Abstract
The present invention relates to human fibroblast growth factor (FGF-1) muteins having decreased mitogenicity, dimers of human FGF-1 muteins, as well as such human FGF-1 muteins and dimers of such muteins for use in reducing blood glucose level, particularly for use in the treatment of diabetes.
Claims
1. A human fibroblast growth factor 1 (FGF-1) mutein having decreased mitogenicity, characterized in that it comprises two point mutations: at amino acid position S114 and at amino acid position L150, wherein the numbering of amino acid positions is based on a full-length wild-type FGF-1 protein sequence as shown in Sequence No. 1.
2. The human FGF-1 mutein according to claim 1, characterized in that the point mutation at position S114 is S114A mutation.
3. The human FGF-1 mutein according to claim 1 or 2, characterized in that the point mutation at position L150 is L150D mutation
4. The human FGF-1 mutein according to any one of claims 1 to 3, characterized in that it comprises S114A point mutation and L150D point mutation.
5. The FGF-1 mutein according to any one of claims 1 to 4, characterized in that it further comprises at least one stabilizing mutation at amino acid position selected from: Q55, S62 and H108.
6. The FGF-1 mutein according to claim 5, characterized in that the at least one stabilizing mutation at amino acid position Q55, S62 or H108 is a point mutation selected from Q55P, S62I and H108G, respectively.
7. The FGF-1 mutein according to any one of claims 1 to 6, characterized in that it comprises three stabilizing mutations: Q55P, S62I and H108G.
8. The FGF-1 mutein according to claim 7, characterized in that it has an amino acid sequence presented in Sequence No. 10.
9. The FGF-1 mutein according to any one of claims 1 to 8, characterized in that it additionally comprises N-terminal deletion of at least 19 contiguous amino acids of the full-length FGF-1 protein.
10. The FGF-1 mutein according to claim 9, characterized in that it comprises N-terminal deletion of E3 to G21 amino acids of the full-length FGF-1 protein.
11. The FGF-1 mutein according to any one of claims 1 to 10, characterized in that it comprises S114A point mutation, L150D point mutation and N-terminal deletion of E3 to G21 amino acids of the full-length FGF-1 protein.
12. The FGF-1 mutein according to claim 11, characterized in that it has an amino acid sequence presented in Sequence No. 18.
13. A dimer of human fibroblast growth factor 1 (FGF-1) muteins having decreased mitogenicity as defined in any one of claims 1 to 12.
14. The dimer according to claim 13, characterized in that it is a homodimer.
15. The dimer according to claim 13 or 14, characterized in that muteins forming the dimer are connected by a linker, preferably an amino acid linker.
16. The dimer according to claim 15, characterized in that a linker is an amino acid sequence GGGGGGGGSGGGG.
17. The dimer according to claim 16, characterized in that it has an amino acid sequence presented in Sequence No. 21.
18. The dimer according to claim 16, characterized in that it has an amino acid sequence presented in Sequence No. 22.
19. The human fibroblast growth factor 1 (FGF-1) mutein having decreased mitogenicity as defined in any one of claims 1 to 12 for use in reducing blood glucose level.
20. The human fibroblast growth factor 1 (FGF-1) mutein having decreased mitogenicity as defined in any one of claims 1 to 12 for use in the treatment of diabetes, in particular type 2 diabetes.
21. The FGF-1 mutein for use according to claim 19 or 20, characterized in that it has an amino acid sequence presented in Sequence No. 10.
22. The FGF-1 mutein for use according to claim 19 or 20, characterized in that it has an amino acid sequence presented in Sequence No. 18.
23. The dimer of mutated human fibroblast growth factor 1 (FGF-1) proteins having decreased mitogenicity as defined in any one of claims 13 to 18 for use in reducing blood glucose level.
24. The dimer of mutated human fibroblast growth factor (FGF-1) proteins having decreased mitogenicity as defined in any one of claims 13 to 18 for use in the treatment of diabetes, in particular type 2 diabetes.
25. The dimer of mutated FGF-1 proteins for use according to claim 23 or 24, characterized in that it has an amino acid sequence presented in Sequence No. 21.
26. The dimer of mutated FGF-1 muteins for use according to claim 24 or 25, characterized in that it has an amino acid sequence presented in Sequence No. 22.
Description
BRIEF DESCRIPTION OF THE FIGURES
[0120] FIG. 1 shows a proliferation assay measured by MTT method using the commercial CellTiter 96 Non-Radioactive Cell Proliferation Assay kit (Promega). NIH3T3 cells were seeded at 2000 cell/well on a 96-well plate using Dulbecco's Modified Eagle's Medium (DMEM) supplemented with 10% calf serum (CS). On the next day the cells were starved for 5 hours in DMEM without CS; subsequently, proteins were added at the specified concentrations, and were incubated for 48 hours. The MTT assay was performed in accordance with the manufacturer's protocol. The results of the experiment showed an increase in proliferation of both forms of FGF1 without significant differences between the truncated formFGF1 protein (155aa) and the full-length formFGF1(155aa) protein.
[0121] FIG. 2 shows the results of a proliferation assay measured by MTT method using the commercial CellTiter 96 Non-Radioactive Cell Proliferation Assay kit (Promega). NIH3T3 cells were seeded at 2000 cells/well on a 96-well plate using DMEM supplemented with 10% CS. On the following day the cells were starved for 5 hours in DMEM without CS; subsequently, proteins were added at the specified concentrations, and were incubated for 48 hours. The MTT assay was performed according to the manufacturer's protocol. The results of the experiment showed a significant induction of proliferation by the full-length form of the wild-type FGF1 protein (155aa) and insignificant induction of proliferation by S114A mutant (with reduced affinity to CK2), both in the truncated formFGF1(155aa; S114A) and the full-length formFGF1(155aa; S114A).
[0122] FIG. 3 shows the results of a proliferation assay measured by MTT method using the commercial CellTiter 96 Non-Radioactive Cell Proliferation Assay kit (Promega). NIH3T3 cells were seeded at 2000 cells/well on a 96-well plate using DMEM supplemented with 10% CS. On the following day the cells were starved for 5 hours in DMEM without CS; subsequently, proteins were added at the specified amounts, and were incubated for 48 hours. The MTT assay was performed according to the manufacturer's protocol. The results of the experiment showed a significant induction of proliferation by the full form of the wild-type FGF1 protein (155aa) and insignificant induction of proliferation by L150D mutant (with reduced affinity to the FGFR1 receptor), both in the truncated formFGF1(155aa; L150D) and the full-length formFGF1(155aa; L150D).
[0123] FIG. 4 shows the results of a proliferation assay measured by MTT method using the commercial CellTiter 96 Non-Radioactive Cell Proliferation Assay kit (Promega). NIH3T3 cells were seeded at 2000 cells/well on a 96-well plate using DMEM supplemented with 10% CS. On the following day the cells were starved for 5 hours in DMEM without CS; subsequently, proteins were added at the specified concentrations, and were incubated for 48 hours. The MTT assay was performed according to the manufacturer's protocol. The results of the experiment showed a significant induction of proliferation by the full form of the wild-type FGF1 protein (155aa) and insignificant induction of proliferation by S114A/L150D mutant (with reduced affinity to CK2 and the FGFR1 receptor), both in the truncated formFGF1(155aa; S114A/L150D) and the full-length formFGF1(155aa; S114A/L150D).
[0124] FIG. 5 shows the results of a proliferation assay measured by MTT method using the commercial CellTiter 96 Non-Radioactive Cell Proliferation Assay kit (Promega). NIH3T3 cells were seeded at 2000 cells/well on a 96-well plate using DMEM supplemented with 10% CS. On the next day the cell were starved for 5 hours in DMEM without CS; subsequently, proteins were added in the specified concentrations, and were incubated for 48 hours. The MTT assay was performed according to the manufacturer's protocol. The results of the experiment showed a significant induction of proliferation by the wild-type form of FGF1 protein (155aa) and insignificant induction of proliferation by Q55P/S62I/H108G/S114A/L150D mutant (the mutant with increased thermal stability and reduced affinity to CK2 and the FGFR1 receptor), both in the truncated formFGF1(155aa; Q55P/S62I/H108G/S114A/L150D), and the full-length formFGF1(155aa; Q55P/S62I/H108G/S114A/L150D).
[0125] FIG. 6 shows the results of a proliferation assay measured by MTT method using the commercial CellTiter 96 Non-Radioactive Cell Proliferation Assay kit (Promega). NIH3T3 cells were seeded at 2000 cells/well on a 96-well plate using DMEM supplemented with 10% CS. On the following day the cells were starved for 5 hours in DMEM without CS; subsequently, proteins were added at the specified concentrations, and were incubated for 48 hours. The MTT assay was performed according to the manufacturer's protocol. The proteins used in the experiment are mutants of the full-length FGF1(155aa) protein. The results of the experiment showed a very high induction of proliferation by the mutant with increased thermal stability-FGF1(155aa; Q55P/S62I/H108G). Moreover, the experiment showed an induction of proliferation by the wild-type form of FGF1(155aa) protein and insignificant induction of proliferation by mutants: FGF1(155aa; Q55P/S62I/H108G/S114A/L150D), FGF1(155aa; S114A), FGF1(155aa; L150D) and FGF1(155aa; S114A/L150D).
[0126] FIG. 7 shows the results of a proliferation assay measured by MTT method using the commercial CellTiter 96 Non-Radioactive Cell Proliferation Assay kit (Promega). NIH3T3 cells were seeded at 2000 cells/well on a 96-well plate using DMEM supplemented with 10% CS. On the following day the cells were starved for 5 hours in DMEM without CS; subsequently, proteins were added at the specified concentrations, and were incubated for 48 hours. The MTT test was performed according to the manufacturer's protocol. The proteins used in the experiment are mutants of the full-length form of the FGF1(155aa) protein. The results of the experiment showed a very high induction of proliferation by the mutants with increased thermal stability-FGF1(155aa; Q55P/S62I/H108G), FGF1(155aa; Q55P/S62I/H108G/S153A), FGF1(155aa; Q55P/S62I/H108G/S153R). Moreover, the experiment showed an induction of proliferation by the wild-type form of FGF1(155aa) protein, and mutants: FGF1(155aa; S153A), FGF1(155aa; S153R). FGF1(155aa; Q55P/S62I/H108G/S114A/L150D) mutant showed an insignificant induction of proliferation.
[0127] FIG. 8 shows the results of a proliferation assay measured by MTT method using the commercial CellTiter 96 Non-Radioactive Cell Proliferation Assay kit (Promega). NIH3T3 cells were seeded at 2000 cells/well on a 96-well plate using DMEM supplemented with 10% CS. On the following day the cells were starved for 5 hours in DMEM without CS; subsequently, proteins were added at the specified concentrations, and were incubated for 48 hours. The MTT assay was performed according to the manufacturer's protocol. The proteins used in the experiment are mutants of the full-length form of the FGF1(155aa) protein. The results of the experiment showed a very high induction of proliferation by the mutants: FGF1(155aa; S153R) and FGF1(155aa; Q55P/S62I/H108G/S114A/L150D). Moreover, the experiment showed an induction of proliferation by the wild-type form of the FGF1(155aa) protein and mutants: FGF1(155aa; S153A), FGF1(155aa; Q55P/S62I/H108G). FGF1(155aa; Q55P/S62I/H108G/S114A/L150D) mutant showed an insignificant induction of proliferation.
[0128] FIG. 9 shows the results of a proliferation assay measured by MIT method using the commercial CellTiter 96 Non-Radioactive Cell Proliferation Assay kit (Promega). NIH3T3 cells were seeded at 2000 cells/well on a 96-well plate with the use of Dulbecco's Modified Eagle's Medium (DMEM) with 10% addition of CS. On the following day the cells were starved for 5 hours in DMEM without CS; subsequently, proteins were added at the specified concentrations, and were incubated for 48 hours. The MTT assay was performed according to the manufacturer's protocol. The results of the experiment showed an increase in the proliferation of both forms of FGF1 without significant differences between the truncated formFGF1(155aa) protein and the truncated dimeric formFGF1_DIMER(155aa) protein.
[0129] FIG. 10 shows the results of a proliferation assay measured by MTT method using the commercial CellTiter 96 Non-Radioactive Cell Proliferation Assay kit (Promega). NIH3T3 cells were seeded at 2000 cells/well on a 96-well plate using DMEM supplemented with 10% CS. On the following day the cells were starved for 5 hours in DMEM without CS; subsequently, proteins were added at the specified concentrations (with an addition of heparin, 10 U/ml, or without it), and were incubated for 48 hours. The MTT assay was performed according to the manufacturer's protocol. The results of the experiment showed a positive effect of heparin on the increase in proliferation for the wild-type form of the FGF1(155aa) protein. The lack of heparin had an effect on the intensity of proliferation but did not stop it. The results unambiguously indicate that heparin does not influence the effectiveness of proliferation induction by FGF1(155aa). Moreover, the experiment showed an insignificant induction of proliferation by FGF1(155aa; Q55P/S62I/H108G/S114A/L150D) mutant, both with or without the addition of heparin.
[0130] FIG. 11 shows the results of a glucose uptake assay measured by the chemiluminescence method using the commercial Glucose Uptake-Glo Assay kit (Promega). In the test, adipocytes differentiated from mouse fibroblasts 3T3-L1 were used according to the protocol provided by the manufacturer, Glucose Uptake-Glo (Promega). The differentiated adipocytes were seeded at 50000 cells/well on a 96-well Poly-D-Lysine coated plate using DMEM supplemented with 10% foetal bovine serum (FBS). On the following day, proteins were added at the specified concentrations in DMEM without FBS, and were incubated for 16 hours. The glucose uptake assay was performed according to the manufacturer's protocol. The results of the experiment showed an increase in glucose uptake for both forms of FGF1, without significant differences between the truncated formFGF1(155aa) protein and the full-length FGF1(155aa) protein.
[0131] FIG. 12 shows the results of a glucose uptake assay measured by the chemiluminescence method using the commercial Glucose Uptake-Glo Assay kit (Promega). In the test, adipocytes differentiated from mouse fibroblasts 3T3-L1 were used according to the protocol provided by the manufacturer, Glucose Uptake-Glo (Promega). The differentiated adipocytes were seeded at 50000 cells/well on a 96-well Poly-D-Lysine coated plate using DMEM supplemented with 10% FBS. On the following day, proteins were added at the specified concentrations in DMEM without FBS, and were incubated for 16 hours. The glucose uptake assay was performed according to the manufacturer's protocol. The results of the experiment showed an increase in glucose uptake for the wild-type form of the FGF1(155aa) protein and a significant increase in glucose uptake (exceeding the action of FGF1(155aa)) for S114A mutant, both in the truncated formFGF1(155aa; S114A) and the full-length formFGF1(155aa; S114A).
[0132] FIG. 13 shows the results of a glucose uptake assay measured by the chemiluminescence method using the commercial Glucose Uptake-Glo Assay kit (Promega). In the test, adipocytes differentiated from mouse fibroblasts 3T3-L1 were used according to the protocol provided by the manufacturer, Glucose Uptake-Glo (Promega). The differentiated adipocytes were seeded at 50000 cells/well on a 96-well Poly-D-Lysine coated plate using DMEM supplemented with 10% FBS. On the following day, proteins were added at the specified concentrations in DMEM without FBS, and were incubated for 16 hours. The glucose uptake assay was performed according to the manufacturer's protocol. The results of the experiment showed an increase in glucose uptake for the wild-type form of the FGF1(155aa) protein and a significant decrease in glucose uptake (compared to FGF1(155aa)) for L150D mutant, both in the truncated formFGF1(155aa; L150D) and the full-length formFGF1(155aa; L150D).
[0133] FIG. 14 shows the results of a glucose uptake assay measured by the chemiluminescence method using the commercial Glucose Uptake-Glo Assay kit (Promega). In the test, adipocytes differentiated from mouse fibroblasts 3T3-L1 were used according to the protocol provided by the manufacturer, Glucose Uptake-Glo (Promega). The differentiated adipocytes were seeded at 50000 cells/well on a 96-well Poly-D-Lysine coated plate using DMEM supplemented with 10% FBS. On the following day, proteins were added at the specified concentrations in DMEM without FBS, and were incubated for 16 hours. The glucose uptake assay was performed according to the manufacturer's protocol. The results of the experiment showed an increase in glucose uptake for the wild-type form of the FGF1(155aa) protein and a significant decrease in glucose uptake (compared to the FGF1(155aa)) for S114A/L150D mutant, both in the truncated formFGF1(155aa; S114A/L150D), and the full-length formFGF1(155aa; S114A/L150D).
[0134] FIG. 15 shows the results of a glucose uptake assay measured by the chemiluminescence method using the commercial Glucose Uptake-Glo Assay kit (Promega). In the test, adipocytes differentiated from mouse fibroblasts 3T3-L1 were used according to the protocol provided by the manufacturer, Glucose Uptake-Glo (Promega). The differentiated adipocytes were seeded at 50000 cells/well on a 96-well Poly-D-Lysine coated plate using DMEM supplemented with 10% FBS. On the following day, proteins were added at the specified concentrations in DMEM without FBS, and were incubated for 16 hours. The glucose uptake assay was performed according to the manufacturer's protocol. The results of the experiment showed a significant increase in glucose uptake for the wild-type form of the FGF1(155aa) protein and an increase in glucose uptake for the 1000 ng/ml dose of Q55P/S62I/H108G/S114A/L150D mutant, both in the truncated formFGF1(155aa; Q55P/S62I/H108G/S114A/L150D) and the full-length formFGF1(155aa; Q55P/S62I/H108G/S114A/L150D).
[0135] FIG. 16 shows the results of a glucose uptake assay measured by the chemiluminescence method using the commercial Glucose Uptake-Glo Assay kit (Promega). In the test, adipocytes differentiated from mouse fibroblasts 3T3-L1 were used according to the protocol provided by the manufacturer, Glucose Uptake-Glo (Promega). The differentiated adipocytes were seeded at 50000 cells/well on a 96-well Poly-D-Lysine coated plate using DMEM supplemented with 10% FBS. On the following day, proteins were added at the specified concentrations in DMEM without FBS and were incubated for 16 hours. The glucose uptake assay was performed according to the manufacturer's protocol. The proteins used in the experiments are mutants of the full-length FGF1(155aa) protein. The results of the experiment showed a very high glucose uptake for the mutant with increased thermal stability-FGF1(155aa; Q55P/S62I/H108G). Moreover, it showed an increased glucose uptake for FGF1(155aa), FGF1(155aa; Q55P/S62I/H108G/S114A/L150D and FGF1(155aa; S114A). FGF1(155aa; L150D) and FGF1(155aa; S114A/L150D) exhibited an insignificant or no increase in glucose uptake.
[0136] FIG. 17 shows the results of a glucose uptake assay measured by the chemiluminescence method using the commercial Glucose Uptake-Glo Assay kit (Promega). In the test, adipocytes differentiated from mouse fibroblasts 3T3-L1 were used according to the protocol provided by the manufacturer, Glucose Uptake-Glo (Promega). The differentiated adipocytes were seeded at 50000 cells/well on a 96-well Poly-D-Lysine coated plate using DMEM supplemented with 10% FBS. On the following day, proteins were added at the specified concentrations in DMEM without FBS, and were incubated for 16 hours. The glucose uptake assay was performed according to the manufacturer's protocol. The proteins used in the experiments are mutants of the full-length FGF1(155aa) protein. The results of the experiment showed a very high induction of glucose uptake for the mutants with increased thermal stability FGF1(155aa; Q55P/S62I/H108G), FGF1(155aa; Q55P/S62I/H108G/S153A), FGF1(155aa; Q55P/S62I/H108G/S153R). Moreover, the experiment showed an increased glucose uptake for FGF1(155aa) and FGF1(155aa; Q55P/S62I/H108G/S114A/L150D), whereas FGF1(155aa; Q55P/S62I/H108G/S153A) and FGF1(155aa; Q55P/S62I/H108G/S153R) showed the highest induction of glucose uptake.
[0137] FIG. 18 shows the results of a glucose uptake assay measured by the chemiluminescence method using the commercial Glucose Uptake-Glo Assay kit (Promega). In the test, adipocytes differentiated from mouse fibroblasts 3T3-L1 according to the protocol provided by the manufacturer, Glucose Uptake-Glo (Promega). The differentiated adipocytes were seeded at 50000 cells/well on a 96-well Poly-D-Lysine coated plate using DMEM supplemented with 10% FBS. On the following day, proteins were added at the specified concentrations in DMEM without FBS, and were incubated for 16 hours. The glucose uptake assay was performed according to the manufacturer's protocol. The results of the experiment showed an increase in glucose uptake for both dimeric forms of FGF1, without significant differences between the truncated form of the FGF1_DIMER(155aa) protein and the full-length form of the FGF1_DIMER(155aa) protein.
[0138] FIG. 19 shows the results of an analysis of protein expression and phosphorylation of proteins of FGF:FGFR dependent pathway performed by Western Blot (WB) method with membrane scanning performed by the chemiluminescence method using ChemiDock (Bio-Rad) equipment. In the test, adipocytes differentiated from mouse fibroblasts 3T3-L1 according to the protocol provided by the manufacturer, Glucose Uptake-Glo (Promega). The differentiated adipocytes were seeded at 50000 cells/well on a 96-well Poly-D-Lysine coated plate using DMEM supplemented with 10% FBS. On the following day, the cells were starved for 5 hours in DMEM without FBS; subsequently proteins were added at the concentration of 100 ng/ml, and were incubated for 10 minutes to study phosphorylation of proteins of FGF:FGFR-dependent pathway, and for 16 hours to study the expression of protein Glut1. After incubation the proteins were washed with cold PBS, and cell lysis was performed using a RIPA buffer with an addition of protease and phosphatase inhibitors. The proteins were separated by SDS-PAGE electrophoresis and were transferred onto a nitrocellulose membrane. Blocked membranes were incubated for 16 hours at the temperature of 4 with primary antibodies in 1:1000 dilution. Anti-rabbit-HRP secondary antibodies were incubated for 1 hour at room temperature. Clarity Max ECL buffers were used to obtain a chemiluminescent signal. The control of protein concentration was performed by means of Stain-Free signal analysis. The results of the experiment showed an increase in the phosphorylation of the FGFR1 receptor and the FGFR1 receptor dependent proteins, i.e. pErk1/2 and pFRS2a for FGF1(155aa), FGF1(155aa; Q55P/S62I/H108G) and FGF1(155aa; S114A). FGF1(155aa; Q55P/S62I/H108G/S114A/L150D insignificantly activated FGF1:FGFR pathway whereas FGF1(155aa; L150D) and FGF1(155aa; S114A/L150D) showed lack of activation.
[0139] FIG. 20 shows the results of the in vivo study of the blood glucose concentration reducing effect in a diabetic strain of db/db mice (BKS.Cg+Leprdb/+Leprdb/OlaHsd) after administration of 0.5 mg/kg m.c. of the wild-type FGF1(155aa) protein with and without the addition of 10 U/ml heparin. The control group was db/db mice receiving a carrier. The results unambiguously indicate a positive effect of FGF1(155aa) on reducing glucose to normoglycemia and lack of heparin action on the effectiveness of glucose reduction in db/db mice.
[0140] FIG. 21 shows the results of the in vivo study of the blood glucose concentration reducing effect in a diabetic strain of db/db mice (BKS.Cg+Leprdb/+Leprdb/OlaHsd) after administration of 1 mg/kg m.c. of wild-type FGF1 protein (155aa) and FGF1(155aa; S114A) mutant. The control group was db/db mice receiving a carrier. The results show a positive effect of FGF1(155aa) and FGF1(155aa; S114A) mutant on reducing glucose to normoglycemia in db/db mice.
[0141] FIG. 22 shows the results of the in vivo study of the blood glucose concentration reducing effect in a diabetic strain of db/db mice (BKS.Cg+Leprdb/+Leprdb/OlaHsd) after administration of 1 mg/kg m.c. of the wild-type FGF1 protein (155aa) and FGF1(155aa; L150D) mutant. The control group was db/db mice receiving a carrier. The results indicate a positive effect of FGF1(155aa) on reducing glucose to normoglycemia in db/db mice and lack of effective action of FGF1(155aa; L150D) mutant.
[0142] FIG. 23 shows the results of the in vivo study of the blood glucose concentration reducing effect in a diabetic strain of db/db mice (BKS.Cg+Leprdb/+Leprdb/OlaHsd) after administration of 1 mg/kg m.c. of the FGF1 protein (155aa) and FGF1(155aa; Q55P/S62I/H108G) mutant. The control group was db/db mice receiving a carrier. The results indicate a positive effect of FGF1(155aa) and FGF1(155aa; Q55P/S62I/H108G) on reducing glucose to normoglycemia in db/db mice, with the effectiveness of action of FGF1(155aa; Q55P/S62I/H108G) mutant being higher compared to FGF1(155aa).
[0143] FIG. 24 shows the results of the in vivo study of the blood glucose concentration reducing effect in a diabetic strain of db/db mice (BKS.Cg+Leprdb/+Leprdb/OlaHsd) after administration of 1 mg/kg m.c. of the FGF1(155aa) protein and Q55P/S62I/H108G/S114A/L150D mutants. The control group was db/db mice receiving a carrier. The results indicate a positive effect of FGF1(155aa) and Q55P/S62I/H108G/S114A/L150D mutants on reducing glucose to normoglycemia in db/db mice without significant differences between the truncated formFGF1(155aa; Q55P/S62I/H108G/S114A/L150D) and the full-length formFGF1(155aa; Q55P/S62I/H108G/S114A/L150D).
[0144] FIG. 25 shows the results of the in vivo study of the blood glucose concentration reducing effect in a diabetic strain of db/db mice (BKS.Cg+Leprdb/+Leprdb/OlaHsd) after administration of 1 mg/kg m.c. of the FGF1 protein (155aa), FGF1(155aa) and dimeric form of the truncated mutein-FGF1_DIMER(155aa). The control group was db/db mice receiving a carrier. The results indicate a positive effect of FGF1(155aa) and FGF1(155aa) on reducing glucose to normoglycemia in db/db mice and lack of effective action of the truncated dimeric formFGF1_DIMER(155aa).
[0145] FIG. 26 shows the results of the in vivo study of the blood glucose concentration reducing effect in a diabetic strain of mice db/db (BKS.Cg+Leprdb/+Leprdb/OlaHsd) after administration of 5 mg/kg m.c. of FGF1 protein (155aa), mutant FGF1(155aa; Q55P/S62I/H108G/S114A/L150D) and dimeric form FGF1_M8_DIMER(155aa). The control group was db/db mice receiving a carrier. The results indicate a positive effect of FGF1(155aa) and FGF1(155aa; Q55P/S62I/H108G/S114A/L150D) on reducing glucose to normoglycemia in db/db mice and a lower, compared to the monomeric version, effectiveness of action of the dimeric formFGF1_M8_DIMER(155aa).
[0146] FIG. 27 shows the results of the in vivo study of the blood glucose concentration reducing effect in a diabetic strain of db/db mice (BKS.Cg+Leprdb/+Leprdb/OlaHsd) after administration of 1 mg/kg m.c. of the FGF1(155aa) protein, and three doses: 1, 2.5, 5 mg/kg m.c. for FGF1(155aa; Q55P/S62I/H108G/S114A/L150D). The control group was db/db mice receiving a carrier. The results indicate a positive effect of FGF1(155aa) and FGF1(155aa; Q55P/S62I/H108G/S114A/L150D) on reducing glucose to normoglycemia in db/db mice with clear dose-dependence for FGF1(155aa; Q55P/S62I/H108G/S114A/L150D).
EXAMPLES
[0147] All the procedures, assays and experimental analyses described below were performed with the use of commercially available test kits, reactants and apparatus, following recommendations of the manufacturers of the applied kits, reagents and apparatus, unless otherwise clearly indicated herein; the inventors used standard, commonly known methods applied in the field to which the present invention belongs.
Example 1
[0148] Production of constructs encoding FGF1 protein variants according to the invention and disclosure, and their expression and purification.
[0149] Wild-type variant FGF1 WTcDNA (GenBank accession number NM 001354952.2, 468 bp, encoding human fibroblast growth factor 1 (FGF-1), was optimized to an expression in E. coli cells and a gene flanked with the restriction sites for enzymes: Ndel at 5 terminus and XhoI at 3 terminus (Gene Synthesis, Thermo Fisher Scientific) was synthetized. The synthetic gene was cloned into a modified expression vector pCPBT0010 prepared by Celon Pharma, using the above specified restriction enzymes. The designed construct for FGF1(155aa) protein did not have labels facilitating purification and additional amino acids at N or C terminuses.
[0150] All muteins constructed in accordance with the invention, i.e. mutated sequences of proteins FGF1(155aa) (short variants) and FGF1(155aa) (full-length variants) comprising one, two, three, four of five point mutations were obtained in a standard way in a PCR reaction, unless indicated otherwise, in accordance with the methodology of directed mutagenesis and then transformation to DH5 E. coli, in accordance with the methodology described by Hanahan et al. (Hanahan, D., Jessee, J., & Bloom, F. R. (1991). Plasmid transformation of Escherichia coli and other bacteria. Methods in Enzymology, 204, 63-113. https://doi.org/10.1016/0076-6879(91)04006-a).
[0151] Construction of muteins: FGF1(155aa; L150D), FGF1(155aa; S114A)single mutants were prepared on the basis of the FGF1 sequence (1-155 aa, Sequence No. 1) in a PCR reaction in accordance with the methodology of Site-Directed Point Mutagenesis described by Stratagene. All amplification reactions (50 l) comprised a polymerase buffer, dNTP, template DNA, primers complementary to sense and antisense strands carrying a mutated codon and Q5 Hot Start High-Fidelity polymerase (New England Biolab). Reaction mixture was transformed to DH5 E. coli cells in accordance with the Hanahan's methodology. (Hanahan et al., 1991).
[0152] Construction of mutein: FGF1(155aa; S114A/L150D), FGF1(155aa; Q55P/S62I/H108G/S114A/L150D)a mutein with two point mutations was prepared by modifying cDNA codons of the gene for the wild-type FGF1(aa 1-155) protein. DNA sequences were optimized for an expression in E. coli cells, genes were synthesized and prepared in accordance with the above description.
[0153] Construction of mutein: FGF1(155aa; Q55P/S62I/H108G/S114A/L150D)A multiple mutein was prepared by modifying cDNA codons of the gene for the wild-type FGF1(aa 1-155) protein (Sequence nr 1). DNA sequences were optimized for an expression in E. coli cells, genes were synthesized and prepared in accordance with the above description.
[0154] Construction of muteins: FGF1(155aa; S153A), FGF1(155aa; S153R)single mutants were prepared on the basis of the FGF1 sequence (1-155 aa, Sequence No. 1) in a PCR reaction in accordance with Site-Directed Point Mutagenesis described by Stratagene. All amplification reactions (50 l) comprised a polymerase buffer, dNTP, template DNA, primers complementary to sense and antisense strands carrying a mutated codon and Q5 Hot Start High-Fidelity polymerase (New England Biolab). Reaction mixture was transformer to DH5 E. coli cells in accordance with the Hanahan's methodology (Hanahan et al., 1991).
[0155] Construction of muteins: FGF1(155aa; Q55P/S62I/H108G/S153A); FGF1(155aa; Q55P/S62I/H108G/S153R)Multiple muteins were prepared by modifying cDNA codons of the gene for the wild-type FGF1(aa 1-155) protein (Sequence No. 1). DNA sequences were optimized for an expression in E. coli cells, genes were synthesized and prepared in accordance with the above description.
[0156] Construction of mutein: FGF1(155aa; S153D/S154D)a single mutant was prepared on the basis of the FGF1 sequence (1-155 aa, Sequence nr 1) in a PCR reaction in accordance with the methodology of Site-Directed Point Mutagenesis described by Stratagene. All amplification reactions (50 l) comprised a polymerase buffer, dNTP, template DNA, primers complementary to sense and antisense strands carrying a mutated codon and Q5 Hot Start High-Fidelity polymerase (New England Biolab). Reaction mixture was transformed to DH5 E. coli cells in accordance with the Hanahan's methodology (Hanahan et al., 1991).
[0157] Construction of mutein: FGF1(155aa; Q55P/S62I/H108G/S153D/S154D)multiple muteins were prepared by modifying cDNA codons of the gene for the wild-type FGF1(aa 1-155) protein (Sequence nr 1). DNA sequences were optimized for an expression in E. coli cells, genes were synthesized and prepared in accordance with the above description.
[0158] Construction of muteins: FGF1(155aa; Q55P/S62I/H108G/S153D/S154D); FGF1(155aa; S114A); FGF1(155aa; L150D); FGF1(155aa; S114A/L150D); FGF1(155aa; Q55P/S62I/H108G/S153A); FGF1(155aa; Q55P/S62I/H108G/S153R); FGF1(155aa; S153A), FGF1(155aa; S153R)constructs for these mutated proteins included cDNA encoding information for the wild-type form of the FGF1 protein truncated at N-terminus and its truncated muteins. DNA sequences were optimized for an expression in E. coli cells, synthetic genes were ordered and prepared in accordance with the above description.
Expression and Purification of FGF1 Protein Variants According to the Invention and Disclosure
[0159] FGF1 protein and its variants in the long FGF1(155aa) version and truncated FGF1(155aa) version were expressed in E. coli cells in TB (Terrific Broth; Sigma) medium supplemented with antibiotics. Kanamycin comprising TB medium was inoculated with an overnight pre-culture. Protein expression was induced with IPTG and then the cultures were grown for 20 h. Bacterial cultures were centrifuged for 15 min. at 6.000g at 4 C. The pellets obtained were suspended in a lysis buffer and were incubated for 30 min, and then they were subjected to sonication on ice for 5 min. Filtered soluble fractions (supernatant) obtained by centrifuging the samples for 30 min at 20.000g at 4 C. were loaded on a heparin column (Heparin 6 FastFlow, Cytiva). Resin-unbound proteins were washed out with buffer A and FGF1 elution was performed using a gradient in buffer B. Eluted FGF1 proteins were subjected to a further process of gel filtration (HiLoad 16/600 Superdex 75 prep grade column, Cytivia). The proteins produced were subjected to a quantitative analysis (A280) using extinction coefficient and qualitative analyses: purity using SDS/PAGE, capillary electrophoresis, thermal shift, SEC-HPLC, MS. The proteins were aliquoted, frozen in liquid nitrogen and stored at 80 C.
Example 2
[0160] Confirming the effectiveness of the exogenous recombinant human FGF1 protein.
[0161] The test as carried out was aimed at confirming the antidiabetic activity of the obtained FGF1(155aa) in accordance with the invention. db/db mice were administered subcutaneously 0.5 mg/kg m.c. of the FGF1 protein (155aa) with an addition of heparin (the unchanged FGF1 protein, that is why heparin was used to protect the protein against degradation and inactivation). The protein effectively reduces glucose concentration up to 30 h after injection. db/db mice which received only a carrier were used as the control group (FIG. 20).
Example 3
[0162] Confirming the effectiveness of FGF1(155aa) and the truncated variant-FGF1(155aa; Q55/S62/H108)without heparin addition.
[0163] The test examined the antidiabetic effectiveness of FGF1(155aa) protein without the of heparin and effectiveness of the truncated variant-presence FGF1(155aa; Q55/S62/H108). Ultimately, proteins without the presence of heparin in the buffer were to be administered, because administration of heparin causes side effects and that is why it cannot be administered to human subjects.
[0164] The test was performed d with the use of proteins: FGF1(155aa) and FGF1(155aa; Q55/S62/H108) variant at 1 mg/kg m.c. concentration. Both proteins exhibited antidiabetic activity but the mutein with stabilizing mutations (the stable variant)FGF1(155aa; Q55/S62/H108)maintained the effect considerably longer (FIG. 23). Based on the obtained results it was concluded that stabilizing the FGF1 prolongs antidiabetic activity.
[0165] By introducing the known stabilizing mutations described in the literature, preferably all such three stabilizing mutations as described above, to the mutated human FGF1 protein according to the invention, a thermally stabilized protein has been obtained which also has a high antidiabetic potential. This also increases the pharmacological usability of this protein, because a stable protein can be efficiently produced on a large scale. However, the mitogenic effect, which is one of the main factors limiting the clinical use of rhFGF1, still has not been eliminated.
Example 4
[0166] Reducing the mitogenic potential of the human FGF1 protein.
[0167] The inventors investigated the mitogenic potential of the truncated human FGF1 protein with introduced S114A mutation in a system without heparin in the cell line NIH 3T3 using a method described by Skjerpen at al. (Skjerpen at al., 2002). In the experiments conducted in the cell line NIH 3T3, the FGF1(155aa; S114A) protein does not stimulate proliferation of cells compared to the unchanged FGF1 (FIG. 2). The results obtained are different from the results obtained by Skjerpen et al. It is necessary to take into account two variables that differentiate both experiments. First, in the study by Skjerpen et al. the experiments were conducted in a system that included also heparin which stabilized the protein. In the Applicant's experiments the effect was observed for the protein which was not additionally stabilized with heparin because protein is administered in vivo without heparin due to its toxic effect on an organism. Moreover, the Applicant demonstrated that the presence of heparin is not necessary to demonstrate the mitogenic potential of the wild-type protein (FIG. 10). Second, Skjerpen et al. worked with a full-length protein whereas the Applicant performed their experiment using both versions: the truncated FGF1(155aa) and the full-length FGF1(155aa).
Example 5
[0168] Impairment of interaction with the FGFR1 receptor by introduction of a mutation in another site than the heparin binding site.
[0169] The activation and dimerization of the FGF1 receptor (FGFR1) is necessary to induce mitogenic response. Impairment of ligand binding to the receptor weakens interaction, reduces signal transduction and does not promote cell proliferation.
[0170] Variants of FGF-1 were generated which were characterized by impaired binding to the FGFR1 receptor. The mutation was developed within the domain responsible for affinity to the receptor but the modified residue is not crucial for this binding. The Leucine (L) residue at C-terminus was substituted with aspartic acid (D) (L150D).
[0171] Subsequently, the applicant verified whether a protein obtained in this way, which poorly binds to the receptor, has a weaker mitogenic potential. To this end an experiment was performed with NIH 3T3 cells, which were exposed to the action of the studied proteins for 48 hours. Subsequently, MTT reactant was added to the cells and the level of absorption was measured. Reduced proliferation of cells treated with FGF1(155aa; L150D), FGF1(155aa; L150D) was observed compared to the cells treated with FGF1(155aa) (FIG. 3).
[0172] Having obtained the effect of weakened mitogenicity, the inventors verified whether the protein was still characterized by a high antidiabetic activity. In the in vitro assays, 3T3-L1 cells differentiated to adipocytes were treated with the FGF1 protein or a variant: FGF1(155aa; L150D), FGF1(155aa; L150D). After 16 hours from administration, a glucose uptake assay was performed in accordance with the manufacturer's protocol (Glucose Uptake Glo Assay, Promega). In the in vitro assay without addition of heparin the protein induced a slightly increased glucose uptake (FIG. 13).
[0173] The in vitro results proved to be promising, both in the assessment of mitogenic potential and antidiabetic effectiveness of the new variant-FGF1(155aa; L150D). That is why the new protein with L150D mutation was tested for the ability to reduce blood glucose level in vivo, in the db/db mice animal model. 1 mg/kg dose of the protein was administered subcutaneously to the mice. Subsequently, at several points of time, blood glucose level was measured in the mice by means of a glucometer by puncturing the tip of the tail. Unfortunately, FGF1(155aa; L150D) variant failed to reduce blood glucose in vivo (FIG. 22).
[0174] The reason for the lack of effect consists probably in the physical properties of the proteinthis mutant is characterized by low stability, its denaturation temperature is about 28 C. Thus, the protein administered subcutaneously is immediately unfolded and heparans naturally present on the surface of cells are not sufficient to stabilize the protein. At the physiological body temperature of mice (about 37 C.) the protein undergoes denaturation and losses its activity.
Example 6
[0175] Activity of FGF1 protein variants with double mutation (FGF1(155aa; S114A/L150D)) In the experiment, the activity of a mutant comprising S114A mutation and L150D mutation in NIH 3T3 cells was tested in an MTT assay. It was fund out that such a double mutant exhibits decreased mitogenic properties (FIG. 4). Both S114A mutation and L150D mutation as well as their combination reduce proliferation compared to the wild-type protein. Interestingly, the most decreased mitogenic effect was observed for the stabilized protein, i.e. one additionally comprising the three stabilizing mutations according to the invention FGF1(155aa; Q55P/S62I/H108G), as will be shown below. Additionally, the protein with five mutationsFGF1(155aa; Q55P/S62I/H108G/S114A/L150D)according to the invention exhibits antidiabetic activity, contrary to the L150D and S114A/L150D variants without stabilizing mutations.
[0176] In the experiment testing a signal cascade responsible for triggering a proliferation path, FGF1(155aa; L150D), FGF1(155aa; S114A/L150D) variants are less effective in activating the FGFR1 receptor, which results in a weaker signal transduction and lower phosphorylation of Akt/PKB and ERK kinases (FIG. 19).
[0177] In the glucose uptake assay, a key assay in the context of antidiabetic activity of the FGF-1 protein, FGF1(155aa; L150D), FGF1(155aa; S114A/L150D) variants according to the invention do not induce glucose uptake by mouse adipocytes, even with the highest dose of 1000 ng/ml (FIG. 14). They do not show antidiabetic effectiveness without stabilizing mutations; that is why variants with multiple point mutations were obtained, in particular with five point mutations, both with two mutations reducing mitogenicity (S114A/L150D) and three stabilizing mutations (Q55P/S62I/H108G), which show antidiabetic effectiveness.
Example 7
[0178] Activity of FGF-1 muteins with multiple mutations. FGF1(155aa; Q55P/S62I/H108G/S114A/L150D) and FGF1(155aa; Q55P/S62I/H108G/S114A/L150D) variants
[0179] This protein combines all the desired properties: it is stable (denaturation temperature is above 60 C.), it has decreased mitogenicity compared to the wild-type. Wild-type FGF1 protein, even if not stabilized by an addition of heparin to the solution, stimulates proliferation of NIH 3T3 cells. FGF1(155aa; Q55P/S62I/H108G/S114A/L150D) variant at the studied concentrations does not show such an effect, irrespective of the content of heparin or lack thereof (FIG. 10). Moreover, the mutants show antidiabetic activity, both in vitro (in mouse 3T3-L1 cells differentiated to adipocytes) (FIGS. 15, 16, 17) and in vivo. The effect of reduced blood glucose in db/db mice was maintained for the subsequent 30 h after administration of FGF1(155aa; Q55P/S62I/H108G/S114A/L150D) variant and for 48 h after administration of the wild-type protein. Mutated protein is effective, but its activity is shorter than that of the wide-type protein (administered in the same dose to the animals) This results probably from the mechanism of action of FGF1 and the role of FGFR1 receptor in the effect of reducing blood glucose (the mutant has an impaired binding to the FGFR1 due to the presence of L150D mutation). An increase in a dose of mutein to 5 mg/kg m.c. results in achievement of activity time comparable to the wild-type FGF1 administered in a 1 mg/kg m.c. dose. The results of conducted antidiabetic activity assays are presented herein (FIGS. 24,26, 27).
Example 8
[0180] Modifications of the full-length FGF1(155aa; Q55P/S62I/H108G/S114A/L150D) variants FGF1(155aa; Q55P/S62I/H108G/S114A/L150D) variant developed on the full-length FGF1 (155 aa) sequence and the full-length wild-type FGF1(155aa) variant were analyzed to find out whether the 19 aa fragment at N-terminus of the protein has an influence on the biological activity of the protein and its muteins.
[0181] The N-terminal deletion of E3-G21 amino acids in the full-length sequence did not change the effect of glucose reduction in vivo (FIG. 24) and the effect was similar to that in the case of analogous full-length muteins (FIG. 15).
Example 9
[0182] Dimeric proteins of the truncated FGF1_DIMER(155aa) variant and the full-length FGF1_DIMER(155aa) variant
[0183] Two molecules of FGF1 were connected by an amino-acid linker in a standard manner. Dimerization was performed in the case of both short and full-length protein sequences. The activity of all these forms is similar in the in vitro conditions (FIG. 18), but in vivo the dimeric FGF1_DIMER(155aa) version was less active (FIG. 25).
Example 10
[0184] The results showing the activity of the FGF1 protein variants with single mutation: FGF1(155aa; S114A), FGF1(155aa; L150D) and double mutation: FGF1(155aa; S114A/L150D) FGF1(155aa; S114A) and FGF1(155aa; L150D) mutations limit the cell mitogenic activity. The data are presented for NIH 3T3 cells in the MTT assay: both mutation FGF1(155aa; S114A) and FGF1(155aa; L150D) as well as their combination FGF1(155aa; S114A/L150D) reduce proliferation compared to the wild-type protein, i.e. they decrease the mitogenic potential of the mutated protein (FIGS. 2, 3, 4, 6). Interestingly, the most decreased mitogenic effect was observed for the additionally stabilized protein i.e. one additionally comprising the stabilizing mutation according to the invention FGF1(155aa; Q55P/S62I/H108G/S114A/L150D) (FIGS. 6, 7, 8, 10). Additionally, such stabilized FGF1 protein (155aa; Q55P/S62I/H108G/S114A/L150D) exhibits antidiabetic activity, contrary to the variants with single mutations: FGF1(155aa; S114A), FGF1(155aa; L150D) and double mutations FGF1(155aa; S114A/L150D) (FIGS. 21, 22, 24). In the experiment for signal cascade responsible for triggering a proliferation path, FGF1(155aa; S114A), FGF1(155aa; L150D) and FGF1(155aa; S114A/L150D) variants are less effective in activating the FGFR1 receptor, which results in a weaker signal and lower phosphorylation of Akt/PKB and ERK kinases (FIG. 19).
[0185] In the glucose uptake assay, a key assay in the context of antidiabetic activity of the FGF1 protein, FGF1(155aa; S114A), FGF1(155aa; L150D) and FGF1(155aa; S114A/L150D) variants according to the invention do not induce glucose uptake by mouse adipocytes, even in the highest dose of 1000 ng/ml (FIGS. 12, 13, 14, 16). They do not show antidiabetic effectiveness without stabilizing mutations; that is why variants with multiple point mutations were obtained, in particular with five point mutations, both with two mutations reducing mitogenicity (S114A/L150D) and three stabilizing mutations (Q55P/S62I/H108G), which, unexpectedly, additionally show antidiabetic activity.
Example 11
[0186] Activity of FGF1 muteins with single mutations at S153 position of the hFGF1 The inventors studied the activity of developed muteins comprising a point mutation located at position S153 in a domain responsible for FGF1 ligand affinity to the receptor, including combinations with the stabilizing mutations according to the disclosure (Q55P, S62I and H108G). The results of studies carried out have showed that the developed muteins at position S153 are characterized by both decreased mitogenicity and the effect of reduced blood glucose level (S153A and S153R-FIGS. 7, 8, 17). Moreover, the multiple variants additionally comprising three stabilizing mutations are characterized by a high thermal stability, resistance to proteolytic degradation and, as a result, a longer half-life in an organism.
