Means and methods for treating angiogenesis, fibrosis and cancer related diseases with protein oligomers comprising NC-1-Fc

Abstract

The present invention relates to a method for producing a protein oligomer comprising at least two, and preferably three, heterodimeric human NC-1-Fc proteins, the method comprises: a) culturing a host cell expressing (i) and (ii) under conditions allowing the formation of a protein oligomer comprising at least two, and preferably three heterodimeric human NC-1-Fc proteins, said (i) being a fusion protein comprising, from the N-to C-terminus, human NC-1 from collagen 18 fused to human IgG1 Fc having a "pestle" mutation, said fusion protein comprising, from the N-to C-terminus, a human NC-1-Fc protein derived from collagen 18, said human NC-1-Fc protein derived from collagen 18, and preferably a human NC-1-Fc protein derived from collagen 18, said human NC-1-Fc protein having a "pestle" mutation. The present invention relates to a method for preparing a protein oligomer comprising (i) a human IgG1 Fc having a "pestle" mutation fused to human NC-1 from collagen 18, or a human IgG1 Fc having a "pestle" mutation fused to human NC-1 from collagen 18, said (ii) being a human IgG1 Fc having a "mortar" mutation wherein (i) said fusion protein and (ii) said human IgG1 Fc having a "mortar" mutation are expressed in a ratio of 2: 1 or higher, and b) obtaining a protein oligomer comprising at least two, and preferably three, heterodimeric human NC-1-Fc proteins.

Description

[0001] This application is a divisional application of the application with an international application date of April 16, 2019, international application number PCT / EP2019 / 059747, Chinese national phase application number 201980038342.3, and invention name "Means and methods for treating angiogenesis, fibrosis and cancer-related diseases using protein oligomers containing NC-1-Fc".

[0002] The present invention relates to a method for producing a protein oligomer comprising at least two, and preferably three, heterodimeric human NC-1-Fc proteins, the method comprising: a) culturing a host cell expressing (i) and (ii) under conditions allowing the formation of a protein oligomer comprising at least two, and preferably three, heterodimeric human NC-1-Fc proteins, wherein (i) is a fusion protein comprising, from N- to C-terminus, human NC-1 from collagen 18 fused to human IgG1 Fc with a "knob" mutation, or human IgG1 Fc with a "knob" mutation fused to human NC-1 from collagen 18, and (ii) is a human IgG1 Fc with a "hole" mutation, wherein (i) the fusion protein and (ii) the human IgG1 Fc with a "hole" mutation are expressed in a ratio of 2:1 or higher, and b) obtaining a protein oligomer comprising at least two, and preferably three, heterodimeric human NC-1-Fc proteins from the host cell of step a). The present invention further relates to a method for producing a protein oligomer comprising at least two, and preferably three, monomeric human NC-1-Fc proteins, the method comprising: a) culturing a host cell expressing a fusion protein under conditions allowing the formation of a protein oligomer comprising at least two, and preferably three, monomeric human NC-1-Fc proteins, the fusion protein comprising, from N- to C-terminus: human NC-1 from collagen 18 fused to human IgG1 Fc, or human IgG1 Fc fused to human NC-1 from collagen 18, wherein the human IgG1 Fc comprises at least one monomer mutation, preferably the monomer mutation F405R, and one or more half-life extension mutations, preferably the half-life extension mutations M252Y, S254T and T256E, and b) obtaining a protein oligomer comprising at least two, and preferably three monomeric human NC-1-Fc proteins from the host cell of step a). Further, the present invention provides a protein oligomer comprising at least two, and preferably three heterodimeric human NC-1-Fc proteins, wherein the protein oligomer is used to treat, improve or prevent diseases associated with angiogenesis, wherein the heterodimeric human NC-1-Fc protein comprises (i) a fusion protein comprising, from N- to C-terminus: human NC-1 from collagen 18 fused to human IgG1 Fc having a "knob" mutation, or human IgG1 Fc fused to human NC-1 from collagen 18, and (ii) human IgG1 Fc having a "hole" mutation, wherein the "knob" mutation and the "hole" mutation in the human IgG1 Fc are as defined in the present application. The present invention also relates to a monomeric NC-1-Fc protein comprising human NC-1 from collagen 18 and an Fc domain from human IgG1, wherein the Fc domain from human IgG1 comprises at least one monomer mutation, preferably the monomer mutation F405R, and one or more half-life extension mutations, preferably the half-life extension mutations M252Y, S254T and T256E.The present invention also relates to a protein oligomer comprising at least two monomeric human NC-1-Fc fusion proteins, which is used as a drug, preferably for treating diseases associated with angiogenesis, fibrosis or diseases associated with fibrosis, vascular endothelial growth factor (VEGF)-related diseases or matrix metalloproteinase (MMP)-related diseases.

[0003] The original recombinant endostatin produced in 1997 by the Folkman laboratory (O'Reilly et al., Cell 88(2):277-85, 1997) was an insoluble aggregate generated in E. coli that was injected subcutaneously into mice. Subsequently, soluble endostatin was produced in yeast. Entremed monomeric endostatin lacks a large portion of the critical N-terminal Zn2+ binding domain. This monomeric endostatin showed a modest response in a phase 1 trial and was so expensive to produce that only a few trials were conducted (for a review, see Abdollahi DRU 2015 PMID: 15939343).

[0004] Chinese scientists developed Endostar by adding a His tag to the N-terminus, resulting in better solubilization, and discovered a refolding procedure to produce endostatin monomers in bacterial culture, which were effective in non-small cell lung cancer (US7078485B2) and obtained approval from the Chinese FDA.

[0005] Therefore, even the production of endostatin is a challenge. Dr. Javaherian developed Fc-endostatin with Lexigen (later Merck KGa), initially to circumvent two obstacles, namely using conventional mature antibody production platforms to increase the expression of large amounts of endostatin for clinical trial level production. In addition, later studies revealed a better half-life (Lee et al. CCR 14 (5): 1487-93, 2008), which is another argument for the half-life of Fc-endostatin for recombinant endostatin molecules of only 30 minutes to 1 hour.

[0006] The production of collagen 18 NC-1, which includes the endostatin domain, was a huge challenge from the outset due to the aggregation of this molecule at the oligomerization domain. Therefore, there are no in vivo data on the efficacy of recombinant NC-1, and only a few studies have been conducted on the molecule that was considered to be the precursor of endostatin so far.

[0007] Dr. Javaherian and a few other labs were able to express very small amounts of recombinant NC-1 via a His-tag or Flag-tag, but this was far from enough for preclinical dosing in animals. Therefore, generating amounts of NC-1 relevant for in vivo studies remained a major challenge.

[0008] In view of the foregoing, there is a need in the art to develop efficient methods for producing NC-1 in quantities sufficient for preclinical and clinical studies.

[0009] The technical problem on which the present invention is based can be seen as providing means and methods that meet the above needs. This technical problem has been solved by the embodiments characterized in the claims and below.

[0010] Therefore, the present invention relates to a method for producing a protein oligomer comprising at least two, and preferably three heterodimeric human NC-1-Fc proteins, the method comprising:

[0011] a) culturing a host cell expressing (i) and (ii) under conditions that allow the formation of a protein oligomer comprising at least two, and preferably three, heterodimeric human NC-1-Fc proteins, wherein (i) is a fusion protein comprising, from N- to C-terminus: human NC-1 from collagen 18 fused to human IgG1 Fc with a "knob" mutation, or human IgG1 Fc with a "knob" mutation fused to human NC-1 from collagen 18, and (ii) is a human IgG1 Fc with a "hole" mutation, wherein the fusion protein (i) and the human IgG1 Fc (ii) with a "hole" mutation are expressed at a ratio of 2:1 or higher, and

[0012] b) obtaining a protein oligomer comprising at least two, and preferably three, heterodimeric human NC-1-Fc proteins from the host cell of step a).

[0013] In the method of the present invention, a protein oligomer is produced. The protein oligomer comprises at least two heterodimeric human NC-1-Fc proteins, but can also comprise more than two heterodimeric human NC-1-Fc proteins, such as three, four, five, six or even more heterodimeric human NC-1-Fc proteins. Preferably, the protein oligomer comprises three heterodimeric human NC-1-Fc proteins.

[0014] Fig. 9 A heterodimeric human NC-1-Fc protein and a protein oligomer comprising three heterodimeric human NC-1-Fc proteins are described.

[0015] The heterodimeric human NC-1-Fc protein comprises (i) a fusion protein comprising, from N-terminus to C-terminus: human NC-1 from collagen 18 fused to human IgG1 Fc with a "knob" mutation, or a fusion protein comprising, from N-terminus to C-terminus: human IgG1 Fc with a "knob" mutation fused to human NC-1 from collagen 18. The Fc "knob-hole" (KiH) strategy is well known in the art and is described elsewhere herein. If human NC-1 from collagen 18 is located at the N-terminus of the fusion protein and human IgG1 Fc with a "knob" mutation is located at the C-terminus of the fusion protein, then the fusion protein (i) is also referred to herein as "NC-1-Fc-knob". Therefore, "Fc-knob-NC-1" corresponds to a fusion protein in which human IgG1 Fc with a "knob" mutation is located at the N-terminus of the fusion protein and human NC-1 from collagen 18 is located at the C-terminus of the fusion protein.

[0016] The heterodimeric human NC-1-Fc protein further comprises (ii) an Fc domain from IgG1 having a "hole" mutation. The Fc domain (ii) is also referred to herein as "Fc-hole".

[0017] The heterodimeric human NC-1-Fc protein comprising (i) NC-1-Fc-knob or Fc-knob-NC-1 and (ii) Fc-hole is also referred to herein as “NC-1-Fc-KiH” or “NC-1-Fc(KiH)”.

[0018] NC-1 and IgG1 Fc are human or derived from human. The term "derived from" is explained elsewhere herein. The "knob" and "hole" mutations in the Fc domain of IgG1 allow the formation of Fc dimers.

[0019] (i) it comprises, from N- to C-terminus: human NC-1 from collagen 18 fused to human IgG1 Fc with a "knob" mutation, or human IgG1 Fc with a "knob" mutation fused to human NC-1 from collagen 18, and (ii) a fusion protein of human IgG1 Fc with a "hole" mutation is expressed in a host cell under conditions that allow the formation of protein oligomers, wherein the protein oligomer comprises at least two or more heterodimeric human NC-1-Fc proteins. To this end, the host cell can be co-transfected, for example, with two different expression vectors, one expression vector comprising a nucleic acid sequence encoding a fusion protein under the control of a suitable promoter, the fusion protein comprising, from N- to C-terminus, human NC-1 from collagen 18 fused to human IgG1 Fc with a "knob" mutation, or human IgG1 Fc with a "knob" mutation fused to human NC-1 from collagen 18, and the other expression vector comprising human IgG1 Fc with a "hole" mutation under the control of a suitable promoter. However, it is obvious to those skilled in the art that the host cells can also be transfected with a single expression vector comprising a nucleic acid sequence encoding a fusion protein (i) and (ii) under the control of a suitable promoter, wherein the fusion protein (i) comprises, from N- to C-terminus, human NC-1 from collagen 18 fused to a human IgG1 Fc with a "knob" mutation, or fused to human NC-1 from collagen 18 with a "knob" mutation, and the (ii) is a human IgG1 Fc with a "hole" mutation.

[0020] The present inventors have found that in order to prevent dimerization of two human IgG1 Fc domains with a "hole" mutation ("hole-hole Fc dimer") and to achieve efficient production of heterodimeric human NC-1-Fc protein (the heterodimeric human NC-1-Fc protein then forms protein oligomers with desired biological activity, as explained elsewhere herein and shown in the following examples), it is important to express the human IgG1 Fc with a "hole" mutation in fusion proteins (i) and (ii) in a ratio of 2:1 or higher, such as 3:1 or 4:1 or even higher, wherein the fusion protein (i) comprises, from N- to C-terminus, human NC-1 from collagen 18 fused to a human IgG1 Fc with a "knob" mutation, or human IgG1 Fc with a "knob" mutation fused to human NC-1 from collagen 18. In other words: the fusion protein comprising, from N- to C-terminus, human NC-1 from collagen 18 fused to human IgG1 Fc with a "knob" mutation, or human IgG1 Fc with a "knob" mutation fused to human NC-1 from collagen 18 (i) must be overexpressed in the host cell compared to human IgG1 Fc with a "knob" mutation (ii). An ideal ratio of 2:1 or more of the fusion protein comprising, from N- to C-terminus, human NC-1 fused to human IgG1 Fc with a "knob" mutation, or human IgG1 Fc with a "knob" mutation fused to human NC-1 from collagen 18 (i) to human IgG1 Fc with a "knob" mutation (ii) can be achieved, for example, by using suitable promoters directing expression with different promoter strengths.

[0021] To achieve the desired ratio, the inventors utilized a double amount of the "knob" plasmid for a transient expression system. They also utilized a "knob": hole" multiplicity of infection (MOI) of up to 4:1 by using two different viral particles, each carrying one construct. In an alternative approach, one can also include the NC-1-Fc-knob or Fc-knob-NC-1 construct and the Fc-hole in a dual promoter system for transient transfection and stable cell generation, as described above. Then, it is not the amount of DNA (transient) or the MOI (virus) that needs to exceed 2:1 to favor the NC-1-Fc-knob or Fc-knob-NC-1 construct, but this can be controlled by different promoter strengths, such as CMV "knob" versus EF1 "hole", etc.

[0022] Preferably, the fusion protein comprising human NC-1 from collagen 18 fused to human IgG1 Fc with a "knob" mutation, or human IgG1 Fc with a "knob" mutation fused to human NC-1 from collagen 18 (i) and (ii) human IgG1 Fc with a "knob" mutation is expressed at a ratio higher than 2:1 from N- to C-terminus. Ratios equal to or higher than 5:1 may be counterproductive due to aggregation of the NC-1 oligomerization domain in the "knob" construct. Therefore, the preferred ratio is 2:1, 3:1 or 4:1.

[0023] Finally, a protein oligomer containing at least two or more heterodimeric human NC-1-Fc proteins is obtained from a host cell by methods known in the art; see, for example, Sambrook et al., Molecular cloning: a laboratory manual / Sambrook, Joseph; Russell, David W. --. 3rd ed. -- NewYork: Cold Spring Harbor Laboratory, 2001. Ausubel et al., Current Protocols in Molecular Biology. In order to obtain a protein oligomer containing at least two heterodimeric human NC-1-Fc proteins from a host cell or supernatant, the inventors performed protein A purification and size exclusion chromatography, indicating that there are two peaks in a ratio of 1:1. The first peak is the expected heterodimeric construct, and the second peak is composed of "hole-hole dimer". Therefore, the inventors moved the "knob" to "hole" to 2:1 to increase the production of NC-1-Fc-knob, which almost reduced the formation of the second peak (Fc-hole-dimer). In any case, the heterodimer NC-1-Fc-KiH could be well separated by the first peak in the fraction collector.

[0024] In a preferred embodiment of the method of the present invention for producing a protein oligomer comprising at least two heterodimeric human NC-1-Fc proteins, the method further comprises a step c), wherein the formation of (preferably) trimers is tested by cross-linking non-denatured heterodimeric human NC-1-Fc proteins. The art fully describes how to perform cross-linking experiments on non-denatured proteins. The inventors have performed cross-linking to ensure that the native protein structure (especially tertiary and quaternary) is not lost during electrophoresis. Therefore, cross-linking is performed to identify the in vivo composition of the molecule, and it is found that the molecule forms trimers in the same manner as the natural or native NC-1 domain. The method is preferably used in the production process of therapeutic substances to ensure that the NC-1-trimer is indeed formed during the production process of future production lines.

[0025] In another preferred embodiment of the method of the present invention for producing a protein oligomer comprising at least two heterodimeric human NC-1-Fc proteins, human IgG1 Fc comprises the "knob" mutation S354C / T366W, and human IgG1 Fc comprises the "hole" mutation Y349C / T366S / L368A / Y407V.

[0026] Preferably, the human IgG1 Fc with the "knob" mutation S354C / T366W comprises SEQ ID NO: 25, and the human IgG1 Fc with the "hole" mutation Y349C / T366S / L368A / Y407V comprises SEQ ID NO: 26.

[0027] More preferably, the fusion protein comprising, from N- to C-terminus, human NC-1 from collagen 18 fused to a human IgG1 Fc with a "knob" mutation (i) comprises or consists of SEQ ID NO: 27, 29, or 31, and (ii) human IgG1 Fc with a "hole" mutation comprises or consists of SEQ ID NO: 28, 30, or 32.

[0028] Particularly preferred is a fusion protein comprising, from N- to C-terminus, human NC-1 from collagen 18 fused to a human IgG1 Fc with a "knob" mutation (i) comprising or consisting of SEQ ID NO: 31, and (ii) human IgG1 Fc with a "hole" mutation comprising or consisting of SEQ ID NO: 32. (See Fig. 9 and Table 1, molecule #1).

[0029] In another embodiment of the method of the present invention for producing a protein oligomer comprising at least two heterodimeric human NC-1-Fc proteins, KiH-modified human IgG4 Fc is used instead of KiH-modified human IgG1 Fc. The "knob" and "hole" mutations in the Fc domain of IgG4 allow the formation of Fc dimers. For example, in US 2013 / 0177555A1 Figure 6 , the corresponding amino acid sequence of human IgG4 Fc is shown.

[0030] The shift from endostatin to NC-1 as an anti-angiogenic and anti-cancer agent was the first major step of the inventors of this application; see WO 2013 / 026913 and WO 2017 / 093569.

[0031] Since this discovery, the inventors have now spent nearly a decade to realize a system for large-scale production of NC-1, and many constructs and classical tags have failed, as demonstrated herein and in the following examples.

[0032] The inventors' primary goal was to advance in the field of immunoglobulin IgG Fc conjugation, achieving numerous biotechnological, pharmacokinetic and biological advantages.

[0033] However, the generation of NC-1-Fc is not a straight-forward process and is not at all an obvious matter to pursue for one skilled in the art.

[0034] The reason for this is that there are two oligomerization forces, the first being the trimerization and aggregation forces induced by the NC-1 oligomerization (or association) domain, and the second being the dimerization forces induced by the IgG-Fc. The inventors have spent a great deal of time using all the different prior art techniques, i.e., conjugation of NC-1 to the C- or N-terminus of standard IgG Fc fusions, as previously done with endostatin, and all of these approaches have failed, as explained in the following examples.

[0035] One has to consider that in the classical Fc tagging approach, one would generate an N-terminally or C-terminally conjugated NC-1 molecule that would form dimers without the NC-1 oligomerization domain. However, with the NC-1 oligomerization domain, one would receive aggregates and very pure expression.

[0036] To the best of the inventors' knowledge, there is no literature on the successful and efficient production of NC-1-Fc fusion protein and oligomers containing the fusion protein.

[0037] As described herein, the inventors have studied different approaches to circumvent this major problem and after more than a decade of research were able to come up with a successful strategy. A core point is the generation of a dimeric Fc monovalent NC-1 or a heterodimeric NC-1-Fc. The inventors decided to use a mutation in the Fc that prevents the dimerization of NC-1-Fc with NC-1-Fc using the Fc "knob-hole structure" (KiH) strategy. Therefore, the NC-1-Fc knob will only dimerize with an empty Fc-hole. Although knob-knob dimerization is prohibited, hole-hole dimerization may still occur at a lower level. In fact, the inventors found two peaks, one peak is the expected NC-1-Fc-KiH heterodimer, and the second peak is the Fc-hole-hole dimer (see the Examples below). This suggests that NC-1-Fc-knob is more difficult for cells to express. The inventors circumvented this problem again by increasing the expression of NC-1-Fc-knob relative to Fc-hole (e.g., 2:1, 3:1, 4:1 or higher), which resulted in primarily one peak of NC-1-Fc-KiH heterodimers with excellent expression efficacy. However, in a size exclusion column (SEC), the inventors realized that the molecules under physiological conditions were much larger than those expected based on the NC-1-Fc-KiH heterodimers (see the following examples). Subsequent cross-linking experiments confirmed that the inventors had surprisingly achieved the production of trimeric molecules, i.e., trimers of NC-1-Fc-KiH heterodimers. This molecule retains all the excellent properties of Fc plus trimeric NC-1. Surprisingly, this trimeric NC-1-Fc-KiH molecule showed the same efficacy in binding to unique oligomeric NC-1 binding partners such as fibronectin, VEGF, and MMP-2 / 9. With the excellent expression rate and availability of sufficient quantities of recombinant NC-1, the inventors have now been able to use this NC-1 molecule for in vivo experiments in different disease models after a long time. In addition to the data in the lung cancer (LLC) model and the pulmonary fibrosis model, further experiments are underway that will greatly improve the inventors' understanding of this molecule.

[0038] Notably, the separate expression of NC-1-Fc-knob and NC-1-Fc-hole and subsequent dimerization through a redox system failed to produce an efficient NC-1-Fc construct. Therefore, steric hindrance of the heterodimeric Fc is crucial for the success and lack of aggregation. Next, the inventors aimed to evaluate whether adding a part to the "empty" Fc-hole would affect the expression of the molecule. To this end, they conjugated the classic Fab fragment of an antibody to the Fc-hole and co-expressed it with the NC-1-Fc-knob. Interestingly, the addition of steric hindrance by the Fab part NC-1-Fc-KiH-Fab reduced the expression efficiency compared to the NC-1-Fc-KiH heterodimer, as shown in the Examples and Table 6.

[0039] At the same time, the trimeric NC-1-Fc-KiH heterodimer reported by the inventors is the first NC-1-based Fc construct that can be expressed in sufficient quantities for preclinical and clinical studies, and given the 3D complexity of both NC-1 and Fc, engineering such a molecule was not, and is not, a trivial process as can be deduced from the current literature.

[0040] The present invention further relates to a method for producing a protein oligomer comprising at least two, and preferably three monomeric human NC-1-Fc proteins, the method comprising:

[0041] a) culturing a host cell expressing a fusion protein under conditions that allow the formation of a protein oligomer comprising at least two, and preferably three monomeric human NC-1-Fc proteins, the fusion protein comprising, from N- to C-terminus, human NC-1 from collagen 18 fused to human IgG1 Fc, or human IgG1 Fc fused to human NC-1 from collagen 18, wherein the human IgG1 Fc comprises at least one monomer mutation, preferably the monomer mutation F405R, and one or more half-life extension mutations, preferably the half-life extension mutations M252Y, S254T and T256E, and

[0042] b) obtaining a protein oligomer comprising at least two, and preferably three monomeric human NC-1-Fc proteins from the host cell of step a).

[0043] Also in the method of the present invention, a protein oligomer is produced. The protein oligomer comprises at least two monomeric human NC-1-Fc proteins, but can also comprise more than two monomeric human NC-1-Fc proteins, such as three, four, five, six or even more monomeric human NC-1-Fc proteins. Preferably, the protein oligomer comprises three monomeric human NC-1-Fc proteins.

[0044] Fig.10 A monomeric human NC-1-Fc protein and a protein oligomer comprising three monomeric human NC-1-Fc proteins are shown.

[0045] The monomeric human NC-1-Fc protein comprises a fusion protein comprising, from N-terminus to C-terminus, human NC-1 from collagen 18 fused to human IgG1 Fc, or human IgG1 Fc fused to human NC-1 from collagen 18. The monomeric human NC-1-Fc protein is also referred to herein as "NC-1-Fc (monomer)".

[0046] Human IgG1 Fc comprises one or more monomer mutations, which means that the human IgG1 Fc domain is transformed into a monomer Fc, in which one, two, three, four, five or even more key amino acid residues located at the position on the human IgG1 Fc dimer interface are mutated. Human IgG1 Fc with one or more such monomer mutations is no longer able to dimerize with another human IgG1 Fc domain, but only forms a monomer. An example of such a mutation is the monomer mutation F405R.

[0047] Obviously, a fusion protein (which comprises, from N-terminus to C-terminus, human NC-1 from collagen 18 fused to human IgG1 Fc, or human IgG1 Fc fused to human NC-1 from collagen 18, wherein the human IgG1 Fc comprises one or more monomer mutations) is no longer able to dimerize with another human IgG1 Fc. However, this fusion protein is able to oligomerize with other fusion proteins, wherein the other fusion proteins comprise, from N-terminus to C-terminus, human NC-1 from collagen 18 fused to human IgG1 Fc, or human IgG1 Fc fused to human NC-1 from collagen 18, wherein the human IgG1 Fc comprises one or more monomer mutations via the oligomerization domain of NC-1, as described in the following examples, Fig.10 and as shown in Table 4 (molecule #4, SEQ ID NO: 37).

[0048] Human IgG1 Fc further comprises one, two, three, four, five or more half-life extension mutations, such as mutations M252Y, S254T and / or T256E (YTE). It is reported that half-life extension mutations can enhance binding to neonatal Fc receptor (FcRn) and increase serum half-life compared to wild-type IgG1 (Dall'Acqua et al., J Biol Chem. 2006 Aug 18; 281(33):23514-24. Epub 2006 Jun 21).

[0049] As will be appreciated by those skilled in the art, IgG1 is not responsible for generating monomeric Fc, and IgG4 can also be used here. For example, MedImmune provides a method that can be used in the context of NC-1. IgG4 naturally splits into half antibodies and then rapidly reassembles, a process known as Fab arm exchange, and was first described by van der Neut Kolfschoten et al., Science. 2007 Sep 14; 317(5844): 1554-7; 317(5844): 1554-7, and then by Labrijn et al., Nat Biotechnol. 2009 Aug; 27(8): 767-71. doi: 10.1038 / nbt.1553. Epub 2009 Jul20. For this reason, MedImmune selected IgG4 as the starting point for generating monomeric Fc, because it is easier to prevent such dimerization than IgG1. However, it turns out that some of the mutations found work on both human IgG4 and IgG1 (even mouse IgG1 works in this form). Human IgG4 can be selected as a blocking antibody (i.e., no ADCC or CDC) and IgG1 to activate the immune response. In the case of NC-1-Fc, the inventors have no evidence to show that how much oligomerization of NC-1 can retain the affinity of the compound and maintain the ADCC potential of the compound. If ADCC does not work, then a strategy based on IgG4 may also be adopted.

[0050] Thus, in another embodiment of the methods of the invention, human IgG4 Fc comprising at least one monomer mutation and one or more half-life extending mutations can be used instead of human IgG1 Fc; see, e.g., US 2013 / 0177555.

[0051] In addition to human IgG1 Fc containing the monomer mutation F405R and the half-life extension mutations M252Y, S254T and T256E (Shan L,Colazet M, Rosenthal KL, Yu XQ, Bee JS, Ferguson A, et al. (2016) Generation and Characterization of an IgG4 Monomeric Fc Platform. PLoS ONE 11(8):e0160345. https: / / doi.org / 10.1371 / journal.pone.0160345), the inventors have so far tested the method reported by Wilkinson et al. (MAbs. 2013 May 1; 5(3): 406–417; US20130177555) as an alternative.

[0052] Studies presented by Ying et al. provide other possible alternative approaches (The Journal of BiologicalChemistry 287, 19399-19408; and Tianlei Ying, Yang Feng, Yanping Wang, WeizaoChen & Dimiter S. Dimitrov (2014). Monomeric IgG1 Fc molecules displaying unique Fc receptor interactions can be used to treat inflammation-mediated diseases mAbs, 6:5, 1201-1210), US 2013 / 0177555 and US 9,200,060.

[0053] Preferably, the human IgG1 Fc comprises the monomer mutation F405R and the half-life extending mutations M252Y, S254T and T256E (YTE).

[0054] The aforementioned fusion protein is expressed in a host cell under conditions that allow the formation of a protein oligomer, wherein the protein oligomer comprises at least two or more monomeric human NC-1-Fc proteins.

[0055] Preferably, the monomeric human NC-1-Fc protein comprises SEQ ID NO: 37 or consists of it.

[0056] Finally, protein oligomers of human NC-1-Fc proteins containing at least two or more monomers are obtained from host cells by methods known in the art; see, for example, Sambrook et al., Molecular cloning: a laboratory manual / Sambrook, Joseph; Russell, David W. --. 3rd ed. -- NewYork: Cold Spring Harbor Laboratory, 2001; Ausubel et al., Current Protocolsin Molecular Biology. As shown in the following examples, the inventors have accepted trimer molecules composed of three NC1-Fc (monomers).

[0057] In a preferred embodiment, the method of the invention is an in vitro method.

[0058] The present inventors were able to successfully prepare monomeric NC-1-Fc fusion proteins using monomeric mutant F405R and YTE mutant IgG1 Fc domains carrying half-life extension mutations M252Y, S254T and T256E (see, e.g., WO2013 / 096291A2) (see Fig.10 and Table 4, Molecule #4, SEQ ID NO: 37). Oligomerization via the NC-1 oligomerization domain results in the formation of trimeric NC-1, each with a monomeric Fc domain. This fusion protein can be efficiently produced by transient expression (30 mg / l). Under standard conditions, the monomeric Fc is unable to induce antibody-dependent cell-mediated cytotoxicity (ADCC). It is currently being investigated if trimerization via the Fc domain of NC-1 can increase avidity, possibly leading to antibody-dependent cell-mediated cytotoxicity (ADCC).

[0059] In summary, the inventors have finally managed to achieve a system for large-scale production of NC-1 after many constructs and classical labels have failed for nearly a decade so far. In addition, the inventors unexpectedly found that compared with homodimeric Fc-endostatin (Fc-ES, US8703908B2, US20130165634A1), the protein oligomers with heterodimeric NC-1-Fc or monomeric NC-1-Fc presented herein showed higher efficacy in initial preclinical tumor models and fibrosis studies, as shown in the following examples and Figure 7 and 8 Confirmed.

[0060] As used herein, the term "protein" or "polypeptide" or "(poly)peptide" or "peptide" (all terms are used interchangeably if not otherwise indicated) includes isolated and / or purified (poly)peptides that are substantially free of other host cell polypeptides. The term "peptide" as referred to herein comprises at least 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 45, 50, 60, 70, 80, 90, 100, 150, 200, 250, 300 or even more amino acid residues in which an alpha carboxyl group of one is bound to an alpha amino group of another. As used and contemplated herein, post-translational modifications of a protein or peptide are modifications of the newly formed protein or peptide and may involve deletions, substitutions and additions of amino acids, chemical modifications of certain amino acids, such as amidation, acetylation, phosphorylation, glycosylation, pyroglutamate formation, oxidation / reduction of the sulfonate group on methionine, or the addition of similar small molecules.

[0061] The term "protein" or "peptide" as used herein encompasses peptidomimetics. As known in the art, peptidomimetics are compounds whose basic elements (pharmacophores) mimic natural peptides or proteins in 3D space and retain the ability to interact with biological targets (such as fibronectin, VEGF, MMP-2 and / or MMP-9)) and produce the same biological effects (e.g., anti-fibrotic activity, anti-angiogenic activity, anti-invasive / anti-metastatic activity, vascular permeability reduction activity, anti-inflammatory and anti-tumorigenic activity, antibody-dependent cell-mediated cytotoxicity (ADCC) induction); see, for example, Vagner et al. 2008, Current Opinion in Chemical Biology 12, Pages 292–296 for a review. Peptide mimetics are designed to avoid some problems associated with natural peptides, such as stability to proteolysis (duration of biological activity) and poor bioavailability. Certain other properties, such as selectivity to the above-mentioned biological targets or the efficacy of biological activity, such as the above-mentioned biological activity, can usually be substantially improved.

[0062] As used herein, protein or peptide modifications include synthetic embodiments of the (poly)peptides described herein. In addition, analogs (non-peptide organic molecules), derivatives (chemically functionalized (poly)peptide molecules obtained starting with the disclosed (poly)peptide sequences) and variants (homologues) of these proteins can be used for the means and methods described herein and for medical and diagnostic purposes. Each (poly)peptide disclosed herein consists of a sequence of amino acids, which can be naturally occurring or other L- and / or D-amino acids. (Poly)peptides can be modified by a variety of chemical techniques to produce derivatives having substantially the same biological activity as unmodified (poly)peptides (e.g., anti-fibrotic activity, anti-angiogenic activity, anti-invasive / anti-metastatic activity, vascular permeability reduction activity, anti-inflammatory and anti-tumorigenic activity, and / or induction of antibody-dependent cell-mediated cytotoxicity (ADCC)) and optionally having other desired properties. For example, the carboxylic acid groups of proteins, whether at the carboxyl terminal or on the side chain, can be provided in the form of a salt of a pharmaceutically acceptable cation or esterified to form a C1-C16 ester, or converted to an amide of the formula NR1R2, wherein R1 and R2 are each independently H or C1-C16 alkyl, or combined to form a heterocyclic ring, such as a 5-membered or 6-membered ring. Whether at the amino terminal or on the side chain, the amino group of the polypeptide can be in the form of a pharmaceutically acceptable acid addition salt, such as HCl, HBr, acetic acid, benzoic acid, toluenesulfonic acid, maleic acid, tartaric acid and other organic salts, or can be modified to a C1-C16 alkyl or dialkylamino group or further converted to an amide. The hydroxyl groups of the polypeptide side chains can be converted to C1-C16 alkoxy or C1-C16 ester using recognized techniques. The phenyl and phenol rings of the polypeptide side chains can be substituted with one or more halogen atoms (e.g., fluorine, chlorine, bromine or iodine), or substituted with C1-C16 alkyl, C1-C16 alkoxy, carboxylic acid and its ester, or amide of such carboxylic acid. The methylene groups of the polypeptide side chains can be extended to homologous C2-C4 alkylene groups. The thiol can be protected with any of a number of recognized protecting groups (e.g., acetamide groups). Those skilled in the art will also recognize methods of introducing cyclic structures into the (poly)peptides of the present invention to select and provide conformational constraints on the structure (which results in enhanced stability).

[0063] The proteins or (poly)peptides mentioned herein may also be fusion proteins. The term "fusion protein" as used herein refers to a chimeric protein (literally, composed of parts from different sources) that is produced by connecting two or more genes that originally encoded separate proteins. Translation of the fusion gene produces a single or multiple polypeptides having functional properties derived from each original protein. For example, the method of the present invention for producing a protein oligomer comprising at least two heterodimeric human NC-1-Fc proteins uses a fusion protein comprising, from N- to C-terminus, human NC-1 from collagen 18 fused to human IgG1 Fc with a "knob" mutation. This fusion protein is then heterodimerized with an IgG1 Fc domain comprising an appropriate "hole" mutation, such as Fig. 9 To provide another example of a fusion protein, the method of the present invention for producing a protein oligomer comprising at least two monomeric human NC-1-Fc proteins uses a fusion protein comprising, from N- to C-terminus, human NC-1 from collagen 18 fused to human IgG1 Fc, wherein the human IgG1 Fc comprises a monomer mutation F405R and half-life extension mutations M252Y, S254T and T256E.

[0064] Fusion proteins as defined herein can be prepared by chemical synthesis or recombinant molecular biology techniques well known to those skilled in the art. This applies, mutatis mutandis, to the isolation of fusion proteins from host cells or supernatants. See, e.g., Sambrook et al., Molecular cloning: a laboratory manual / Sambrook, Joseph; Russell, David W. --. 3rd ed. -- New York: Cold Spring Harbor Laboratory, 2001; Ausubel, Current Protocols in Molecular Biology, Green Publishing Associates and Wiley Interscience, NY (1994).

[0065] The fusion protein defined herein can also contain a purification tag, a label or another therapeutic agent, such as an anti-fibrotic agent, an anti-angiogenic agent and / or an anti-tumorigenic agent, etc. "Labels" as referred to herein are detectable compounds or compositions that are directly or indirectly conjugated to another molecule, such as a fusion protein or protein oligomer as defined herein, to facilitate detection of the molecule. Specific non-limiting examples of labels include fluorescent labels, enzymatic connections and radioisotopes known in the art. In one embodiment, there may be a protease cleavage site and / or a joint (i.e., a protease cleavage site; or a joint; or both a protease cleavage site and a joint; or a joint comprising a protease cleavage site) between NC-1 and the Fc domain. If desired, the protease cleavage site can be used to cut off the Fc domain by treatment with a protease such as enterokinase or thrombin. For example, the Fc domain can be used as a label for expression and purification, and NC-1 can be separated after cleavage with a protease. As is well known to those skilled in the art, in addition to the basic role of connecting functional domains together (such as in flexible and rigid linkers) or releasing free functional domains in vivo (such as in cleavable linkers in vivo), the linker can also provide many other advantages for the production of the fusion protein, such as improving biological activity, improving expression yield and achieving desired pharmacokinetic characteristics. The linker can be, for example, a protein / peptide linker, such as a polyglycine linker or other linkers known in the art (see, for example, Chen et al., Adv Drug Deliv Rev. 2013; 65(10): 1357–1369). Obviously, the linker can be designed in a manner that includes a protease cleavage site. As will be appreciated by those skilled in the art, the fusion protein defined herein does not include a protease cleavage site and / or linker for clinical use, or a protease cleavage site and / or linker removed from the fusion protein before and for clinical use.

[0066] Collagen 18 consists of a central, interrupted triple-helical domain flanked by larger non-triple-helical globular structures at the N-terminus (NC-11 domain) and C-terminus (NC-1 domain) (Oh et al., PNAS 1994, 91, 4229; Oh et al., Genomics 1994, 19, 494; Abe et al. 1993, Biochem. Biophys. Res. Commun. 196, 576). Type XVIII collagen belongs to a unique and novel subclass of the collagen superfamily, for which the name "MULTIPLEXIN family" has been proposed.

[0067] Oh et al. have described the cloning of mouse and human collagen 18 proteins (supra). The nucleotide and amino acid sequences of mouse collagen 18 are shown in accession number NM_001109991.1, while the corresponding human sequence is shown in NM_030582.3. In addition, the amino acid sequences of mouse and human collagen 18 are shown in SEQ ID NO: 1 and 2, respectively.

[0068] As used herein, the "NC-1 domain" (or simply NC-1 or NC1) is derived from or from the C-terminus of collagen 18 and includes (i) an N-terminal association region (about 50 amino acid residues) (ii) a central protease-sensitive hinge region (about 70 amino acid residues) and / or (iii) a C-terminal stabilized endostatin domain (about 180 amino acid residues) (Sasaki et al., 1998, EMBO J. 17, 4249; WO 2013 / 026913 and WO 2017 / 093569). As used herein, the NC-1 domain (or simply NC-1 or NC1) preferably includes (i) an N-terminal association region, (ii) a central protease-sensitive hinge region and (iii) a C-terminal stabilized endostatin domain. In other preferred embodiments, the NC-1 domain used herein comprises (i) an N-terminal association region and (iii) a C-terminal stabilized endostatin domain, wherein the central protease-sensitive hinge region (ii) is missing. Due to the lack of this region sensitive to proteases, such NC-1 domains without (ii) a central protease-sensitive hinge region are particularly useful for clinical and therapeutic purposes.

[0069] The amino acid sequence of the NC-1 domain of mouse collagen 18 is depicted in SEQ ID NO: 3, while the corresponding sequence of the NC-1 domain of human collagen 18 sequence is shown in SEQ ID NO: 4. Preferably, the NC-1 domain of collagen 18 is human or derived from human.

[0070] As used herein, a (poly)peptide "derived from" the NC-1 domain means that such a (poly)peptide is identical to or differs from the corresponding amino acid sequence of a natural (poly)peptide in the NC-1 domain by 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 15, 20, 25, 30, 35, 40, 50 or even more amino acid residues, while at least maintaining (or even exceeding) the biological activity of the corresponding NC-1 domain (as described elsewhere herein), such as oligomerization properties, anti-fibrotic activity, anti-angiogenic activity, anti-invasion / anti-metastasis activity, vascular permeability reduction activity, anti-inflammatory and / or anti-tumorigenic activity. For example, compared with the natural human NC-1 domain, the NC-1 domain derived from human collagen 18 can carry 1, 2, 3, 4, 5 or even more mutations. Such mutations are well known in the art and include, for example, substitutions, additions and / or deletions in the nucleic acid or amino acid sequence of natural (or wild-type) human NC-1. Tests for determining such biological activity are described in, for example, WO 2013 / 026913 and WO 2017 / 093569. The term "derived from" a (poly)peptide of a human NC-1 domain is referred to as a variant of a human NC-1 domain. Preferably, the variant sequence has at least 50%, at least 60%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98% or at least 99% identity over the entire length with the specific amino acid sequence of the human NC-1 domain shown in SEQ ID NO: 4. Means and methods for performing sequence alignment and determining sequence identity are known in the art and are described elsewhere herein.

[0071] The "association domain" or "association region" or "oligomerization domain" (all terms are interchangeable) of the human NC-1 domain of about 10 to about 60 amino acid residues of the amino acid sequence shown in SEQ ID NO: 4 is responsible for the non-covalent trimerization of NC-1 monomers to form a globular trimer. Therefore, the association domain acts as a trimerization domain. The "hinge region" sensitive to proteolytic cleavage comprises about 61 to about 129 amino acid residues of the amino acid sequence shown in SEQ ID NO: 4. The compact "endostatin domain" comprises about 130 to about 308 amino acid residues of the amino acid sequence shown in SEQ ID NO: 4; see, for example, Sasaki, loc. cit.; Kuo 2001, JCB 152, 1233; Tjin et al. 2005, Cancer Res 65, 3656. The endostatin domain contains a zinc binding site that mediates the binding to zinc and is located at the N-terminus of endostatin (Ding et al., 1998, PNAS 95, 10443; US 7,524,811). Interestingly, this zinc binding site has been shown to be responsible for the anti-tumor / anti-angiogenic activity of endostatin (Boehm et al., 1998, Biochem. Biophys. Res. Commun. 252, 190). The association region in the NC-1 domain and the endostatin domain are connected by a hinge region (see Sasaki et al., supra). The hinge region has been found to be cleaved by, for example, matrix metalloproteinases (MMPs), such as MMP-3, -7, -9, -13 and -20 (Heliasvaara et al., Exp Cell Res 2005, 307,192).

[0072] The term "Fc domain" or "Fc region" as used herein refers to a fragment crystallizable region, which is the tail region of an antibody or immunoglobulin that interacts with cell surface receptors, namely Fc receptors and some proteins of the complement system. This property allows antibodies to activate the immune system. In IgG, IgA and IgD antibody isotypes, the Fc domain is composed of two identical protein fragments derived from the second and third constant domains of the two heavy chains of the antibody; IgM and IgE Fc domains contain three heavy chain constant domains (CH domains 2-4) in each polypeptide chain. The Fc domain of IgG carries a highly conserved N-glycosylation site. Glycosylation of the Fc fragment is essential for Fc receptor-mediated activity. The N-glycans attached to this site are mainly complex types of core-fucosylated biantennary structures. In addition, a small amount of these N-glycans also carry bipartite GlcNAc and α-2,6 linked sialic acid residues. It has been found that the fusion of the Fc domain of an immunoglobulin to a protein enhances the production and secretion of the fusion protein in mammalian cells (see, e.g., Lo et al., 1998, Protein Eng. 11, 495, Capon et al., 1989, Nature 337, 525). In addition, the connection of angiogenesis inhibitors to the Fc domain of an immunoglobulin has been shown to increase the half-life of the inhibitors (see, e.g., Capon et al. 1989, Nature 337, 525; Gordon et al., 2001, J. Clin. Oncol. 19, 843; Holash et al., 2002, Proc. Natl. Acad. Sci. USA 99, 11393). However, the Fc domain can be used not only for purification, solubilization and / or detection purposes, but also to advantageously change the biological and pharmacokinetic properties of the fusion protein or protein oligomer, as described herein and in the following examples. In one embodiment, if necessary, the Fc domain can be cut off by treating with a protease such as enterokinase or thrombin. It is apparent to those skilled in the art that, in principle, any IgG isotype can be used to generate the protein oligomer or fusion protein defined herein. Even subfragments or single chains of the Fc domain of IgG can be used to extend the half-life or oligomerization of protein oligomer or fusion protein described herein.

[0073] Preferably, the Fc domain mentioned herein is from human IgG or derived from human IgG, such as human IgG1 or IgG4, more preferably from human IgG1 or derived from human IgG1 (Bergers and Javaherian Science 1999; Lee et alClin Canc Res 2008). The human Fc domain from IgG1 comprises or consists of the amino acid sequence shown in SEQ ID NO: 6 or SEQ ID NO: 24.

[0074] As used herein, "derived from" an Fc domain of human IgG1 or human IgG4 means that such an Fc domain is identical to the corresponding amino acid sequence of a natural (poly)peptide in a human IgG1 Fc domain or a human IgG4 Fc domain or may differ in 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 15, 20, 25, 30, 35, 40, 50 or even more amino acid residues, while changing, maintaining or even exceeding the biological activity of the corresponding natural human IgG1 Fc domain or human IgG4 Fc domain, such as dimerization properties, Fc receptor binding (e.g., binding to FcRn) and modifying the pharmacokinetics of the construct and the induction of antibody-dependent cell-mediated cytotoxicity (ADCC). For example, compared to a natural (or wild-type) human IgG1 Fc domain, an Fc domain derived from a human IgG1 Fc domain may carry 1, 2, 3, 4, 5 or even more mutations. Such mutations are well known in the art and include, for example, substitutions, additions and / or deletions in the nucleic acid or amino acid sequence of a native (or wild-type) human IgG Fc domain. The term "derived from" a (poly)peptide of a human IgG1 Fc domain comprises a variant of a human IgG1 domain. Preferably, the variant sequence has at least 50%, at least 60%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98% or at least 99% identity over the entire length to the specific amino acid sequence of the human IgG1 domain shown in SEQ ID NO: 6 or 24. The term "derived from" a (poly)peptide of a human IgG4 Fc domain comprises a variant of a human IgG1 domain. Preferably, the variant sequence is identical to, for example, that of US20130177555A1. Figure 6 The specific amino acid sequence of the human IgG4 domain shown in has at least 50%, at least 60%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98% or at least 99% identity over its entire length.

[0075] On the one hand, the Fc domain is a "knob-in-hole structure" (KiH)-modified Fc domain. As demonstrated in the following examples, the inventors have used the "knob-in-hole structure" (KiH)-modified Fc domain to produce a protein oligomer comprising a heterodimer NC-1-Fc. The knob-in-hole structure is a well-proven heterodimerization technology for the third constant domain of an antibody. Basically, the concept relies on the modification of the interface between the two CH3 domains where most interactions occur. A bulky residue is introduced into the CH3 domain of an antibody heavy chain, which acts like a key. In the other heavy chain, a "hole" is formed that can accommodate this large residue, imitating a lock. The resulting heterodimer Fc portion can be further stabilized by an artificial disulfide bond. In the process of optimizing the heterodimerization interface, a phage display library is used to evaluate and ultimately optimize various rational designs (including spatial complementarity, KiH, disulfide bonds, and salt bridges juxtaposed with residues of opposite charge on either side of the CH3 domain). Correct heavy chain association resulting in greater than 97% heterodimerization can be achieved by introducing the following six mutations: S354C, T366W in the “knob” heavy chain and Y349C, T366S, L368A, Y407V in the “hole” heavy chain (Klein et al., MAbs. 2012 Nov 1; 4(6): 653–663; Ridgway et al., Protein Eng. 1996 Jul; 9(7): 617-21). Furthermore, properties of antibodies with KiH mutations such as (thermo)stability, FcγR binding and effector functions (e.g., antibody-dependent cell-mediated cytotoxicity (ADCC), FcRn binding) and pharmacokinetic (PK) behavior are not affected. Non-covalent interactions as well as disulfide bonds in the hinge region drive assembly into heterodimers and minimize combinatorial heterogeneity. Suitable KiH engineered Fc domains are described, for example, in SEQ ID NOs. 25, 26, 28 and 30. For example, SEQ ID NO: 25 shows the amino acid sequence of a human IgG1 Fc with the "knob" mutations S354C / T366W, and SEQ ID NO: 26 describes the amino acid sequence of a human IgG1 Fc with the "hole" mutations Y349C / T366S / L368A / Y407V.

[0076] SEQ ID NO: 27 shows the amino acid sequence of a fusion protein comprising human NC-1 fused to human IgG1 Fc with a "knob" mutation (S354C / T366W) via an enterokinase cleavage site and a linker (from N- to C-terminus). The fusion protein is capable of heterodimerizing with human IgG1 Fc with a "hole" mutation Y349C / T366S / L368A / Y407V (SEQ ID NO: 28). Such heterodimers are Fig. 9 shown.

[0077] SEQ ID NO: 29 shows the amino acid sequence of a fusion protein comprising a human IgG1 Fc with a "knob" mutation (S354C / T366W) fused to human NC-1 (from N- to C-terminus) via a linker and an enterokinase site. The fusion protein is capable of heterodimerization with a human IgG1 Fc with a "hole" mutation Y349C / T366S / L368A / Y407V (SEQ ID NO: 30).

[0078] The inventors have found that the Fc domain modified by the "knob-in-hole structure" (KiH) is particularly useful for producing protein oligomers containing at least two or more heterodimeric human NC-1-Fc proteins, as confirmed in the following examples. Therefore, the Fc domain modified by the KiH of human immunoglobulin, such as human IgG1 or human IgG4, is preferably used to produce the protein oligomers.

[0079] On the other hand, the Fc domain is monomeric, i.e., modified by mutation of key residues located on the dimerization interface of IgG1 or IgG4 Fc. How to produce monomeric IgG1 or monomeric IgG4 is known in the art; see, for example, Wang et al., Front. Immunol., 13 November 2017, https: / / doi.org / 10.3389 / fimmu.2017.01545. The inventors have found that monomeric Fc domains are particularly useful in preparing protein oligomers comprising two or more monomeric NC-1-Fc fusion proteins, as demonstrated in the following examples. Preferably, the Fc domain is from IgG1, and the mutation F405R corresponding to the monomer mutation and the mutation M252Y, S254T and T256E corresponding to the half-life extension mutation are included in the monomeric NC-1-Fc fusion protein. Further monomeric NC-1-Fc fusion proteins will be identified using Selexis's RCB technology.

[0080] The term "oligomer" generally refers to a macromolecular complex formed in biochemistry by non-covalent bonding of some macromolecules such as proteins or nucleic acids. By definition, a dimer is a macromolecular complex formed by two molecules such as proteins or peptides that are usually non-covalently bound. This complex can also be formed by a protein domain, which is a part of the protein sequence and can evolve, function and exist independently of the rest of the protein chain. Homodimers are formed by two identical molecules. The underlying process is called homodimerization. Heterodimers are constructed by two different molecules, which are formed by heterodimerization. As known in the art, most dimers or trimers in biochemistry are not connected by covalent bonds except for disulfide bonds. Some proteins contain specialized domains to ensure dimerization, trimerization or oligomerization, which are so-called dimerization, trimerization or oligomerization domains, as defined herein, and are well known in the art. As an example, dimerization can be mediated by the Fc domain of an immunoglobulin or by disulfide bonds or both or other means known in the art. For example, Fig. 9 A heterodimeric human NC-1-Fc protein is shown in FIG. Here, the NC-1-Fc knob dimerizes with the Fc hole.

[0081] Therefore, a trimer is a macromolecular complex formed by three peptides, proteins or protein domains that are usually non-covalently bound. Homotrimers are formed by three identical molecules, while heterotrimers are composed of three different molecules. For example, collagen 18 is a homotrimeric protein. A tetramer consists of four molecules, a pentamer consists of five molecules, and so on. In these cases, as described above, the formation of the complex is usually mediated by an oligomerization domain. For example, for trimerization, the native association region (or oligomerization domain) within the NC-1 domain can be used to mediate the trimerization of NC-1 of collagen 18, because the native association region within the NC-1 domain of collagen 18 acts as a trimerization domain.

[0082] In the context of the present invention, "oligomer" is understood as a "protein oligomer" that contains some monomer units, for example, two, three, four, five or even more monomer units. Therefore, the oligomer can be, for example, a dimer, a trimer, a tetramer, a pentamer, etc. Preferably, the oligomer is a homodimer, a homotrimer, etc. The monomer unit (or simply monomer) can be, for example, a heterodimeric human NC-1-Fc protein or a monomer NC-1-Fc that can form a trimer through the oligomerization domain of NC-1. Then, the monomer unit "heterodimeric human NC-1-Fc protein" forms a trimer via the oligomerization (or association) domain of NC-1; see Fig. 9 Likewise, the monomer unit "monomer NC-1-Fc" forms a trimer via the oligomerization domain of NC-1, such as Fig.10 shown.

[0083] The protein oligomer of the present invention has at least one of the following biological activities mediated by oligomeric NC-1, preferably at least two, more preferably at least three, four, five, and particularly preferably all: anti-fibrotic activity, anti-angiogenic activity, anti-invasion / anti-metastasis activity, vascular permeability reduction activity, anti-inflammatory and / or anti-tumorigenic activity. Tests for the above biological activities and properties are known in the art; see, for example, WO 2013 / 026913 and WO 2017 / 093569. In addition, the protein oligomer of the present invention preferably has the ability to induce antibody-dependent cell-mediated cytotoxicity (ADCC) via a dimerized or oligomerized Fc domain. The means and methods for testing ADCC are well described in the literature (see, e.g., Yamashita et al., Scientific Reports volume 6, Article number: 19772 (2016); doi:10.1038 / srep19772; Parekh et al., MAbs. 2012 May 1; 4(3): 310–318. doi:10.4161 / mabs.19873; Chung et al., J Immunol Methods. 2014 May;407:63-75. doi:10.1016 / j.jim.2014.03.021. Epub 2014 Apr 3; Tada et al. (2014) PLoS ONE 9(4):e95787. https: / / doi.org / 10.1371 / journal.pone.0095787; https: / / www.essenbioscience.com / media / uploads / files / 8000-0580-A00-Immune_Cell_Killing--Tumor_Spheroids_Assay_Protocol.pdf).

[0084] Protein oligomers as defined herein can be produced by chemical synthesis or recombinant molecular biology techniques and can be isolated from host cells or cell culture supernatants by methods well known in the art; see, for example, Sambrook et al., Molecular cloning: a laboratory manual / Sambrook, Joseph; Russell, David W. --. 3rd ed. -- New York: Cold Spring Harbor Laboratory, 2001.

[0085] As used herein, the term "protein oligomer" also includes protein preparations that contain protein oligomers and additionally other proteins, agents or compounds. For example, the protein oligomers as defined herein can be administered to a subject in need, preferably a human subject, using one or more other anti-fibrotic, anti-angiogenic and / or tumorigenic proteins, compounds or agents in a combination regimen. Drug combinations are generally more effective than using one drug alone to combat the diseases defined herein. For example, the protein oligomers as defined herein can be used in combination with angiostatin or angiostatin fusion protein (such as angiostatin linked to the Fc domain of an immunoglobulin), or with inhibitors of other pathways associated with the fibrotic process (including, for example, TGF-β, PDGF, VEGF, mTOR, CTGF, integrins, matrix metalloproteinases, anti-inflammatory agents such as cyclooxygenase, IKK / NFkB, JAK / STAT and / or steroid inhibitors of Pi3K signaling).

[0086] The present invention further relates to a protein oligomer comprising at least two, preferably three heterodimeric human NC-1-Fc proteins for treating, ameliorating or preventing diseases associated with angiogenesis, wherein the heterodimeric human NC-1-Fc includes: (i) a fusion protein comprising, from N- to C-terminus, human NC-1 from collagen 18 fused to human IgG1 Fc comprising the "knob" mutation S354C / T366W, or human IgG1 Fc fused to human NC-1 from collagen 18 comprising the "knob" mutation S354C / T366W, and (ii) human IgG1 Fc having the "hole" mutation Y349C / T366S / L368A / Y407V.

[0087] Preferably, the human IgG1 Fc with the "knob" mutation S354C / T366W comprises SEQ ID NO: 25, and the human IgG1 Fc with the "hole" mutation Y349C / T366S / L368A / Y407V comprises SEQ ID NO: 26.

[0088] More preferably, the fusion protein (i), which comprises, from N- to C-terminus, human NC-1 from collagen 18 fused to human IgG1 Fc with a "knob" mutation comprises SEQ ID NO: 27, 29, or 31 or consists thereof, and (ii) human IgG1 Fc with a "hole" mutation comprises SEQ ID NO: 28, 30, or 32 or consists thereof.

[0089] Particularly preferred is a fusion protein (i) comprising or consisting of SEQ ID NO: 31, said fusion protein comprising, from N- to C-terminus, human NC-1 from collagen 18 fused to a human IgG1 Fc with a "knob" mutation, and (ii) a human IgG1 Fc with a "hole" mutation comprising or consisting of SEQ ID NO: 32 (see Fig. 9 and Table 1, molecule #1).

[0090] As shown in the following examples and Figure 7 and Figure 8 As demonstrated, it has been found that such protein oligomers are particularly useful as pharmaceutical compositions for treating, improving or preventing diseases associated with angiogenesis.

[0091] As used herein, a "pharmaceutical composition" can be used in a therapeutically effective dose for the treatment of a non-human or preferably a human of the diseases mentioned herein. A "subject" referred to herein is preferably a human suffering from the diseases mentioned herein.

[0092] The protein oligomers comprising at least two heterodimeric human NC-1-Fc proteins defined herein, the monomeric human NC-1-Fc proteins of the present invention or the protein oligomers comprising at least two monomeric human NC-1-Fc proteins of the present invention are active ingredients of pharmaceutical compositions or drugs (the two terms are used interchangeably), and in one aspect, are administered in conventional dosage forms prepared by combining the drug with a standard pharmaceutical carrier according to conventional procedures. These procedures may involve mixing, granulating, and compressing or dissolving the ingredients suitable for the desired formulation. It should be understood that the form and characteristics of the pharmaceutically acceptable carrier or diluent depend on the amount of the active ingredient combined therewith, the route of administration, and other well-known variables.

[0093] The carrier must be acceptable in the sense that it is compatible with the other ingredients of the preparation and harmless to its recipient. The pharmaceutical carrier used can include solid, gel or liquid. The example of a solid carrier is lactose, terra alba, sucrose, talc, gelatin, agar, pectin, gum arabic, magnesium stearate, stearic acid, etc. The example of a liquid carrier is phosphate buffered saline, syrup, oil, water, emulsion, various types of wetting agents, etc. Similarly, the carrier or diluent can include time-delay materials well known in the art, such as single glyceryl monostearate or glyceryl distearate or together with wax. The suitable carrier includes those mentioned above and other carriers well known in the art, referring to, for example, Remington's Pharmaceutical Sciences, Mack Publishing Company, Easton, Pennsylvania.

[0094] The diluent is selected so as not to affect the biological activity of the pharmaceutical composition. The biological activity has been defined elsewhere herein. Examples of such diluents are distilled water, physiological saline, Ringer's solution, dextrose solution and Hank's solution. In addition, the pharmaceutical composition or formulation may also include other carriers, adjuvants or non-toxic, non-therapeutic, non-immunogenic stabilizers and the like.

[0095] Preferably, the protein oligomer comprising at least two heterodimeric human NC-1-Fc proteins defined herein, the monomeric human NC-1-Fc protein of the present invention or the protein oligomer comprising at least two monomeric human NC-1-Fc proteins of the present invention is formulated into a pharmaceutical composition, which can be administered by standard routes. Typically, the pharmaceutical composition can be administered topically, transdermally, intraperitoneally, intracranial / intrathecally, intravitreal, intraventricularly, intracerebrally, intravaginally, intrauterinely, orally, rectally or parenterally (e.g., intravenously, intraspinally, subcutaneously or intramuscularly).

[0096] Preferably, the protein oligomer comprising at least two heterodimeric human NC-1-Fc proteins defined herein, the monomeric human NC-1-Fc protein of the present invention or the protein oligomer comprising at least two monomeric human NC-1-Fc proteins of the present invention is administered intravenously, subcutaneously, intracranial / intravarianally, intravitreally or intraperitoneally.

[0097] The therapeutically effective dose refers to the amount of the protein oligomer containing at least two heterodimeric human NC-1-Fc proteins defined herein, the monomeric human NC-1-Fc protein of the present invention or the protein oligomer containing at least two monomeric human NC-1-Fc proteins of the present invention used in the pharmaceutical composition, which prevents, improves or treats the symptoms associated with the diseases mentioned in this specification. The therapeutic efficacy and toxicity of the compound can be determined by standard drug procedures in cell culture or experimental animals, such as ED50 (the therapeutically effective dose in 50% of the population) and LD50 (the lethal dose for 50% of the population). The dose ratio between therapeutic and toxic effects is the therapeutic index, which can be expressed as the ratio LD50 / ED50.

[0098] The dosage regimen will be determined by the attending physician and other clinical factors. As is well known in the medical field, the dosage for any one patient depends on many factors, including patient size, body surface area, age, specific compound to be administered, sex, time and route of administration, general health, and other drugs administered simultaneously. Progress can be monitored by regular assessments.

[0099] Preferably, the protein oligomer comprising at least two heterodimeric human NC-1-Fc proteins defined herein, the monomeric human NC-1-Fc protein of the present invention or the protein oligomer comprising at least two monomeric human NC-1-Fc proteins of the present invention is administered at a concentration of about 1 to 100 mg / kg body weight. More preferably, the concentration is about 5 to 75 mg / kg or about 10 to 50 mg / kg, most preferably about 15 mg / kg body weight.

[0100] The drugs or pharmaceutical compositions mentioned herein are administered at least once to treat or improve or prevent the diseases described in this specification. However, the drugs may be administered more than once, for example, twice, three times, four times, five times, six times or even more frequently. For example, the drugs or pharmaceutical compositions referred to herein may be administered once a day or every second, third, fourth, fifth or sixth day or once a week. It may also be administered once every second week, every three weeks or every four weeks.

[0101] Preferred target plasma concentrations are about 5 to 100, 10 to 50, or 15 micrograms per ml of blood.

[0102] Specific pharmaceutical compositions are prepared in a manner well known in the pharmaceutical art and contain at least one of the above-mentioned active compounds mixed with or otherwise associated with a pharmaceutically acceptable carrier or diluent. To prepare these specific pharmaceutical compositions, the active compound is usually mixed with a carrier or diluent. The resulting preparation should be suitable for the mode of administration. Dosage recommendations should be stated in the prescriber or user instructions so that dosage adjustments can be predicted based on the recipient in question.

[0103] In another aspect of the present invention, the pharmaceutical composition may contain, in addition to the protein oligomer comprising at least two heterodimeric human NC-1-Fc proteins as defined herein, a drug other than the monomeric human NC-1-Fc protein of the present invention or the protein oligomer comprising at least two monomeric human NC-1-Fc proteins of the present invention, which is added to the drug during the drug formulation process. For example, it can be used with angiostatin in a combination regimen. In addition, on the other hand, a combination with recently approved fibrosis regulators such as VEGF / PDFG RTKi (e.g., Nindetanib), specific and nonspecific inhibitors of TGF-beta signaling (Perfinidone) and regulators of integrin signaling (Cilengitide, or anti-alphaV abizumab) or regulators of inflammation (leukocyte infiltration, cytokine inhibitors, antibodies against subsets) is envisioned. Therefore, in the preferred embodiment of the protein oligomer comprising at least two heterodimeric human NC-1-Fc proteins defined herein, the monomeric human NC-1-Fc protein of the present invention or the protein oligomer comprising at least two monomeric human NC-1-Fc proteins of the present invention, the protein oligomer or monomeric human NC-1-Fc protein further comprises angiostatin (US 8,206,718). In a specific embodiment, angiostatin is Fc-angiostatin or angiostatin-Fc fusion protein, preferably a human fusion protein.

[0104] It should be understood that the preparation of the pharmaceutical composition is carried out under GMP standardized conditions, etc. to ensure the quality, drug safety and efficacy of the pharmaceutical preparation.

[0105] The term "treatment" as used herein means improving or even eliminating one or more symptoms associated with the diseases mentioned herein by administering a protein oligomer or fusion protein as defined herein to a subject in need thereof.

[0106] The term "alleviation" mentioned herein refers to the behavior of improving or ameliorating the disease mentioned herein by administering the protein oligomer or fusion protein described herein. Improvement can also be regarded as the slowing down or stopping of the progression of the disease.

[0107] The term "prevention" as used herein refers to avoiding the occurrence or recurrence of the diseases mentioned herein by administering the protein oligomer or fusion protein as defined herein.

[0108] The angiogenesis-related disease is preferably selected from the group consisting of: angiogenesis-dependent cancers, including solid tumors, melanoma, tumor metastases, hematogenous tumors (such as leukemia), benign tumors (such as hemangioma), acoustic neuroma, neurofibroma, trachoma, pyogenic granuloma; rheumatoid arthritis; psoriasis; ocular angiogenesis diseases such as diabetic retinopathy, retinopathy of prematurity, macular degeneration, corneal graft rejection, neovascular glaucoma, retrolental fibroplasia, rubeosis; Osler-Webber syndrome; myocardial angiogenesis; plaque neovascularization; telangiectasia; hemophilic joints; angiofibroma; wound granulation; diseases with excessive or abnormal stimulation of endothelial cells such as intestinal adhesions, atherosclerosis, scleroderma, hypertrophic scars (keloids); diseases with angiogenesis as a pathological consequence, for example cat scratch disease (Rochele minalia quintosa) and ulcers (Helicobacter pylori).

[0109] Diseases associated with angiogenesis are described and characterized in WO 2013 / 026913. The definitions and embodiments of WO 2013 / 026913 apply to the present application with necessary changes. Diseases described in WO 2017 / 093569 as - or which may be considered as - angiogenesis-related diseases are explicitly excluded and abandoned from the scope of medical use of the present invention, wherein the human NC-1-Fc protein oligomers comprising at least two heterodimers defined herein are used to treat, improve or prevent diseases associated with angiogenesis as defined herein.

[0110] The present invention also relates to a monomeric NC-1-Fc protein, which comprises human NC-1 from collagen 18 and an Fc domain from human IgG1, wherein the Fc domain from human IgG1 comprises at least one monomer mutation, and one or more half-life extension mutations. For example, the monomer mutation may be F405R, and the half-life extension mutation may be M252Y, S254T and / or T256E.

[0111] Preferably, the present invention relates to a monomeric NC-1-Fc protein comprising, from N- to C-terminus, human NC-1 from collagen 18 and an Fc domain from human IgG1, or an Fc domain from human IgG1 and human NC-1 from collagen 18, wherein the Fc domain from human IgG1 comprises a monomeric mutation F405R and half-life extension mutations M252Y, S254T and T256E. More preferably, the monomeric NC-1-Fc protein comprises or consists of SEQ ID NO: 37 (see Fig.10 , Table 4, molecule #4).

[0112] In another embodiment, the monomeric NC-1-Fc protein comprises human NC-1 from collagen 18 and an Fc domain from human IgG4, wherein the Fc domain from human IgG4 comprises at least one monomer mutation and one or more half-life extending mutations.

[0113] The present invention further provides a protein oligomer comprising at least two monomeric human NC-1-Fc fusion proteins comprising human NC-1 from collagen 18 and an Fc domain from human IgG1 or human IgG4, wherein the Fc domain from human IgG1 or IgG4 comprises at least one monomer mutation and one or more half-life extension mutations.

[0114] Preferably, the monomer mutation in human IgG1 Fc is F405R, and the half-life extension mutations are M252Y, S254T and T256E.

[0115] In addition, the present invention provides a protein oligomer, which is prepared by a method for producing a protein oligomer comprising at least two monomeric human NC-1-Fc proteins, wherein the method comprises:

[0116] a) culturing a host cell expressing a fusion protein under conditions that allow the formation of a protein oligomer comprising at least two, and preferably three monomeric human NC-1-Fc proteins, the fusion protein comprising, from N- to C-terminus, human NC-1 from collagen 18 fused to human IgG1 Fc, or an Fc domain from human IgG1 and human NC-1 from collagen 18, wherein the human IgG1 Fc comprises at least one monomer mutation, preferably the monomer mutation F405R, and one or more half-life extension mutations, preferably the half-life extension mutations M252Y, S254T and T256E, and

[0117] b) obtaining a protein oligomer comprising at least two, and preferably three monomeric human NC-1-Fc proteins from the host cell of step a).

[0118] Preferably, the protein oligomer comprising at least two monomeric human NC-1-Fc fusion proteins binds to fibronectin, VEGF, MMP-2 and / or MMP-9, more preferably binds to each of these proteins.

[0119] Preferably, the Fc domain contained in the protein oligomer comprising at least two monomeric human NC-1-Fc fusion proteins is capable of inducing antibody-dependent cell-mediated cytotoxicity (ADCC).

[0120] The present invention also relates to a protein oligomer comprising at least two monomeric NC-1-Fc proteins, wherein the monomeric NC-1-Fc protein comprises, from N- to C-terminus, human NC-1 from collagen 18 and an Fc domain from human IgG1, or an Fc domain from human IgG1 and human NC-1 from collagen 18, wherein the Fc domain from human IgG1 comprises at least one monomer mutation and one or more half-life extension mutations, which is used as a drug.

[0121] It has been found that the protein oligomer of the present invention comprising three monomeric NC-1-Fc proteins is surprisingly useful as a pharmaceutical composition for treating, ameliorating or preventing fibrosis in a pulmonary fibrosis model, as shown in the following examples and Fig.16 (See NC-1-Fc (monomer)).

[0122] The present invention also relates to a protein oligomer comprising at least two monomeric NC-1-Fc proteins, wherein the monomeric NC-1-Fc protein comprises human NC-1 from collagen 18 and an Fc domain from human IgG4 at the N- to C-terminus, or an Fc domain from human IgG4 and human NC-1 from collagen 18, wherein the Fc domain from human IgG4 comprises at least one monomer mutation and one or more half-life extension mutations, for use as a drug.

[0123] Preferably, the protein oligomer comprises at least two monomeric human NC-1-Fc fusion proteins, which fusion proteins comprise, from N- to C-terminus, human NC-1 from collagen 18 and an Fc domain from human IgG1, or an Fc domain from human IgG1 and human NC-1 from collagen 18, wherein the Fc domain from human IgG1 comprises a monomer mutation F405R and half-life extension mutations M252Y, S254T and T256E.

[0124] In addition, the present invention relates to a protein oligomer produced by the method of the present invention for producing a protein oligomer comprising at least two monomeric human NC-1-Fc proteins, which is used as a medicament.

[0125] Preferably, the drug is used to treat, improve or prevent a disease selected from the following:

[0126] (i) Angiogenesis-related diseases including the following: angiogenesis-dependent cancers, including solid tumors, melanoma, tumor metastases, hematogenous tumors (such as leukemia), benign tumors (such as hemangiomas), acoustic neuromas, neurofibromas, trachoma, pyogenic granulomas; rheumatoid arthritis; psoriasis; ocular angiogenesis diseases such as diabetic retinopathy, retinopathy of prematurity, macular degeneration, corneal graft rejection, neovascular glaucoma, retrolental fibroplasia, rubeosis; Osler-Webber syndrome; myocardial angiogenesis; plaque neovascularization; telangiectasia; hemophilic joints; angiofibromas; wound granulation; diseases with excessive or abnormal stimulation of endothelial cells such as intestinal adhesions, atherosclerosis, scleroderma, hypertrophic scars (keloids); diseases with angiogenesis as a pathological consequence, such as cat scratch disease (Rocheleminalia quintosa) and ulcers (Helicobacter pylori);

[0127] (ii) fibrosis or fibrosis-related diseases, including skin fibrosis, preferably scleroderma; keloids or keloids; hypertrophic scars; morphea; fibrosis due to graft-versus-host disease; subepithelial fibrosis; endomyocardial fibrosis; uterine fibrosis; myelofibrosis; retroperitoneal fibrosis; nephrogenic systemic fibrosis; postoperative scars; asthma; cirrhosis / liver fibrosis; fibrosis due to abnormal wound healing; glomerulonephritis; multifocal fibrosclerosis; radiation-induced fibrosis, preferably radiation-induced pneumonitis or radiation-induced pulmonary fibrosis; chemotherapy-induced or drug-induced fibrosis, for example due to mTOR or EGFR kinase inhibition; common or idiopathic pulmonary fibrosis; fibrosis due to autoimmune diseases, such as lupus, fibrosis / fibrogenesis associated with tumors and cancer, chronic inflammation associated with organ fibrosis, for example by viral stimulation or transplantation; organ fibrosis as a terminal stage of chronic kidney disease, long-term dialysis or diabetes;

[0128] (iii) vascular endothelial growth factor (VEGF)-related diseases, including benign pathophysiological conditions that depend on dysregulated VEGF levels, such as wet macular degeneration, endometriosis, bronchial asthma and diabetes, enhanced VEGF-induced vascular permeability (e.g., enhanced permeability after brain tissue irradiation, "radionecrosis"), changes in vascular tone (e.g., hypertension), rheumatoid arthritis, as well as malignant VEGF-dependent diseases, such as renal cell carcinoma and other VEGF-addicted tumors, VEGF-dependent ascites development, VEGF-dependent immune system suppression, such as recruitment and microenvironment training of bone marrow-derived cells (BMDCs), bone marrow-derived suppressor cells (MdSCs), immature dendritic cells, etc.; and

[0129] (iv) Matrix metalloproteinase (MMP)-associated diseases, including both benign and malignant diseases in which MMP activation contributes to pathophysiology, for example, MMP activation is intrinsic to local tumor invasion and cancer metastasis in tumors with high rates of local treatment failure (e.g., glioblastoma, pancreatic cancer, lung cancer), as well as acquired enhanced MMP activation due to overt immune responses induced by treatment-induced selective pressures (e.g., tumor hypoxia and fibrosis after radiation therapy), autoimmune diseases, and chronic inflammatory diseases.

[0130] The definitions, embodiments and explanations of the therapeutic use of the present invention regarding the protein oligomer comprising at least two heterodimeric NC-1-Fc fusion proteins apply mutatis mutandis to the protein oligomer comprising at least two monomeric NC-1-Fc fusion proteins of the present invention.

[0131] In a preferred embodiment, a protein oligomer comprising at least two heterodimeric human NC-1-Fc proteins, or a protein oligomer comprising at least two monomeric human NC-1-Fc fusion proteins is administered intravenously, intracranial / intravarianally, intravitreally, subcutaneously or intraperitoneally, preferably at a concentration of 0.1-1 mg / kg / day.

[0132] In another preferred embodiment, a protein oligomer comprising at least two heterodimeric human NC-1-Fc proteins, or a protein oligomer comprising at least two monomeric human NC-1-Fc fusion proteins, has one or more biological activities selected from the group consisting of: anti-fibrotic activity, anti-angiogenic activity, anti-invasion / anti-metastasis activity, vascular permeability reducing activity, anti-inflammatory and anti-tumorogenic activity, and the ability to induce antibody-dependent cell-mediated cytotoxicity (ADCC).

[0133] In another embodiment, a protein oligomer comprising at least two heterodimeric human NC-1-Fc proteins, or a protein oligomer comprising at least two monomeric human NC-1-Fc fusion proteins, further comprises angiostatin, thrombospondin, anti-PD-1 / PD-L1 antibodies or another therapy for treating, improving or preventing angiogenesis-related diseases, fibrosis or fibrosis-related diseases, VEGF-related diseases or MMP-related diseases as defined herein.

[0134] In addition to protein oligomers comprising at least two monomeric human NC-1-Fc proteins, drugs for the treatment of vascular endothelial growth factor (VEGF)-related diseases that can be used include, for example, other modulators of vascular permeability (e.g., enhanced permeability after brain tissue irradiation, "radionecrosis") and vascular tone (such as the endothelin antagonist macitentan, AT1 / ACE inhibitors), β2-sympathomimetics and corticosteroids used in asthma, immunosuppressants in chronic inflammatory / autoimmune diseases, chemotherapy and radiotherapy for different VEGF-dependent tumors and ascites, kinase inhibitors used in, for example, renal cell carcinoma (mTORi, such as RAD001, multi-kinase inhibitors pazopanib / cinetinib / axitinib, immunomodulators, such as checkpoint inhibitors anti-PD-1 / PD-l1). In addition to protein oligomers comprising at least two monomeric human NC-1-Fc proteins, drugs for the treatment of matrix metalloproteinase-related diseases that can also be used include, for example, locally invasive tumors treated with radio(chemo)therapy with high local regional treatment failure rates, such as glioblastoma, pancreatic cancer, anti-inflammatory and immunosuppressive therapies (anti-TNFα antibodies / infliximab, mycophenolic acid, cyclophosphamide, etc.), tumor invasion or pseudoprogression after cancer treatment, such as anti-angiogenic therapy in recurrent gliomas, and treatment of metastatic diseases (such as breast cancer) with high MMP-2 / MMP-9 activity (i.e., hormone therapy tamoxifen, trastuzumab in HER2+ disease, chemotherapy).

[0135] The present invention further describes a polynucleotide encoding a heterodimeric human NC-1-Fc protein or a monomeric human NC-1-Fc protein.

[0136] As used herein, the term "polynucleotide" or "nucleic acid" refers to single-stranded or double-stranded DNA molecules and RNA molecules. The term includes genomic DNA, cDNA, hnRNA, mRNA and all naturally occurring or artificially modified derivatives of these molecular species. Polynucleotides can be linear or circular molecules in one aspect. In addition, in addition to the nucleic acid sequence encoding heterodimeric human NC-1-Fc protein or monomeric human NC-1-Fc protein, the polynucleotide can also additionally contain sequences required for correct transcription and / or translation, such as 5' or 3'-UTR sequences. In view of the degeneracy of the genetic code, optimized codons can be used to encode nucleic acid sequences of heterodimeric human NC-1-Fc protein or monomeric human NC-1-Fc protein. Thus, optimal expression in, for example, a host cell can be achieved.

[0137] It should be understood that the present invention also covers variants of specific amino acid sequences of heterodimeric human NC-1-Fc protein or monomeric human NC-1-Fc protein or nucleic acid sequences encoding them, as long as these variant sequences also allow the formation of protein oligomers. The protein oligomer formed by the variant preferably has at least one of the following biological activities, preferably at least two, more preferably at least three, and particularly preferably all: anti-fibrosis activity, anti-angiogenesis activity, anti-invasion / anti-metastasis activity, vascular permeability reduction activity, anti-inflammatory and / or anti-tumorigenic activity. In addition, the protein oligomer formed by the variant preferably has the ability to induce antibody-dependent cell-mediated cytotoxicity (ADCC).

[0138] In one aspect, the sequence variant used herein differs from the specific amino acid sequence or specific nucleic acid sequence specified above in that one, two, three, four, five or more amino acids or nucleotide substitutions, additions and / or deletions. On the other hand, the variant sequence has at least 50%, at least 60%, at least 70%, at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, and at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98% or at least 99% identity with the specific nucleic acid sequence or amino acid sequence of the two chains of the heterodimeric human NC-1-Fc protein or the monomeric human NC-1-Fc protein over the entire length or at least half the length of the specific sequence. Preferably, the variant sequence is at least 50%, at least 60%, at least 70%, at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 95%, at least 96%, at least 97%, at least 98% or at least 99% identical to SEQ ID NO: 27, 28, 29, 30, 31, 32 or 37 over the entire length. The term "identical" as used herein refers to sequence identity, which is characterized by determining the number of identical amino acids between sequences, wherein the sequences are aligned to obtain the highest level match. It can be calculated using public techniques or methods such as BLASTP or FASTA (Altschul 1990, J Mol Biol 215, 403) encoded in a computer program. In one aspect, the percent identity value is calculated over the entire amino acid sequence or over a sequence segment of at least 50% of the query sequence. A range of programs based on various algorithms are available to technicians for comparing different sequences. In this case, the algorithms of Needleman and Wunsch or Smith and Waterman give particularly reliable results. For sequence alignment, the program PileUp (Higgins 1989, CABIOS 5, 151) or the programs Gap and BestFit (Needleman 1970, J Mol Biol 48; 443; Smith 1981, Adv Appl Math 2, 482) can be used, which are part of the GCG software package (Genetics Computer Group 1991, 575 Science Drive, Madison, Wisconsin, USA 53711). In another aspect of the invention, the sequence identity values ​​expressed as percentages (%) are determined over the entire sequence region using the program GAP with the following settings: Gap weight: 50, Length weight: 3, Average matches: 10.000 and Average mismatches: 0.000, which, unless otherwise stated, should always be used as standard settings for sequence alignments.

[0139] The present invention further describes a vector comprising a polynucleotide encoding a heterodimeric human NC-1-Fc protein or a monomeric human NC-1-Fc protein.

[0140] Preferably, the vector is an expression vector.

[0141] The term "vector" preferably includes bacteriophage, plasmid, viral or retroviral vectors and artificial chromosomes, such as bacterial or yeast artificial chromosomes. In addition, the term also relates to a targeting construct that allows the targeting construct to be randomly or site-specifically integrated into genomic DNA. In one aspect, such a target construct comprises a DNA of sufficient length for homologous or heterologous recombination as described in detail below. In one aspect, the vector comprising the mentioned polynucleotide also comprises a selection marker for propagation and / or selection in a host cell. The vector can be incorporated into a host cell by various techniques well known in the art. For example, a plasmid vector can be introduced in a precipitate (such as a calcium phosphate precipitate or a rubidium chloride precipitate), or a complex is formed with a charged lipid or introduced in a carbon-based cluster (such as fullerene). Alternatively, a plasmid vector can be introduced by heat shock or electroporation techniques. If the vector is a virus, it can be packaged in vitro using a suitable packaging cell line before being applied to a host cell. Retroviral vectors can be replicative or replication-deficient. In the latter case, viral propagation usually occurs only in a complementary host / cell.

[0142] In addition, in one aspect, the above-mentioned polynucleotides are effectively connected to expression control sequences, thereby allowing expression in prokaryotic or eukaryotic host cells or their separated fractions in the vector. Therefore, in one aspect, the vector is an expression vector. The expression of the polynucleotides includes transcribing the polynucleotides into translatable mRNA. Regulatory elements that ensure expression in host cells are well known in the art. In one aspect, they include regulatory sequences that ensure transcription initiation and / or poly-A signals that ensure transcription termination and transcript stabilization. Other regulatory elements may include transcription enhancers and translation enhancers. Possible regulatory elements that allow expression in prokaryotic host cells include, for example, lac-, trp- or tac-promoters in Escherichia coli, and examples of regulatory elements that allow expression in eukaryotic host cells are AOX1- or GAL1-promoters in yeast, or CMV-, SV40-, RSV-promoters (Rous sarcoma virus) in mammals and other animal cells, CMV-enhancers, SV40-enhancers or globin introns. In addition, inducible expression control sequences can be used for expression vectors. Such inducible vectors may contain tet or lac operator sequences or sequences that can be induced by heat shock or other environmental factors. Suitable expression control sequences are well known in the art. In addition to the elements responsible for transcription initiation, these regulatory elements may also contain transcription termination signals, such as SV40-poly-A sites or tk-poly-A sites downstream of the polynucleotide. In this context, suitable expression vectors are known in the art, such as Okayama-Berg cDNA expression vectors pcDV1 (Pharmacia), pBluescript (Stratagene), pCDM8, pRc / CMV, pcDNA1, pcDNA3 (Invitrogen) or pSPORT1 (Invitrogen). Preferably, the vector is an expression vector and a gene transfer or targeting vector. Expression vectors from viruses such as retroviruses, vaccinia viruses, adeno-associated viruses, herpes viruses or bovine papilloma viruses can be used to deliver polynucleotides or vectors to target cell populations. Recombinant viral vectors can be constructed using methods well known to those skilled in the art; see, for example, the techniques described in Sambrook, Molecular Cloning A Laboratory Manual, Cold Spring Harbor Laboratory (2001) NY and Ausubel, Current Protocols in Molecular Biology, Green Publishing Associates and Wiley Interscience, NY (1994).

[0143] The present invention further describes a host cell comprising a polynucleotide encoding a heterodimeric human NC-1-Fc protein or a monomeric human NC-1-Fc protein.

[0144] The term "host cell" as used herein includes prokaryotic and eukaryotic host cells. In one aspect, the host cell is a bacterial cell. In one aspect, the bacterial host cell is an E. coli host cell. Such bacterial host cells can be used, for example, to replicate the polynucleotides or vectors mentioned.

[0145] In one aspect, the eukaryotic host cell is a cell comprising a polynucleotide encoding a heterodimeric human NC-1-Fc protein or a monomeric human NC-1-Fc protein or a vector wherein the polynucleotide or vector is expressed in the host cell to generate a vector of a protein oligomer comprising a heterodimeric human NC-1-Fc protein or a monomeric human NC-1-Fc protein. The polynucleotide can be introduced transiently or stably into the host cell. In one aspect, the eukaryotic host cell can be a cell of a eukaryotic host cell line that stably expresses the polynucleotide. On the other hand, the host cell is a eukaryotic host cell that has been transiently transfected with a polynucleotide or a vector and expresses the polynucleotide. In another aspect, the cell is a cell that has been genetically modified to produce a protein. How to genetically modify these cells by molecular biology techniques is well known to those skilled in the art.

[0146] sequence

[0147] The sequence shows:

[0148] SEQ ID NO: 1: Mouse collagen 18

[0149] SEQ ID NO: 2: Human collagen 18

[0150] SEQ ID NO: 3: NC-1 domain of mouse collagen 18

[0151] SEQ ID NO: 4: NC-1 domain of human collagen 18

[0152] EQ ID NO: 5: Mouse Fc domain

[0153] SEQ ID NO: 6: Human Fc domain

[0154] SEQ ID NO: 7: rat superstatin

[0155] SEQ ID NO: 8: Mouse fibronectin motif

[0156] EQ ID NO: 9: Murine N-terminal zinc binding domain of endostatin

[0157] SEQ ID NO: 10: Human N-terminal zinc binding domain endostatin

[0158] SEQ ID NO: 11: Mouse RGD motif

[0159] SEQ ID NO: 12: Human RGD motif

[0160] SEQ ID NO: 13: human superstatin

[0161] SEQ ID NO: 14: Human superstatin with His replaced by Ala at positions 1 and 3

[0162] SEQ ID NO: 15: Human superstatin, in which Gln at position 7 is replaced by Cys

[0163] SEQ ID NO: 16: Human superstatin, the "RGD" motif of which is replaced by the "RAD" motif

[0164] EQ ID NO: 17: Mouse integrin binding motif of fibronectin

[0165] SEQ ID NO: 18: Murine endostatin (murine)

[0166] SEQ ID NO: 19: Human endostatin (human)

[0167] SEQ ID NO: 20: mP1 peptide (murine; N-terminal)

[0168] SEQ ID NO: 21: E4 peptide of endostatin (human; C-terminal)

[0169] SEQ ID NO: 22: hP1 peptide of endostatin (human; N-terminal)

[0170] SEQ ID NO: 23: Enterokinase cleavage site

[0171] SEQ ID NO: 24: Fc sequence of wild-type human IgG1

[0172] SEQ ID NO: 25: Fc sequence "knob" human IgG1 Fc: S354C / T366W

[0173] SEQ ID NO: 26: Fc sequence "mortar" human IgG1 Fc Y349C / T366S / L368A / Y407V

[0174] SEQ ID NO: 27: NC-1-enterokinase site linker human IgG1 Fc with "knob" mutation (S354C / T366W)

[0175] SEQ ID NO: 28: Human IgG1 Fc with "hole" mutations Y349C / T366S / L368A / Y407V

[0176] SEQ ID NO: 29: Human IgG1 Fc with "knob" mutation (S354C / T366W)-linker-enterokinase site-NC-1

[0177] SEQ ID NO: 30: Human IgG1 Fc with "hole" mutations Y349C / T366S / L368A / Y407V

[0178] SEQ ID NO: 31: NC-1-Fc fusion protein with "knob" mutations S354C / T366W; Chain 1 of heterodimeric NC-1-Fc (molecule #1)

[0179] SEQ ID NO: 32: Fc with "hole" mutations Y349C / T366S / L368A / Y407V; chain 2 of heterodimeric NC-1-Fc (molecule #1)

[0180] SEQ ID NO: 33: NC-1-Fc fusion protein with "knob" mutations S354C / T366W; chain 1 of homodimeric NC-1-Fc (molecule #2)

[0181] SEQ ID NO: 34: Fc with "hole" mutations Y349C / T366S / L368A / Y407V; chain 2 of homodimeric NC-1-Fc (molecule #2)

[0182] SEQ ID NO: 35: Fc-endostatin fusion protein with "knob" mutations S354C / T366W; Chain 1 of heterodimeric Fc-endostatin (molecule #3)

[0183] SEQ ID NO: 36: Fc-endostatin fusion protein with "hole" mutations Y349C / T366S / L368A / Y407V; Chain 2 of heterodimeric Fc-endostatin (molecule #3)

[0184] SEQ ID NO: 37: Monomeric NC-1-Fc, wherein the Fc comprises the monomer mutation F405R and the YTE half-life extension mutations M252Y / S254T / T256E (Molecule #4)

[0185] SEQ ID NO: 38: Monomeric Fc-endostatin, wherein the Fc comprises the monomer mutation F405R and the YTE half-life extension mutations M252Y / S254T / T256E (Molecule #5) BRIEF DESCRIPTION OF THE DRAWINGS

[0186] These figures show:

[0187] Figure 1 : SDS-PAGE (left) and Western blot (right) analysis of mouse NC-1 and human NC-1 from cell culture supernatants.

[0188] Lane M: marker

[0189] Lanes 1-3: Cell culture supernatants on days 2, 4, and 5 after transfection under reducing conditions

[0190] Lanes 4-6: Cell culture supernatants at days 2, 4, and 5 after transfection under non-reducing conditions

[0191] Lane NC1: negative control under reducing conditions

[0192] Lane NC2: negative control under non-reducing conditions

[0193] Primary antibody: mouse anti-His mAb (Genscript)

[0194] Figure 2 : SDS-PAGE (left) and Western blot (right) analysis of cell lysates of mouse NC-1 and human NC-1.

[0195] Lanes 1-3: Cell lysates at 2, 4, and 5 days after transfection under reducing conditions

[0196] Lane 4: Cell debris on day 4 after transfection under reducing conditions

[0197] Lanes 5-7: Cell lysates at 2, 4, and 5 days after transfection under non-reducing conditions

[0198] Lane 8: Cell debris on day 4 after transfection under non-reducing conditions

[0199] Lane NC1: negative control under reducing conditions

[0200] Lane NC2: negative control under non-reducing conditions

[0201] Primary antibody: mouse anti-His mAb (Genscript)

[0202] Figure 3: Transient mammalian (Expi) expression of heterodimeric NC-1-Fc using two plasmids encoding NC-1-Fc-knob and Fc-hole resulted in the formation of two peaks in SEC after protein A purification. The components of each peak were displayed by SDS-PAGE under reducing and non-reducing conditions. Peak 1 consists of two bands under reducing conditions, containing NC-1-Fc-knob at 62kDa and Fc-hole at 30kDa, respectively. In contrast, peak 2 consists only of a single Fc hole band under reducing conditions, indicating that Fc-hole dimers are generated. Therefore, peak 1 corresponds to the 88 kDa heterodimeric NC-1-Fc (KiH) protein under non-reducing conditions, while the size of the Fc-hole dimer peak 2 corresponds to approximately 50kDa protein under non-reducing conditions.

[0203] Figure 4 : After protein A purification of the protein supernatant, SEC-HPLC was performed to detect the size of the protein expressed under native conditions. In order to better estimate the size of the expressed protein (two peaks), well-known spikes in controls consisting of IgG1 / 2B, albumin, human Fc and IgG1-Fab were used. As expected, peak 2, consisting of Fc-hole dimer, ran simultaneously with the Fc domain. However, peak 1 seemed to be larger than the standard IgG (150 kDa). This was unexpected because Figure 3 The reduced gel clearly showed that peak 1 contained both the NC-1-Fc-knob and Fc-hole portions, resulting in a heterodimeric NC-1-Fc protein of 88 kDa.

[0204] Figure 5 : Transduction of mammalian cells with NC-1-Fc-knob vs. Fc-hole plasmids at a ratio of 2:1 enhanced the formation of heterodimeric NC-1-Fc-KiH and almost reduced the formation of Fc-hole dimers.

[0205] Figure 6 : Cross-linking studies conclusively confirmed the SEC observations that the heterodimeric NC-1-Fc must be much larger than the 88 kDa monomer molecule found in the non-reducing gel (lane 2). In fact, the heterodimeric NC-1-Fc (KiH) forms trimers (lane 3) under physiological conditions. Details of the cross-linking procedure can be found in Fig.17 Description.

[0206] Figure 7:NC-1-Fc-KiH has excellent anticancer effect in Lewis lung cancer model. Tumor growth kinetics of isogenic Lewis lung cancer (LLC) model in C57bl6 mice. Compared with all recombinant monomeric endostatin molecules and Fc-endostatin (Fc-ES) forming dimers on the Fc part, the trimeric NC-1-Fc (KiH) shows higher activity. Three monomeric endostatin compounds were used, including heterodimeric Fc (KiH) endostatin (Fc-(KiH)-ES), which forms monomeric endostatin with dimer Fc, i.e., the reference backbone construct of heterodimeric NC-1-Fc (KiH), Endo (monomeric recombinant Escherichia coli His-tagged endostatin approved in China), and Entemed endostatin (ES-Entremed) (monomeric recombinant Pichia pastoris endostatin in Phase I / II trials in the United States / Europe). All drugs were injected subcutaneously with endostatin at an equivalent dose of 20 µg / mouse / day. Bars represent mean values ​​of n:7 + / - SEM.

[0207] Figure 8 : Tumor growth inhibition. At day 9 after the start of treatment, the trimeric NC-1-Fc (KiH) performed best, with a 51% inhibitory effect on the growth of LLC tumors (B). This effect was significantly more pronounced compared to all monomeric endostatin molecules (p < 0.001). Interestingly, NC-1-Fc (KiH) was more effective than the dimeric Fc-ES (A, p < 0.01). Therefore, in addition to oligomerization, other biological properties of the entire NC-1 sequence vs. only the N-terminal endostatin domain may be related to the anticancer activity of this molecule.

[0208] Fig. 9 :Heterodimer NC-1-Fc (molecule #1)

[0209] Fig.10 :Monomer NC-1-Fc (molecule #4)

[0210] Fig.11 :Homodimer NC-1-Fc (molecule #2)

[0211] Fig.12 :Heterodimer Fc-endostatin (molecule #3)

[0212] Fig.13 :Monomeric Fc-endostatin (Molecule #5)

[0213] Fig.14 : Similar to Fc-endostatin (molecule #6), using the N- and C-terminal NC-1-Fc fusion protein of the IgG1 wild-type backbone

[0214] Fig.15 : (A) Heterodimeric protein consisting of NC-1-Fc(knob) and Fc(hole) fused to different parts including IgG Fab fragment (molecule #7). (B)

[0215] Fig.16 : In vivo treatment with a protein oligomer comprising monomeric NC-1-Fc (NC-1-Fc(monomer)) (SEQ ID NO: 37) in a lung fibrosis model (see Example 4).

[0216] Fibrosis data n per group: 10

[0217] IR dose: 15 Gy

[0218] Monomer: Pattern structure

[0219] * P<0.05, **p<0.01

[0220] Dashed line 16 weeks after IR (red) vs. non-irradiated (black).

[0221] Fig.17 : NC-1-Fc-monomer forms trimers, which shows the non-reducing gel results, where

[0222] Lane 1: Spectral Multicolor High Range Protein Ladder (Thermo Scientific #26625) 10 μl

[0223] Lane 2: monomer hFC-NC1 25 μg

[0224] Lane 3: 25 μg of monomeric hFC-NC1 cross-linked

[0225] Even under non-reducing conditions, protein association and 3D properties are partially lost by gel electrophoresis. Therefore, cross-linking of native proteins is required to decipher the composition of NC-1-Fc-mono. To this end, the cross-linking protocol described by (Kuo, Javaherian et al. JCB 2001, PMID: 11257123) was used with some modifications. Briefly, 50 μg of protein was used in 50 μl of PBS containing 50 μM zinc chloride. To this was added 5 μl of the cross-linker EGS (ethylene glycol-bis(succinic acid N-hydroxysuccinimide ester) dissolved in DMSO at a concentration of 10 mM. After the samples were incubated at 37°C for 1 / 2 hour, the reaction was terminated with 4 μl of 1 M Tris, pH 8. Finally, the samples were subjected to SDS-PAGE under non-reducing conditions. The inventors' data clearly showed that NC-1-Fc (monomer) formed trimers on the NC-1 oligomerization domain as found after cross-linking in lane 3. These data are consistent with the observation of the size of NC-1-Fc (monomer) under native conditions in SEC.

[0226] Fig.18 The structures of the heterodimeric NC-1-Fc and the trimer on the oligomerization domain of NC-1 are shown;

[0227] Fig.19 The structure of the homodimeric NC-1-Fc is shown;

[0228] Fig. 20 The structure of heterodimeric Fc-endostatin is shown;

[0229] Fig.21 The structure of the prototype Fc-endostatin, which is an N-terminal fusion protein linked to wild-type IgG1 Fc via an EK digestion site, is shown;

[0230] Fig. 22 The structures of monomeric NC-1-Fc and the trimer formed by the NC-1 oligomerization domain are shown;

[0231] Fig.23 The structure of monomeric Fc-endostatin is shown;

[0232] Fig.24 N- and C-terminal NC-1-Fc fusion proteins using an IgG1 wild-type backbone are shown, similar to Fc-endostatin;

[0233] Fig.25 Shown are the structures of heterodimeric proteins composed of NC-1-Fc (knob) and Fc (hole) fused to different parts including an IgG Fab fragment, as well as the low yield, aggregation, and formation of large amounts of product on SEC after Protein A purification.

[0234] sheet

[0235] Table 1: Heterodimer NC-1-Fc (Molecule #1)

[0236] Table 2: Homologous NC-1-Fc (Molecule #2)

[0237] Table 3: Heterodimer Fc-endostatin (Molecule #3)

[0238] Table 4: Monomeric NC-1-Fc (molecule #4)

[0239] Table 5: Monomeric Fc-endostatin (Molecule #5)

[0240] Table 6: Other constructs designed but failed

[0241] The present invention will now be illustrated by examples, however, these examples should not be construed as limiting the scope of the present invention. Example

[0242] Example 1: Recombinant expression of NC-1 is not feasible in preclinical and clinical studies

[0243] The production of collagen 18 NC-1 was a huge challenge from the outset due to the aggregation of the molecule at the oligomerization domain. Therefore, there are no in vivo data on the efficacy of NC-1 and only a few studies have been conducted on the molecule that was hitherto considered to be a precursor of endostatin.

[0244] Dr. Javaherian and a few other laboratories were able to express very little recombinant NC-1 through the His tag and Flag tag, respectively, but this was far from enough for preclinical dosing in animals. Therefore, the production of NC-1 quantities relevant to preclinical and clinical studies poses a challenge. After confirming this hypothesis, the inventors provide data on ultra-low yields of His-tagged NC-1 expression in humans and mice below.

[0245] Designed and synthesized the DNA sequence encoding mouse NC-1 and human NC-1.Then the complete sequence, i.e., EcoRI-Kozak sequence-leader sequence-His6 tag-mouse NC-1-terminator codon-HindIII (SEQ ID NO. 40), and EcoRI-Kozak sequence-leader sequence-His6 tag-human NC-1-terminator codon-HindIII (SEQ ID NO. 42) were subcloned into the pcDNA3.4 vector for Expi293F cell expression. Expi293F cells were grown in serum-free Expi293FTM expression medium (Thermo Fisher Scientific). The cells were maintained in conical flasks (Corning Inc., Acton, MA) at 37 ° C, 8% CO2 on an orbital shaker (VWR Scientific, Chester, PA). One day before transfection, cells were seeded in Corning conical flasks at an appropriate density. On the day of transfection, DNA and transfection reagent were mixed at an optimal ratio and then added to the conical flask together with the cells prepared for transfection. Recombinant plasmids encoding mouse NC-1 and human NC-1 target proteins were transiently transfected into 40 ml suspension Expi293F cell cultures, respectively. Cell culture supernatants collected on days 2, 4, and 5 were used for protein expression evaluation. About 1 ml of cell culture supernatant and cell pellet samples were collected on days 2, 4, and 5 after transfection and analyzed by SDS-PAGE and Western blot analysis to evaluate the expression levels of mouse NC-1 and human NC-1, as shown in Figure 2. Figure 1 and 2 The primary antibody used for Western blotting was mouse-anti-his mAb (GenScript, Cat. No. A00186).

[0246] The present inventors have attempted to express and detect mouse NC-1 and human NC-1 in suspension Expi293F cell cultures.

[0247] For mouse NC-1, a weak band of interest was positively detected from the cell culture supernatant by Western blot analysis, e.g. Figure 1 As shown in B. By reducing condition SDS-PAGE and Western blot analysis, a target signal with an estimated molecular weight of about 40 kDa (Cal.MW.38.17) was observed from the cell debris. Figure 2 As shown in A and B (black arrows). This indicates insufficient secretion of the target protein. The estimated expression level of the target protein is ~0.1 mg / L. This indicates insufficient secretion of the target protein.

[0248] For human NC-1, a distinct target band with an estimated molecular weight of about 40 kDa (C Cal. MW. 37.47) was detected from the cell culture supernatant and cell debris by SDS-PAGE and Western blot analysis under reducing conditions, as shown in Figure 1 C and D and Figure 2 C and D. The estimated expression level of the target protein is low <3 mg / L. Compared with mouse NC-1, human NC-1 has better expression, but still at low levels for most downstream evaluations (e.g., in vivo studies).

[0249] This data suggests that recombinant expression of NC-1 is challenging and not feasible for preclinical and clinical studies.

[0250] Example 2: Modification of collagen 18 NC-1-Fc construct

[0251] The main purpose of the present invention is to advance in the field of immunoglobulin IgG Fc conjugation, achieving numerous biotechnological, pharmacokinetic and biological advantages.

[0252] However, the generation of Fc-NC-1 is not a straightforward process as is evident from the fact that the inventors have spent nearly a decade thus far to realize a system for large-scale production of NC-1 and that many constructs and classical tags have failed.

[0253] The reason for this is that there are two oligomerization forces, the first being the trimerization force induced by the NC-1 oligomerization domain and the second being the dimerization force induced by the IgG-Fc. The inventors have spent a great deal of time with all the different prior art techniques, i.e., conjugation of the C- or N-terminus of NC-1 to Fc, as previously done with endostatin, and all of these approaches have failed, as shown for example in Tables 2 and 6.

[0254] One has to consider that in the classical Fc tagging approach, an N-terminal or C-terminal conjugated Fc-NC-1 molecule is generated, which without the NC-1 oligomerization domain forms dimers. However, with the NC-1 oligomerization domain one receives aggregates and very pure expression.

[0255] To the best of the inventors' knowledge, there is no literature on successful NC-1-Fc conjugation.

[0256] As described below, the inventors have studied different methods to circumvent this major problem and have discovered a successful strategy after more than ten years of research. A core point is to generate monovalent NC-1-dimer Fc or heterodimeric NC-1-Fc. The inventors decided to use mutations in Fc that prevent dimerization of NC-1-Fc with NC-1-Fc using the Fc "knob-hole structure" (KiH) strategy. Therefore, the NC-1-Fc knob will only dimerize with the empty Fc-hole. Although knob-knob dimerization is prohibited, hole-hole dimerization may still occur. In fact, the inventors found two peaks, one peak is the expected NC-1-Fc-KiH heterodimer, and the second peak is the Fc-hole-Fc-hole dimer. This shows that NC-1-Fc-knob is more difficult for cells to express. The inventors circumvented this problem again by increasing the expression of NC-1-Fc-knob to Fc-hole (at an initial ratio of 2:1, then 4:1), which mainly resulted in a peak of heterodimeric NC-1-Fc-KiH with excellent expression efficacy. However, in the size exclusion column, the inventors realized that the molecule under physiological conditions was much larger than the molecule expected according to the NC-1-Fc-KiH heterodimer. Subsequent cross-linking experiments confirmed that they had surprisingly achieved the production of trimeric molecules, i.e., trimers of NC-1-Fc-KiH heterodimers. This molecule retains all the excellent properties of Fc plus trimeric NC-1. Surprisingly, this trimeric NC-1-Fc-KiH molecule exhibited the same efficacy in binding to unique oligomeric NC-1 binding partners (such as fibronectin, VEGF and MMP-2 / -9). With excellent expression rates, the inventors have now been able to use this NC-1 molecule for in vivo experiments in different models after a long time. In addition to data in lung cancer (LLC) models and pulmonary fibrosis models, further experiments are underway that will greatly improve their understanding of this molecule.

[0257] Notably, the separate expression of NC-1-Fc-knob and NC-1-Fc-hole and subsequent dimerization via a redox system failed to produce an efficient NC-1-Fc construct. Therefore, steric hindrance of the heterodimeric Fc is crucial for the success and lack of aggregation. Next, the inventors aimed to evaluate whether adding a part to the "empty" Fc-hole would affect the molecule. To this end, they conjugated the classical Fab fragment of an antibody to the Fc-hole and co-expressed it with the NC-1-Fc-knob. Interestingly, the addition of steric hindrance via the Fab part NC-1-Fc-KiH-Fab reduced the expression efficiency compared to the NC-1-Fc-KiH heterodimer.

[0258] In summary, the trimeric NC-1-Fc-KiH heterodimer (SEQ ID NO: 31 / 32; Table 1; Fig. 9; Molecule #1) appears to be the first NC-1 based Fc construct reported by the inventors, and given the 3D complexity of NC-1 and Fc, engineering this molecule was not, and is not, a trivial process as can be deduced from the current literature.

[0259] Example 2A: Heterodimer NC-1-Fc (SEQ ID NO: 31 / 32)

[0260] Transient mammalian (Expi) expression of heterodimeric NC-1-Fc-KiH using two plasmids encoding NC-1-Fc-knob and Fc-hole resulted in the formation of two peaks in SEC after protein A purification. The components of each peak were shown by SDS-PAGE under reducing and non-reducing conditions. Figure 3 Peak 1 consists of two bands under reducing conditions, containing NC-1-Fc-knob at 62 kDa and Fc-hole at 30 kDa. In contrast, peak 2 consists of only a single Fc hole band under reducing conditions, indicating the generation of Fc hole dimers. Thus, peak 1 corresponds to the 88 kDa heterodimeric NC-1-Fc-KiH protein under non-reducing conditions, while the size of Fc-hole dimer peak 2 corresponds to a protein of approximately 50 kDa under non-reducing conditions, see Figure 3 .

[0261] The inventors ran SEC-HPLC on these two isolated peaks and compared them with other proteins of known size (see Figures 4 to 6 ). As expected, peak 2 runs the same size as the Fc domain. Peak 1 appears to be larger than a standard IgG. This is unexpected because Figure 3 The reduced gel clearly showed that peak 1 contained both the knob and hole parts (ie, NC-1-Fc-knob at 62 kDa and Fc-hole at 30 kDa).

[0262] Cross-linking studies ultimately confirmed that the inventors were able to generate, for the first time, a trimer of NC-1 on top of the Fc backbone, ie, a trimer of NC-1-Fc-KiH (or NC-1-Fc(KiH)) under physiological conditions.

[0263] By increasing the amount of NC-1-Fc-knob vs. Fc-hole (starting at a 2:1 ratio, then 4:1), they were able to further optimize the production of this NC-1-Fc-KiH trimer construct by nearly eliminating hole-hole formation (peak 2); see Figure 5 .

[0264] Binding studies to date have demonstrated that the NC-1-Fc-KiH trimer retains the central properties of NC-1, such as binding to fibronectin as well as VEGF and MMP-2 / -9.

[0265] Cross-linking experiments confirmed the formation of trimeric NC-1-Fc-KiH heterodimers under physiological conditions, see Figure 6 .

[0266] Example 2B: Monomer NC-1-Fc (SEQ ID NO: 37)

[0267] The present inventors also generated monomeric NC-1-Fc (SEQ ID NO: 37; Fig.10 ; molecule #4). IgG1 Fc includes the monomer mutation F405R and the YTE half-life extension mutations M252Y / S254T / T256E. When expressed, high yields of monomeric NC-1-Fc (also referred to herein as NC-1-Fc (monomer)) can be obtained. Fig.17 As shown in lane 3 of the present invention, after cross-linking, NC-1-Fc (monomer) forms a trimer on the NC-1 oligomerization domain; see also Example 5. As demonstrated in Example 4, a protein oligomer comprising monomeric NC-1-Fc (NC-1-Fc (monomer)) (SEQ ID NO: 37) exhibits excellent anti-fibrotic activity in a pulmonary fibrosis model.

[0268] Example 3: Anticancer Effects of Collagen 18 NC-1-Fc Construct

[0269] As a tumor model, a subcutaneous syngeneic model of Lewis lung carcinoma (LLC) in a C57bl6 mouse background has been used. Mice have been treated with equivalent doses of:

[0270] - heterodimer NC-1-Fc(KiH) (SEQ ID NO: 31 / 32) that forms a trimer on the NC-1 oligomerization domain (NC-1-Fc(KiH)),

[0271] - heterodimeric Fc(KiH)endostatin (Fc-(KiH)-ES) (SEQ ID NO: 35 / 36), which together with the dimeric Fc forms a reference construct of monomeric endostatin, ie heterodimeric NC-1-Fc(KiH),

[0272] - Fc-endostatin (Fc-ES) which forms a dimer on the Fc part,

[0273] - Endostatin, a monomeric recombinant E. coli His-tagged endostatin approved in China,

[0274] -Entemed endostatin (ES-Entremed), a monomeric recombinant Pichia pastoris endostatin in Phase I / II trials in the US / Europe.

[0275] The tumor growth delay data clearly demonstrate that the trimeric NC-1-Fc(KiH), a drug designed based on NC-1, outperforms all other constructs in tumor growth inhibition in a direct back-to-back comparison in the prototype LLC model; see Figure 7 .

[0276] Notably, the superior activity of trimeric NC-1-Fc(KiH) with 51% inhibition of tumor growth at day 9 after the start of treatment compared to 30% inhibition of dimeric Fc-ES (p < 0.01) clearly indicates that not only oligomerization (trimer vs. dimer) but also other biological properties of the entire NC-1 sequence vs. only the N-terminal endostatin domain contribute to the therapeutic effect of trimeric NC-1-Fc(KiH); see Figure 8 .

[0277] Example 4: Anti-fibrotic effect of collagen 18 NC-1-Fc construct

[0278] In vivo treatment, protein oligomers containing monomeric NC-1-Fc (NC-1-Fc (monomer)) (SEQ ID NO: 37) have been generated in a lung fibrosis model, and the results show that Fig.16 middle.

[0279] The inventors have previously shown that the N-terminus and oligomerization of endostatin, demonstrated by NC-1 mimics (oligomeric Fc-endostatin, FcES), are critical for preventing the development of pulmonary fibrosis induced by ionizing radiation (IR; WO2017 / 093569). In a back-to-back comparison with FcE and three monomeric endostatin constructs (i.e., Entremed endostatin, Endostatin, and Fc(monomer)-ES), the inventors aimed to discover the potential of novel protein oligomers comprising recombinant monomeric NC-1-Fc (NC-1-Fc(monomer)) (SEQ ID NO: 37) to inhibit and ultimately reverse fibrosis. The C57b16 radiation-induced pulmonary fibrosis (RILF) model was used. Treatment was initiated 16 weeks after 15Gy whole chest irradiation, with mean lung density increasing from -380 hounsfield units (HU, black dashed line) in control mice to -320 HU (red dashed line) in irradiated lungs ( Fig.16 A). All compounds were administered subcutaneously at equimolar doses daily for 8 weeks. Oligomeric constructs FcES and NC-1-Fc (monomer) both reduced lung density ( Fig.16 B), increase lung volume ( Fig.16 C) and improved CT-based quantitative lung fibrosis indices ( Fig.16D) shows anti-fibrotic effects) (FI, Zhou et al. 2017, 16; https: / / www.ncbi.nlm.nih.gov / pubmed / 29116014).

[0280] Interestingly, in this intervention trial, the novel protein oligomer comprising recombinant monomer NC-1-Fc (NC-1-Fc (monomer)) (SEQ ID NO: 37) was the only compound capable of reversing fibrosis. The anti-fibrotic effect of the protein oligomer of the present invention was significantly stronger than that of FcES (p < 0.01). All three monomer constructs are shown as patterned bars. The bars represent the average of n: 10 + / - SEM.

[0281] Example 5: NC-1-Fc (monomer) (SEQ ID NO. 37) forms a trimer

[0282] Even under non-reducing conditions, protein association and 3D properties are partially lost by gel electrophoresis. Therefore, cross-linking of native proteins is required to decipher the composition of NC-1-Fc-monomers. For this purpose, the cross-linking protocol described by (Kuo, Javaherian et al. JCB 2001, PMID: 11257123) was utilized with some modifications. Briefly, 50 μg of protein were used in 50 μl of PBS containing 50 μM zinc chloride. To this was added 5 μl of the cross-linker EGS (ethylene glycol-bis(succinic acid N-hydroxysuccinimide ester) dissolved in DMSO at a concentration of 10 mM. After incubation of the samples at 37°C for 1 / 2 hour, the reaction was terminated with 4 µl of 1 MTris, pH 8. Finally, the samples were subjected to SDS-PAGE under non-reducing conditions. The inventors' data clearly show that, as in Fig.17 As found in lane 3 of the cross-linked, NC-1-Fc (monomer) formed a trimer on the NC-1 oligomerization domain. These data are consistent with the observation of the size of NC-1-Fc (monomer) under native conditions in SEC.

[0283] Table 1: Heterodimer NC-1-Fc

[0284]

[0285] Table 2: Homodimer NC-1-Fc

[0286]

[0287] Table 3: Heterodimer Fc-endostatin

[0288]

[0289] Table 4: Monomer NC-1-Fc

[0290]

[0291] Table 5: Monomeric Fc-endostatin

[0292]

[0293] Table 6: Additional constructs that failed

[0294]

Claims

1. A monomeric NC-1-Fc fusion protein comprising human NC-1 from collagen 18 and a monomeric Fc domain from human IgG1 or IgG4, wherein the monomeric Fc domain from human IgG1 or IgG4 comprises at least one monomer mutation, preferably the monomer mutation F405R, and one or more half-life extension mutations, preferably the half-life extension mutations M252Y, S254T and T256E.

2. The monomeric NC-1-Fc fusion protein of claim 1, comprising an amino acid sequence that is preferably at least 70%, at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94% or at least 95% identical to the amino acid sequence shown in SEQ ID NO: 37 over its entire length.

3. A protein oligomer comprising at least two monomeric human NC-1-Fc fusion proteins as claimed in claim 1 or 2.

4. The protein oligomer of claim 3, wherein the monomeric Fc has no FcγR-dependent effector function, preferably antibody-dependent cell-mediated cytotoxicity (ADCC) or antibody-dependent cellular phagocytosis (ADCP) or complement-dependent cytotoxicity (CDC). The protein oligomer of claim 3 , wherein the protein oligomer binds to fibronectin, MMP-2 and / or MMP-9. The protein oligomer according to claim 3 , for use as a medicament.

7. The protein oligomer according to claim 6, for use as a drug for treating, ameliorating or preventing a disease selected from the group consisting of: (i) fibrosis or diseases associated with fibrosis, including skin fibrosis, preferably scleroderma; keloids or keloids; hypertrophic scars; morphea; fibrosis due to graft-versus-host disease; subepithelial fibrosis; endomyocardial fibrosis; uterine fibrosis; myelofibrosis; retroperitoneal fibrosis; nephrogenic systemic fibrosis; postoperative scars; asthma; cirrhosis / liver fibrosis; fibrosis due to abnormal wound healing; glomerulonephritis; multifocal fibrosclerosis; radiation-induced fibrosis, preferably radiation-induced pneumonitis or radiation-induced pulmonary fibrosis; chemotherapy-induced or drug-induced fibrosis, for example due to mTOR or EGFR kinase inhibition; common or idiopathic pulmonary fibrosis; fibrosis caused by autoimmune diseases, such as lupus, fibrosis / fibrogenesis associated with tumors and cancer, chronic inflammation associated with organ fibrosis, for example by viral stimulation or transplantation; organ fibrosis as a terminal stage of chronic kidney disease, long-term dialysis or diabetes; and (ii) Matrix metalloproteinase (MMP)-associated diseases, including both benign and malignant diseases in which MMP activation contributes to pathophysiology, for example, MMP activation during local tumor invasion and cancer metastasis inherent in tumors with high rates of local treatment failure such as glioblastoma, pancreatic cancer, and lung cancer, as well as acquired enhanced MMP activation due to overt immune responses induced by treatment-induced selective pressures (e.g., tumor hypoxia and fibrosis after radiotherapy), autoimmune diseases, and chronic inflammatory diseases.

8. The protein oligomer for use according to claim 7, further comprising angiostatin, thrombospondin, anti-PD-1 / PD-L1 antibody or another therapy for treating, improving or preventing the fibrosis or fibrosis-related diseases or the MMP-related diseases.

9. A method for producing the protein oligomer of claim 3, comprising: a) culturing a host cell expressing a monomeric human NC-1-Fc fusion protein, wherein the monomeric human NC-1-Fc fusion protein comprises, from N- to C-terminus, human NC-1 from collagen 18 fused to a monomeric Fc domain from human IgG1 or human IgG4, or a monomeric Fc domain from human IgG1 or human IgG4 fused to human NC-1 from collagen 18, under conditions allowing the formation of protein oligomers comprising at least two, preferably three, monomeric human NC-1-Fc fusion proteins, wherein the monomeric Fc domain from human IgG1 or human IgG4 comprises at least one monomer mutation, preferably the monomer mutation F405R, and one or more half-life extension mutations, preferably the half-life extension mutations M252Y, S254T and T256E, and b) obtaining a protein oligomer comprising at least two and preferably three monomeric human NC-1-Fc fusion proteins from the host cell of step a).

10. The method according to claim 9, wherein the monomeric human NC-1-Fc fusion protein comprises an amino acid sequence that is preferably at least 70%, at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94% or at least 95% identical to the amino acid sequence shown in SEQ ID NO: 37 over the entire length.

Images