Fusion protein containing an EGFR-binding domain and a masking domain

Abstract

A fusion protein is provided that includes an EGFR binding domain, a masking domain, and a linker that joins the masking domain to the EGFR binding domain.

Description

[Technical field]

[0001] The present disclosure relates to the selective targeting of EGFR expressed on tumor cells. [Background technology]

[0002] In cancer, overexpression of oncogenic receptors is common, providing an opportunity to design tumor-selective drugs. However, in many cases, these receptors are also abundantly expressed in healthy tissues, limiting therapeutic potential in terms of safety and efficacy. The epidermal growth factor receptor (EGFR or ErbB-1) is a tyrosine kinase receptor involved in the tumorigenesis of many cancers, often due to overexpression. Therapeutic targeting of this receptor is characterized by the challenge of systemic toxicity due to its abundant endogenous expression. Designing drugs that are highly active within the tumor microenvironment, but less active in the circulation and in healthy tissues, has the potential to significantly improve toxicity profiles and efficacy, allowing the selection of more invasive procedures that would otherwise be limited by systemic toxicity. Summary of the Invention

[0003] Affibody molecules are small (58 amino acid residues, 6.5 kDa), trihelical affinity proteins that are promising alternatives to antibody-based drugs due to their efficient tissue penetration, high stability, simple modularity of functional domains, and ease of production in prokaryotic hosts. Novel binding specificities arise from large combinatorial libraries by randomization of surface-exposed residues on helices 1 and 2 of the affibody molecule (Figure 1). Common selection platforms include phage and cell surface display, the latter offering the possibility to isolate binding populations and distinguish closely related affinities using fluorescence-assisted cell sorting (FACS). Displayed on the gram-positive Staphylococcus carnosus, 10 9Affibody libraries with variant sizes (Lofblom et al. Appl. Microbiol. Biotechnol. 2017, 101(23-24), 8293-8307) have been used previously to generate novel affibody binders. The libraries were synthesized using trinucleotide codons to avoid sequence bias and a stop codon. The system utilizes the XM cell wall anchoring sequence from Staphylococcus aureus protein A to display affibody libraries. Cells displaying affibody variants are incubated with soluble fluorescently labeled targets of interest and subsequently sorted using FACS based on fluorescent signal. To normalize cell surface affibody expression using differentially labeled fluorescent albumins, the displayed protein constructs contain two albumin binding domains to provide a linear correlation between affinity and fluorescent signal. Magnetic-assisted cell sorting (MACS) can be used prior to FACS as a pre-enrichment step to reduce the complexity of the library using magnetic beads with immobilized targets to capture cells displaying binding affibody variants.

[0004] The present disclosure is based on the development of conditionally activated affibody-based prodrugs targeting EGFR. To achieve conditional activation, Staphylococcus carnosus cell surface display was used to select anti-idiotypic affibody molecules that mask the binding interface of existing EGFR-targeting affibody molecules (ZEGFR:2377, see Friedman et al., J. Mol. Biol. 2008, 376(5), 1388-1402; Tolmachev et al. Eur. J. Nucl. Med. Mol. Imaging 2010, 37(3), 613-622 and WO 2007 / 065635). ZEGFR:2377 is hereinafter referred to as "ZEGFR".

[0005] A masking domain ("ZB05") was isolated that showed high binding propensity in flow cytometry (FC) on the surface of S. carnosus. The ZB05 affibody was produced as a soluble monomer and characterized for binding to ZEGFR and thermal stability. Rapid association and dissociation rates were observed from kinetic evaluation using SPR, favoring the utility of the masking domain. Furthermore, the protein showed high thermostability (T m 64.1°C) and the ability to refold after heat denaturation, a typical trait seen in monomeric affibody molecules.

[0006] The binding contributions of various residues in ZB05 were investigated by mutation. The results of the mutagenesis studies revealed flexibility and interchangeability of residues at several randomized positions with respect to retained binding (see Figure 10).

[0007] To further explore the potential of ZB05 as a masking domain, we designed a proof-of-concept pro-affibody (POC-PA) construct in which ZB05 was fused to ZEGFR-ABP (ABP stands for "albumin-binding protein") using a TEV-cleavable linker. We first analyzed its ability to mask EGFR binding using FC in the presence of fluorescently labeled soluble EGFR as a displayed protein on staphylococcal cells. No binding to EGFR could be observed in cells displaying the intact protein, but binding to EGFR was restored after treatment of the cells with TEV-protease.

[0008] The construct was produced as a soluble molecule and evaluated for cleavage by TEV protease. After 1 hour of incubation with TEV protease, the protein was completely digested and showed a clear band of the correct size on an SDS-page gel. Binding to recombinant immobilized EGFR using surface plasmon resonance (SPR) was similarly masked by ZB05 for intact POC-PA and restored for cleaved POC-PA. Equal injection amounts of intact and cleaved POC-PA were confirmed using separate surfaces immobilized with human serum albumin.

[0009] For binding to endogenously expressed EGFR, intact and truncated POC-PA were tested against H292 human mucoepidermoid lung carcinoma cells and A431 human squamous cell carcinoma cells, which express moderate to high levels of EGFR, respectively. Binding to cells alone could be observed with intact POC-PA as well as truncated POC-PA. Nevertheless, truncated POC-PA increased the signal for both cell lines and was comparable to the construct POC-PA-DM, in which ZB05 was exchanged for a dummy masking domain that has no specificity for ZEGFR.

[0010] Using radioactive labels, 111 The ability of [In]In-labeled prodrugs and their variants with non-cleavable linkers ("dummy linkers") and unmasked controls to target EGFR-expressing matriptase-positive cancer cells in vivo without binding to EGFR-expressing hepatocytes was evaluated. 111 For labeling with In, a maleimide derivative of the DOTA-chelator was conjugated to the C-terminus of all constructs. This site-specific approach ensures that the radiolabeling reflects the distribution of the ZEGFR-ABD035 fusion. Furthermore, site-specific labeling provides homogeneously labeled proteins rather than a mixture of proteins with different numbers of conjugated chelators at different positions. All radiolabeled constructs had high radiochemical purity and showed excellent stability. In vitro studies showed [111 It was demonstrated that binding of the In]In-labeled unmasked control was significantly reduced by saturation of the receptor using unlabeled ZEGFR-ABD035 fusion and cetuximab, indicating that binding was specific. 111 In vitro binding of [In]In-labeled prodrugs and dummy linkers was much lower and primarily nonspecific, suggesting that incorporation of the anti-idiotype masking domain effectively prevents binding of these constructs to EGFR in vitro.

[0011] [ 111 Initial in vivo evaluation of [In]In-labeled prodrug, dummy linker and unmasked controls demonstrated that the concentrations of all these proteins in blood were significantly higher than those of non-ABD035 fusion proteins. 111 It was demonstrated that the concentration of ABD035 in the liver was much higher than that of In-ZEGFR and the renal uptake was much lower. This phenomenon demonstrates the presence of binding of ABD035 to mouse albumin in vivo. This prevents glomerular filtration of the constructs and their reabsorption in the proximal tubules. Thus, the bioavailability of the constructs is higher and the potential renal toxicity is lower than that of non-ABD035 fused ZEGFR based targeting constructs. The most important observation from this experiment concerns the hepatic uptake. The uptake of the unmasked control in the liver was high, 17.2±1 and 12.7±1.8% ID / g at 4 and 24 hours after injection, respectively. This is expected due to the prominent expression of EGFR on hepatocytes and the high affinity of ZEGFR to mouse EGFR. The prodrug and dummy linker had much lower hepatic uptake, indicating that the incorporation of the masking domain served its purpose.

[0012] Data from studies in mice bearing EGFR-expressing H292 xenografts demonstrated extended residence times in the circulation for all constructs and 111 Low hepatic uptake of the In-labeled prodrug and dummy linker was confirmed. Importantly, the [ 111The uptake of the In]In-prodrug was significantly higher than in the EGFR-negative Ramos xenografts. This was also confirmed by imaging experiments, demonstrating that the uptake in H292 xenografts was EGFR-specific. Another interesting finding was that the tumor uptake was equally high for the prodrug and the dummy linker.

[0013] [ 111 Additional experiments were performed to elucidate the role of matriptase in tumor uptake of [In]In-labeled prodrugs. Tumor uptake was compared in mice bearing H292 and A431 xenografts simultaneously. Results of these experiments demonstrated that uptake in H292 xenografts with high matriptase expression was twice as high as that in A431 xenografts with low matriptase expression. At the same time, EGFR expression per cell was three times higher in A431 cells than in H292 cells. Taken together, these data suggest that matriptase mediates tumor uptake of [In]In-labeled prodrugs. 111 These results suggest that the EGFR-dependent uptake of the In[In]In-labeled prodrug may play a role in tumor uptake. At the same time, the uptake in A431 xenografts was higher than that in EGFR-negative Ramos xenografts, suggesting that a matriptase-independent mechanism is also at work.

[0014] In conclusion, the data show that the masked EGFR binder has a significantly lower liver uptake than the unmasked version, and interestingly, the tumor uptake of the masked EGFR binder was at the same level as that of the unmasked version, regardless of the protease-cleavable linker. Thus, the masking domain significantly improves the tumor-to-liver ratio.

[0015] Accordingly, the following itemized listing of embodiments of the present disclosure is provided: 1. A fusion protein comprising an EGFR binding domain, a masking domain, and a linker connecting the masking domain to the EGFR binding domain, The masking domain is represented by the amino acid sequence IX 10 SX12 X 13 X 14 X 15 X 16 WWX 19 X 20 X 21 X 22 X 23 X 24 X 25 X 26 KX 28 X 29 X 30 X 31 YX 33 X 34 Contains V, Independently of each other, X 10 is R, K, M, N or Q; X 12 is A or a substitution, X 13 is E or non-existent, X 14 is T or S, X 15 is E or a substitution, X 16 is I or a substitution, X 19 is L, substituted or absent; X 20 is P, a substitution or absence; X 21 is N, substituted or absent; X 22 is L, substituted or absent; X 23 is T or a substitution, X 24 is A, F, I, K, L, M, T or Y, X 25 is D, G, I or W; X 26 is Q or a substitution, X 28 is W, A, F, I, L, M, Q, R, S, T or V, X 29 is A or a substitution, X 30is F or a substitution, X 31 is I or L, X 33 is K or a substitution, and X 34 is L or a substitution, However, X 12 , X 15 , X 16 , X 19 , X 20 , X 21 , X 22 , X 23 , X 26 , X 29 , X 30 , X 33 and X 34 Six or fewer of X are substitutions, 19 ~X 22 are absent, and The EGFR binding domain has the amino acid sequence EX2X3X4AX6X7EIX 10 X 11 LPNLNX 17 X 18 QX 20 X 21 AFIX 25 SLX 28 Contains D, Independently of each other, X2 is M, F, V, L, I or S; X3 is W, D, E or L; X4 is I, V, G, S, M, L, A, T, N, D or W; X6 is W, V, L, I, M or S; X7 is D, E, N or K; X 10 is R, G, H or K, X 11 is D, N, E, Y or S; X 17 is G, W or A, X 18 is W, G or A; X 20 is M, L, F, A or E, X21 is T, D, N, A or Q; X 25 is A, S, N, G or L, and X 28 is L, W, V, F or A; Fusion proteins. 2. In the masking domain, X 12 , X 15 , X 16 , X 19 , X 20 , X 21 , X 22 , X 23 , X 26 , X 29 , X 30 , X 33 and X 34 and / or X 19 ~X 22 2. The fusion protein of item 1, wherein one or more of: 3. In the masking domain, X 12 , X 15 , X 16 , X 19 , X 20 , X 21 , X 22 , X 23 , X 26 , X 29 , X 30 , X 33 and X 34 and / or X 19 ~X 22 3. The fusion protein of item 2, wherein none of 4. The masking domain has the amino acid sequence X5X6X7X8IX 10 SX 12 X 13 X 14 X 15 X 16 WWX 19 X 20 X 21 X 22 X 23 X 24 X 25 X 26 KX 28 X29 X 30 X 31 YX 33 X 34 Contains V, Independently of each other, X5 is Y or a substitution; X6 is A or a substitution; X7 is K or a substitution; and X8 is E or a substitution; 4. The fusion protein according to any one of items 1 to 3, wherein no more than two of X5 to X8 are substituted. 5. The masking domain has the amino acid sequence X1X2X3X4X5X6X7X8IX 10 SX 12 X 13 X 14 X 15 X 16 WWX 19 X 20 X 21 X 22 X 23 X 24 X 25 X 26 KX 28 X 29 X 30 X 31 YX 33 X 34 Contains V, Independently of each other, X1 is V or a substitution; X2 is D or substituted; X3 is A or a substitution; X4 is K or a substitution; 5. The fusion protein according to item 4, wherein no more than four of X1 to X8 are substituted. 6. The fusion protein of item 5, wherein in the masking domain, no more than two of X1 to X8 are substitutions. 7. The masking domain is an amino acid sequence of IX 10 SX 12 X 13 X 14 X 15 X 16 WWX 19 X 20 X 21 X22 X 23 X 24 X 25 X 26 KX 28 X 29 X 30 X 31 YX 33 X 34 VX 36 X 37 X 38 X 39 X 40 X 41 X 42 X 43 X 44 Including, Independently of each other, X 36 is D or a substitution, X 37 is D or a substitution, X 38 is P or a substitution, X 39 is S or a substitution, X 40 is Q or a substitution, X 41 is S or a substitution, X 42 is S or a substitution, X 43 is E or a substitution, X 44 is L or a substitution, However, X 36 ~X 44 7. The fusion protein of any one of items 1 to 6, wherein no more than four of the amino acids are substitutions. 8. In the masking domain, X 36 ~X 44 8. The fusion protein of any one of items 1 to 7, wherein no more than two of the following are substitutions: 9. The masking domain is an amino acid sequence of IX 10 SX 12 X 13 X 14 X 15 X 16 WWX 19 X 20 X 21 X22 X 23 X 24 X 25 X 26 KX 28 X 29 X 30 X 31 YX 33 X 34 VX 36 X 37 X 38 X 39 X 40 X 41 X 42 X 43 X 44 X 45 X 46 X 47 X 48 X 49 X 50 X 51 X 52 X 53 X 54 Including, Independently of each other, X 45 is L or a substitution, X 46 is S or a substitution, X 47 is E or a substitution, X 48 is A or a substitution, X 49 is K or a substitution, X 50 is K or a substitution, X 51 is L or a substitution, X 52 is N or a substitution, X 53 is D or a substitution, X 54 is S or a substitution, However, X 45 ~X 54 9. The fusion protein of item 7 or 8, wherein no more than five of the residues are substitutions. 10. In the masking domain, X 36 ~X 54 10. The fusion protein of item 9, wherein seven or fewer of the residues are substitutions. 11. In masking domains, X 36 ~X 54 11. The fusion protein of item 10, wherein no more than five of the residues are substitutions. 12. The masking domain is selected from the group consisting of the amino acid sequence IX 10 SX 12 X 13 X 14 X 15 X 16 WWX 19 X 20 X 21 X 22 X 23 X 24 X 25 X 26 KX 28 X 29 X 30 X 31 YX 33 X 34 VX 36 X 37 X 38 X 39 X 40 X 41 X 42 X 43 X 44 X 45 X 46 X 47 X 48 X 49 X 50 X 51 X 52 X 53 X 54 X 55 X 56 X 57 X 58 Including, Independently of each other, X 55 is Q or a substitution, X 56 is A or a substitution, X 57 is P or a substitution, X 58 is K or a substitution, However, X 55 ~X 58 12. The fusion protein of any one of items 9 to 11, wherein no more than two of the following are substitutions: 13. In masking domains, X 36 ~X 58 13. The fusion protein of item 12, wherein seven or fewer of the residues are substitutions. 14. In masking domains, X 36 ~X 58 14. The fusion protein of item 13, wherein five or fewer of the following are substitutions: 15. In masking domains, X 10 15. The fusion protein of any one of items 1 to 14, wherein 16. In masking domains, X 13 16. The fusion protein of any one of items 1 to 15, wherein 17. In masking domains, X 14 17. The fusion protein of any one of items 1 to 16, wherein 18. In masking domains, X 24 18. The fusion protein of any one of items 1 to 17, wherein 19. In masking domains, X 25 19. The fusion protein of any one of items 1 to 18, wherein 20. In masking domains, X 28 is W. 21. In masking domains, X 31 The fusion protein of any one of items 1 to 20, wherein 22. The masking domain has the amino acid sequence IRSX 12 TX 15 X 16 WWX 19 X 20 X 21 X 22 X 23 ADX 26 KWX 29 X 30 IYX 33 X 34 22. The fusion protein of any one of items 1 to 21, comprising V. 23. The fusion protein of any one of items 1 to 22, wherein the masking domain comprises the amino acid sequence IRSATEIWWLPNLTADQKWAFIYKLV (SEQ ID NO: 1). 24. The fusion protein of any one of items 1 to 23, wherein in the EGFR binding domain, X3 is W. 25. The fusion protein of any one of items 1 to 24, wherein in the EGFR binding domain, X6 is V or W. 26. In the EGFR binding domain, X 10 is R or G. 27. In the EGFR binding domain, X 17 is W or G. 28. In the EGFR binding domain, X 18 is W or G, preferably W. 29. In the EGFR binding domain, X 21 is T or D, preferably T. 30. The EGFR binding domain has the amino acid sequence EX2WX4AWX7EIRX 11 LPNLNGWQX 20 TAFIX 25 SLX 28 Contains D, Independently of each other, X2 is M, V, L or I; X4 is I, V, G, S, M, L, A, T, N or D; X7 is D, E, N or K; X 11 is D, N, E, Y or S; X 20 is M, L or F; X 25 is A, S or G, and X 28 is L or V. 31. The fusion protein of any one of items 1 to 30, wherein in the EGFR binding domain, X2 is M. 32. The fusion protein of any one of items 1 to 31, wherein in the EGFR binding domain, X4 is I, V, G or S. 33. In the EGFR binding domain, X 11 is D, N or E. 34. In the EGFR binding domain, X 20 The fusion protein of any one of items 1 to 33, wherein 35. In the EGFR binding domain, X 25 is A or S. 36. In the EGFR binding domain, X 28 is L. 37. In the EGFR binding domain, X2 is M and X 20 is M and X 28 is L. 38. The fusion protein of any one of items 1 to 37, wherein the EGFR-binding domain comprises the amino acid sequence EMWIAWEIRDLPNLNGWQMTAFIASLLD (SEQ ID NO: 2). 39. The fusion protein of any one of items 1 to 38, wherein the EGFR-binding domain comprises the amino acid sequence VDNKFNKEMWIAWEEIRDLPNLNGWQMTAFIASLLD (SEQ ID NO: 3). 40. The fusion protein of any one of items 1 to 39, wherein the EGFR-binding domain comprises the amino acid sequence EMWIAWEEIRDLPNLNGWQMTAFIASLLDDPSQSANLLAEAKKLNDAQAPK (SEQ ID NO: 4). 41. The fusion protein of any one of items 1 to 40, wherein the EGFR binding domain comprises the amino acid sequence VDNKFNKEMWIAWEEIRDLPNLNGWQMTAFIASLLDDPSQSANLLAEAKKLNDAQAPK (SEQ ID NO: 5). 42. The fusion protein of any one of items 1 to 41, further comprising a half-life extending region, such as an Fc binding region or an albumin binding region (ABR). 43. The fusion protein of any one of claims 1 to 42, further comprising an albumin binding region (ABR) comprising an amino acid sequence selected from the following: a) LAX3AKX6X7AX9X 10 ELDX 14 YGVSDX 20 YKX 23 LIX 26 X 27 AKT VEGVX 35 ALX 38 X 39 X 40 ILX 43 X 44 X 45 X 46 (where, independently of each other, X3 is selected from E, S, Q and C; X6 is selected from E, S, V and C; X7 is selected from A, L and S; X9 is selected from L and N; X 10 is selected from A, S and R; X 14 is selected from A, S, C and K; X 20 is selected from Y and F; X 23 is selected from N, D and R; X 26 is selected from N, D and E; X 27 is selected from N and K; X 35 is selected from K and E; X 38 is selected from I and K, X 39 is selected from D, E and L; X 40 is selected from A, E and H; X 43 is selected from A and K; X 44 is selected from A, S and E; X 45 is L or non-existent, X 46 is P or non-existent) and b) an amino acid sequence having at least 95% identity to the sequence defined in a). 44. ABR has the amino acid sequence LAX3AKX6X7AX9X 10 ELDX 14 YGVSDX 20 YKX 23 LIX 26 X 27 AKT VEGVX 35 ALX 38 X 39 X 40 Including ILAALP, and independently of each other, X3 is selected from E and S; X6 is selected from E and V; X7 is selected from A and L; X9 is selected from L and N; X 10 is selected from A and R; X 14 is selected from A, S, C and K, preferably from A, S and K; X 20 is selected from Y and F; X 23 is selected from N, D and R; X 26 is selected from N and D; X 27 is selected from N and K; X 35 is selected from K and E; X 38 is selected from I and K, X 39 is selected from D and L; X 40 is selected from A, E and H; X 45 is L or non-existent, X 46 The fusion protein of item 43, wherein A is P or is absent. 45. The therapeutic conjugate according to item 43, wherein the ABR comprises an amino acid sequence selected from the group consisting of: LAEAKVLANR ELDKYGVSDF YKRLINKAKT VEGVEALKLH ILAALP (SEQ ID NO:6); LAEAKEAANA ELDSYGVSDF YKRLIDKAKT VEGVEALKDA ILAALP (SEQ ID NO: 7); GLAEAKEAAN AELDSYGVSD FYKRLIDKAK TVEGVEALKD AILAALP (SEQ ID NO: 8); LAEAKVLANR ELDKYGVSDY YKNLINNAKT VEGVKALIDE ILAALP (SEQ ID NO: 9); and LAEAKVLALR ELDKYGVSDY YKDLIDKAKT VEGVKALIDE ILAALP (SEQ ID NO: 10). 46. ​​The fusion protein of any one of items 42 to 45, wherein the half-life extending region is located C-terminal to the EGFR binding domain. 47. The fusion protein of any one of items 1 to 46, wherein the linker is a protease-cleavable linker. 48. The fusion protein of item 47, wherein the protease-cleavable linker comprises a sequence selected from the group consisting of GFLG (SEQ ID NO:11), Glutamic acid-Valine-Citrulline, GILGVP (SEQ ID NO:13), GPLGIAGQ (SEQ ID NO:14), VHMPLGFLGP (SEQ ID NO:15), SGGPGPAGMKGLPGS (SEQ ID NO:16), PLGLAG (SEQ ID NO:17), LALGPG (SEQ ID NO:18), KRALGLPG (SEQ ID NO:19), GGGRR (SEQ ID NO:20), LSGRSDNH (SEQ ID NO:21), PMAKK (SEQ ID NO:22), RQARVVNG (SEQ ID NO:23), MSGRSANA (SEQ ID NO:38), HSSKLQL (SEQ ID NO:24) and RRSSYYSG (SEQ ID NO:25). 49. The fusion protein of any one of items 1 to 48, wherein the length of the linker is at least 12 amino acid residues, for example 12 to 60 amino acid residues, for example 20 to 50 amino acid residues. 50. The fusion protein of any one of items 1 to 49, wherein the linker does not contain a cysteine ​​(C) residue. 51. The fusion protein of any one of items 1 to 50, wherein the masking domain does not contain a cysteine ​​(C) residue. 52. The fusion protein of any one of items 1 to 51, wherein the EGFR-binding domain does not contain a cysteine ​​(C) residue. 53. The fusion protein of any one of items 1 to 52, wherein the masking domain is located N-terminal to the EGFR binding domain. 54. A therapeutic conjugate comprising the fusion protein of any one of items 1 to 53 and a cytotoxic agent, such as a cytotoxic molecule, a cytotoxic peptide, a cytotoxic protein or a cytotoxic radionuclide. 55. A cytotoxic agent is 177 Lu, 90 Y, 188 Re; 186 Re; 166 Ho, 153 Sm, 67 Cu, 64 Cu, 149 Tb, 161 Tb, 47 Sc; 225 Ac; 212 Pb; 213 Bi, 212 Bi, 227 Th, 223 Ra; 58m Co, 131 I, 76 As, 77 As and 211 The therapeutic conjugate according to item 54, wherein the cytotoxic radionuclide is selected from the group consisting of At. 56. The therapeutic conjugate according to item 54, wherein the cytotoxic agent is a cytotoxic molecule selected from the group consisting of doxorubicin (DOX), duocarmycin (DUO), docetaxel (DTX), monomethylauristatin E (MMAE), monomethylauristatin F (MMAF), paclitaxel (PTX), mertansine (DM1), emtansine (DM1), ravtansine (DM4), soravtansine (DM4), pyrrolobenzodiazepines (PBDs) and calicheamicin. 57. The therapeutic conjugate according to item 54, wherein the cytotoxic agent is a proteinaceous toxin selected from the group consisting of Pseudomonas exotoxin (PE), diphtheria toxin (DT), ricin toxin A chain (RTA) and deBouganin. 58. A therapeutic conjugate according to any one of items 54 to 57 for use in a therapeutic treatment method. 59. The therapeutic conjugate for use according to item 58, wherein the therapeutic treatment method is a method of treating a subject suffering from a cancer, such as a cancer that overexpresses EGFR. 60. The therapeutic conjugate for use according to item 59, wherein the cancer is selected from the group consisting of lung cancer, preferably non-small cell lung cancer, prostate cancer, breast cancer, colon and rectal cancer, head and neck cancer, esophagogastric cancer, liver cancer, glioblastoma, cervical cancer, ovarian cancer, bladder cancer, renal cancer and pancreatic cancer. [Brief description of the drawings]

[0016] [Figure 1] 1 is a schematic diagram of an affibody molecule showing the 14 surface-exposed amino acid positions that were randomized in the Staphylococcal display library.

[0017] [Diagram 2] Schematic diagram of the proof-of-concept pro-affibody (POC-PA) construct. The POC-PA construct comprises an anti-idiotype masking domain with specificity for an EGFR-binding affibody molecule, a TEV protease cleavable linker, an EGFR-binding affibody domain, and an albumin binding protein.

[0018] [Diagram 3] Characterization of the ZB05 affibody masking domain: (A) melting curves (VTM) (left) and circular dichroism spectroscopy (right) showing the refolding ability of ZB05 after thermal denaturation; (B) SPR sensorgrams showing the binding interaction of B05 with immobilized ZEGFR (left) and negative control human serum albumin (HSA) (right).

[0019] [Figure 4] Characterization of POC-PA produced as a soluble monomer: (A) SDS-page showing purity after HSA affinity purification and size of TEV-protease and cleavage products after treatment with TEV-protease; and (B) analysis of protease-dependent binding of POC-PA to immobilized human EGFR (top) and unaffected binding to immobilized HSA (bottom) using SPR.

[0020] [Diagram 5] Flow cytometry analysis of EGFR binding on H292 and A431 cells for intact POC-PA and POC-PA precleaved by TEV protease. A control construct (POC-PA-DM) with a dummy masking domain (ZE01) was included for comparison.

[0021] [Figure 6] Binding specificity of [111In]In-labeled prodrug, dummy linker and unmasked control in (A) A431 cells and (B) H292 cells. Cells were incubated at room temperature with radiolabeled compounds with or without presaturation using a 100-fold molar excess of unmasked control without label. Y-axis corresponds to measured total activity of cells as a percentage of total activity added to each well. Asterisk (*) corresponds to significant difference (p<0.05, t-test).

[0022] [Figure 7] Comparison of [111In]In-labeled unmasked control binding to H292 and A431 cells in vitro. Cells were incubated at room temperature with radiolabeled compound (1.64 x 106 CPM) with or without pre-saturation using a 100-fold molar excess of the same compound, blocked or unblocked, respectively, lacking the radiolabel or the anti-EGFR antibody cetuximab. An additional set of dishes was used to count the number of cells per dish at the time of the experiment.

[0023] [Figure 8] (A) Comparison of [111In]In-prodrug uptake in H292 (EGFR positive) and Ramos (EGFR negative) xenografts at 4 hours pi and (B) 48 hours pi (t-test); (C) Comparison of uptake in H292 (EGFR positive, high matriptase levels) and A431 (EGFR positive, low matriptase levels) xenografts in the same animal at 48 hours pi (t-test). Symbols (x) indicate uptake in H292 xenografts; symbols (+) indicate uptake in Ramos xenografts; symbols (●) indicate uptake in A431 xenografts.

[0024] [Figure 9] (A) Micro-Single-Photon Emission Computed Tomography / Computed Tomography (microSPECT / CT) imaging of Balb / c nu / nu mice with EGFR-positive H292 xenografts at 4 hours pi and (C) 48 hours pi with [111In]In-labeled unmasked control (left) and prodrug (right). (B) Micro-Single-Photon Emission Computed Tomography / Computed Tomography (microSPECT / CT) imaging of Balb / c nu / nu mice with EGFR-negative Ramos xenografts at 4 hours pi and (D) 48 hours pi with [111In]In-labeled prodrug. T arrow points to tumor. L arrow points to liver.

[0025] [Figure 10]1 is a table showing the substitutions allowed at each randomized position of the ZB05 scaffold that retain binding to ZEGFR. Allowed substitutions that retained binding were determined from a 2-fold enrichment in the binding population compared to a naive library with at least 50% depletion for the corresponding variant in the non-binding population. The randomized positions of the ZB05 sequence are marked by arrows.

[0026] [Figure 11] A representative FACS sort of the ZB05 mutagenesis library containing a total of 253 different variants (left) and flow cytometry analysis of the binding population (right) are shown.

[0027] [Figure 12] The complete amino acid sequence of the proof-of-concept pro-affibody (POC-PA) prepared and tested in Example 1 is shown. In the masking domain ZB05, the subsequence specifically relevant for binding to ZEGFR (SEQ ID NO: 1) is highlighted in grey. In the TEV-cleavable linker, the subsequence recognized by TEV is highlighted in grey. In the ZEGFR (i.e., EGFR-binding domain), the subsequence specifically relevant for binding to EGFR (SEQ ID NO: 2) is highlighted in grey.

[0028] [Figure 13] The complete amino acid sequences of the (proof of concept) prodrugs used in Example 2 are shown. In the masking domain ZB05, the subsequence specifically relevant for binding to ZEGFR (SEQ ID NO:1) is highlighted in grey. In the matriptase cleavable linker, the subsequence recognized by matriptase (SEQ ID NO:21) is highlighted in grey. In the ZEGFR (i.e., EGFR binding domain), the subsequence specifically relevant for binding to EGFR (SEQ ID NO:2) is highlighted in grey.

[0029] [Figure 14] 4 shows an SDS-page gel after cleavage of PA as described in Example 3 below. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0030] As a first aspect of the present disclosure, there is provided a fusion protein comprising an EGFR-binding domain, a masking domain, and a linker connecting the masking domain to the EGFR-binding domain.

[0031] The masking domain binds to the EGFR binding domain, thereby restricting binding to EGFR under certain conditions.

[0032] The masking domain is the amino acid sequence IX 10 SX 12 X 13 X 14 X 15 X 16 WWX 19 X 20 X 21 X 22 X 23 X 24 X 25 X 26 KX 28 X 29 X 30 X 31 YX 33 X 34 Contains V, Independently of each other, X 10 is R, K, M, N or Q; X 12 is A or a substitution, X 13 is E or absent, X 14 is T or S, X 15 is E or a substitution, X 16 is I or a substitution, X 19 is L, substituted or absent; X 20 is P, a substitution or absence, X 21 is N, substituted or absent, X 22 is L, substituted or absent; X 23 is T or a substitution, X 24 is A, F, I, K, L, M, T or Y, X 25 is D, G, I or W, X 26 is Q or a substitution, X 28 is W, A, F, I, L, M, Q, R, S, T or V, X 29 is A or a substitution, X 30 is F or a substitution, X 31 is I or L, X 33 is K or a substitution, X 34 is L or a substitution.

[0033] position 12 , X 15 , X 16 , X 19 , X 20 , X 21 , X 22 , X 23 , X 26 , X 29 , X 30 , X 33 and X 34 The amino acid residues in are not believed to form part of a binding site that actively interacts with the EGFR binding domain. Therefore, at these positions, substitutions (compared to ZB05) and possibly substitutions (X 19 ~X 22 ) and even deletions are permitted. However, the three-dimensional structure may be modified by substitutions or (X 19 ~X 22 In the case of X, the binding ability is considered to be lost when the X is deleted. 12 , X 15 , X 16 , X 19 , X 20 , X 21 , X22 , X 23 , X 26 , X 29 , X 30 , X 33 and X 34 Up to six of the X's may be substitutions. 19 ~X 22 Up to two of may be absent.

[0034] In one embodiment of the masking domain, X 12 , X 15 , X 16 , X 19 , X 20 , X 21 , X 22 , X 23 , X 26 , X 29 , X 30 , X 33 and X 34 Preferably, no more than four of X 12 , X 15 , X 16 , X 19 , X 20 , X 21 , X 22 , X 23 , X 26 , X 29 , X 30 , X 33 and X 34 No more than two of the are substitutions.

[0035] In an alternative or complementary embodiment, X 19 ~X 22 Preferably, no more than one of X 19 ~X 22 Neither of these is non-existent.

[0036] In one embodiment, the masking domain comprises additional amino acid residues (X5 to X8) at the N-terminus. Thus, the masking domain has the amino acid sequence X5X6X7X8IX 10 SX 12 X 13 X 14 X 15 X 16WWX 19 X 20 X 21 X 22 X 23 X 24 X 25 X 26 KX 28 X 29 X 30 X 31 YX 33 X 34 V, Independently of each other, X5 is Y or a substitution; X6 is A or a substitution; X7 is K or a substitution; X8 is E or a substitution; However, two or less of X5 to X8 are substitutions.

[0037] In one embodiment, the masking domain comprises additional amino acid residues (X1-X4) at the N-terminus. Thus, the masking domain has the amino acid sequence X1X2X3X4X5X6X7X8IX 10 SX 12 X 13 X 14 X 15 X 16 WWX 19 X 20 X 21 X 22 X 23 X 24 X 25 X 26 KX 28 X 29 X 30 X 31 YX 33 X 34 V, Independently of each other, X1 is V or a substitution; X2 is D or a substitution; X3 is A or a substitution; X4 is K or a substitution; However, four or less, for example two or less, of X1 to X8 are substitutions.

[0038] In one embodiment, the masking domain comprises an additional amino acid residue (X 36 ~X 44 ) The masking domain therefore comprises the amino acid sequence IX 10 SX 12 X 13 X 14 X 15 X 16 WWX 19 X 20 X 21 X 22 X 23 X 24 X 25 X 26 KX 28 X 29 X 30 X 31 YX 33 X 34 VX 36 X 37 X 38 X 39 X 40 X 41 X 42 X 43 X 44 It may include, Independently of each other, X 36 is D or a substitution, X 37 is D or a substitution, X 38 is P or a substitution, X 39 is S or a substitution, X 40 is Q or a substitution, X 41 is S or a substitution, X 42 is S or a substitution, X 43 is E or a substitution, X 44 is L or a substitution, However, X 36 ~X 44 of which no more than four, e.g., no more than two, are substitutions.

[0039] In one embodiment, the masking domain further comprises an additional amino acid residue at the C-terminus (X 45 ~X 54 ) The masking domain therefore comprises the amino acid sequence IX 10 SX 12 X 13 X 14 X 15 X 16 WWX 19 X 20 X 21 X 22 X 23 X 24 X 25 X 26 KX 28 X 29 X 30 X 31 YX 33 X 34 VX 36 X 37 X 38 X 39 X 40 X 41 X 42 X 43 X 44 X 45 X 46 X 47 X 48 X 49 X 50 X 51 X 52 X 53 X 54 It may include, Independently of each other, X 45 is L or a substitution, X 46 is S or a substitution, X 47 is E or a substitution, X 48 is A or a substitution, X 49 is K or a substitution, X 50 is K or a substitution, X 51 is L or a substitution, X 52 is N or a substitution, X53 is D or a substitution, X 54 is S or a substitution, However, X 45 ~X 54 of which no more than five, e.g., no more than three, are substitutions.

[0040] In one embodiment of the masking domain, X 36 ~X 54 Of these, no more than seven, e.g., no more than five, are substitutions.

[0041] The masking domain is composed of four additional domains at the C-terminus (X 55 ~X 58 The masking domain can thus be further extended to the amino acid residues of the amino acid sequence IX 10 SX 12 X 13 X 14 X 15 X 16 WWX 19 X 20 X 21 X 22 X 23 X 24 X 25 X 26 KX 28 X 29 X 30 X 31 YX 33 X 34 VX 36 X 37 X 38 X 39 X 40 X 41 X 42 X 43 X 44 X 45 X 46 X 47 X 48 X 49 X 50 X 51 X 52 X 53 X 54 X 55 X 56 X 57 X 58 It may include, Independently of each other, X 55 is Q or a substitution, X 56 is A or a substitution, X 57 is P or a substitution, X 58 is K or a substitution, However, X 55 ~X 58 No more than two of the are substitutions.

[0042] In one embodiment of the masking domain, X 36 ~X 58 Of these, no more than seven, e.g., no more than five, are substitutions.

[0043] In a preferred embodiment of the masking domain, X 10 is R, X 13 is non-existent, X 14 is T, X 24 is A, X 25 is D, X 28 is W, and / or X 31 is I.

[0044] Thus, the masking domain has the amino acid sequence IRS 12 TX 15 X 16 WWX 19 X 20 X 21 X 22 X 23 ADX 26 KWX 29 X 30 IYX 33 X 34 It may contain V.

[0045] In one embodiment, the masking domain does not contain any cysteine ​​(C) residues.

[0046] In one embodiment, the masking domain comprises the amino acid sequence IRSATEIWWLPNLTADQKWAFIYKLV (SEQ ID NO:1), similar to ZB05.

[0047] In one embodiment, the masking domain comprises the amino acid sequence VDAKYAKEIRSATEIWWLPNLTADQKWAFIYKLVDDPSQSSELLSEAKKLNDSQAPK (SEQ ID NO: 26), which is the sequence of ZB05.

[0048] The EGFR binding domain of the fusion protein of the first aspect has the amino acid sequence EX2X3X4AX6X7EIX 10 X 11 LPNLNX 17 X 18 QX 20 X 21 AFIX 25 SLX 28 Contains D, Independently of each other, X2 is M, F, V, L, I or S; X3 is W, D, E or L; X4 is I, V, G, S, M, L, A, T, N, D or W; X6 is W, V, L, I, M or S; X7 is D, E, N or K; X 10 is R, G, H or K, X 11 is D, N, E, Y or S, X 17 is G, W or A, X 18 is W, G or A, X 20 is M, L, F, A or E, X 21 is T, D, N, A or Q, X 25 is A, S, N, G or L, X 28 is L, W, V, F or A.

[0049] The rationale behind this amino acid sequence (and the embodiments presented below) is explained in WO 2007 / 065635, which is incorporated herein by reference.

[0050] In a preferred embodiment of the EGFR binding domain, X3 is W, X6 is V or W; X 10 is R or G, X 17 is W or G, X 18 is W or G, preferably W, and / or X 21 is T or D, preferably T.

[0051] In one embodiment, the EGFR binding domain has the amino acid sequence EX2WX4AWX7EIRX 11 LPNLNGWQX 20 TAFIX 25 SLX 28 Contains D, Independently of each other, X2 is M, V, L or I, preferably M; X4 is I, V, G, S, M, L, A, T, N or D, preferably I, V, G or S; X7 is D, E, N or K; X 11 is D, N, E, Y or S, preferably D, N or E; X 20 is M, L or F, preferably M; X 25 is A, S or G, preferably A or S; X 28 is L or V, preferably L.

[0052] In a preferred embodiment of the EGFR binding domain, X2 is M and X 20 is M and X 28 is L.

[0053] In particularly preferred embodiments, the EGFR binding domain comprises an amino acid sequence selected from the following: (i) EMWIAWEIRDLPNLNGWQMTAFIASLLD (SEQ ID NO: 2); and (ii) an amino acid sequence having at least 90% identity, for example at least 95% identity, to a sequence defined in (i).

[0054] Optionally, additional N-terminal amino acid residues are present, such that the EGFR binding domain comprises an amino acid sequence selected from the following: (iii) VDNKFNKEMWIAWEEIRDLPNLNGWQMTAFIASLLD (SEQ ID NO: 3); and (iv) An amino acid sequence having at least 90% identity, for example at least 95% identity, to a sequence defined in (iii).

[0055] Optionally, additional C-terminal amino acid residues are present, such that the EGFR binding domain comprises an amino acid sequence selected from the following: (v) EMWIAWEEIRDLPNLNGWQMTAFIASLLDDPSQSANLLAEAKKLNDAQAPK (SEQ ID NO: 4); and (vi) An amino acid sequence having at least 90% identity, for example at least 95% identity, to a sequence defined in (v).

[0056] In one embodiment, additional amino acid residues are present at the N-terminus and C-terminus such that the EGFR binding domain comprises an amino acid sequence selected from the following: (vii) VDNKFNKEMWIAWEEIRDLPNLNGWQMTAFIASLLDDPSQSANLLAEAKKLNDAQAPK (SEQ ID NO: 5); and (viii) An amino acid sequence having at least 90% identity, for example at least 95% identity, to a sequence defined in (vii).

[0057] Sequence (vii) is the sequence of the EGFR binding affibody (also called ZEGFR) used in the Examples section below.

[0058] In one embodiment, the EGFR binding domain does not contain any cysteine ​​(C) residues.

[0059] In one embodiment of the fusion protein of the first aspect, the masking domain is located N-terminal to the EGFR binding domain. Thus, in this embodiment, the linker connects the C-terminus of the masking domain to the N-terminus of the EGFR binding domain.

[0060] In one embodiment of the fusion protein of the first aspect, the linker is a protease-cleavable linker.Preferably, such a linker is cleavable by one or more of the following proteases: cathepsin B; MMP-2; MMP-9; MMP-7; urokinase-type plasminogen activator; matriptase; and prostate-specific antigen (PSA).These proteases are found in the tumor microenvironment.

[0061] The above proteases recognize the cleavage sites listed below.

[0062] Cathepsin B GFLG (SEQ ID NO: 11; see reference [1]); and Glutamate-valine-citrulline (see references [2]).

[0063] MMP-2 / -9 GILGVP (SEQ ID NO: 13; see reference [1]); GPLGIAGQ (SEQ ID NO: 14; see reference [1]); VHMPLGFLGP (SEQ ID NO: 15; see reference [3]); SGGPGPAGMKGLPGS (SEQ ID NO: 16; see reference [4]); PLGLAG (SEQ ID NO: 17; see reference [5]); and LALGPG (sequence number 18; see reference [5]).

[0064] MMP-7: KRALGLPG (SEQ ID NO: 19; see reference [1])

[0065] Urokinase-type plasminogen activator GGGRR (SEQ ID NO: 20; see reference [1]); and LSGRSDNH (sequence number 21; see reference [6]).

[0066] Matriptase LSGRSDNH (SEQ ID NO:21; see references [3] and [8]); PMAKK (SEQ ID NO:22; see reference [3]); RQARVVNG (SEQ ID NO: 23; see reference [3]); and MSGRSANA (sequence number 38; see reference [7]).

[0067] Prostate-specific antigen (PSA) HSSKLQL (SEQ ID NO: 24; see reference [1]); and RRSSYYSG (sequence number 25; see reference [1]).

[0068] As a result, one embodiment of a protease-cleavable linker comprises at least one of these cleavage site sequences listed above.

[0069] References (proteases and cleavage sites) JPEG2025507294000002.jpg115160

[0070] To link the masking sequence and the EGFR binding sequence together, the length of the linker is typically at least 12 amino acid residues, for example at least 20 amino acid residues. The maximum length can be, for example, 50 or 60 amino acid residues.

[0071] In one embodiment, the linker does not contain any cysteine ​​(C) residues.

[0072] The fusion protein of the first aspect may further comprise a half-life extending region, such as an Fc binding region or an albumin binding region (ABR). The half-life extending region may, for example, be located C-terminal to the EGFR binding domain.

[0073] Alternatively, the half-life extending group is not part of the fusion protein, but is otherwise connected to the fusion protein. In this alternative embodiment, the fusion protein and the half-life extending group together form a construct that may include further moieties or groups.

[0074] Various strategies for extending the half-life of proteins are described in a review article by Kontermann (EXPERT OPINION ON BIOLOGICAL THERAPY, 2016 VOL. 16, NO. 7, 903-915).

[0075] In one embodiment, the fusion protein of the first aspect comprises an ABR comprising an amino acid sequence selected from: a) LAX3AKX6X7AX9X 10 ELDX 14 YGVSDX 20 YKX 23 LIX 26 X 27 AKT VEGVX 35 ALX 38 X 39 X 40 ILX 43 X 44 X 45 X 46 (where, independently of each other, X3 is selected from E, S, Q and C; X6 is selected from E, S, V and C; X7 is selected from A, L and S; X9 is selected from L and N; X 10 is selected from A, S and R; X14 is selected from A, S, C and K; X 20 is selected from Y and F; X 23 is selected from N, D and R; X 26 is selected from N, D and E; X 27 is selected from N and K; X 35 is selected from K and E; X 38 is selected from I and K, X 39 is selected from D, E and L; X 40 is selected from A, E and H; X 43 is selected from A and K, X 44 is selected from A, S and E; X 45 is L or absent, X 46 is P or non-existent) and b) an amino acid sequence having at least 95% identity to the sequence defined in a).

[0076] The rationale behind this ABR sequence (and the embodiments described below) is explained in WO 2021 / 180727.

[0077] ABR has the amino acid sequence LAX3AKX6X7AX9X 10 ELDX 14 YGVSDX 20 YKX 23 LIX 26 X 27 AKT VEGVX 35 ALX 38 X 39 X 40 ILAALP, each independently preferably comprising: X3 is selected from E and S; X6 is selected from E and V; X7 is selected from A and L; X9 is selected from L and N; X 10 is selected from A and R; X 14 is selected from A, S, C and K, preferably from A, S and K, X 20 is selected from Y and F; X 23 is selected from N, D and R; X 26 is selected from N and D; X 27 is selected from N and K; X 35 is selected from K and E; X 38 is selected from I and K, X 39 is selected from D and L; X 40 is selected from A, E and H; X 45 is L or absent, X 46 is P or non-existent.

[0078] One embodiment of the ABR does not contain any cysteine ​​(C) residues.

[0079] In one embodiment, the ABR comprises an amino acid sequence selected from the group consisting of: LAEAKVLANR ELDKYGVSDF YKRLINKAKT VEGVEALKLH ILAALP (SEQ ID NO:6, ABD035); LAEAKEAANA ELDSYGVSDF YKRLIDKAKT VEGVEALKDA ILAALP (SEQ ID NO:7); GLAEAKEAAN AELDSYGVSD FYKRLIDKAK TVEGVEALKD AILAALP (SEQ ID NO: 8, PEP07914 of WO2012004384); LAEAKVLANR ELDKYGVSDY YKNLINNAKT VEGVKALIDE ILAALP (SEQ ID NO: 9, ABD wt ); and LAEAKVLALR ELDKYGVSDY YKDLIDKAKT VEGVKALIDE ILAALP (SEQ ID NO: 10).

[0080] As a second aspect of the present disclosure, there is provided a therapeutic conjugate comprising the fusion protein of the first aspect and a cytotoxic agent, such as a cytotoxic molecule, a cytotoxic peptide, a cytotoxic protein, or a cytotoxic radionuclide. In the case of a cytotoxic peptide or a cytotoxic protein, the entire therapeutic conjugate may be the fusion protein.

[0081] The cytotoxic molecule can be, for example, doxorubicin (DOX), duocarmycin (DUO), docetaxel (DTX), monomethylauristatin E (MMAE), monomethylauristatin F (MMAF), paclitaxel (PTX), mertansine (DM1), emtansine (DM1), ravtansine (DM4), soravtansine (DM4), pyrrolobenzodiazepine (PBD) or calicheamicin.

[0082] In one embodiment, the cytotoxic molecule is DM1, MMAE, MMAF or DM4 (see review by Tarcsa et al. Drug Discovery Today: Technologies; Volume 37, December 2020, Pages 13-22, and Khongorzul et al. Mol Cancer Res; 18(1) January 2020).

[0083] Examples of cytotoxic proteins include Pseudomonas exotoxin (PE) and its engineered variants, such as PE38, diphtheria toxin (DT) and devouganin (see Antignani et al, Biomolecules; 2020 Sep 17; 10(9): 1331).

[0084] Other examples of cytotoxic proteins include targeting domains against immune regulatory targets such as CD3, CD47, PD-1, PD-L1, CTLA-4, 4-1BB and OX40 (see review by Blanco et al., Clin Cancer Res 2021 Oct 15;27(20):5457-5464).

[0085] Cytotoxic radionuclides include, for example: 177 Lu, 90 Y, 188 Re; 186 Re; 166 Ho, 153 Sm, 67 Cu, 64 Cu, 149 Tb, 161 Tb, 47 Sc; 225 Ac; 212 Pb; 213 Bi, 212 Bi, 227 Th, 223 Ra; 58m Co, 131 I, 76 As, 77 As and 211 At may be selected from the group consisting of:

[0086] Preferred cytotoxic radionuclides are 177 Lu, 90 Y and 188 Re (see review by Rondon et al., Cancers (Basel); 2021 Nov 7; 13(21): 5570).

[0087] 177 Lu, 90 Y, 188 Re; 186 Re; 166 Ho, 153 Sm, 67 Cu, 64 Cu, 149 Tb, 161 Tb, 47 Sc; 225 Ac;212 Pb; 213 Bi, 212 Bi, 227 Th, 223 Ra and 58m Co is a radiometal that can be bound to the fusion protein by chelator-based conjugation. The chelator is preferably covalently bound to a cysteine ​​residue of the fusion protein, optionally via a thiol-reactive linker. Binding to an amine of an amino acid residue of the fusion protein is also possible, but generally less preferred.

[0088] The chelating agent may be selected from the group consisting of DOTA and its derivatives (eg, maleimide derivatives of DOTA), bridged macrocyclic chelating agents and sterically restricted acyclic chelating agents.

[0089] 177 Lu, 90 Y, 166 Ho, 153 Sm, 149 Tb, 161 Tb, 47 Sc; 225 Ac; 212 Pb; 213 Bi, 212 Bi, 227Th and 58m A particularly suitable chelating agent for Co is DOTA and its derivative DOTAGA.

[0090] 67 Cu and 64 For Cu, a cross-linked chelator such as CB-TE2A is a good choice.

[0091] 188 Re and 186 For Re, chelators based on cysteine- or mercaptoacetyl-containing peptides are preferred.

[0092] 131 I, 76 As, 77 As and 211At is a non-metallic radionuclide that can be attached to the fusion protein by covalent conjugation.

[0093] Radioiodine labeling can be achieved using ((4-hydroxyphenyl)ethyl)maleimide (HPEM), which can be attached to cysteine ​​(C) residues of the fusion protein. 76 As and 77 As (and 74 As) can be directly coupled to the (neo)reduced thiol group of a cysteine ​​(C) residue of the fusion protein.

[0094] 211 In the coupling of At, N-[4-(tri-n-butylstannyl)phenethyl]-maleimide can be used as a linker. 211 At is first coupled to a linker by astatodestanylation to form 4-astato-phenethyl-maleimide (AtPEM), which can then be coupled to a cysteine ​​(C) residue of the fusion protein.

[0095] As a result, the cytotoxic radionuclide is preferably linked to the fusion protein via a cysteine ​​(C) residue of the fusion protein. To avoid cross- / side reactions, the fusion protein in such cases preferably contains only one cysteine ​​(C) residue.

[0096] The cysteine ​​(C) residue linking the cytotoxic radionuclide to the fusion protein may be located at the terminal position. Preferably, the cysteine ​​(C) residue is the C-terminal residue of the fusion protein. In one embodiment, the cysteine ​​(C) residue is the C-terminal residue of an amino acid sequence extending from the C-terminus of ABR. Such an amino acid sequence may be, for example, EEEC (SEQ ID NO: 33) or GSSC (SEQ ID NO: 34).

[0097] Adoptive transfer of immune cells (e.g., T cells, NK cells, and macrophages) expressing engineered chimeric antigen receptors (CARs) has shown outstanding potential in human cancer treatment in recent years. Adoptive transfer of CD19-directed CAR T cells has led to complete and sustained remission in patients with refractory and relapsed B-cell malignancies. The extracellular portion of the CAR contains an affinity protein for a tumor antigen (TA). The affinity protein is typically fused to a transmembrane domain and an intracellular stimulatory domain. Upon cell binding to the TA on the cancer cell, endogenous downstream signaling molecules are recruited and signaled, resulting in T-cell activation and killing of the cancer cell. However, despite the extraordinary responses observed for B-cell malignancies, the potent cell killing and extremely long serum circulation time (possibly lifelong) of engineered T cells in patients may pose challenges in controlling toxicity in healthy organs and long-term side effects. Therefore, this treatment is primarily used for cancers with favorable TA expression profiles. For this reason, considerable efforts have focused on strategies to control the activity of CAR T cells (see also review by Krug et al., Cancers (Basel) 2021 Dec 30;14(1):183).

[0098] The fusion protein of the first aspect can be used to protease-activate CAR immune cells. Such CAR immune cells have the potential to be preferentially active in protease-containing tumors. Alternatively, the corresponding protease can be co-administered with the infusion of CAR immune cells for activation of cells during a defined time frame. After the treatment is completed, the protease is not available in the patient and the cells remain inactive. Both strategies have the potential to reduce toxicity and long-term side effects and make the treatment available for more cancer forms.

[0099] Thus, in a variation of the second aspect, the fusion protein of the first aspect is expressed on the surface of a cell for use in cell therapy. From the above description, it follows that the cell may be a T cell, a NK cell or a macrophage, in particular a T cell or a NK cell.

[0100] As a third aspect of the present disclosure, there is provided a therapeutic conjugate of the second aspect for use in a method of therapeutic treatment.

[0101] Therapeutic treatment method is typically the method of treating the subject suffering from cancer, such as the cancer that overexpresses EGFR.The examples of the cancer that overexpresses EGFR include lung cancer (particularly non-small cell lung cancer), prostate cancer, breast cancer, colon and rectum cancer, head and neck cancer, gastroesophageal cancer, liver cancer, glioblastoma, cervical cancer, ovarian cancer, bladder cancer, renal cancer and pancreatic cancer.

[0102] In one embodiment, the method of treatment comprises a diagnostic step of quantifying the degree of EGFR expression in the tumor, and administration of a therapeutic conjugate in case of overexpression of EGFR in the tumor.

[0103] Quantification of the degree of EGFR expression can be, for example, based on imaging, for example, by using an imaging agent that comprises the fusion protein of the first aspect. In such an imaging agent, the fusion protein can be coupled to a radionuclide suitable for imaging. When used for imaging, the fusion protein typically does not comprise a half-life extension domain.

[0104] Following administration of the imaging agent containing radionuclides, the patient is scanned to detect, visualize and / or quantify EGFR expression.Scanning is typically tomography, preferably positron emission tomography (PET) or single photon emission computed tomography (SPECT).In the latter case, CZT-based camera technology can be used.

[0105] In one embodiment, the radionuclide suitable for imaging is 18 F, 124 I, 76 Br, 68 Ga, 44 Sc, 61 Cu, 64 Cu, 89 Zr, 55 Co, 41 Ti, 66 Ga, 86Y, 110m In, 123 I, 131 I, 99m Tc, 111 In and 67 A preferred group is selected from the group consisting of: 18 F, 68 Ga, 99m Tc and 111 Another preferred group consists of 18 F, 68 Ga and 111 It consists of In.

[0106] 18 In radiolabeling with F, the conjugated moiety ( 18 Forming a covalent bond to F) can be coupled to the fusion protein. Examples of resulting structures include N-(2-(4-[ 18 F]-fluorobenzamido)ethyl)maleimide ([ 18 F]FBEM), 4-[ 18 F]-Fluorobenzaldehyde ([ 18 F]-FBA), and [ 18 F]-Fluorophenyloxadiazole methyl sulfone ([ 18 Another option is the use of [F]-FPOS in combination with a triazacyl chelator. 18 F]Aluminum monofluoride.

[0107] 123 I, 124 I, 131 I and 76 In the case of radiolabeling with Br, conjugated groups may also be used. Examples of the resulting structures are iodo- / bromo-benzoates and iodo- / bromo-hydroxyphenylethyl mealeimides.

[0108] 68 Ga, 67 Ga, 66 Ga, 44 Sc, 55 Co, 41 Ti, 86 Y, 110mIn and 111 For radiolabeling with In, it is preferred to couple a chelating agent to the HBP. Examples of chelating agents include DOTA, NOTA, NODAGA and DOTAGA and their derivatives.

[0109] 61 Cu and 64 For Cu, a cross-linked chelator such as CB-TE2A is a good choice.

[0110] 99m For radiolabeling with Tc, a variety of chelators can be used, such as chelators based on hexahistidine (H6), and cysteine- or mercaptoacetyl-containing peptides.

[0111] 18 F, 124 I, 76 Br, 68 Ga, 44 Sc, 61 Cu, 64 Cu, 89 Zr, 55 Co, 41 Ti, 66 Ga, 86 Y or 110m In the case of In, the scanning technique is preferably PET.

[0112] 123 I, 131 I, 99m Tc, 111 In or 67 In the case of Ga, the scanning technique preferably involves SPECT, for example using a CZT-based camera.

[0113] The radionuclide is preferably coupled to a terminus of the fusion protein, for example to the C-terminus of the fusion protein.

[0114] In one embodiment, the fusion protein includes an extension that forms a chelator for a radionuclide. In one example, the chelator-forming moiety is 99mIt may comprise the sequence HHHHHH (SEQ ID NO: 35) which is capable of binding to Tc. An alternative to HHHHHH is HEHEHE (SEQ ID NO: 30).

[0115] As a fourth aspect of the present disclosure, there is provided a method of treating a subject suffering from cancer comprising administering a therapeutic conjugate of the second aspect.

[0116] In a fifth aspect of the present disclosure, there is provided a pharmaceutical composition comprising the therapeutic conjugate of the second aspect and a pharma- ceutically acceptable carrier.

[0117] The composition may be adapted, for example, for intravenous administration. The composition may therefore be aqueous. The aqueous composition is preferably buffered, for example phosphate buffered. As an example, the composition may be based on phosphate buffered saline. The aqueous composition may contain human serum albumin (HSA). HSA scavenges free radicals and prevents radiolytic damage to the therapeutic conjugate. The amount of HSA may be 10-150 mg / ml, for example 50-100 mg / ml, preferably 75 mg / ml. EXAMPLES

[0118] Example 1 material and method Bacterial culture and library expression

[0119] S. carnosus cells were grown overnight in B2 medium (1% casein hydrolysate, 2.5% yeast extract, 0.5% D-glucose, 2.5% NaCl, 0.08% K2HPO4, pH = 7.4) supplemented with 10 μg / mL chloramphenicol. For the isolation of binders, 10 9 A combinatorial S. carnosus library expressing different affibody variants was used.

[0120] Targeted biotinylation

[0121] ZEGFR-His6-Cys was purified by IMAC and biotinylated using EZ-Link™ Maleimide-PEG2-Biotin (Thermo Scientific) according to the manufacturer's recommendations. EGFR (His tag) (Sino Biological Inc., China) was biotinylated using Biotin-XX Microscale Protein Labelling Kit (Invitrogen, USA) according to the manufacturer's recommendations. Protein concentration was determined using absorbance at 280 nm.

[0122] Library selection against ZEGFR using MACS

[0123] Streptavidin-coated Dynabeads (Invitrogen, USA) (500 μL) were washed twice with 800 μL PBS-P (phosphate-buffered saline supplemented with 0.1% Pluronic® F108 NF Surfactant, pH 7.4; BASF Corporation, USA) and incubated with 150 nM biotinylated ZEGFR for 1 h at room temperature under gentle rotation. The ZEGFR-coated magnetic beads were then washed with PBS-P supplemented with 2 mM EDTA (PBSP-E). A number of cells (10 10 The cells were washed with PBSP-E, the pellet was resuspended in PBSP-E and incubated for 5-10 min. 8A final concentration of 1000 cells / mL was achieved. Cells were incubated with 1.25 mg of ZEGFR-coated magnetic beads with gentle rotation for 1 h at room temperature followed by 5 min on ice. The tube was placed on a magnetic rack for 4 min to capture the beads and the supernatant was removed. The beads were then resuspended in 30 mL of ice-cold PBS-P. Capturing and washing of the beads was repeated three times. Finally, cells were resuspended in 50 mL of B2 medium supplemented with 10 μg / mL chloramphenicol. The culture was incubated overnight at 37 °C and 150 rpm. Serial dilutions of samples taken before and after magnetic sorting were used to calculate population size. Finally, the library was incubated with 225 nM Alexa Fluor 647-HSA conjugate and 33.3 nM streptavidin R-phycoerythrin conjugate (SAPE) in PBS-P. Cells were washed twice with ice-cold PBS-P and resuspended in 300 μL of ice-cold PBS-P. Cells were analyzed using a Gallios™ flow cytometer (Beckman Coulter, CA, USA). The laser protocols used for detection were FL6-660 / 20 nm for the detection of Alexa Fluor 647-HSA and FL2-575 / 20 nm for the detection of SAPE.

[0124] Library selection against ZEGFR using FACS

[0125] Subsequently, fluorescence-activated cell sorting (FACS) was used to isolate the output cells from MACS selection (library size ≈ 10 5The cells were then sorted. A coverage of 100x the new library size was used for overnight culture. The cells were washed twice with ice-cold PBS-P and mixed with biotinylated ZEGFR suspended in PBS-P at a concentration of 150 nM. The samples were incubated at room temperature for 1 h with gentle rotation. The cells were then washed twice using ice-cold PBS-P. To visualize and sort the cell library by FACS, the cells were incubated with 225 nM Alexa Fluor 647-HSA conjugate and 33.3 nM SAPE on PBS-P. Finally, the cells were washed twice with ice-cold PBS-P and resuspended in 300 μL of ice-cold PBS-P. The cells were then sorted using an Astrios FC cell sorter (Beckman Coulter, USA). The laser protocols used for detection were 561-585 / 40 height-log for SAPE detection and 640-671 / 30 height-log for Alexa Fluor 647-HSA detection. 6 Cells were aliquoted into 1.5 mL of B2 medium and incubated at 37 °C for 1 h with gentle rotation. Cells were then mixed with B2 medium supplemented with 10 μg / mL chloramphenicol to a final volume of 3 mL and incubated at 37 °C overnight with gentle rotation. Finally, libraries were tested by FC and aliquoted as glycerol stocks and stored at -80 °C for further experiments.

[0126] Selection and generation of candidate affibodies

[0127] Individual affibody candidates were selected by plating the library on agar plates supplemented with 10 μg / mL chloramphenicol and growing single colonies overnight on TSB-Y medium (Merck, Germany) supplemented with 10 μg / mL chloramphenicol. Cells were analyzed by FC and sent for sequencing (Microsynth Seqlab, Germany). The coding sequences of the selected candidate affibodies were cloned into the pETb26+ bacterial expression vector (Addgene, USA) by restriction cloning. Escherichia coli BL21* cells were transformed with the previously cloned plasmids by standard heat shock treatment. A single colony from the transformation was inoculated and grown overnight in 10 mL TSB-Y medium supplemented with 50 ng / mL kanamycin. After 16 h, the overnight culture was diluted 1:100 in TSB-Y supplemented with 50 ng / mL kanamycin to a final volume of 100 mL. The culture was incubated at OD 600The cells were induced with IPTG to a final concentration of 1 mM at ≈1 and incubated overnight at 27 °C and 200 rpm. After 16 h, the cells were harvested by centrifugation at 5000 x g for 8 min and stored at -20 °C. The cells were lysed by sonication for 1.5 min (1 s ON / 1 s OFF) followed by centrifugation at 4 °C and 10000 x g for 20 min. The supernatant was filtered (0.45 μm) and the protein of interest was purified using immobilized metal ion affinity chromatography (IMAC). For this purpose, a PD-10 column was packed with 3 ml of TALON Metal Affinity Resin. Washing buffer (50 mM Na2HPO4, 500 mM NaCl, 15 mM imidazole, pH 8) and elution buffer (50 mM Na2HPO4, 500 mM NaCl, 300 mM imidazole, pH 8) were used according to the manufacturer's recommendations (GE Healthcare, Sweden). Finally, the buffer was changed to PBS using a PD-10 desalting column according to the manufacturer's recommendations (GE Healthcare, Sweden). The purified proteins were analyzed by mass spectrometry (MS) (4800 MALDI TOF / TOF, Applied Biosystems, USA) and SDS-page gels (NuPAGE, Invitrogen, USA).

[0128] Assessment of secondary structure, thermal stability and refolding ability of ZB05 using circular dichroism

[0129] Circular dichroism spectroscopy was performed using a Chirascan spectropolarimeter (Applied Photophysics, UK) with a 1 mm path length to analyze the alpha-helical content, thermal stability, and refolding ability of the constructs at a concentration of 0.4 mg / mL. Thermal stability was assessed by measuring the change in ellipticity at 221 nm upon heating (5 °C / min) from 20 °C to 95 °C. Melting temperatures (Tm) were approximated by data obtained from variable temperature measurements (VTM) by curve fitting using the Boltzmann Sigmoidal model (GraphPad Prism version 7, USA). Refolding ability was assessed by comparing the spectra obtained from measurements at wavelengths ranging from 195 to 260 nm at 20 °C before and after thermal denaturation.

[0130] Affinity screening by surface plasmon resonance

[0131] Target binding was measured for soluble ZB05 affibody masking candidates and POC-PA molecules using a Biacore 8K SPR instrument (GE Healthcare, Sweden). Approximately 676 RU of ZEGFR and 1000 RU of HSA were immobilized by amine coupling on a dextran CM-5 sensor chip according to the manufacturer's recommendations (GE Healthcare, Sweden). PBS-T (phosphate buffered saline supplemented with 0.05% Tween 20, pH 7.4) was used as running buffer. Analytes ZB05 and ZEGFR were injected at five different concentrations (200, 100, 50, 25 and 12.5 mM) for 150 s followed by dissociation for 150 s. Intact and cleaved POC-PA were injected at a concentration of 100 nM for 150 s followed by dissociation for 150 s. Experiments were performed at 25 °C with a flow rate of 100 μL / min. The chip was regenerated by injecting 10 mM HCl for 30 seconds. Binding kinetics were analyzed by Biacore evaluation software using the Langmuir 1:1 kinetic model.

[0132] Mutagenesis studies of ZB05 masking candidates, and binding analysis

[0133] Each randomized position in the ZB05 sequence was individually mutated to each amino acid except cysteine, generating a total of 253 sequences. Oligos for each sequence were synthesized and pooled (Twist Bioscience, USA). The pooled oligos were cloned into the S. carnosus display vector pSC2 using NEB Builder (New England Biolabs, USA). The resulting plasmids were transformed into Stellar™ Electrocompetent Cells (Takara Bio, Japan) for amplification and extracted using a Qiagen Maxi prep kit (Qiagen, USA). The amplification vectors were then transformed into electrocompetent S. carnosus TM300 cells following standard electroporation protocols.

[0134] S. carnosus cells were grown overnight in B2 medium supplemented with 10 μg / mL chloramphenicol. After 16 h, cells were washed twice with PBS-P and mixed with biotinylated ZEGFR resuspended in PBS-P at a concentration of 150 nM. Samples were incubated for 1 h at room temperature with gentle rotation. Afterwards, cells were washed twice using ice-cold PBS-P. To visualize and sort the library by FACS, cells were incubated with 225 nM Alexa Fluor 647-HSA conjugate and 33.3 nM SAPE in PBS-P. Finally, cells were washed twice with ice-cold PBS-P and resuspended in 300 μL of ice-cold PBS-P. Cells were subsequently sorted for binding and non-binding using an Astrios FC cell sorter (Beckman Coulter, USA). The laser protocols used for detection were 561-585 / 40 height-log for SAPE detection and 640-671 / 30 height-log for Alexa Fluor 647-HSA detection. A total of 106 Cells were aliquoted into 1.5 mL of B2 medium and incubated for 1 h at 37 °C with gentle rotation. Afterwards, cells were mixed with B2 medium supplemented with 10 μg / mL chloramphenicol to a final volume of 3 mL and incubated overnight at 37 °C with gentle rotation. Finally, the sorted bound and unbound populations, as well as the naive library, were tested using FC and used for next-generation sequencing.

[0135] Deep sequencing of libraries from mutagenesis studies

[0136] S. carnosus cells were grown overnight in B2 medium supplemented with 10 μg / mL chloramphenicol. Plasmids were purified from each library population using a Qiagen Miniprep kit according to the manufacturer's instructions (Qiagen, USA). Samples were prepared for deep sequencing by PCR amplifying the plasmids with primers containing TrueSeq adapters and specific indexes (Illumina, USA). Sequencing was performed at Scilifelab (National Genomics Infrastructure, Sweden) using a MiSeq 300 cycle instrument (Illumina, USA). Output FASTQ files were analyzed by Geneious version 2020.1.1 (Geneious, New Zealand). NGS was used to sequence the bound and unbound populations of mutant ZB05 variants as well as the naive mutagenesis library. Data were normalized to the occurrence of the amino acid at each position in the naive library. Acceptable substitutions that retained binding to ZEGFR were determined from a 2-fold enrichment in the binding population compared to the naive library with at least 50% depletion for the corresponding variant in the non-binding population.

[0137] POC-PA subcloning into staphylococcal display vector

[0138] The genes encoding the EGFR-binding affibody molecule ZEGFR and the anti-idiotypic affibody molecule ZB05 were cloned by restriction cloning into the pSC2 vector, separated by a TEV protease substrate sequence and linked to an albumin binding protein (ABP). The TEV protease substrate consisted of the sequence ENLYFQG (SEQ ID NO: 36) flanked by G4S repeats to extend the length of the linker (the exact sequence of the linker containing the TEV protease substrate is shown in FIG. 12). A construct containing an anti-ZHER2 affibody masking domain ("ZE01", which does not bind EGFR) was cloned into pSC2 and used as a control. The resulting plasmids were transformed into E. coli TOP10 for plasmid amplification and subsequently into electrocompetent S. carnosus TM300 cells following standard electroporation protocols (Lofblom et al. J. Appl. Microbiol. 2007, 102(3), 736-747).

[0139] The complete amino acid sequence of the POC-PA prodrug is shown in FIG.

[0140] POC-PA activation by TEV protease on the S. carnosus cell surface

[0141] S. carnosus cells displaying the various prodrug constructs were grown overnight according to standard protocols previously described. Approximately 10 7Bacterial cells (10 μL of overnight culture) were washed twice with 800 μL of 1× PBS-P and pelleted by centrifugation at 4 °C and 6000 rpm for 6 min. Cells were resuspended in either assay buffer (50 mM Tris-HCl, 0.5 mM EDTA, 1 mM DTT, pH 8) supplemented with 5 units of TEV protease or assay buffer alone (control) and incubated for 1 h at 30 °C. Bacterial cells were washed three times with PBS-P and incubated in 100 μL of PBS-P supplemented with 50 nM biotinylated EGFR receptor or PBS-P alone for 45 min at 37 °C with gentle rotation. Finally, cells were washed twice with PBS-P, incubated with Alexa Fluor 647-HSA and streptavidin R-phycoerythrin conjugate and analyzed by FC.

[0142] Soluble POC-PA activation by TEV protease

[0143] The various prodrug constructs were cloned into the pET26b+ expression vector and produced in E. coli BL21* cells. Proteins were purified using an automated purification system (AKTA, GE Healthcare, Sweden). For this purpose, a PD-10 desalting column packed with 10 mL of HSA-Sepharose, 1x TST pH 8, 5 mM, NH4Ac pH 5.5, and 0.5 M HAc was used according to the manufacturer's instructions. Proteins were lyophilized and stored at -20 °C. POC-PA (ZB05-TEV) was added to assay buffer (50 mM Tris-HCl, 0.5 mM EDTA, 1 mM DTT, pH 8) supplemented with 33 μg of TEV. 基質 The (-ZEGFR-ABP) protein was resuspended for 1 h at 30 °C. The buffer was changed to PBS-0.1% BSA using a PD-10 desalting column according to the manufacturer's recommendations (GE Healthcare, Sweden).

[0144] mammalian cell culture

[0145] H292 (human mucoepidermoid lung carcinoma) and A431 (human squamous cell skin carcinoma) (American Type Culture Collection, ATCC, via LGC Promochem, Sweden) cell lines were used for the cytotoxicity assay. Cells were cultured in Roswell Park Memorial Institute (RPMI) medium (Flow laboratories, UK) supplemented with 10% fetal bovine serum (Sigma-Aldrich, St. Louis, MO., USA) and a mixture of penicillin 100 IU / mL and streptomycin 100 μg / mL (PEST, Biokrom Kg, Berlin, Germany) (for H292), and in Dulbecco's Modified Eagle's Medium (DMEM) medium (Flow, Irvine, UK) supplemented with 10% fetal bovine serum (for A431). Cells were grown at 37 °C in a 5% CO2 atmosphere.

[0146] Flow cytometric analysis of EGFR binding on H292 and A431 cells

[0147] A431 and H292 cells were cultured as described above according to the manufacturer's recommendations (ATCC, USA). Trypsinized cells (5·10 5 ) and washed in 500 μL PBS-0.1% BSA. Cells were incubated in 100 μL PBS-0.1% BSA supplemented with 100 nM POC-PA (previously treated with or without TEV protease) for 1 h at room temperature. Cells were washed once more and incubated with 225 nM Alexa Fluor 647-HSA in PBS-0.1% BSA for 45 min on ice. After the final washing step, cells were resuspended in 300 μL PBS-0.1% BSA and analyzed by FC using a Gallios™ flow cytometer (Beckman Coulter, USA).

[0148] result Using staphylococcal display by randomization of 14 surface-exposed residues, we successfully generated an anti-idiotypic affibody masking domain, denoted ZB05, with specificity for the binding surface of ZEGFR (Figure 1). The purpose of the masking domain is to block binding of EGFR-targeting affibody. We constructed a proof-of-concept pro-affibody (POC-PA) molecule consisting of the masking domain fused to the N-terminus of the EGFR-binding affibody molecule ZEGFR via a TEV-protease cleavable linker and an albumin-binding protein (ABP) (Figures 2 and 12). We selected a substrate sequence for TEV protease to facilitate initial characterization and included the ABP as a purification tag and characterization tool (although in therapeutic applications, the ABP has a half-life extension effect).

[0149] Isolation of affibody masking candidates

[0150] A naive combinatorial S. carnosus library with a theoretical size of 10 affibody variants was used to isolate anti-idiotypic affibodies with affinity for the binding surface of the EGFR-binding affibody molecule ZEGFR. Magnetic-assisted cell sorting (MACS) was first used twice to reduce the library complexity before several rounds of FACS. The estimated maximum cell diversity after the second MACS selection round was 4.8 10 5variants. To further enrich the library of binders, three rounds of FACS selection were performed. In the first round of FACS, cells were sorted from the output of the second MACS round using a 0.42% gate to generate library F1A. In the second round of FACS, F1A was sorted using a 3.02% gate to obtain F2A1. Finally, a third round of FACS was performed on the F2A library using a gate encompassing 12.77% of the population to obtain library F3A1.2. The output from each selection round was analyzed by flow cytometry (FC) to confirm enrichment for binding to ZEGFR. Negative controls were used to exclude possible streptavidin binders (data not shown). As can be seen from the flow cytometry analysis of the sorted output, the number of putative binders increased after each selection round.

[0151] Affinity testing of individual masking domain candidates

[0152] More than 30 sequenced candidates (both repetitive and unique) were analyzed for binding to ZEGFR by FC. One of these candidates (ZB05) was found to show high binding propensity to ZEGFR, and ZB05 was selected as an anti-idiotype masking domain candidate for the construction of a pro-affibody targeting EGFR. The sequence of ZB05 is shown in Figure 10.

[0153] Mutagenesis studies of ZB05 masking candidates, and binding analysis

[0154] Mutagenesis studies of the ZB05 masking domain were performed to assess the binding contribution of the amino acids at each randomized position and how substitution to any other amino acid would affect binding to ZEGFR. The original sequence and randomized positions of ZB05 are shown in Figure 10. FACS was used to sort the binding and non-binding populations of a mutagenesis library containing 253 different ZB05 variants. A representative sort of the library and subsequent flow cytometry analysis of the binding population is shown in Figure 11. NGS was used to sequence the binding and non-binding populations as well as the naive mutagenesis library. The results are summarized in Figure 10. Position 13 was included in the mutagenesis library despite being deleted in the original ZB05 sequence. From the data, positions that appear to be important for binding can be identified. These include amino acids at positions 9, 11, 17, 18, 27, 32 and 35, where substitution to any other amino acid abolished binding. Positions 14 and 31 have limited flexibility, allowing threonine to serine and isoleucine to leucine substitutions, respectively. Position 13 is deleted in the original ZB05 sequence, allowing only substitution with glutamic acid. The deletion appears to be beneficial for binding, since 40% of the non-binding population contained an amino acid at the deleted position, compared to only 0.6% of the binding population. This is not surprising, since the insertion of an amino acid at position 13 shifts the spatial location of amino acids with high apparent binding contributions to positions 14, 17 and 18 on helix 1. The remaining positions (10, 24, 25 and 28) are highly variable with several different amino acid substitutions that preserve binding.

[0155] Generation and characterization of the anti-idiotype masking domain ZB05

[0156] The ZB05 masking candidate was cloned into the pET26b+ vector and produced in E. coli BL21* cells as a soluble monomer with a predicted molecular weight of 7.6 kDa, containing a C-terminal His tag. The protein was purified by IMAC and subsequently analyzed using mass spectrometry (MS) and SDS-page to confirm mass identity and assess purity, respectively (data not shown). Circular dichroism spectroscopy was used to investigate the secondary structure and to determine the thermal stability and refolding ability. The results of variable temperature measurements (VTM) are shown in Figure 3. Melting temperatures (T m ) was calculated to be 64.1 °C. The protein exhibited the predicted alpha-helical secondary structure content and could be fully refolded after thermal denaturation (Figure 3A). A surface plasmon resonance (SPR)-based biosensor assay was performed to test the binding affinity of ZB05 to ZEGFR. Five concentrations of ZB05 were injected over a sensor chip containing immobilized ZEGFR. HSA was immobilized and screened for nonspecific binding. The results confirm the binding of ZB05 to ZEGFR with no nonspecific binding to the negative control surface (Figure 3B).

[0157] Analysis of POC-PA activity on the cell surface of S. carnosus

[0158] To generate the proof-of-concept pro-affibody (POC-PA), both ZB05 and ZEGFR sequences were subcloned into a Staphylococcus aureus display vector containing a TEV substrate coding sequence and a linker with a C-terminal albumin binding protein (ABP). TEV-protease was chosen to facilitate initial analysis of the interaction between EGFR-binding ZEGFR molecules and masked ZB05 affibody molecules. In later experiments, the protease substrate sequence could be easily exchanged to accommodate any protease specificity. POC-PA (ZB05-TEV 基質POC-PA (ZB05-ZEGFR-ABP) was displayed on staphylococci and analyzed for binding to recombinant human EGFR using FC before and after treatment with TEV-protease. As we were unable to detect binding to EGFR for intact POC-PA in this particular experiment (data not shown), the results demonstrated the masking ability of ZB05 to prevent the binding interaction of ZEGFR with its target. However, after treatment with TEV-protease, ZEGFR binding was restored (data not shown). Overall, the results demonstrated that protease cleavage is a requirement for POC-PA to bind EGFR in solution. Furthermore, another construct consisting of anti-ZHER2 affibody masking domain and ZEGFR binding domain (POC-PA-DM) was evaluated under the same conditions to verify that masking by ZB05 is conferred by its specificity for ZEGFR and is not due to steric hindrance or nonspecific interactions (data not shown). This result indicates that the interaction between the two domains of POC-PA is due to the specific anti-idiotypic binding of ZB05 to ZEGFR.

[0159] Generation of soluble POC-PA and assessment of TEV cleavage

[0160] The purified protein was analyzed by both MS (data not shown) and SDS-page gel to confirm the size and purity of the sample, respectively (Figure 4A). Protein samples were analyzed before and after incubation with TEV protease, which confirmed the cleavage of the prodrug into two products, the ZB05 masking domain (8146 Da) and the ZEGFR binding domain (30531 Da) fused to the ABP (Figure 4A). SPR was used to analyze the binding of both intact and cleaved POC-PA to immobilized EGFR and HSA. The sensorgrams are shown in Figure 4B. The results demonstrate the masking ability of ZB05, which prevents the interaction of intact POC-PA with EGFR in this particular experimental setting. After cleavage by TEV-protease, POC-PA was able to bind with high affinity to EGFR. The binding of ABP to HSA was not affected by the presence of ZB05 or by treatment with TEV-protease. The results confirm those obtained from FC, thus confirming the masking ability of ZB05.

[0161] EGFR binding analysis of POC-PA on H292 and A431 cells using flow cytometry

[0162] We analyzed the binding of POC-PA to native EGFR on two human cancer cell lines, H292 (human mucoepidermoid lung carcinoma) and A431 (human squamous cell carcinoma), with moderate and high EGFR expression, respectively, for both intact POC-PA and POC-PA precleaved by TEV-protease (Figure 5). To evaluate the effect of steric hindrance on the binding capacity of EGFR-specific affibodies, we included the construct POC-PA-DM, which has a dummy (non-ZEGFR-binding) masking affibody domain (ZE01). The signal of cleaved POC-PA overlapped with that of POC-PA-DM in both cell lines and was notably different from intact uncleaved POC-PA, indicating a specific blockade of EGFR binding by the ZB05 masking domain and that protease-mediated activation improves EGFR binding by removal of ZB05. However, Figure 5 also demonstrates that uncleaved POC-PA successfully binds to EGFR on cancer cells, indicating that cleavage of the linker connecting the EGFR-binding domain to the masking domain is not necessarily required for therapeutic applications.

[0163] Example 2 (Biodistribution of ZEGFR-Based Prodrugs) There are several differences between the prodrug used in this in vivo biodistribution study and the POC-PA prodrug used in Example 1: · In the linker, the TEV-substrate sequence is replaced by the matriptase-substrate sequence; ·ABD035 is used in the albumin binding region (ABR) instead of ABP (ABD035 is a smaller domain with higher affinity); At the C-terminus, a triglutamyl (EEE) linker is added followed by a cysteine ​​(SEQ ID NO: 33) for conjugation of a radiometal chelator (the triglutamyl (EEE) linker increases hydrophilicity); an N-terminal HEHEHE-tag is added; and The spacer linking ZEGFR to ABR is GGGGS (SEQ ID NO:32) instead of VDLQAC (SEQ ID NO:28).

[0164] The complete amino acid sequence of the prodrug of Example 2 (SEQ ID NO:12) is shown in FIG.

[0165] material and method cell culture

[0166] Human epidermoid carcinoma cell line A431 (EGFR positive, low matriptase expression), lung mucoepidermoid carcinoma cell line H292 (EGFR positive, high matriptase expression), and lymphoma cell line Ramos (EGFR negative) were obtained from the American Type Culture Collection (ATCC, via LGC Promochem, Boras, Sweden). They were maintained in RPMI-1640 medium supplemented with 10% fetal bovine serum (FBS) (Sigma-Aldrich), 2 mM L-glutamine, and a mixture of 100 IU / mL penicillin and 100 μg / mL streptomycin. Cells were grown in a humidified incubator at 37 °C with a 5% CO2 atmosphere.

[0167] Labeling and in vitro stability

[0168] Indium chloride [ 111 [In]InCl3 was purchased from Ibsen. Buffers were prepared in high quality Milli-Q water and purified overnight from metal contaminants using Chelex 100 resin (Bio-Rad Laboratories, Hercules, CA, USA) and 0.2 μm filtered. The three compounds were stored at -20 °C in 0.2 M ammonium acetate, pH 6.0.

[0169] Compounds (21–25 μg in 88–97 μL of 0.2 M ammonium acetate, pH 6.0) were added to [ 111Radiolabeling was performed by mixing [In]InCl3 (30 MBq in 40 μL of 0.05 M HCl). The mixture was vortexed thoroughly and incubated at 80 °C for 60 min. After incubation, a 500-fold molar excess of EDTA (160 μg in 16 μL of 0.2 M ammonium acetate, pH 6.0) was added and the mixture was incubated at 80 °C for 10 min. The reaction mixture was then purified using a NAP-5 size-exclusion column pre-equilibrated and eluted with phosphate-buffered saline (PBS). The radiochemical yield and radiochemical purity of the compounds were determined using iTLC eluted with 0.2 M citric acid, pH 2.0. In this system, [ 111 In]In-labeled compounds remain at the point of application and free [ 111 [In]In moves with the solvent front. The distribution of activity between the strips was measured using a Storage Phosphor System (CR-35 BIO Plus, Elysia-Raytest, Bietigheim-Bissingen, Germany) and analyzed using AIDA Image Analysis software (Elysia-Raytest, Bietigheim-Bissingen, Germany).

[0170] In vitro stability was assessed in PBS in the presence of a 1000-fold molar excess of ethylenediaminetetraacetic acid (EDTA). After purification, samples of labeled compounds (1-1.6 μg, 50 μL in 1% BSA in PBS) were mixed with EDTA (20 μg, 2 μL in PBS) and incubated at room temperature for up to 48 h. Samples were taken at 1 h, 4 h, 24 h, and 48 h for iTLC analysis.

[0171] Binding specificity in EGFR-expressing cells in vitro

[0172] In vitro binding specificity testing was performed according to a previously described method (Wallberg et al., Cancer Biother Radiopharm. 2008;23(4):435-42). In summary, 1 day before the experiment, 10 6EGFR-positive A431 and H292 cells were seeded in 6-well plates at a density of cells / well. During the day of the experiment, a solution of radiolabeled compound (10 nM) was added to the cell plate. For blocking, 1000 nM of unlabeled ZEGFR-ABD035-DOTA was added 15 min before the radiolabeled compound to saturate the receptor in a set (n=3) of wells at room temperature. The same volume of culture medium was added to another set of dishes to make up the volume. The cells were incubated for 60 min at 37 °C. Afterwards, the medium was removed and the cells were washed using 1 mL of PBS wash. The cells were then detached by trypsin-EDTA solution and harvested. Cell-associated activity was measured using an automated γ-spectrometer (2480 Wizard; Wallac, Finland).

[0173] Additional in vitro binding experiments were performed to compare EGFR expression levels in H292 and A431 in vitro. Cells were incubated with radiolabeled ZEGFR-ABD035-DOTA (10 nM) (without blocking) and cetuximab (1000 nM) or unlabeled ZEGFR-ABD035-DOTA (1000 nM) (with blocking) as described above. The same amount of radioactivity (1.64 × 10 6 Counts per minute (CPM) were added to each cell culture dish. In addition, cells were counted in additional culture dishes (n=3) for each cell line. CPM of unblocked cells / 10 6 CPM / 10 blocked by cetuximab from cells 6 Specific cell binding activity was calculated minus cells.

[0174] Animal testing

[0175] Animal studies were approved by the Local Ethics Committee for Animal Research in Uppsala.

[0176] Female NMRI and Balb / c nu / nu mice were supplied by Scanbur A / S (Karlslunde, Denmark) and allowed a 1-week adaptation period before the start of experimental procedures.

[0177] Biodistribution in NMRI mice

[0178] To test whether masking of binding sites would prevent hepatic uptake, biodistribution studies were performed in 24 female NMRI mice at 4 and 24 hours post-injection. For each time point, groups of 4 mice were injected with equimolar amounts (133 pmol) of each compound: 3 μg of [ 111 In]In-labeled prodrug, 3 μg of [ 111 1.8 μg of [In]In-labeled dummy linker or 111 [In]In-labeled unmasked control was injected intravenously (iv) (40 kBq in 100 μL of 1% BSA in PBS per mouse). In the prodrug, the linker (linking ZB05 to ZEGFR) was based on the sequence LSGRSDNH (SEQ ID NO: 21) extended on both sides by G4S repeats (the exact sequence of this linker is shown in FIG. 13). The total length of this linker was therefore 39 amino acid residues. LSGRSDNH is recognized, for example, by matriptase and is therefore protease cleavable. In its non-cleavable variant (i.e., the "dummy linker"), the linker was instead two repeats of GGGS (SEQ ID NO: 37) extended on both sides by three repeats of GGGGS (SEQ ID NO: 32). The total length of the non-cleavable linker was therefore also 39 amino acid residues.

[0179] Mice were euthanized by an overdose of anesthesia solution (30 μl of solution per gram of body weight, ketamine 10 mg / mL and xylazine 1 mg / mL) and sacrificed by cardiac puncture. Blood, salivary glands, lungs, liver, spleen, pancreas, small intestine, large intestine, kidneys, muscle, and bones were collected and weighed. The digestive tract (including its contents) and the remaining carcass were also collected. An automated γ-spectrometer (2480 Wizard; Wallac, Finland) was used to measure the activity of the organs, as well as the standard of the injected solution. Intake values ​​were calculated as a percentage of the injected dose per gram of sample weight (%ID / g), except for the digestive tract (including its contents) and the remaining carcass, which were calculated as a percentage of the injected dose per whole sample (%ID).

[0180] transplant

[0181] For H292 tumor implantation, female Balb / c nu / nu mice were implanted with 10 × 10 cells in 100 μL of culture medium mixed 1:1 with Matrigel into the right hind limb. 6 H292 cells were subcutaneously injected. Experiments were performed 13–16 days after implantation. For Ramos implantation, female Balb / c nu / nu mice were injected with 5 × 10 cells in 100 μL of culture medium into the left hind limb. 6 Ramos were subcutaneously injected. Experiments were performed 13–15 days after implantation. For A431 tumor implantation, female Balb / c nu / nu mice were injected with 10 × 10 cells in 100 μL of culture medium into the left hind leg. 6 A431 cells were subcutaneously injected into the H292 xenografts. Experiments were performed 16 days after implantation. The mean animal weights were 17.6 ± 1.3 g in the H292 group, 18.6 ± 1.3 g in the Ramos group, and 17.3 ± 1.8 g in the A431 group. The mean tumor weights were 0.38 ± 0.23 g for H292 xenografts, 0.19 ± 0.11 g for Ramos xenografts, and 0.11 ± 0.06 g for A431 xenografts.

[0182] To assess the biodistribution of the three compounds over time, 24 mice bearing H292 xenografts were randomized into six groups (n=4). Two groups received 133 pmol of 111 In]In-labeled prodrug,111 In]In-labeled dummy linker or [ 111 In]In-labeled unmasked control (40 kBq in 100 μL of 1% BSA in PBS per mouse) was injected iv, respectively. Organs and tumors were harvested 4 and 48 hours after injection, weighed, and assayed for activity as described above.

[0183] In vivo specificity of EGFR binding (H292 vs. Ramos 4 hours and 48 hours)

[0184] To assess in vivo specificity, eight mice bearing EGFR-positive H292 xenografts and eight mice bearing EGFR-negative Ramos xenografts were administered 133 pmol of [ 111 In]In-labeled prodrug was injected. Organs and tumors were harvested at 4 and 48 hours pi, weighed, and assayed for activity as described above.

[0185] In vivo assay of matriptase function (H292 and A431 in the same animal, 48 hours)

[0186] To test the relationship between tumor uptake of the prodrug and matriptase levels, five mice bearing H292 xenografts (EGFR positive, high matriptase expression) in the right hind limb and A431 xenografts (EGFR positive, low matriptase expression) in the left hind limb were treated with 133 pmol of [ 111 In]In-labeled prodrug (40 kBq in 100 μL of 1% BSA in PBS per mouse) was injected iv. Mice were euthanized at 48 hours pi and organs, half of the liver and parts of the A431 and H292 xenografts, were harvested, weighed, and assayed for activity as described above. Another half of the liver and parts of the A431 and H292 xenografts were harvested, formalin fixed, and embedded in paraffin for histopathological examination.

[0187] Imaging (H292 and Ramos, 4 hours 48 hours)

[0188] To image EGFR expression, whole-body SPECT / CT scanning was performed using a nanoScan SPECT / CT (Mediso Medical Imaging Systems, Budapest, Hungary). 111 As a control, one mouse bearing an H292 xenograft was injected iv with [In]In-labeled prodrug (14.7 μg, 1.9 MBq). 111 To demonstrate in vivo specificity, one mouse bearing a Ramos xenograft was injected iv with [In]In-labeled unmasked control (8.8 μg, 1.1 MBq). 111 In]In-labeled prodrug (11.2 μg, 2.9 MBq) was injected iv. Mice were imaged at 4 and 48 hours pi. 111 Imaging of [In]In-labeled compounds and image reconstruction were performed as previously described (Rinne et al. J. Mol Sci. 2020 Feb 15;21(4):1312).

[0189] statistical analysis

[0190] Statistical analysis was performed using GraphPad Prism (version 9.0.0; GraphPad Software, Inc., La Jolla, CA, USA). In vitro and in vivo data were analyzed using unpaired two-tailed t-tests. A p-value <0.05 was considered statistically significant.

[0191] result Labeling and Stability

[0192] Radiolabeling of three compounds with Indium-111 was performed with radiochemical yields ranging from 10% to 21%. Purification after size-exclusion column afforded purities greater than 97% (Table 1). No release of activity was observed during incubation with excess EDTA (Table 2).

[0193] [Table 1]

[0194] [Table 2]

[0195] In vitro testing

[0196] Binding specificity studies were performed using A431 and H292 cell lines. Both cell lines overexpress EGFR. For both cell lines, a significant (p<0.05, t-test) decrease in activity was observed for the unmasked control of the blocked group. This confirmed EGFR-mediated binding of the unmasked control to A431 and H292 cells. For the A431 cell line, no significant difference was observed for the prodrug or dummy linker between the blocked and unblocked groups (Figure 6A). However, for the H292 cell line, a small but significant (p<0.05, t-test) difference was observed for the prodrug or dummy linker between the blocked and unblocked groups (Figure 6B). [ 111 Specific binding of the In]In-labeled unmasked control was three-fold higher than that of H292 cells (FIG. 7), suggesting that EGFR expression by this cell line was also three-fold higher.

[0197] Biodistribution in NMRI mice

[0198] Data on the biodistribution of the three compounds in NMRI mice are shown in Table 3. Blood activity concentrations for all three compounds were >15%ID / g at 4 hours pi and >6%ID / g at 24 hours pi. For comparison, the non-ABD035 fusion 111 The blood concentrations of In-ZEGFR were 0.34±0.02 and 0.12±0.03%ID / g at 4 and 24 hours, respectively (Tolmachev et al. Eur J Nucl Med Mol Imaging. 2010 Mar;37(3):613-22), indicating that fusion to ABD035 results in extended circulation time. 111The biodistribution of In]In-labeled prodrug and dummy linker was very similar, except for a small but significant (P<0.05, t-test) difference in renal uptake at 4 hours pi, and significant differences in salivary gland and hepatic uptake at 24 hours pi. 111 Both the In]In-labeled prodrug and the dummy linker had significantly lower hepatic uptake than the unmasked control at 4 and 24 hours pi.

[0199] [Table 3-1] [Table 3-2]

[0200] Biodistribution in Balb / c nu / nu mice bearing H292 xenografts

[0201] To assess tumor uptake over time, mice bearing H292 xenografts were 111 In]In-labeled compounds were injected. Biodistribution data at 4 and 48 hours pi are shown in Table 4. Equal levels of tumor uptake (p>0.3, t-test) were observed for the three compounds at both time points. 111 In]In-labeled prodrugs and dummy linkers were measured at both 4 and 48 hours pi. 111 [In]In-labeled unmasked controls had significantly (p<0.05, t-test) higher blood activity, higher uptake in organs such as lung, kidney and muscle, and significantly (p<0.05, t-test) lower hepatic uptake.

[0202] [Table 4-1] [Table 4-2]

[0203] In vivo specificity

[0204] To determine in vivo EGFR binding specificity, biodistribution was compared in mice bearing H292 and Ramos xenografts at 4 and 48 hours pi. At both time points, EGFR-positive H292 xenografts had significantly (p<0.05, unpaired t-test) higher prodrug uptake than EGFR-negative Ramos xenografts (Figure 8A,B). Uptake in H292 xenografts was 10±2%ID / g and 13±2%ID / g, whereas uptake in Ramos xenografts was 3±1%ID / g and 2±0%ID / g at 4 and 48 hours pi, respectively.

[0205] To assess the uptake of the prodrug in the presence of matriptase, biodistribution was compared in mice bearing both H292 and A431 xenografts at 48 hours pi. Uptake in matriptase-positive H292 xenografts (11 ± 2% ID / g) was significantly (p < 0.05, paired t test) higher than in matriptase-negative A431 xenografts (6 ± 1% ID / g) (Figure 8C).

[0206] Imaging

[0207] In H292 xenografts, 111 In]In-labeled prodrug and unmasked control (Figure 9A, Figure 9C), as well as [ 111 Experimental microSPECT / CT imaging of the In]In-labeled prodrug (FIGS. 9B, 9D) confirmed the biodistribution data. Imaging revealed that at 4 and 48 hours pi, 111 It was now possible to clearly visualize EGFR-expressing H292 xenografts containing both [In]In-labeled prodrug and unmasked controls, whereas Ramos xenografts showed no significant difference in [ 111 In] was not visible following injection of In-labeled prodrug.

[0208] Example 3 (Biodistribution of ZEGFR-Based Prodrugs) Four new proaffibody (PA) variants were designed to improve the cleavage efficiency by matriptase. A new protease recognition sequence (MSGRSANA, SEQ ID NO: 38, "ZW") developed by Zymeworks was used in PA-ZW-(G4S)1, PA-ZW-(G4S)2, PA-ZW-(G4S)3 and PA-ZW3 to replace the original LSGRSDNH sequence used above:

[0209] PA-ZW-(G4S)1:(HE)3-ZB05-(G4S)-(ZW)-(G4S)-ZEGFR-ABD035-EEEC; PA-ZW-(G4S)2:(HE)3-ZB05-(G4S)2-(ZW)-(G4S)2-ZEGFR-ABD035-EEEC; PA-ZW-(G4S)3:(HE)3-ZB05-(G4S)3-(ZW)-(G4S)3-ZEGFR-ABD035-EEEC; and PA-ZW3:(HE)3-ZB05-ZW3-ZEGFR-ABD035-EEEC.

[0210] Thus, in the new PAs, the linker interconnecting ZB05 and ZEGFR contains a single ZW sequence flanked by one to three G4S sequences, or a concatemer of three ZW substrate sequences.

[0211] New variants were generated, purified using affinity chromatography, lyophilized and dissolved in PBS. Four identical samples were prepared for each variant. Each sample contained 10 uM PA protein and 20 nM recombinant human matriptase (cat.Nr.3946-SEB) in a total volume of 50 uL PBS. Samples were incubated at 37 °C for 1, 3, 5 or 24 h and then frozen at -20 °C. Frozen samples containing intact (int.) uncleaved protein from the same stock used for sample preparation were thawed and analyzed by SDS-PAGE.

[0212] As shown in Figure 14, efficient matriptase cleavage was observed for variants with ZW substrate sequences. Near complete cleavage of the variant with a single ZW sequence can be seen after 5 hours. Complete cleavage of the concatemer containing the ZW3 sequence can be seen after only 1 hour.

Claims

1. A fusion protein comprising an EGFR binding domain, a masking domain, and a linker connecting the masking domain to the EGFR binding domain, The masking domain is selected from the group consisting of the amino acid sequence IX １０ SX １２ X １３ X １４ X １５ X １６ WWX １９ X ２０ X ２１ X ２２ X ２３ X ２４ X ２５ X ２６ KX ２８ X ２９ X ３０ X ３１ YX ３３ X ３４ Contains V, Independently of each other, X １０ is R, K, M, N or Q; X １２ is A or a substitution, X １３ is E or absent, X １４ is T or S, X １５ is E or a substitution; X １６ is I or a substitution, X １９ is L, substituted or absent; X ２０ is P, a substitution, or absent; X ２１ is N, substituted or absent; X ２２ is L, substituted or absent; X ２３ is T or a substitution, X ２４ is A, F, I, K, L, M, T or Y; X ２５ is D, G, I or W; X ２６ is Q or a substitution, X ２８ is W, A, F, I, L, M, Q, R, S, T or V; X ２９ is A or a substitution, X ３０ is F or a substitution, X ３１ is I or L, X ３３ is K or a substitution, and X ３４ is L or substituted; However, X １２ , X １５ , X １６ , X １９ , X ２０ , X ２１ , X ２２ , X ２３ , X ２６ , X ２９ , X ３０ , X ３３ and X ３４ six or fewer of X are substitutions, １９ ~X ２２ are absent, and The EGFR binding domain has the amino acid sequence EX ２ X ３ X ４ AX ６ X ７ EIX １０ X １１ LPNLNX １７ X １８ QX ２０ X ２１ AFIX ２５ SLX ２８ Contains D, Independently of each other, X ２ is M, F, V, L, I or S; X ３ is W, D, E or L; X ４ is I, V, G, S, M, L, A, T, N, D or W; X ６ is W, V, L, I, M or S; X ７ is D, E, N or K; X １０ is R, G, H or K; X １１ is D, N, E, Y or S; X １７ is G, W or A, X １８ is W, G or A; X ２０ is M, L, F, A or E; X ２１ is T, D, N, A or Q; X ２５ is A, S, N, G or L, and X ２８ is L, W, V, F or A; Fusion proteins.

2. In the masking domain, X １２ , X １５ , X １６ , X １９ , X ２０ , X ２１ , X ２２ , X ２３ , X ２６ , X ２９ , X ３０ , X ３３ and X ３４ and / or X １９ ~X ２２ 2. The fusion protein of claim 1, wherein one or more of:

3. In the masking domain, X １２ , X １５ , X １６ , X １９ , X ２０ , X ２１ , X ２２ , X ２３ , X ２６ , X ２９ , X ３０ , X ３３ and X ３４ and / or X １９ ~X ２２ The fusion protein of claim 2 , wherein none of

4. In the masking domain, X １０ is R and X １３ is non-existent and X １４ is T and X ２４ is A and X ２５ is D and X ２８ is W, and / or X ３１ The fusion protein of claim 1 or 2, wherein

5. The masking domain has the amino acid sequence IRSX １２ TX １５ X １６ WWX １９ X ２０ X ２１ X ２２ X ２３ ADX ２６ KWX ２９ X ３０ IYX ３３ X ３４ 3. The fusion protein of claim 1 or 2, comprising V.

6. 3. The fusion protein of claim 1, wherein the masking domain comprises the amino acid sequence IRSATEIWWLPNLTADQKWAFIYKLV (SEQ ID NO: 1).

7. In the EGFR binding domain, X ３ is W and X ６ is V or W, and X １０ is R or G, and X １７ is W or G, and X １８ is W or G, and / or X ２１ is T or D.

8. The EGFR binding domain has the amino acid sequence EX ２ WX ４ AWX ７ EIRX １１ LPNLNGWQX ２０ TAFIX ２５ SLX ２８ Contains D, Independently of each other, X ２ is M, V, L or I; X ４ is I, V, G, S, M, L, A, T, N or D; X ７ is D, E, N or K; X １１ is D, N, E, Y or S; X ２０ is M, L or F, X ２５ is A, S or G, and X ２８ The fusion protein according to claim 1 or 2, wherein is L or V.

9. In the EGFR binding domain, X ２ is M and X ４ is I, V, G or S, and X １１ is D, N or E, and X ２０ is M and X ２５ is A or S, and / or X ２８ The fusion protein of claim 1 or 2, wherein is L.

10. 3. The fusion protein of claim 1 or 2, wherein the EGFR binding domain comprises the amino acid sequence EMWIAWEEIRDLPNLNGWQMTAFIASLLD (SEQ ID NO: 2).

11. 3. The fusion protein of claim 1 or 2, further comprising a half-life extending region, such as an Fc binding region or an albumin binding region (ABR).

12. The fusion protein of claim 1 or 2, wherein the linker is a protease-cleavable linker.

13. 13. The fusion protein of claim 12, wherein the protease-cleavable linker comprises a sequence selected from the group consisting of GFLG (SEQ ID NO:11), glutamic acid-valine-citrulline, GILGVP (SEQ ID NO:13), GPLGIAGQ (SEQ ID NO:14), VHMPLGFLGP (SEQ ID NO:15), SGGPGPAGMKGLPGS (SEQ ID NO:16), PLGLAG (SEQ ID NO:17), LALGPG (SEQ ID NO:18), KRALGLPG (SEQ ID NO:19), GGGRR (SEQ ID NO:20), LSGRSDNH (SEQ ID NO:21), PMAKK (SEQ ID NO:22), RQARVVNG (SEQ ID NO:23), HSSKLQL (SEQ ID NO:24), and RRSSYYSG (SEQ ID NO:25).

14. A therapeutic conjugate comprising the fusion protein of claim 1 or 2 and a cytotoxic agent such as a cytotoxic molecule, cytotoxic peptide, cytotoxic protein or cytotoxic radionuclide.

15. 15. The therapeutic conjugate of claim 14 for use in a method of therapeutic treatment of a subject suffering from cancer.

Images