CN115785248A - FGFR1 c-targeted FGF allosteric and application thereof - Google Patents

Abstract

The invention discloses an FGF1 variant, which can reduce the binding versatility of FGF1 receptors and reduce the carcinogenic effect of FGF1, and provides application of the FGF1 variant, which can reduce blood sugar, relieve insulin resistance and improve liver function without causing side effects such as influence on appetite, weight reduction, influence on bladder health and the like.

Description

FGFR1 c-targeted FGF allosteric and application thereof

Technical Field

The invention relates to the technical field of biology, in particular to an FGF allosteric effector targeting FGFR1c and application thereof.

Background

The incidence of type 2 diabetes (T2D), characterized by insulin resistance, lipolysis and increased blood glucose, is on the rise worldwide, in proportion to the increase in obese populations, creating a significant socio-economic burden. Hyperglycemia, if not treated in the immediate future, can lead to serious complications, including metabolic disorders, vascular disease, chronic inflammation, blindness, and amputation. The primary therapy for managing hyperglycemia in T2D is Thiazolidinediones (TZDs) and their derivatives. These synthetic heterocyclic compounds bind to and activate peroxisome proliferator-activated receptor gamma (PPAR γ) in adipose tissue, up-regulate glucose uptake and expression of adipokine-related genes, thereby promoting insulin sensitivity. PPAR γ is a pleiotropic transcription factor that affects gene expression in many other tissues, as well as many other tissues including immune cells, brain and skeletal muscle. Given the diversity of genes and tissues targeted by PPAR γ, it is not surprising that TZDs carry many adverse side effects, including weight gain, macular edema, hepatic steatosis, bone fractures, heart attack, and increased risk of stroke.

Fibroblast growth factor 1 (FGF 1), an initiating member of the FGF ligand family, was identified as an evolutionary conserved target of PPAR γ that can maintain White Adipose Tissue (WAT) homeostasis and remodeling to address nutritional challenges (e.g., high fat diet). Loss of adipose FGF1 results in a worsening of the diabetic phenotype with a concomitant abnormal adipocyte morphology and inflammatory injury. This finding led to the idea that fatty FGF1 might promote PPAR γ intervention to improve glucose control and insulin sensitivity. In fact, acute peripheral injection of recombinant FGF1 alleviates hyperglycemia and insulin resistance in leptin receptor knockout mice (db/db) and Diet Induced Obesity (DIO) mice by promoting insulin-dependent glucose uptake in adipose tissue. Recent data indicate that FGF1 exerts its antidiabetic effect by inhibiting hepatic glucose production through down-regulation of lipolytic phosphodiesterase. Although acute peripheral injection of FGF1 results in transient normoglycemia, lateral ventricular (icv) administration of FGF1 has been shown to cause sustained relief of T2D rodent hyperglycemia by acting on the hypothalamus. We have recently found that skeletal muscle, as another target tissue, can independently intervene with insulin signaling in the antihyperglycemic effects of FGF 1. We found that FGF1 upregulates the expression of the muscle-specific glucose transporter GLUT4 and cell surface translocation by binding, dimerizing and activating the c-spliced isoform of FGF receptor 1 (FGFR 1). Activated FGFR1c stimulates the AMPK signaling pathway, which is required for the known antihyperglycemic activity of metformin, adiponectin, leptin, or other natural products that function as AMPK activators.

As a downstream target of PPAR gamma, FGF1 is expected to avoid the side effects of osteoporosis, urinary retention, myocardial hypertrophy and the like caused by the direct action of Thiazolidinediones (TZDs) on PPAR gamma, and simultaneously, FGF1 has good development and clinical application potential in the treatment of type 2 diabetes due to the prominent effect of reducing blood sugar and enhancing insulin sensitivity shown by FGF 1. However, the available treatments to convert FGF 1to T2D face two problems: 1) Carcinogenesis of FGF 1; FGF1 is a potent mitogen capable of stimulating proliferation of a variety of cell types of mesenchymal and epithelial origin; 2) FGF1 receptor binding versatility: FGF1 is a versatile FGFR ligand capable of binding and activating seven major FGFR subtypes, namely the b and c subtypes of FGFR1-3 (FGFR 1b, FGFR2b, FGFR3b, FGFR1c, FGFR2c, FGFR3 c) and FGFR4.

Disclosure of Invention

It is an object of the present invention to address at least the above problems and to provide at least the advantages described hereinafter.

It is yet another object of the present invention to provide an FGF1 variant capable of reducing the versatility of FGF1 receptor binding and reducing the carcinogenic effects of FGF 1;

provides an application of FGF1 variant, which can reduce blood sugar, relieve insulin resistance and improve liver function without side effects such as influence on appetite, weight reduction and influence on bladder health.

To achieve these objects and other advantages in accordance with the present invention, there is provided an FGF1 variant having an amino acid sequence which lacks the entire flexible N-terminal tail, including the residues Met1to Asn22, and the immediately preceding two core residues, including Try23to Lys24, and an alteration in the three heparin binding sites (Lys 127Asp, lys128 gin and Lys133 Val) as compared to the amino acid sequence of FGF 1.

A polynucleotide encoding said FGF1 variant is provided.

A vector or cell comprising the polynucleotide is provided.

A pharmaceutical combination is provided comprising said variant and a pharmaceutically acceptable carrier.

Provides an FGF1 variant for use in the preparation of a medicament for reducing the versatility of FGF1 receptor binding and for reducing the carcinogenic effects of FGF 1.

There is provided the use of an FGF1 variant for the preparation of a medicament for alleviating insulin resistance without affecting appetite, body weight and bladder health.

There is provided the use of an FGF1 variant for the preparation of a medicament for lowering blood glucose without affecting appetite, body weight and bladder health.

There is provided the use of an FGF1 variant in the manufacture of a medicament for improving insulin resistance and improving liver function without affecting appetite, body weight and bladder health.

The invention at least comprises the following beneficial effects:

the flexible N-terminal tail of FGF1 intervenes in the receptor binding versatility by interacting with the different FGFR subtypes with high degeneracy. FGF1 strictly relies on its N-terminal tail to bind FGFR2b, FGFR2c and FGFR3c, however this region is largely unnecessary for FGFR1c binding, in contrast to the first two core residues (Try 23to Lys 24) which have constant interaction with all FGFRs, and therefore FGF1 variants with the entire flexible N-terminal tail of FGF1 deleted and the first two core residues immediately adjacent are specific low affinity ligands, which retain full anti-diabetic effect in preclinical studies in healthy diabetic mice and monkeys, without side effects affecting appetite, weight loss and bladder health.

Additional advantages, objects, and features of the invention will be set forth in part in the description which follows and in part will become apparent to those having ordinary skill in the art upon examination of the following or may be learned from practice of the invention.

Drawings

FIG. 1 is FGF1 ^ΔHBS Graphs inducing edema, inflammatory cell infiltration, thickening of the lamina propria of the bladder mucosa, and severe weight loss;

FIG. 2 is a schematic representation of a designed FGF1 ^ΔHBSΔNT ；

FIG. 3 is FGF1 ^WT And variants thereof binding affinity maps to FGFRs;

FIG. 4 is FGF1 ^ΔHBSΔNT FGFR1c map specifically and partially dimerized and activated;

FIG. 5 shows FGF1 in vivo ^ΔHBSΔNT A FGFR1c map selectively activatable;

FIG. 6 shows FGF1 in normal C57BL/6J mice and healthy cynomolgus monkeys ^ΔHBSΔNT Does not result in bladder damage or weight loss patterns;

FIG. 7 is FGF1 ^ΔHBS And FGF1 ^ΔHBSΔNT Both do not induce hypoglycemia in acute and chronic conditions;

FIG. 8 shows FGF1 ^ΔHBSΔNT GLUT4 expression and translocation driven by AMPK lowers glucose levels, which lower levels in combination with FGF1 ^ΔHBS Equivalent;

figure 9 is a graph of the effect of AMPK or FGFR1 inhibitors on acute blood glucose lowering and GLUT4 expression and translocation;

FIG. 10 shows long-term administration of FGF1 ^ΔHBSΔNT Lowering blood glucose and promoting insulin sensitivity, at levels consistent with FGF1 ^ΔHBS A comparative drawing;

FIG. 11 shows long-term FGF1 injection ^ΔHBSΔNT Effects on high fat induced obesity (DIO) and leptin receptor knock-out (db/db) mice feeding, body weight and bladder pathology;

FIG. 12 is FGF1 ^ΔHBSΔNT And FGF1 ^ΔHBS By inhibitionMacrophage accumulation in white adipose tissue to relieve insulin resistance;

FIG. 13 is FGF1 ^ΔHBSΔNT Therapeutic effect on T2D non-human primates.

Among them, (A-C) C57BL/6J mice in FIG. 1 were intraperitoneally injected with FGF1 daily ^ΔHBS (2.5 mg/kg) or PBS, change in food intake (A), body weight (B) and bladder tissue H after 2 weeks of injection&E, dyeing (C). (D) Healthy cynomolgus monkey is intraperitoneal injection FGF1 every day ^ΔHBS (1.5 mg/kg) or PBS, bladder tissue staining 4 weeks after injection. The bladder transitional epithelium, lamina propria, and detrusor are outlined with dashed lines. Arrows indicate the mucosal lamina propria. The (a-B) data are shown as mean ± SEM (n = 6). * P<Pbs control, bidirectional analysis of variance (retest) was performed first, followed by Bonferroni statistical method;

FIG. 2 (A) is based on FGF1 ^ΔHBS Contemplated FGF1 variants (FGF 1) ^ΔHBSΔNT ) Schematic structural diagram of (1). (B) The dimer FGF1-FGFR1c-HS is constructed by superimposing the FGF1-FGFR1c (PDB ID:3 OJV) monomeric structure onto the FGF2-FGFR1c-HS (PDB ID:1FQ 9) dimeric structure. (C-E) Crystal Structure of FGF1-FGFR2b (PDB ID:3 OJM) (C) Crystal Structure of FGF1-FGFR2C (PDB ID:1 DJS) (D), crystal Structure of FGF1-FGFR3C (PDB ID:1FY 7) (E). FGF1 is shown symbolically in the figure, with selected residues appearing as rod-like NTs indicating the position of the N-terminus of FGF 1), a close-up view showing the interface between FGF1 and different FGFRs is shown. Hydrogen bonds are indicated by dashed lines. The residues involved in hydrophobic interactions are rod-shaped. (F-I) representative SPR sensorgram to show FGF1 ^WT (F),FGF1 ^ΔHBS (G),FGF1 ^ΔNT (H) And FGF1 ^ΔHBSΔNT (I) Binding affinity to the binding regions (D2-D3 regions) of ligands FGFR1c, FGFR2b, FGFR2c, FGFR3c, and FGFR4.

In FIG. 3 (A-B), FGFRs (FGFR 1c, FGFR2B, FGFR2c, FGFR3c or FGFR 4) were passed through a column containing FGF1 at increasing concentrations ^WT (A) Or FGF1 ^ΔHBS (B) The equilibrium binding response obtained for the sensor chip of (a) is plotted as a function of the analyte (FGFR) concentration. (C-D) by flowing FGFR1C in increasing concentrations through a column containing FGF1 ^ΔNT (C) Or FGF1 ^ΔHBSΔNT (D) Equilibrium binding reaction mapping analysis obtained by sensor chipFunction of the concentration of the substance (FGFR 1 c). The equilibrium dissociation constant (Kd value) is derived from the saturation binding curve;

FIG. 4 (A) FGF1 in L6 rat myoblasts expressing human FGFR in vitro ^WT ,FGF1 ^ΔHBS ,FGF1 ^ΔNT or FGF1 ^ΔHBSΔNT PLA analysis figure, scale bar, 25 μm inducing dimerization of FGFRs (FGFR 1c, FGFR2b, FGFR2c, FGFR3b, FGFR3c and FGFR 4). (B) PLA was quantified by counting the number of positive spots per nucleus. Data are shown as mean ± SEM (n = 8-10). * P<0.01；***p<0.001；****p<0.0001 ns, no significance, firstly carrying out common one-way variance analysis, and then carrying out Tukey method statistics; . (C) In L6 cells expressing human FGFR1c, FGFR2b, FGFR2c, FGFR3b, FGFR3c or FGFR4 in vitro, with increasing doses (0-60 nM) of FGF1 ^ΔHBS Or FGF1 ^ΔHBSΔNT Immunoblotting (left panel) and semi-quantitative analysis (right panel) of induced phosphorylation levels of PLC γ and FRS 2. Data are shown as mean ± SEM of three independent measurements. * p is a radical of<0.05,**p<0.01,***p<0.001,****p<0.0001 ns, no significance, firstly carrying out common one-way variance analysis, and then carrying out Tukey method statistics;

FIG. 5 (A-D) mRNA levels of FGFRs in White Adipose Tissue (WAT) of C57BL/6J mice (WAT) (A), mRNA levels of FGFRs in liver (B), mRNA levels of FGFRs in muscle (C) and mRNA levels of FGFRs in bladder tissue (D) determined by RT-PCR. (E-I) intraperitoneal injection of FGF1 into C57BL/6J mice ^ΔHBS ,FGF1 ^ΔHBSΔNT (0.5 mg/kg), or saline (E) 15 minutes later, the level of ERK1/2 phosphorylation in the tissue was determined by immunoblotting. Quantitative immunoblotting of liver (F), quantitative immunoblotting of WAT (G), quantitative immunoblotting of muscle (H) and quantitative immunoblotting of bladder (I). The (a-D, F-I) data are shown as mean ± SEM (n = 8-10). * P<0.01；***p<0.001；****p<0.0001,ns, not significant, (F-I) common one-way variance analysis is carried out firstly, and then Tukey method statistics is carried out;

in FIG. 6, C57BL/6J mice were treated by daily intraperitoneal injection of FGF1 ^WT Or a variant thereof (2.5 mg/kg), injected for two weeks and administered PBS or FGF1 for four weeks to cynomolgus monkeys (A-F) ^ΔHBSΔNT Analysis (1.5 mg/kg/day) (G-I). (A-B) FGF1 ^WT ,FGF1 ^ΔNT ,FGF1 ^ΔHBS Or FGF1 ^ΔHBSΔNT Representative H of bladder tissue staining of treated normal C57BL/6J mice (n = 6)&E (A), ki67 and PCNA (B). The bladder transitional epithelium, mucosal lamina propria, and detrusor are outlined with dashed lines. Arrows indicate the mucosal lamina propria (a) and the proliferation marker (B). Scale bar: panel A (50 μm), and panel B (200 μm). (C-D) FGF1 ^WT ,FGF1 ^ΔNT ,FGF1 ^ΔHBS Or FGF1 ^ΔHBSΔNT Semi-quantitative statistics of PCNA and Ki67 positive cells in bladder tissue of C57BL/6J mice after treatment. (E-F) Food Intake (FI) changes (E) and C57BL/6J mouse Body weight (Body weight, BW) changes (F). (G-I) four weeks after PBS or FGF1 ^ΔHBSΔNT (1.5 mg/kg/day) H of bladder tissue of healthy cynomolgus monkey&E staining (G), change in food intake (H), change in body weight (I), scale bar 100 μm in graph G. The (C-D) data are shown as mean. + -. SEM. * P<0.01,****p<0.0001 ns, no significance, firstly carrying out common one-way variance analysis, and then carrying out Tukey method statistics;

FIG. 7 (A-B) 8 week old Normal Male C57BL/6J mice daily injected intraperitoneally with PBS, FGF1, respectively ^ΔHBS Or FGF1 ^ΔHBSΔNT (2.5 mg/kg body weight), and then the blood glucose level in an acute environment (A) and the blood glucose level in a chronic environment (B) were measured. Data are shown as mean ± SEM (n = 6-7);

FIG. 8 (A-B) Single intraperitoneal injection of FGF1 ^ΔHBS, FGF1 ^ΔHBSΔNT (0.5 mg/kg), or fasting blood glucose levels (A) in db/db mice and (B) in DIO mice after PBS (control). (C) Positron emission tomography/computed tomography (PET-CT) imaging of 18F-FDG uptake in organs of db/db mice 15 minutes after caudal intravenous injection of radiolabeled fluorodeoxyglucose (18F-FDG) was performed in db/db mice. All mice were intraperitoneally injected with FGF1 prior to 18F-FDG treatment ^ΔHBS ,FGF1 ^ΔHBSΔNT (0.5 mg/kg), or PBS for 2 hours. The image in the left figure is a side view. FGF1 for (D-E) ^ΔHBS ,FGF1 ^ΔHBSΔNT (0.5 mg/kg) or immunofluorescence of hind limb muscle (D) and GLUT4 translocation quantification (E) in db/db mice treated with PBS, proportional size, 50 μm. (F) With FGF1 ^ΔHBS ,FGF1 ^ΔHBSΔNT (0.5 mg/kg) or in the muscle of the hind limb of db/db mice treated with PBSGLUT4 and phosphorylated AMPK and AKT levels. . Data in (A-C, E-F) are shown as mean. + -. SEM. * p is a radical of<0.05,**p<0.01,***p<0.001,****p<0.0001; ns, not significant, (a, B) bi-directional analysis of variance (repeated measures) first followed by Bonferroni statistics (n = 5-6); (C) Carrying out common bidirectional variance analysis, and then carrying out Sidak method statistics (n = 4); (E, F) carrying out common one-way variance analysis and Tukey method statistics (n = 5-6);

FIG. 9 (A) db/db mice pretreated with AMPK inhibitor (Compound C) or FGFR1 inhibitor (PD 166866) followed by intraperitoneal injection of FGF1 ^ΔHBS ,FGF1 ^ΔHBSΔNT (0.5 mg/kg), or PBS. After 3 hours, blood glucose levels were measured. (B-C) AMPK inhibitor (B) or FGFR1 inhibitor (C) pre-treat GLUT4 protein and AMPK α phosphorylation levels in the lower limb muscles of db/db mice. Quantification of egg Western blots is shown on the right. (D) db/db mice (AMPK inhibitor, FGFR1 inhibitor or PBS pretreatment) were injected intraperitoneally with FGF1 ^ΔHBS ,FGF1 ^ΔHBSΔNT After 3 hours, GLUT4 translocation in lower limb muscles was analyzed by immunofluorescence staining. The scale bar is 50 μm. The (a-C) data are shown as mean ± SEM (n = 6). * p is a radical of<0.05；***p<0.001；**p<0.01；****p<0.0001; ns is not significant, common one-way variance analysis is firstly carried out, and then Tukey method statistics is carried out;

FIG. 10 (A-C) db/db mice injected daily intraperitoneally with FGF1 ^ΔHBS ,FGF1 ^ΔHBSΔNT (0.5 mg/kg), or PBS, injected for 20 days. Change in blood glucose level (a), glucose Tolerance Test (GTT) (B) and integrated area under the curve of change in blood glucose level (C) over 180 minutes (AUC). (D-I) DIO mice daily intraperitoneal injection of FGF1 ^ΔHBS ,FGF1 ^ΔHBSΔNT (0.5 mg/kg), or PBS, for 4 weeks. Change in blood glucose level (D), GTT (E) and AUC under the curve of change in blood glucose level (F) over 120 min. (G, H) Insulin Tolerance Test (ITT) (G) AUC under the curve of change in blood glucose level (H) over 120 minutes. (I) plasma insulin levels. Data in (A-F, H, I) are shown as mean. + -. SEM. * p is a radical of formula<0.05,***p<0.001,****p<0.0001; ns, not significant, (a, B, D, E) bi-directional analysis of variance (repeated measures) followed by Bonferroni statistics (n = 5-6); (C, F, H, I) carrying out common one-way analysis of variance and Tukey method statistics (n = 6);

FIG. 11 (A-C) DIO mice daily intraperitoneal injection of FGF1 ^ΔHBS ,FGF1 ^ΔHBSΔNT (0.5 mg/kg) or PBS for 4 weeks. Changes in food intake (A) and body weight (B), (C) H of bladder tissue&And E, dyeing. (D-F) daily injection of FGF1 ^ΔHBS ,FGF1 ^ΔHBSΔNT (0.5 mg/kg) or analysis of db/db mice treated with PBS for 4 weeks. Changes in food intake (D) and body weight (E), (F) H of bladder tissue&And E, dyeing. Data (a, B, D, E) are presented as mean ± SEM (n = 6). * P<0.01,***p<0.001; ns, not significant: (B, E) two-way analysis of variance (repeated measurements), followed by Bonferroni. The bladder transitional epithelium, lamina propria, and detrusor are outlined with dashed lines. Arrows indicate the mucosal lamina propria.

FIG. 12 (A-I) db/db mice intraperitoneally injected daily with FGF1 ^ΔHBS ,FGF1 ^ΔHBSΔNT (0.5 mg/kg), or PBS, for 4 weeks. (A, B) plasma TNF alpha levels (A) and plasma IL-6 levels (B). (C, D) macrophage marker mRNA level of F4/80 (C) and macrophage marker mRNA level of Cd68 (D) in WAT as determined by RT-PCR. (E) Immunofluorescence quantification of F4/80 and CD11c double positive cells in white fat. (F-G) mRNA and protein levels of CCL2 determined by RT-PCR (F) and Western blot (G). Data are shown as mean ± SEM. * p is a radical of<0.05,**p<0.01,***p<0.001,****p<0.0001; ns, not significant, (a-D, F-G) first performed a general one-way anova followed by Tukey's method statistics (n = 6);

in FIG. 13, 5 male cynomolgus monkeys with idiopathic diabetes were administered 0.15mg/kg of FGF1 per day ^ΔHBSΔNT Treatment was for 4 weeks and was monitored for an additional 12 days during the elution phase. (A) study design of non-human primates. (B-D) changes in fasting blood glucose levels (B), changes in food intake (C), changes in body weight (D) by FGF1 ^ΔHBSΔNT Percent change from baseline after dosing is indicated. (E) Intraperitoneal Glucose Tolerance Test (GTT) was performed on days 0, 14 and 28. (F) integrated area under the curve of change in blood glucose level (AUC). (G-L) FGF1 ^ΔHBSΔNT The metabolic parameters before and after administration are expressed as percentage change in plasma insulin (G), percentage change in HbA1c (H), percentage change in Triglyceride (TG) (I), percentage change in total cholesterol (T-CHO) (J), alanine aminotransferase(ALT) percent change (K), and aspartate Aminotransferase (AST) (L). The (B, D, F-L) data are shown as mean ± SEM (n = 5). * p is a radical of formula<0.05,**p<Day 0.01vs 0, data analyzed by paired t-test, A, p<0.05 was considered statistically significant.

Detailed Description

The present invention is further described in detail below with reference to examples so that those skilled in the art can practice the invention with reference to the description.

1. Experiment of

1.1 expression and purification of proteins

Have disclosed targeting FGF1 ^ΔHBS (residues Met1-Asp 155) pET30 a-based bacterial expression constructs as templates for PCR to generate FGF1 with deletion of the first 24N-terminal residues ^ΔHBSΔNT . The qualified BL21 (DE 3) Escherichia coli is transformed to construct FGF1 ^WT And variants thereof (FGF 1) ^ΔNT ,FGF1 ^ΔHBS or FGF1 ^ΔHBSΔNT ) The expression construct of (1). 1mM isopropyl beta-D-1-thiogalactoside (IPTG) induced protein expression for 4 hours at 37 ℃ and cells were harvested by centrifugation. Cells were lysed using an emulsified Emulsiflex-C3 high-capacity homogenizer (Ontawa Ottostein, ontariostein, ontario, canada). FGF1 was purified by size exclusion chromatography (HiLoadTM 16/600Superdex TM75, amysia) using a heparin affinity chromatography column in sequence ^WT And FGF1 ^ΔNT . FGF1 purification by cation exchange column (Source S, GE Healthcare, piscataway, N.J.) and gel filtration chromatography (HiLoadTM 16/600Superdex TM 75) ^ΔHBS And FGF1 ^ΔHBSΔNT 。

The extracellular ligand-binding regions (D2-D3 domains) of human FGFR1c (D142-R365), FGFR2b (D132-P364), FGFR2c (N149-E368), FGFR3c (D147-E365) and FGFR4 (Q144-D355) were expressed in E.coli and then refolded in vitro from isolated bacterial inclusion bodies as described previously. The folded FGFRs were obtained on a heparin affinity chromatography column, eluted with pH =7.5,1M NaCl in 25mM Hepes-NaOH, and further purified by size exclusion chromatography (HiLoadTM 16/600superdex tm 75) using a buffer containing 1M NaCl, pH =7.5, 25mM Hepes-NaOH, with the purity of all proteins estimated to be >98% based on SDS-PAGE analysis.

1.2 surface plasmon resonance

Surface Plasmon Resonance (SPR) experiments were performed on a BIAcore T200 system (GE healthcare) in HBS-EP buffer (10 mM Hepes-NaOH, pH 7.4,150mM NaCl,3mM EDTA,0.005% surfactant P20). Recombinant FGF1 was conjugated using an amine coupling kit (BR-1000-50, GE healthcare) as described in (J.Niu et al, proc Natl Acad Sci U S A117,29025-29034 (2020), by inhibiting the receptor dimerization capacity of FGF 19) ^WT ,FGF1 ^ΔNT ,FGF1 ^ΔHBS Or FGF1 ^ΔHBSΔNT Fixed on CM5 biosensor chip (BR 1005-30, GE medical) to obtain FGF1 ^WT ,FGF1 ^ΔNT ,FGF1 ^ΔHBS Or FGF1 ^ΔHBSΔNT And (3) a chip. Briefly, FGF1 ^WT Or a variant thereof, was immobilized to 400-480 Response Units (RU) by activating the chip with sodium acetate at pH 5.5. The surface of the chip is sealed by 1M ethanolamine-HCl with the pH value of 8.5, a control flow channel is blank, after the preparation of the chip is finished, FGFRs ligand binding regions (FGFR 1c, FGFR2b-2c, FGFR3c or FGFR 4) with continuously increased concentration are prepared in HBS-EP buffer solution, and flow through the sensor chip within 180s at the speed of 30 muL/min. After each injection, HBS-EP buffer was flowed over the sensor chip at 30. Mu.L/min for 60s to monitor dissociation. The sensor chip surface was regenerated by 2.0M NaCl in 10mM Hepes NaOH (pH 7.5). The nonspecific response of the control flow channel was subtracted from the response recorded for the FGF1 flow channel. The data processing adopts BIA-Evaluation software. The equilibrium dissociation constant (KD) was obtained from the fitted saturated binding curve.

1.3 ortho-ligation (PLA)

Extracellular expression of human FGFR1c, FGFR2b, FGFR2, FGFR3c or FGFR 4L 6 cells was cultured in 12-well plates (1.5 × 105/well). L6 cells were starved for 8h in DMEM medium without FBS, then sequentially with 60nM FGF1 ^WT 、FGF1 ^ΔHBS 、FGF1 ^ΔNT Or FGF1 ^ΔHBSΔNT Stimulation was carried out for 15 minutes. After treatment, the samples were rinsed three times with PBS and fixed with pre-cooled 4% paraformaldehyde for 30 minutes. Samples were deposited on slides, blocked with 5% BSA at 37 ℃ for 1 hour, and incubated with anti-FGFR 1 (ab 824)Mixtures of anti-FGFR 2 (sc-6930, san Crux biotechnology and CST23328, cell Signaling Technology), anti-FGFR 3 (sc-13121, san Crux biotechnology and CST4574, cell Signaling Technology) or anti-FGFR 4 (ab 41948, abcam and sc-136988, san Crux biotechnology) primary antibodies were incubated overnight in a humidity chamber at 4 ℃ and then PLA was performed as described (J.Niu et al., mitosis inhibited by inhibiting the receptor dimerization capacity of FGF 19. Proc Natl Acad Sci U S A117, 25-29034 (2020)). The resulting data was analyzed to determine the total number of PLA signals per cell.

1.4 rat myoblast-based assays

A rat myoblast cell line (L6) is established, which expresses wild-type human FGFR1c, FGFR2b, FGFR3c, FGFR3b, FGFR3c or FGFR4 in vitro, and is maintained. Prior to each experiment, cells were stripped in free RPMI 1640 medium with FBS for 5 hours, then FGF1 ^WT ,FGF1 ^ΔHBS Or FGF1 ^ΔHBSΔNT Stimulation was carried out for 10 minutes. Cells were lysed in RIPA buffer containing protease inhibitors and centrifuged at 4000xg for 10 min at 4 ℃. Protein concentration in the soluble fraction was determined using the bicinchoninic acid (BCA) kit and samples were analyzed by western blot. Density analysis of immunoblots was performed using Image J software (National Institutes of Health, bethesda, md., USA).

1.5 animals and animal welfare

Male C57BL/6J mice, male db/db mice (C57 BLKS/J-leprdb/leprdb), diet Induced Obesity (DIO) mice were purchased from Nanjing university model animal research center. Animals were housed in a Specific Pathogen Free (SPF) environment controlled animal facility (22 ± 1 ℃,50-60% humidity, 12-h light/dark cycle, light at 7 am), with free access to food and water. All steps were approved by the animal care and use committee of the university of medical science, wenzhou.

Clinically healthy, co-living adult cynomolgus monkeys (3-5 years old, weight range 2.90-4.50 kg) were purchased from Guangzhou Xiangguan Biotech, inc., china). Monkeys were individually identified by collar tags and housed in a conventional clean GLP certification facility of the department of drug safety evaluation of tianjin institute of medicine, china. Preclinical evaluations were performed in a good laboratory practice facility certified by the FDA in china. Animals were kept in a 12 hour light/dark cycle with constant temperature (16-26) deg.C and humidity (40-70%). The study using cynomolgus monkeys was performed according to the current guidelines for animal welfare (the Institute for Laboratory Animals (ILAR) guidelines for care and use of laboratory animals, 1996, animal welfare act, revised in 1966, 1970, 1976 and 1985, parts 1, 2 and 3 of 9 CFR). The procedures used were approved by the animal Care and use Committee of the institute of Tianjin pharmaceutical research, china.

Medium-aged (10-16 years old) cynomolgus monkeys (6 males, average body weight 9.6 + -0.7 kg) with idiopathic diabetes (fasting plasma glucose ≥ 8.0 mM) were purchased from Liangguan Biotech, inc., jiangsu, china. Animals were individually housed in a 12 hour light/dark cycle environment with controlled temperature and humidity. All procedures involving diabetic monkeys were approved by the animal care and use committee of the national jiangsu liangcuan biotechnology limited.

1.6FGF1 ^ΔHBSΔNT Safety assessment in normal C57BL/6J mice and healthy cynomolgus monkeys

Eight-week-old normal C57BL/6J mice of similar body weight were randomly divided into four groups, each of which was intraperitoneally injected with PBS, FGF1 daily ^ΔHBS ,FGF1 ^ΔNT And FGF1 ^ΔHBSΔNT (2.5 mg/kg body weight/day). Food consumption and body weight were monitored every other day. Two weeks after dosing, mice were euthanized by intraperitoneal injection with sodium pentobarbital, excised for major organs and tissues (including heart, kidney, liver, fat and bladder), and passed through H&E staining for morphological changes.

Clinically healthy monkeys were divided into three groups, each group consisting of one male and one female. One group was administered with normal saline, and the other two groups were administered with 1.5mg/kg of FGF1 by subcutaneous injection each day ^ΔHBS And FGF1 ^ΔHBSΔNT . Food consumption and body weight were monitored at the indicated time points. After the treatment period, the animals were euthanized and their hearts, livers, lungs, kidneys and bladders were harvested and passed through H&E staining for morphological changes.

1.7FGF1 ^ΔHBS Or FGF1 ^ΔHBSΔNT Research on hypoglycemic effect of T2D mice

8 week old db/db or DIO mice with comparable body weight and blood glucose levels were randomized into three groups (PBS, FGF 1) ^ΔHBS And FGF1 ^ΔHBSΔNT ). Blood samples were collected via tail vein blood and blood glucose levels were measured using a FreeStyle whole blood glucose monitor (Abbott Diabetes Care inc., alameda, CA). To evaluate the acute hypoglycemic effect, mice were injected intraperitoneally with FGF1 in a single injection ^ΔHBS Or FGF1 ^ΔHBSΔNT (0.5 mg/kg weight). To test the long-term therapeutic effect, mice were injected intraperitoneally with PBS, FGF1 daily ^ΔHBS Or FGF1 ^ΔHBSΔNT Over 20 days, an intraperitoneal glucose tolerance test was performed. In this experiment, overnight fasted mice were administered a glucose solution (2.0 g/kg body weight, abdominal cavity). Blood samples were then collected at 0, 15, 30, 60, 120 and 180 minutes and blood glucose levels were determined as described above. The area under the curve (AUC) was calculated by the trapezoidal method of the glucose tolerance curve using GraphPad Prism 7 software. Food consumption and body weight were monitored every other day during the long-term treatment. At the end of the experiment, mice were euthanized by intraperitoneal injection with sodium pentobarbital, bladder tissue was excised, and passed through H&E staining for morphological changes.

1.8 pharmacokinetic evaluation

Adult male SD rats (220-250 g) with single intraperitoneal injection of FGF1 ^ΔHBS And FGF1 ^ΔHBSΔNT (all at 1.0mg/kg body weight), FGF1 was measured ^ΔHBS Or FGF1 ^ΔHBSΔNT The in vivo half-life of (c). Detailed pharmacokinetic analysis is as described previously (l.zhao et al, paracrine-endocrine FGF chimeras as effective therapies for metabolic diseases. EBioMedicine 48,462-477 (2019)). In particular, protein levels were measured using human FGF1 immunoassay ELISA kit (biological engineering ltd, bosch, switzerland, northHubei). Pharmacokinetic parameters were calculated using drug and statistical software (DAS, v 2.0, chinese mathematic pharmacology professional committee). The elimination half-life (t 1/2) was calculated by the formula t1/2=0.693/Ke, where Ke is the elimination rate constant.

1.9 Micropositron emission computed tomography (MicroPET CT) imaging and analysis

Night forbiddingDb/db mice fed with injections of PBS, FGF1 ^ΔHBS Or FGF1 ^ΔHBSΔNT (0.5 mg/kg body weight). After 2 hours, mice were given a single dose (177.6 ± 20.65mCi, i.v.) of 18F-FDG via the caudal vein, then anesthetized (by inhalation of 1.5% isoflurane in oxygen at 2L/min) and scanned for 1 hour in an Inveon preclinical multi-modal PET-CT (siemens munich, germany). The images are reconstructed using a three-dimensional ordered subset expectation-maximization (3D-OSEM) algorithm and a maximum a posteriori probability (MAP) algorithm in sequence. Under CT image guidance, a 3D region of interest (ROI) is rendered on the liver and muscle; radioactivity of the corresponding tissues was measured using Inveon Research Workplace (Siemens Munich, germany). After scanning, the mice were sacrificed by cervical dislocation. Liver and muscle (including anterior, posterior and dorsal muscles) were excised and radiotracer uptake (ID%/g) in these tissues was determined by gamma-counting. An aliquot of the injected dose was set aside and counted simultaneously with the tissue samples to correct for radioactive decay.

1.10 analysis of GLUT4 translocation and expression in skeletal muscle.

Male db/db mice of 8 weeks of age were randomly divided into three groups (PBS, FGF 1) ^ΔHBS Or FGF1 ^ΔHBSΔNT ). Intravenous injection of PBS, FGF1 ^ΔHBS Or FGF1 ^ΔHBSΔNT (0.5 mg/kg body weight). After 6 hours (when the most pronounced acute hypoglycemic effect of the FGF1 variant was seen), the mice were euthanized and skeletal muscle was harvested. Muscle tissue was fixed in 4% (wt/vol) formaldehyde in PBS, paraffin embedded, and sectioned at 4 μm. Immunofluorescence was used to detect GLUT4 translocation in skeletal muscle. GLUT4 levels in skeletal muscle were analyzed by Western blot.

1.11FGF1 ^ΔHBSΔNT Determination of the Effect on the spontaneous T2D cynomolgus monkey Metabolic Regulation

FGF1 evaluation with 5 female spontaneous T2D cynomolgus monkeys ^ΔHBSΔNT The therapeutic effect of (1). After 1 week of adaptation, diabetic monkeys were injected subcutaneously with FGF1 ^ΔHBSΔNT (0.15 mg/kg) once daily for 4 weeks. Fasting blood glucose levels, body weight and food intake were monitored every other day. An Oral Glucose Tolerance Test (OGTT) was performed and plasma insulin levels were determined after two and four weeks of treatment. For OGTT, diabetic monkeysFasting was performed overnight (16 hours) and then with oral glucose solution (1.75 g/kg body weight). Blood glucose levels were measured at 0, 15, 30, 60, 90, 120 and 180 minutes post-glucose administration using an Accu-Chek device (Mannheim Roche, germany). The levels of Triglycerides (TGs), total cholesterol (T-CHO), alanine Aminotransferase (ALT) and aspartate Aminotransferase (AST) were determined by an Advia 2400 automated biochemical analyzer (Siemens, germany). Determination of FGF1 ^ΔHBSΔNT Serum tumor biomarker levels in diabetic cynomolgus monkeys before and after treatment to check the safety of FGF1 variants.

1.13 quantitative real-time PCR and Western blot analysis

Total RNA was extracted from mouse tissues using TRIzol reagent (Walthersmemer Feishal, mass.) and reverse transcribed to complementary DNA using Prime Script RT kit (Takara). Quantitative real-time PCR was performed on a StepOnePlus real-time PCR system (Applied Biosystems QuantStudio 3) using SYBR Premix Ex Taq (Takara) and appropriate primers (listed in Table S1). Beta-actin was used as an internal control to normalize the difference in the amount of total RNA added to each reaction.

Table S1 gene primer for RT-PCR

The tissue was lysed in RIPA lysis buffer (25 mM Tris-HCl, pH7.6, 150mM NaCl,1% NP-40,1% sodium deoxycholate, 0.1% SDS) containing protease and phosphatase inhibitors (Waltheromemer femtole, mass.). The total amount of protein was quantified using BCA kit (protein detection kit, china shanghai bi yunnan biotechnology limited). 40 micrograms of cell or tissue lysis protein were separated by 8-12% SDS-PAGE and electrotransferred to nitrocellulose membrane. Phosphorylation of FGFR1 (ab 59194; abcam), phosphorylation of PLC γ (CST 14008, cell Signaling Technology) with an anti-glyceraldehyde-3-phosphate dehydrogenase antibody (GAPDH) (CST 2118, cell Signaling Technology); PLC γ (CST 5690, cell Signaling Technology), phosphorylated FRS2 (CST 3864, cell Signaling Technology); FRS2 (sc-17841, santa Cruz Biotechnology); phosphorylated ERK1/2 (CST 4370, cell Signaling Technology); phosphorylated AMPK α (CST 2535, cell Signaling Technology); or AMPK alpha (CST 5831, cell Signaling Technology).

The immunoreactive bands were detected by incubation with a secondary antibody (Dallas Santa Cruz Biotechnology, tex.), labeled with horseradish peroxidase, and visualized using Enhanced Chemiluminescence (ECL) reagents (Heracles Bio-Rad, calif.). The amount of immunoreactive protein was analyzed using Image J software (besserda national institutes of health, maryland, usa) and normalized against a control group.

1.14 statistical analysis

The in vitro experiments were repeated three times. All data are expressed as mean ± SEM, analyzed using one-or two-way analysis of variance and t-test using the statistical software NASDAQ: SPSS from SPSS corporation. A p-value less than 0.05 was considered statistically significant.

2 results and analysis

2.1 injection of FGF1 ^ΔHBS Causing severe bladder injury in normal mice and cynomolgus monkeys

To re-evaluate FGF1 ^ΔHBS Potential clinical utility of the invention, we performed a single dose toxicology study in healthy C57BL/6J mice intraperitoneally at 2.5mg/kg body weight for two weeks. Consistent with previous data disclosed by us (L.ZHao et al, paracrine-endocrine FGF chimeras as effective therapies for metabolic diseases. EBiomedicine 48,462-477 (2019)), FGF1 was administered ^ΔHBS The animals exhibited appetite suppression, resulting in severe weight loss (as shown in a and B in figure 1). Histomorphometric analysis showed administration of FGF1 ^ΔHBS The heart, liver, lung and kidney of the animals have no obvious pathological changes. However, we detected severe edema, inflammatory cell infiltration and thickening of the mucosal lamina propria (C in fig. 1) in the bladder tissue with the potential risk of inducing bladder cancer. Similar results were obtained in healthy cynomolgus monkeys (see D in figure 1). These toxic side effects significantly hinder FGF1 ^ΔHBS Can be used for treating type II diabetes (T2D).

2.2FGF1 ^ΔHBSΔNT Rational design and in vitro characterization of FGFR1c specific partial agonists

Recently, we reported that paracrine FGF4 also exerts potent hypoglycemic effects in mice by activating homologous FGFR1c in skeletal muscle, similar to FGF 1. However, in contrast to FGF1, FGF4 did not adversely affect appetite and food intake in mice. Notably, unlike FGF1, FGF4 interacts with all seven major FGFRs with versatility, regardless of alternative splicing, FGF4 does not recognize the b-splice subtype of FGFRs. Therefore, we speculate that the versatility of FGF1 may be the culprit in influencing animal appetite and food intake. As regards the oncogenic activity of FGF1 observed in the bladder, we conclude that FGF1 is present according to a threshold model ^ΔHBS May not be sufficiently impaired. In other words, FGF1 ^ΔHBS The variant may still be able to promote the formation of FGFR dimers long enough and stable to elicit the mitotic response necessary in bladder tissue.

Based on these considerations, we designed an N-terminal deleted FGF1 ^ΔHBS (abbreviated as FGF 1) ^ΔHBSΔNT ) It lacks the entire flexible N-terminal tail of FGF1 (residue Met1to Asn 22) and the immediately first two core residues (Try 23to Lys 24), and changes in the three heparin binding sites (Lys 127Asp, lys128Gln and Lys133 Val) (as shown in fig. 2). By comparison of X-ray crystallography analysis of complexes formed by four different FGF1 s with their receptors FGFR1c, FGFR2B, FGFR2c and FGFR3c, we found that the flexible N-terminal tail of FGF1 intervenes in the binding versatility of the receptor through highly degenerate interactions with different FGFR subtypes (see fig. 2B-E). In particular, FGF1 strictly binds FGFR2b, FGFR2c and FGFR3c depending on its N-terminal tail, however this region is largely unnecessary for FGFR1c binding. In contrast, the first two core residues (Try 23to Lys 24) have constant interactions with all FGFRs (as in fig. 2B-E). Thus, FGF1 ^ΔHBSΔNT Are predicted to be specific low affinity ligands for FGFR1 c.

To test whether this strategy would produce the desired effect, we first passed FGF1 ^WT And FGF1 ^ΔHBS As a pairFGF1 was analyzed by Surface Plasmon Resonance (SPR) spectroscopy ^ΔHBSΔNT Binding affinity for FGFR1c, FGFR2b, FGFR2c, FGFR3c and FGFR4. As a third control, we also made FGF1 variants (denoted FGF 1) ^ΔNT ) Which lacks residues 1-24 but contains the entire HS binding site (see fig. 2A). FGF1 despite impaired HS binding affinity ^ΔHBS With FGF1 ^WT Similar affinities multifunctional bind all five FGFR subtypes (see figure 2f, g and figure 3a, b). This is the expected result, since the HS and FGFR binding sites of FGF1 are structurally different (see fig. 2B). In contrast, FGF1 ^ΔNT And FGF1 ^ΔHBSΔNT The ability to bind FGFR2b, FGFR2c, FGFR3c and FGFR4 was lost while still binding to FGFR1c, although the binding affinity was attenuated by 3-fold (fig. 2h, i and fig. 3c, d). Thus, deletion of 1-24N-terminal residues will result in FGF1 ^ΔHBSΔNT Conversion to FGFR1c specific partial agonists.

2.3FGF1 ^ΔHBSΔNT Reduced FGFR dimerization capacity

Next, we compared the three FGF1 variants (FGF 1) by the proximity ligation technique (PLA) ^ΔNT ，FGF1 ^ΔHBS And FGF1 ^ΔHBSΔNT ) The ability to induce FGFR dimerization in L6 myoblasts stably expressing a single FGFR subtype. With FGF1 ^WT In contrast, FGF1 ^ΔHBS Showing a generally reduced ability to dimerize all six FGFRs in situ. In contrast, FGF1 ^ΔNT And FGF1 ^ΔHBSΔNT FGFR2b, FGFR2c, FGFR3b, FGFR3c and FGFR4 were not dimerized while the ability to dimerize FGFR1c was selectively retained (fig. 4a, b). Notably, these three variants showed a decreasing gradient of activity inducing in situ dimerization of FGFR1c, in the order FGF1 ^ΔNT >FGF1 ^ΔHBS >FGF1 ^ΔHBSΔNT (FIG. 4A, B). Thus, a decrease in binding affinity of HS to the receptor will further decrease FGF1 ^ΔHBSΔNT Receptor dimerization potential. . As can be seen from the PLA experiments, these quantitative data demonstrate that the order of progressive loss of dimerization potential of the variants is FGF1 ^ΔNT >FGF1 ^ΔHBS >FGF1 ^ΔHBSΔNT 。

Stably expressing six homologiesThe receptor activation assay data generated in the context of the L6 cell line for each of the FGFRs reflects the PLA assay data. In particular, with FGF1 ^WT In contrast, FGF1 was determined by tyrosine phosphorylation of two direct downstream aptamers of PLC γ 1 and FRS2 α, FGFRs ^ΔHBS The ability to activate all six FGFRs is generally diminished (fig. 4C). Similar to the PLA data, FGF1 ^ΔHBSΔNT Failure to activate FGFR2b, FGFR2C, FGFR3b, FGFR3C, and FGFR4 while selectively retaining a considerable ability to activate FGFR1C (fig. 4C). Notably, with FGF1 ^ΔHBS In contrast, FGF1 ^ΔHBSΔNT Further loss of ability to induce FGFR1C activation (fig. 4C). Taken together, our data show that deletion of 1-24N-terminal residues will result in FGF1 ^ΔHBSΔNT Conversion to a low affinity but specific ligand for FGFR1 c.

To confirm this conclusion in physiologically relevant environments, we next investigated FGF1 in vivo ^ΔHBS And FGF1 ^ΔHBSΔNT Receptor selectivity and signaling capacity. First, we analyzed the tissue distribution of FGFR subtypes (i.e., fgfr1C, fgfr2b-2C, fgfr3b-3C, and Fgfr 4) in C57BL/6J mice using quantitative RT-PCR. Consistent with previous reports, we found that WAT and liver predominantly expressed Fgfr1c and Fgfr4, respectively (fig. 5a, b). In contrast, skeletal muscle co-expresses Fgfr1C and Fgfr4 (FIG. 5C), while the bladder co-expresses Fgfr1C and Fgfr3b (FIG. 5D). Is injected into the abdominal cavity with FGF1 ^ΔHBS And FGF1 ^ΔHBSΔNT MAPK pathway activation analysis in mice showed that it was able to interact with FGF1 ^ΔHBS Different, FGF1 ^ΔHBSΔNT Failure to activate the MAPK pathway in the liver suggests FGF1 ^ΔHBSΔNT It could not act on FGFR4 (FIG. 5E, F). In the FGFR1c expressing tissues WAT, muscle and bladder, compared to FGF1 ^ΔHBS ，FGF1 ^ΔHBSΔNT The MAPK pathway was activated, although its capacity was significantly reduced (fig. 5e, g-I). Based on these results, we conclude that FGF1, unlike multifunctional ones ^ΔHBS ，FGF1 ^ΔHBSΔNT FGFR1c can be selectively bound and activated, although its ability is greatly reduced.

2.4 in mice and monkeys, compared to FGF1 ^ΔHBS ，FGF1 ^ΔHBSΔNT Has high safety.

Based on FGF1 ^ΔHBSΔNT For the unique binding specificity and weak dimerization capacity of FGFR1c, we predicted FGF1 ^ΔHBSΔNT Will be less toxic than FGF1, which is multifunctional ^ΔHBS A parent molecule. To validate this hypothesis, we performed a two-week single dose (2.5 mg/kg body weight) toxicology study in normal C57BL/6J mice. We have found that FGF1 ^ΔHBSΔNT Treatment did not cause significant histological changes in mouse bladder tissue, whereas FGF1 caused ^WT ，FGF1 ^ΔHBS And FGF1 ^ΔNT All caused severe edema, inflammatory cell infiltration and thickening of the inherent layer of bladder mucosa (fig. 6A). In addition, FGF1 by measurement of Proliferating Cell Nuclear Antigen (PCNA) and Ki67 levels ^ΔHBSΔNT Did not cause any proliferation in the bladder (FIGS. 6B-D), whereas FGF1 ^ΔNT Or FGF1 ^ΔHBS After treatment, the result was an increase in PCNA and Ki67 levels, although to a lesser extent than FGF1 ^WT (FIGS. 6B-D). More importantly, FGF1, unlike the parent molecule ^ΔHBSΔNT Treatment had no effect on appetite/food intake and body weight in C57BL/6J mice (fig. 6e, f). In addition, FGF1 is useful in acute and chronic settings ^ΔHBS And FGF1 ^ΔHBSΔNT Neither induced hypoglycemia (fig. 7).

By FGF1 ^ΔHBSΔNT The incentive for high safety performance in mice, followed by a one month toxicology study on healthy cynomolgus monkeys. FGF1 as assessed by histological analysis ^ΔHBSΔNT No pathological abnormalities were caused in the cynomolgus monkey bladder (fig. 6G) and the effect on appetite/food intake and body weight was also insignificant (fig. 6h, i). The evaluation of various other safety parameters such as body temperature, urine and serum biochemical markers, hemagglutination test and electrocardiogram and the like further proves that FGF1 ^ΔHBSΔNT High security (Table S2-4).

Table S2 rectal temperature (. Degree.C.) of Normal cynomolgus monkey during one month toxicology study

Urine examination during Table S3 Normal Macaca fascicularis one month toxicology study

LEU: (ii) a leukocyte; KET: a ketone; NIT: a nitrite salt; UBG: urobilinogen; BIL: bilirubin; and (2) PRO: urine protein; GLU: glucose; SG: specific gravity of urine; BLD: hematuria.

Change in serum biochemical parameters during Table S4 Normal cynomolgus monkey one month toxicology study

Note: * : p <0.05,. Times.0.01 compared to PBS treated group. ALT: alanine aminotransferase; AST: aspartate aminotransferase; ALP: alkaline phosphatase BUN: blood urea nitrogen; and (3) CREA: creatinine; TP: total protein; ALB: albumin; GLU: glucose; TCHO: total cholesterol; TBIL: total bilirubin; TG: a triglyceride; CK: creatine kinase; GGT: gamma-glutamyl transpeptidase; LDH: a lactate dehydrogenase; k: serum potassium ions; na: serum sodium ions; cl: serum chloride ion.

2.5FGF1 ^ΔHBSΔNT Remain with FGF1 ^ΔHBS Similar ability to regulate blood glucose

FGF1 has been identified ^ΔHBSΔNT Compared with FGF1 ^ΔHBS With ultra-high safety, we next tested FGF1 in commonly used T2D animal models db/db mice ^ΔHBSΔNT The ability to regulate blood glucose. With FGF1 ^ΔHBS Similarly, single administration of FGF1 ^ΔHBSΔNT (intraperitoneal injection, 0.5mg/kg body weight), 1 hourThe blood glucose level can be normalized. Normoglycemia lasted 6 hours, with blood glucose levels tending to rise over 6 hours, but still lower than the PBS-treated control (fig. 8A). FGF1 in an insulin resistant diet-induced obesity (DIO) mouse model ^ΔHBSΔNT A similar antihyperglycemic effect was exerted (fig. 8B). FGF1 in SD rats ^ΔHBSΔNT Shows a half-life of-1.496 + -0.29 hrs, similar to FGF1 ^ΔHBS (ii) half-life (. About.1.944. + -. 0.37 hrs) (Table S5).

Table S5 SD rat intraperitoneal injection FGF1 ^WT 、FGF1 ^ΔHBS And FGF1 ^△HBSΔNT Pharmacokinetics of (0.5 mg/kg body weight)

Recently, we have identified skeletal muscle as the main target organ for mediating potent antihyperglycemic activity of paracrine FGFs including FGF 1. More specifically, we found that the blood glucose regulating effect of FGF1 or FGF4 was dependent on up-regulation of GLUT4 (glucose transporter-4) expression and membrane translocation downstream of AMPK signaling that stimulated FGFR1c activation. Accordingly, we compared FGF1 ^ΔHBSΔNT And FGF1 ^ΔHBS The ability to stimulate glucose uptake in skeletal muscle. db/db FGF1 injection ^ΔHBSΔNT (Abdominal) or FGF1 ^ΔHBS Control, 2 hours later, 18F-fluoro-2-deoxy-2-D-glucose ([ 18F ]]FDG) trace amount. After 4 hours, glucose uptake in muscle was analyzed using positron emission tomography/computed tomography (PET-CT). With FGF1 ^ΔHBS Likewise, FGF1 ^ΔHBSΔNT Treatment resulted in 18F in the forelimb, hindlimb and back muscles]Enrichment of FDG (fig. 8C). Immunofluorescence and Western blot showing FGF1 ^ΔHBSΔNT And FGF1 ^ΔHBS It had an equivalent effect in promoting expression and membrane translocation of GLUT4 (fig. 8D-F). Likewise, FGF1, as determined by phosphorylation on Thr172 ^ΔHBSΔNT Having FGF1 in promoting AMPK alpha signaling ^ΔHBS Similar capabilities (fig. 8F). Phosphorylation of AMPK was not significant in insulin signaling as assessed by AKT phosphorylationIncreased (fig. 8F). This means that FGF1 is administered in an acute environment ^ΔHBS And FGF1 ^ΔHBSΔNT The effect on glycemic control is independent of insulin signaling. Pretreatment with potent inhibitors of the AMPK signaling pathway blocked GLUT4 expression promoting activity while reversing the hypoglycemic effects of both FGF1 variants (fig. 9a, b, d). These data are further confirmed by mice lacking AMPK α 2 (the predominant AMPK subtype in skeletal muscle). In these animals, FGF1 ^ΔHBS And FGF1 ^ΔHBSΔNT The acute hypoglycemic effect of (1) was severely impaired (fig. 9). At the same time, the FGFR1 kinase antagonist PD166866 also blocked the acute hypoglycemic effects of both FGF1 variants (fig. 9a, c, d). From the above, we conclude that FGF1 ^ΔHBSΔNT Retains complete hypoglycemic effect through FGFR1-AMPK signaling, without causing parent molecule FGF1 ^ΔHBS Adverse side effects.

2.6 Long-term administration of FGF1 ^ΔHBSΔNT Can relieve hyperglycemia and insulin resistance, and has no adverse reaction

Next we compared FGF1 in db/db mice in a long term (chronic) treatment setting ^△HBSΔNT And FGF1 ^ΔHBS Has anti-diabetic activity of (1). daily intraperitoneal injection of FGF1 in db/db mice ^△HBSΔNT And FGF1 ^ΔHBS And injected for 20 days. Long-term administration of FGF1 ^ΔHBSΔNT Leading to a sustained normoglycemia, which is associated with FGF1 ^ΔHBS The effect of the treatment was similar (fig. 10A). Intraperitoneal glucose tolerance test (ipGTT) after the last injection proves that FGF1 ^ΔHBSΔNT In the control of glucose with FGF1 ^ΔHBS Equally valid (FIG. 10B, C). In addition, insulin resistant DIO mice were chronically administered FGF1 ^ΔHBSΔNT Or FGF1 ^ΔHBS Can induce sustained hypoglycemic action similar to db/db mice (FIG. 10D) and demonstrate FGF1 by GTT, insulin resistance test (ITT) and plasma insulin level determination ^ΔHBSΔNT Significantly improved insulin sensitivity (FIG. 10E-I).

Consistent with the toxicological findings of healthy mice and monkeys, long-term administration of FGF1 ^ΔHBSΔNT The db/db mice and DIO mice had little or no effect on food intake and body weight (FIGS. 11A, B, D, E) and did not cause pathological changes in the bladder (FIG. 11C,f) In that respect We conclude that FGF1 lowers blood glucose levels and increases insulin sensitivity in acute or chronic settings ^ΔHBSΔNT With FGF1 ^ΔHBS Is as effective and does not cause FGF1 ^ΔHBS Adverse side effects.

2.7FGF1 ^ΔHBSΔNT Amelioration of insulin resistance by reducing accumulation of C-C chemokine ligand 2 (CCL 2) intervention by pro-inflammatory Adipose Tissue Macrophages (ATM)

We have recently found that chronic treatment of T2D mice with paracrine FGF1 or FGF4 improves insulin resistance by inhibiting Adipose Tissue Macrophage (ATM) infiltration and inflammation. In this study, FGF1 ^ΔHBS And FGF1 ^ΔHBSΔNT Plasma levels of tumor necrosis factor alpha (TNF α) and interleukin 6 (IL 6) (fig. 12a, b), as well as mRNA levels of adipose tissue F4/80 and Cd68 (macrophage infiltration marker) (fig. 12c, d) were significantly reduced in treated mice. Co-workers, immunofluorescence analysis showed, FGF1 ^ΔHBSΔNT Reduction in the number of F4/80 and CD11 c-positive macrophages with FGF1 ^ΔHBS As effective (fig. 12E). It is well known that CCL2 promotes obesity-induced insulin resistance by promoting infiltration of circulating monocytes and activation and proliferation of resident macrophages in adipose tissue, we measured the expression of C-C chemokine ligand 2 (CCL 2) in white adipose tissue (fig. 12F-G), and the results indicate FGF1 ^ΔHBSΔNT And FGF1 ^ΔHBS The level of CCL2 in white adipose tissue of treated DIO mice was significantly reduced (fig. 12F-G).

2.8FGF1 ^ΔHBSΔNT Can improve insulin resistance and liver function in T2D primate model, but does not cause adverse reaction

We next evaluated FGF1 ^ΔHBSΔNT The potential to treat idiopathic type 2 diabetes cynomolgus monkeys. Animals were injected subcutaneously with FGF1 at 0.15mg/kg daily ^ΔHBSΔNT Injection was performed for 4 weeks (fig. 13A). In the fasting state, FGF1 ^ΔHBSΔNT The mean decreased blood glucose levels by 23.5% (fig. 13B) and did not affect appetite and food intake (fig. 13C), although it slightly lost body weight (fig. 13D). Oral Glucose Tolerance Test (OGTT) showed FGF1 four weeks after injection treatment ^ΔHBSΔNT Shadow of blood sugarThe effect of the noise is most pronounced (fig. 13e, f). Serum insulin levels also tended to decrease, indicating an increase in insulin sensitivity (fig. 13G). In addition, FGF1 compared to Pre-treatment ^ΔHBSΔNT Serum levels of hemoglobin A1C (HbAIc) were also reduced (fig. 13H). FGF1 ^ΔHBSΔNT Serum Triglyceride (TG) and total cholesterol (T-CHO) levels were also reduced (fig. 13i, j), and serum levels of alanine Aminotransferase (ALT) and aspartate aminotransferase were significantly reduced, improving liver function (fig. 13k, l). Consistent with the results obtained with healthy mice and monkeys, spontaneously diabetic monkeys were injected with 4-week-continuous injections of FGF1 ^ΔHBSΔNT The level of clinical cancer biomarkers was not affected, which means that there was no carcinogenesis (Table S6). Therefore, we conclude that FGF1 ^ΔHBSΔNT Has important potential as a safe and effective molecule for treating T2D.

Table S6 FGF1 ^ΔHBSΔNT Serum levels of tumor biomarkers in diabetic cynomolgus monkeys before and after treatment

While embodiments of the invention have been described above, it is not limited to the applications set forth in the description and the embodiments, which are fully applicable in various fields of endeavor to which the invention pertains, and further modifications may readily be made by those skilled in the art, it being understood that the invention is not limited to the details shown and described herein without departing from the general concept defined by the appended claims and their equivalents.

Claims (8)

1. FGF1 variant, characterized in that the amino acid sequence of the FGF1 variant lacks the entire flexible N-terminal tail comprising the residues Met1to Asn22 and the immediately preceding two core residues comprising Try23to Lys24 compared to the amino acid sequence of wild type FGF1, and the three heparin binding sites (Lys 127Asp, lys128Gln and Lys133 Val) are altered.
2. 2. A polynucleotide encoding the FGF1 variant of claim 1.
3. 3. A vector or cell comprising the polynucleotide of claim 2.
4. 4. A pharmaceutical combination comprising the variant of claim 1 and a pharmaceutically acceptable carrier.
5. Use of a FGF1 variant for the preparation of a medicament for reducing FGF1 receptor binding versatility and reducing the carcinogenic effects of FGF 1.
6. Use of a fgf1 variant for the manufacture of a medicament for alleviating insulin resistance while not affecting appetite, body weight and bladder health.
7. Use of a fgf1 variant for the preparation of a medicament for lowering blood glucose without affecting appetite, body weight and bladder health.
8. Use of a fgf1 variant for the manufacture of a medicament for improving insulin resistance and improving liver function without affecting appetite, body weight and bladder health.

Images