CN115515969A - Application of FGF4 and Fgfr1 binding ligand in diabetes - Google Patents

Abstract

Fibroblast growth factor 4 (FGF 4) and Fgfr1 binding ligands are provided for use in the treatment of metabolic disorders, methods of treatment using fibroblast growth factor 4 (FGF 4) and Fgfr1 binding ligands, and use of fibroblast growth factor 4 (FGF 4) and Fgfr1 binding ligands in the manufacture of a medicament.

Description

Application of FGF4 and Fgfr1 binding ligand in diabetes

Technical Field

The present invention relates to FGF4 and Fgfr1 binding ligands for use in the treatment of metabolic diseases, methods of treatment using FGF4 and Fgfr1 binding ligands, and the use of FGF4 and Fgfr1 binding ligands in the manufacture of a medicament.

Background

The Fibroblast Growth Factor (FGF) family contains 22 structurally related polypeptides that act in an autocrine, paracrine or endocrine manner, mediating many key events in mammalian development, tissue homeostasis/repair, and body metabolism (Beenken and Mohammadi,2009 itoh and Ornitz, 2011. Over the past decades, there has been substantial evidence that endocrine members FGF19 and FGF21, as well as the paracrine member FGF1, have great potential for the treatment of obesity and type 2 diabetes (T2D) (degrolamo et al, 2016 fu et al, 2004 huang et al, 2017 jonker et al, 2012 lan et al, 2017. Notably, a single dose of FGF1 administered to the Central Nervous System (CNS) induces sustained diabetes remission in T2D rodents, in contrast to the transient effects seen following central administration of endocrine FGFs (Morton et al, 2013 perry et al, 2015 sarrouf et al, 2010 scarlet et al, 2016. However, the mechanism by which FGF1 induces this sustained action remains unclear. FGF1 administration was associated with transient reduction of food intake, stimulation of hepatic glucose uptake, and preservation of islet β cells (Scarlett et al, 2019, tennant et al, 2019. Studies also suggest that the assembly of monocytes, glial cells and neuronal networks (PNN) at the arcuate nucleus and adjacent intermediate bulge (ARC-ME) sites may play an important role in the sustained hypoglycemic effect of FGF1 (Alonge et al, 2020 bentsen et al, 2020. For example, central administration of FGF1 rapidly induces activation of MAPK-ERK signaling in monocytes and results in increased coverage of dendritic spines by processes from perisynaptic astrocytes in the ARC neurons of ob/ob mice (Bentsen et al, 2020, brown et al, 2019. Although not essential for initial anorexia and glucose lowering, melanocortin signaling and PNN within the ARC are partially essential for sustained action (Alonge et al, 2020. However, the key receptors and cellular mechanisms that mediate this sustained action remain unclear. Also, it is not known whether this surprising antidiabetic activity is unique to FGF1 in the FGF family. These gaps in knowledge have prevented the development of more effective methods for FGF-based treatment of T2D.

FGF mediates multiple biological functions by binding, dimerizing and thereby activating seven subtypes of homologous FGF receptors (FGFRs), each FGFR subtype having a unique spatio-temporal expression profile (Fon Tacer et al, 2010 goetz and Mohammadi,2013 kurosu et al, 2007. It has recently been found that different FGFs can modulate the intensity of FGFR dimerization to perform corresponding cellular functions in metabolic regulation or cell proliferation (Huang et al, 2017). Thus, it is speculated that other FGFs may exist that have anti-diabetic potential similar to or more pronounced than FGF1.

Fgfr1 is abundantly expressed in white adipose tissue and in the Central Nervous System (CNS) (Fon Tacer et al, 2010). Previous studies have shown that Fgfr1 is a key receptor target in white adipose tissue that mediates the regulation of FGF metabolism (Adams et al, 2012 ye et al, 2016. Fgfr1 in the CNS is also thought to be a receptor target mediating the regulation of FGF19 and FGF21 metabolism (Lan et al, 2017). However, it remains unclear whether Fgfr1 mediates the sustained hypoglycemic effect of FGF1 in the CNS. Glucose Sensing Neurons (GSNs) include glucose-stimulated neurons (GEs) and glucose-inhibited neurons (GIs). GSNs have been found to sense changes in extracellular glucose levels and provide feedback to regulate blood glucose (Anand et al, 1964, marty et al, 2007. Therefore, based on the important roles of Fgfr1 and GSN in metabolic regulation, it is hypothesized that FGF may exert its sustained anti-hyperglycemic effect by acting on GSN and by activating Fgfr1 on the relevant cells.

In this study, several representative members of the FGF paracrine family were screened to test their antihyperglycemic effect in the mouse db/db model of type 2 diabetes, and it was found that central administration of FGF4 sustains relief of hyperglycemia. Notably, FGF4 has significantly better anti-diabetic efficacy than FGF1 in diet-induced obese mice. Mechanistic studies suggest that FGF4 needs to mediate its lowering of glucose levels in T2D rodent models through Fgfr1 and GSN in the hypothalamus. Acute perfusion of FGF4 inhibits GI neurons, while chronic administration of FGF4 re-sensitizes GI and GE neurons and resets the ratio of GI to GE neurons. Notably, non-invasive nasal administration of FGF4 was able to induce sustained diabetes remission. Thus, the data indicate that FGF4 is a new biological agent with high transforming potential and can achieve long-lasting remission of diabetes.

Disclosure of Invention

The present invention relates to Fgfr1 ligands (e.g. FGF 4) for use in the treatment of metabolic disorders, methods of treatment using Fgfr1 ligands, and the use of Fgfr1 ligands in the manufacture of a medicament.

In a first aspect, the invention provides a method for treating a metabolic disease, the method comprising the step of administering to a patient in need thereof a therapeutically effective amount of a ligand of Fgfr1.

In a second aspect, the invention provides a ligand of Fgfr1 for use in the treatment of a metabolic disease.

In a third aspect, the invention provides the use of a ligand of Fgfr1 in the manufacture of a medicament for the treatment of a metabolic disorder.

Preferably, the Fgfr1 ligand is FGF4.

Preferably, the metabolic disorder is selected from the group consisting of obesity, diabetes, insulin resistance, hyperinsulinemia, glucose intolerance and hyperglycemia. In an embodiment, the diabetes is type 1 diabetes or type 2 diabetes, particularly type 2 diabetes.

The Fgfr1 ligand is isolated. The isolated Fgfr1 ligand can be comprised in a pharmaceutical composition. In an embodiment, FGF4 is the only active ingredient in the pharmaceutical composition.

Preferably, the Fgfr1 ligand or the pharmaceutical composition comprising the Fgfr1 ligand is administered intracerebrally or intranasally. Thus, the pharmaceutical composition is a pharmaceutical composition suitable for intracerebral administration or intranasal administration, and the medicament of the third aspect is a medicament suitable for intracerebral administration or intranasal administration.

Preferably, the intracerebral administration is chronic intracerebral administration.

The pharmaceutical composition may further comprise pharmaceutically acceptable formulating agents such as carriers, adjuvants, solubilizers, stabilizers and/or antioxidants. In an embodiment, the pharmaceutical composition further comprises liposomes suitable for intranasal administration.

In embodiments, the frequency of intracerebral administration is less than once every 2 weeks, preferably less than once every 5 weeks, for example once every 6 to 16 weeks.

In another embodiment, the frequency of intranasal administration is less than once every 2 days, preferably less than once every 5 days, for example, once every 6 to 9 days.

Drawings

Figure 1 shows that a single injection of FGF4 into the CNS reduces hyperglycemia in diabetic mice.

(A) Representative FGFs are from each subfamily.

(B-C) changes in blood glucose (B) and body weight (C) following a single icv administration of paracrine FGF (3.0 μ g each) to db/db mice.

(D-E) changes in blood glucose (left) and area under the curve (AUC) (right) in db/db mice subjected to ipGTT 7 days (D) or 50 days (E) after icv FGF1 or FGF4 (3.0 μ g), as indicated.

(F) By intravenous injection ¹⁸ Representative micro-PET/CT images (left) and quantitative analysis (right) were performed 30 minutes after F-FDG in db/db mice that received icv FGF1 or FGF4 (3.0. Mu.g) 3 days before.

(G) Expression of mRNA encoding glucose regulatory enzymes in the liver of db/db mice 7 days after icv FGF4 (3.0. Mu.g).

(H) 7 days after icv FGF4 (3.0. Mu.g), glycogen in the liver and skeletal muscle of db/db mice was stained with periodic acid-Schiff (PAS). Scale bar, 100 μm.

(I) Change in blood glucose after a single icv injection of FGF4 (3.0. Mu.g) in DIO mice.

(J) Blood glucose (left) and AUC (right) changes in DIO mice subjected to ipGTT 7 days post icv FGF4 (3.0 μ g).

(K) Change in body weight after a single dose of icv FGF4 (3.0. Mu.g) in DIO mice.

(L-M) changes in physical Activity (L) and thermogenesis (M) in DIO mice 7 days after icv FGF4 (3.0. Mu.g).

Data are shown as mean ± s.e.m,. P ≦ 0.05 for the corresponding control group; * P is less than or equal to 0.01; * P is less than or equal to 0.001; compared with FGF1 group, # p is less than or equal to 0.05; # p is not more than 0.001,ns, not significant.

Figure 2 shows that FGF4 requires Fgfr1 in hypothalamic neurons to exert anti-hyperglycemic effects.

(A-C) Fgfr1 is expressed by the hypothalamus in which mRNA hybridizes in situ. Scale bar, 1mm (A) and 500 μm (B, C).

(D) For injection of AAV8-CMV-creGFP into Fgfr1 ^flox/flox A hypothalamic Fgfr1 knockout protocol was generated in mice.

(E) Hypothalamic expression of AAV 8-CMV-creGFP. Scale bar, 200 μm.

(F)Fgfr1 ^CMV-Cre The blood glucose of the mice was significantly increased. Virus was injected at t = 0.

(G) In Fgfr1 ^CMV-Cre Blocking the antihyperglycemic effect of FGF4 in mice.

(H) Designed by injecting AAV8-hSyn-Cre into Fgfr1 ^flox/flox Fgfr1 was knocked out in hypothalamic neurons in mice.

(I-J) Fgfr1 under different dietary conditions (Normal food or HFD) ^hSyn-Cre Changes in blood glucose (I) and body weight (J) in mice and GFP control.

(K-N)Fgfr1 ^hSyn-Cre Changes in blood glucose (K, L) and body weight (M, N) after icv FGF4 (3.0 μ g) in mice (K, M) and GFP control (L, N) mice.

(O-P) changes in blood glucose (O) and body weight (P) after FGF4 (1.5. Mu.g) was injected into VMH.

(Q) change in blood glucose (left) and AUC (right) in DIO mice subjected to ipGTT 24 hours after FGF4 (1.5 μ g) was injected into VMH.

VMH, ventral hypothalamus; DMH, dorsal hypothalamus; LH, lateral hypothalamus; ARC, hypothalamic arcuate nucleus; 3V, third ventricle. Data are shown as mean ± s.e.m,. P ≦ 0.05 for the corresponding control group; * P is less than or equal to 0.01; * P.ltoreq.0.001,ns, not significant.

Figure 3 shows that FGF4 restores the function of VMH GSN in diabetic mice.

(A-B) representative firing rates associated with VMH GE neurons (A) and GI neurons (B) from wt, db/db and db + FGF4 mice respond to changes in low glucose (0.1 mM). db + FGF4 means db/db mice injected with FGF4 (3.0. Mu.g, 7 to 10 days post-icv).

(C) Relative percentages of GSN cell types in wt, db/db and db + FGF4 mice. INS, glucose insensitive.

(D-F) GE neurons respond to mean firing rate change (D), peak firing rate change (E) and peak time (F) of 0.1mM glucose.

(G-I) mean firing rate change (G), peak firing rate change (H), and peak time (I) of GI neurons in response to 0.1mM glucose.

(J) The experiment was designed to test the acute effect of FGF4 on neurons in db/db mice.

(K) Representative traces show inhibition of GI neurons by FGF4 perfusion.

(L) mean firing rate changes associated with GI neurons and their percentage of perfusion by FGF4.

(M) the assay was designed to test the effect of blocking Fgfr1 by PD 166866.

(N) representative trace shows that D166866 inhibits FGF 4-inhibited GI neurons.

(O) mean firing rate change and its percentage in GI neurons treated with FGF4 plus PD 166866.

(P) Central administration of GI inhibitors to lower blood glucose in db/db mice.

(Q) the experiment was designed to test the acute effect of FGF4 on neurons in DIO mice.

(R) representative traces show that FGF4 perfusion inhibits GI neurons.

(S) mean firing rate changes of GI neurons and their percentage perfused by FGF4.

(T) the experiment was designed to test the effect of PD166866 on GI neurons stimulated or inhibited by FGF4.

(U) the Fgfr1 inhibitor partially abrogates the effect of FGF4 on GI neurons.

(V) mean firing rate change and percentage thereof in FGF 4-inhibited GI neurons treated with PD 166866.

Data are shown as mean ± s.e.m,. P ≦ 0.05 for the corresponding group; * P is less than or equal to 0.01; * P is less than or equal to 0.001.

Fig. 4 shows that nasal delivery of FGF4 to the brain relieves hyperglycemia.

(A) Representative pictures of liposome nanoparticles loaded with gelatin solution (control) and FGF 4-gelatin solution (FGF 4). Scale bar, 100nm.

(B) Distribution of liposome nanoparticle diameters.

(C-D) changes in blood glucose (C) and body weight (D) of db/db mice following intranasal delivery of FGF4 (30 μ g).

(E-F) changes in blood glucose (left) and AUC (right) in db/db mice subjected to ipGTT (E) or ipITT (F) 24 hours after intranasal FGF4 delivery.

(G) Blood glucose in db/db mice after repeated intranasal delivery of FGF4.

Data are shown as mean ± s.e.m,. P.ltoreq.0.05 compared to the corresponding control group; * P is less than or equal to 0.01; * P is less than or equal to 0.001.

FIG. S1 shows the phenotypes in wt and db/db mice after FGF injection, which are related to FIG. 1.

(A) There was no significant change in blood glucose following a single icv administration of endocrine FGF21 (3.0 μ g) in db/db mice.

(B-C) blood glucose (B) and body weight (C) in response to increasing FGF4 dose.

(D) icv FGF4 (3.0 μ g) indicates the change in blood glucose (left) and AUC (right) in db/db mice subjected to ipGTT 7 days after dose.

(E) db/db mice had no decrease in blood glucose after ip injection of FGF4 (3.0. Mu.g).

(F) Transient decrease in blood glucose after ip injection of FGF4 (30. Mu.g).

(G) Change in blood glucose after icv FGF4 (3.0 μ g) in wt mice.

(H) Blood glucose (left) and area under the curve (AUC) (right) changes in wt mice subjected to ipGTT 7 days post icv FGF4 (3.0 μ g).

(I) Change in body weight after a single icv administration of FGF4 (3.0. Mu.g) to wt mice.

(J-O) changes in plasma hormone (J), food intake (K), body fat content (L), physical activity level (M), respiratory exchange ratio (RER, N) and thermogenesis (O) in db/db mice on the indicated days following injection of icv FGF4 (3.0 μ g). The 12-light-dim light cycle is indicated.

(P) blood glucose (left) and AUC (right) changes in db/db mice subjected to ipITT 10 days after icv FGF4 (3.0 μ g).

(Q) representative micro-PET/CT images and intravenous injection ¹⁸ Quantitative analysis 15 min (left) and 60 min (right) after F-FDG was performed in db/db mice that had received icv FGF1 or FGF4 (3.0. Mu.g) 3 days ago.

Data are shown as mean ± s.e.m,. P.ltoreq.0.05 compared to the corresponding control group; * P is less than or equal to 0.01; * P is less than or equal to 0.001; compared with FGF4 (0.3. Mu.g) group, # # # p ≦ 0.001,ns, not significant.

Figure S2 shows the phenotype in DIO mice after FGF injection, which correlates with figure 1.

(A-B) changes in blood glucose (A) and body weight (B) following a single injection of FGF1 (3.0. Mu.g) in DIO mice.

(C) Blood glucose (left) and AUC (right) changes were made 7 days after icv FGF4 (3.0 μ g) in ipott DIO mice.

(D-F) Change in body fat content (D), food intake (E) and RER (F) after a single icv FGF4 (3.0 μ g) in DIO mice. Days post injection and 12 hours light-dim light cycle are indicated.

(G) Transient reduction of blood glucose after ip injection of 30 μ g FGF4 in DIO mice.

Data are shown as mean ± s.e.m,. P ≦ 0.05 for the corresponding control group; * P is less than or equal to 0.001.

Figure S3 shows that hypothalamic Fgfr1 is required for FGF 4-mediated glucose lowering, in relation to figure 2.

(A) The hypothalamus expresses Fgfr1. Scale bar, 1mm (left); 100 μm (right).

(B)Fgfr1 ⁺ The proportion of cells, expressed as the number of Fgfr1+ cells divided by all cells from wt mice (Dapi +).

(C) Relative expression of Fgfr1 in different hypothalamic regions was normalized to VMH (n =8 slides from 5 mice).

(D) For injection of AAV8-CMV-CreGFP into Fgfr1 ^flox/flox Protocol for generating hypothalamic Fgfr1 knockdown in mice.

(E) AAV8-CMV-creGFP is expressed in the hypothalamus of different brain regions, brain region B.

(F)GFP ⁺ Cells and all cells from the indicated hypothalamic nuclei (Dapi) ⁺ ) The ratio of (a) to (b).

(G) Reduction of Fgfr1 mRNA after CMV-Cre expression.

(H) Weight change in CMV-GFP and CMV-Cre mice.

(I) Blood glucose (left) and AUC (right) of CMV-Cre mice subjected to ipGTT 6 weeks after virus injection.

(J) Change in body weight of CMV-cre mice after icv FGF4 (3.0. Mu.g).

(K-L) change in blood glucose (K) and body weight (L) in CMV-GFP mice after icv FGF4 (3.0. Mu.g).

Data are shown as mean ± s.e.m,. P ≦ 0.05 for the corresponding control group; * P is less than or equal to 0.01; * P is less than or equal to 0.001.

FIG. S4 shows a graph at Lepr ⁺ And lack of Fgfr1 in astrocytes but none block the hypoglycemic effect of FGF4.

(A-B) Lepr changes in body weight (A) and blood glucose (B) in CKO mice.

(C-D) Lepr that experienced either ipGTT (C) or ipITT (D) changes in blood glucose (left) and AUC (right) in CKO mice (14 weeks of age).

(E) Lepr weight change in CKO mice under HFD.

(F) Lepr CKO mice blood glucose at HFD at week 12.

(G-H) Lepr changes in blood glucose (G) and body weight (H) in CKO mice after icv FGF4 (3.0. Mu.g).

(I-J) Gfap: weight change in CKO mice under diet food (I) or HFD (J).

(K) Gfap CKO mice blood glucose at HFD at week 12.

(L-M) Gfap changes in blood glucose (L) and body weight (M) of CKO mice after icv FGF4 (3.0. Mu.g).

Data are shown as mean ± s.e.m,. P.ltoreq.0.05 compared to the corresponding control group; * P is less than or equal to 0.01; * P is less than or equal to 0.001; ns, not significant.

Figure S5 shows the phenotype associated with figure 2 following Fgfr1 delay in hypothalamic neurons.

(A) By injecting AAV8-hSyn-CreGFP into Fgfr1 ^flox/flox Fgfr1 knock-outs were designed in mice in hypothalamic neurons.

(B-C) changes in body weight (B) and blood glucose (C) of syn-GFP and syn-cre mice.

(D) AAV8-hSyn-CreGFP is expressed in the hypothalamus of different brain regions, brain region B.

(E) Expression of Fgfr1 in Syn-GFP and Syn-Cre mice.

(F) Reduction of Fgfr1 expression levels following Syn-Cre expression.

Data are shown as mean ± s.e.m,. P ≦ 0.01 for the corresponding control group; * P is less than or equal to 0.001.

FIG. S6 shows the effect of FGF4 on VMH GE neurons from db/db mice, in relation to FIG. 3.

(A) The experiments were designed to test the acute effect of FGF4 on neurons in db/db mice.

(B-C) shows representative traces of either excitation (B) or inhibition (C) of GE neurons perfused by FGF4.

(D) The mean firing rate change associated with VMH GE neurons and their percentage of perfusion by FGF4.

(E) Schematic diagram showing the experimental design to test the effect of the Fgfr1 inhibitor PD 166866.

(F-G) mean firing rate changes in FGF 4-stimulated (F) or FGF 4-inhibited (G) GE neurons treated with PD 166866.

ACSF, artificial cerebrospinal fluid. Data are shown as mean ± s.e.m.

Figure S7 shows the effect of FGF4 on VMH GE neurons from DIO mice, in relation to figure 3.

(A) The experiment was designed to test the response of GSN to high (5.0 mM) glucose and low (0.1 mM) glucose.

(B-D) mean firing rate changes associated with GE (B) neurons, GI (C) neurons, and INS (D) neurons by treatment with 5.0mM glucose and 0.1mM glucose.

(E) The experiment was designed to test the acute effect of FGF4 on brain sections.

(F) Representative traces show the firing, inhibition, and insensitivity of GE neurons perfused by FGF4.

(G) Mean firing rate changes associated with GE neurons and their percentage of perfusion by FGF4.

(H) The experiment was designed to test the effect of the Fgfr1 inhibitor PD166866 on GE neurons.

(I) PD166866 partially blocks the FGF 4-stimulated and inhibited GE neurons.

(J) Change in mean firing rate and percentage of FGF 4-stimulated GE neurons treated with PD 166866.

Data are shown as mean ± s.e.m.

Detailed Description

As used herein, the terms "a" or "an" mean one or more, unless expressly specified otherwise.

The term "isolated" molecule refers to a molecule that is substantially free of any other contaminating molecules or other contaminants that are found in its natural environment and which would interfere with its use in production or its therapeutic, diagnostic, prophylactic or research use. Thus, typically an isolated polypeptide of the invention is a purified polypeptide and free of pathogens and pyrogens, having a purity of at least 80%, preferably at least 90%, more preferably at least 95%, in particular having a pharmaceutical purity, i.e. at least 98%. Preferably, the isolated polypeptide of the invention may be substantially free of other polypeptides, in particular polypeptides from animals.

The term "FGF4", an abbreviation for fibroblast growth factor 4, means full length FGF4 or an active fragment thereof, e.g. a fragment from Ala67 to Leu 206. Preferably, FGF4 is human FGF4. Human FGF4 is a 176 AA long protein derived by cleavage of the N-terminal 30 AA of the precursor protein. FGF-4 contains a single N-linked glycosylation site. The unglycosylated FGF-4 is a polypeptide truncated by two NH 2-terminal residues with better heparin affinity and biological activity than the wild-type protein.

Recombinant FGF4 protein is commercially available (e.g., from Sigma Aldrich, where recombinant FGF4 protein is produced in baculovirus; or from Invitrogen, where recombinant FGF4 protein is produced in E.coli).

The term "pharmaceutically acceptable carrier" as used herein refers to one or more formulation materials suitable to effect or enhance FGF4 delivery.

The term "therapeutically effective amount" refers to an amount of FGF4 to support observable levels of one or more biological activities, such as the ability to lower blood glucose, insulin, triglyceride, or cholesterol levels; reducing body weight; or improve glucose tolerance, energy expenditure or insulin sensitivity. In the present invention, the "patient" is preferably a human, but may also be an animal, preferably a mammal such as a dog, pig, cow, horse, rat, mouse, guinea pig, and the like.

For human subjects, "obesity" can be defined as a body weight that is more than 20% higher than the ideal body weight for a given population (r.h. williams. Endocrinology textbook 1974, pages 904-906).

"diabetes" is characterized by excessive urination and persistent thirst and includes both types. Type II diabetes is characterized by excessive glucose production despite the availability of insulin, and the body's glucose levels remain too high due to insufficient glucose clearance. In this embodiment, the diabetes may be type 1 diabetes, and the effect of FGF4 on the treatment of type 2 diabetes is believed to be much stronger than the effect of type 1 diabetes. The present inventors have found that FGF4 is more beneficial for the treatment of type 2 diabetes.

The term "insulin resistance" refers to a state where a normal amount of insulin produces a subnormal biological response.

The term "hyperinsulinemia" is defined as higher than normal levels of insulin in the blood.

The term "glucose intolerance" as used herein is defined as an abnormal sensitivity to glucose.

The term "hyperglycemia" is defined as excess sugar (glucose) in the blood.

The pharmaceutical composition may comprise one or more pharmaceutically acceptable formulations for modifying, maintaining or preserving, for example, pH, permeability, viscosity, clarity, color, isotope, odor, sterility, stability, dissolution or release rate, adsorption or permeation of the composition. Suitable agents include, but are not limited to, amino acids (such as glycine, glutamine, aspartic acid, arginine, or lysine), antimicrobials, antioxidants (such as ascorbic acid, sodium sulfite, or sodium bisulfite), buffers (such as borates, bicarbonates, tris-HCl, citric acid, phosphates, or other organic acids), co-raw materials (such as mannitol or glycine), chelating agents (such as ethylenediaminetetraacetic acid (EDTA)), complexing agents (such as caffeine, polyvinylpyrrolidone, beta-cyclodextrin, or hydroxypropyl beta-cyclodextrin), fillers, monosaccharides, disaccharides, and other carbohydrates (such as glucose, mannose, or dextrin), proteins (such as serum albumin, gelatin, or immunoglobulin), colorants, flavors and diluents, emulsifiers, hydrophilic polymers (such as polyvinylpyrrolidone), low molecular weight polypeptides, salt-forming counter-ions (such as sodium), preservatives (such as phenylmethanoxy chloride, benzoic acid, salicylic acid, thimerosal, phenylethyl alcohol, methyl terephthalic acid, propyl terephthalic acid, chlorhexidine, sorbic acid, or hydrogen peroxide), solvents (such as glycerol, propylene glycol, or polyethylene glycol), sugar alcohol (such as mannitol or sorbitol), suspending agents (such as suspending agents, suspending agents or suspending agents; PEG, sorbitan esters, polysorbates such as polysorbate 20 or polysorbate 80, triton, trimethylamine, lecithin, cholesterol or tyloxapol), stability sensitizers (such as sucrose or sorbitol), sensitizers (such as alkali metal halides-preferably sodium or potassium chloride or mannitol sorbitol), delivery vehicles, diluents, adjuvants and/or pharmaceutical adjuvants (pharmaceutical science of remgton, a.r. gennaro, editors, mikawa).

The primary vehicle or carrier in the pharmaceutical composition may be aqueous or non-aqueous. For example, a suitable vehicle or carrier for injection may be water or a physiological saline solution.

The pharmaceutical compositions of the present invention may be administered by any route known to those skilled in the art, such as oral administration, rectal administration, sublingual administration, intrapulmonary administration, intraperitoneal administration, transdermal administration, iontophoretic administration, vaginal administration, intracerebral administration, and intranasal administration. However, intraperitoneal administration of large doses of FGF4 only resulted in a temporary reduction of blood glucose levels in db/db mice. Preferably, the pharmaceutical composition of the invention is administered by intracerebral administration or intranasally, according to the mechanism discovered by the inventors.

The pharmaceutical compositions for intranasal administration of the present invention may be in the form of liposomes. In embodiments, FGF4 is loaded into a 100nm liposome.

A single intracerebral or intranasal administration of FGF4 may result in a sustained chronic hypoglycemic effect. The present inventors found that a single intracerebral administration of FGF4 might even produce a hypoglycemic effect lasting more than 4 months. At least the frequency of intracerebral administration may be less than once every 2 weeks, preferably less than once every 5 weeks, for example, once every 6 to 16 weeks. In addition, the frequency of intranasal administration may be less than once every 2 days, preferably less than once every 5 days, for example, once every 6 to 9 days.

The dosage administered will vary depending on the formulation, the desired time course, and the patient to be treated, and the physician can determine the feasible dosage in treatment based on the actual circumstances (e.g., the condition, weight, etc. of the patient). For an average adult, the dosage of the pharmaceutical composition of the present invention may be 1ng to 10mg of FGF4 or an variant thereof per kg of body weight of the adult. For the injection route, the dose is preferably 10ng to 1mg per kg body weight, more preferably 100ng to 100ug per kg body weight, still more preferably 1ug to 10ug such as 2ug, 5ug, 7ug or 8ug per kg body weight.

Publications cited in this application are provided herein for purposes of illustration and are incorporated herein by reference as if fully set forth herein.

For a better understanding of the present invention, it will now be described in more detail with reference to specific examples. It should be noted that these examples are merely illustrative of the present invention and should not be construed as limiting the scope of the present invention. Various modifications and alterations of this invention will become apparent to those skilled in the art from this description.

Examples of the invention

The following examples are illustrative of embodiments of the present invention and various uses thereof. The following examples are for illustrative purposes only and should not be construed as limiting the scope of the invention in any way. The materials used in the examples are commercially available.

EXAMPLE 1 materials and methods

Experimental model and subject details

Mouse model

Adult Male db/db (B6. BKS (D) -Lepr) ^db /J,8-12 weeks old) was purchased from seikagaku limited company (nanjing, china). C57BL/6J (wt) mice were from Chighui laboratory animal feeding Co., ltd. Fgfr1 ^flox/flox Mice (accession number 007671) were obtained from Jackson laboratories. By Fgfr1 ^flox/flox Mice and Lepr ^Cre/+ Mice (# 008320, jackson laboratory) and Gfap ^Cre/+ Mice (gifts from J.W.Zhou) were crossed to generate Fgfr1 conditional knockout mice (Lepr: CKO, lepr stands for leptin receptor; GFAP: CKO, GFAP stands for glial fibrillary acidic protein). Mice were housed under constant temperature and humidity in a 12 hour light/12 hour dark cycle (light, 8PM-8 AM). Mice were randomly assigned to experimental groups and water and either standard laboratory food or a high fat diet (HFD; D12492, high fat diet, usa) were ad libitum (ad libitum) obtained unless otherwise stated. All animal procedures were performed strictly following animal studies: in vivo experimental guidelines are reported. Care and use of animals is provided by Shanghai university of science and technology, shanghai model animals Co., ltd, zhang JiangLaboratory national protein science facilities animal facilities and government regulations review and approval.

Details of the method

Expression constructs and recombinant protein production

Human FGF1 (Met 1-Asp 155), FGF4 (Ala 67-Leu 206), FGF7 (Cys 32-Thr 194), FGF8a (Gln 23-Arg 204), FGF20 (Met 1-Thr 211) and FGF21 (His 29-Ser 209) were expressed and purified according to published protocols (benken et al, 2012, bellosta et al, 2001. Briefly, a cDNA fragment encoding FGF4 (Ala 67-Leu 206) was PCR amplified and subcloned into pET-15b at the NcoI site and the Xhol site, respectively. Coli BL-21 (DE 3) cells transformed with the FGF4 construct were cultured in LB medium (2% glucose and 30kg/mL kanamycin) at 37 ℃ and 200rpm in an incubation shake flask. Cells were induced at an optical density of 0.8-1.0 (measured as 600) with 1.0mM isopropyl-L-thio-D-galactopyranoside (IPTG) for 4 hours at 37 ℃. Then, the cells were collected and lysed in 25mM Na/K phosphate buffer (pH 7.5) containing 300mM NaCl using an Emulsiflex-C3 (Avena protein, ontario, canada) high volume homogenizer. The lysates were clarified by centrifugation at 20,000rpm for 30 min at 4 ℃ and the soluble recombinant proteins were purified by a heparin affinity column (Hitrap heparin HP, GE Healthcare, piscataway, NJ, column volume 5 ml) eluting with a linear gradient of NaCl (0M-2.0M) in HEPES (25mm, ph 7.5). Purity of the protein of interest will be determined by 12% SDS-PAGE analysis; further, the fractions were combined, concentrated and further purified using gel molecular sieves (Superdex TM-75GE healthcare, piscataway, NJ) in a mobile phase buffer (1.0M NaCl,25mM Tris-HCl, pH 8.0). The purity of the recombinant protein was estimated to be > 98%. Other FGFs were prepared in a similar manner.

Brain stereotaxic injection

FGF (. About.3.0. Mu.g, unless otherwise specified) or AAV virus (. About.200 nl) was delivered by a drawn glass pipette and pen-type oil pressure syringe with custom designed robot (MO-10, tritech research, USA). Injection coordinates were calculated from Paxinos & Franklin mouse brain coordinates (3 rd edition) by injection under isoflurane general anesthesia using a small animal stereotaxic instrument (David Kopf instrument, # PF-3983 rwd life sciences, #68030; qiano family nanjing biotechnology limited, # SH 01A). The feedback heater is used to keep the mouse warm during surgery. Mice were recovered in warm racks before being transferred to containment cages. The coordinates are as follows: 1) For intraventricular (i.c.v.) administration, lateral Ventricle (LV): bregma is-0.40 mm behind; 1.10mm on the side, 2.50mm below the skull surface; 2) For hypothalamic delivery, ventral hypothalamic nuclei (VMH): bregma is-1.65 mm; 0.40mm lateral, 5.70mm below the skull surface, anatomy is shown directly on the relevant figures. For pharmacological studies, cannulas of mice (RWD life sciences) were placed in the LV.

Physiological measurement

Glucose levels were determined by a hand-held glucometer (agco-dukele golgi glucometer, roche, switzerland). Unless otherwise stated, blood samples were collected from the tail vein of mice fed ad libitum. In the food intake assay, mice were implanted with cannulae in the lateral ventricle (RWD life sciences, 62002, 62102) and the cannulae were fixed to the skull with dental cement (C & B metabolic rapid adhesive cement system). All intubated mice were individually housed for 7-10 days post-surgery and recovered with food ad libitum and water. FGF4 (3.0 μ g) or saline was delivered to the mice via cannula and the food was weighed daily. Body composition was measured using quantitative magnetic resonance spectroscopy (minisec body composition analyzer LF50, bruker, karlsruhe, german). Heat production, physical activity and RER (respiratory exchange rate) were monitored with a comprehensive laboratory animal monitoring system (CLAMS, columbus instruments). For plasma analysis, samples were collected from mice fed ad libitum. Plasma insulin, corticosterone, ACTH, epinephrine, and non-esterified free fatty acid (NEFA) concentrations were determined by the Elisa kit according to the manufacturer's instructions (# E0448m, # E0540Ge, # E0836m, # E0858Ge from EIAab, wuhan, china, # a042-2-1 from the institute of bioengineering, tokyo, china).

Intraperitoneal glucose tolerance test (ipGTT) and intraperitoneal insulin tolerance test (ipITT)

ipGTTs were performed in mice overnight after intraperitoneal injection of 1.0g/kg body weight of glucose (20% glucose) followed by fasting for 16 hours. ipITT was performed in fasted 6-hour mice by injection of 0.75 units/kg body weight (for wt mice) or 1.0 units/kg (for db/db and HFD mice) insulin (# 40112es25, yeasenl2016, china). Blood samples were collected from the tail vein at the indicated times.

Microelectronic emission computed tomography (micro PET-CT) imaging and analysis

MicroPET-CT analysis was performed in db/db mice 3 days after icv injection of FGF (. About.3.0. Mu.g). Receiving ¹⁸ Overnight fasted mice of F-FDG (259.1 + -17.76 μ Ci, i.v.) were transiently anesthetized with 1.5% isoflurane in oxygen at 2L/min and scanned in Inveon preclinical multimodal PET-CT (Siemens, munich, germany) for 1 hour. The image is reconstructed using a three-dimensional ordered subset expectation maximization (3 DOSEM) algorithm and then using a Maximum A Posteriori (MAP). Under the guidance of CT images, a 3D region of interest (ROI) is rendered on the brain, liver and muscles. The radioactivity of the corresponding tissues was measured using the Inveon research workplace (Siemens, munich, germany).

AAV vectors

AAV vectors carrying specific genes (titre > 10) ¹² ) Purchased from Taitol Bioscience (Shanghai, china), including AAV2/8-CMV _ bGI-Cre-EGFP-pA, AAV2/8-CMV _ bGI-EGFP-WPRE-pA, AAV2/8-hSyn-Cre-EGFP-WPRE-pA, AAV2/8-hSyn-EGFP-WPRE-pA.

Reverse transcription and quantitative real-time PCR

Total RNA was extracted from hypothalamus or liver using Trizol reagent (Invitrogen, usa). Reverse transcription PCR was performed using PrimeScript RT kit (RR 047A, takara, japan), and ViiA was performed using SYBR PCR kit (RR 820A, takara, japan) ^TM 7 real-time PCR System (applied biosystems, USA), real-time PCR was determined according to the manufacturer's instructions. Actin was used as an internal control for real-time PCR amplification.

Immunohistochemistry and RNA in situ hybridization

Mice were anesthetized and perfused with PBS followed by 10% formalin. Brains were excised, fixed overnight and cut into-50 μm sections. After collection and three washes with PBST (0.3% Triton X-100 in PBS), cryostat sections were incubated withBlocking solution (PBS containing 2.5% goat serum and 0.3% Triton X-100) was incubated together, then treated with primary antibody (chicken anti-GFP, ab13970, abcam,1 1000) overnight at 4 ℃, and then the samples were incubated with secondary antibody (goat anti-rabbit Alexa 488 secondary antibody, invitrogen, a 11039) for 2 hours at room temperature. Images were captured on a Zeiss LSM800 confocal microscope or Olympus VS120 virtual microscope slide scanning system. According to the manufacturer's instructions, use

2.5HD kit (# 322350, advanced cell diagnostics) for Fgfr1 RNA in situ detection. The Fgfr1 probe was designed by advanced cytodiagnostics and the DapB probe was used as a negative control.

Electrophysiological recording slice

Briefly, hypothalamic sections containing VMH regions were prepared from adult males wt, db/db and db/db with icv FGF4. Oxygen content in ice-cold condition (95% ₂ And 5% of CO ₂ ) Before cardiac perfusion, mice were anesthetized with sodium pentobarbital (100 mg/kg i.p.), a high sucrose solution consisting of (in mM): 213 sucrose, 26NaHCO ₃ 、1.25NaH ₂ PO ₄ 、2.0Na ₂ HPO ₄ 2.5KCl and 2MgSO ₄ (osmolality adjusted to 300mOsm/kg, pH 7.4). Brains were removed quickly and placed in ice-cold oxygenated high sucrose solution. The coronal hypothalamic sections (250 μm) were cut with a vibrating microtome (VT 1200S, leica microsystems, germany). Sections were incubated at 31 ℃ in oxygenated artificial cerebrospinal fl fluid (ACSF in mM:124NaCl,5KCl, 1.24KH) prior to transfer to the recording chamber ₂ PO ₄ ,1.3MgSO ₄ ,26NaHCO ₃ ,2.4CaCl ₂ and 2.5glucose, pH 7.2-7.4,300-310 mOsm/kg)>For 1 hour. ACSF was perfused at 1.2 ml/min. The hypothalamic sections were observed by infrared differential interference contrast (ID-DIC) video microscopy using an upright microscope (AxioDetector, zeiss, germany) and an infrared camera (IR-1000, DAGE-MTI, USA).

The recording pipette was pulled from the borosilicate glass capillary tube using a P-97 pipette puller (Sutter instruments, usa). To use electricitySpontaneous discharges were recorded extracellularly in sections by flow-clamping, using solutions (in mM: 130 potassium gluconate, 1 CaCl) ₂ ，10EGTA，10HEPES，2.0Mg-ATP，2.0Na ₃ -GTP) (pH 7.2-7.4,295 mosm/kg) fill pipettes. The voltage signals were recorded using EPC-10 amplifier and PatchMaster10 acquisition software (HEKA, germany). These signals were digitized at 10kHz and filtered with a 2kHz low pass filter. To determine the glucose sensitivity of each recorded VMH neuron, 2.5mM glucose was replaced with 0.1mM glucose in ACSF, as shown in the figure. The spontaneous firing activity of the neurons was continuously recorded and no holding current was applied to the neurons during the extracellular glucose change. After identification, neurons were recorded for 20 min during perfusion with ACSF containing 2.5mM glucose. Control ACSF was then replaced by experimental ACSF with or without 10 μ M FGFR1 selective inhibitor (PD 166866) containing 9 μ g/ml FGF4 for about 5 minutes according to the experiment. The experimental ACSF was then replaced with the control ACSF for 20 minutes to wash the drug away.

For GE/GI neurons, the peak in firing rate change was defined as the minimum/maximum firing rate minus 100% (baseline, t) in 0.1mM glucose perfusion ₀ ). The response curves of the GE/GI neurons to 0.1mM glucose were fitted by 2nd order smoothing. The time to peak of GE/GI neurons is defined as t ₀ (0.1 mM glucose perfusion) to the time duration that the FR reaches the minimum/maximum. The criteria for excitation or inhibition were set to give a rate change of-30% (Toda et al, 2016).

Preparation, characterization and treatment of FGF 4-loaded liposomes

FGF4 was dissolved in a 20% poloxamer 188 covalently bound heparin copolymer solution and further diluted with a 2.0% gelatin solution to a final concentration of 1.68mg/ml. The mixture was then sonicated (110w, 15 ℃,15 s) and lyophilized to obtain a powder containing FGF4. Thereafter, the powder was dispersed in a tert-butanol solution containing Hydrogenated Soybean Phospholipids (HSPC) and cholesterol. The mixed solution was sonicated (90w, 25 ℃,20 s) and lyophilized again to obtain a powder containing FGF 4-liposomes. Finally, the FGF 4-liposomes were reconstituted in double distilled water to form a liposome suspension. In these procedures, a gelatin solution without FGF4 was used as a blank.

The morphology of the blank and FGF4 loaded liposomes was observed by Transmission Electron Microscopy (TEM) (1230, jeol Jem, tokyo, japan) using negative staining with 1% phosphotungstic acid. Zeta potential of blanks and FGF 4-liposomes Zeta potential by dynamic light Scattering Using zeta potential/particle Sizer Nicomp as described previously ^TM 380 Nils (pss. Nicomp, san babara, ca, usa) (Zhao et al, 2016).

Then, db/db mice were intranasally administered equal volumes of blank or FGF 4-liposomes (30 μ g FGF 4) to achieve nasal to brain delivery, respectively. Blood glucose was measured at the indicated times and ipGTTs/ipITT experiments were performed as described above.

References to methods only

Benken, a., eliseenkova, a.v., ibrahimi, o.a., olsen, s.k., and Mohammadi, m. (2012). The plasticity of the N-terminus of fibroblast growth factor 1 (FGF 1) interacting with FGF receptors forms the basis of FGF1 versatility. J Biol Chem 287,3067-3078.

Bellosta, p., iwahori, a., plotnikov, a.n., eliseenkova, a.v., basilico, c., and Mohammadi, m. (2001) identification of heparin binding sites in receptors and fibroblast growth factor 4 based on structural mutation. Mol Cell Biol 21,5946-5957.

Kaliina, j., byron, s.a., makarenkova, h.p., olsen, s.k., eliseenkova, a.v., larochelle, w.j., dhanabal, m., blais, s.ornitz, d.m., day, l.a., et al (2009). Homodimerization produces receptor binding of the fibroblast growth factor 9 subfamily and heparan sulfate-dependent diffusion in the extracellular matrix. Mol Cell biol29,4663-4678.

Kharitonnkov, a., shiyanova, t.l., koester, a., ford, a.m., micarovic, r., galbreth, e.j., sandusky, g.e., hammond, l.j., moyers, j.s., owens, r.a., et al. (2005). A novel metabolic regulator FGF-21.J Clin Invest 115,1627-1635.

Olsen, s.k., li, j.y., bromleigh, c., eliseenkova, a.v., ibreahimi, o.a., lao, z., zhang, f., lindardt, r.j., joyner, a.l., and Mohammadi, m. (2006). Alternative splicing modulates the structural basis of FGF8 tissue activity in the brain. Genes Dev 20,185-198.

Structure-based mutational analysis in Sher, i., yeh, b.k., mohammadi, m., adir, n., and Ron, d. (2003) FGF7 identified new residues associated with FGFR2IIb specific interactions. FEBS Lett 552,150-154.

Toda, c., kim, j.d., improvizzeri, d., cuzzocura, s., liu, z.w., and Diano, s. (2016.) regulation of UCP2 mitochondrial fission and glucose-responsive control of the ventral medial nucleus. Cell 164,872-883.

Zhao, y.z., zhang, m., tian, x.q., zheng, l., and Lu, c.t. (2016) early intervention in diabetic cardiomyopathy was performed using basic fibroblast growth factor nanoliposomes in combination with ultrasound technology. Int J Nanomedicine 11,675-686.

Quantitative and statistical analysis

Statistical information

Mice were randomly assigned to each group, and the age and body weight of the mice were matched. Animals were not excluded from the study unless otherwise indicated. Two sets of comparisons were made using unpaired Student's t-test. RM one-way ANOVA was used for multiple group comparisons, two-way ANOVA followed by Sidak's or Tukey's multiple comparisons for groups mixed by time factor design. Data were analyzed using Prism 8.0 and expressed as mean ± s.e.m.

Example 2 results

Single injection of FGF4 into CNS for reducing hyperglycemia

First, four human recombinant FGFs (rFGF 4, rFGF7, rFGF8a, and rFGF 20) were purified as representative members of their paracrine subfamilies (fig. 1A). These rFGF's were then administered into the CNS of db/db mice and tested for their anti-hyperglycemic activity. Human recombinant FGF1 and FGF21 (rFGF 1 and rFGF 21) were used as controls. Notably, single dose administration of rFGF4 (FGF 4,3.0 μ g) into the lateral ventricles of db/db mice significantly reduced blood glucose for > 7 weeks without affecting body weight, similar to the effect of FGF1 (fig. 1B and fig. 1C). In contrast, rFGF7, rFGF8a, rFGF20 and rFGF21 lack this antihyperglycemic activity (fig. 1B, fig. 1C and fig. S1A). Interestingly, intracerebroventricular (icv) administration of FGF4 (hereinafter icv FGF 4) had a more pronounced effect on glucose tolerance than FGF1 (measured using the intraperitoneal glucose tolerance test, ipGTT) 7 days after injection (fig. 1D). This difference was no longer evident 50 days after injection (fig. 1E). As expected, higher doses of FGF4 were more effective in lowering blood glucose and improving glucose tolerance, as well as not affecting body weight (fig. S1B-S1D). As little as 0.3 μ g of FGF4 was sufficient to lower blood glucose for 4 days (fig. S1B). Like FGF1 (Scarlett et al, 2016), intraperitoneal (ip) administration of FGF4 (3.0 μ g) did not affect blood glucose in db/db mice (FIG. S1E), while the 30 μ g dose induced a transient decrease (FIG. S1F). Notably, central administration of FGF1 or FGF4 to wild-type (wt) mice did not affect blood glucose, body weight, or glucose tolerance (fig. S1G through S1L and (Scarlett et al, 2016)), suggesting that the risk of hypoglycemia induced by these two FGFs is negligible.

Next, the peripheral response after central FGF administration was recorded. Plasma hormones from db/db mice were measured 7 days after icv FGF4 and no significant change in insulin or corticosterone concentration was found, while adrenocorticotropic hormone (ACTH) levels were significantly increased (fig. S1J). These results indicate that the glucose lowering effect of FGF4 is unlikely due to an increase in insulin or a decrease in stress hormones associated with the hypothalamic-pituitary-adrenal (HPA) axis. Furthermore, there was a transient decrease in food intake after icv FGF4 (fig. S1K), similar to the effect after icv FGF1 (Scarlett et al, 2016). Meanwhile, body fat/lean content, physical activity and Respiratory Exchange Rate (RER) were not changed 7 days after icv FGF4 (fig. S1 to S1N). However, heat production increased (fig. S10). In addition, insulin resistance was significantly improved (as measured by the intraperitoneal insulin resistance test (IPITT)) 10 days after icv FGF4 (fig. S1P). To identify the target organs responsible for handling glucose, micro PET/CT studies were performed on db/db mice 3 days after icv FGF4 or FGF1. Deep enrichment of radiolabeled fluorodeoxyglucose (18F-FDG) was observed in the liver, skeletal muscle and brain of FGF4 and FGF1 treated mice (fig. 1F and S1Q). Consistent with the results, hepatic glycolytic genes Pklr, gck and Gys2 were up-regulated, while gluconeogenic gene Pck1 was down-regulated (fig. 1G). Liver and skeletal muscle glycogen also increased significantly (fig. 1H). Taken together, these data indicate that transient anorexia, elevated caloric production, and increased glucose uptake and use are important peripheral responses for centrally administered FGF to lower blood glucose.

To test whether these FGFs were able to induce sustained diabetes remission in other diabetes models, a hyperglycemic mouse model of diet-induced obesity (DIO) was developed by feeding a High Fat Diet (HFD). Significantly, central administration of FGF4 significantly reduced blood glucose, and the effect was maintained for > 2 weeks (fig. 1I). In contrast, FGF1 did not alleviate hyperglycemia in DIO mice nor affected their body weight (fig. S2A and S2B), consistent with previous reports (Scarlett et al, 2016). As expected, FGF4 can significantly improve glucose tolerance and insulin sensitivity in DIO mice (fig. 1J and S2C). In addition, glucose lowering in DIO mice was independent of body weight, fat/lean tissue content, food intake, physical activity and RER (fig. 1K, 1L and S2D-S2F). Like db/db mice, thermogenesis was increased (FIG. 1M), and ip administration of 30 μ G FGF4 only transiently reduced blood glucose in DIO mice (FIG. S2G). Thus, unlike FGF1, central injection of FGF4 induces sustained remission in diabetic DIO mice.

Fgfr1 in hypothalamic neurons is essential for FGF4 to exert anti-hyperglycemic function

To determine whether Fgfr1 in the CNS is a target for FGF, the expression pattern of Fgfr1 was first characterized by mRNA in situ hybridization. Fgfr1 ⁺ Relative to Dapi ⁺ Quantification of cells revealed that Fgfr1 is expressed in most cells throughout the hypothalamus, including VMH, dorsolateral hypothalamus (DMH), and ARC (fig. 2A-2C and S3A-S3C). The highest levels of expression were observed in VMH (fig. 2C and S3C). Then, the mixture is passed to Fgfr1 ^flox/flox Mice were injected in the hypothalamus with adeno-associated virus expressing Cre under it by a constitutive CMV promoter (AAV 8-CMV-CreGFP) to non-selectively knock out Fgfr1 in the hypothalamus (fig. 2D). The CMV promoter drives the expression of creGFP in-33% of the cells within VMH, in-24% of the cells in ARC, and in-53% of the cells in DMH (FIGS. S3E and S3F, resulting in an overall reduction of-40% of Fgfr1 expression in the hypothalamus (FIG. S3G). Note that mice lacking Fgfr1 in the hypothalamus (Fgfr 1) ^CMV-Cre ) Hyperglycemia and glucose tolerance were shown 6 weeks after virus injectionAnd (fig. 2F and S3I). Interestingly, these mice began to gain weight 3 weeks after injection (fig. S3H), which means that Fgfr1 also has an important regulatory effect on body weight. To directly test the function of Fgfr1 in FGF 4-mediated antihyperglycemic, FGF4 was centrally injected into hyperglycemic Fgfr1 ^CMV-Cre In mice. Icv FGF4 was found not to alleviate hyperglycemia, demonstrating that the ability of FGF4 to reduce glucose was blocked (fig. 2G) and body weight was not affected (fig. S3J). Therefore, fgfr1 is a key receptor in the hypothalamus to which FGF4 is centrally administered to lower blood glucose.

To determine which hypothalamic cell type mediates the effect of FGF4 on blood glucose, the Lepr-Cre line was used to knock out Fgfr1 in cells expressing the leptin receptor (Lepr) because Lepr is known to regulate metabolism (Xu et al, 2018 zhang et al, 1994. Surprisingly, specific Fgfr1 conditional knockdown in these cells did not affect blood glucose or body weight in standard food and HFD fed mice (FIGS. S4A-S4F). This knockout also did not affect the ability of icv FGF4 to reduce blood glucose in HFD fed mice (fig. S4G and S4H). Then, fgfr1 in astrocytes was knocked out using the Gfap-Cre line, and it was found that it did not block the glucose lowering effect of FGF4 (FIGS. S4I to S4M). AAV8-hSyn-CreGFP was then injected into Fgfr1 ^flox/flox In the hypothalamus of mice (FIG. S5A), fgfr1 (Fgfr 1) was selectively deleted in hypothalamic neurons ^{hSyn -Cre} ). Interestingly, fgfr1 ^hSyn-Cre Mice showed a significant increase in body weight, but blood glucose was unaffected (fig. S5B and S5C), indicating that the effects of Fgfr1 loss on body weight gain and hyperglycemia were separable. The mice were then replaced with HFD and Fgfr1 was found ^hSyn-Cre Body weight and blood glucose in mice were dramatically increased compared to GFP control (fig. 2H to fig. 2J). More importantly, icv FGF4 does not reduce Fgfr1 ^hSyn-Cre Hyperglycemia in mice (FIG. 2K), in sharp contrast to GFP control (FIG. 2L). Post icv FGF4 or GFP control body weights were not affected in Fgfr1 hson-cre mice (fig. 2M and fig. 2N). Taken together, these results suggest that the glucose lowering effect of FGF4 requires Fgfr1 in hypothalamic neurons.

Fgfr1 showed the highest level of expression in VMH (FIG. 2C), which led to the hypothesis that VMH was the primary target of FGF4. Therefore, FGF4 was directly injected into VMH of DIO mice to test its effect. Indeed, half the icv dose (1.5 μ g) was sufficient to alleviate hyperglycemia and improve glucose tolerance without affecting body weight (fig. 2O to fig. 2Q), suggesting that VMH may be the primary target region for FGF4 to exert its glucose lowering effect.

FGF4 restoration of VMH GSN function in diabetic mice

It is well known that GSN in VMH regulates blood glucose and may be important for the role of FGF4. To test the function of GSNs, their self-release rates (FR) were recorded using slice preparations in wt mice, db/db mice and FGF4 treated db/db mice (7-10 days post icv FGF 4). As expected, the conversion from normal glucose (2.5 mM) to low glucose (0.1 mM) inhibited GE neurons and stimulated GI neurons (fig. 3A and 3B). Then, the number of GE, GI and glucose Insensitive (INS) neurons were analyzed and it was found that GE neurons were present in excess of db/db, whereas GI neurons were present in deficiency of db/db mice compared to wt mice (fig. 3C). The relative numbers of GI neurons were restored to wt levels by the focused administration of FGF4 in db/db mice (FIG. 3C). The peak response and time to peak of the GSN in response to low glucose was further analyzed. For the GE neurons, the peak response to low glucose appeared to be smaller in db/db mice compared to the wt control, although the difference was not significant. The peak response increased after treatment of db/db with FGF4 (FIG. 3D and FIG. 3E). The time to peak was significantly prolonged in db/db mice compared to wt, and was fully restored by FGF4 treatment (fig. 3F). In GI neurons, the peak responses were indistinguishable between wt, db/db and FGF 4-treated db/db mice (FIGS. 3G and 3H). The time to peak was prolonged in db/db mice compared to wt, and was also restored by FGF4 treatment (fig. 3I). Taken together, these results indicate that glucose sensitivity and the relative amounts of GSNs are altered in diabetic conditions and are restored by FGF4 treatment.

In the inspired of these findings, an attempt was made to determine whether GSN is sensitive to FGF4 if FGF4 is directly applied (fig. 3J). Notably, most of the VMH GI neurons recorded (7/9, 78%) were inhibited by FGF4 perfusion, while 11% were either stimulated or insensitive to FGF4 (fig. 3K and fig. 3L). In contrast, FGF4 infusion stimulated or inhibited an equivalent number of GE neurons (fig. S6A-S6D). To test whether this acute effect is dependent on Fgfr1, PD166866, a selective Fgfr1 inhibitor, was added (Panek et al, 1998), which prevented Fgfr1 during slice recording (FIG. 3M). Remarkably, the inhibitor removed almost all FGF 4-induced neural responses (13/14, 93%); fig. 3N and fig. 3O), which indicates that FGF4 requires intact Fgfr1 for inhibition of gastrointestinal neurons. In contrast, PD166866 had essentially no effect on FGF 4-stimulated GE neurons and partially blocked FGF 4-inhibited GE neurons (fig. S6E to S6G).

In view of the ability of FGF4 to inhibit GI neurons, it was tested whether central administration of GI inhibitors could mimic the effect of FGF4 in alleviating hyperglycemia. As expected, a single icv injection of N' -nitro-L-arginine methyl ester hydrochloride (L-name), which inhibits GI neurons by inhibiting nNOS (Canabal et al, 2007), significantly ameliorates hyperglycemia for 2 days (fig. 3P). Recently, anoctamin 4 was identified as a chloride channel in the glucose-sensing pathway involving VMH GI neurons (He et al, 2020). Its inhibitor, caCciNH-A01, was found to also lower blood glucose in db/db mice for 2 days (FIG. 3P).

To explore whether FGF4 has a similar acute effect on VMH GSNs in DIO mice, the above experiments in db/db mice were therefore repeated in DIO mice. It was first demonstrated that GE neurons could be easily identified by switching to high glucose (5 mM) or low glucose, whereas GI neurons were more sensitive to low glucose and showed a small response to high glucose (fig. S7A to S7D). Therefore, for convenience, low glucose is used to distinguish between GSN subtypes. As expected, most of the recorded VMHGI neurons (6/11, 55%) were inhibited by FGF4 perfusion, while 2/11 was stimulated and 3/11 was insensitive (FIG. 3Q to FIG. 3S). To test whether Fgfr1 is required for this acute effect, fgfr1 was blocked using PD166866 (fig. 3T), resulting in 60% (3/5) of FGF 4-inhibited GI neuronal silencing (fig. 3U and fig. 3V). Thus, as in db/db mice, most GI neurons are acutely inhibited by FGF4, this acute effect being mediated primarily by Fgfr1. In contrast, a similar number of VMH GE neurons were found to be (6/18) stimulated, (7/18) inhibited, or (5/18) insensitive to FGF4 perfusion (FIGS. S7E-S7G). In addition, PD166866 blocked 60% (3/5) FGF 4-stimulated GE neurons and 50% (1/2) FGF 4-inhibited GE neurons (fig. S7H to S7J). Taken together, these results indicate that FGF4 inhibits most GI neurons in DIO mice, and that most inhibition requires Fgfr1, whereas FGF4 has an equal chance to acutely excite or inhibit GE neurons, and Fgfr1 mediates only half of this effect.

Nasal to brain delivery of FGF4 to alleviate hyperglycemia

Finally, to explore the therapeutic potential of FGF in the treatment of diabetic patients. A nasal to brain delivery system that has been used for brain-targeted drug delivery in humans has been tried by reference (Chapman et al, 2013). FGF4 was loaded into 100-nm liposomes (FIGS. 4A and 4B) and they were administered to db/db mice by the nasal drip route. Notably, nasal to brain delivery significantly reduced blood glucose in db/db mice for nearly a week without affecting body weight (fig. 4C and 4D). Thus, nasally delivered FGF4 improved glucose tolerance and insulin resistance (fig. 4E and 4F). To determine whether repeated central administration of FGF4 would cause resistance, FGF4 liposomes were administered intranasally to db/db mice every 2 weeks-3 weeks, and FGF4 was found to be able to reduce blood glucose repeatedly and efficiently without diminishing antihyperglycemic effects (fig. 4G). Therefore, intranasal administration is an effective way to deliver FGF4 for long-term anti-diabetic effects.

Discussion of the related Art

In this study, representative FGF ability to reduce blood glucose was evaluated systematically. Strong evidence is given that central or intranasal administration of FGF4 can effectively and chronically alleviate hyperglycemia in diabetic db/db mice without inducing hypoglycemia (fig. 1 and 4). Notably, FGF4 had more potent antidiabetic activity than FGF1 in diabetic DIO mice (figure 1). FGF4 is a well-established essential factor in various stages of embryonic development, including during implantation, morphogenesis and organogenesis (Beenken and Mohammadi,2009 itoh and Ornitz, 2008). This study extends the role of FGF4 in treating T2D in a sustained manner.

Attempts to identify receptors and cell classes that underlie the glucose lowering effect of FGFType, and Fgfr1 in the hypothalamus was found to be a key factor (fig. 2G). Fgfr1 is widely expressed in different hypothalamic cell types (including neurons and glia). Neuronal Fgfr1 was shown to be critical for FGF 4-mediated diabetes remission as its deletion substantially prevented the glucose lowering effect of FGF4 in DIO mice (fig. 2K). In contrast, fgfr1 is in lepr ⁺ Expression in cells and astrocytes was not necessary for FGF4 mediated glucose lowering (fig. S4). Fgfr1 was also required for sustained action of FGF1, as FGF1 was unable to lower blood glucose in DIO mice (FIG. S2A) and (Scarlett et al, 2016). However, based on the strong binding affinity between FGF1 and Fgfr1 (Suh et al, 2014), it is reasonable to assume that FGF1 will also function via Fgfr1.

GSNs are important players in glucose homeostasis. The lowering of blood glucose stimulates GI neurons, which leads to sympathetic nerve activation of the liver, adrenal medulla and pancreas, thereby increasing glucose production (Shimazu and Minokoshi, 2017). In contrast, an increase in blood glucose stimulates GE neurons, then activates sympathetic nerves innervating BAT, heart and skeletal muscles to increase glucose use (Shimazu and Minokoshi, 2017). However, the role of GSNs in diabetes pathology and prevention has not been fully explored. Here, FGF4 was shown to directly inhibit GI neurons by Fgfr1 in VMH, and pharmacological inhibition of GI neurons was sufficient to alleviate hyperglycemia (fig. 3). Consistent with known GI neurological function, FGF4 administration decreased hepatic glucose production and increased hepatic glycolysis and glycogen formation (fig. 1F-1H). Thus, the data collectively suggest that FGF4 exerts its antihyperglycemic function by inhibiting GI neurons. For GE neurons, FGF4 was observed to have a mixed effect on GE neurons, and only partial effects were dependent on Fgfr1. However, a subset of FGF 4-stimulated GE neurons may still mediate this glucose lowering effect, since activation of GE leads to glucose lowering. Whether GE neurons are involved in the antihyperglycemic effects of FGF4 remains to be investigated further.

In addition to GSN, fgfr1 is expressed in other cell types of the hypothalamus, including tanycytes. Heat shock protein 25 and pERK activation were reported in villous cells following central FGF1 administration (Brown et al, 2019, scarlett et al, 2016). In addition, tanycytes is also sensitive to glucose changes (Benford et al, 2017, fraying et al, 2011). Whether FGF requires Fgfr1 signaling in stem cells to affect blood glucose is also an important topic for future studies.

With respect to the peripheral response associated with central administration of FGF, insulin and glucagon levels did not change immediately after central administration of FGF1 or FGF4 (fig. S1J and (Scarlett et al, 2019)), suggesting that these hormones may not be the major drivers for the induction of diabetes remission. In contrast, transient decrease in food intake (fig. S1K and (Tennant et al, 2019)), increase in glucose intake (fig. 1F to 1H), and increase in energy expenditure (fig. 1M and S10) may be important contributing factors.

In conclusion, FGF4 was found to be effective and long-term in relieving hyperglycemia in diabetic db/db and DIO mice. FGF4 accomplishes this function through Fgfr1 and GSNs in the hypothalamus. FGF4 acutely inhibits GI neurons to achieve hypoglycemic effects, and re-sensitizes GSNs and restores their relative numbers over a long period of time. The identification of FGF4 as a potent and persistent antidiabetic factor may provide a new avenue for the treatment of T2D and related metabolic diseases.

Reference documents:

adams, a.c., yang, c., coskun, t., cheng, c.c., gimeno, r.e., luo, y, and kharitonnkov, a. (2012). Breadth of FGF21 metabolism is controlled by FGFR1 in adipose tissue. Mol Metab 2,31-37.

Alonge, k.m., mirzadeh, z., scarlett, j.m., logsdon, a.f., brown, j.m., cabrales, e., chan, c.k., kaiyala, k.j., bentsen, m.a., banks, w.a., et al (2020). In rats, fibroblast growth factor 1-induced diabetes is continuously alleviated requiring self-assembly of the hypothalamic peripheral neural network.

Anand, b.k., khina, g.s., sharma, k.n., dua, s., and Singh, b. (1964). The effect of glucose. Am J Physiol 207,1146-1154.

Beenken, a., and Mohammadi, m. (2009) FGF family: biology, pathophysiology, and therapy. Nat Rev Drug Discov 8,235-253.

Benford, h., bolborrea, m., pollatzek, e., lossow, k., hermans-Borgmeyer, i., liu, b., meyerhof, w., kasparov, s., and Dale, n. (2017). Glia 65,773-789.

Bentsen, m.a., rausch, d.m., mirzadeh, z., muta, k.k., scarlett, j.m., brown, j.m., herranz-Perez, v., baquero, a.f., thompson, j.e., alonge, k.m., et al (2020) transcriptomics analysis linked different hypothalamic cell types to the sustained remission of fibroblast growth factor 1-induced diabetes. Nat Commun 11,4458.

Brown, j.m., scarlett, j.m., matsen, m.e., nguyen, h.t., seer, a.l., jorgensen, r., morton, g.j., and Schwartz, m.w. (2019.) hypothalamic arcus-median eminence is a target for sustained remission of fibroblast growth factor 1-induced diabetes. Diabetes 68,1054-1061.

Canabal, d.d., song, z., potian, j.g., beuve, a., mcArdle, j.j., and Routh, v.h. (2007). Glucose, insulin and leptin signaling pathways modulate nitric oxide synthesis in hypothalamic ventral glucose-inhibited neurons. Am J Physiol Regul Integr Comp Physiol 292, R1418-1428.

Chapman, c.d., frey, w.h.,2nd, craft, s., danielyan, l., hallschmid, m., schioth, h.b., and Benedict, c. (2013). Pharm Res 30,2475-2484.

Degirolamo, C., sabba, C., and Moschetta, A. (2016.) therapeutic effects of endocrine fibroblast growth factors FGF19, FGF21, and FGF 23. Nat Rev Drug Discov 15,51-69.

Fon Tacer, k., bookout, a.l., ding, x., kurosu, h., john, g.b., wang, l., goetz, r., mohammadi, m., kuro-o, m., magelsdorf, d.j., et al. (2010) research resources: comprehensive expression profiles of fibroblast growth factor system in adult mice. Mol Endocrinol 24,2050-2064.

Frayling, c., britton, r., and Dale, n. (2011). J Physiol 589,2275-2286.

Fu, l., john, l.m., adams, s.h., yu, x.x., tomlinson, e., renz, m., williams, p.m., soriano, r., corpuz, r., moffat, b., et al (2004) fibroblast growth factor 19 increases metabolic rate, reverses diet and leptin-deficient diabetes. Endocrinology 145,2594-2603.

Goetz, r., and Mohammadi, m. (2013). Nat Rev Mol Cell Bio 14,166-180.

He, y, xu, p, wang, c, xia, y, yu, m, yang, y, yu, k, cai, x, qu, n, saito, k, et al, (2020) estrogen receptor α expression in ventral VMH regulates glucose balance. Nat Commun 11,2165.

Huang, z., tan, y., gu, j., liu, y., song, l., niu, j., zhao, l., srinivasan, l., lin, q., deng, j., et al. (2017). The mitotic and metabolic functions of FGF1 are decoupled by modulating FGF1-FGF receptor dimer stability. Cell Rep 20,1717-1728.

History of functional evolution of the mouse family of Fgf genes (2008). Dev Dyn 237,18-27.

Itoh, n., and Ornitz, d.m. (2011). From molecular evolution to roles in development, metabolism, and disease. J Biochem 149,121-130.

Jonker, j.w., suh, j.m., atkins, a.r., ahmadian, m., li, p., whiyte, j.he, m., jukuilon, h.s., yin, y.q., phillips, c.t., et al (2012) PPARgamma-FGF1 axis is necessary in adaptive fat remodeling and metabolic balance. Nature 485,391-394.

Kharitonnkov, a., shiyanova, t.l., koester, a., ford, a.m., micarovic, r., galbreth, e.j., sandusky, g.e., hammond, l.j., moyers, j.s., owens, r.a., et al.27 (2005). FGF-21 as a novel metabolic modulator. Journal of Clinical Investigation 115,1627-1635.

Tissue-specific expression of Kurosu, h., choi, m., ogawa, y., dickson, a.s., goetz, r., eliseenkova, a.v., mohammadi, m., rosenblatt, k.p., kliewer, s.a., and Kuro-o, m. (2007), betaKlotho and Fibroblast Growth Factor (FGF) receptor subtypes determines the metabolic activity of FGF19 and FGF 21.J Biol Chem 282,26687-26695.

Lan, t, morgan, d.a., rahmouni, k, sonoda, j, fu, x.r., burgess, s.c., holland, w.l., kliewer, s.a., and magelsdorf, d.j. (2017). FGF19, FGF21, and FGFR1/betaKloth activating antibodies act on the nervous system to regulate body weight and blood glucose. Cell Metabolism 26,709- +.

Marty, n., dallaporta, m., and Thorens, b. (2007). Physiology (Bethesda) 22,241-251.

Morton, g.j., matsen, m.e., bracy, d.p., meek, t.h., nguyen, h.t., stefanovski, d., bergman, r.n., wasserman, d.h., and Schwartz, m.w. (2013) FGF19 action in the brain: inducing insulin independent blood sugar reduction. J Clin Invest 123,4799-4808.

Oomura, y., ono, t., ooyama, h., and Wayner, m.j. (1969). Nature 222,282-284.

Ornitz, d.m. (2005). FGF signaling in endochondral skeletal development. Cytokine Growth Factor Rev16,205-213.

In vitro biological properties and anti-angiogenic effects of Panek, r.l., lu, g.h., dahring, t.k., butley, b.l., connolly, c., hamby, j.m., and Brown, k.j. (1998). PD 166866: selective inhibitors of FGF-1 receptor tyrosine kinase. J Pharmacol Exp Ther 286,569-577.

Perry, r.j., lee, s, ma, l, zhang, d, schlesssinger, j, and Shulman, g.i. (2015). FGF1 and FGF19 cure diabetes by inhibiting the hypothalamic-pituitary-adrenal axis. Nat Commun 6,6980.

The role of fibroblast growth factor 21 in the brain, sarrouf, d.a., thaler, j.p., morton, g.j., german, j., fischer, j.d., agimoto, k., and Schwartz, m.w. (2010): increase the energy expenditure and insulin sensitivity in obese rats. Diabetes 59,1817-1824.

Scarlett, j.m., muta, k., brown, j.m., rojas, j.m., matsen, m.e., achaya, n.k., seer, a., ingvorsen, c., jorgensen, r., hoeg-Jensen, t., et al (2019) peripheral mechanisms in brain mediate FGF1 sustained antidiabetic effects. Diabetes 68,654-664.

Central injection of fibroblast growth factor 1 induces sustained relief of diabetic hyperglycemia in rodents. Nat Med 22,800-806.

Shimazu, t., and Minokoshi, y. (2017). Systemic blood glucose regulation by glucose-sensing neurons in the hypothalamic ventral medial nucleus (VMH). J Endocr Soc 1,449-459.

Steinbusch, l., labouebe, g., and Thorens, b. (2015). Homeostatic and in feature regulation brain glucose sensing. Trends Endocrinol Metab 26,455-466.

The endocrine of Suh, j.m., jonker, j.w., ahmadian, m., goetz, r., lackey, d., osborn, o., huang, z., liu, w., yoshihara, e., van Dijk, t.h., et al (2014) FGF1 produces a new modality and effective insulin sensitizer. Nature 513,436-439.

Central and peripheral administration of Tennant, k.g., lindsley, s.r., kirigiti, m.a., true, c., and kievt, p. (2019) fibroblast growth factor 1 improved insulin secretion in diabetic mouse models. Diabetes.

Xu, j., bartolome, c.l., low, c.s., yi, x., chien, c. -h., wang, p., and Kong, d. (2018). Nature.

Ye,M.,Lu,W.,Wang,X.,Wang,C.,Abbruzzese,J.L.,Liang,G.,Li,X.,and Luo,Y.

(2016) FGF21-FGFR1 coordinates phospholipid homeostasis, lipid droplet function and endoplasmic reticulum stress in obesity. Endocrinology157,4754-4769.

Zhang, y., proenca, r., maffei, m., barone, m., leopold, l., and Friedman, j.m. (1994). Positional cloning of the mouse adiposity gene and its human homolog. Nature 372,425-432.

Claims (18)

1. A method for treating a metabolic disease comprising the step of administering a therapeutically effective dose of a Fgfr1 ligand or a pharmaceutical composition comprising the Fgfr1 ligand to a patient in need thereof via intracerebral or intranasally, wherein the metabolic disease comprises obesity, diabetes (particularly type 2 diabetes), insulin resistance, hyperinsulinemia, glucose intolerance or hyperglycemia.

2. The method of claim 1, wherein the ligand is FGF4.

3. The method of claim 2, wherein the intracerebral administration is chronic intracerebral administration.

4. The method according to claim 3, wherein the frequency of intracerebral administration is less than once every 2 weeks, preferably less than once every 5 weeks, such as once every 6 to 16 weeks.

5. The method according to claim 2, wherein the frequency of intranasal administration is less than once every 2 days, preferably less than once every 5 days, such as once every 6 to 9 days.

6. The method of claim 1, wherein the pharmaceutical composition further comprises a pharmaceutically acceptable pharmaceutical agent, such as a liposome.

7. An Fgfr1 ligand or a pharmaceutical composition comprising the Fgfr1 ligand for use in the treatment of a metabolic disease by intracerebral or intranasal administration, wherein the metabolic disease is obesity, diabetes (particularly type 2 diabetes), insulin resistance, hyperinsulinemia, glucose intolerance or hyperglycemia.

8. The Fgfr1 ligand or the pharmaceutical composition comprising the Fgfr1 ligand of claim 7 wherein the ligand is FGF4.

9. The Fgfr1 ligand or the pharmaceutical composition comprising the Fgfr1 ligand of claim 8, wherein the intracerebral administration is chronic intracerebral administration.

10. The Fgfr1 ligand or the pharmaceutical composition comprising the Fgfr1 ligand of claim 9, wherein the frequency of intracerebral administration is less than once every 2 weeks, preferably less than once every 5 weeks, such as once every 6 to 9 weeks.

11. The Fgfr1 ligand or the pharmaceutical composition comprising the Fgfr1 ligand of claim 8, wherein the frequency of intranasal administration is less than once every 2 days, preferably less than once every 5 days, such as once every 6 to 9 days.

12. The Fgfr1 ligand or the pharmaceutical composition comprising the Fgfr1 ligand of claim 7 wherein the pharmaceutical composition further comprises a pharmaceutically acceptable pharmaceutical agent such as a liposome.

13. Use of an Fgfr1 ligand or a pharmaceutical composition comprising the Fgfr1 ligand in the manufacture of a medicament for the treatment of a metabolic disease by intracerebral or intranasal administration, wherein the metabolic disease is obesity, diabetes (particularly type 2 diabetes), insulin resistance, hyperinsulinemia, glucose intolerance or hyperglycemia.

14. The use according to claim 13, wherein the ligand is FGF4.

15. The use of claim 14, wherein the intracerebral administration is chronic intracerebral administration.

16. Use according to claim 15, wherein the frequency of intracerebral administration is less than once every 2 weeks, preferably less than once every 5 weeks, such as once every 6 to 9 weeks.

17. Use according to claim 14, wherein the frequency of intranasal administration is less than once every 2 days, preferably less than once every 5 days, such as once every 6 to 9 days.

18. The use according to claim 13, wherein the pharmaceutical composition further comprises a pharmaceutically acceptable pharmaceutical formulation, such as a liposome.

Images