection. Tumor volume was measured every 2 days after injecting with
Figure 19. In vivo tumor therapy with iv. injection of RBDV-Ig Fc. Female C57/BL6 (6-8 weeks of age) was inoculated with 1×106 cells subcutaneously in 200μl PBS. When the tumor average volume was up to 30 mm3, mice were intravenously (i.v.) injected with 150μg protein and injected again after five days. ▼ means the day with protein injection. Tumor volume was measured every 2 days after injecting with