Heterodimeric fc-fused cytokine and pharmaceutical composition comprising the same

The present invention is a heterodimeric Fc fusion protein comprising a first Fc region and a second Fc region of an immunoglobulin Fc pair and a physiologically active protein, wherein one or more subsystems of the physiologically active protein are the first. Linked to one or more ends of the N or C ends of the Fc region and / or the second Fc region, the CH3 domains of the first Fc region and the second Fc region promote the formation of Fc heterodimers. Heterodimer Fc fusion proteins that are mutated to do so, as well as pharmaceutical compositions comprising heterodimeric Fc fusion proteins.

The heterodimeric Fc fusion protein according to the invention allows it to retain the activity of a naturally occurring physiologically active protein composed of two or more different subunit proteins, thereby making the fusion protein. Assembled because each subunit of the protein can be fused separately to each strand of the Fc of the heterodimer of the immunoglobulin so that the naturally occurring morphology and structure can be maintained to the highest possible extent. It has the advantage that it can exhibit intact biological activity by forming a protein.

When the heterodimeric Fc fusion protein according to the present invention is used, the in vivo half-life of the bioactive protein contained in the heterodimeric Fc fusion protein is set to Fc so that its bioactivity in vivo can be sustained for a long period of time. There is an advantage in that it can be significantly increased due to the long half-life mediated by.

Further, the heterodimeric Fc fusion protein according to the present invention has a structure in which one or more subunits of a physiologically active protein are fused to the N-terminal or C-terminal of the immunoglobulin heterodimeric Fc, and the heterodimeric Fc fusion protein. Is readily purified after its expression as compared to wild Fc-based fusion proteins.

Naturally occurring human antibodies (immunoglobulin G (IgG), IgM, IgD, IgE and IgA) each exist as an assembly of two heavy chains with the same amino acid sequence and two light chains with the same sequence. In this regard, homodimerization between two identical heavy chains is not between the constant region terminal domains (CH3 domains of IgG, IgD and IgA, CH4 domains of IgM and CH2 and CH4 domains of IgE). It is induced by covalent interactions and disulfide bonds between hinge domains.

Antibody heterodimer Fc technology prefers engineered Fc fragments with CH3 variant pairs to form Fc heterodimers over naturally occurring antibodies (IgG, IgM, IgA, IgD and IgE) over Fc homodimers. Thus, it is a technique for producing a heterodimer crystalline fragment (Fc) of an immunoglobulin heavy chain constant region by modification to the CH3 domain interface by different mutations on each domain. More specifically, it is genetically engineered so that the two Fc fragments form heterodimers with minimal sequence changes, but they have tertiary structure very similar to that of naturally occurring antibodies. A technology that induces mutations in two different CH3 domains of Fc (US Pat. No. 7,695,936; and Korean Pat. No. 1,522,954). Heterodimer Fc technology is a platform technology for the production of bispecific antibodies, and the previously known CH3 domain mutants that induce Fc heterodimer formation are rationally designed based on the structure of the antibody. Mostly generated by introducing asymmetric mutant pairs into the CH3 domain interface (Spreter Von Kreudenstein et al., 2014). Pioneering studies include knob-into-hole technology from Genentech (Ridgway et al., 1996), Zymeworks (ZW1; Von Kreudenstein et al., 2013), Xencor (HA-TF; Moore GL et al., 2011). Year), and many multinational pharmaceutical companies, including EMD Serono (SEEDbody; Davis JH et al., 2010), have developed and reported on platform technology.

In particular, the A107 variant used in the present invention is a high yield Fc heterodimer screened from a human antibody heterodimer Fc library constructed using a yeast cell surface presentation system and is sterically complementary. Hydrophobic interaction (K409W)_CH3A-D399V / F405T_CH3B) Is induced in charged amino acids to form a hydrogen bond (K370E) while maintaining hydrophobic core integrity at the CH3 domain interface._CH3A-E357N_CH3B) Is a heterodimer Fc variant that promotes the formation of heterodimers (Choi et al. 2016; Korean Patent Application No. 2015-0142181).

All previously reported heterodimer Fc variants, including the A107 variant, are based on IgG1, which accounts for the largest proportion of human antibody isotypes, and isotypes other than IgG1 (IgG2, IgG3, IgG4, IgA, IgM and IgE). No variants of this have been reported yet.

This is because the majority of therapeutic antibodies marketed under the approval of the US Food and Drug Administration (FDA) use the IgG1 isotype (Irani et al., 2015). In recent years, the effector function has been enhanced for immunomodulatory antibodies or receptor agonist fusion proteins that do not need to have large antibody effector functions such as antibody-dependent cellular cytotoxicity (ADCC) or complement-dependent cytotoxicity (CDC). Developments are underway for therapeutic proteins based on IgG2 or IgG4 that are significantly lower than those of IgG1.

On the other hand, most bioactive proteins have the drawback of having a small size and therefore a short in vivo half-life. In order to solve this drawback, there have been attempts to conjugate PEG (polyethylene glycol) or the like or fuse it into the antibody Fc (crystalline fragment) region. However, it is not yet possible to develop a bioactive protein whose activity is maintained efficiently and sufficiently for a long period of time.

In particular, for proteins composed of two or more different subunits, where two or more different subunits form a protein complex to exhibit physiological activity, wild-type Fc fusion proteins Due to the homodimer nature of Fc, it has not been possible to develop an Fc fusion protein that is formed to have the original protein complex structure that naturally exists with wild Fc. Therefore, wild-type Fc is not suitable for Fc fusion for heterodimer or heterooligomeric proteins that adequately exhibit the activity of the original proteins and maintain their activity well for long periods of time.

Against this technical background, we have constructed heterodimer variants containing Fc regions derived from not only IgG1 but also previously unreported other isotype antibodies such as IgG2, IgG3 and IgG4. , These heterodimeric variants are composed of two or more different subunits, one of the proteins that exhibits physiological activity by forming a protein complex with two or more subunits. Alternatively, multiple subunits were used to develop a novel therapeutic fusion protein in the form of a heterodimer Fc fusion protein in which multiple subunits are genetically fused to the end of the Fc region, thereby completing the invention.

In particular, in the present invention, interleukin-12 (IL-12) can preferably be used as a protein composed of two different subunits, p35 and p40, where the two subunits are IL. -12 Shows bioactivity by forming proteins.

IL-12 can directly kill tumors by increasing the activity of immune cells such as cytotoxic T lymphocytes (CTL) or natural killer cells (NK) among immune cells, or In the tumor microenvironment where the immune response is inhibited, tumor formation can be inhibited by activating the immune response through the secretion of proinflammatory cytokines such as interferon gamma (IFN-γ). Therefore, IL-12 has been extensively studied as an anticancer cytokine (Lasek et al., 2014). However, in the development of therapeutic methods using IL-12, the short half-life of cytokines itself requires frequent administration, which can lead to toxicity. For this reason, studies have been conducted to fuse IL-12 with antibodies or Fc for its use as long-acting IL-12 (Tugues et al., 2015). However, in these studies, unlike the endogenous monovalent form of IL-12, the fused IL12 protein was due to the fusion of wild-type Fc-based antibodies that form homodimers by interaction between CH3 domains. The problem of avidity arises, for which reason IL-12 fused to wild-type Fc-based antibodies exhibits inferior physiological activity to endogenous IL-12, or IL-12 to immune cells. Undesirable localization emerges due to the increased binding promoted by avidity (Tzeng et al., 2015; Dumont et al., 2006).

Therefore, as shown in FIGS. 1 (A) to 1 (C), a single selective tag for further purification is used in efforts to generate monovalent fusion proteins using wild-type antibodies or Fc regions. Methods have been used to construct fusion proteins through strategies such as fusing only to the C-terminus of the Fc region of the Fc region, or fusing the Fc region and protein to each other after purification separately with high purity. However, this method is not only very expensive to produce large amounts of protein, but also requires research to optimize the further purification process.

However, the use of the heterodimer Fc fusion protein according to the present invention makes it possible to easily produce a monovalent heterodimer Fc fusion protein as shown in FIG. 2 without the need for further optimization of the purification process. To do.

<p><patcit num="1"><text>U.S. Pat. No. 7,695,936</text></patcit><patcit num="2"><text>Korean Patent No. 1,522,954</text></patcit></p>

<p> An object of the present invention is to provide a novel heterodimeric Fc fusion protein, which is composed of one, two or more different subunits, thereby forming an assembled protein. This allows it to exhibit intact biological activity and thus maintain the natural bioactivity of its fusion protein in vivo for extended periods of time.</p><p> In particular, the heterodimeric Fc fusion protein according to the invention has two or more subunits so that the fusion protein can maintain its naturally occurring form and structure to the highest possible extent. It is formed so that it can retain the activity of naturally occurring bioactive proteins that aggregate to form proteins and exhibit bioactivity.</p><p> Furthermore, the heterodimeric Fc fusion protein according to the present invention has an in vivo half-life of the bioactive protein contained in the heterodimeric Fc fusion protein by Fc so that its bioactivity in vivo can be sustained for a long period of time. It has the advantage that it can be significantly increased due to the long half-life mediated.</p><p> Another object of the present invention is to provide a pharmaceutical composition comprising the heterodimer Fc fusion protein described above, and a composition and therapeutic method for treating a disease, particularly cancer, using the same.</p>

<p> In order to achieve the above object, the present invention is a heterodimeric Fc fusion protein containing a first Fc region and a second Fc region of an immunoglobulin Fc pair and a physiologically active protein, and there are two physiologically active proteins. Or composed of many different subunits, two or more different subunits exhibit physiological activity by forming a protein complex, the physiologically active protein subunits are the first Fc region and / Or linked or genetically fused to one or more ends of the N or C ends of the second Fc region, the CH3 domains of the first Fc region and the second Fc region are Fc heterozygous. Provided is a heterodimeric Fc fusion protein that is mutated to promote dimer formation.</p><p> The present invention also provides a pharmaceutical composition comprising the heterodimer Fc fusion protein described above, as well as compositions and therapeutic methods for treating diseases, particularly cancers, using the same.</p>

<figref num="1">Figures 1 (A) to 1 (C) illustrate conventional strategies for obtaining monomer and heterodimer fusion proteins using wild-type Fc of human IgG antibodies. (A) Epo-Fc dimer vs. Epo-Fc monomer based on wild-type Fc. (B) Glycosylated Fc-fused GLP-1 / GCG monomeric peptide produced by LAP Covery technology. (C) Fc-FSH tandem homodimer vs. Fc-FSH heterodimer based on wild-type Fc. Epo, erythropoietin; GLP-1 / GCG, glucagon-like peptide-1 / glucagon; FSH, follicle-stimulating hormone. Figure 1 (D) constructs an antibody-cytokine (immunocytokine) by fusing a monomeric cytokine (IL2) to an IgG-type antibody containing a knock-into-hole (KiH) heterodimer Fc variant according to previous literature. An example of doing so is shown.</figref>

<figref num="2">Figures 2 (A) and 2 (B) illustrate the morphology of monomer and heterodimer fusion proteins that can be constructed using the heterodimer Fc. Potential use of heterodimer Fc for the production of Fc fusion monomers or heterodimer proteins to present their naturally occurring forms of fusion partners. The Fc fusion monomer can be easily produced by fusion of the monomeric protein to the N-terminus or C-terminus of one heterodimer Fc chain. FIG. 2 (C) illustrates a fusion protein formed by fusing a heterodimer to an IgG-type human antibody containing a heterodimer Fc. The Fc fusion heterodimer can be produced by the separate fusion of two subunits of the heterodimer protein to each strand of the N-terminal or C-terminal heterodimer Fc.</figref>

<figref num="3">FIG. 3 shows the sequence alignment of the CH3 domain of human IgG isotype antibodies (hIgG1, hIgG2, hIgG3, hIgG4) and the mutated residues (K370E / K409W) in the A107 heterodimer Fc variant._CH3A-E357N / D399V / F405T_CH3B) Is highlighted.</figref>

<figref num="4">FIG. 4 performs structural modeling of heterodimeric Fc variants for each isotype by using sequences with mutations induced at the positions selected in FIG. 3 and wild-types the resulting modeled structure. The results of analysis in comparison with the A107 variant based on type IgG1 are shown.</figref>

<figref num="5">FIG. 5 is a schematic diagram of a vector for expressing the heterodimer Fc for each isotype constructed by sequence and structural analysis in animal cells. Heterodimer Fc variants for each isotype containing mutated hinge regions were cloned into vectors using restriction enzymes (NotI / HindIII).</figref>

<figref num="6">FIG. 6 shows scFv-Fc for assessing the ability of heterodimer Fc variants to form heterodimers by differences in dimer size between expressed proteins._CH3A/ Fc_CH3BThe expression system is shown graphically.</figref>

<figref num="7">FIG. 7 shows scFv-Fc fused to a single chain variable fragment (scFv) constructed to evaluate the heterodimerization yield of antibody Fc by the CH3 variant pair as shown in FIG. , Is a schematic diagram for cloning into the pcDNA3.1 vector, which is an animal cell expression vector.</figref>

<figref num="8">FIG. 8 cotransfects HEK293F cells with a CH3 variant pair-introduced animal cell expression vector constructed by the expression system shown in FIGS. 5 and 7 to assess the formation of heterodimerization shown in FIG. The vector is then transiently expressed and purified, then 5 μg of protein is separated by SDS-PAGE under non-reducing conditions to assess the formation of heterodimerization, size and Coomassie blue staining. The results of protein analysis by the combination of are shown. Wild-type Fc with wild-type CH3 was used as a negative control.</figref>

<figref num="9">FIG. 9 shows the results of protein separation by SDS-PAGE by the method shown in FIG. 8 and then Western blotting using an anti-human IgG-AP conjugated antibody.</figref>

<figref num="10">FIG. 10 (A) is a schematic diagram showing the morphology of the endogenous IL-12 cytokine used as a control in the present invention without Fc fused. FIG. 10 (B) shows the divalent (bi) IL-12-Fc fusion protein obtained by fusing IL-12 cytokines to wild-type IgG4 Fc with an amino acid linker and used as a comparative example in the present invention. It is a schematic diagram which shows the form. FIG. 10 (C) shows the monovalent (mono) obtained by fusing the IL-12 cytokine to the IgG4-based γ4-A107 variant among the heterodimeric Fc variants for each isotype according to the invention. ) It is a schematic diagram showing the morphology of the IL-12-Fc fusion protein.</figref>

<figref num="11">11 (A) and 11 (B) are schematic diagrams of a vector for expressing and purifying the fusion protein of the example of the present invention (FIG. 10 (C)) in animal cells.</figref>

<figref num="12">FIG. 12 is a schematic diagram of a vector for expressing and purifying the fusion protein of the example of the present invention (FIG. 10 (B)) in animal cells.</figref>

<figref num="13">FIG. 13 shows that HEK293F cells were cotransfected with the animal cell expression vectors of FIGS. 11 (A) and 11 (B) constructed using human and mouse interleukin genes to transiently express the genes. After purification, 5 μg of protein was separated by SDS-PAGE under non-reducing conditions, and the results of protein analysis by combination of size and Coomassie blue staining are shown.</figref>

<figref num="14">FIG. 14 shows the results of analyzing the fusion protein of FIG. 13 by size exclusion chromatography (SEC).</figref>

<figref num="15">Figure 15 shows monovalent hIL on normal PMBC without IL-12 receptor, and PHA-activated PBMC that induced IL-12 receptor by treatment with mitogen PHA (phytohemaglutinin). The results of FACS analysis performed to analyze the binding affinity of -12-Fc and wild-type divalent hIL-12-Fc are shown.</figref>

<figref num="16">FIG. 16 shows Fc (A107), recombinant human IL-12 (rhIL-12), and divalent hIL-12-Fc on the proliferation of PHA-activated PBMCs induced by mitogen PHA treatment. And the results of the WST-1 cell proliferation assay performed to measure the effects of various concentrations of monovalent hIL-12-Fc are shown.</figref>

<figref num="17">FIG. 17 shows the results of an ELISA performed to measure the concentration of IFN-γ in the resulting culture supernatant as shown in FIG.</figref>

<figref num="18">Figure 18 shows that mIL-12 cross-reacts with human IL-12R on activated human T and NK cells, so by treatment with normal PMBC, which does not have the IL-12 receptor, and mitogen PHA. The results of flow cytometric analysis performed to measure the binding affinity of monovalent mIL-12-Fc and divalent mIL-12-Fc on PHA-activated PBMCs that induced the IL-12 receptor are shown. ..</figref>

<figref num="19">FIG. 19 shows Fc (A107), recombinant mouse IL-12 (rmIL-12), and divalent mIL-12-Fc affecting the proliferation of PHA-activated PBMCs induced by IL-12 receptor by mitogen PHA treatment. And the results of the WST-1 cell proliferation assay performed to measure the effects of various concentrations of monovalent mIL-12-Fc are shown.</figref>

<figref num="20">FIG. 20 (A) shows CT26 during intraperitoneal administration of Fc (A107), rmIL-12, divalent mIL-12-Fc and monovalent mIL-12-Fc.^HER2The changes in tumor volume in Balb / c mice transplanted with / Neu tumor, and photographs of tumor-bearing mice after sacrifice at the end of administration are shown. Injection of mIL12-Fc protein was initiated 11 days after tumor cell inoculation when the tumor volume reached 100 mm3. FIG. 20 (B) is a graph showing changes in mouse body weight measured at the time shown in the experimental procedure shown in FIG. 20 (A).</figref>

<figref num="21-1">Figure 21 (A) shows CT26^{HER2 / Neu}Tumor volume of Balb / c mice transplanted with 300 mm³The results of measuring changes in mouse tumor volume measured during intraperitoneal administration of various concentrations of divalent mIL-12-Fc and monovalent mIL-12-Fc twice a week when reaching. FIG. 21 (B) is a graph showing changes in individual mouse tumor volumes treated with the mIL12-Fc protein at the time indicated in the experimental procedure shown in FIG. 21 (A). FIG. 21 (C) shows photographs of tumors taken from tumor-bearing mice 3 days after the last dose of FIG. 21 (A). FIG. 21 (D) is a graph showing changes in mouse body weight measured at the time shown in the experimental procedure shown in FIG. 21 (A).</figref>

<figref num="21-2">FIG. 21 (E) is a graph showing the results of measuring alanine aminotransferase (ALT) (a hepatotoxic marker) in blood collected from the facial veins of mice one day after the final administration of FIG. 21 (A). Is.</figref>

<figref num="22A">FIG. 22 (A) shows CD4 in the spleen of mice sacrificed 3 days after the last dose of FIG. 21 (A).⁺T cells, CD8⁺It is a graph which shows the result of having measured the increase in the number of T cells and NK cells.</figref>

<figref num="22B">FIG. 22 (B) shows total immune cells, CD4, infiltrating the tumor of the sacrificed mouse 3 days after the third dose of FIG. 21 (A).⁺T cells and CD8⁺It is a graph which shows the number of T cells.</figref>

<figref num="23A">Figure 23 (A) shows CT26 treated with mIL-12-Fc protein.^HER^{2 / neu}The results of an ELISA performed to measure serum levels of IFN-γ in tumor-bearing mice are shown. After coagulating the blood collected from the facial vein of the mouse 24 hours after the final administration of FIG. 21 (A), the mouse serum was separated.</figref>

<figref num="23B">Figure 23 (B) shows CT26^{HER2 / Neu}Tumor volume of Balb / c mice transplanted with cancer cells is 300 mm³When divalent mIL-12-Fc and monovalent mIL-12-Fc were intraperitoneally administered at an equimolar concentration of 1 μg of rmIL-12, the facial veins of mice were administered 1, 3 and 5 days later. It is a graph which shows the result of the ELISA performed to measure the concentration of IFN-γ in the serum separated from the blood collected from.</figref>

<figref num="23C">Figure 23 (C) shows CT26^{HER2 / Neu}It is a graph which shows the result of having measured the cytotoxic effect of the cytotoxic T cell isolated from the spleen of the mouse which was sacrificed 3 days after the last administration of FIG. 21 (A) with respect to the cancer cell.</figref>

<figref num="23D">FIG. 23 (D) shows CT26 expressing the tumor antigen 3 days after the third administration of FIG. 21 (A).^{HER2 / Neu}CD8 of the spleen isolated from CT26-HER2 / neu-bearing tumor mice treated with mIL-12-Fc protein analyzed by culturing for 4 hours with 4T1 cells expressing cancer cells and tumor antigens.⁺Shows cytotoxic activity of T cells.</figref>

<figref num="23E">Figure 23 (E) shows CT26^{HER2 / Neu}It is a graph which shows the result of having measured the cytotoxic effect of the natural killer cell isolated from the spleen of the mouse which was sacrificed 3 days after the 3rd administration of FIG. 21 (A) with respect to a cancer cell.</figref>

<figref num="24A">FIG. 24 (A) shows CD8 isolated from the spleen isolated from tumor-bearing mice sacrificed 3 days after the last dose of FIG. 21 (A).⁺It is a graph which shows the result of having measured the number of effector T cells.</figref>

<figref num="24B">FIG. 24 (B) shows CD8 in the spleen isolated from tumor-bearing mice sacrificed 3 days after the last dose of FIG. 21 (A).⁺It is a graph which shows the result of having measured the number of effector memory T cells.</figref>

<figref num="24C">FIG. 24 (C) shows CD8 in the spleen isolated from tumor-bearing mice sacrificed 3 days after the last dose of FIG. 21 (A).⁺It is a graph which shows the result of having measured the number of central memory T cells.</figref>

<figref num="24D">FIG. 24 (D) shows CT26 in surviving Balb / c mice 120 days after administration of 1 μg of monovalent IL-12-Fc in FIG. 21 (A).^{HER2 / Neu}The results obtained by retransplanting cancer cells and measuring changes in tumor volume in mice are shown.</figref>

<figref num="24E">FIG. 24 (E) shows CD8 in the spleen isolated from tumor-bearing mice sacrificed 3 days after the third dose of FIG. 21 (A).⁺Memory precursor effector cells between T cells (KLRG1)^-IL-7R⁺) And short-lived effector cells (KLRG1)⁺IL-7R^-The results of flow cytometry performed to analyze the proportion of) are shown.</figref>

<figref num="25A">FIG. 25 (A) shows CD8 in splenocytes isolated from mice sacrificed 3 days after the third dose of FIG. 21 (A).⁺It is a graph which shows the result of the flow cytometry analysis performed to measure the ratio of T cells (which showed high expression of the transcription factor T-bet which inhibits the differentiation of memory cells).</figref>

<figref num="25B">FIG. 25 (B) shows CD8 in splenocytes isolated from mice sacrificed 3 days after the third dose of FIG. 21 (A).⁺It is a graph which shows the result of the flow cytometry analysis performed to measure the proportion of T cells (showing high expression of Eomes and low expression of T-bet).</figref>

<figref num="25C">Figure 25 (C) shows CT26^{HER2 / Neu}Tumor volume of Balb / c mice transplanted with cancer cells is 300 mm³Twenty-four hours after intraperitoneal administration of divalent mIL-12-Fc and monovalent mIL-12-Fc at a concentration equivalent to 1 μg of rmIL-12, the tumor inflow region (inguinal region) ) CD8 isolated from lymph nodes⁺It is a graph which shows the result of the flow cytometry analysis performed to measure the expression level of phosphorylated STAT4 in a T cell.</figref>

<figref num="25D">FIG. 25 (D) shows CD8 in the tumor influx region (inguinal) lymph node 72 hours after the single intraperitoneal administration of FIG. 25 (C).⁺It is a graph which shows the result of the flow cytometry analysis performed to measure the ratio of T cells (expressing T-bet which inhibits the differentiation of memory cells).</figref>

<figref num="25E">Figure 25 (E) shows CD8 isolated from spleen and inguinal lymph nodes in normal Balb / c mice.⁺Results of flow cytometric analysis performed to measure pSTAT4 expression levels when T cells were stimulated with monovalent mIL-12-Fc and divalent mIL-12-Fc cross-reacted with anti-Fc antibody. It is a graph which shows.</figref>

<figref num="25F">Figure 25 (F) shows CD8 isolated from spleen and inguinal lymph nodes in normal Balb / c mice.⁺T-bet expression CD8 when T cells were stimulated with monovalent mIL-12-Fc and divalent mIL-12-Fc cross-reacted with anti-Fc antibody⁺It is a graph which shows the result of the flow cytometry analysis performed to measure the proportion of T cells.</figref>

<figref num="26">FIG. 26 is an overall schematic showing the mechanism by which monovalent mIL-12-Fc induces the differentiation of memory precursor effector cells and the mechanism by which divalent mIL-12-Fc induces the differentiation of short-lived effector cells. is there.</figref>

Unless otherwise stated, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art of the present invention. In general, the nomenclature used herein, and the experimental methods described below, are well known and commonly used in the art.

In one aspect, the invention is a heterodimeric Fc fusion protein comprising a first Fc region and a second Fc region of an immunoglobulin Fc pair and a physiologically active protein, wherein there are two or more physiologically active proteins. Composed of different subunits, two or more different subunits exhibit physiological activity by forming a protein complex, and one or more subsystems of the physiologically active protein are the first Fc region and / Or linked to one or more ends of the N or C ends of the second Fc region, the CH3 domains of the first Fc region and the second Fc region are mutated to promote the formation of heterodimers. With respect to the heterodimer Fc fusion protein to be made.

As used herein, the term "Fc region" or "heavy chain constant region" means a region that includes the immunoglobulin CH2 domain, CH3 domain and hinge domain. However, in the case of IgE, this term means a region that includes the CH2 domain, CH3 domain, CH4 domain and hinge domain.

As used herein, the expression "the first Fc region and the second Fc region are mutated to promote heterodimer formation" means that naturally occurring antibodies have the same sequence of two Fc regions. Some of these Fc region sequences are mutated, thus allowing the heterodimer through a specific non-covalent interaction between the first Fc region and the second Fc region. It means that the formation can be promoted, or the formation of homodimers can be reduced, or preferably it can hardly occur.

Preferably, "the first Fc region and the second Fc region are mutated to promote the formation of heterodimers" is contained in the "first Fc region and the second Fc region from immunoglobulins". Each of the CH3 domains can be mutated to promote the formation of Fc heterodimers. "

In the present invention, the "heterodimer Fc or Fc heterodimer" includes a first Fc region and a second Fc region, and the first Fc region and the second Fc region are the first Fc region and the second Fc region. It means a heterodimer in which the CH3 domain of the Fc region of the Fc region is mutated to promote the formation of the Fc heterodimer.

In the present invention, each of the first Fc region and the second Fc region can be derived from an Fc region selected from the group consisting of human IgG1, IgG2, IgG3, IgG4, IgM, IgA, IgD and IgE. , Preferably, each of the first Fc region and the second Fc region is derived from IgG1, IgG2, IgG3 or IgG4.

In addition, the first Fc region and the second Fc region can be derived from isotype antibodies.

In another aspect, mutations in the CH3 domain can include one or more mutations selected from the following group (all mutation positions in the invention are numbered according to the EU index):

(1) Substitution of the amino acid residue at position K370 in the CH3 domain of the first Fc region; and substitution of the amino acid residue at position E357 and / or S364 of the CH3 domain of the second Fc region; and / or (2) ) Substitution of the amino acid residue at position K409 in the CH3 domain of the first Fc region; and substitution of the amino acid residue at position F405 and / or D399 in the CH3 domain of the second Fc region.

Preferably, the substitution of the amino acid residue at position K370 in the CH3 domain of the first Fc region may be K370E, K370R, K370M, K370D or K370H, and the amino acid residue at position E357 in the CH3 domain of the second Fc region. The substitution of the group may be E357N, E357D, E357A, E357I, E357G or E357M, and the substitution of the amino acid residue at position S364 of the CH3 domain of the second Fc region may be S364T or S364W.

Furthermore, the substitution of the amino acid residue at position K409 in the CH3 domain of the first Fc region may be K409W, and the substitution of the amino acid residue at position F405 in the CH3 domain of the second Fc region is F405T. Often, the substitution of the amino acid residue at position D399 in the CH3 domain of the second Fc region may be D399V.

Amino acid residue mutations such as K370E mean that K at position 370 is mutated to E, and all amino acid residue mutations in the present invention are used interchangeably as described above.

Most preferably, mutations in the CH3 domain of the first Fc region or the second Fc region include one or more mutations selected from the following group (mutation locations are numbered according to the EU index): Can: (1) Substitution of amino acid residue at position K370 in the CH3 domain of the first Fc region K370E, K370R, K370M, K370D or K370H; (2) Amino acid at position E357 in the CH3 domain of the second Fc region Substitution of residues E357N, E357D, E357A, E357I, E357G or E357M, and substitution of the amino acid residue at position S364 in the CH3 domain of the second Fc region S364T or S364W; (3) Substitution of the CH3 domain of the first Fc region Substitution of amino acid residue at position K409 K409W; and (4) Substitution of amino acid residue at position F405 in the CH3 domain of the second Fc region F405T, and substitution of the amino acid residue at position D399 in the CH3 domain of the second Fc region Replacement D399V.

The CH3 domains of the first Fc region and the second Fc region can further contain the following residues: (i) Cysteine substituted at the Y349 position of the CH3 domain of the first Fc region (C); And (ii) Cysteine (C) substituted at the S354 position in the CH3 domain of the second Fc region.

In yet another embodiment, the mutation in the CH3 domain can include one or more mutations selected from the following group: (1) the amino acid residue at position K360 of the CH3 domain in the first Fc region. Substitution; and substitution of the amino acid residue at position E347 in the CH3 domain of the second Fc region; and / or (2) substitution of the amino acid residue at position K409 in the CH3 domain of the first Fc region; and the second Fc Substitution of amino acid residues at positions F405 and D399 in the CH3 domain of the region.

Preferably, the substitution of the amino acid residue at position K360 in the CH3 domain of the first Fc region may be K360E, and the substitution of the amino acid residue at position E347 in the CH3 domain of the second Fc region may be E347R. ..

The substitution of the amino acid residue at position K409 in the CH3 domain of the first Fc region may be K409W, the substitution of the amino acid residue at position F405 in the CH3 domain of the second Fc region may be F405T, and the substitution of the amino acid residue at position F405 may be F405T. The substitution of the amino acid residue at position D399 in the CH3 domain of the Fc region of Fc region may be D399V.

Most preferably, mutations in the CH3 domain of the first Fc region or the second Fc region include one or more mutations selected from the following group (mutation locations are numbered according to the EU index): Can: (1) Substitution of amino acid residue at position K360 in the CH3 domain of the first Fc region K360E; (2) Substitution of the amino acid residue at position E347 of the CH3 domain of the second Fc region E347R; (3) ) Substitution of the amino acid residue at position K409 in the CH3 domain of the first Fc region K409W; and (4) Substitution of the amino acid residue at position F405 in the CH3 domain of the second Fc region F405T, and in the second Fc region Substitution of amino acid residue at position D399 in the CH3 domain D399V.

The CH3 domains of the first Fc region and the second Fc region can further contain the following residues: (i) Cysteine substituted at the Y349 position of the CH3 domain of the first Fc region (C); And (ii) Cysteine (C) substituted at the S354 position in the CH3 domain of the second Fc region.

Preferably, each of the CH3 domains contained in the first Fc region and the second Fc region from the immunoglobulin according to the present invention is an amino acid sequence selected from the group consisting of the amino acid sequences represented by the following SEQ ID NOs: Can have: (1) SEQ ID NO: 1 and SEQ ID NO: 2; (2) SEQ ID NO: 3 and SEQ ID NO: 4; (3) SEQ ID NO: 5 and SEQ ID NO: 6; (4) SEQ ID NO: 8 and SEQ ID NO: 9; (5) SEQ ID NO: 11 and SEQ ID NO: 12; and (6) SEQ ID NO: 14 and SEQ ID NO: 15.

In particular, the first Fc region and the second Fc region from the immunoglobulin according to the invention preferably have the sequences of the IgG4 CH3 domains shown in Table 1 below.<tables num="1"><img file="JP2019536734A_D0001.tif" /></tables>

In the heterodimeric Fc fusion protein according to the invention, the subunit of the physiologically active protein can be linked to only any one terminal at the N-terminus or C-terminus of the first Fc region or the second Fc region. One or more different subsystems of a physiologically active protein can be linked to each of the N-terminus and C-terminus of the first Fc region and the second Fc region (Fig. 2 (B) and See 2 (C)).

"A subsystem of a physiologically active protein is linked to only any one terminal at the N-terminus or C-terminus of the first Fc region or the second Fc region." Linking to only any one of the four terminus of the N- or C-terminus of the first Fc region or the second Fc region, and the remaining subsystems of the bioactive protein (s) It means that it is linked by a linker to a subsystem of a physiologically active protein that is linked to any one terminal at the N-terminus or C-terminus of the first Fc region or the second Fc region. The linker is preferably an amino acid linker, but is not limited thereto.

In addition, "one or more different subunits of a single bioactive protein are linked to each of the N-terminus and C-terminus of the first Fc region and the second Fc region" is single. One or more different subsystems of a physiologically active protein are linked to the N-terminus of each of the first Fc region and the second Fc region, or one or more of a single physiologically active protein. Different subunits are linked to the C-terminus of each of the first and second Fc regions, or one or more different subsystems of a single bioactive protein are in the first Fc region. And means that it is linked to the N-terminus and C-terminus of the second Fc region, respectively.

In the heterodimer Fc fusion protein according to the invention, the subunits of the physiologically active protein can be linked to the N-terminus and / or C-terminus of the first Fc region and / or the second Fc region by gene fusion.

In yet another embodiment, the subunits of the bioactive protein can be linked to the first Fc region and the second Fc region through a linker. The linker is preferably an amino acid linker, but is not limited thereto.

In yet another embodiment, in the heterodimeric Fc fusion protein according to the invention, the bioactive protein is composed of two or more different subunits and the two or more different subunits are protein complexes. It is characterized by exhibiting physiological activity by forming a body.

"A bioactive protein is composed of two or more different subunits, and two or more different subunits exhibit bioactivity by forming a protein complex." It means that two or more subunits exhibit bioactivity when forming a protein complex.

Physiologically active proteins are interleukin-12 (IL-12), interleukin-23 (IL-23), interleukin-27 (IL-27), interleukin-35 (IL-35) and follicle-stimulating hormone (FSH). ), But not limited to. Moreover, it will be apparent to those skilled in the art that any bioactive protein suitable for the purposes of the present invention can be used in the present invention.

Most preferably, the bioactive protein according to the invention is IL-12.

A protein that is composed of two or more different subunits and exhibits bioactivity by forming a protein complex according to the invention in which two or more different subunits are bioactive is a preferred bioactive protein. IL-12 will be described in detail here as an example.

IL-12 is composed of two subunits, p35 (IL-12A) and p40 (IL-12B), and the bioactive form of IL-12 is p70, a heterodimer of p35 and p40. IL-12 should be present in the form of p70, a heterodimer of p35 and p40, for IL-12 to exhibit its activity in natural systems.

In the present invention, the morphology of the heterodimer Fc fusion protein according to the present invention has been embodied in order to mimic the morphology of naturally occurring IL-12 as much as possible.

Specifically, as described above, one or more subsystems of a physiologically active protein are linked to one or more ends of the N-terminus or C-terminus of the first Fc region and the second Fc region. In the heterodimeric Fc fusion protein containing the first Fc region and the second Fc region according to the present invention, (i) one or more subsystems constituting the physiologically active protein are the first Fc region or the first Fc region. Only one of the N-terminus or C-terminus of the Fc region of 2 can be linked, and the remaining subsystems of the physiologically active protein (s) can be linked by a linker, or (ii) simply. One or more different subunits of a bioactive protein can be linked to the N-terminus and / or C-terminus of each of the first and second Fc regions, respectively.

In the above cases, an example of IL-12 will be described below.

In the case of (i), the p35 or p40 subunit of IL-12 can be linked to only any one end of the N-terminus or C-terminus of the first Fc region or the second Fc region, and the rest. The subunits are linked by a linker to the p35 or p40 subunit linked to any one of the N-terminus or C-terminus of the first Fc region or the second Fc region to form a heterodimer Fc fusion protein. (See Figures 2 (B) and 2 (C)).

In the case of (ii), any one selected from the p35 or p40 subunits of IL-12 can be linked only to the N-terminus or C-terminus of the first Fc region, and the other subunits Heterodimer Fc fusion proteins can be formed by linking only to the N-terminus or C-terminus of the second Fc region (see Figures 2 (B) and 2 (C)).

It was found that this morphology exhibited in vitro bioactivity similar to that of conventional recombinant IL-12 protein while maintaining the original heterodimer morphology that exists naturally (Fig. 2 (B), 2 (Fig. 2 (B)), 2 ( See C) and 10 (C)).

Therefore, the preferred immunoglobulin heterodimer Fc fusion protein according to the invention is that the physiologically active protein is IL-12, and that the p35 or p40 subunit of IL-12 is the first Fc region or the second Fc region. It is linked to only one of the N-terminus or C-terminus of the subunit, and the remaining subunits are linked to any one of the N-terminus or C-terminus of the first Fc region or the second Fc region. That the subunits are linked by a linker, or that the p35 and p40 subunits of IL-12 are linked to each of the N-terminus and C-terminus of the first and second Fc regions, respectively. It is a feature.

In another aspect, in the heterodimer Fc fusion protein according to the invention, the hinge domain contained at the N-terminal of each of the first Fc region and the second Fc region is mutated with a cysteine residue contained in the hinge domain. It can be characterized by being.

Preferably, the cysteine residue mutation in the hinge domain is such that all cysteine residues in the upper hinge region are replaced with serine residues, except for the cysteine residue in the core hinge domain for heterodimer formation. However, the scope of the present invention is not limited thereto.

Furthermore, in the present invention, the first Fc region and the second Fc region may be included in a complete antibody form consisting of human IgG1, IgG2, IgG3, IgG4, IgM, IgA, IgD and IgE.

In the present invention, the term "complete antibody form" refers to the CH2 domain, CH3 domain and hinge domain (including the CH4 domain in IgE) in the Fc region of IgG, IgA and IgD, as well as the CH1 domain, VH domain, CL. Means an intact antibody that further includes a domain and a VL domain.

In yet another aspect, the invention relates to a pharmaceutical composition comprising a heterodimer Fc fusion protein according to the invention. The use of the pharmaceutical composition according to the invention can depend on the use of the bioactive protein contained in the heterodimer Fc fusion protein.

Preferably, the bioactive protein contained in the heterodimer Fc fusion protein according to the invention may be IL-12 or one or more subunits thereof. Therefore, the present invention provides a pharmaceutical composition for the treatment of cancer, comprising a heterodimer Fc fusion protein containing IL-12 as a bioactive protein.

Cancers that can be treated with a pharmaceutical composition for the treatment of cancer, including IL-12 as a physiologically active protein or a heterodimer Fc fusion protein containing one or more subunits, are colorectal cancer, You can choose from the group consisting of melanoma, breast cancer, pancreatic cancer, kidney cancer, prostate cancer, ovarian cancer, small bowel cancer, esophageal cancer, cervical cancer, lung cancer, lymphoma and blood cancer. , Not limited to that.

The pharmaceutical composition according to the invention can further comprise a pharmaceutically acceptable carrier. The term "pharmaceutically acceptable carrier" refers to a substance that can be added to the active ingredient to aid in the formulation or stabilization of the preparation and does not cause a significant toxicological effect on the patient.

As used herein, the term "pharmaceutically acceptable carrier" refers to a carrier or diluent that does not irritate the patient and does not impair the biological activity and properties of the heterodimeric Fc fusion proteins according to the invention. Point to. Aseptic and biocompatible carriers are used as pharmaceutically acceptable carriers in compositions formulated as liquid solutions. Pharmaceutically acceptable carriers are saline, sterile water, Ringer's solution, buffered saline, albumin injection solution, dextrose solution, maltodextrin solution, glycerol, ethanol or a mixture of two or more thereof. Good. In addition, the compositions of the present invention may optionally contain other conventional additives, including antioxidants, buffers and bacteriostatic agents. Furthermore, the compositions of the present invention can be formulated as injectable forms such as aqueous solutions, suspensions or emulsions with diluents, dispersants, surfactants, binders and lubricants. In addition, the compositions according to the invention can be formulated in the form of pills, capsules, granules or tablets. Other carriers are available in the literature [Remington's Pharmaceutical Sciences (EW) Martin)].

Pharmaceutically acceptable carriers include sterile aqueous solutions or dispersions and sterile powders for the immediate preparation of sterile injectable solutions or dispersions. The use of such vehicles and agents for pharmaceutically active substances is known in the art. The composition is preferably formulated for parenteral injection. The composition can be formulated as a solid, solution, microemulsion, liposome, or other ordered structure suitable for high drug concentrations. The carrier may be a solvent or dispersion medium containing, for example, water, ethanol, polyol (eg, glycerol, propylene glycol and liquid polyethylene glycol) and suitable mixtures thereof. In some cases, the composition is isotonic. It can contain agent), such as sugars, polyhydric alcohols, such as sorbitol or sodium chloride. Aseptic injectable solution is prepared by sterile microfiltration followed by the required amount of heterodimer Fc fusion protein in a suitable solvent, optionally with one or a combination of the components listed above. be able to. Generally, the dispersion is prepared by incorporating the active compound into a sterile vehicle containing the basic dispersion medium and other components required from those listed above. In the case of sterile powders for the preparation of sterile injectable solutions, the preferred method of preparation is to vacuum dry and freeze to give a powder with the active ingredient and any additional desired ingredients from that solution that has been pre-filtered and sterilized. It's dry.

In addition, the pharmaceutical compositions according to the invention can be administered orally or parenterally to a patient at dosages and frequencies that can vary depending on the severity of the patient. The composition can be administered to the patient in need, either as a bolus or by continuous infusion. In another example, the pharmaceutical composition according to the invention can be administered rectal, intravenously, subcutaneously, intrauterine, or intracerebrovascularly, but not limited to.

In addition, pharmaceutical compositions for cancer treatment containing an immunoglobulin heterodimer Fc fusion protein containing IL-12 can be used for combination therapy with other anti-cancer drugs. Other anti-cancer agents are preferably cytotoxic T cells and / or natural killer (NK) cells, but are not limited to all anti-cancer agents that can be used in the art of the present invention. Can be used for combination therapy.

In particular, pharmaceutical compositions for cancer treatment containing immunoglobulin heterodimer Fc fusion proteins, including IL-12, are used for combination therapy with cytotoxic T cells and / or natural killer (NK) cells. When it can induce: (1) increased cytokine secretion stimulated by T cells or natural killer (NK) cells; (2) antibody-dependent cell-mediated cell injury (ADCC) or cytotoxic T lymph Increased sphere (CTL) response; (3) Increased number of cytotoxic T lymphocytes (CTL) and / or natural killer cells; (3) Increased peritumor lymphocyte introduction; or (4) In vivo Increased IL-12R beta 1 and IL-12R beta 2 signaling in lymphocytes.

In yet another aspect, the invention relates to a method of treating or preventing a disease comprising administering to a patient in need of treatment a pharmaceutical composition comprising a heterodimer Fc fusion protein according to the invention.

As with the composition, the disease that can be treated or prevented depends on the use of the bioactive protein contained in the heterodimer Fc fusion protein. Preferably, when one or more subsystems of the physiologically active protein contained in the heterodimeric Fc fusion protein according to the invention is one or more subsystems of IL-12, the present invention relates to cancer. Especially from the group consisting of colonic rectal cancer, melanoma, breast cancer, pancreatic cancer, kidney cancer, prostate cancer, ovarian cancer, small bowel cancer, esophageal cancer, cervical cancer, lung cancer, lymphoma and hematological cancer. Provided are methods of cancer treatment or prevention for patients with selected cancers.

<p> Hereinafter, the present invention will be described in more detail by way of examples. It will be apparent to those skilled in the art that these examples are for illustration purposes only and should not be considered as limiting the scope of the invention.</p><p> (Example 1: Design of antibody Fc CH3 domain variant for heterodimer formation for each human immunoglobulin isotype (sequencing)) A CH3 domain that plays a major role in interactions for heterodimer formation to generate heterodimer Fc fragments for each human immunoglobulin isotype by introducing CH3 domain mutations that favor heterodimer formation. The amino acid sequence similarity between them was first analyzed as follows. In this regard, previous literature or patent documents (Choi et al. 2016; Korean patent application) such that heterodimerization of CH3A: CH3B drives Fc variants to form heterodimers with high yields. According to the strategy published in No. 2015-0142181), the heterodimer CH3A: CH3B pair (in the present invention, CH3A and CH3B mean the CH3 domain of the first Fc region and the CH3 region of the second Fc region, respectively. A heterodimer Fc variant (A107) was generated by introducing an asymmetric mutation at the CH3 homodimer interface. Figure 3 aligns and compares the sequences of the CH3 domains for each human antibody immunoglobulin G (IgG) isotype. Each amino acid sequence was obtained from the International Immunogenetic Information System (IMGT; URL: http://www.imgt.org/). In particular, among various allotypes, the sequence of G3m (s, t), which was reported to maintain serum half-life at levels similar to those of other IgG isotypes, was used for IgG3 (Stapleton). NM et al., 2011).</p><p> Sequencing results differ from those in IgG1, IgG2 and IgG3, where IgG4 is conserved in all isotypes, except that the amino acid at position 409 of the position where the A107 mutation is introduced is arginine. It was shown to have. Therefore, a position having the same amino acid SEQ ID NO: was selected as the position for introducing the A107 mutant pair into an isotype other than IgG1. The positions of all amino acids in the present invention are numbered according to the EU index.</p><p> (Example 2: Design of immunoglobulin Fc CH3 domain variant for heterodimer formation for each human immunoglobulin isotype (structural modeling)) Before actually constructing the CH3 domain variant for each isotype In addition, whether or not the A107 mutant pair could be stably introduced at the position selected in Example 1 so as to form a heterodimer was determined by using the variant sequence introduced in each mutation as shown in FIG. Predicted through structural modeling. Structural modeling uses an online modeling server (URL: https://swissmodel.expasy.org/; Biasini M et al.) Using a known immunoglobulin Fc heterodimer variant structure (PDB ID: 4X98) as a template. , 2014). To observe the structural changes in the CH3 domain and the location of the A107 mutation after the introduction of the mutation, each of the resulting structures was overlaid using Pymol software capable of visualizing the protein structure. In the stacked structure, even when the A107 mutation is introduced into each isotype, it is constructed based on the IgG1 isotype, CH3A: CH3B. It was found that the structure was maintained without major changes compared to the modeled structure of the conventional A107 variant forming the Fc heterodimer. In particular, it was shown that the orientation of the introduced A107 mutant amino acid residues was almost consistent, and that the distance of interaction between the mutated amino acids was maintained at similar levels (see Figure 4). I want).</p><p> (Example 3: Construction of A107 heterodimer Fc isotype variant for each human immunoglobulin isotype) An A107 heterodimer Fc isotype variant designed through sequencing of Example 1 and structural modeling of Example 2 Animal cell expression vector pcDNA3 to have the signal sequence-hinge-CH2-CH3 using NotI / HindIII restriction enzymes and synthetic oligonucleotides (Macrogen, Korea) by site-directed mutagenesis methods performed by those skilled in the art. Clone in frame in .1 (+) (Invitrogen, USA) (see Figure 5).</p><p> In the hinge domain used, serine residues in the upper hinge region, other than the cysteine residues in the core hinge region for heterodimer formation, were left in order to block the formation of disulfide bonds during protein fusion. Substituted with a group. In particular, in the case of IgG3, IgG3's high antibody effector function (ADCC and CDC) is maintained by only the C-terminal 15 amino acids of the core hinge domain out of the 47 amino acids of the G3m (s, t) allotype hinge domain. That was found in the literature (Dall'Acqua WF et al., 2006). Therefore, only the C-terminal 15 amino acids of the sequence shown in FIG. 5 were used.</p><p> Table 2 below shows the amino acid sequence information of the CH3 region in the Fc variant pair of the wild type and the A107 heterodimer of the present invention.<tables num="2-1"><img file="JP2019536734A_D0002.tif" /></tables><tables num="2-2"><img file="JP2019536734A_D0003.tif" /></tables></p><p> (Example 4: Evaluation of heterodimerization ability of A107 heterodimer Fc variant for each human immunoglobulin isotype) The A107 heterodimer Fc isotype variant constructed in Example 3 is a wild-type A107 variant. ScFv-Fc, which is primarily used to assess the heterodimerizing ability of Fc variants in the same type of study to determine if they actually have a heterodimerizing ability similar to that of the one._CH3A/ Fc_CH3BAn expression system was used (Choi et al., 2013). Figure 6 shows scFv-Fc_CH3A/ Fc_CH3BIt is a schematic diagram which shows the expression system. scFv-Fc_CH3A/ Fc_CH3BThe antibody purified in the expression system is scFv-Fc._CH3AHomo dimer (103kDa), scFv-Fc_CH3A/ Fc_CH3BHeterodimer (78 kDa) and Fc_CH3BSince they show different molecular weights among homodimers (53 kDa), the degree of heterodimer formation can be compared by SDS-PAGE.</p><p> Fc_CH3BAs the vector, the vector constructed in Example 3 was used. In addition, scFv to Fc_CH3ABy introducing only to the N-terminus of, ie, pcDNA3.1 (+)-scFv-hinge-CH2-CH3A (scFv-Fc)_CH3A) The vector was cloned by providing the format. Figure 7 shows scFv-Fc_CH3A/ Fc_CH3BAnimal cell expression vector used in the expression system pcDNA3.1 (+)-scFv-hinge-CH2-CH3A (scFv-Fc)_CH3A) Is a schematic diagram. The scFv antibody used is an antibody obtained by ligating the VH and VL regions of hAY4a, an affinity-enhanced version of the humanized antibody hAY4 that specifically binds to DR4 (Lee, Park et al. 2010). Cloning was performed using NotI and BsiWI restriction enzymes located just before the hinge domain. Wild-type Fc was constructed in the same format (scFv-Fc / Fc) as a control for variants.</p><p> (Example 5: Expression and purification of antibody containing A107 heterodimer Fc variant for each human immunoglobulin isotype) Constructed scFv-Fc_CH3AAnd Fc_CH3BCo-expression of is temporarily transfected into HEK293-F cells (Invitrogen) with a mixture of expression vector (1: 1 ratio) and polyethyleneimine (PEI) (Polyscience) and contains serum-free FreeStyle293 expression medium. Performed by culturing the cells in a shaking flask. The detailed method is as follows.</p><p> 2.0 x 10 HEK293-F cells in 100 ml medium for 200 mL transfection in a shaking flask (Corning)⁶Seed at a density of cells / ml, 8% CO₂Incubated at 150 rpm below. To generate each humanized antibody, place 250 μg total (2.5 μg / ml) of heavy and light chain plasmids for each antibody in 10 ml of FreeStyle 293 expression medium (Invitrogen) (125 μg for light chain and heavy chain). Dilute with 125 μg) and mix the medium with 10 ml of medium containing 750 μg of diluted PEI therein (7.5 μg / ml). The medium mixture was incubated at room temperature for 10 minutes. The incubated medium mixture was then added to 100 ml of seeded cells and 8% CO₂Incubate at 150 rpm below for 4 hours, then add the remaining 100 ml of FreeStyle 293 expression medium to it, followed by incubation for 5-7 days. During this incubation, cells produced antibodies, ie antibodies containing Fc variants, were secreted from the cells and accumulated in the medium. For this reason, proteins were purified using a Protein A Sepharose column (GE Healthcare) from cell culture supernatants collected by centrifugation at 2500 rpm for 20 minutes after cell culture. In this case, the purification step was performed with reference to the standard protocol provided by the protein A column company. Purified protein was quantified by measuring the absorbance at a wavelength of 562 nm using the solution in the BCA protein assay kit (Thermo) and determining the amount by the standard curve created.</p><p> (Example 6: Evaluation of heterodimerization ability of A107 heterodimer Fc variant for each human immunoglobulin isotype) 5 μg antibody containing the A107 heterodimer Fc variant of each isotype purified in Example 5 Was analyzed by 12% SDS-PAGE under non-reducing conditions (Fig. 8). The CH3A variant homodimer was observed at 103 kD; the CH3B variant homodimer was observed at 53 kD; the CH3B variant monomer was observed at 25 kD; the CH3A variant and CH3B variant heterodimer was observed at 78 kD. It was observed. Western blotting was also performed to more accurately determine the degree of homodimerization. 0.1 μg of the protein, which was smaller than that in the 12% SDS-PAGE analysis, was isolated under non-reducing conditions, and the protein was then subjected to anti-human IgG-AP conjugated antibody (Sigma) by conventional methods known in the art. Western blotting was performed by processing (Fig. 9).</p><p> As can be seen in FIGS. 8 and 9, in the IgG1 heterodimer introduced with the wild-type CH3 domain as a control, each homodimer of CH3A and CH3B and the CH3A: CH3B heterodimer were all observed by SDS-PAGE, but IgG1 The A107 heterodimer Fc variant for each human immunoglobulin isotype obtained by introducing the A107 heterodimerization mutation into IgG2, IgG3 and IgG4, except for the previously reported IgG1 All heterodimers were formed with yields similar to or higher than those of the A107 variant based on. At this time, for IgG4 variants, Fc monomers containing CH3A or CH3B (half Fc) were also observed, which is one of the properties of naturally occurring IgG4 and the occurrence of Fab arm exchange in blood. It results from the property of forming half Fc relative to the hinge domain (particularly serine at position 228 of the core hinge region) before (Liu H et al., 2012).</p><p> (Example 7: Construction of human / mouse IL-12 fusion protein) The isotype variants of Examples 1-6 are at a level similar to that of the previously reported IgG1-based A107 heterodimer Fc variant. It was found to retain the ability to heterodimerize. Among these isotype variants, an IgG4-based variant (γ4-A107) was used to construct a long-lasting IL-12 fusion protein. Naturally occurring IL-12 is composed of two subunits, the p35 subunit (p35; IL-12A) and the p40 subunit (p40; IL-12B), which interact to activate. Form a heterodimer with. The formation of this heterodimer is achieved because the two subunits are more strongly and stably coupled by a single disulfide bond between the two subunits. Therefore, two subunits of IL12 (p35 and p40) were genetically fused to the N-terminus of each heterodimer Fc chain to maintain the heterodimeric form of naturally occurring cytokines.</p><p> As a heterodimer Fc variant for the construction of the fusion protein, γ4-A107, which is based on IgG4 and can form a heterodimer by introducing an A107 mutation, was used. As previously reported, the antibody effector function of IgG1 (such as ADCC / CDC) rather promotes in vivo clearance in the construction of immunocytokines, which are fusions of antibodies and cytokines. For this reason, fusion proteins were constructed using IgG4 isotypes, which show little ADCC / CDC function compared to IgG1 (Gillies SD et al., 1999).</p><p> FIG. 10 shows the IL-12 recombinant protein of the present invention, a monovalent IL-12 fusion protein (monovalent IL-12-Fc) obtained using γ4-A107, and wild-type Fc. The schematic diagram of the obtained divalent IL-12 fusion protein (divalent IL-12-Fc) is shown. In particular, FIG. 10 (C) shows the fusion protein constructed by introducing the CH3 variant pair in the present invention. Human IL-12 (hIL-12, Uniprot entry names P29460, P29459; SEQ ID NOs: 17-18) and mouse IL-12 (mIL-12, Uniprot entry names P43432, P43431; Each DNA sequence of SEQ ID NOs: 19-20) was amplified and each amplification product was converted to a γ4-A107 variant by using a NotI / BsiWI restriction enzyme as shown in FIGS. 11 (A) and 11 (B). It was cloned in-frame into the containing animal cell expression vector. The resulting proteins were named monovalent hIL-12-Fc and monovalent mIL-12-Fc, respectively. In particular, a flexible peptide linker consisting of 15 amino acids was added between the p35 subunit and the hinge domain so that the human / mouse p35 subunit could fully interact with the p40 subunit (flexibility (G).₄S)₃Linker). As a comparative example of the proteins shown in FIG. 10 (C), human IL-12 (hIL-12) and mouse IL-12 (mIL-12) were each fused to wild-type IgG4 Fc (wt IgG4). Valuable hIL-12-Fc and divalent mIL-12-Fc were constructed. To fuse a single Fc to IL-12, which is active only in the heterodimer form, the two subunits of IL-12 were linked together by a 15 amino acid peptide linker and NotI / BsiWI restriction as shown in FIG. It was then cloned in-frame into an animal cell expression vector containing the γ4-A107 variant by the use of an enzyme. A comparative example is the fusion protein used in previous studies to make the IL-12 fusion protein (Lisan S. Peng et al., 1999).</p><p> Table 3 below shows the amino acid sequences for the mature form of the human and mouse IL-12 subunits used to construct the fusion protein.<tables num="3-1"><img file="JP2019536734A_D0004.tif" /></tables><tables num="3-2"><img file="JP2019536734A_D0005.tif" /></tables></p><p> (Example 8: Expression / purification of IL-12 fusion protein) The monovalent IL-12-Fc fusion protein of FIG. 10 (C) was prepared by the method described in Example 5 for human / mouse IL-12.p40-. It was expressed / purified from γ4-A107A and human / mouse IL-12.p35-γ4-A107B expression vectors (1: 1 ratio). The divalent IL-12-Fc fusion protein of FIG. 10 (B) was expressed / purified through a single transfection of a human / mouse scIL-12-IgG4 Fc (wt) expression vector. All fusion proteins were expressed / purified in an amount of 12-13 mg per 100 ml HEK293F cell culture.</p><p> 5 μg of each of the purified monovalent IL-12-Fc and divalent IL-12-Fc fusion proteins were analyzed by 12% SDS-PAGE under non-reducing conditions (Fig. 13). The monomer of the IL-12.p40-CH3A variant was observed at 60 kD; the homodimer of the IL-12.p40-CH3A variant was observed at 120 kD; the monomer of the IL-12.p35-CH3B variant was observed at 50 kD. The homodimer of the IL-12.p35-CH3B variant was observed at 100 kD; the heterodimer of the IL-12.p40-CH3A variant and the IL-12.p35-CH3B variant was observed at 110 kD. It was. However, for proteins obtained by linking human and mouse interleukin subunits, bands of slightly different sizes were observed, and it has been found in the literature that these bands result from different glycosylation patterns. (Lo et al., 2007). In addition, monomers were observed in all IgG4-based IL-12 fusion proteins, similar to the description in Example 6 above. A monovalent IL-12-Fc fusion protein obtained using a heterodimer Fc variant, similar to previous reports that the p35 subunit is not naturally expressed in monomeric form without the help of the p40 subunit. In, only the p40 subunit linked CH3A monomer was observed (Gillies et al., 1998 b).</p><p> FIG. 14 shows the results of analyzing the fusion protein by size exclusion chromatography (SEC). Oligomers were partially observed from the monovalent hIL-12-Fc fusion protein.</p><p> (Example 9: Evaluation of binding affinity of monovalent hIL-12-Fc fusion protein to IL-12 receptor) IL-12 receptor of monovalent hIL-12-Fc expressed and purified in Example 8 The binding affinity to the body was compared and analyzed with that of divalent hIL-12-Fc.</p><p> FIG. 15 shows FACS performed to determine that the constructed monovalent hIL-12-Fc may exhibit binding affinity for the IL-12 receptor as compared to divalent hIL-12-Fc. -Shows the results of Calibur (BD Biosciences) analysis.</p><p> Specifically, 5 ml of Ficoll (GE Healthcare) was placed in a 15 ml test tube to isolate immune cells (PBMC) from human peripheral blood. The collected blood is mixed 1: 1 with PBS (pH 7.4) and shaken, then 10 ml of blood is taken and Ficoll at 750 g for 20 minutes "without brake" to avoid mixing with Ficoll. Centrifugated in a containing test tube. Next, the buffy coat formed on Ficoll was collected and washed twice with PBS (pH 7.4) to obtain PBMC containing T cells, B cells, NK cells and monocytes. The isolated normal PBMC did not express IL-12R in such a large amount that IL-12 binding could be observed. For this reason, cells were stimulated for 72 hours by treatment with mitogen PHA (Sigma-Aldrich) so that T and NK cells could be activated. It has been reported that when cells are treated with PHA, IL-12 receptors are expressed on T and NK cells during immune cell division. PBMC 1x10⁶Add to RPMI1640 medium containing 10% FBS at a cell / ml density, add mitogen PHA to it at a concentration of 10 μg / ml, then cell at 37 ° C for 72 hours, 5% CO.₂Incubated in an incubator. Normal PBMC and PHA-activated PBMC were washed with cold PBS (pH 7.4) and 5 × 10 per sample.⁵Cells were prepared. Each of Fc (A107), divalent hIL-12-Fc and monovalent hIL-12-Fc was added to each sample at a concentration of 1 μM and incubated at 4 ° C for 30 minutes, then cold PBS (pH 7. It was washed in 4). Each sample was incubated with FITC-conjugated human anti-IgG4 secondary antibody (Sigma-Aldrich) at 4 ° C for 30 minutes, washed with PBS (pH 7.4), and then by flow cytometry (FACS Calibur, BD Bioscience). analyzed. After the analysis, a histogram graph of each sample was obtained to evaluate the binding affinity of monovalent hIL-12-Fc for the IL-12 receptor.</p><p> The results of the analysis show that divalent hIL-12-Fc and monovalent hIL-12-Fc do not bind to normal PBMCs that do not express IL-12 receptors, but PHA-activated PBMCs that express IL-12 receptors. It was shown that it was bound only to. Therefore, it was found that the binding affinity of monovalent hIL-12-Fc to the IL-12 receptor is equal to that of divalent hIL-12-Fc.</p><p> (Example 10: Evaluation of the ability of the monovalent hIL-12-Fc fusion protein to induce PBMC proliferation) The IL-12 portion of the IL-12 fusion protein is actually recombined by its binding to the IL-12 receptor. Whether or not it retains the same physiological activity as that of IL-12 (rIL-12) was investigated using recombinant human IL-12 (rhIL-12, Thermo Fisher Scientific) as a control.</p><p> FIG. 16 shows the WST-1 cell proliferation assay performed to examine the cell proliferation ability of Fc (A107), rhIL-12, divalent hIL-12-Fc and monovalent hIL-12-Fc in PHA-activated PBMC. The result of is shown.</p><p> Specifically, PBMCs (2 × 10) activated by PHA in the same manner as described in Example 9.⁴Cells, 50 μl) were added to 96-well plates (SPL, Korea) and subsequently diluted serially in RPMI1640 medium containing 10% FBS, 50-0.4 pM Fc (A107), rhIL-12, divalent hIL-12-Fc. And 50 μl of each of the monovalent hIL-12-Fc was added. Next, 5% CO₂The cells were cultured under 37 ° C for 72 hours. For the cell proliferation assay, 10 μl WST-1 (water-soluble tetrazolium salt, Sigma-aldrich) reagent was then added to each well, incubated at 37 ° C for 4 hours, and used with a microplate reader (Molecular Devices). The absorbance at 570 nm was measured.</p><p> As a result, it was shown that monovalent hIL-12-Fc has a PBMC proliferative capacity similar to or higher than that of rhIL-12.</p><p> (Example 11: Evaluation of ability of monovalent hIL-12-Fc fusion protein to induce IFN-γ secretion from PBMC) Fig. 17 shows Fc (A107), rhIL-12, divalent hIL-12-Fc and The results of an ELISA performed to measure the amount of IFN-γ secreted from PHA-activated PBMC by monovalent hIL-12-Fc are shown.</p><p> Specifically, in order to measure the concentration of IFN-γ in the culture supernatant cultured for 72 hours in Example 10, a 96-well plate (Thermo Fisher Scientific, Korea) for ELISA was used for human IFN-γ. It was coated with a capture antibody (Thermo Fisher Scientific) for 12 hours, washed with PBST, and then blocked with 1% BSA (PBS with 1% bovine serum albumin) for 1 hour at room temperature. After washing with PBST (PBS with 0.1% Tween-20), the culture supernatant obtained in Example 2 was diluted 5-fold with 1% BSA, 100 μl of dilution was added to each well, and 2 at room temperature. Incubated for hours. After washing with PBST, each well was incubated with biotin-conjugated IFN-γ detection antibody (Thermo Fisher Scientific) for 1 hour at room temperature. After washing with PBST (PBS with 0.1% Tween-20), each well is avidin-conjugated horseradish peroxidase (HRP) (Thermo Fisher). Incubate with Scientific) for 30 minutes at room temperature, wash with PBST (PBS with 0.1% Tween-20), then 3,3', 5,5'-tetramethylbenzidine substrate (TMB, sigma-aldrich) Processed with. Absorbance at 405 nm was measured using a microplate reader.</p><p> The results showed that the ability of monovalent hIL-12-Fc to induce IFN-γ secretion from PBMCs was similar to or higher than that of rhIL-12.</p><p> (Example 12: Evaluation of binding affinity of monovalent mIL-12-Fc to IL-12 receptor) To IL-12 receptor of monovalent mIL-12-Fc expressed / purified in Example 8 The binding affinity of the divalent mIL-12-Fc was compared and analyzed.</p><p> Figure 18 shows the flow cytometry performed to determine that the constructed monovalent mIL-12-Fc exhibits binding affinity to the IL-12 receptor compared to divalent mIL-12-Fc. The result of metric is shown.</p><p> Specifically, it was reported that mouse IL-12 binds not only to the mouse IL-12 receptor but also to the human IL-12 receptor. Therefore, the analysis was performed in the same manner as described in Example 9. The results of the analysis show that divalent mIL-12-Fc and monovalent mIL-12-Fc do not bind to normal PBMCs that do not express IL-12 receptors, but PHA-activated PBMCs that express IL-12 receptors. It was shown that it was bound only to. Therefore, it was shown that the binding affinity of monovalent mIL-12-Fc to the IL-12 receptor is the same as that of divalent mIL-12-Fc.</p><p> (Example 13: Evaluation of monovalent mIL-12-Fc ability to induce PBMC proliferation) Fig. 19 shows Fc (A107), recombinant mouse IL-12 (rmIL-) on cell proliferation of PHA-activated PBMC. 12), the results of the WST-1 cell proliferation assay performed to investigate the effects of divalent mIL-12-Fc and monovalent mIL-12-Fc capacity are shown.</p><p> Specifically, PBMCs (2 × 10) activated by PHA in the same manner as described in Example 9.⁴Cells, 50 μl) were added to 96-well plates and subsequently diluted serially in RPMI1640 medium containing 10% FBS, 50-0.4 pM Fc (A107), rmIL-12, divalent mIL-12-Fc and monovalent mIL-. 50 μl of each of 12-Fc was added. Next, 5% CO₂The cells were cultured at 37 ° C for 72 hours under the conditions and then the WST assay was performed in the same manner as described in Example 10. As a result, it was shown that monovalent mIL-12-Fc has the same ability to induce PBMC proliferation as rmIL-12.</p><p> (Example 14: Evaluation of the ability of monovalent mIL-12-Fc to inhibit in vivo tumor growth) In Example 13, the ability of monovalent mIL-12-Fc to induce the growth of PHA-activated PBMC was evaluated. .. We investigated whether the same effect of monovalent mIL-12-Fc also appears in vivo.</p><p> Figures 20 (A) and 20 (B) show 100 mm of live mice.³The results of measuring the tumor growth inhibitory activity of monovalent mIL-12-Fc on tumors are shown.</p><p> Specifically, CT26 was a 4-week-old female Balb / c mouse (NARA Biotech, Korea) shaved and diluted in 150 μL PBS.^{HER2 / Neu}Colorectal cancer cells (1 x 10)⁶Cells / mice) were subcutaneously transplanted into mice. Similar tumor volume (average volume: 100-120 mm)³) Was randomly grouped, and 1 μg of each of Fc (A107), rmIL-12) (Thermo Fisher Scientific), divalent mIL-12-Fc and monovalent mIL-12-Fc was placed in the abdominal cavity of each mouse. A total of 6 injections (twice a week) were given at a dose corresponding to the equimolar amount of IL-12. Tumors were measured twice a week and tumor volume (V) was calculated using the following formula: V = length x width²/2。 </p><p> As shown in FIG. 20 (A), administration of 1 μg of rmIL-12 did not affect the inhibition of tumor growth compared to controls, but equimolar concentrations of monovalent mIL-12-Fc and divalent. mIL-12-Fc inhibited tumor growth. Furthermore, as shown in FIG. 20 (B), administration of monovalent mIL-12-Fc and divalent mIL-12-Fc showed little or no change in mouse body weight compared to controls. , Monovalent mIL-12-Fc and divalent mIL-12-Fc were shown to be non-toxic.</p><p> Figures 21 (A), 21 (B) and 21 (C) show 300 mm of live mice.³The results of measuring the tumor growth inhibitory activity of various concentrations of monovalent mIL-12-Fc on tumors are shown.</p><p> Specifically, CT26 was a 4-week-old female Balb / c mouse (NARA Biotech, Korea) shaved and diluted in 150 μL PBS.^{HER2 / Neu}Colorectal cancer cells (1 x 10)⁶Cells / mice) were subcutaneously transplanted into mice. Similar tumor volume (average volume: 300 mm³) Are randomly grouped, and each of the divalent mIL-12-Fc and monovalent mIL-12-Fc is placed intraperitoneally in each mouse at a concentration of 0.1 to 2 μg of rmIL-12 and an equimolar concentration 6 times in total. Injected (twice a week). Tumors were measured twice a week and tumor volume (V) was calculated using the following formula: V = length x width²/2。 </p><p> As shown in FIGS. 21 (A), 21 (B) and 21 (C), at a dose corresponding to an equimolar amount of 1 μg of IL-12 or less, monovalent mIL-12-Fc is divalent IL. Compared with -12-Fc, it showed a high effect of inhibiting the growth of large tumors. At a concentration corresponding to an equimolar amount of 0.25 μg of IL-12, divalent mIL-12-Fc showed an inhibitory effect on tumor growth, but did not remove the tumor. However, under the same dosing regimen, monovalent mIL-12-Fc showed a tumor-removing effect in 40% of mice. In addition, at a concentration corresponding to the equimolar amount of 0.5 μg IL-12 in which divalent mIL-12-Fc failed to remove the tumor, monovalent mIL-12-Fc was administered in mice even when administered only 5 times. Tumors were removed in 73%.</p><p> (Example 15: Evaluation of in vivo toxicity of monovalent mIL-12-Fc) Figure 21 (D) shows body weight changes to determine in vivo toxicity of monovalent mIL-12-Fc administered at various concentrations. The result of measuring.</p><p> Specifically, as shown in FIG. 21 (A), whether or not the body weight was reduced was observed by measuring the body weight of the mice to which the monovalent mIL-12-Fc was administered twice a week. Mice gained weight with increasing tumor volume in the control group, but mice treated at all concentrations of divalent mIL-12-Fc and monovalent mIL-12-Fc lost weight compared to before administration. It was shown that it was not shown. Therefore, it was determined that monovalent mIL-12-Fc did not induce weight loss and therefore did not have significant in vivo toxicity.</p><p> FIG. 21 (D) shows the results of measuring alanine aminotransferase (ALT), which is a hepatotoxic marker.</p><p> Specifically, blood was collected from the facial vein of the mouse shown in FIG. 21 (A) 24 hours after the last administration. Blood was allowed to stand at room temperature for 2 hours to induce blood coagulation, then centrifuged at 8000 rpm for 10 minutes, and supernatant serum was collected. Blood was drawn from the facial veins of mice 24 hours after the last dose of IL-12-Fc fusion protein to measure serum ALT levels. Blood was allowed to stand at room temperature for 2 hours to induce blood coagulation, then centrifuged at 8000 rpm for 10 minutes, and supernatant serum was collected. To measure the concentration of ALT in serum, take a substrate solution (mixture of alanine and α-ketoglutarate) for ALT measurement in a 15 ml test tube and incubate for 5 minutes in a constant temperature water bath at 37 ° C. did. Serum isolated from the blood of tumor-transplanted mice treated with each of divalent mIL-12-Fc and monovalent mIL-12-Fc was diluted 10-fold, 200 μl of dilution was added to the substrate solution, and shaken. Incubated for 30 minutes in a constant temperature water bath at 37 ° C. 1 ml of a coloring reagent (2,4-dinitrophenyl-1-hydrazone) was added to a test tube taken out from a water bath at a constant temperature, and the test tube was allowed to stand at room temperature for 20 minutes. Next, 10 ml of 0.4N sodium hydroxide solution was added to the test tube and mixed, and then the test tube was allowed to stand at room temperature for 10 minutes. Absorbance at 505 nm was measured using a photoelectric spectrophotometer (GeneQuant100, GE Healthcare). The standard curve created by adding the standard curve reagent instead of serum was used to convert the ALT to units. Blood-derived sera collected from mice treated with divalent mIL-12-Fc or monovalent mIL-12-Fc are ALTs similar to those of sera isolated from blood samples of control or normal Balb / c mice. It was shown to show activity. This suggests that when divalent mIL-12-Fc or monovalent mIL-12-Fc is administered to tumor-transplanted mice at an equimolar concentration of 0.5 μg or 1 μg of IL-12, it does not induce hepatotoxicity. To do.</p><p> (Example 16: Evaluation of the ability of monovalent mIL-12-Fc to induce immune cell proliferation in vivo) As shown in Example 15, divalent mIL-12-Fc and monovalent mIL-12-Fc When administered at concentrations corresponding to an equimolar amount of 2 μg of IL-12, divalent mIL-12-Fc and monovalent mIL-12-Fc all removed tumors, but they were more than 1 μg of IL-12. When administered at low molar concentrations, the tumor growth inhibitory effect of monovalent mIL-12-Fc was significantly higher than that of divalent mIL-12-Fc. In fact, the high tumor growth inhibitory effect of monovalent mIL-12-Fc is the IL-12 receptor-bearing NK cells, CD4.⁺T cells and CD8⁺Analysis was performed to determine if it was associated with an increase in the number of unique effector cells such as T cells.</p><p> FIG. 22 (A) shows CD4 in the spleen of mice sacrificed 3 days after the last dose of FIG. 21 (A).⁺T cells, CD8⁺The result of measuring the increase in the number of T cells and NK cells is shown.</p><p> Specifically, after the procedure shown in FIG. 21 (A), 34 days after tumor transplantation, the mouse spleen was dissected, crushed using a wide mesh in a Petri dish, and then 10 ml 2 Washed with% FBS-containing medium. Next, 1 ml of erythrocyte lysis buffer was added to lyse the erythrocytes, and the resulting cells were washed with PBS to prepare a splenocyte suspension, and the number of cells was counted with a blood cell counter. Add APC, FITC, PE or PE-cy5 conjugated anti-CD45, anti-CD3, anti-CD4, anti-CD8 and anti-CD49b antibodies to spleen lymphocytes, which are then stained at 4 ° C for 30 minutes and cold PBS (pH 7). It was washed with .4) and then analyzed by flow cytometry (FACS Calibur, BD Bioscience) and Flow jo (Thermo Fisher Scientific). Each sample was analyzed by dot plot and CD45⁺CD3⁺CD4⁺Cell population, CD45⁺CD3⁺CD8⁺Cell population and CD45⁺CD3^-CD49b⁺Each cell population is CD4⁺T cells, CD8⁺Defined as T cells and NK cells, calculate their ratio to total splenocytes, multiply by the number of cells counted by the blood cell counter, and increase CD4 after administration of monovalent mIL-12-Fc.⁺T cells, CD8⁺The numbers of T cells and NK cells were analyzed.</p><p> As a result, monovalent mIL-12-Fc was CD4 in tumor-transplanted mice compared to controls.⁺T cells and CD8⁺It was found that the number of T cells was increased in a concentration-dependent manner. However, divalent mIL-12-Fc was CD8 only in the group receiving 0.5 μg of IL-12 at a concentration corresponding to the equimolar amount.⁺In the group that increased the number of T cells and administered it at a concentration corresponding to an equimolar amount of 1 μg IL-12, it was CD4.⁺T cells and CD8⁺It did not increase the number of T cells. Consistent with previous studies (Cerwenka and Lanier, 2016; Schreiber et al., 2011) that NK cells do not form memory cells in tumor-transplanted mice, monovalent mIL-12-Fc 34 days after tumor transplantation And it was observed that the number of NK cells in the group treated with divalent mIL-12-Fc was similar to that in the control group. As a result, the monovalent mIL-12-Fc was CD4.⁺T cells and CD8⁺It has been shown to cause greater expansion of T cells, which explains a stronger inhibition of tumor growth compared to divalent mIL-12-Fc.</p><p> Adaptive immune cells infiltrating the tumor (CD4)⁺T cells and CD8⁺Monovalent mIL-12-Fc increases the number of adaptive immune cells infiltrating tumors, based on a report (Schreiber et al., 2011) that increasing the number of T cells) is important in inhibiting tumor growth. Analyzed whether or not. When monovalent mIL-12-Fc was administered 6 times, there were many mice without tumors. For this reason, monovalent mIL-12-Fc was administered three times and then the number of immune cells infiltrating mouse tumors was analyzed.</p><p> FIG. 22 (B) shows total immune cells, CD4, infiltrating the tumor of the sacrificed mouse 3 days after the third dose of FIG. 21 (A).⁺T cells and CD8⁺The result of measuring the number of T cells is shown.</p><p> Specifically, after the procedure shown in FIG. 21 (A), mouse tumors were dissected and weighed 24 days after tumor transplantation. Tumors were then disrupted using wire mesh and collagenase (100 μg / ml) in Petri dishes and centrifuged at 50 g for 5 minutes in 10 ml of 2% FBS-containing medium to remove parenchymal tissue. Next, 1 ml of erythrocyte lysis buffer was added to lyse the erythrocytes, and the resulting cells were washed with PBS to prepare a cell suspension, and the number of cells was counted with a hemocytometer. APC, FITC or PE-cy5 conjugated anti-CD45, anti-CD3, anti-CD4 and anti-CD8 antibodies were added to cells isolated from the tumor and then stained at 4 ° C for 30 minutes and cold PBS (pH 7. It was washed with 4) and then analyzed by flow cytometry (FACS Calibur, BD Bioscience) and Flow jo (Thermo Fisher Scientific). Each sample was analyzed by dot plot, then CD45⁺Cell population, CD45⁺CD3⁺CD4⁺Cell population and CD45⁺CD3⁺CD8⁺Cell population and CD45⁺CD3^-CD49b⁺Cell populations, total tumor infiltrating immune cells, tumor infiltrating CD4, respectively⁺T cell and tumor infiltration CD8⁺Defined as a T cell. Calculate the ratio of these cells to cells isolated from all tumors, multiply by the number of cells counted by the blood cell counter, and then increase total tumor infiltrating immune cells after administration of monovalent mIL-12-Fc. , Tumor infiltration CD4⁺T cell and tumor infiltration CD8⁺The number of T cells was analyzed.</p><p> As a result, as compared with the control, divalent mIL-12-Fc and monovalent mIL-12-Fc were infiltrated into the tumor, total immune cells, CD4.⁺T cells and CD8⁺It was found that the number of T cells was increased in a concentration-dependent manner. At equimolar concentrations, monovalent mIL-12-Fc is a tumor-infiltrated total immune cell, CD4, compared to divalent mIL-12-Fc.⁺T cells and CD8⁺Significantly increased T cells. As a result, monovalent mIL-12-Fc was CD4 in the tumor.⁺T cells and CD8⁺It has been shown to cause greater infiltration of T cells, which explains a stronger inhibition of tumor growth compared to divalent mIL-12-Fc.</p><p> (Example 17: Evaluation of the effect of monovalent mIL-12-Fc on cytokine secretion and increased cytotoxicity from immune cells in vivo) IL-12 is an IFN-γ from T cells and NK cells. It is known to inhibit the growth of cancer cells by increasing secretion (Trinchieri, 2003). In addition, IL-12 exhibits anti-cancer effects by enhancing the direct cytotoxic effects of cytotoxic T cells and natural killer cells on cancer cells. Therefore, the high anticancer activity of monovalent IL-12-Fc increases the serum IFN-γ concentration in tumor-transplanted mice, as well as the direct cells of cytotoxic T cells and natural killer cells against cancer cells. An analysis was performed to determine if it was due to an enhanced injurious effect.</p><p> FIG. 23 (A) shows the results of an ELISA performed to measure the concentration of IFN-γ in serum isolated from blood collected from the facial veins of mice 24 hours after the final administration of FIG. 21 (A).</p><p> Specifically, blood was collected from the facial veins of mice 24 hours after the final administration of the mIL-12-Fc fusion protein shown in FIG. 20 (A). Blood was allowed to stand at room temperature for 2 hours to induce blood coagulation, then centrifuged at 8000 rpm for 10 minutes, and supernatant serum was collected. To measure the concentration of IFN-γ in serum, a 96-well plate for ELISA (Thermo Fisher Scientific) was coated with mouse IFN-γ capture antibody for 12 hours and PBS with PBST (0.1% Tween-20). ), Then blocked with 1% BSA (PBS with 1% bovine serum albumin) at room temperature for 1 hour. After washing with PBST (PBS with 0.1% Tween-20), serum was diluted 10-fold with 1% BSA and incubated at room temperature for 2 hours. After washing with PBST (PBS with 0.1% Tween-20), each well was subjected to biotin-conjugated mouse IFN-γ detection antibody (Thermo Fisher). Scientific) and room temperature for 1 hour. After washing with PBST (PBS with 0.1% Tween-20), each well is avidin-conjugated horseradish peroxidase (HRP) (Thermo Fisher). Incubate with Scientific) for 30 minutes at room temperature, wash with PBST (PBS with 0.1% Tween-20), then treat with 3,3', 5,5'-tetramethylbenzidine substrate (TMB, sigma-aldrich). did. Absorbance at 450 nm was measured using a microplate reader. As shown in FIG. 23 (A), the serum IFN-γ concentration of the mice treated with divalent mIL-12-Fc did not increase as compared with that of the control group. However, increased serum IFN-γ levels were observed in mice treated with monovalent mIL12-Fc proportional to the dose of 1 mg rmIL12 up to an equimolar amount compared to that of the control group. .. In addition, monovalent mIL-12-Fc tumors have increased the secretion of IFN-γ, which is known to have an inhibitory effect on the growth of some cancer cells. It was shown that there was a formation inhibitory effect.</p><p> Serum levels of IFN-γ were lower in tumor-transplanted mice treated with divalent mIL12-Fc (Fig. 23 (A)). Therefore, to determine if divalent mIL-12-Fc has a low ability to induce IFN-γ secretion from NK and T cells, monovalent mIL-12-Fc and divalent mIL-12- Serum IFN-γ levels were measured at the indicated time points after a single dose of Fc.</p><p> Figure 23 (B) shows CT26^{HER2 / Neu}IFN in serum at various time points after a single intraperitoneal administration of divalent mIL-12-Fc and monovalent mIL-12-Fc to Balb / c mice transplanted with colorectal cancer cells The results of the ELISA performed to measure the concentration of -γ are shown.</p><p> Specifically, CT26^{HER2 / Neu}Tumor volume of Balb / c mice transplanted with colorectal cancer cells is 300 mm³Divalent mIL-12-Fc and monovalent mIL-12-Fc were intraperitoneally administered at a concentration equivalent to 1 μg of rmIL-12. After 1, 3 and 5 days, blood was drawn from the facial veins of the mice. Blood was allowed to stand at room temperature for 2 hours to induce blood coagulation, centrifuged at 8000 rpm for 10 minutes, and supernatant serum was collected. To measure the concentration of IFN-γ in serum, a 96-well plate for ELISA (Thermo Fisher Scientific) was coated with mouse IFN-γ capture antibody for 12 hours and PBS with PBST (0.1% Tween-20). ), Then blocked with 1% BSA (PBS with 1% bovine serum albumin) at room temperature for 1 hour. After washing with PBST (PBS with 0.1% Tween-20), serum was diluted 10-fold with 1% BSA and incubated at room temperature for 2 hours. After washing with PBST (PBS with 0.1% Tween-20), each well was subjected to biotin-conjugated mouse IFN-γ detection antibody (Thermo Fisher). Scientific) and room temperature for 1 hour. After washing with PBST (PBS with 0.1% Tween-20), each well is incubated with avidin-conjugated horseradish peroxidase (HRP) (Thermo Fisher Scientific) for 30 minutes at room temperature to have PBST (0.1% Tween-20). It was washed with PBS) and then treated with 3,3', 5,5'-tetramethylbenzidine substrate (TMB, sigma-aldrich). Absorbance at 450 nm was measured using a microplate reader. As shown in FIG. 23 (B), in the tumor-transplanted mice, the group to which the divalent mIL-12-Fc was administered had a serum IFN-γ similar to that of the monovalent mIL-12-Fc group until the 5th day. The concentrations showed that divalent mIL-12-Fc did not have a defect inherent in its ability to induce the secretion of IFN-γ from effector cells.</p><p> Figure 23 (C) shows CT26^{HER2 / Neu}It is a graph which shows the result of having measured the cytotoxic effect of the cytotoxic T cell isolated from the spleen of the mouse which was sacrificed 3 days after the last administration of FIG. 21 (A) with respect to the cancer cell.</p><p> Specifically, 72 hours after the final administration of the cytokine of FIG. 21 (A), mice were sacrificed, the spleen dissected from it and crushed in a 60 mm dish containing 70 micron mesh and PBS. Erythrocytes were lysed by adding erythrocyte lysis buffer to the cells obtained by centrifugation. The cells were then washed with PBS and incubated with APC-conjugated anti-CD3 antibody (Thermo Fisher Scientific) and PE-conjugated anti-CD8 antibody at 4 ° C for 30 minutes. After washing the cells with PBS, use FACS Aria III (BD biosciences, Korea) to use cytotoxic T cells (CD3).⁺CD8⁺) Was isolated. Target CT26^{HER2 / Neu}CT26 to measure the cytotoxic effects of cytotoxic T cells on cancer cells^{HER2 / Neu}Cancer cells were stained with calcein AM (Thermo Fisher Scientific Inc., 10 μM). CT26^{HER2 / Neu}Cancer cells (2 x 10)⁶) Is suspended in 2 ml DPBS, mixed with 2 μl calcein AM (10 mM), then 5% CO₂Incubated for 45 minutes at 37 ° C below. After washing with 10 ml of RPMI 1640 containing 10% FBS, cells were 2 × 10 per well.⁴In addition to each well of the 96-well plate at cell density, cytotoxic T cells (1 x 10)⁵/ 100 μl / well) is added to each well and 5% CO₂Incubated at 37 ° C below for 4 hours. CT26 showing green fluorescence^{HER2 / Neu}Live cancer cells and CT26 showing no green fluorescence^{HER2 / Neu}Cancer-dead cells were analyzed by flow cytometry, and the cytotoxic effects of cytotoxic T cells were expressed as a percentage. Tumors treated with divalent mIL-12-Fc Tumors treated with monovalent mIL-12-Fc compared to cytotoxic T cells isolated from transplanted mice or cytotoxic T cells isolated from controls Cytotoxic T cells isolated from transplanted mice are targeted by CT26^{HER2 / Neu}It was shown to have a higher cytotoxic effect on cancer cells. Furthermore, it was shown that the tumorigenicity inhibitory effect of monovalent mIL-12-Fc was attributed to the direct cytotoxic effect of some cytotoxic T cells on cancer cells.</p><p> FIG. 23 (D) shows tumors to determine whether the cytotoxic effects of cytotoxic T cells enhanced by administration of monovalent IL-12-Fc to tumor-transplanted mice are tumor antigen-specific. CT26 expressing an antigen^{HER2 / Neu}The cytotoxic effects of cytotoxic T cells isolated from the spleen of mice slaughtered 3 days after the third dose in Figure 21 (A) using cancer cells and 4T1 cells that do not express tumor antigens. The measurement result is shown.</p><p> Specifically, 72 hours after the third dose of monovalent IL-12-Fc in Figure 20 (A), mice were sacrificed, the spleen dissected from it, and a 60 mm dish containing 70 micron mesh and PBS. Crushed in. Target CT26^{HER2 / Neu}CT26 by the method used in FIG. 21 (C) to measure the cytotoxic effects of cytotoxic T cells on cancer cells and non-target 4T1 cells.^{HER2 / Neu}Cancer cells and 4T1 cancer cells were stained with calcein AM (Thermo Fisher Scientific Inc., 10 μM). After 3 washes with RPMI 1640 containing 10 ml of 10% FBS, cells were placed 2 x 10 per well.⁴In addition to each well of the 96-well plate at cell density, cytotoxic T cells (1 x 10)⁵/ 100 μl / well) is added to each well and 5% CO₂Incubated for 4 hours in a 37 ° C incubator below. CT26 showing green fluorescence^{HER2 / Neu}Live cancer cells and CT26 showing no green fluorescence^{HER2 / Neu}Dead cancer cells or 4T1 cancer cells were analyzed by flow cytometry, and the cytotoxic effects of cytotoxic T cells were expressed as a percentage. As a result, it was shown that the cytotoxic effect of cytotoxic T cells enhanced by the administration of monovalent mIL-12-Fc is target cell-specific.</p><p> Figure 23 (E) shows CT26^{HER2 / Neu}The results of measuring the cytotoxic effect of natural killer cells isolated from the spleen of mice sacrificed 3 days after the third administration of FIG. 21 (A) on cancer cells are shown.</p><p> Specifically, three days after the third dose of cytokine in FIG. 21 (A), mice were sacrificed, the spleen dissected from it and crushed in a 70 mm dish containing 70 micron mesh and PBS. Erythrocytes were lysed by adding erythrocyte lysis buffer to the cells obtained by centrifugation. The cells were then washed with PBS and incubated with APC-conjugated anti-CD3 antibody (Thermo Fisher Scientific) and PE-conjugated anti-CD49b antibody at 4 ° C for 30 minutes. After washing the cells with PBS, use FACS Aria III (BD biosciences, Korea) to natural killer cells (CD3).^-CD49b⁺) Was isolated. Target CT26^{HER2 / Neu}CT26 to measure the cytotoxic effects of natural killer cells on cancer cells^{HER2 / Neu}Cancer cells were stained with calcein AM (Thermo Fisher Scientific Inc., 10 μM). CT26^{HER2 / Neu}Cancer cells (2 x 10)⁶) Is suspended in 2 ml DPBS, mixed with 2 μl calcein AM (10 mM), then 5% CO₂Incubated for 45 minutes at 37 ° C below. After washing with 10 ml of RPMI 1640 containing 10% FBS, cells were 2 × 10 per well.⁴Natural killer cells (1x10) in addition to each well of the 96-well plate at cell density⁵/ 100 μl / well) is added to each well and 5% CO₂Incubated at 37 ° C below for 4 hours. CT26 showing green fluorescence^{HER2 / Neu}Live cancer cells and CT26 showing no green fluorescence^{HER2 / Neu}Dead cancer cells were analyzed by flow cytometry and the cytotoxic effects of natural killer cells were expressed as a percentage. Tumor-transplanted mice treated with monovalent mIL-12-Fc compared to natural killer cells isolated from tumor-transplanted mice treated with divalent mIL-12-Fc or cytotoxic T cells isolated from the control group. Natural killer cells isolated from target CT26^{HER2 / Neu}It was shown to have a higher cytotoxic effect on cancer cells. Furthermore, it was shown that the tumorigenicity inhibitory effect of monovalent mIL-12-Fc was attributed to the direct cytotoxic effect of some natural killer cells on cancer cells.</p><p> (Example 18: Effector CD8 in vivo⁺T cells and memory CD8⁺Evaluation of the ability of monovalent mIL-12-Fc to form T cells) The generation of adaptive immunity in tumor-transplanted mice is the effector memory CD8.⁺T cells and memory CD8⁺Evaluate by whether T cells are produced. Tumor removal effect of monovalent mIL-12-Fc, effector memory CD8⁺T cells and memory CD8⁺It was measured whether it was attributed to the production of T cells.</p><p> Figures 24 (A), 24 (B) and 24 (C) show effector CD8 produced when monovalent mIL-12-Fc was administered to tumor-bearing mice.⁺T cells, effector memory CD8⁺T cells and memory CD8⁺The result of measuring the number of T cells is shown.</p><p> Specifically, after the procedure shown in FIG. 21 (A), 34 days after tumor transplantation, the mouse spleen was dissected, crushed using a wire mesh in a Petri dish, and then 10 ml of 2% FBS-containing medium. Washed with. Next, 1 ml of erythrocyte lysis buffer was added to lyse the erythrocytes, and the resulting cells were washed with PBS to prepare a splenocyte suspension, and the number of cells was counted with a blood cell counter. APC, FITC, PE or PE-cy5 conjugated anti-CD3, anti-CD8, anti-CD62L and anti-IL-7 receptor (IL-7R) antibodies were added to splenocytes, which were then stained at 4 ° C for 30 minutes. , Washed with cold PBS (pH 7.4), then analyzed by flow cytometry (FACS Calibur, BD Bioscience) and Flow jo (Thermo Fisher Scientific). Analyze each sample by dot plot and CD3⁺CD8⁺CD62L^lowIL-7R^lowCell population, CD3⁺CD8⁺CD62L^lowIL-7R^hiCell population and CD3⁺CD8⁺CD62L^hiIL-7R^hiEach cell population, effector CD8⁺T cells, effector memory CD8⁺T cells and memory CD8⁺Defined as T cells, the ratio to the total splenocytes was calculated, multiplied by the number of cells counted by the blood cell counter, and the effector CD8 increased after administration of monovalent mIL-12-Fc.⁺T cells, effector memory CD8⁺T cells and memory CD8⁺The number of T cells was analyzed.</p><p> As a result, monovalent mIL-12-Fc was found in effector memory CD8 in tumor-transplanted mice compared to controls.⁺T cells and memory CD8⁺It was found that the number of T cells was increased in a concentration-dependent manner. However, divalent mIL-12-Fc was administered to the effector memory CD8 only in the group administered at a concentration corresponding to an equimolar amount of 0.5 μg of IL-12.⁺T cells and memory CD8⁺Effector memory CD8 in the group in which the number of T cells was increased and it was administered at a concentration corresponding to an equimolar amount of 1 μg of IL-12.⁺T cells and memory CD8⁺It did not increase the number of T cells. Therefore, the higher tumorigenicity inhibitory effect of monovalent mIL-12-Fc compared to divalent mIL-12-Fc is effector memory CD8.⁺T cells and memory CD8⁺It was found to be attributed to an increase in the number of T cells.</p><p> FIG. 24 (D) shows CT26 in surviving mice 120 days after administration of 1 μg of monovalent IL-12-Fc in FIG. 21 (A).^{HER2 / Neu}The results obtained by retransplanting cancer cells and measuring changes in tumor volume in mice are shown.</p><p> Specifically, 120 days after the final administration of 1 μg of monovalent IL-12-Fc to female Balb / c mice (NARA Biotech, Korea) in FIG. 21 (A), the surviving mice were shaved and 150 μL. CT26 diluted in PBS^{HER2 / Neu}Cells (1x10⁶Cells / mice) were subcutaneously transplanted into mice. Tumors were then measured twice a week without further administration of 1 μg of monovalent IL-12-Fc and tumor volume (V) was calculated using the following formula: V = length x width²/ 2. As a result, it was found that the tumors of the mice that survived the administration of 1 μg of monovalent mIL-12-Fc began to shrink from the 11th day as compared with the control group. Therefore, when monovalent mIL-12-Fc was administered to tumor-transplanted mice, it was the effector memory CD8.⁺T cells and memory CD8⁺It was found to have produced T cells, so it would be eliminated even when the tumor was re-transplanted into mice.</p><p> (Example 19: Memory precursor effector CD8 in vivo⁺Evaluation of the ability of monovalent mIL-12-Fc to form T cells) In Examples 16 and 18, CD8 in tumor-transplanted mice⁺T cells, effector memory CD8⁺T cells and central memory CD8⁺It was observed that the effect of divalent mIL-12-Fc on increasing the number of T cells was lower than that of monovalent mIL-12-Fc. Activated CD8⁺Effector CD8 after the effector stage where T cells destroy tumor cells directly⁺T cells become memory precursor effector cells (MPEC), then memory CD8⁺It was reported that it partially differentiated into T cells and most differentiated into short-lived effector cells (SLEC). Therefore, CD8 activated by administration of divalent mIL-12-Fc⁺Memory CD8 produced when T cells differentiate into short-lived effector cells⁺Analysis was performed to determine if the number of T cells was small and therefore they could not remove the tumor.</p><p> FIG. 24 (E) shows CD8 present in the spleen of mice sacrificed 3 days after the third dose of FIG. 21 (A).⁺Memory precursor effector cells in T cells (KLRG1)^-IL-7R⁺) And short-lived effector cells (KLRG1)⁺IL-7R^-The result of analyzing the ratio of) is shown.</p><p> Specifically, after the procedure shown in FIG. 21 (A), 24 days after tumor transplantation, the mouse spleen was dissected, crushed using a wire mesh in a Petri dish, and then 10 ml of 2% FBS-containing medium. Washed with. Next, 1 ml of erythrocyte lysis buffer was added to lyse the erythrocytes, and the resulting cells were washed with PBS to prepare a cell suspension. APC, FITC, PE or PE-cy5 conjugated anti-CD3, anti-CD8, anti-KLRG1 and anti-IL-7 receptor (IL-7R) antibodies are added to spleen cells and then stained at 4 ° C for 30 minutes. The cells were washed with cold PBS (pH 7.4) and then analyzed by flow cytometry (FACS Calibur, BD Bioscience) and Flow jo (Thermo Fisher Scientific). Analyze each sample by dot plot and CD3⁺CD8⁺KLRG1^-IL-7R⁺Cell population and CD3⁺CD8⁺KLRG1_⁺IL-7R^-Cell populations were defined as memory precursor effector cells and short-lived effector cells, respectively, and their proportions compared to total splenocytes were analyzed.</p><p> As a result, it was found that monovalent mIL-12-Fc increased the proportion of memory precursor effector cells in tumor-transplanted mice in a concentration-dependent manner as compared with the control. However, administration of divalent mIL-12-Fc did not increase the proportion of memory precursor effector cells compared to controls, but rather increased the number of short-lived effector cells. Therefore, compared to divalent mIL-12-Fc, monovalent mIL-12-Fc promotes the production of memory precursor effector cells, thereby effector memory CD8.⁺T cells and memory CD8⁺It was found to have significantly increased the number of T cells, indicating that it has a higher effect on tumor removal.</p><p> (Example 20: Evaluation of the effect of monovalent mIL-12-Fc on the expression of transcription factors involved in the induction of memory cell differentiation) CD8⁺CD8 when T cells are administered with high concentrations of IL-12, or when activated by frequent administration of IL-12 for 2 days or longer⁺Increased expression of the transcription factor T-bet, which allows T cells to differentiate into short-lived effector cells, CD8⁺It has been reported that the expression of the transcription factor eomesodermin (Eomes), which allows T cells to differentiate into memory precursor effector cells, is reduced. Therefore, CD8 in which monovalent mIL-12-Fc and divalent mIL-12-Fc differentiate into short-lived effector cells.⁺CD8 to change the proportion of T cells⁺Analysis was performed to determine whether to differentially regulate T-bet and Eomes expression in T cells.</p><p> Figures 25 (A) and 25 (B) show CD8 in the spleen of mice sacrificed 3 days after the third dose of Figure 21 (A).⁺T cells (high expression of T-bet that inhibits memory cell differentiation) and CD8⁺The results of a flow cytometric analysis performed to measure the proportion of T cells (which show low expression of Eomes, which promotes memory cell differentiation) are shown.</p><p> Specifically, after the procedure shown in FIG. 21 (A), 24 days after tumor transplantation, the mouse spleen was dissected, crushed using a wire mesh in a Petri dish, and then 10 ml of 2% FBS-containing medium. Washed with. Next, 1 ml of erythrocyte lysis buffer was added to lyse the erythrocytes, and the resulting cells were washed with PBS to prepare a cell suspension. Spleen cells were stained with PE-cy5 or FITC-conjugated anti-CD3 and anti-CD8 antibodies at 4 ° C for 30 minutes and washed with cold PBS (pH 7.4). The cells were then fixed and permeabilized with Foxp3 / Transcription Factor Staining Buffer Set (Thermo Fisher Scientific) (an intranuclear transcription factor staining reagent). The cells were then stained for 30 minutes at 4 ° C by PE or efluor660 conjugated anti T-bet or anti Eomes antibody, then flow cytometry on permeabilization buffer (FACS Calibur, BD Bioscience) and flow Saito Analyzed by Flow jo (Thermo Fisher Scientific) for metric data analysis. Analyze each sample by dot plot and CD3⁺CD8⁺T-bet^{high high}Cell population and CD3⁺CD8⁺Eomes⁺T-bet^lowThe proportion of cell population was analyzed. As a result, monovalent mIL-12-Fc was CD3 compared to the control.⁺CD8⁺T-bet^{high high}Decrease the proportion of cell population in a concentration-dependent manner, CD3⁺CD8⁺Eomes⁺T-bet^lowIt was found that the proportion of the cell population was increased in a concentration-dependent manner. However, divalent mIL-12-Fc was CD3 only in the group receiving 0.5 μg of IL-12 at a concentration corresponding to the equimolar amount.⁺CD8⁺T-bet^{high high}Reduce the proportion of cell population, CD3 in the group⁺CD8⁺Eomes⁺T-bet^lowIncreased proportion of cell population. Furthermore, in the group in which divalent mIL-12-Fc was administered at a concentration corresponding to an equimolar amount of 1 μg of IL-12, divalent mIL-12-Fc was CD3.⁺CD8⁺T-bet^{high high}CD3 also has the effect of reducing the proportion of cell population⁺CD8⁺Eomes⁺T-bet^lowIt also showed no effect on increasing the proportion of cell population. Therefore, compared to the divalent mIL-12-Fc, the monovalent mIL-12-Fc is an effector memory CD8.⁺T cells and memory CD8⁺CD3 to significantly increase the number of T cells⁺CD8⁺T-bet^{high high}Reduce the proportion of cell population, CD3⁺CD8⁺Eomes⁺T-bet^lowIt was found to have a higher tumor repellent effect by increasing the proportion of cell population.</p><p> CD8 in the presence of T cell receptor and co-occurrence stimulus signals⁺When T cells are stimulated with inflammatory cytokines such as IL-12, phosphorylation of STAT4 is increased, and phosphorylated STAT4 (pSTAT4) migrates to the nucleus and binds to the T-bet enhancer, thereby T- It is known to increase the expression of bet. Therefore, CD8 to short-lived effector cells that occurred when divalent mIL-12-Fc was administered at a concentration corresponding to an equimolar amount of 1 μg of IL-12.⁺T cell differentiation is such that administration of divalent mIL-12-Fc at a concentration corresponding to an equimolar amount of 1 μg of IL-12 compared to monovalent mIL-12-Fc causes tumor influx in tumor-transplanted mice. Analysis was performed to determine if the cause was increased expression of pSTAT4 and T-bet when T cells were activated in the regional lymph nodes.</p><p> Figure 25 (C) shows CT26^{HER2 / Neu}Tumor volume of Balb / c mice transplanted with 300 mm³Twenty-four hours after intraperitoneal administration of divalent mIL-12-Fc and monovalent mIL-12-Fc at a concentration corresponding to an equimolar amount of 1 μg of rmIL-12, the tumor inflow region CD8 isolated from lymph nodes⁺The results of flow cytometric analysis performed to measure the expression level of phosphorylated STAT4 in T cells are shown.</p><p> Specifically, as described with respect to FIG. 23 (B), CT26^{HER2 / Neu}Tumor volume of Balb / c mice transplanted with colorectal cancer cells is 300 mm³Divalent mIL-12-Fc and monovalent mIL-12-Fc were administered intraperitoneally to mice at a concentration equivalent to 1 μg of rmIL-12. After 24 hours, the lymph nodes in the tumor influx area of the mice were dissected, crushed using a wire mesh in a Petri dish, and then washed with 10 ml of 2% FBS-containing medium. Next, 1 ml of erythrocyte lysis buffer was added to lyse the erythrocytes, and the resulting cells were washed with PBS to prepare a cell suspension in this way. Inflow region lymph node cells were stained with PE-cy5 or FITC-conjugated anti-CD3 and anti-CD8 antibodies at 4 ° C for 30 minutes, washed with PBS (pH 7.4) and then fixed with cold methanol. Inflow region lymph node cells are then washed with cold PBS (pH 7.4), stained with APC-conjugated anti-pSTAT4 antibody at 4 ° C for 30 minutes, washed with cold PBS (pH 7.4), and then flow. Analyzed by cytometry (FACS Calibur, BD Bioscience) and Flow jo (Thermo Fisher Scientific). Analyze each sample by dot plot and CD3⁺CD8⁺The expression levels of pSTAT4 in T cells were compared. As a result, bivalent mIL-12-Fc was CD8 in the tumor influx region lymph nodes of tumor-transplanted mice compared to monovalent mIL-12-Fc.⁺It showed an effect of increasing the expression of pSTAT4 when T cells were activated.</p><p> FIG. 25 (D) shows CD8 in the tumor influx area lymph nodes 72 hours after the single intraperitoneal administration of FIG. 25 (C).⁺The results of flow cytometry performed to measure the proportion of T cells (expressing T-bet that inhibits memory cell differentiation) are shown.</p><p> Specifically, as described with respect to FIG. 23 (B), CT26^{HER2 / Neu}Tumor volume of Balb / c mice transplanted with colorectal cancer cells is 300 mm³Divalent mIL-12-Fc and monovalent mIL-12-Fc were administered intraperitoneally to mice at a concentration corresponding to an equimolar amount of 1 μg of rmIL-12. After 72 hours, the lymph nodes in the tumor influx area of the mice were dissected, crushed using a wire mesh in a Petri dish, and then washed with 10 ml of 2% FBS-containing medium. Next, 1 ml of erythrocyte lysis buffer was added to lyse the erythrocytes, and the resulting cells were washed with PBS to prepare a cell suspension in this way. Inflow region lymph node cells were stained with PE-cy5 or FITC-conjugated anti-CD3 and anti-CD8 antibodies at 4 ° C for 30 minutes, washed with PBS (pH 7.4), and Foxp3 / transcription factor staining buffer set (Thermo Fisher). It was fixed using Scientific) (an intracellular transcription factor staining reagent) and then permeabilized. The cells were then stained with PE or APC-conjugated anti-T-bet antibody at 4 ° C for 30 minutes, then flow cytometry (FACS Calibur, BD Bioscience) and Flow jo (Thermo Fisher) in permeabilized buffer. Scientific) Analyzed by analysis. CD3 expressing T-bet by analyzing each sample by dot plot⁺CD8⁺The proportion of T cells was compared. As a result, bivalent mIL-12-Fc was CD8 in the influx region lymph nodes of tumor-transplanted mice compared to monovalent mIL-12-Fc.⁺It showed an action to increase the expression of T-bet when T cells were activated. Therefore, CD8 to short-lived effector cells that occurred when divalent mIL-12-Fc was administered at a concentration corresponding to an equimolar amount of 1 μg of IL-12.⁺T cell differentiation was such that administration of divalent mIL-12-Fc compared to monovalent mIL-12-Fc resulted in pSTAT4 and pSTAT4 and when T cells were activated in the tumor influx region lymph nodes of tumor-transplanted mice. It was found that the cause was increased expression of T-bet.</p><p> Figures 25 (E) and 25 (F) show that monovalent mIL-12-Fc expresses two L-12 molecules, thus CD8.⁺Expression of pSTAT4 and T-bet in cells when cross-reacting with anti-Fc antibodies, such as divalent mIL-12-Fc, which can stimulate T cells with two L-12 molecules, causes the cells to react twice. The results of measuring whether or not the level increases to a level similar to the level shown when treated with the valence mIL-12-Fc are shown.</p><p> Specifically, spleen and tumor influx lymph nodes were dissected from normal Balb / c mice, crushed using a wire mesh in Petri dishes, and then washed with 10 ml of 2% FBS-containing medium. Next, 1 ml of erythrocyte lysis buffer was added to lyse the erythrocytes, and the resulting cells were washed with PBS to prepare a cell suspension in this way. Lymph node cells were stained with PE-conjugated anti-CD8 antibody at 4 ° C for 30 minutes, washed with cold PBS (pH 7.4), incubated with anti-PE microbeads (Miltenyi Biotec) for 15 minutes, MACS separator and LS. CD8 using column (Miltenyi Biotec)⁺T cells were isolated from it. Add 100 μl of 0.5 μg / ml anti-CD3 antibody to each well of a 96-well round bottom plate, incubate it at 4 ° C for 12 hours, wash with PBS to remove anti-CD3 antibody that is not attached to the plate. It was removed and 50 μl of 2 μg / ml anti-CD28 antibody was added to each well. Next, monovalent mIL-12-Fc and divalent mIL-12-Fc were reacted with various concentrations of anti-Fc antibody at 4 ° C for 30 minutes, then 20 pM IL-12 at equimolar concentrations, respectively. Added to the well. Next, CD8⁺T cells (4 x 10)⁴/ Wells) were added to each well and incubated in an incubator at 37 ° C for 3 hours to measure pSTAT4 expression and 3 days to measure T-bet expression. To measure the expression of pSTAT4 and T-bet, cells were stained by the methods described for FIGS. 25 (C) and 25 (D) and then analyzed by flow cytometry. Analyze each sample by dot plot and CD8⁺The expression levels of pSTAT4 or T-bet in T cells were compared. As a result, monovalent mIL-12-Fc, CD8⁺When T cells were cross-reacted with anti-Fc antibodies so that they could be stimulated by two L-12 molecules, the expression levels of pSTAT4 and T-bet in the cells showed that the cells were divalent mIL-12-Fc. It was shown to have increased to the levels shown when treated with.</p><p> In conclusion, as shown in FIG. 26, the monovalent mIL-12-Fc is CD8 compared to the divalent mIL-12-Fc.⁺CD8 so that T cells can differentiate into memory precursor effector cells, then effector memory cells and central memory cells.⁺Induces low expression of pSTAT4 and T-bet in T cells. Thus, monovalent mIL-12-Fc can remove tumors from tumor-transplanted mice even at low concentrations (corresponding to an equimolar amount of 0.5 μg IL-12), thus prolonging the lifespan of the mice. However, divalent mIL-12-Fc allows CD8 to differentiate into short-lived effector cells and prevent the development of memory cells.⁺Induces high expression of pSTAT4 and T-bet in T cells. Therefore, when divalent mIL-12-Fc is administered at the same molar concentration as monovalent mIL-12-Fc, it cannot completely remove the tumor from tumor-transplanted mice. Therefore, divalent mIL-12-Fc is administered at a higher concentration (corresponding to an equimolar amount of 2 μg IL-12) and cytotoxic CD8 in the effector stage, which directly destroys tumor cells.⁺Divalent mIL-12-Fc can eliminate tumors only when T cells are expanded and proliferated.</p>

The heterodimeric Fc fusion protein according to the invention allows it to retain the activity of a naturally occurring physiologically active protein composed of two or more different subunit proteins, thereby making the fusion protein. Assembled because each subunit of the protein can be fused separately to each strand of the heterodimer Fc of the immunoglobulin so that the naturally occurring morphology and structure can be maintained to the highest possible extent. It has an advantage in that it exhibits physiological activity by forming a protein. In addition, the in vivo half-life of the bioactive protein contained in the heterodimer Fc fusion protein is due to the long half-life mediated by the heterodimer Fc so that its bioactivity in vivo can be sustained for a long period of time. Can be significantly increased.

Furthermore, the heterodimer Fc fusion protein according to the present invention has an advantage in that the heterodimer Fc fusion protein can be easily produced in its natural composition without the need for further optimization of the purification process. ..

Although the present invention has been described in detail with respect to specific features, it will be apparent to those skilled in the art that this description is for preferred embodiments only and does not limit the scope of the invention. Therefore, the substantive scope of the present invention is defined by the appended claims and their equivalents.