TITLE: NOVEL COMPOUNDS FOR TREATMENT OF OBESITY

TECHNICAL FIELD

The present invention relates to novel compounds for treatment of obesity and related conditions.

INCORPORATION-BY-REFERENCE OF THE SEQUENCE LISTING

The Sequence Listing, entitled "SEQUENCE LISTING", is 9676 bytes, was created on 23-MAY-2018 and is incorporated herein by reference.

BACKGROUND

Obesity is the result of a long-term positive energy balance whereby energy intake surpasses expenditure. The central nervous system (CNS) plays a major role in the maintenance of body weight within a narrow range by regulating energy intake. To regulate energy intake, signals, both neuronal and humoral, arising from peripheral organs involved in food intake convey information on hunger and/or satiety to the brain. Gut-derived hormones, such as prandially secreted glucagon like peptide 1 (GLP-1), have been identified as players in the regulation of feeding by relaying meal-related

information on nutritional status to the brain (Journal of Endocrinology (2014), vol.

221(1), p. T1-T16).

The presence of GLP-1 receptors in the CNS and findings from animal and human studies indicate that the GLP-1 receptor agonist (GLP-1RA) induced satiety and weight effects are, at least in part, mediated by their actions on the brain, however, brain penetration of GLP-1 and GLP-lRA's is severely limited by the largely impermeable blood-brain barrier (BBB) (Journal of Clinical Investigation (2014), vol. 124(10), p. 4473- 4488).

One approach to facilitate delivery of protein therapeutics across the BBB is to take advantage of receptor mediated transcytosis, an endogenous endocytic process in which a ligand is transported across an endothelial cell barrier. Among the many strategies to overcome this obstacle is to utilise transcytosis trafficking pathways of endogenous receptors expressed at the brain capillary endothelium such as the transferrin receptor (TfR) (Science Translational Medicine (2011), vol. 3(84), p. 1-8).

WO 2004/019872 A2 relates to pharmaceutical compositions comprising a transferrin (Tf) protein exhibiting reduced glycosylation fused to at least one therapeutic protein or peptide, and methods of using same. Prophetic Example 1 is entitled "GLP-1- transferrin fusion proteins".

The preparation of transferrin fusion proteins with analogues of GLP-1 and exendin-4 are disclosed in the article "Transferrin Fusion Technology: A Novel Approach to Prolonging Biological Half-Life of Insulinotropic Peptides" by Byung-Joon Kim et a/, in Journal of Pharmacology and Experimental Therapeutics, Vol. 334 No. 3 (2010), p. 682- 691. These fusions are designated GLP-l-Tf and Ex-4-Tf, respectively, and were made with a non-glycosylated form of human transferrin. According to the abstract, GLP-l-Tf activated the GLP-1 receptor, was resistant to inactivation by peptidases and had a half- life of approximately 2 days. The abstract also makes it clear that the fusion proteins did not cross the blood-brain barrier, whereas, for example, the GLP-l-Tf retained the acute glucose-dependent insulin-secretory properties of native GLP-1 in diabetic animals and had a profound effect on proliferation of pancreatic beta-cells.

US 2010/0077498 A1 relates to compositions and methods for blood-brain barrier delivery in the mouse.

SUMMARY

The present invention relates to novel compounds and their use for treatment of obesity and related conditions.

In one aspect the invention relates to a construct comprising a compound targeting areas in the brain involved in the regulation of body weight and an allosteric ligand to a receptor located in the blood-brain barrier (BBB), as well as pharmaceutically acceptable salts, amides or esters thereof.

In other aspects the invention relates to a composition comprising such construct and at least one pharmaceutically acceptable excipient, the construct for use as a medicament in general and specifically in the prevention or treatment of obesity and related conditions.

In one aspect, the construct of the invention exhibits an improved binding to areas of the brain that are involved in the regulation of body weight. The binding is in particular improved in areas that are protected by the BBB.

In one aspect the compound targeting areas in the brain involved in the regulation of body weight is a GLP-1 receptor agonist (GLP-1RA).

Thus, in one aspect, the invention serves to increase the accessibility of a systemically dosed GLP-1RA to appetite regulating areas in the brain.

Also, or alternatively, the invention serves to target additional GLP-1 receptor (GLP-1R) populations in the brain, for example those that are protected by the BBB. In one aspect these effects on appetite and body weight are obtained while keeping the glucose homeostasis at the same level as with systemic treatment with classic GLP-1 therapy (f.ex. with GLP-1 derivatives like liraglutide and semaglutide).

Also, or alternatively, the construct of the invention increases the binding of a systemically dosed GLP-1RA to brain regions expressing the GLP-1R. This can be demonstrated, for example, by a study as described in Example 16 (see Fig. 16A).

Also, or alternatively, the construct of the invention increases the binding of a systemically dosed GLP-1RA to brain regions expressing the GLP-1R which are protected by the BBB. This can be demonstrated, for example, by a study as described in Example 16 (see Fig. 16C).

Also, or alternatively, in one aspect the construct has the effect of lowering food intake. This can, for example, be demonstrated by a study as described in Example 20, part I.

Also, or alternatively, in one aspect the construct has the effect of causing weight loss. This can be demonstrated, for example, by a study as described in Example 20, part III.

In one aspect the receptor located in the BBB is the transferrin receptor (TfR).

The TfR is highly expressed in BBB endothelium and is a carrier protein for transferrin. Transferrin is an iron-binding blood plasma glycoprotein that controls the level of free iron (Fe) in biological fluids. Transferrin is needed for the import of iron into the cell and is regulated by intracellular iron concentration. It imports iron by

internalising the transferrin-iron complex through receptor-mediated endocytosis. Full transcytosis via this receptor route thus seems unlikely, and this is one of the challenges to overcome to use TfR as a delivery system.

BRIEF DESCRIPTION OF DRAWINGS

A better understanding of the features of the invention will be obtained by reference to the following detailed description and the accompanying drawings of which:

Fig. 1A shows the structure of an anti TfR-Fab, prepared as described in Example

1,

Fig. IB shows the structure of a fluorescently-labelled version of the Fab in Fig. 1A which is used in Example 19,

Fig. 2 shows the structure of a control Fab which does not bind to the TfR, prepared as described in Example 2,

Fig. 3 shows the structure of an anti TfR-Fab, prepared as described in Example

3, Fig. 4 shows the structure of a control Fab which does not bind to the TfR, prepared as described in Example 4,

Fig. 5 shows the structure of a fusion protein of a control Fab and a GLP-1 analogue, prepared as described in Example 5,

Fig. 6 shows the structure of a GLP-1 anti TfR-Fab fusion protein, prepared as described in Example 6,

Fig. 7 shows the structure of a GLP-1 control Fab fusion protein, prepared as described in Example 7,

Fig. 8 shows the structure of a fluorescence labelled GLP-1 anti TfR-Fab fusion protein, prepared as described in Example 8,

Fig. 9 shows the structure of a fluorescence labelled GLP-1 control Fab fusion protein, prepared as described in Example 9,

Fig. 10 shows the structure of a GLP-1 analogue, prepared as described in Example 10,

Fig. 11 shows the structure of a GLP-1 anti TfR-Fab conjugate, prepared as described in Example 11,

Fig. 12 shows the structure of a GLP-1 control Fab conjugate, prepared as described in Example 12,

Fig. 13 shows the structure of a derivative of a GLP-1 analogue, prepared as described in Example 13,

Fig. 14 shows the structure of a GLP-1 control Fab conjugate, prepared as described in Example 14,

Fig. 15 shows the structure of a GLP-1 anti TfR-Fab conjugate, prepared as described in Example 15,

Fig. 16A shows the result of in vivo imaging of fluorescently labelled GLP-1 Fab fusion proteins after acute administration to mice, more in particular the total intensity of the fluorescence signals after quantification in all brain regions expressing the GLP-1R, for an "active" construct which binds to the TfR as well as for an "inactive" control construct that does not bind to the TfR,

Fig. 16B shows the result of in vivo imaging of fluorescently labelled GLP-1 Fab fusion proteins after acute administration to mice, more in particular the total intensity of the fluorescence signals after quantification in circumventricular organs expressing the GLP-1R but devoid of a BBB, for an "active" construct which binds to the TfR as well as for an "inactive" control construct that does not bind to the TfR,

Fig. 16C shows the result of in vivo imaging of fluorescently labelled GLP-1 Fab fusion proteins after acute administration to mice, more in particular the total intensity of the fluorescence signals after quantification in exemplary brain structures which are protected by the BBB, for an "active" construct which binds to the TfR as well as for an "inactive" control construct that does not bind to the TfR,

Fig. 16D shows the result of in vivo imaging of fluorescently labelled GLP-1 Fab fusion proteins after acute administration to mice, more in particular the total intensity of the fluorescence signals after quantification in brain regions that do not express the GLP- 1R, for an "active" construct which binds to the TfR as well as for an "inactive" control construct that does not bind to the TfR,

Fig. 17A shows the result of an acute in vivo study in lean mice, more in particular the cumulative food intake (FI) in g over a period of up to 24 hours after administration of an "active" GLP-1 anti TfR-Fab conjugate,

Fig. 17B shows the result of an acute in vivo study in lean mice, more in particular the plasma exposure levels in pM measured over a period of up to 24 h after dosing of an "active" GLP-1 anti TfR-Fab conjugate,

Fig. 18A shows the result of an acute in vivo study in lean mice, more in particular the cumulative food intake (FI) in g over a period of 24 hours after

administration of either an "active" GLP-1 anti TfR-Fab conjugate or an "inactive" GLP-1 control-Fab conjugate,

Fig. 18B shows the result of an acute in vivo study in lean mice, more in particular the plasma exposure levels in pM measured over a period of up to 24 h after dosing of either an "active" GLP-1 anti TfR-Fab conjugate or an "inactive" GLP-1 control Fab conjugate,

Fig. 19A shows the result of a sub-chronical in vivo study in diet-induced obese (DIO) mice, more in particular weight loss in % measured over the treatment period of 19 days with once daily administration of either an "active" GLP-1 anti TfR-Fab conjugate or an "inactive" control Fab conjugate (for further details see Example 20, part III), and Fig. 19B shows the result of a sub-chronical in vivo study in diet-induced obese (DIO) mice, more in particular plasma exposure in pM 2 hours after dosing on the last day of the treatment period (19 days with once daily administration of either an "active" GLP-1 anti TfR-Fab conjugate or an "inactive" control Fab conjugate.

DESCRIPTION

In what follows, Greek letters may be represented by their symbol or the corresponding written name, for example: a = alpha; b = beta; e = epsilon; g = gamma; d = delta; w = omega; etc. Also, the Greek letter of m may be represented by "u", e.g. in mI = ul, or in mM = uM. In what follows chemical terminology is as usual in the art, for example the designation -C(O)- refers to a di-radical of a carbonyl group:

Chem.

and (COOH) as for example used in the designation -CH(COOH)- refers to a mono-radical of a carboxylic acid group of Chem. 8:

r

which is attached to the carbon atom shown to the left (CH).

When used herein the word "a" generally means "one or more". For example, the construct of the invention (being defined so as to comprise "a" compound targeting areas in the brain involved in the regulation of body weight and "an" allosteric ligand to a receptor located in the blood-brain barrier (BBB)) may incorporate one or more of such compounds and ligands.

Unless otherwise indicated in the specification, terms presented in singular form generally also include include the plural situation.

The invention also relates to constructs, compounds, ligands, GLP-1 RA's, GLP-1 analogues, methods of preparation, compositions and uses as disclosed herein, wherein open ended terms like "comprises" and "comprising" are replaced with closed terms such as "consists of", "consisting of", and the like.

The present invention relates to novel compounds and their use for treatment of obesity and related conditions. The compounds of the invention exhibit an improved binding to areas of the brain that are involved in the regulation of body weight. The binding is in particular improved in areas that are protected by the blood-brain barrier (BBB).

In one aspect the invention relates to a construct comprising a compound targeting areas in the brain involved in the regulation of body weight and an allosteric ligand to a receptor located in the BBB, as well as pharmaceutically acceptable salts, amides or esters thereof.

Construct

The term "construct" - comprising a compound targeting areas in the brain involved in the regulation of body weight and an allosteric ligand to a receptor located in the BBB - refers to a chemical entity (molecule) in which the compound and the ligand are covalently connected, optionally via a linker.

In one embodiment the construct consists of a compound targeting areas in the brain involved in the regulation of body weight, an allosteric ligand to a receptor located in the BBB, and an optional linker.

In one embodiment the compound and the ligand are directly attached to each other (linker absent).

In another embodiment, the compound and the ligand are attached to each other via a linker.

Fusion protein

In one embodiment, the construct is a fusion protein.

A fusion protein is a compound which can be produced by a process involving a step of recombinant expression of the entire fusion protein in a host cell.

In one embodiment, the amino acid sequence of the fusion protein consists solely of coded amino acids.

In one embodiment the protein resulting from the recombinant expression step is identical to the desired fusion protein, in which case the fusion protein is, or may be, produced by a "fully recombinant" process. In this case no further chemical modifications are needed of the fusion protein, and the expressed fusion protein can be isolated and purified and constitutes the final, desired product.

One non-limiting example of a fusion protein is the fusion protein of Example 6 (Fig. 6), in which a GLP-1 analogue is fused, via a peptidic linker (4x(GlyGlyGlySer)), to the N-terminus of the VL domain of an anti TfR-Fab. Each of the amino acid sequences of each of these three parts consists entirely of coded amino acids.

Conjugate

In one embodiment the construct is a conjugate.

In one embodiment a "conjugate" is not a fusion protein.

In one embodiment, the covalent bonds between the compound targeting areas in the brain involved in the regulation of body weight, the allosteric ligand, and the optional linker may be formed by any chemical process, except by a fully recombinant process.

In one embodiment, the structural formula of the conjugate includes at least one non-coded amino acid. One non-limiting example of a non-coded amino acid is the amino acid Aib (see e.g. the conjugate of Example 11 (Fig. 11).

Also, or alternatively, the structural formula of the conjugate includes at least one element which is not an amino acid. One non-limiting example of such element is a chemical linker such as Chem. 1 : -CH2-C(0)- (see, e.g., the conjugate of Example 11 (Fig. 11) in which a linker of Chem. 1 connects the epsilon amino group of Lys 485 and the thiol group of Cys 377. Another non-limiting example of such element is a chemical linker of Chem. 2: -CH2-C(0)-Ado-gGlu-Ado (see, e.g., the conjugate of Example 15 in which a linker of Chem. 2 connects the epsilon-amino group of Lys 438 and the thiol group of Cys 377.

In some embodiments, a conjugate can be, or is, produced by a partly recombinant process (e.g., a protein part of a desired conjugate can be, or is, produced in a recombinant expression step, and remaining part(s) of the desired conjugate can be, or are, added in one or more separate process steps, after the recombinant expression step, in one or more chemical reaction steps outside of the host cell. Such production processes may be referred to as "semi-recombinant" processes.

Areas in the brain involved in the regulation of body weight

Non-limiting examples of areas in the brain involved in the regulation of body weight are referred to by Bloemendaal et al in "Effects of glucagon-like peptide 1 on appetite and body weight: focus on the CNS", Thematic Review, Journal of Endocrinology (2014), 221, T1-T16, see, e.g., Fig. 1 at p. T8.

In some embodiments, areas in the brain involved in the regulation of body weight include, without limitation, lateral septal nucleus (LS), dentate gyrus (DG), medial habenula (MH), and/or parabrachial nucleus (PB) (see Fig. 16C and Example 16).

In some embodiments, areas in the brain involved in the regulation of body weight include, without limitation, nucleus accumbens (ACB), dorsomedial nucleus of the hypothalamus (DMH), lateral hypothalamic area (LHA), and/or lateral habenula (LH).

Regulation of body weight

In one embodiment, regulation of body weight refers to reduction of body weight.

In one embodiment reduction of body weight may be due to one or more of the following : Reduced food intake, reduced caloric intake, reduced appetite, increased energy expenditure, and/or change in food preferences.

Thus, in some embodiments, the compound part of the construct of the invention is targeting areas in the brain involved in a) reduction of food intake, b) reduction of caloric intake, c) reduction of appetite, d) increase of energy expenditure, and/or e) change in food preferences.

In some embodiments the invention relates to a method for weight

management. In some embodiments the invention relates to a method for reduction of appetite. In some embodiments the invention relates to a method for reduction of food intake.

Generally, all subjects suffering from obesity are also considered to be suffering from overweight. In some embodiments the invention relates to a method for treatment or prevention of obesity. In some embodiments the invention relates to use of the construct of the invention for treatment or prevention of obesity. In some embodiments the subject suffering from obesity is human, such as an adult human or a paediatric human (including infants, children, and adolescents). Body mass index (BMI) is a measure of body fat based on height and weight. The formula for calculation is BMI = weight in kilograms/height in meters ² . A human subject suffering from obesity may have a BMI of >30; this subject may also be referred to as obese. In some embodiments the human subject suffering from obesity may have a BMI of >35 or a BMI in the range of >30 to <40. In some embodiments the obesity is severe obesity or morbid obesity, wherein the human subject may have a BMI of >40.

In some embodiments the invention relates to a method for treatment or prevention of overweight, optionally in the presence of at least one weight-related comorbidity. In some embodiments the invention relates to use of the construct of the invention for treatment or prevention of overweight, optionally in the presence of at least one weight-related comorbidity. In some embodiments the subject suffering from overweight is human, such as an adult human or a paediatric human (including infants, children, and adolescents). In some embodiments a human subject suffering from overweight may have a BMI of >25, such as a BMI of >27. In some embodiments a human subject suffering from overweight has a BMI in the range of 25 to <30 or in the range of 27 to <30. In some embodiments the weight-related comorbidity is selected from the group consisting of hypertension, diabetes (such as type 2 diabetes), dyslipidaemia, high cholesterol, and obstructive sleep apnoea.

In some embodiments the invention relates to a method for reduction of body weight. In some embodiments the invention relates to use of the construct of the invention for reduction of body weight. A human to be subjected to reduction of body weight according to the present invention may have a BMI of >25, such as a BMI of >27 or a BMI of >30. In some embodiments the human to be subjected to reduction of body weight according to the present invention may have a BMI of >35 or a BMI of >40. The term "reduction of body weight" may include treatment or prevention of obesity and/or overweight.

Compounds Compounds targeting areas in the brain involved in the regulation of body weight include, without limitation, gut-derived hormones having been identified as players in the regulation of feeding by relaying meal-related information on nutritional status to the brain.

In one embodiment, the compound targeting areas in the brain involved in the regulation of body weight is a protein. In one embodiment, the compound is a

polypeptide. In one embodiment, the compound is a peptide.

In one embodiment, the compound targeting areas in the brain involved in the regulation of body weight is a hormone. In one embodiment, the compound is a gut- derived hormone. In one embodiment, the hormone is selected from the group consisting of eicosanoids, steroids and amino acid/protein derivatives. In one embodiment, the hormone is an amino acid/protein derivative.

In one embodiment, the compound targeting areas in the brain involved in the regulation of body weight is a GLP-1 receptor agonist (GLP-1RA).

In one embodiment, the compound targeting areas in the brain involved in the regulation of body weight is an insulinotropic agent. This term refers to the ability to initiate a signal transduction pathway resulting in insulinotropic actions or physiological effects as is known in the art.

In one embodiment, the compound targeting areas in the brain involved in the regulation of body weight is targeting pathways in the brain involved in the regulation of body weight. In one embodiment, the compound is targeting receptors in the brain involved in the regulation of body weight.

Receptor located in the BBB

A receptor located in the blood-brain barrier (BBB) is an endogenous receptor expressed at the brain capillary endothelium.

In one embodiment, the receptor located in the BBB is the transferrin receptor

(TfR).

Allosteric/orthosteric

An "orthosteric ligand" of a receptor is an endogenous, or primary, ligand which modulates the activation of the receptor. The site at which an orthosteric ligand binds to the receptor is designated an "orthosteric site" of the receptor.

An "allosteric ligand" of a receptor refers to a ligand that binds at a different site from that of the endogenous (orthosteric) ligand, viz. to an "allosteric site" of the receptor.

As an example, transferrin (Tf) is the orthosteric ligand which modulates the activation of the transferrin receptor (TfR), and Tf binds to an orthosteric site of the TfR. An allosteric ligand of the TfR binds at a different site from that of the orthosteric ligand (Tf), viz. to an allosteric site of the TfR.

GLP-1 receptor agonist

A receptor agonist may be defined as a compound that binds to a receptor and elicits a response typical of the natural ligand. A full agonist may be defined as one that elicits a response of the same magnitude as the natural ligand (see e.g. "Principles of Biochemistry ", AL Lehninger, DL Nelson, MM Cox, Second Edition, Worth Publishers, 1993, page 763).

Thus, a "GLP-1 receptor agonist" (GLP-1RA) may be defined as a compound which is capable of binding to the GLP-1 receptor (GLP-1R) and capable of activating it. And a "full" GLP-1RA may be defined as a GLP-1 receptor agonist which is capable of eliciting a magnitude of GLP-1 receptor response that is similar to native GLP-1. In one embodiment the GLP-1 receptor is the human GLP-1 receptor.

The capability of activating the human GLP-1 receptor may suitably be determined in a medium containing membranes expressing the human GLP-1 receptor, and/or in an assay with whole cells expressing the human GLP-1 receptor.

For example, purified plasma membranes from a stable transfected cell line expressing the human GLP-1 receptor may be stimulated with the compound in question, and the potency of cAMP production measured, e.g. based on competition between endogenously formed cAMP and exogenously added biotin-labelled cAMP, which may be captured using a specific antibody.

Also, or alternatively, the response of the human GLP-1 receptor may be measured in a reporter gene assay, e.g. in a stably-transfected BHK cell line that expresses the human GLP-1 receptor and contains the DNA for the cAMP response element (CRE) coupled to a promoter and the gene for firefly luciferase (CRE luciferase). When cAMP is produced as a result of activation of the GLP-1 receptor this in turn results in the luciferase being expressed. Luciferase may be determined by adding luciferin, which by the enzyme is converted to oxyluciferin and produces bioluminescence, which is measured and is a measure of the in vitro potency. One non-limiting example of such an assay is described in Example 17.

The term half maximal effective concentration (ECso) generally refers to the concentration which induces a response halfway between the baseline and maximum, by reference to the dose response curve. ECso is used as a measure of the potency of a compound and represents the concentration where 50% of its maximal effect is observed. The in vitro potency of the constructs and GLP-1RA compounds of the invention may be determined as described above, and the ECso of the compound or construct in question determined. The lower the ECso value, the better the potency.

In some embodiments, the GLP-1RA of the construct of the invention is a GLP-1 analogue. In some embodiments, the GLP-1RA of the construct of the invention is a GLP- 1 derivative.

Non-limiting examples of GLP-1 analogues are those of Examples 6, 10, and 13 (Fig. 6, Fig. 10, Fig. 13), viz. amino acids 1-31 of SEQ ID NO: 11, amino acids 1-28 of SEQ ID NO: 13, and amino acids 1-31 of SEQ ID NO: 15, respectively.

A non-limiting example of a GLP-1 derivative is disclosed in Examples 10 and 13 (SEQ ID NO: 13, SEQ ID NO: 15). These are GLP-1 analogues with a covalently attached linker.

Linker

The construct of the invention may include a linker which connects the compound targeting areas in the brain involved in the regulation of body weight with the allosteric ligand to a receptor located in the BBB. The linker is in other words optional.

In one embodiment the linker is peptidic. A peptidic linker is composed of amino acids. In one embodiment the peptidic linker consists of amino acids. In one embodiment the linker consists of coded amino acids.

Non-limiting examples of peptidic linkers are: GQAPGQAPGQAPGQAPGQAPK (SEQ ID NO: 13) and GGGSGGGSGGGSGGGS (residues 32-47 of SEQ ID NO: 11), see Examples 11 and 6 (Fig. 11 and Fig. 6, respectively)).

In another embodiment the linker is non-peptidic, or chemical. While a non- peptidic linker may include amino acid residues it does not consist entirely of amino acids.

Non-limiting examples of non-peptidic linkers are: Chem. 1 : -CH2-C(0)-, see Example 11 (Fig. 11), and Chem. 2: -CH2-C(0)-Ado-gGlu-Ado-, where Ado is a di-radical of 8-amino-3,6-dioxaoctanoic acid, Chem. 3 : -NH-(CH2)2-0-(CH2)2-0-CH2-C(0)-, and gGlu is a di-radical of L-gamma-glutamic acid, Chem. 4: -NH-CH(C00H)-(CH2)2-C(0)-, see Example 15 (Fig. 15).

Amino acid, peptide, protein

An amino acid may be defined as a compound which comprises an amine group and a carboxylic acid group, and optionally one or more additional groups often referred to as a side chain. The amine group may, e.g., be a primary or secondary amino group. Proline is one non-limiting example of an amino acid comprising a secondary amino group. A sub-group of amino acids are the a-amino acids where the nitrogen atom of the primary or secondary amino group is bonded to the a-carbon atom.

An amino acid residue is a radical of an amino acid as incorporated into a peptide or protein.

Peptides are distinguished from polypeptides and proteins on the basis of size. In the present context, unless otherwise specified, a peptide consists of up to and including 50 amino acid residues, whereas a polypeptide and a protein consists of more than 50 amino acid residues. A protein may consist of one or more polypeptide chains that may be arranged in a biologically functional way. In particular embodiments, the peptide consists of (i) at least 2 amino acid residues, (ii) at least 5 amino acid residues, (iii) at least 10 amino acid residues, (iv) at least 20 amino acid residues, (v) at least 30 amino acid residues, or (vi) at least 40 amino acid residues. In further particular embodiments, the polypeptide consists of (i) no more than 2000 amino acid residues, (ii) no more than 1500 amino acid residues, (iii) no more than 1000 amino acid residues, or (iv) no more than 600 amino acid residues. In still further particular embodiments, the protein consists of (i) no more than 2000 amino acid residues, (ii) no more than 1500 amino acid residues, (iii) no more than 1000 amino acid residues, (iv) no more than 600 amino acid residues, or (v) no more than 500 amino acid residues.

Amino acids may be classified based on origin, as coded amino acids or non- coded amino acids. The term "coded" amino acids (sometimes also referred to as "natural" amino acids) may be defined by reference to IUPAC (Table 1 in section 3AA-1) : (http://www.chem.qmul.ac.Uk/iupac/AminoAcid/AAln2.html#AAl), which gives structure, trivial name, systematic name, one- and three-letter symbols for these 20 amino acids. These are all a-amino acids.

One non-limiting example of a non-coded amino acid is Aib, also known as a- aminoisobutyric acid and with the following structure:

Chem. 6:

GLP-1 peptides and analogues

The term "GLP-1 peptide" as used herein refers to the human Glucagon-Like Peptide-1 (GLP-l(7-37)), the sequence of which is included in the sequence listing as SEQ ID NO: 16, or an analogue thereof. The peptide having the sequence of SEQ ID NO: 16 may also be designated "native" GLP-1.

The term "GLP-1 analogue" or "analogue of GLP-1" as used herein refers to a peptide, or a compound, which is a variant of GLP-l(7-37) (SEQ ID NO: 16).

In the sequence listing, the first amino acid residue of SEQ ID NO: 16 (histidine) is assigned no. 1. However, in what follows, according to established practice in the art, this histidine residue is referred to as no. 7, and subsequent amino acid residues are numbered accordingly, ending with glycine no. 37. Therefore, generally, any reference herein to an amino acid residue number or a position number of the GLP-l(7-37) sequence is to the sequence starting with His at position 7 and ending with Gly at position 37.

GLP-1 analogues of the constructs and derivatives of the invention may be described by reference to i) the number of the amino acid residue in native GLP-l(7-37) which corresponds to the amino acid residue which is changed (i.e., the corresponding position in native GLP-1), and to ii) the actual change.

In other words, a GLP-1 analogue is a GLP-l(7-37) peptide in which a number of amino acid residues have been changed when compared to native GLP-l(7-37) (SEQ ID NO: 16). These changes may represent, independently, one or more amino acid substitutions, additions, and/or deletions.

As is apparent from the above examples, amino acid residues may be identified by their full name, their one-letter code, and/or their three-letter code. These three ways are fully equivalent.

The expressions "a position equivalent to" or "corresponding position" may be used to characterise the site of change in a variant GLP-l(7-37) sequence by reference to native GLP-l(7-37) (SEQ ID NO: 16). Equivalent or corresponding positions, as well as the number of changes, are easily deduced, e.g. by simple handwriting and eyeballing; and/or a standard protein or peptide alignment program may be used, such as "align" which is based on a Needleman-Wunsch algorithm. This algorithm is described in

Needleman, S.B. and Wunsch, C.D., (1970), Journal of Molecular Biology, 48: 443-453, and the align program by Myers and W. Miller in "Optimal Alignments in Linear Space" CABIOS (computer applications in the biosciences) (1988) 4: 11-17. For the alignment, the default scoring matrix BLOSUM62 and the default identity matrix may be used, and the penalty for the first residue in a gap may be set at -12, or preferably at -10, and the penalties for additional residues in a gap at -2, or preferably at -0.5. The term "peptide", as e.g. used in the context of the GLP-1 analogues of the derivatives of the invention, refers to a compound which comprises a series of amino acids interconnected by amide (or peptide) bonds.

The GLP-1 peptides of the invention comprise at least five constituent amino acids connected by peptide bonds. In particular embodiments the GLP-1 peptide comprises at least 10, preferably at least 15, more preferably at least 20, even more preferably at least 25, or most preferably at least 28 amino acids.

In additional particular embodiments, the peptide is a) composed of, or b) consists of, i) 29, ii) 30, iii) 31, or iv) 32 amino acids.

In a still further particular embodiment the peptide consists of amino acids interconnected by peptide bonds.

In what follows, all amino acids of the GLP-1 peptide for which the optical isomer is not stated is to be understood to mean the L-isomer (unless otherwise specified).

Pharmaceutically acceptable salt, amide ester

The construct of the invention may be in the form of a pharmaceutically acceptable salt, amide, or ester. This is also the case for its constituent compound and ligand parts as such.

Salts are e.g. formed by a chemical reaction between a base and an acid, e.g. : 2NH ₃ + H2SO4 (NH ₄ )2S04.

The salt may be a basic salt, an acid salt, or it may be neither nor (i.e. a neutral salt). Basic salts produce hydroxide ions and acid salts hydronium ions in water.

The salts may be formed with added cations or anions between anionic or cationic groups, respectively. These groups may be situated in the compound part, in the ligand part, and/or in the optional linker part of the constructs of the invention.

Non-limiting examples of anionic groups of the constructs of the invention include free carboxylic groups, e.g. a free carboxylic acid group at a C-terminal amino acid of the construct, and free carboxylic groups at internal acid amino acid residues such as Asp and Glu.

Non-limiting examples of cationic groups include a free amino group at the N- terminus, if present, as well as any free amino group of internal basic amino acid residues such as His, Arg, and Lys.

The ester of the construct of the invention may, e.g., be formed by the reaction of a free carboxylic acid group with an alcohol or a phenol, which leads to replacement of at least one hydroxyl group by an alkoxy or aryloxy group

The ester formation may involve the free carboxylic group at the C-terminus, and/or any free carboxylic group internally in the construct. The amide of the construct of the invention may, e.g., be formed by the reaction of a free carboxylic acid group with an amine or a substituted amine, or by reaction of a free or substituted amino group with a carboxylic acid.

The amide formation may involve the free carboxylic group at the C-terminus, any free carboxylic group internally in the construct, the free amino group at the N- terminus of the construct, and/or any free or substituted amino group internally in the construct.

In a particular embodiment, the construct is in the form of a pharmaceutically acceptable salt. In another particular embodiment, the construct is in the form of a pharmaceutically acceptable amide, preferably with an amide group at the C-terminus. In a still further particular embodiment, the construct is in the form a pharmaceutically acceptable ester.

PRODUCTION PROCESSES

The production of proteins and peptides like the control Fab's, anti TfR-Fab's and GLP-1 analogues disclosed herein is well known in the art.

GLP-1 analogues may for instance be produced by classical peptide synthesis, e.g., solid phase peptide synthesis using t-Boc or Fmoc chemistry or other well established techniques, see, e.g., Greene and Wuts, "Protective Groups in Organic Synthesis", John Wiley & Sons, 1999, Florencio Zaragoza Dorwald, "Organic Synthesis on solid Phase", Wiley-VCH Verlag GmbH, 2000, and "Fmoc Solid Phase Peptide Synthesis", Edited by W.C. Chan and P.D. White, Oxford University Press, 2000.

Also, or alternatively, they may be produced by recombinant methods, viz. by culturing a host cell containing a DNA sequence encoding the analogue and capable of expressing the peptide in a suitable nutrient medium under conditions permitting the expression of the peptide. Non-limiting examples of host cells suitable for expression of these peptides are: Escherichia coli, Saccharomyces cerevisiae, as well as mammalian BHK or CHO cell lines.

Those constructs of the invention which include non-coded amino acids may e.g. be produced as described in the experimental part. Or see e.g., Hodgson et al: "The synthesis of peptides and proteins containing non-natural amino acids", Chemical Society Reviews, vol. 33, no. 7 (2004), p. 422-430; and WO 2009/083549 Al entitled "Semi- recombinant preparation of GLP-1 analogues".

Specific examples of methods of preparing a number of constructs are included in the experimental part. PHARMACEUTICAL COMPOSITIONS

In one embodiment, the composition of the invention is a pharmaceutical composition.

Pharmaceutical compositions comprising a construct of the invention or a pharmaceutically acceptable salt, amide, or ester thereof, and a pharmaceutically acceptable excipient may be prepared as is known in the art.

The term "excipient" broadly refers to any component other than the active therapeutic ingredient(s). The excipient may be an inert substance, an inactive substance, and/or a not medicinally active substance.

The excipient may serve various purposes, e.g. as a carrier, vehicle, diluent, tablet aid, and/or to improve administration, and/or absorption of the active substance.

The formulation of pharmaceutically active ingredients with various excipients is known in the art, see e.g. Remington: The Science and Practice of Pharmacy (e.g. 19 ^th edition (1995), and any later editions).

Non-limiting examples of excipients are: Solvents, diluents, buffers,

preservatives, tonicity regulating agents, chelating agents, and stabilisers.

Examples of formulations include liquid formulations, i.e. aqueous formulations comprising water. A liquid formulation may be a solution, or a suspension. An aqueous formulation typically comprises at least 50% w/w water, or at least 60%, 70%, 80%, or even at least 90% w/w of water.

Alternatively, a pharmaceutical composition may be a solid formulation, e.g. a freeze-dried or spray-dried composition, which may be used as is, or whereto the physician or the patient adds solvents, and/or diluents prior to use.

A pharmaceutical composition may comprise a buffer.

A pharmaceutical composition may comprise a preservative.

A pharmaceutical composition may comprise a chelating agent.

A pharmaceutical composition may comprise a stabiliser.

A pharmaceutical composition may comprise one or more surfactants.

A pharmaceutical composition may comprise one or more protease inhibitors.

The construct of the invention may be administered in the form of a

pharmaceutical composition. It may be administered to a patient in need thereof at several sites.

The route of administration may be, for example, parenteral.

A composition may be administered in several dosage forms, for example as a solution; a suspension; an emulsion; a microemulsion; multiple emulsions; an injection solution; an infusion solution. Systemic or parenteral administration may be performed by subcutaneous, intramuscular, intraperitoneal, or intravenous injection by means of a syringe, optionally a pen-like syringe, or by means of an infusion pump.

The treatment with a construct according to the invention may also be combined with one or more additional pharmacologically active substances.

The treatment with a construct according to the invention may also be combined with a surgery that influences the glucose levels, and/or lipid homeostasis such as gastric banding or gastric bypass.

PARTICULAR EMBODIMENTS

The invention is further described by the following non-limiting embodiments.

1. A construct comprising a compound targeting areas in the brain involved in the regulation of body weight and an allosteric ligand to a receptor located in the blood-brain barrier (BBB); or a pharmaceutically acceptable salt, amide or ester thereof.

2. The construct of embodiment 1, wherein the compound is targeting pathways in the brain involved in the regulation of body weight.

3. The construct of any of embodiments 1-2, wherein the compound is targeting receptors in the brain involved in the regulation of body weight.

4. The construct of any of embodiments 1-3, wherein the compound is a peptide, a polypeptide, or a protein.

5. The construct of any of embodiments 1-4, wherein the compound is a hormone.

6. The construct of embodiment 5, wherein the hormone is selected from the group consisting of eicosanoids, steroids and amino acid/protein derivatives.

7. The construct of embodiment 5, wherein the hormone is an amino acid/protein derivative.

8. The construct of any of embodiments 1-7, wherein the compound is a gut-derived hormone.

9. The construct of any of embodiments 1-8, wherein the compound is an

insulinotropic agent.

10. The construct of any of embodiments 1-9, wherein the compound is a GLP-1

receptor agonist (GLP-1RA).

11. The construct of any of embodiments 1-10, wherein the receptor located in the BBB is the transferrin receptor (TfR).

12. The construct of embodiment 11, wherein the allosteric ligand binding to the TfR is a peptide, a polypeptide, or a protein. 13. The construct of any of embodiments 11-12, wherein the allosteric ligand binding to the TfR is an anti-transferrin receptor Fab (anti TfR-Fab).

14. The construct of any of embodiments 1-13 which is a fusion protein or a

conjugate.

15. The construct of any of embodiments 1-14, which further comprises a linker.

16. The construct of embodiment 15, wherein the linker is peptidic or non-peptidic.

17. The construct of any of embodiments 1-16, which is a GLP-1RA.

18. The construct of any of embodiments 1-17, which is a full GLP-1RA.

19. The construct of any of embodiments 1-18, which has an in vitro potency

corresponding to an ECso at or below 1000 pM, determined using the assay of Example 17.

20. The construct of any of embodiments 1-19, which is capable of binding to the GLP-1R.

21. The construct of any of embodiments 1-20, which has a GLP- 1R binding affinity corresponding to an ICso at or below 100 nM, determined using the assay of Example 18.

22. The construct of any of embodiments 1-21, which is capable of binding to the transferrin receptor (TfR).

23. The construct of any of embodiments 1-22, which is capable of inhibiting the

uptake of a TfR targeted antibody.

24. The construct of embodiment 23, which inhibits uptake of the TfR targeted

antibody corresponding to an ICso at or below 100 nM, determined using the assay of Example 19.

25. The construct of embodiment 24, for which the ICso value is lower than the ICso value of a control construct which incorporates the same GLP-1RA but a ligand which does not bind to the TfR, wherein both ICso values are determined using the assay of Example 19.

26. The construct of any of embodiments 11-25, which exhibits an increased binding to brain regions expressing the GLP-1R as compared to a control construct which only differs from the construct by incorporating a ligand which does not bind to the TfR.

27. The construct of embodiment 26, wherein the binding to the brain regions

expressing the GLP-1R is determined using in vivo imaging of fluorescently labelled constructs.

28. The construct of embodiment 27, which results in higher fluorescence signals as compared to a control construct in at least a majority of the brain regions expressing the GLP-1R, wherein the brain regions expressing the GLP-1R are as set out in Fig. 16A.

29. The construct of embodiment 28, wherein the fluorescence signals are higher in at least 50%, 60%, 70%, 80%, or preferably at least 90% of the GLP-1R expressing brain regions.

30. The construct of any of embodiments 26-29, wherein the binding to brain regions expressing the GLP-1R is determined essentially as described in Example 16.

31. The construct of any of embodiments 11-30, which exhibits an increased binding to GLP-1R expressing brain regions that are protected by the BBB as compared to a control construct which only differs from the construct by incorporating a ligand which does not bind to the TfR.

32. The construct of embodiment 31, which results in higher fluorescence signals as compared to a control construct in at least a majority of the BBB protected brain regions expressing the GLP-1R.

33. The construct of embodiment 32, wherein the BBB protected brain regions

expressing the GLP-1R are LS, DG, MH, and PB as set out in Fig. 16C.

34. The construct of embodiment 32, wherein the BBB protected brain regions

expressing the GLP-1R are ACB, DMH, LHA, and LH.

35. The construct of any of embodiments 32-34, wherein the fluorescence signals are higher in at least 50%, 60%, 70%, 80%, or at least 90% of the GLP-1R expressing brain regions.

36. The construct of any of embodiments 32-35, wherein the fluorescence signal is at least two, three, four, five, six, seven, eight, nine, or at least ten times higher than the fluorescence signal for the control construct.

37. The construct of any of embodiments 32-36, wherein the binding to GLP-1R

expressing brain regions that are protected by the BBB is determined essentially as described in Example 16.

38. The construct of any of embodiments 1-37, which lowers food intake.

39. The construct of embodiment 38, which lowers food intake dose dependently.

40. The construct of any of embodiments 38-39, which causes a lower food intake as compared to a control which only differs from the construct by incorporating a ligand which does not bind to the TfR.

41. The construct of any of embodiments 38-40, wherein food intake is cumulative food intake in g over a period of 24 hours. 42. The construct of any of embodiments 38-41, wherein food intake is determined in an acute study in lean mice, essentially as described in Example 20, part I or part

II.

43. The construct of any of embodiments 1-42, which causes weight loss.

44. The construct of embodiment 43, which causes a higher weight loss as compared to a control which only differs from the construct by incorporating a ligand which does not bind to the TfR.

45. The construct of any of embodiments 43-44, wherein weight loss is determined in a sub-chronical study in DIO mice, essentially as described in Example 20, part

III.

46. A composition comprising a construct of any of embodiments 1-45 and at least one pharmaceutically acceptable excipient.

47. The construct of any of embodiments 1-45 or the composition of embodiment 46 for use as a medicament.

48. The construct of any of embodiments 1-45 or the composition of embodiment 46 for use in the prevention or treatment of obesity.

49. A method for the prevention or treatment of obesity, comprising administering the construct according to any of embodiments 1-45 or the composition of

embodiment 46.

Examples

This experimental part starts with a list of abbreviations, and is followed by a section including general methods for detecting and characterising the constructs of the invention. Then follows a number of examples which relate to the preparation of specific construct parts and constructs, and at the end a number of examples have been included relating to the activity and properties of these analogues and derivatives (section headed pharmacological methods).

The examples serve to illustrate the invention.

List of Abbreviations

Ac acetyl

ACB nucleus accumbens

ad lib ad libitum

Ado 8-amino-3,6-dioxaoctanoic acid

AOB accessory olfactory bulb

AP area postrema APN anterior pretectal nucleus

ARH arcuate hypothalamic nucleus

AU arbitrary unit

AUC area under the curve

BBB blood-brain barrier

BBB-R blood-brain barrier receptor

BGG bovine gamma globulin

BHK baby hamster kidney

BMA basomedial amygdalar nucleus

BID bis in die (twice daily)

Boc t-butyloxycarbonyl

BrAc bromo acetamide

BST bed nuclei of the stria terminalis

CAMP cyclic AMP

CeA central amygdalar nucleus

CH constant region heavy chain

CHO Chinese hamster ovary

CL constant region light chain

CLND chemiluminescent nitrogen detection

CNS central nervous system

COA cortical amygdalar area

CRE cAMP response element

CU cuneate nucleus

CV coefficient of variation

Da Dalton

DAPI 4',6-diamidino-2-phenylindole

DBE dibenzyl ether

DCM dichloromethane

DF diafiltration

DG dentate gyrus

dPhO demineralised water

DIC diisopropylcarbodiimide

DIO diet-induced obese

DMEM dulbecco modified eagle medium

DMF dimethylformamide

DMH dorsomedial nucleus of the hypothalamus DMSO dimethyl sulfoxide

DMX dorsal motor nucleus of the vagus nerve

DPP-4 dipeptidyl peptidase 4

DR dorsal nucleus raphe

ECU external cuneate nucleus

EDTA ethylenediaminetetraacetic acid

Fab fragment antigen-binding (antigen-binding fragment)

FI food intake

Fmoc 9-fluorenylmethyloxy carbonyl

G418 geneticin

GENd geniculate group, dorsal thalamus

gGiu glutamic acid for incorporation/when attached to another molecule via an amide bond involving the carboxylic acid group at the gamma position of the side chain of Glu

GLP-1 glucagon like peptide-1

GLP-1R glucagon like peptide-1 receptor

GLP-1RA glucagon like peptide-1 receptor agonist

H hours

HBSS Hanks' balanced salt solution

HEK human embryonic kidney

HEPES 4-(2-hydroxyethyl)-l-piperazineethanesulfonic acid

HFIP l,l, l,3,3,3-hexafluoro-2-propanol or hexafluoroisopropanol

HPLC high performance liquid chromatography

HSA human serum albumin

IgG4 subclass four of immunoglobulin G

10 inferior olivary complex

/. v. intravenous

L liter

LCMS liquid chromatography mass spectroscopy

LH lateral habenula

LHA lateral hypothalamic area

lib libitum

LOCI luminescence oxygen channeling immunoassay

LoQ lower limit of quantification

LPO lateral preoptic area

LRNm lateral reticular nucleus, magnocellular LS lateral septal nucleus

LSFM light sheet fluorescence microscopy

ME median eminence

MH medial habenula

MM medial mammillary nucleus

MPT medial pretectal area

MQ de-ionised water

dRN midbrain reticular nucleus

Mtt 4-methyltrityl

MV medial vestibular nucleus

NaP sodium phosphate

NBF neutral buffered formalin

NI nucleus incertus

NLL nucleus of the lateral lemniscus

NTS nucleus of the solitary tract

OEG same as Ado

O/N overnight

OtBu tert butyl ester

OV vascular organ of the lamina terminalis

Oxyma Pure® cyano-hydroxyimino-acetic acid ethyl ester

PB parabrachial nucleus

Pbf 2,2,4,6,7-pentamethyldihydrobenzofuran-5-sulfonyl

PBS phosphate buffered saline

Pen/strep Penicillin Streptomycin

PFA paraformaldehyde

QC quality control

QD quaque die (once daily)

QW quaque (once weekly)

Rt retention time

RT room temperature

s.c. sub cutaneous

SDS-PAGE sodium dodecyl sulfate polyacrylamide gel electrophoresis

SEM standard error of mean

SEC-HPLC size exclusion high performance liquid chromatography

SFO subfornical organ

SPPS solid phase peptide synthesis tBu tert. butyl

TCEP tris(2-carboxyethy I) phosphine

TEA triethanolamine

Tf transferrin

TFA trifluoroacetic acid

TfR transferrin receptor

TFT tangential flow filtration

THF tetrahydrofuran

TIC total ion count

TIPS triisopropylsilane

TNP 2,4,6-trinitrophenol

Tris tris(hydroxymethyl)aminomethane or 2-amino-2-hydroxymethyl- propane-l,3-diol

Trt triphenylmethyl (trityl)

Tween20 polyethylene glycol sorbitan monolaurate

UF ultrafiltration

ULoQ upper limit of quantification

VL light chain variable domain

WGA wheat germ agglutinin

General Methods of Detection and Characterisation

1. LC-MS methods

Method: LCMS_34

System LC-system : Waters Acquity UPLC H Class

Column : : Waters Acquity BEH, C-18, 1.7 m, 2.1 mm x 50 mm

Detector: : Waters Xevo G2-XS QTof

Detector Ionisation method : ES

setup Scanning range: 50 - 4000 amu

Operating mode : MS resolution mode

positive/ne : positive mode

Voltage: Capillary 3.00 kV

Sample cone 40 V

Source 80 V

Temperature: Source 150C

Method: LCMS_36

This protocol outlines the MS method depicted by the protocol title, and is used to link the method description to the results of the analysis. The method is used to measure the purity of the analyte by a chromatographic separation of the sample, and to identify and quantify the components by their mass spectra and total ion count (TIC) LC-system : Waters Acquity UPLC H Class. Step gradient: 5 % to 25% B 1 min 25-65 % B 6 min

Purity: Total ion current (TIC) AUC of analyte peak, in percent of total AUC excl solvent peak, as reported by system software. Identity: Mass of each analyte mass peak expressed as m/z from highest to lowest. Scanning range is the range scanned in the method used. Detection method is e.g linear reflector

Example 1:

Transient expression of anti TfR-Fab

This example describes the transient expression of the anti Tfr-Fab shown in Fig. 1A (SEQ ID NO : 3).

Light chain (SEQ ID NO : 1) :

DIQMTQSPASLSASLEEIVTITCQASQDIGNWLAWYQQKPGKSPQLLIYGATSLADGVPS

RFSGSRSGTQFSLKISRVQVEDIGIYYCLQAYNTPWTFGGGTKLELKRTVAAPSVFI FPP

SDEQLKSGTASVVCLLNNFYPREAKVQWKVDNALQSGNSQESVTEQDSKDSTYSLSS TLT

LSKADYEKHKVYACEVTHQGLSSPVTKSFNRGEC

Heavy chain (SEQ ID NO : 2) :

EVQLVESGGGLVQPGNSLTLSCVASGFTFSNYGMHWIRQAPKKGLEWIAMIYYDSSKMNY ADTVKG

RFTISRDNSKNTLYLEMNSLRSEDTAMYYCAVPTSHYVVDVWGQGVSVTVSSASTKG PSVFPLAPCS

RSTSESTAALGCLVKDYFPEPVTVSWNSGCLTSGVHTFPAVLQSSGLYSLSSVVTVP SSSLGTKTYTC

NVDHKPSNTKVDKRVESK

The anti TfR-Fab was designed as a chimeric Fab with variable domains as published in US 2010/0077498 Al, and human CL and IgG4 CH I domains. The CH I domain carries an A158C mutation to generate a cysteine handle for chemical

conjugation.

Nucleic acid sequences encoding the anti TfR-Fab were cloned into a pTT5- derived expression vector. To express the Fab, HEK293 6E cells were cultured to the cell density of lxlOE6 cells/ml and transfected with the expression vector plasmid DNA using lipofectin (Invitrogen, cat. no. 12347-500). The expression supernatant was harvested by centrifugation and filtering of the cell culture 5 days after the transfection.

The cell culture supernatant containing secreted anti TfR-Fab was applied to a protein G Sepharose 4FF affinity column (GE Healthcare) equilibrated in phosphate buffered saline (PBS). The bound Fab was eluted with 0.1 M glycine-HCI, pH 2.8.

Fractions were collected and neutralized immediately with 1/20 volume 2 M Tris-HCI, pH 9.0. The pooled fractions were then diluted into 25 mM NaOAc pH 5.0 and applied to a SP HP column (GE Healthcare). The bound Fab was eluted with a 100-400 mM linear gradient of NaCI in 25 mM NaOAc pH 5.0 and buffer-exchanged to PBS on a G25 desalting column (GE Healthcare). Purified protein was sterilized by filtration through a 0.2 mm filter unit (Sartorius). The purity of anti TfR-Fab was analysed by SDS-PAGE and size-exclusion HPLC. The Fab identity was confirmed by mass spectrometry.

The purity was above 95% by SDS-PAGE and 99.1% by SEC-HPLC (A280 nm). The observed mass was 47198.3 Da, which is consistent with the theoretical mass (47198.5 Da).

Example 2:

Transient expression of control Fab

This example describes the transient expression of a control Fab shown in Fig. 2, which does not bind to the TfR. The control Fab of Fig. 2 has the amino acid sequence of SEQ ID NO: 5.

Light chain (SEQ ID NO : 4) :

DIVMTQTPLSLSVTPGQPASISCRSSQSLLHSNGNTYLHWYLQKPGQSPQLLIYKVSNRF SGVPDRF

SGSGSGTDFTLKISRVEAEDVGVYYCSQSTHVPWTFGQGTKVEIKRTVAAPSV

FIFPPSDEQLKSGTASVVCLLNNFYPREAKVQWKVDNALQSGNSQESVTEQDSKDST YSL

SSTLTLSKADYEKHKVYACEVTHQGLSSPVTKSFNRGEC

A control Fab was designed to be with the variable domains of the light chain of anti-TNP antibody (SEQ ID NO : 4) and the same heavy chain as in the anti-TfR antibody of Example 1 (SEQ ID NO : 2).

A similar procedure as described in Example 1 was used to clone and express the control Fab.

The purity was above 95% by SDS-PAGE and 99.2% by SEC-HPLC (A280 nm). The observed mass was 47787.4 Da, which is consistent with the theoretical mass (47787.0 Da).

Example 3:

Generation of a stable cell line expressing anti TfR-Fab

This example describes the generation of a stable cell line expressing the anti TfR-Fab shown in Fig. 3 (same as shown in Fig. 1A, SEQ ID NO : 3).

A stable cell line expressing the anti TfR-Fab of Example 1 was generated using a CHO K1SV GS KO cell line with GS selection (Lonza). The Fab light chain (SEQ ID NO: 1) and heavy chain (SEQ ID NO: 2) with an N terminal CD33 signal peptide (SEQ ID NO:

17) were subcloned into a pEE17.4 vector (Lonza). The DNA was linearized and transfected into CHO K1SV GS KO cell lines by electroporation. The transfected cell micropools were propagated in CD CHO medium (Thermo Fisher Scientific) with a MSX (Sigma-Aldrich) concentration of 25 mM for two weeks. The cells were then transferred to 96 deep well plates without MSX and the highest expressing 24 micropools were selected by Fortebio analysis using protein G sensor (Fortebio).

To achieve higher expression stability, single cell clones were isolated with the selected high expressing micropools by limited dilutions. The cells were grown for 2-3 weeks and 96 single clones were picked and screened by Fortebio. The top 24 clones were further propagated in 24 well plates and ranked by Fortebio with protein G sensor and SDS PAGE analysis. Six highest expressing cell clones each were used for further stability test. Cells before and after 30 days propagation were cultured in Fed-batch culture for the expression level evaluation by HPLC (Fed-batch culture is a 12 day culture with two nutrient feeds on day 3 and day 7). Two single clones were stable with expression levels after 30 days' passage of above 70%.

The stable cell lines expressing the Fabs were revived and propagated. The cell culture was then inoculated into a 13 L Sartorius bioreactor with the initial volume of 10.5L in CD-CHO medium and the cell density to be 0.3xl0E6 cells/ml. The fed-batch cell culture was maintained at 36.5°C, Dissolved Oxygen level of 40%, and pH of 7 (±0.3). Nutrient feeding was made based on the cell density and glucose level. The cell culture was harvested after 13 days using Millipore depth-filter MD0HC054H1.

The anti-TfR Fab was purified, sterilised, and analysed as described in Example

1.

The purity was above 95% by SDS-PAGE and 96.7% by SEC-HPLC (A280 nm). The observed mass was 47199.3 Da, which is consistent with the theoretical mass (47198.5 Da).

Example 4:

Generation of a stable cell line expressing control Fab

This example describes the generation of a stable cell line expressing the control Fab shown in Fig. 4 (same as shown in Fig. 2, SEQ ID NO: 5).

A stable cell line expressing the control Fab of Example 2 was generated as described in Example 3, except that the Fab light chain of SEQ ID NO: 4 was used instead of that of SEQ ID NO: 1. Also the measures to achieve higher expression stability described in Example 3 were followed. Three single clones were stable with expression levels above 70% after 30 days' passage.

The stable cell lines expressing the control Fabs were further cultivated in a 13 L Sartorius bioreactor as described in Example 3.

The cell culture supernatant containing secreted control Fab was concentrated and buffer exchanged to 25 mM NaOAc pH 5.0 using ultrafiltration/diafiltration (UF/DF) in tangential flow filtration (TFF) system (Millipore), then applied to a SP HP column (GE Healthcare). The bound Fab was eluted with a 0-350 mM linear gradient of NaCI in 25 mM NaOAc pH 5.0 and buffer-exchanged to PBS using UF/DF in TFF system (Millipore).

Purified control Fab was sterilized by filtration through a 0.2 mm filter unit (Sartorius). The purity was analysed by SDS-PAGE and size-exclusion HPLC. The Fab identity was confirmed by mass spectrometry.

The purity was above 95% by SDS-PAGE and 98.7% by SEC-HPLC (A280 nm). The observed mass was 47787.5 Da, which is consistent with the theoretical mass (47787.0 Da).

Example 5:

Transient expression of GLP-1 anti TNP-Fab fusion protein

This example describes the expression of a fusion protein shown in Fig. 5 (SEQ ID NO : 8) of a control Fab and a GLP-1 analogue. The control Fab is an anti-TNP Fab, where TNP refers to 2,4,6-trinitrophenol.

Light chain (SEQ ID NO : 6) :

HGEGTFTSDVSSYLEEQAAREFIAWLVKGRPGGGSGGGSGGGSGGGSDIVMTQTPLSLSV TPGQP

ASISCRSSQSLLHSNGNTYLHWYLQKPGQSPQLLIYKVSNRFSGVPDRFSGSGSGTD FTLKISRVEA

EDVGVYYCSQSTHVPWTFGQGTKVEIKRTVAAPSVFIFPPSDEQLKSGTASVVCLLN NFYPREAKVQ

WKVDNALQSGNSQESVTEQDSKDSTYSLSSTLTLSKADYEKHKVYACEVTHQGLSSP VTKSFNRGE

C

Heavy chain (SEQ ID NO : 7) :

QVQLQESGPGLVKPSETLSLTCTVSGGSISSGYWNWIRQPPGKGLEWIGTISYSGDTYYN PSLKSRV

TISVDTSKNQFSLKLSSVTAADTAVYYCARYGSYVFDYWGQGTTVTVSSASTKGPSV FPLAPCSRST

SESTAALGCLVKDYFPEPVTVSWNSGALTSGVHTFPAVLQSSGLYSLSSVVTVPSSS LGTKTYTCNV

DHKPSNTKVDKRVESKHHHHHHHHHH The two chain Fab fusion protein is comprised of a light chain (SEQ ID NO: 6) and a heavy chain (SEQ ID NO: 7). In the light chain an active GLP-1 analogue is fused through a flexible 4x(GlyGlyGlySer) linker, to the N-terminus of the VL domain of the anti TNP-Fab of Example 2 (SEQ ID NO: 4). In the heavy chain the C-terminus of the CH I domain is extended to include a lOxHistidine purification tag. Thus, the light chain may be referred to as "GLP-l-(GGGS)4-aTNP light chain" and the heavy chain as "aTNP-(His)io heavy chain".

Nucleic acid sequences encoding the anti TNP-Fab fusion were cloned into a pTT- derived expression vector as described in Example 1.

To express the fusion protein, Expi293F cells (ThermoFisher Scientific, cat. no. A14527) were grown in suspension culture in Expi293 Expression™ medium

(ThermoFisher Scientific, cat. no. A1435101) to a cell density of 3xlOE6 cells/ml.

Transient transfection was performed with a mix of ExpiFectamine™ 293 Reagent (ExpiFectamine™ 293 Transfection Kit, ThermoFisher Scientific, cat. no. A14525) and plasmid DNA encoding the two protein chains. Transfection Enhancers 1 and 2 from the ExpiFectamine™ 293 Transfection Kit were added the day after transfection. The expression supernatant was harvested by centrifugation and filtering of the cell culture 4 days after the transfection.

The cell culture supernatant containing secreted GLP-1 anti TNP-Fab fusion protein was applied to 2x 5 ml HisTrap Excel columns (GE Healthcare) equilibrated in 50 mM NaP (sodium phosphate), 300 mM NaCI, 10 mM Imidazole pH 7.5. The bound fusion protein was eluted with 50 mM NaP, 300 mM NaCI, 500 mM Imidazole pH 7.5. Fractions were pooled and applied to a Superdex 200 16_600 column (GE Healthcare) equilibrated in PBS. The purified protein was sterilized by filtration through a 0.2 pm filter unit. The purity of the fusion protein was analysed by non-reducing SDS-PAGE and produced a single band indicating >90% purity and size-exclusion HPLC showed 12% high molecular weight species. The fusion protein identity was confirmed by mass spectrometry. The measured intact average mass 52977 Da and peptide mass mapping of tryptic peptides (data not shown) showed that the N-terminal glutamic acid on the VH I domain (SEQ ID NO: 7) was in fact a pyroglutamic acid in the purified protein.

Example 6:

Transient expression of GLP-1 anti TfR-Fab fusion protein

This example describes the expression of a fusion protein shown in Fig. 6 (SEQ ID NO: 11) of an anti TfR-Fab and a GLP-1 analogue. Light chain (SEQ ID NO : 9) :

HGEGTFTSDVSSYLEEQAAREFIAWLVKGRPGGGSGGGSGGGSGGGSDIQMTQSPASLSA SLEEIV

TITCQASQDIGNWLAWYQQKPGKSPQLLIYGATSLADGVPSRFSGSRSGTQFSLKIS RVQVEDIGIY

YCLQAYNTPWTFGGGTKLELKRADAAPTVSIFPPSSEQLTSGGASVVCFLNNFYPKD INVKWKIDGS

ERQNGVLNSWTDQDSKDSTYSMSSTLTLTKDEYERHNSYTCEATHKTSTSPIVKSFN RNEC

Heavy chain (SEQ ID NO : 10) :

EVQLVESGGGLVQPGNSLTLSCVASGFTFSNYGMHWIRQAPKKGLEWIAMIYYDSSKMNY ADTVKG

RFTISRDNSKNTLYLEMNSLRSEDTAMYYCAVPTSHYVVDVWGQGVSVTVSSAKTTP PSVYPLAPGS

AAQTNSMVTLGCLVKGYFPEPVTVTWNSGSLSSGVHTFPAVLQSDLYTLSSSVTVPS STWPSETVTC

NVAHPASSTKVDKKIVPRDCGHHHHHHHHHH

The two chain Fab fusion protein is comprised of a light chain (SEQ ID NO : 9) and a heavy chain (SEQ ID NO : 10). In the light chain an active GLP-1 analogue is fused through a flexible 4x(GlyGlyGlySer) linker to the N-terminus of the VL domain of an anti TfR-Fab. In the heavy chain the C-terminus of the CH I Fab domain is extended to include a lOxHistidine purification tag. Thus, the light chain may be referred to as "GLP-1- (GGGS) ₄ -aTfR light chain" and the heavy chain as ^M aTfR-(His)io heavy chain".

Nucleic acid sequences encoding the anti TfR-Fab fusion were cloned into a pTT- derived expression vector as described in Example 1.

The fusion protein was expressed in expi293F cells and the supernatant harvested as described in Example 5.

The cell culture supernatant containing secreted GLP-1 anti TfR-Fab fusion protein was applied to 2x 5 ml HisTrap Excel columns (GE Healthcare) and further purified and sterilised as described in Example 5 with the addition of a stringency wash with 50 mM NaP (sodium phosphate), 300 mM NaCI, 30 mM Imidazole pH 7.5 before elution from the HisTrap Excel columns.

The purity of the fusion protein was analysed by SDS-PAGE and size-exclusion HPLC showing greater than 95% purity. The fusion protein identity was confirmed by mass spectrometry. The average intact mass was measured to be 53243.7 Da by mass spectrometry, thereby confirming the identity of the purified protein.

Example 7:

Transient expression of GLP-1 control Fab fusion protein

This example describes the expression of a fusion protein shown in Fig. 7 (SEQ ID NO : 12) of another control Fab and a GLP-1 analogue. The plasmid DNA encoding the light chain (SEQ ID NO: 6), of this this protein is the same as used in Example 5 and the plasmid DNA encoding the heavy chain (SEQ ID NO: 10) of this protein is the same as used in Example 6. This leads to a hybrid construct with no TfR binding albeit still with an active GLP-1 element similar to the GLP-1 anti-TfR Fab of Example 6 (SEQ ID NO: 11).

Nucleic acid sequences encoding the control Fab fusion were cloned into a pTT- derived expression vector as described in Example 1.

The fusion protein was expressed in expi293F cells and the supernatant harvested as described in Example 5.

The cell culture supernatant containing secreted fusion protein was applied to 2x 5ml HisTrap Excel columns (GE Healthcare) and further purified and sterilised as described in Example 5.

The purity of the Fab fusion protein was analysed by SDS-PAGE and size- exclusion HPLC showing greater than 95% purity. The fusion protein identity was confirmed by mass spectrometry. The average intact mass was measured to be 53662.9 Da by mass spectrometry, thereby confirming the identity of the purified protein.

Example 8:

Preparation of fluorescence labelled GLP-1 anti TfR-Fab fusion protein

This example describes the fluorescence labelling of the GLP-1 anti TfR-Fab fusion protein of Example 6 (SEQ ID NO: 11). The fluorescence labelled compound is shown in Fig. 8 (please note that the labelling is randomly attached and not just on the N-terminal as depicted).

To 2 ml of the GLP-1 anti TfR-Fab fusion protein of Example 6 (containing about 92 nmol) was added 120 nmol VivoTag750-NHS, which is an amine reactive (NHS ester) near-infrared fluorochrome for labelling biomolecules for in vivo imaging applications, commercially available from PerkinElmer.

Usually this fluorescent label is fairly quantitatively attached to Fabs but in this case only very limited labelling was obtained. 600 nmol VivoTag750-NHS was added and the coupling proceeded overnight. Next day, the solution was applied to a PD-10 column (GE Healthcare) running in PBS and the labelled material collected in 1.6 ml PBS. The majority of the blue colour eluted later as hydrolysed tag. The concentration was determined to 32 nmol/ml by CLND (Chemiluminescent Nitrogen Detection) and the solution was aliquoted into 5 Eppendorf tubes with 320 pL in each and frozen.

LCMS showed a mixture 0, 1,2,3 labels per molecule (average about 1). Example 9:

Preparation of fluorescence labelled GLP-1 control Fab fusion protein

This example describes the fluorescence labelling of the GLP-1 control Fab fusion protein of Example 7 (SEQ ID NO: 12). The fluorescence labelled compound is shown in Fig. 9 (please note that the labelling is randomly attached and not just on the N-terminal as depicted).

To 2 ml of the GLP-1 control Fab fusion protein of Example 7 (SEQ ID NO: 12) (containing about 138 nmol) was added 160 nmol VivoTag750-NHS (PerkinElmer).

As described in Example 8 an additional amount of the fluorochrome had to be added, in this case 900 nmol. The coupling and collection of the labelled material was as described in Example 8 (in this case the labelled material was collected in 2 ml PBS). The majority of the blue colour eluted later as hydrolysed tag. The concentration was determined to 49 nmol/ml by CLND as described in Example 8, and the solution was aliquoted into 5 Eppendorf tubes with 400 pL in each and frozen.

LCMS showed a mixture of 0, 1,2,3 labels per molecule (average about 1).

Example 10:

Synthesis of GLP-1 peptide used for conjugation

This example describes the synthesis of a GLP-1 peptide shown in Fig. 10.

The peptide of Fig. 10 without the bromo acetamide (BrAc) derivatisation at the epsilon amino group of the C-terminal Lys residue at position 58 has SEQ ID NO: 13 (position 58 corresponds to position 52 of SEQ ID NO: 13) .

The peptide of SEQ ID NO: 13 may be defined as the analogue (8Aib, 22E, 26R, 34R) of GLP-l(7-37) with a C-terminal peptidic linker (viz., residues no. 38-58 added onto GLP-l(7-37), where residues 38-58 correspond to residues 32-52 of SEQ ID NO:

13). The peptide linker is: G-Q-A-P-G-Q-A-P-G-Q-A-P-G-Q-A-P-G-Q-A-P-K and has SEQ ID NO: 14.

The Fmoc-protected amino acid derivatives used were the standard

recommended : Fmoc-Ala-OH, Fmoc-Arg(Pbf)-OH, Fmoc-Asn(Trt)-OH, Fmoc-Asp(OtBu)- OH, Fmoc-Gln(Trt)-OH, Fmoc-Glu(OtBu)-OH, Fmoc-Gly-OH, Fmoc-His(Trt)-OH, Fmoc-lle- OH, Fmoc-Leu-OH, Fmoc-Lys(Boc)-OH, Fmoc-Phe-OH, Fmoc-Pro-OH, Fmoc-Ser(tBu)-OH, Fmoc-Thr(tBu)-OH, Fmoc-Trp(Boc)-OH, Fmoc-Tyr(tBu)-OH, Fmoc-Val-OH and Fmoc- Lys(Mtt)-OH, Fmoc-Aib-OH supplied from e.g. Anaspec, Bachem, Iris Biotech, or

NovabioChem. Where nothing else is specified the natural L-form of the amino acids were used. The N-terminal amino acid was employed with N-terminal Boc protection at the alpha amino group (e.g. Boc-His(Trt)-OH). SPPS was performed using Fmoc based chemistry on a SymphonyX Solid Phase Peptide Synthesizer from Protein Technologies (Tucson, AZ 85714 U.S.A.). Preloaded Fmoc-Lys(Mtt)-wang LL resin (Novabiochem, loading 0.33 mmol/g) was used for the peptide synthesis. Fmoc-deprotection was achieved with 20% piperidine in DMF. Peptide double couplings for lh were performed in the first 20 cycles and single couplings for 2h with capping (using acetic acid anhydride) in the rest. A mixture of Amino acid/Oxyma Pure/DIC solutions (0.3 M/0.3 M/0.3 M in DMF at a molar excess of 4 fold) were used for all couplings.

The resin was transferred to a glass reactor fitted with a filter. HFIP/DCM/TIPS 75:22:3 solution (70 ml) was added, stirred for 20 min, filtered and repeated IX with additional 60 ml, and washed with DCM (3x65 ml) and DMF (3x65 ml).

A mixture of bromoacetic acid (3.0 g, 12 eq, 22 mmol) and DIC (1.39 g, 6 eq, 11 mmol) in DMF/DCM 1 : 1 (50 ml) was standing for 10 min and poured onto the resin. Standing O/N, stirring. Filtered and washed with DMF (4x60ml) and DCM (4x60ml).

The peptide was cleaved from the resin by stirring in a mixture of (95 TFA : 2.5 TIPS : 2.5 water, 80 ml) for 3h at RT. Filtered and the resin was washed with a small amount of TFA.

The solution was split into 12 plastic vials, and dry-ice cooled diethylether (25 ml) was added to each vial and filled up with an additional approximately 20 ml of ether. The precipitate was spinned down.

The supernatant was removed and the product was purified on preparative

HPLC.

LCMS_34: Rt = 2.37 min, M/4 = 1378.3

Example 11:

Synthesis of GLP-1 anti TfR-Fab conjugate

This example describes the synthesis of a conjugate of a GLP-1 peptide and an anti TfR-Fab. The structure of the conjugate is shown in Fig. 11.

The element refers to the tripeptide of His-Aib-Glu (residues 434-436 of

Fig. 11).

The N-terminal part of Fig. 11 (residues 1-433) is identical to the anti TfR-Fab of SEQ ID NO: 3 (see Examples 1 and 3).

The C-terminal part of Fig. 11 (residues 434-485) is identical to SEQ ID NO: 13 (see Example 10).

As shown in Fig. 11, the epsilon amino group of Lys 485 and the thiol group of Cys 377 are interconnected via Chem. 1 : -CH2-C(0)-. To a solution of the anti TfR-Fab of Example 3 (41 ml with 252.4 mg, 5348 nmol of anti TfR-Fab) dissolved in PBS buffer, pH 7.4 was added 522 ul of a 200 mM EDTA buffer.

TCEP (5.05 pmol/ml, 1.05 eq., 1112 pl_, 5615 nmol) was added to the protein solution.

Preparation of TCEP solution: TCEP, 0.5 M, pH 7, MW: 286 g/mol.

25 pl_ of the 0.5M TCEP solution was diluted with 2475 mI_ buffer (20 mM TEA, 2 mM EDTA, pH 8.5 buffer). Standing for about 2-3 h or until fully reduced.

The peptide from Example 10, 3.6 eq, was dissolved in DMSO (1.5 ml) and added to the Fab solution and the flask was rinsed with a little TRIS buffer pH 9. pH was adjusted with TRIS buffer pH 9 to pH 8.3 (start pH = 6.5), standing O/N.

The reaction mixture was concentrated by spinning in a Vivaspin 20, 10K filter to give 13 ml and was purified on an Akta purifier from GE Healthcare.

Column: Hiload 26/600, superdex 200pg SEC

Buffer: PBS-buffer, pH 7.4

The following characteristics of the resulting product were determined :

LCMS_36: Rt = 3.52 min; m/30 = 1751.4; m/32 = 1642.0

Example 12:

Synthesis of GLP-1 control-Fab conjugate

This example describes the synthesis of a conjugate of a GLP-1 peptide and a control Fab. The structure of the conjugate is shown in Fig. 12.

H

The element A refers to the tripeptide of His-Aib-Glu (residues 439-441 of

Fig. 12).

This conjugate was prepared by Cys alkylation of the control Fab of Example 4 with the bromo acetyl peptide of Example 10 (Fig. 10) using the same protocol as described in Example 11.

The N-terminal part of Fig. 12 (residues 1-438) is identical to the control Fab of SEQ ID NO: 5 (see Examples 2 and 4).

The C-terminal part of Fig. 12 (residues 439-490) is identical to SEQ ID NO: 13 (see Example 10).

As shown in Fig. 12, the epsilon amino group of Lys 490 and the thiol group of Cys 382 are interconnected via Chem. 1 : -CH2-C(0)-.

The following characteristics of the resulting product were determined :

LCMS_36: Rt = 3.01 min; m/29 = 1832.1; m/31 = 1713.9 Example 13:

Synthesis of GLP-1 peptide used for conjugation

This example describes the synthesis of a GLP-1 peptide derivative of Fig. 13.

Fig. 13 incorporates the GLP-1 analogue (8G, 22E, 26R, 34R, 36K) of GLP-1(7- 37) (SEQ ID NO: 15).

Fig. 13 also incorporates a non-peptidic linker which can be defined as Chem. 5: -Ado-gGlu-Ado-, where Ado refers to 8-amino-3,6-dioxaoctanoic acid and gGlu to the amino acid Glu when attached to another molecule via an amide bond involving the carboxylic acid group at the gamma position of the side chain of Glu. A di-radical of Ado has the formula of Chem. 3 : -NH-(CH2)2-0-(CH2)2-0-CH2-C(0)-. A di-radical of gGlu has the formula of Chem. 4: -NH-CH(C00H)-(CH2)2-C(0)-). One Ado of the linker is attached to the epsilon amino group of the Lys residue at position 36 of the GLP-1 analogue. The other Ado of the linker is derivatised with bromoacetamide, for later conjugation to an anti-TfR Fab or a control Fab, as described in Examples 15 and 14, respectively.

The Fmoc-protected amino acid derivatives used were the standard

recommended : Fmoc-Ala-OH, Fmoc-Arg(Pbf)-OH, Fmoc-Asn(Trt)-OH, Fmoc-Asp(OtBu)- OH, Fmoc-Gln(Trt)-OH, Fmoc-Glu(OtBu)-OH, Fmoc-Gly-OH, Fmoc-His(Trt)-OH, Fmoc-lle- OH, Fmoc-Leu-OH, Fmoc-Lys(Boc)-OH, Fmoc-Phe-OH, Fmoc-Pro-OH, Fmoc-Ser(tBu)-OH, Fmoc-Thr(tBu)-OH, Fmoc-Trp(Boc)-OH, Fmoc-Tyr(tBu)-OH, Fmoc-Val-OH and Fmoc- Lys(Mtt)-OH, Fmoc-Aib-OH supplied from e.g. Anaspec, Bachem, Iris Biotech, or

NovabioChem. Where nothing else is specified the natural L-form of the amino acids were used. The N-terminal amino acid was employed with N-terminal Boc protection at the alpha amino group (e.g. Boc-Tyr(tBu)-OH for peptides with Tyr at the N-terminus). SPPS was performed using Fmoc based chemistry on a SymphonyX Solid Phase Peptide Synthesizer from Protein Technologies (Tucson, AZ 85714 U.S.A.). Resins employed for the preparation of C-terminal peptide were Fmoc-Gly-wang LL resin (Novabiochem, loading e.g. 0.29 mmol/g). Fmoc-deprotection was achieved with 20% piperidine in DMF. Peptide couplings for lh were performed using Amino acid/Oxyma Pure/DIC solutions (0.3 M/0.3 M/0.3M in DMF at a molar excess of 4 fold). The Mtt group was removed by washing the resin with HFIP/TIS/DCM (75:2.5:22.5, v/v/v) (3 x 10 mL x 30 min) before washing with DCM (5 x 10 mL). Synthesis was continued with Fmoc-Ado-OH and Fmoc- gGlu(OtBu)-OH as described above. The resin was transferred to a glass reactor fitted with a filter. A mixture of bromoacetic acid (0.840 g, 12 eq, 6 mmol) and DIC (0.400 ml, 5.2 eq, 2.6 mmol) in DMF (16 ml) was standing for 10 min and poured onto the resin.

Standing O/N. Filtered and washed with DMF (4x6 ml) and DCM (4x6 ml).

The peptide was cleaved from the resin by stirring in a mixture of (95 TFA : 2.5 TIPS : 2.5 Water, 80 ml) for 3h at RT. Filtered and the resin was washed with a small amount of TFA. The solution was split into 12 plastic vials and dry-ice cooled diethylether (25 ml) was added to each vial and filled up with an additional approximately 20ml of ether. The precipitate was spinned down. The supernatant was removed and the product was purified on preparative HPLC.

LCMS_34: Rt = 2.45 min, M/3 = 1328.2

Example 14:

Synthesis of GLP-1 control Fab conjugate

This example describes the synthesis of a conjugate of a GLP-1 peptide and a control Fab. The structure of the conjugate is shown in Fig. 14.

This conjugate was prepared by Cys alkylation of the control Fab of Example 4 with the bromo acetyl peptide derivative of Example 13 (Fig. 13) using the same protocol as described in Example 11, except that the peptide derivative of Example 13 was used instead, in an amount of 3.6 eq.

The N-terminal part of Fig. 14 (residues 1-438) is identical to the control Fab of SEQ ID NO: 5 (see Examples 2 and 4).

The C-terminal part of Fig. 14 (residues 439-469) is identical to Fig. 13, except for the fact that the distal Ado of the linker of Chem. 5: -Ado-gGlu-Ado- which is bromo acetamide derivatised in Fig. 13, is here connected to the thiol group of Cys 382 via a Chem. 1 chemical linker (Chem. 1 : -CH2-C(0)-). The entire linker thus is Chem. 2: -CH2- C(0)-Ado-gGlu-Ado-.

The following characteristics of the resulting product were determined :

LCMS_36: Rt = 3.14 min; m/30 = 1720.0; m/32 = 1612.6

Example 15: Synthesis of GLP-1 anti TfR-Fab conjugate

This example describes the synthesis of a conjugate of a GLP-1 peptide and an anti TfR-Fab. The structure of the conjugate is shown in Fig. 15.

This conjugate was prepared by Cys alkylation of the anti TfR-Fab of Example 3 with the bromo acetyl peptide derivative of Example 13 (Fig. 13) using the same protocol as described above for Example 14. The N-terminal part of Fig. 15 (residues 1-433) is identical to the anti TfR-Fab of SEQ ID NO: 3 (see Examples 1 and 3).

The C-terminal part of Fig. 15 (residues 434-464) is identical to Fig. 13, except for the fact that the distal Ado of the linker of Chem. 5: -Ado-gGlu-Ado- which was bromo acetamide derivatised in Fig. 13 is here connected to the thiol group of Cys 377 via a chemical linker (Chem. 1 : -CH2-C(0)-). The entire linker thus is Chem. 2: -CH2- C(0)-Ado-gGlu-Ado-.

The following characteristics of the resulting product were determined :

LCMS_36: Rt = 3.55 min; m/30 = 1700.4; m/32 = 1594.2

Pharmacoloqical methods

Example 16:

In vivo imaging of fluorescently labelled GLP-1 TfR fusion proteins

This example examines the access to various GLP-1 receptor expressing areas of the brain of an "active" and an "inactive" GLP-l-Fab fusion protein ("active" in the sense that it binds to the TfR, "inactive" in the sense that it does not bind to the TfR). The two fusion proteins include one and the same GLP-1 receptor agonist compound but differ in the Fab part. This is for the purpose of determining if there is a substantial difference between the active and the inactive fusion protein in their access to brain areas that are protected by the blood brain barrier (BBB). The study is an acute study in mice.

Compound administration:

Mice (C57BL/6J, male, Taconic, n=4/group) received a single i. v. dose of the fluorescence labelled GLP-1 anti TfR-Fab fusion protein of Example 8 ("GLP- 1-Active TfR"), 120 nmol/kg, or of the fluorescence labelled GLP-1 control Fab fusion protein of Example 9 ("GLP-l-Inactive TfR"), 120 nmol/kg. The mice were anesthetised with isoflurane 6 hours ("6H") following administration of compound and euthanised by transcardiac perfusion with 10 ml heparinised (10 U/ml) saline followed by 10 ml 10% neutral buffered formalin (NBF). Brains were removed and immersed into 10% NBF and stored at 4°C O/N until further processed.

Clearing of tissue:

The brains were dehydrated and saturated in dibenzyl ether (DBE) to minimise scattering of light during scanning by removal of water, and by matching of the refractive index. Brain tissue was dehydrated at room temperature in graded tetrahydrofuran (THF) diluted in demineralised H2O (w/v) 50/80/96/99%/2 x 100% 6-12 hours for each step. The brains were then cleared at room temperature in 3 x DBE for 6-12 hours each step.

Scanning with Light Sheet Fluorescence Microscopy (LSFM) :

Brain samples were scanned using an UltraMicroscope II LSFM system (Lavision Biotec, Bielefeld, Germany) in 10.32 pm isotropic resolution. Data acquisition was performed using a 620/60 nm excitation filter and 680/30 nm emission filter for imaging autofluorescence, and a 710/75 nm excitation filter and 780/40 nm emission filter for imaging specific signals.

Image analysis:

Images are single plane images of a z plane (2D). Spectral unmixing was performed to minimise the contribution of tissue auto-fluorescence in the images of the specific signals presented. The unmixing was carried out in Imaris (Release 7.6.5, Zurich, Switzerland) on imported tiff-files. The estimated auto-fluorescence contribution in the specific channel was calculated and removed based on ratios of voxel intensities between selected voxels in the unspecific recording and the corresponding voxels in the specific channel. A ratio was computed for 40 sets of voxels selected based on the histogram of the unspecific recording. The unmixing algorithm was written in Matlab (Release 2012b, MathWorks, Natick, Massachusetts, United States) and applied as a XT-plugin in Imaris.

Images were registered to a common reference atlas space for visualisation and quantification using the Elastix software library version 4 (Klein, Stefan, et al. "Elastix: a toolbox for intensity-based medical image registration." IEEE transactions on medical imaging 29.1 (2010) : 196-205). A source image Is(x) was registered to the target atlas image IT(X) by finding a coordinate transformation T(x) that made Is(T(x)) spatially aligned with IT(X) . We used an affine coordinate transformation for initialisation followed by a nonrigid b-spline coordinate transformation which was iteratively optimised with respect to the mutual information between the source and the target. Registration parameters for mapping of LSFM data were similar to those used in Renier et al.

Following image registration, a manual quality check was made for correct registration. If parts of the registered brains were out of alignment, the measurement connected to these individual regions were removed. Thus, for some brain regions, the number of samples varied between 3 and 4 within each group. Quantification was performed by summation of the unmixed intensity value of all voxels within the individual brain regions, indicating the total fluorescence signal from the compounds in these regions.

The total signal is reported in arbitrary unit ("AU"). Total signal is a summation of the unmixed voxel intensity values in individual brain regions. The total signal thus indicates the total fluorescence signal from the compounds in these regions. (Renier, Nicolas, et al. "Mapping of Brain Activity by Automated Volume Analysis of Immediate Early Genes."

Cell (2016)).

Results:

The results of the image analyses are shown in Figures 16A, 16B, 16C, and 16D.

In Fig. 16A, the total intensity of the fluorescently labelled compounds ("GLP-1- Active TfR" and "GLP-l-Inactive TfR") was quantified in brain regions expressing the GLP- 1 receptor (GLP-1R). The total signal is reported in arbitrary unit (AU) and represents averages. Fig. 16B shows intensity of fluorescently labelled compounds in examples of circumventricular organs expressing the GLP-1R but devoid of a BBB. Fig. 16C shows intensity of fluorescently labelled compounds in examples of brain structures expressing the GLP-1R which are protected by the BBB. Fig. 16D shows intensity of fluorescently labelled compounds in examples of brain regions that do not express the GLP-1R. The abbreviations used for the various brain regions are defined in the list of abbreviations.

Fig 16A shows that in the GLP-1R expressing regions the fluorescence signals were higher from the GLP-l-Active TfR compared to the GLP-l-Inactive TfR.

The differences between the active and inactive TfR compounds were smaller in the circumventricular organs expressing the GLP-1R but devoid of a BBB (Fig. 16B). The TfR is expressed both in capillaries and brain parenchyma. Thus, the slightly higher contribution from the active TfR compound in these regions indicates contribution from the TfR binding of the compounds in addition to binding by the GLP-1R.

The signals were also quantified in brain regions that do not express the GLP-1R (Fig 16D). Here the signals were almost the same with both compounds where the slightly higher contribution from the active TfR compound again indicates contribution from the TfR binding of the compounds.

An example of brain structures which are protected by the BBB are given in Fig 16C. Here the signal was highly increased from the GLP-l-Active TfR compared to the GLP-l-Inactive TfR indicating that the GLP-1 component was carried to regions not readily accessible for GLP-1 by itself. Although binding to the TfR may also contribute to this signal, the difference between the two compounds in regions expressing the GLP-1R is much larger than in regions without the GLP-1R (Fig 16D). This suggests that the majority of the signal originated from binding to the GLP-1R.

Conclusion : The distinct and marked difference in signal strength between the active and the inactive TfR compounds which is seen in the brain structures that are protected by the BBB (Fig. 16C) shows that the coupling of a GLP-1R agonist to an anti-TfR Fab can be used as a means to target GLP-1R agonists to areas of the brain which are not readily accessible by GLP-1R agonists. Some of these areas, such as the ACB, DMH, LHA, and LH have been shown to be important in regulation of body weight and reward related food intake.

Example 17: GLP-1 in vitro potency

The purpose of this example is to test the GLP-1 activity of the active and inactive GLP-1 fusions and conjugates disclosed herein (where the terms active/inactive refer to the ability to bind/not bind, respectively, to the transferrin receptor (TfR)). The GLP-1 activity is determined as GLP-1 in vitro potency and is the measure of human GLP- 1 receptor activation in a whole cell assay.

Principle:

In vitro potency was determined by measuring the response of the human GLP-1 receptor in a reporter gene assay. The assay was performed in a stably-transfected BHK cell line that expresses the human GLP-1 receptor and contains the DNA for the cAMP response element (CRE) coupled to a promoter and the gene for firefly luciferase (CRE luciferase). When the human GLP-1 receptor is activated it results in the production of cAMP, which in turn results in the luciferase protein being expressed. When assay incubation is completed the luciferase substrate (luciferin) is added and the enzyme converts luciferin to oxyluciferin to produce bioluminescence. The luminescence is measured as the readout for the assay.

In order to test the binding of the compounds to albumin, the assay may be performed in the absence of serum albumin as well as in the presence of a considerably higher concentration of serum albumin (1.0% final assay concentration). An increase of the in vitro potency, ECso value, in the presence of serum albumin would indicate an affinity to serum albumin and represents a method to predict a protracted

pharmacokinetic profile of the test substance in animal models.

Cell culture and preparation :

The cells used in this assay (clone FCW467-12A/KZ10-1) were BHK cells with BHKTS13 as a parent cell line. The cells were derived from a clone (FCW467-12A) that expresses the human GLP-1 receptor and were established by further transfection with CRE luciferase to obtain the current clone. The cells were cultured at 5% CO2 in Cell Culture Medium. They were aliquoted and stored in liquid nitrogen. Before each assay an aliquot was taken up and washed twice in PBS before being suspended at the desired concentration in the assay specific buffer. For 96-well plates the suspension was made to give a final concentration of 5xl0 ³ cells/well.

Materials:

The following chemicals were used in the assay: Pluronic F-68 (10% v/v) (Gibco 2404), human serum albumin (HSA) (Sigma A9511), ovalbumin (Sigma A5503), DMEM w/o phenol red (Gibco 11880-028), 1 M Hepes (Gibco 15630), Glutamax lOOx (Gibco 35050) and steadylite plus (PerkinElmer 6016757).

Buffers:

Cell Culture Medium consisted of DMEM medium with 10% w/v FBS (Fetal Bovine Serum; Invitrogen 16140-071), 1 mg/ml G418 (Invitrogen 15140-122), 240 nM MTX (methotrexate; Sigma M9929) and 1% w/v pen/strep (penicillin/streptomycin; Invitrogen 15140-122).

Assay Medium consisted of DMEM w/o phenol red, lOmM Hepes and lx

Glutamax. The 1% Assay Buffer consisted of 2% ovalbumin, 0.2% Pluronic F-68 and 2 % HSA in Assay Medium. The 0% Assay Buffer consisted of 2% w/v ovalbumin and 0.2% v/v Pluronic F-68 in Assay Medium.

Procedure:

1. Cell stocks were thawed in a 37 °C water bath.

2. Cells were washed three times in PBS.

3. The cells were counted and adjusted to 5xl0 ³ cells/50 pi (1x10 ^s cells/ml) in Assay Medium. A 50 mI aliquot of cells was transferred to each well in the assay plate.

4. Stocks of the test compounds were diluted to a concentration of 0.2 mM in 0% Assay Buffer for the 0% HSA CRE luciferase assay and 1% Assay Buffer for the 1% HSA CRE luciferase assay. Compounds were diluted 10-fold to give the following concentrations: 2xl0 ⁷ M, 2xl0 ⁸ M; 2xl0 ⁹ M, 2xlO ¹⁰ M, 2c10 ^h M, 2xl0 ¹² M, 2xl0 ¹³ M, and 2xl0 ¹⁴ M.

5. A 50 mI aliquot of compound or blank was transferred from the dilution plate to the assay plate. Compounds were tested at the following final concentrations: lxlO ⁷ M, lxlO ⁸ M; lxlO ⁹ M, lxlO ¹⁰ M, lxlO ¹¹ M, lxlO ¹² M, lxlO ¹³ M, and lxlO ¹⁴ M.

6. The assay plate was incubated for 3 h in a 5% CO2 incubator at 37 °C. 7. The assay plate was removed from the incubator and allowed to stand at room temperature for 15 min.

8. A 100 pi aliquot of steadylite plus reagent was added to each well of the assay plate (reagent was light sensitive).

9. Each assay plate was covered with aluminum foil to protect it from light and shaken for 30 min at room temperature.

10. Each assay plate was read in a microtiter plate reader.

Calculations and Results:

The data from the plate reader were transferred to GraphPad Prism software.

The software performed a non-linear regression (log(agonist) vs response). ECso values which were calculated by the software and reported in pM are shown in Table 1 below. A minimum of two replicates was measured for each sample on separate assay plates. The reported values are averages of the duplicates. For some examples more than one assay was run and in this case the data reported are the averages of the averaged duplicates.

Table 1 : In vitro potency

All compounds have potency data that confirms that they are GLP-1 receptor agonists. All compounds had a good GLP-1 in vitro potency corresponding to an ECso at 0% HSA of below 1000 pM.

Example 18: GLP-1 receptor binding The purpose of this example is to test the GLP-1 receptor binding activity in vitro of the active and inactive GLP-1 fusions and conjugates disclosed herein (where the terms active/inactive refer to the ability to bind/not bind, respectively, to the transferrin receptor (TfR)). The receptor binding is a measure of the affinity of the compounds for the human GLP-1 receptor.

Principle:

The binding to the human GLP-1 receptor was measured in a competitive binding assay. In this type of assay a labelled ligand (in this case ¹²⁵ I-GLP-1) was bound to the receptor. Each test compound was added in a series of concentrations to isolated membranes containing the human GLP-1 receptor and displacement of the labelled ligand was monitored. The receptor binding was reported as the concentration at which half of the labelled ligand was displaced from the receptor, the ICso value. In order to test the binding of the derivatives to albumin, the assay may be performed in a very low concentration of serum albumin (max. 0.001% (w/v) final assay concentration as well as in the presence of a considerably higher concentration of serum albumin (2.0% (w/v) final assay concentration). An increase of the ICso value in the presence of serum albumin indicates an affinity to serum albumin.

Materials:

The following chemicals were used in the assay: human serum albumin (HSA) (Sigma A1653), DMEM w/o phenol red (Gibco 11880-028), Pen/strep (Invitrogen 15140- 122), G418 (Invitrogen 10131-027), 1 M HEPES (Gibco 15630), EDTA (Invitrogen 15575- 038), PBS (Invitrogen 14190-094), fetal calf serum (Invitrogen 16140-071), EGTA, MgCh (Merck 1.05832.1000), Tween 20 (Amresco 0850C335), SPA particles (wheat germ agglutinin (WGA) SPA beads, Perkin Elmer RPNQ0001), [ ¹²⁵ I]-GLP-l]-(7-36)NH ₂

(produced in-house), OptiPlate™-96 (Packard 6005290).

Buffer 1 consisted of 20 mM Na-HEPES plus 10 mM EDTA and pH was adjusted to 7.4. Buffer 2 consisted of 20 mM Na-HEPES plus 0.1 mM EDTA and pH was adjusted to 7.4. Assay buffer consisted of 50 mM HEPES supplemented with 5 mM EGTA, 5 mM MgCh, 0.005% w/v Tween 20 and pH was adjusted to 7.4. An 8% albumin stock consisted of HSA dissolved at 8% (w/v) in assay buffer. An 0.02% albumin stock consisted of HSA dissolved at 0.02% (w/v) in assay buffer.

Cell culture and membrane preparation: The cells used in this assay (clone FCW467-12A) were BHK cells with BHKTS13 as a parent cell line. The cells express the human GLP-1 receptor. The cells were grown at 5% CO2 in DMEM, 10% w/v fetal calf serum and 1% w/v Pen/Strep

(Penicillin/Streptomycin). To make a membrane preparation the cells were grown to approximately 80% confluence. The cells were washed twice in phosphate-buffered saline and harvested. The cells were pelleted using a brief centrifugation and the cell pellet was kept on ice. The cell pellet was homogenised with ULTRA-THURRAX™ dispersing instrument for 20-30 seconds in a suitable amount of buffer 1 (e.g., 10 ml). The homogenate was centrifuged for 15 minutes. The pellet was re-suspended

(homogenised) in 10 ml buffer 2 and centrifuged. This step was repeated once more. The resulting pellet was re-suspended in buffer 2 and the protein concentration was determined. The membranes were aliquoted and stored at minus 80°C.

Procedure:

1. For the receptor binding assay in the presence of low HSA (0.001%) 50 pi of the assay buffer was added to each well of an assay plate. Assay continued with step 3.

2. For the receptor binding assay in the presence of high HSA (2.0%) 50 pi of the 8% albumin stock was added to each well of an assay plate. Assay continued with step 3.

3. Test compounds were serially diluted to give the following concentrations: 8xl0 ⁷ M, 8xl0 ⁸ M, 8xl0 ⁹ M, 8x10 ¹⁰ M, 8x10 ¹¹ M, 8x10 ¹² M and 8x10 ¹³ M. Twenty-five mI were added to appropriate wells in the assay plate.

4. Cell membrane aliquots were thawed and diluted to their working concentration. Fifty mI were added to each well in the assay plate.

5. WGA SPA beads were suspended in assay buffer at 20 mg/ml. The suspension was diluted to 10 mg/ml in assay buffer just prior to addition to the assay plate. Fifty mI were added to each well in the assay plate.

6. The incubation was started by adding 25 mI of 480 pM solution of [ ¹²⁵ I]-GLP-l]-(7- 36)NH2 to each well of the assay plate. A 25 mI aliquot was reserved for measuring total counts/well.

7. The assay plate was incubated for 2 h at 30 °C.

8. The assay plate was centrifuged for 10 min.

9. The assay plate was read in a Packard TopCount NXT instrument.

Calculations: The data from the TopCount instrument were transferred to GraphPad Prism software. The software averaged the values for the replicates and performed a non-linear regression. IC50 values were calculated by the software and reported in nM. Results:

The following results were obtained :

Table 2: GLP-1 receptor binding

All derivatives had an IC50 (low albumin) below 60 nM.

Example 19: Transferrin receptor uptake inhibition

The purpose of this example is to test the transferrin receptor binding activity in vitro of exemplary active and inactive GLP-1 fusions and conjugates disclosed herein (where the terms active/inactive refer to the ability to bind/not bind, respectively, to the transferrin receptor (TfR)). The transferrin receptor binding is a measure of affinity of a compound for the mouse transferrin receptor.

Principle:

The capacity of the conjugates of Examples 11-12 and 14-15 to inhibit the uptake of a transferrin receptor targeted antibody was measured in a competitive uptake assay. The two compounds in each set of Fab conjugates (Examples 11 and 12; and Examples 15 and 14) include the same GLP-1R agonist part but differ when it comes to the Fab part, which is "active" and "inactive", respectively, as regards binding to the TfR. Each compound was added in a series of concentrations to a commercially available mouse fibroblast cell line (MEF-1) along with a set concentration of a

fluorescently-labelled Fab fragment (described below). The inhibition of cellular uptake of the labelled Fab fragment by each compound was reported as the concentration at which half-maximal signal was obtained (ICso value).

Materials:

Preparation of fluorescently-labelled Fab fragment:

The fluorescently-labelled Fab fragment was made by Cy5-labelling of the anti TfR-Fab of Example 1. The Cy5-labelled anti TfR-Fab is shown in Fig. IB, where the Cy5 dye is shown attached to Cys at position 377.

226 uL of a 200 mM EDTA buffer was added to a solution of the anti TfR-Fab of Example 1 (6.15 mg/ml, 17.5 ml). 25 uL of a TCEP solution (0.5 M, pH 7.0) was diluted with 2475 ul buffer (20 mM TEA, 2 mM EDTA, pH 8.5). 517uL (1.25eq, 2.9 umol) of the diluted TCEP solution was added to to the anti TfR-Fab solution. The solution was stored at minus 80°C.

An aliquot of 1 ml_ (6.15 mg, 131 nmol) of the above produced solution was bobbled with argon, and 1 mg of Cy5 (Amersham Cy5 Maleimide Mono-Reactive Dye) dissolved in 100 uL dry DMF was added. The mixture was kept under argon night O/N at 5°C.

The product was purified on a Zeba Spinn desalting column, 7K MWCO, 5ml using the following protocol: Column spinned at 1000G for 2 min. 2.4 ml washed out. Washed 3 timed with 2.4 ml PBS buffer, each time spinned at 1000G for 2 min. After the 3. time the product (lml+0.7ml wash) was placed on top of the column and spinned at 1000G for 2min. The product (1.7 ml) was collected. Calc, exact mass (m+z)/z

(dimensionless) : 47859.88. Found exact mass (m+z)/z (dimensionless): 47859.83.

The following additional chemicals were used in the assay:

DMEM (Gibco 31966-026), Pen/strep (Invitrogen 15140-122), G418 (Invitrogen 10131-027), 1 M HEPES (Gibco 15630), EDTA (Invitrogen 15575-038), fetal calf serum (Invitrogen 16140-071), PFA 4% (Chem Cruz Sc-281692), human serum albumin (Sigma A9511), collagen-coated imaging plate (Becton Dickinson 734-0319), DAPI Hoechst 33342 (Sigma B2261), Wheat germ agglutinin-Alexa 488 (Invitrogen W32464), HBSS (Gibco 14025). Procedure:

1. Prior to seeding into the cell plates the MEF-1 cells were grown at 5% CO2 in DMEM, 10% w/v fetal calf serum and 1% w/v penicillin/streptomycin. The day before each assay the cells were plated out at 5000 cells/well.

2. On the assay day the cells were washed with HBSS containing 0.1% (w/v) HSA.

3. Test compounds were serially diluted to give the following concentrations: 2xl0 ⁶ M, 2xl0 ⁷ M, 2xl0 ⁸ M, 2xl0 ⁹ M, 2xlO ¹⁰ M, 2X10 ¹¹ M, 2xl0 ¹² M and 2xl0 ¹³ M. 100 pi of each was added to appropriate wells in the assay plate.

4. To each of the wells the fluorescently-labelled Fab fragment was added to a final concentration of 5 nM.

5. The plates were incubated at 37 °C for 15 min. After incubation the buffer was removed and the cells were washed with HBSS containing 0.1% (w/v) HSA.

6. Fifty pi of cold 4% PFA was added in a fume hood. The plates were incubated for 10 min at room temperature followed by removal of the PFA.

7. Fifty mI of a labelling solution was added to the plates and incubated for 10 min. The labelling solution contained DAPI at 1 :200 (v/v; from 10 mg/ml stock) and WGA- Alexa488 at 1 :200 (v/v; from 5 mg/ml stock). The assay plate was incubated for 10 min at room temperature.

8. The assay plate washed three times with 100 mI of HBSS and 100 mI of HBSS was added to each well after the wash steps.

9. The assay plate was read in a high-content imaging instrument (IN Cell 2200 instrument, GE).

Calculations:

The data from the IN Cell instrument was analysed using the software that was supplied with the instrument. Nuclei were defined using the DAPI-stained nuclei and cell boundaries were constructed based on the WGA-Alexa488 stainings. The amount of fluorescence that was internalised was calculated by defining granules visible in the Cy5 channel of the instrument and corresponded to internalised Cy5-labelled Fab fragment. The number and intensity of granules in each cell were calculated. The readout from the instrument that was used for further calculations was the total area of the granules divided by the number of cells in each acquisition field. A typical experiment contained 32 fields using a 20x microscope objective per well. The data were transferred to GraphPad Prism software and plotted as concentration response curves with log M (M is molarity of the test compounds) on the x-axis and total granule area per cell on the y-axis. The software performed a non-linear regression. IC50 values were calculated by the software and reported in nM.

Results:

The following results were obtained :

Table 3 : Transferrin receptor uptake inhibition

All derivatives with an active Fab (TfR-Fab) had an IC50 value below 20 nM while all derivatives with an inactive Fab (control Fab) had an IC50 of > 1000 nM.

Example 20: In vivo studies in mice

The purpose of this example is to assess whether GLP-1 anti TfR-Fab conjugates are biologically active in vivo, with focus on obesity related features.

Part I of the present example reports a study of the food intake lowering capability of an "active" GLP-1 anti TfR-Fab conjugate of the invention in lean mice, dosed acutely i. v.

Part II reports a comparison of this "active" conjugate with an "inactive" GLP-1 control-Fab conjugate, to determine whether the TfR arm contributes to the results obtained in Part I.

Part III reports a study in diet-induced obese mice, treated sub-chronically, with a view to investigating whether the reduced food intake observed in Parts I and II translates into a weight loss.

I. In vivo studv of GLP-1 anti TfR-Fab conjugate of Example 11

Study design:

Food intake was studied in lean, male C57BI6J mice (Taconic, Denmark) dosed acutely i. v. (0-100 nmol/kg) with the GLP-1 anti TfR-Fab conjugate of Example 11 over a 24 hour period using a biodaq automated monitoring system (Research Diets, New Brunswick, NJ, USA). Briefly, mice were maintained under a 12-hour light dark cycle with food (Altromin, Cat. no.1314, Lippe, Germany) and tap water available ad lib. Prior to dosing, mice (n=4-6 per group) were assigned to 1 of 5 treatment groups that were evenly matched for body weight and placed in the biodaq system 8 days prior to dosing in order to acclimatise to testing and handling procedures. On the day of the experiment, mice were dosed i. v. at time 0 with either vehicle or the compound of Example 11 according to the table below:

Table 4: Groups

*PBS+0.05% polysorbate 80 Plasma Exposure:

After 24 h, blood samples were taken via the sublingual plexus into EDTA tubes containing a protease inhibitor cocktail (3.097 g K3EDTA (MW 406.53), Fluka 03665, dissolved in 50 ml Trasylol, 10.000 KIU (kallikrein inactivator units), 0.5 ml 20 mM Val- Pyr (16.48 mg Val-pyr/4 ml H2O pH is regulated to 7.4 by 1 M HCI) intended to stabilise the peptide during plasma analysis. Afterwards, animals were euthanised by

exsanguination followed by cervical dislocation while under anaesthesia. Blood samples were kept on ice and centrifuged (4 min x 4000 g) within 30 min, at 4°C. 50 pi plasma was then aliquoted into appropriately labelled tubes, placed on dry ice, and stored at minus 80°C, until analysis.

Plasma concentrations of the compound of Example 11 (and for Parts II and III also the compound of Example 12) were measured by LOCI (Luminescent Oxygen Channelling Immunoassay). In brief, the LOCI reagents included two latex bead reagents (donor and acceptor beads) and a biotinylated monoclonal antibody recognising an N- terminus epitope of GLP-1. The donor bead reagent, containing a photosensitive dye, was coated with streptavidin. The second bead reagent, the acceptor beads, was conjugated with another monoclonal antibody specific for the C-terminus of GLP-1, which made up the sandwich. During the assay, the three reactants were combined with the compound of Example 11 (or, in part II and III, either the compound of Example 11 or the compound of Example 12) in the plasma to form a bead-aggregate immuno-complex. Excitations of the complex released singlet oxygen molecules from the donor beads, which were channelled into the acceptor beads and triggered a chemiluminescence response that was measured in the Envision plate reader. The amount of light generated, reported as counts per second (cps), was proportional to the concentration of the analyte.

All calibrators, QC-samples and unknown samples in EDTA plasma were analysed in quadruplicates determinations.

The analytical procedure was as follows:

Working solutions of both acceptor beads/biotinylated antibody and donor beads were prepared in LOCI buffer.

One pL plasma sample/calibrator/QC was applied per well.

15 pL of the working solution containing biotinylated monoclonal antibody recognising an N-terminus epitope of GLP-1 and acceptor beads coated with monoclonal antibody specific for the C-terminus of GLP-1 were added to each well.

The plates were sealed and covered with a black lid.

The assay was allowed to incubate for 1 hour at 18-22°C.

30 pL of streptavid in-coated donor beads diluted in LOCI buffer was added under the green light.

The plate was covered with a black lid and incubated for 30 minutes at 18-22°C. Finally, the plate was read in Envision.

The analysis was carried out using an Envision reader controlled by Envision software. The instrument response (counts per seconds) was exported to LOCI

Calculator, where plasma concentrations were calculated.

Calibrators: A dilution line of the compound of Example 11 calibrator (and for parts II and III also of the compound of Example 12 calibrator) was prepared in a mouse plasma pool covering a range from 400.000 to 42 pmol/L. The mouse plasma pool was included as the 0 pM calibrator (blank). The calibrators were stored in Micronic tube at minus 18°C.

Performance:

Lower limit of quantification (LoQ) was determined to be 40 pmol/L. Upper limit of quantification (ULoQ) was determined to 25.000 pM. The imprecision (CV) was assessed by the measurement of three samples in 12 consecutive analytical runs and 21 determinations of each of the three samples in the same analytical run; the overall CV was < 10%.

Materials:

Reagents:

LOCI buffer: 25 mM Hepes (Sigma H-3375), 50 mM NaCI (Merck 8.22184), 10 mM K-EDTA (Fluka 03664), 2 mg/ml Dextran (Pharmacosmos 551005009007), 0.5% ovalbumin OA (Sigma-A 5573), 0.05% BGG (Sigma G-7516), 0.1% Tween20 (Merck 8.22184), 0.01% Proclin 300 (Sigma-Aldrich 48912-U), 0.01% Gentamycin (Biol. Indust. 03-035-1), 0.2 mg/ml HBR1 (Scan. Lab. 3KC533), pH 7.4

Mouse plasma for dilutions and calibrators:

Female mouse plasma (C57BL/6) from Bioreclamation, K2 EDTA, Cat.

MSEPLEDTA2-C57-F

Assay plates:

Perkin Elmer, PPN 6005359.

Data Analysis:

Graph Pad® software was used for all analyses; data is presented as mean. Statistical analysis was performed using 2-way ANOVA followed by Bonferroni's multiple comparisons test comparing all groups.

Results:

The results are shown in Fig. 17, where Fig. 17A shows cumulative food intake (FI) in g, and Fig. 17B plasma exposure levels in pM measured 24 h after dosing. Data is shown as mean values.

From Fig. 17A it is clear that the compound of Example 11 lowered food intake in a dose-dependent manner. And Fig. 17B shows that the higher the dose the higher the plasma exposure. This confirms that the compound is biologically active in vivo.

II. Head-to-head in vivo study of "active" and "inactive" compounds of Examples 11 and 12

To determine whether the TfR arm in the GLP-1 anti TfR-Fab conjugate of Example 11 contributed to the food intake lowering ability as determined above, a head- to-head comparison was performed with the GLP-1 control-Fab conjugate of Example 12, in which the same GLP-1R agonist is conjugated to an inactive Fab, to serve as a control compound.

Briefly, lean C57BI6J mice were maintained under a 12-h light dark cycle (lights off 11 a.m.), with food and water available ad lib, as in part I above. Also similarly to part I above, mice were habituated to the biodaq food intake monitoring system for 7 days prior to being assigned into 1 of 3 treatment groups that were evenly matched for body weight. On the day of the experiment, mice were dosed with either vehicle or 100 nmol/kg of the compound of Example 11 or the compound of Example 12 according to the table below:

Table 5: Groups

*PBS+0.05% polysorbate 80

Food intake was monitored hourly over the next 24 h, after which mice were sacrificed by exsanguination followed by cervical dislocation while under anaesthesia.

A satellite group of mice were used to generate exposure profiles of each compound. Blood samples were taken via the sublingual plexus into EDTA tubes containing a protease inhibitor cocktail and plasma was collected and analysed for exposure, as described in part I above. For each compound, blood draws were taken via sparse sampling, n = 3 mice per time point, after 1 h, 4 h, 8 h, 10 h, 16 h, 18 h, and 24 h after i.v. dosing. Plasma samples were analysed for exposure as described in part I above.

Data Analysis:

Graph Pad® software was used for all analyses; data is presented as mean. Statistical analysis was performed using 2-way ANOVA followed by Bonferroni's multiple comparisons test comparing all groups.

Results:

The results are shown in Fig. 18, where Fig. 18A shows cumulative food intake (FI) in g, and Fig. 18B shows plasma exposure levels (plasma concentration in pM) measured over 24 h. From Fig. 18A it is clear that both GLP-1 compounds reduced food intake relative to vehicle treated mice over 24 h of observation. However, food intake lowering was significantly greater in mice treated with the compound of Example 11 relative to mice treated with the compound of Example 12. Fig 18B shows that plasma exposure levels were similar or greater in mice treated with the control compound of Example 12 implying that differences in food intake lowering among analogues were unrelated to differences in PK. Rather, these data demonstrate that the active TfR Fab in the compound of Example 11 contributed to greater food intake lowering.

III. Sub-chronical in vivo study in diet-induced obese mice with "active" and "inactive" compounds of Examples 11 and 12. determination of weight loss

To determine whether reductions in acute food intake lowering translated to weight loss in a preclinical model of obesity, we next compared the compounds used in Parts I and II above in DIO mice treated sub-chronically for 19 days. Briefly, Male C57BI6J mice (Charles River, France) (n = 79) were maintained on a 60% high fat diet (D12492, Research Diets, New Brunswick, NJ, USA) beginning at approximately 4 weeks of age. Obese mice were shipped to the test facility at approximately 20 weeks of age, and were housed in a temperature and humidity controlled environment, n= 10 per cage, under a 12-12 h light-dark cycle (Lights off 11 :00 p.m.). After 4 days, mice in the high- fat diet fed group were switched from 60% to 45% high-fat diet (HFD D12451, Research Diets, New Brunswick, NJ, USA) to prevent crumbling and spillage and more accurately measure food intake. All mice were singly housed at this time and allowed to acclimatise to their new environment for two additional weeks. Food and water were available ad lib at all times.

Body composition (Echo MRI 700, Houston, TX, USA) was measured on day 5 prior to dosing, and on day 3 prior to dosing mice were assigned into 1 of 3 treatment groups evenly matched for fat and lean tissue mass.

Food intake and body weight were monitored daily, beginning the day before dosing. Compounds were administered daily each day beginning at 12:30 p.m. starting on day 1 and according to Table 6 below. Mice were initially dosed i. v. on days 1-4 and then s.c. on days 5-19. Blood samples for plasma exposure were collected 2 h after dosing on day 19 and analysed as described in Part I above.

Table 6: Groups

*PBS+0.05% polysorbate 80

Results:

The results are shown in Fig. 19, where Fig. 19A shows mean weight loss (%), and Fig. 19B shows mean plasma concentration (pM) measured 2 h after dosing on day 19.

From Fig. 19A it is clear that both compounds lowered body weight compared to vehicle treated mice. Fig. 19B shows differences in plasma exposure levels were similar among groups, however, DIO mice treated with the "active" compound of Example 11 lost more weight than DIO mice treated with the "inactive" compound of Example 12, showing that the "active" compound with the TfR Fab conferred an additional benefit to body weight lowering compared to the control compound.

While certain features of the invention have been illustrated and described herein, many modifications, substitutions, changes, and equivalents will now occur to those of ordinary skill in the art. It is, therefore, to be understood that the appended claims are intended to cover all such modifications and changes as fall within the true spirit of the invention.