WO2017070617A1 - Diol-modified insulin analogues containing a glucose-regulated conformational switch - Google Patents

Abstract

An insulin analogue contains a specific combination of modifications to the A- and B chains: (i) a phenylboronic acid derivative (optionally including a halogen atom substitution) linked by a connecting element either to the alpha-amino group of Gly^A1, to the epsilon-amino group of non-standard variant D-Lys^A1, to the epsilon-amino group of standard variant L-Lys^A4, to the alpha-amino group of D-Lys^A1 and its epsilon-amino group, to the alpha-amino group of Gly^A1 and the epsilon-amino group of L-Lys^A4, or to any combination of the above three sites; and (ii) one, two or three diol-containing adducts linked to side chains in the wild-type or variant B27-B30 segment, or within a one-residue C-terminal extension of the B chain (B31) or within a two-residue extension of the B chain (B31-B32). A method of treating a patient comprises administration of an effective amount of the insulin analogue or a physiologically acceptable salt thereof to a patient.

Description

DIOL-MODIED INSULIN ANALOGUES CONTAINING A GLUCOSE-REGULATED

CONFORMATIONAL SWITCH

CROSS REFERENCE TO RELATED APPLICATIONS

[0001] This application claims benefit of pending U.S. Provisional Application No.

62/244,531 filed on October 21, 2015.

STATEMENT REGARDING FEDERALLY SPONSORED

RESEARCH OR DEVELOPMENT

[0002] This invention was made with government support under grant number DK040949 awarded by the National Institutes of Health. The U.S. government has certain rights in the invention.

BACKGROUND OF THE INVENTION

[0003] This invention relates to polypeptide hormone analogues that contain a glucose- conformational switch and so exhibit either glucose-responsive binding to cognate cellular receptors or glucose-responsive rate of disassembly of the zinc insulin analogue hexamer or zinc-free insulin analogue hexamer, or a combination of these activities. The analogues of the present invention contain (i) a glucose-binding element attached to the alpha- amino group of residue A 1 or to the side-chain amino-group of a modified residue at position Al or A4 (and optionally also the alpha-amino group of residue Bl) and (ii) one, two, or three diol- containing modifications of side chains at positions chosen from B27, B28, B29, B30, B31 (as an optional B-chain extention) and B32 (as an optional B-chain extention), where the diol- containing modifications either are attached to the thiol functional group of cysteine or cysteines substituted at or near the C-terminal end of the B chain or are attached to the side- chain amino functional group of lysine, orthinine, diamino-butyric acid or diamino-propionic acid substituted at or near the C-terminal end of the B chain; or both types of attachments are utilized within the same insulin analogue.

[0004] Use of insulin is described in relation to the treatment of patients and non-human mammals with Type 1 or Type 2 diabetes mellitus by subcutaneous, intraperitoneal or intravenous injection. The insulin analogues of the present invention may also exhibit other enhanced pharmaceutical properties, such as increased thermodynamic stability, augmented resistance to thermal fibrillation above room temperature, decreased mitogenicity, and/or altered pharmacokinetic and pharmacodynamic properties. More particularly, this invention relates to insulin analogues that confer either rapid action (relative to wild-type insulin in its regular soluble formuation), intermediate action (comparable to NPH insulin formulations known in the art) or protracted action (comparable to basal insulins known in the art as exemplified by insulin detemir and insulin glargine) such that the affinity of the said analogues for the insulin receptor is higher when dissolved in a solution containing glucose at a concentration above the physiological range (> 140 mg/dl; hyperglycemia) than when dissolved in a solution containing glucose at a concentration below the physiological range (< 80 mg/dl; hypoglycemia).

[0005] The engineering of non-standard proteins, including therapeutic agents and vaccines, may have broad medical and societal benefits. Naturally occurring proteins— as encoded in the genomes of human beings, other mammals, vertebrate organisms, invertebrate organisms, or eukaryotic cells in general— may have evolved to function optimally within a cellular context but may be suboptimal for therapeutic applications. Analogues of such proteins may exhibit improved biophysical, biochemical, or biological properties. A benefit of protein analogues would be to achieve enhanced activity (such as metabolic regulation of metabolism leading to reduction in blood-glucose concentration under conditions of hyperglycemia) with decreased unfavorable effects (such as induction of hypoglycemia or its exacerbation).

[0006] An example of a therapeutic protein is provided by insulin. Wild-type human insulin and insulin molecules encoded in the genomes of other mammals bind to insulin receptors is multiple organs and diverse types of cells, irrespective of the receptor isoform generated by alternative modes of RNA splicing or by alternative patterns of post-translational glycosylation. An example of a medical benefit would be the non-standard design of a soluble insulin analogue whose intrinsic affinity for insulin receptors on the surface of target cells, and hence whose biological potency, would depend on the concentration of glucose in the blood stream. [0007] The insulin molecule contains two chains, an A chain, containing 21 residues, and a B chain containing 30 residues. The mature hormone is derived from a longer single-chain precursor, designated proinsulin, as outlined in Figure 1. Specific residues in the insulin molecule are indicated by the amino-acid type (typically in standard three-letter code; e.g., Lys and Ala indicate Lysine and Alanine) and in superscript the chain (A or B) and position in that chain. For example, Alanine at position 14 of the B chain of human insulin is indicated by Ala ; and likewise Lysine at position B28 of insulin lispro (the active component of Humalog; Eli Lilly and Co.) is indicated by Lys^B28. Although the hormone is stored in the pancreatic β-cell as a Zn²⁺-stabilized hexamer, it functions as a Zn²⁺-free monomer in the bloodstream. Administration of insulin has long been established as a treatment for diabetes mellitus.

[0008] A major goal of conventional insulin replacement therapy in patients with diabetes mellitus is tight control of the blood glucose concentration to prevent its excursion above or below the normal range characteristic of healthy human subjects. Excursions above the normal range are associated with increased long-term risk of microvascular disease, including retinapathy, blindness, and renal failure. Hypoglycemia in patients with diabetes mellitus is a frequent complication of insulin replacement therapy and when severe can lead to significant morbidity (including altered mental status, loss of consciousness, seizures, and death; for reviews, see Cryer, 2002 and Cryer, 2008). Indeed, fear of such complications poses a major barrier to efforts by patients (and physicians) to obtain rigorous control of blood glucose concentrations (i.e., excusions within or just above the normal range), and in patients with long-established Type 2 diabetes mellitus such efforts ("tight control") may lead to increased mortality. In addition to the above consequences of severe hypoglycemia (designated neuroglycopenic effects), mild hypoglycemia may activate counter-regulatory mechanisms, including over-activation of the sympathetic nervous system leading to turn to anxiety and tremulousness (symptoms designated adrenergic). Patients with diabetes mellitus may not exhibit such warning signs, however, a condition known as hypoglycemic unawareness. The absence of symptoms of mild hypoglycemia increases the risk of major hypoglycemia and its associated morbidity and mortality. Multiple and recurrent episodes of hypoglycemia are also associated with chronic cognitive decline, a proposed mechanism underlying the increased prevalence of dementia in patients with long-standing diabetes mellitus. There is therefore an urgent need for new diabetes treatment technologies that would reduce the risk of hypoglycemia while preventing upward excursions in blood-glucose concentration above the normal range (Cryer et al., 2003).

[0009] Diverse technologies have been developed in an effort to mitigate the threat of hypoglycemia in patients treated with insulin. Foundational to all such efforts is education of the patient (and also members of his or her family) regarding the symptoms of hypoglycemia and following the recognition of such symptoms, the urgency of the need to ingest a food or liquid rich in glucose, sucrose, or other rapidly digested form of carbohydrate; an example is provided by orange juice supplemented with sucrose (cain sugar). This baseline approach has been extended by the development of specific diabetes-oriented products, such as squeezable tubes containing an emulsion containing glucose in a form that can be rapidly absorbed through the mucous membranes of the mouth, throat, stomach, and small intestine.

Preparations of the counter-regulatory hormone glucagon, provided as a powder, have likewise been developed in a form amenable to rapid dissolution and subcutaneous injection as an emergency treatment of severe hypoglycemia. Insulin pumps have been linked to a continuous glucose monitor such that subcutaneous injection of insulin is halted and an alarm is sounded when hypoglycemic readings of the interstitial glucose concentration are encountered. Such a device-based approach has led to the experimental testing of closed- loop systems in which the pump and monitor are combined with a computer-based algorithm as an "artificial pancreas" (Russell, 2015).

[0010] For more than three decades, there has been interest in the development of glucose-responsive materials for co-administration with an insulin analogue or modified insulin molecule such that the rate of release of the hormone from the subcutaneous depot depends on the interstitial glucose concentration (Brownlee and Cerami, 1979 and 1983). Such systems in general contain a glucose-responsive polymer, gel or other encapsulation material; and may also require a derivative of insulin containing a modification that enables binding of the hormone to the above material. An increase in the ambient concentration of glucose in the interstitial fluid at the site of subcutaneous injection may displace the bound insulin or insulin derivative either by competitive displacement of the hormone or by physical-chemical changes in the properties of the polymer, gel or other encapsulation material. The goal of such systems is to provide an intrinsic autoregulation feature to the encapsulated or gel-coated subcutaneous depot such that the risk of hypoglycemia is mitigated through delayed release of insulin when the ambient concentration of glucose is within or below the normal range. To date, no such glucose-responsive systems are in clinical use.

[0011] A recent technology exploits the structure of a modified insulin molecule, optionally in conjuction with a carrier molecule such that the complex between the modified insulin molecule and the carrier is soluble and may enter into the bloodstream (Zion et al., 2012). This concept differs from glucose -responsive depots in which the polymer, gel or other encapsulation material remains in the subcutaneous depot as the free hormone enters into the bloodstream. An embodiment of this approach is known in the art wherein the A chain is modified at or near its N-terminus (utilizing the oc-amino group of residue Al or via the ε- amino group of a Lysine substituted at positions A2, A3, A4 or A5) to contain an "affinity ligand" (defined as a saccharide moiety), the B chain is modified at its or near N-terminus (utilizing the oc-amino group of residue B 1 or via the ε-amino group of a Lysine substituted at positions B2, B3, B4 or B5) to contain a "monovalent glucose-binding agent." In this description the large size of the exemplified or envisaged glucose-binding agents (monomeric lectin domains, DNA aptamers, or peptide aptomers) restricted their placement to the N- terminal segment of the B chain as defined above. In the absence of exogenous glucose or other exogenous saccharide, intramolecular interactions between the A 1 -linked affinity ligand and Bl -linked glucose-binding agent was envisaged to "close" the structure of the hormone and thereby impair its activity. Only modest glucose -responsive properties of this class of molecular designs were reported (Zion et al., 2012).

[0012] The suboptimal properties of insulin analogs modified at or near residue Al by an affinity ligand and simultaneously modified at or near residue B 1 by a large glucose-binding agent (i.e., of size similar or greater than that of an insulin A or B chain), as disclosed by Zion et al. (2012), are likely to be intrinsic to this class of molecular designs. Indeed, the rationale for such designs relied on the low activity of insulin analogs containing short chemical crosslinks between the oc-amino groups of residues Al and B l (Gliemann and Gammeltoft (1974) as cited by Zion et al., 2012) but overlooked the native or enhanced activity of an insulin analogue containing a peptide linker between these residues of length similar to or exceeding that of an insulin A or B chain (as described by Heath et al., 1992). Thus, the putative "closed" form of the above insulin analogues may not in fact be adequately constrained in conformation to provide significant impairment of receptor binding (relative to the modified insulin in the presence of exogenous glucose) and hence to provide useful or optimal glucose- dependent biological activity. The prescription of the above class of insulin analogues also overlooked the marked reduction in activity, irrespective of free glucose concentration, likely to arise on substitution of residues A2 or A3 (Ile^A2 or Val^A3) by other aliphatic or non- aliphatic amino acids (such as Lysine as taught by Zion et al., 2012); these analogues would be expected to have negligible biological activity and thus not be useful as is long known in the art (Nakagawa and Tager, 1992). Binding of an insulin analogue to the insulin receptor would also be impaired, but to a lesser degree, by substitution of Lysine at position Al.

[0013] In addition to glucose-dependent regulation of the binding of an insulin analogue to the insulin receptor, the insulin analogues of the present invention may also exhibit glucose-dependent modulation of the rate of disassembly of the zinc insulin analogue hexamer or zinc-free insulin analogue hexamer. In crystal structures of such hexamers the C-terminal residue of the B chain is typically near the N-terminal residue of the A chain of the same protomer of the hexamer. Indeed, in a subset of such structures a salt bridge may be observed between the C-terminal main-chain terminal carboxylate of residue B30 (negatively charged) and the N-terminal main-chain alpha-amino group of residue Al (positively charged). Such an electrostatic interaction may contribute to the damping of conformational fluctuations in the zinc insulin hexamer that contribute to disassembly of the complex into smaller subunits (trimers, dimers and monomers). Although we do not wish to be constrained by theory, we envision that a non-covalent interaction between a glucose-binding element attached to the alpha-amino group of residue Al and diol-containing adducts attached to side chains in the segment B27-B30 (optionally B-chain extensions B31 or B31-B32 and optionally including modifications of these residues) would likewise damp conformational fluctuations that facilitate disassembly of the zinc insulin analogue hexamer or zinc-free insulin analogue hexamer. This effect would lead to acceleration of the absorption of subunits (trimers, dimers and monomers) from a subcutaneous depot under conditions of hyperglycemia and similarly to a retardation of absorption under conditions of hypoglycemia. Thus, glucose-dependent regulation of the rate of disassembly of a subcutaneous depot of an injected insulin analogue formulation would enable effective treatment of hyperglycemia but reduce the concentration of insulin analogue reaching the blood stream under conditions of hypoglycemia, thereby protecting the patient from the symptoms and dangers of hypoglycemia relative to treatment with conventional and existing insulin products. The insulin analogues of the present invention may exhibit primarily glucose-dependent binding to the insulin receptor, glucose- dependent modulation of the rate of hexamer disassembly, or both of these activities in similar proportion.

SUMMARY OF THE INVENTION

[0014] Surprisingly, we have found that a fundamentally different class of molecule designs may optimally provide a glucose-dependent conformational switch between closed and open states of the insulin molecule without the above disadvantages. The analogues of the present invention thus contain one or more diol-containing modifications at or near the C- terminal end of the B chain rather than at or near the N-terminal end of the A chain. Further, the analogues of the present invention avoid bulky glucose-binding agents at or near residue Bl and instead employ small chemical entities (phenylboronic acid derivatives, including halogenic derivatives) at or near residue Al. The closed state of the present invention, tethered by a non-covalent interaction between one, two or three diol-containing

modifications at or near the C-terminal end of the B chain and a small chemical entity at or near the N-terminal end of the A chain, is thus different from and unrelated to that disclosed by Zion et al. (2012). The closed state of the present invention exploits a protective hinge in the insulin molecule that opens to engage the insulin receptor (Menting, J.G. et al. (2014)). Mini-proinsulins or single-chain insulin analogues in which a short covalent tether links the C-terminus of the B chain to the N-terminus of the A chain are known in the art to exhibit very low or undetectable activity (Figure 2). Mini-proinsulins are also known in the art to exhibit enhanced dimerization in an acidic co-solvent unfavorable to dimerization of native insulin (20% acetic acid), suggesting a coupling between propensity for dimerization (a key feature of insulin hexamers) and conformational fluctuations affecting the spatial relationship between the C-terminal end of the B chain and N-terminal end of the A chain.

[0015] The insulin analogues of the present invention thus include two functional classes conferred by the same or similar molecular embodiment: (i) those whose affinity for the insulin receptor and hence whose biological potency would be stronger under conditions of hyperglycemia than under conditions of hypoglycemia; and/or (ii) those whose rate of zinc insulin analogue hexamer disassembly or zinc-independent insulin analogue hexamer disassembly is delayed under conditions of hypoglycemia. These insulin analogues would thereby exhibit either (or both) higher biological potency or higher bio-availability from the subcutaneous depot (i.e., from the site of injection under the skin to the blood stream or lymphatic vessels) under conditions of hyperglycemia than under conditions of hypoglycemia, therteby enhancing the general safety, efficacy, simplicity and convenience of insulin replacement therapy. Such an insulin analogue formulation would be compatible with multiple devices (such as insulin vials, insulin pens, and insulin pumps) and could be integrated with modifications to the insulin molecule known in the art to confer rapid-, intermediate-, or prolonged insulin action. In addition, the present glucose-regulated conformational switch in the insulin molecule, engineered between the C-terminus of the B chain and N-terminus of the A chain, could be combined with other glucose-responsive technologies (such as closed-loop systems or glucose-responsive polymers) to optimize their integrated properties. We thus envisage that the products of the present invention will benefit patients with either Type 1 or Type 2 diabetes mellitus both in Western societies and in the developing world.

[0016] It is, therefore, an aspect of the present invention to provide insulin analogues that provide glucose-responsive binding to the insulin receptor and hence glucose-regulated bioactivity. The analogues of the present investion contain two essential elements. The first element is a phenylboronic acid derivative (including a spacer element) at the oc-amino group of Glycine at position Al (Gly^A1) or optionally at either the ε-amino group of D-Lysine as an amino-acid substitution well tolerated at position Al (D-Lys^A1) or the ε-amino group of L- Lysine as a substitution at position A4 (L-Lys^A4). The phenylboronic acid moiety (Figure 4) may be modified within its aromatic ring by substitution of a hydrogen atom of a halogen atom, such as fluorine, chlorine, bromine or iodine (Figure 5). Phenylboronic acid groups bind to diols within diverse molecular entities (Figure 6); the spacer element may contain a linear acyl chain of 3-16 carbon atoms and optionally one or more nitrogen atoms at or near its terminus (Figure 7). The second element is one or more diol-containing modifications at one or more of the positions B27, B28, B29, B30, or as attached to a peptide extension of the B- chain containing one residue (B31) or two residues (B31-B32). The diol-containing modification(s) are attached either (i) to the side-chain thiol function of a cysteine(s) substititued in the B27-B30 segment (or optionally B27-B31 or B27-B32 segments in the case of an an extended B chain) or (ii) to the side-chain amino functional group of a basic side chain substititued in the B27-B30 segment (or optionally B27-B31 or B27-B32 segments in the case of an an extended B chain) where the set of basic side chains includes Lysine, Ornithine, diamino-butyric acid and diamino-propionic acid. The overall structure of insulin analogues of the present invention is shown in schematic form in Figure 3.

[0017] It is an aspect of the present invention that in synthetic peptides employed in trypsin-mediated semisynthesis, a reaction that employs a prefolded £fes-octapeptide[B23- B30] fragment of wild-type insulin or insulin analogue, sites of cysteine substitution in the peptide will be modified by a diol-containing adduct attached to the sulfur atom prior to the semi-synthetic reaction scheme. This order of reaction is to prevent the free thiol group of the peptide (a) from interacting to form covalent disulfide-linked homodimers or (b) from interacting with cystines in the insulin fragment to yield degraded forms of insulin. It is withing the scope of the present invention that cysteine may be substituted by a non-standard amino acid whose side chain contains a free thiol group (as exemplified by homocysteine). It is likewise an aspect of the present invention that in synthetic peptides employed in trypsin- mediated semisynthesis, sites of substitution in the peptide by basic residues (Lys, Orn, diamino-butyric acid and/or diamino-propinionic acid) will be modified by a diol-containing adduct attached to the side-chain amino group prior to the semi-synthetic reaction scheme. The £fes-octapeptide[B23-B30] fragments of insulin or an insulin analogue employed in these preparations will contain a phenylboronic acid moiety linked to the alpha-amino group of residue A 1 or to the side-chain amino group of a basic residue substituted at position Al or A4; the substituted based residue at Al would either a D- or L amino acids whereas the substituted basic residue at A4 can be only an L amino acid.

[0018] The analogues of the present invention may optionally contain an additional phenyboronic acid group (or halogenic derivative thereof) attached (together with a spacer element) to residue B 1 as a mechanism intended to provide glucose-sensitive binding of the insulin analogue to surface lectins in the subcutaneous depot. In addition, the analogues of the present invention may optionally contain substitutions known in the art to confer rapid action (such as Asp^B28, a substitution found in insulin aspart (the active component of Novolog®); [Lys^B28, Pro^B29], pairwise substitutions found in insulin lispro (the active component of

B29 B3 B29

Humalog®); Glu or the combination [Lys , Glu ] as the latter is found in insulin glulisine (the active component of Apridra®), or modifications at position B24 associated with accelerated disassembly of the insulin hexamer {e.g., substitution of Phe^B24 by

Cyclohexanylalanine or by a derivative of Phenylalanine containing a single halogen substitution within the aromatic ring). Alternatively, the analogues of the present invention may optionally contain modifications known in the art to confer protracted action, such as modification of the ε-amino group of Lys^B29 by an acyl chain or acyl-glutamic acid adduct as respectively illustrated by insulin detemir (the active component of Levemir®) and insulin degludec (the active component of Tresiba®); or contain basic amino-acid substitutions or basic chain extensions designed to shift the isoelectric point (pi) to near neutrality as

B31 B32

exemplified by the Arg -Arg extension of insulin glargine (the active component of Lantus®). Analogues of the present invention designed to exhibit such a shifted pi may also contain a substitution of AsnA21, such as by Glycine, Alanine or Serine. Analogues of the present invention may optionally also contain non-beta-branched amino-acid substitutions of

A8

Thr associated with increased affinity for the insulin receptor and/or increased

thermodynamic stability as may be introduced to mitigate deleterious effects of the primary two above design elements (a phenylboronic acid derivative at or near the N-terminus of the A chain and one or more saccharide derivatives at or near the C-terminus of the B chain) on receptor-binding affinity and/or thermodynamic stability. Examples of such A8 substitutions known in the art are His^A8, Lys^A8, Arg^A8, and Glu^A8.

[0019] The insulin analogues of the present invention may exhibit an isoelectric point (pi) in the range 4.0-6.0 and thereby be amenable to pharmaceutical formulation in the pH range 6.8-7.8; alternatively, the analogues of the present invention may exhibit an isoelectric point in the range 6.8-7.8 and thereby be amenable to pharmaceutical formulation in the pH range 4.0-4.2. The latter conditions are known in the art to lead to isoelectric precipitation of such a pi-shifted insulin analogue in the subcutaneous depot as a mechanism of protracted action. An example of such a pi-shifted insulin analogue is provided by insulin glargine, in which a

B31 B32

basic two-residue extension of the B chain (Arg -Arg ) shifts the pi to near-neutrality and thus enables prolonged pharmacokinetic absorption from the subcutaneous depot. In general the pi of an insulin analogue may be modified through the addition of basic or acidic chain extensions, through the substitution of basic residues by neutral or acidic residues, and through the substitution of acidic residues by neutral or basic residues; in this context we define acidic residues as Aspartic Acid and Glutamic Acid, and we define basic residues as Arginine, Lysine, and under some circumstances, Histidine. We further define a "neutral" residue in relation to the net charge of the side chain at neutral pH.

[0020] It is an additional aspect of the present invention that absolute in vitro affinities of the insulin analogue for insulin receptor (isoforms IR-A and IR-B) are in the range 5-100% relative to wild-type human insulin and so unlikely to exhibit prolonged residence times in the hormone-receptor complex; such prolonged residence times are believed to be associated with enhanced risk of carcinogenesis in mammals or more rapid growth of cancer cell lines in culture. It is yet an additional aspect of the present invention that absolute in vitro affinities of the insulin analogue for the Type 1 insulin-like growth factor receptor (IGF-1R) are in the range 5-100% relative to wild-type human insulin and so unlikely either to exhibit prolonged residence times in the hormone/IGF- 1R complex or to mediate IGF-lR-related mitogenesis in excess of that mediated by wild-type human insulin.

[0021] The insulin analogues of the present invention may consist of two polypeptide chains that contain a novel combination of modifications in the A chain and B chain such that the analogue, in the absence of glucose or other exogenous saccharide, exhibits a non-covalent interaction between S- and/or N-linked diol-containing motieties at or near the C-terminal end of the B chain to a small saccharide-binding moiety (derived from phenylboronic acid) at or near the N-terminal end of the A chain. Examples are provided by S-linked derivatives of

B27 B30

Cys and/or Cys by the reagent 1-thioglycerol and are similarly provided by N-linked derivatives of the analogous diol-containing adducts attached to Lys, ornithine, diamino- butyric acid or diamino-propionic acid as substituted at one, two or three positions selected from B27, B28, B29, B30, B31 and B32.

[0022] Although we do not wish to be restricted by theory, we envisage that these two design elements form a non-covalent interaction in the absence of exogenous glucose such that the structure of the hormone is stabilized in a closed and less active conformation. This closed form may also exhibit a slower rate of disassembly of the zinc insulin analogue hexamer or zinc-free insulin analogue hexamer than the open form created on binding of an exogenous saccharide such as glucose. It is known in the art that closure of the distance between residues B30 and Al by chemical tethers or by short intervening peptides (or even by direct peptide bonds between Gly^A1 and either residues B28, B29 or B30) results in a marked loss of affinity of the tethered or single-chain insulin analogue for the insulin receptor. The recent co-crystal structure of insulin bound to a "micro-receptor" fragment of the ectodomain of the insulin receptor has rationalized these findings as the bound hormone exhibits partial detachment of the B24-B27 segment of the alpha-helical core of insulin (Menting et al. 2014). Although not restricted by theory, we envisage that binding of an exogenous glucose molecule (or other free saccharide) to the phenylboronic acid moiety at or near the N-terminus of the A chain would prevent the latter' s interaction with diol-containing adducts at or near the C- terminus of the B chain, thus facilitating partial detachment of the B24-B27 segment to bind to the insulin receptor and/or facilitating conformational fluctuations leading to disassembly of insulin hexamers in a subcutaneous depot.

BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWINGS

[0023] FIGURE 1A is a schematic representation of the sequence of human proinsulin including the A- and B-chains and the connecting region shown with flanking dibasic cleavage sites (filled circles) and C-peptide (open circles).

[0024] FIGURE IB is a structural model of proinsulin, consisting of an insulin-like moiety and a disordered connecting peptide (dashed line).

[0025] FIGURE 1C is a schematic representation of the sequence of human insulin indicating the position of residues B27 and B30 in the B-chain.

[0026] FIGURE 2 is cylinder model of a mini-proinsulin (single-chain insulin) in which a peptide bond links Lys^B29 and Gly^A1; Thr^B3° has been deleted. This analogue is known in the art to have no detectable activity.

[0027] FIGURE 3 is a schematic of the insulin molecule (top) and modifications pertaining to the present invention. The A chain is represented by the shorter horizontal cylinder (powder blue) and the B chain by the longer horizontal cylinder (gray). The canonical disulfide bridges of wild-type insulin are indicated by black lines (see box at bottom right). The A chain is modified by a monomeric glucose-binding moiety and spacer at or near its N- terminus (red cup and black wavy line, respectively; see box at bottom left) and optionally at the alpha-amino group of the B chain (red asterisks in parentheses). The B chain is modified by one or more diol-containing adducts at or near its C-terminus (green triangle), which may be linked to a side-chain sulfur atom of cysteine or homocysteine (S-linked diol) or linked to a side-chain nitrogen atom of a basic amino acid selected from Lys, ornithine, diamino-butyric acid and diamino-propionic acid (N-linked diol). Each adduct may contain one diol functional group , two diol functional groups or multiple diol-functional group s.

[0028] FIGURE 4 provides the molecular structure of phenylboronic acid.

[0029] FIGURE 5A provides a representation of the molecular structure of a halogen- modifed phenylboronic acid, in this case in which a hydrogen atom in the aromatic ring has been replaced by a fluorine atom at a position ortho to the boronic acid moiety. Halogenic modifications of the phenyl ring are known in the art to modulate the pK_a of the boronic acid group.

[0030] FIGURE 5B is a representation of various linking schema for a halogenated phenylboronic acid (PBA) according to the present invention

[0031] FIGURE 6A is a schematic reaction scheme showing how the phenylboronic acid motiety binds to diol functional group s within either saccharides or non-sugar-related chemical elements.

[0032] FIGURE 6B is a schematic representation of diol-containing amino acids and small diols and amino acid scaffolds according to the present invention.

[0033] FIGURE 7 is a representation of the spacer element which may contain a linear acyl chain of 3-16 carbon atoms and optionally one or more nitrogen atoms at or near its terminus.

[0034] FIGURE 8A is a grph showing the decrease in absorption at 574 nm in a cobalt- EDTA Hexamer Disassociation Assay for lispro insulin (KP Insulin). [0035] FIGURE 8B is a grph showing the decrease in absorption at 574 nm in a cobalt- EDTA Hexamer Disassociation Assay for an insulin analogue according to the present invention (T-2020).

DETAILED DESCRIPTION OF THE INVENTION

[0036] The present invention is directed toward an insulin analogue that provides enhanced in vivo glycemic control through (i) glucose-dependent binding to the insulin receptor in the blood stream and in biological tissues and/or (ii) glucose-dependent disassembly of the insulin hexamer (either zinc bound or zinc free) in the subcutaneous depot.

[0037] It is a feature of the present invention that when the blood glucose

concentration is within or below the normal range, a subset of the claimed insulin analogues exhibit low affinities for the insulin receptor relative to wild-type insulin. Although not wishing to be constrained by theory, we envisage that this reduction in affinity is a consequence of a non-covalent interaction between a diol-containing modification at or near the C-terminal end of the B chain and a phenylboronic acid derivative attached to the A chain at or near its N-terminal end. Such a non-covalent interaction would "close" the conformation of the B chain C-terminal beta-strand (residues B24-B28) and thus hinder binding to the insulin receptor, which requires partial detachment of this beta-strand from the alpha-helical core of the insulin molecule. It is a feature of the present invention that the glycemic potency of this subset of the claimed insulin analogues is restored to levels similar to that of wild-type insulin when the blood-glucose concentration is above the normal range due to competitive displacement of the A-chain-linked phenylboronic acid derivative from the B-chain-linked diol-containing motiety or moieties by the excess of exogenous glucose in solution. We envisage that receptor-binding affinities in the presence of glucose at a concentration > 200 mg/ml that are in the range 5-100% would be sufficient to confer such native or near-native glycemic potency in an animal with diabetes mellitus. It is an additional feature of the present invention that these modifications are also likely to reduce the tendency of insulin to undergo fibrillation at or above room temperature and to attenuate the mitogenicity of insulin, a distinct signaling pathway that is undesirable from the perspective of cancer risk and cancer growth. [0038] It is another feature of the present invention that when the blood glucose concentration is within or below the normal range, a subset of the claimed insulin analogues exhibit retarded rates of disassembly of the insulin analogue hexamer (either zinc -bound or zinc-free) relative to wild-type insulin. Although not wishing to be constrained by theory, we envisage that this reduction in rate of diassembly is a consequence of a non-covalent interaction between a diol-containing modification (or set of one, two or three modifications) at or near the C-terminal end of the B chain and a phenylboronic acid derivative attached to the A chain at or near its N-terminal end. Such a non-covalent interaction would "close" the conformation of the B chain C-terminal beta-strand (residues B24-B28) and thus hinder conformational fluctuations that facilitate disassembly of the hexamer, which may be promoted by partial detachment of this beta-strand from the alpha-helical core of the insulin molecule. It is a feature of the present invention that the bio- availability of this subset of the claimed insulin analogues is restored to levels similar to that of wild-type insulin when the blood-glucose concentration is above the normal range due to competitive displacement of the A-chain- linked phenylboronic acid derivative from the B -chain- linked diol-containing motiety or moieties by the excess of exogenous glucose in solution. We envisage that rates of hexamer diassembly in the presence of glucose at a concentration > 200 mg/ml that are at least 50% as fast as that of wild-type insulin would be sufficient to confer such native or near- native glycemic potency in an animal with diabetes mellitus.

[0039] It is also envisioned that insulin analogues may be made with A- and B chain sequences derived from animal insulins, such as porcine, bovine, equine, and canine insulins, by way of non-limiting examples, so long as the A-chain is modified by a phenylboronic acid derivative at or near its N-terminus, and one or more amino-acid side chains at or near the C- terminus of the B chain are modified by S-linked or N-linked diol-containing moieties. S- linked bonds are provided by the thiol functional group of cysteine (or homocysteine) substitutions whereas N-linked bonds are provided by the side-chain amino group of Lys, ornithine, diamino-butyric acid pr diamino-propionic acid, either at the site of a native basic residue in wild-type insulin (LysB29) or as substituted at one of more sites including B27, B28 or B30, within an extended B27-B31 segment or within an extended B27-B32 segment. The latter variant B chains derived from human insulin or animal insulins may thus optionally contain a C-terminal dipeptide extension (with respective residue positions designated B31 and B32) wherein at least one of these C-terminal extended residues is a modified amino acid containing an S-linked or N-linked diol adduct. In addition or in the alternative, the insulin analogue of the present invention may contain a deletion of residues B1-B3 or may be

B28 B29 combined with a variant B chain lacking Proline at position B28 (e.g., [Lys , Pro ] as in

B28 B28

Humalog; or Asp or Glu in combination with Lysine or Proline at position B29) or containing Glutamic Acid at position B29. At position A13 Leucine may optionally be substituted by Tryptophan, and at position A14 Tyrosine may optionally be substituted by Glutamic Acid.

[0040] It is further envisioned that the insulin analogues of the present invention may be derived from Lys-directed proteolysis of a precursor polypeptide in yeast biosynthesis in Pichia pastoris, Saccharomyces cerevisciae, or other yeast expression species or strains. Such strains may be engineered to insert halogen-modified Phenylalanine at position B24 by means of an engineered tRNA synthetase and orthogonal nonsense suppression. The B -domain of the insulin analogues of the present invention may optionally contain non-standard substitutions, such as D-amino-acids at positions B20 and/or B23 (intended to augment thermodynamic stability, receptor-binding affinity, and resistance to fibrillation). The halogenic modification at position B24 may be at the 2-ring position of Phe^B24 (i.e., ortho-F-

B24 B24 B24

Phe , ori/20-Cl-Phe , or ori/20-Br-Phe . Optionally, the analogues may contain iodo- substitutions within the aromatic ring of Tyr^{B 16} and/or Tyr^B26 (3 -mono-iodo-Tyr or [3, 5]-di- iodo-Tyr); intended to augment thermodynamic stability and receptor-binding activity). It is also envisioned that Thr^B27, Thr^B3°, or one or more Serine residues in the C-domain (if not substituted by Cys or a basic residue above as an attachment point for a diol-containing adduct) may be modified, singly or in combination, by a monosaccaride adduct; examples are provided by O-linked N-acetyl- -D-galactopyranoside (designated GalNAc-C^-Ser or GalNAc-C^-Thr), O-linked oc-D-mannopyranoside (mannose- Οβ-Ser or mannose- Οβ-Thr), and/or oc-D-glucopyranoside (glucose- Οβ-Ser or glucose- Οβ-Thr). Because these saccharide moieties contain diol functional group s, they may contribute further to the mechanism of glucose-regulated receptor binding or glucose-regulated hexamer disassembly rates. Points of attachment of monosaccharides, disaccharides or oligosaccharides would be provided by the side chains of Asn, Gin, Ser, and/or Thr, either at naturally occurring residues (ThrB27 and ThrB30) or as introduced as substitutions within the synthetic peptide employed in the semisynthetic preparation of the insulin analogues of the present invention.

[0041] Furthermore, in view of the similarity between human and animal insulins, and use in the past of animal insulins in human patients with diabetes mellitus, it is also envisioned that other minor modifications in the sequence of insulin may be introduced, especially those substitutions considered "conservative." For example, additional substitutions of amino acids may be made within groups of amino acids with similar side chains, without departing from the present invention. These include the neutral hydrophobic amino acids: Alanine (Ala or A), Valine (Val or V), Leucine (Leu or L), Isoleucine (He or I), Proline (Pro or P), Tryptophan (Trp or W), Phenylalanine (Phe or F) and Methionine (Met or M). Likewise, the neutral polar amino acids may be substituted for each other within their group of Glycine (Gly or G), Serine(Ser or S), Threonine (Thr or T), Tyrosine (Tyr or Y), Cysteine (Cys or C), Glutamine (Glu or Q), and Asparagine (Asn or N). Acidic amino acids are Aspartic acid (Asp or D) and Glutamic acid (Glu or E). Introduction of basic amino-acid substitutions (including Lysine (Lys or K), Arginine (Arg or R) and Histidine (His or H)) are not preferred in order to maintain the enhanced net negative charge of this class of analogues. Unless noted otherwise or wherever obvious from the context, the amino acids noted herein should be considered to be L-amino acids. Standard amino acids may also be substituted by non-standard amino acids belonging to the same chemical class.

[0001] The amino-acid sequence of human proinsulin is provided, for comparative purposes, as SEQ ID NO: 1.

SEQ ID NO: 1 (human proinsulin)

Phe-Val-Asn-Gln-His-Leu-Cys-Gly-Ser-His-Leu-Val-Glu-Ala-Leu-Tyr-Leu-Val-Cys- Gly-Glu-Arg-Gly-Phe-Phe-Tyr-Thr-Pro-Lys-Thr-Arg-Arg-Glu-Ala-Glu-Asp-Leu-Gln-Val- Gly-Gln-Val-Glu-Leu-Gly-Gly-Gly-Pro-Gly-Ala-Gly-Ser-Leu-Gln-Pro-Leu-Ala-Leu-Glu- Gly-Ser-Leu-Gln-Lys-Arg-Gly-Ile-Val-Glu-Gln-Cys-Cys-Thr-Ser-Ile-Cys-Ser-Leu-Tyr-Gln- Leu-Glu-Asn-Tyr-Cys-Asn

[0002] The amino-acid sequence of the A chain of human insulin is provided as SEQ ID NO: 2. SEQ ID NO: 2 (human A chain)

Gly-Ile-Val-Glu-Gln-Cys-Cys-Thr-Ser-Ile-Cys-Ser-Leu-Tyr-Gln-Leu-Glu-Asn-Tyr- Cys-Asn

[0003] The amino-acid sequence of the B chain of human insulin is provided as SEQ ID NO: 3.

SEP ID NO: 3 (human B chain)

Phe-Val-Asn-Gln-His-Leu-Cys-Gly-Ser-His-Leu-Val-Glu-Ala-Leu-Tyr-Leu-Val-Cys- Gly-Glu-Arg-Gly-Phe-Phe-Tyr-Thr-Pro-Lys-Thr

The amino-acid sequence of the "KP" B chain of prandial insulin analogue KP-insulin contains substitutions Pro — >Lys and Lys — >Pro as provided in SEQ ID NO: 4.

SEQ ID NO: 4

Phe-Val- Glu-Gln-His-Leu-Cys-Gly-Ser-His-Leu-Val-Glu-Ala-Leu-Tyr-Leu-Val-Cys- Gly-Glu-Arg-Gly-Phe-Phe-Try-Thr-Lys-Pro-Thr

[0004] The 32-residue amino-acid sequence of an extended "KP" B chain of prandial insulin analogue KP-insulin is provided in SEQ ID NO: 5.

SEQ ID NO: 5

Phe-Val- Glu-Gln-His-Leu-Cys-Gly-Ser-His-Leu-Val-Glu-Ala-Leu-Tyr-Leu-Val-Cys- Gly-Glu-Arg-Gly-Phe-Phe-Try-Thr-Lys-Pro-Thr-Glu-Glu

[0005] The 30-residue amino-acid sequence of a variant B chain modified to contain Ornithine at position B29 is provided in SEQ ID NO: 6.

SEQ ID NO: 6

Phe-Val- Glu-Gln-His-Leu-Cys-Gly-Ser-His-Leu-Val-Glu-Ala-Leu-Tyr-Leu-Val-Cys- Gly-Glu-Arg-Gly-Phe-Phe-Try-Thr-Pro-Orn-Thr

[0006] The amino-acid sequence of a variant B chain modified by diol-containing adducts at or near the C-terminus of the B chain are given in SEQ ID NO: 7-28. The insulin analogues of the present invention may optionally contain halogenic derivatives of the aromatic side chains, O-linked saccharides, or N-linked saccharides; the latter two classes of saccharides would provide additional diol-containing elements at or near the C terminus of the B chain and so contribute to the mechanisms of glucose regulation as described above. PheB24 may also optionally be substituted by Cyclohexanylalanine (Cha).

SEQ ID NO: 7

Phe-Val- Glu-Gln-His-Leu-Cys-Gly-Ser-His-Leu-Val-Glu-Ala-Leu-Tyr-Leu-Val-Cys- Gly-Glu-Arg-Gly-Phe-Phe-Tyr-Xaai-Lys-Pro-Xaa₂

Where Xaai is Cys, Lys, Orn, diamino-butyric acid or diamino-propionic acid as an attachment point for a diol-containing moiety; and where Xaa₂ may be Thr or Ala.

[0007] SEQ ID NO: 8

Phe-Val- Glu-Gln-His-Leu-Cys-Gly-Ser-His-Leu-Val-Glu-Ala-Leu-Tyr-Leu-Val-Cys- Gly-Glu-Arg-Gly-Phe-Phe-Tyr-Thr-Xaai-Pro-Xaa₂

Where Xaa_! is Cys, Lys, Orn, diamino-butyric acid or diamino-propionic acid as an attachment point for a diol-containing moiety; and where Xaa₂ may be Thr or Ala.

[0008] SEQ ID NO: 9

Phe-Val- Glu-Gln-His-Leu-Cys-Gly-Ser-His-Leu-Val-Glu-Ala-Leu-Tyr-Leu-Val-Cys- Gly-Glu-Arg-Gly-Phe-Phe-Tyr-Thr-Lys-Xaai-Xaa₂

Where Xaai is Cys, Lys, Orn, diamino-butyric acid or diamino-propionic acid as an attachment point for a diol-containing moiety; and and where Xaa₂ may be Thr or Ala.

[0009] SEP ID NO: 10

Phe-Val- Glu-Gln-His-Leu-Cys-Gly-Ser-His-Leu-Val-Glu-Ala-Leu-Tyr-Leu-Val-Cys- Gly-Glu-Arg-Gly-Phe-Phe-Tyr-Thr-Lys-Pro-Xaai

Where Xaai is Cys, Lys, Orn, diamino-butyric acid or diamino-propionic acid as an attachment point for a diol-containing moiety.

[0010] SEP ID NO: 11

Phe-Val- Glu-Gln-His-Leu-Cys-Gly-Ser-His-Leu-Val-Glu-Ala-Leu-Tyr-Leu-Val-Cys- Gly-Glu-Arg-Gly-Phe-Phe-Tyr-Xaa₁-Pro-Grn-Xaa₂ Where Xaai is Cys, Lys, Orn, diamino-butyric acid or diamino-propionic acid as an attachment point for a diol-containing moiety; and where Xaa₂ may be Thr or Ala.

[0011] SEP ID NO: 12

Phe-Val- Glu-Gln-His-Leu-Cys-Gly-Ser-His-Leu-Val-Glu-Ala-Leu-Tyr-Leu-Val-Cys- Gly-Glu-Arg-Gly-Phe-Phe-Tyr-Thr-Xaai-Orn-Xaa₂

Where Xaai is Cys, Lys, Orn, diamino-butyric acid or diamino-propionic acid as an attachment point for a diol-containing moiety; and and where Xaa₂ may be Thr or Ala.

[0012] SEQ ID NO: 13

Phe-Val- Glu-Gln-His-Leu-Cys-Gly-Ser-His-Leu-Val-Glu-Ala-Leu-Tyr-Leu-Val-Cys- Gly-Glu-Arg-Gly-Phe-Phe-Tyr-Thr-Orn-Xaai-Xaa₂

Where Xaai is Cys, Lys, Orn, diamino-butyric acid or diamino-propionic acid as an attachment point for a diol-containing moiety; and where Xaa₂ may be Thr or Ala.

[0013] SEQ ID NO: 14

Phe-Val- Glu-Gln-His-Leu-Cys-Gly-Ser-His-Leu-Val-Glu-Ala-Leu-Tyr-Leu-Val-Cys- Gly-Glu-Arg-Gly-Phe-Phe-Tyr-Thr-Orn-Pro-Xaai

Where Xaai is Cys, Lys, Orn, diamino-butyric acid or diamino-propionic acid as an attachment point for a diol-containing moiety.

[0014] SEP ID NO: 15

Phe-Val- Glu-Gln-His-Leu-Cys-Gly-Ser-His-Leu-Val-Glu-Ala-Leu-Tyr-Leu-Val-Cys- Gly-Glu-Arg-Gly-Phe-Phe-Tyr-Xaai-Pro-Xaa₂-Xaa₃

Where Xaai is Cys, Lys, Orn, diamino-butyric acid or diamino-propionic acid as an attachment point for a diol-containing moiety; where Xaa₂ is Ala, Lys, Gin, Ser, Thr or Glu; and where Xaa₃ is Cys, Lys, Orn, diamino-butyric acid or diamino-propionic acid as an attachment point for a diol-containing moiety.

[0015] SEP ID NO: 16

Phe-Val- Glu-Gln-His-Leu-Cys-Gly-Ser-His-Leu-Val-Glu-Ala-Leu-Tyr-Leu-Val-Cys- Gly-Glu-Arg-Gly-Phe-Phe-Tyr-Xaai-Xaa₂-Xaa₃-Xaa₄ Where Xa&i is Cys, Lys, Orn, diamino-butyric acid or diamino-propionic acid as an attachment point for a diol-containing moiety; where Xaa₃ is Cys, Lys, Orn, diamino-butyric acid or diamino-propionic acid as an attachment point for a diol-containing moiety; where Xaa₃ is Pro, Asn, Thr, Gin, Ser or Glu; and where Xaa₄ may be Thr or Ala.

[0016] SEQ ID NO: 17

Phe-Val- Glu-Gln-His-Leu-Cys-Gly-Ser-His-Leu-Val-Glu-Ala-Leu-Tyr-Leu-Val-Cys- Gly-Glu-Arg-Gly-Phe-Phe-Tyr-Xaai-Xaa₂-Xaa₃-Xaa₄

Where Xaai is Cys, Lys, Orn, diamino-butyric acid or diamino-propionic acid as an attachment point for a diol-containing moiety; where Xaa₂ is Pro, Ala, Ser, Thr, Asn, Glu or Gin; and where Xaa₃ is Cys, Lys, Orn, diamino-butyric acid or diamino-propionic acid as an attachment point for a diol-containing moiety; and where Xaa₄ may be Thr or Ala.

[0017] SEQ ID NO: 18

Phe-Val- Glu-Gln-His-Leu-Cys-Gly-Ser-His-Leu-Val-Glu-Ala-Leu-Tyr-Leu-Val-Cys- Gly-Glu-Arg-Gly-Phe-Phe-Tyr-Thr-Xaai-Xaa₂-Xaa₃

Where Xaai is Cys, Lys, Orn, diamino-butyric acid or diamino-propionic acid as an attachment point for a diol-containing moiety; where Xaa₂ is Pro, Ala, Asn, Ser, Thr, Gin or Glu; and where Xaa₃ is Cys, Lys, Orn, diamino-butyric acid or diamino-propionic acid as an attachment point for a diol-containing moiety.

[0018] SEQ ID NO: 19

Phe-Val- Glu-Gln-His-Leu-Cys-Gly-Ser-His-Leu-Val-Glu-Ala-Leu-Tyr-Leu-Val-Cys- Gly-Glu-Arg-Gly-Phe-Phe-Tyr-Xaai-Pro-Orn-Xaa₂

Where Xaai is Cys, Lys, Orn, diamino-butyric acid or diamino-propionic acid as an attachment point for a diol-containing moiety; and where Xaa₂ may be Thr or Ala.

[0019] SEQ ID NO: 20

Phe-Val- Glu-Gln-His-Leu-Cys-Gly-Ser-His-Leu-Val-Glu-Ala-Leu-Tyr-Leu-Val-Cys- Gly-Glu-Arg-Gly-Phe-Phe-Tyr-Thr-Xaai-Orn-Xaa₂

Where Xaa_! is Cys, Lys, Orn, diamino-butyric acid or diamino-propionic acid as an attachment point for a diol-containing moiety; and where Xaa₂ may be Thr or Ala. [0020] SEQ ID NO: 21

Phe-Val- Glu-Gln-His-Leu-Cys-Gly-Ser-His-Leu-Val-Glu-Ala-Leu-Tyr-Leu-Val-Cys- Gly-Glu-Arg-Gly-Phe-Phe-Tyr-Thr-Orn-Xaai-Xaa₂

Where Xaai is Cys, Lys, Orn, diamino-butyric acid or diamino-propionic acid as an attachment point for a diol-containing moiety; and where Xaa₂ may be Thr or Ala.

[0021] SEP ID NO: 22

Phe-Val- Glu-Gln-His-Leu-Cys-Gly-Ser-His-Leu-Val-Glu-Ala-Leu-Tyr-Leu-Val-Cys- Gly-Glu-Arg-Gly-Phe-Phe-Tyr-Thr-Orn-Pro-Xaai

Where Xaai is Cys, Lys, Orn, diamino-butyric acid or diamino-propionic acid as an attachment point for a diol-containing moiety.

The amino-acid sequence of a variant B chains both extended by one- or two residues and modified by diol-containing adducts are given in SEQ ID No 23-26.

[0022] SEQ ID NO: 23

Phe-Val- Glu-Gln-His-Leu-Cys-Gly-Ser-His-Leu-Val-Glu-Ala-Leu-Tyr-Leu-Val-Cys- Gly-Glu-Arg-Gly-Phe-Phe-Tyr-Thr-Xaai-Pro-Xaa₂-Xaa₃-Xaa₄

Where Xaai is Asp, Ala, Glu, Gin, Lys, Ser, or Thr; where Xaa₂ may be Thr or Ala; where Xaa₃ is Cys, Lys, Orn, diamino-butyric acid or diamino-propionic acid as an attachment point for a diol-containing moiety; and where Xaa₄ is Cys, Lys, Orn, diamino- butyric acid or diamino-propionic acid as an attachment point for a diol-containing moiety. As a further option, either Xaa₃ or Xaa₄ (but not both) may be either Gly or Glu. As yet another option, Xaa₄ may be absent.

[0023] SEQ ID NO: 24

Phe-Val- Glu-Gln-His-Leu-Cys-Gly-Ser-His-Leu-Val-Glu-Ala-Leu-Tyr-Leu-Val-Cys- Gly-Glu-Arg-Gly-Phe-Phe-Tyr-Xaa Pro-Glu-Xaa₂-Xaa₃-Xaa₄

Where Xaai is Cys, Lys, Orn, diamino-butyric acid or diamino-propionic acid as an attachment point for a diol-containing moiety; where Xaa₂ may be Thr or Ala; where Xaa₃ is Cys, Lys, Orn, diamino-butyric acid or diamino-propionic acid as an attachment point for a diol-containing moiety; and where Xaa₄ is Cys, Lys, Orn, diamino-butyric acid or diamino- propionic acid as an attachment point for a diol-containing moiety. As a further option, either Xaa₃ or Xaa₄ (but not both) may be either Gly or Glu. As yet another option, Xaa₄ may be absent.

[0024] SEQ ID NO: 25

Phe-Val- Glu-Gln-His-Leu-Cys-Gly-Ser-His-Leu-Val-Glu-Ala-Leu-Tyr-Leu-Val-Cys- Gly-Glu-Arg-Gly-Phe-Phe-Tyr-Thr-Xaai-Pro-Xaa₂-Xaa₃-Xaa₄

Where Xaa_! may be Lys, Orn, Glu, Asp, Ala, Gin, Asn, Ser or Thr; where Xaa₂ is Cys, Lys, Orn, diamino-butyric acid or diamino-propionic acid as an attachment point for a diol- containing moiety; where Xaa₃ is Cys, Lys, Orn, diamino-butyric acid or diamino-propionic acid as an attachment point for a diol-containing moiety; and where Xaa₄ is Cys, Lys, Orn, diamino-butyric acid or diamino-propionic acid as an attachment point for a diol-containing moiety. As a further option, either Xaa₃ or Xaa₄ (but not both) may be either Gly or Glu. As yet another option, Xaa₄ may be absent.

[0025] SEQ ID NO: 26

Phe-Val- Glu-Gln-His-Leu-Cys-Gly-Ser-Xaa₄-Leu-Val-Glu-Ala-Leu-Tyr-Leu-Val- Cys-Gly-Glu-Arg-Gly-Phe-Phe-Tyr-Thr-Pro-Xaai-Xaa₂-Xaa₃-Xaa₄

Where Xaai is Cys, Lys, Orn, diamino-butyric acid or diamino-propionic acid as an attachment point for a diol-containing moiety; where Xaa₂ may be Thr or Ala; where Xaa₃ is Cys, Lys, Orn, diamino-butyric acid or diamino-propionic acid as an attachment point for a diol-containing moiety; and where Xaa₄ is Cys, Lys, Orn, diamino-butyric acid or diamino- propionic acid as an attachment point for a diol-containing moiety. As a further option, either Xaa₃ or Xaa₄ (but not both) may be either Gly or Glu. As yet another option, Xaa₄ may be absent.

[0026] The following synthetic peptide sequences are suitable for trypsin-mediated sem- synthesis with wild-type <i<?s-octapeptide[B23-B30]-insulin or a variant thereof containing one or more substitutions in the A- or truncated B chains. Such synthetic peptides may optionally contain halogenic derivatives of the aromatic side chains, O-linked saccharides, or N-linked saccharides; the latter two classes of saccharides would provide additional diol-containing elements at or near the C terminus of the B chain and so contribute to the mechanisms of glucose regulation as described above. PheB24 may also optionally be substituted by

Cyclohexanylalanine. The des-octapeptide[B23-B30] insulin fragment employed in trypsin- mediated semisynthesis would be premodified by a glucose-sensing element at or near the N terminus of the A chain and optionally also at the N-terminus of the B chain (which may be residue Bl or, in the case of N-terminally truncated chains, at B2-B4 as respectively corresponding to successive N-terminal deletions of the B chain).

[0027] SEQ ID NO: 27

Gly-Phe-Phe-Tyr-Xaai-Pro-Xaa₂-Xaa₃

Where Xaai is Cys, Lys, Orn, diamino-butyric acid or diamino-propionic acid as an attachment point for a diol-containing moiety; where Xaa₂ may be Ala, Ser or Glu; and where Xaa₃ is Cys, Lys, Orn, diamino-butyric acid or diamino-propionic acid as an attachment point for a diol-containing moiety. Optionally either Xaai or Xaa₃ (but not both) may a non-basic side/non-thiol-related chain. As a further option, Xaa₃ may be amidated or absent.

[0028] SEQ ID NO: 28

Gly-Phe-Phe-Tyr-Thr-Xaai-Xaa₂-Xaa₃

Where Xaai is Cys, Lys, Orn, diamino-butyric acid or diamino-propionic acid as an attachment point for a diol-containing moiety; where Xaa₂ may be Ala, Pro, Ser or Glu; and where Xaa₃ is Cys, Lys, Orn, diamino-butyric acid or diamino-propionic acid as an attachment point for a diol-containing moiety. Optionally either Xaai or Xaa₃ (but not both) may a non-basic side/non-thiol-related chain. As a further option, Xaa₃ may be amidated or absent.

[0029] SEQ ID NO: 29

Gly-Phe-Phe-Tyr-Thr-Xaai-Xaa₂-Xaa₃

Where Xaai is Cys, Lys, Orn, diamino-butyric acid or diamino-propionic acid as an attachment point for a diol-containing moiety; where Xaa₂ may be Cys, Lys, Orn, diamino- butyric acid or diamino-propionic acid as an attachment point for a diol-containing moiety; and where Xaa₃ is Ala or Thr. Optionally either Xaai or Xaa₃ (but not both) may a non-basic side/non-thiol-related chain. As a further option, Xaa₃ may be amidated or absent. [0030] SEP ID NO: 30

Gly-Phe-Phe-Tyr-Xaai-Pro-Xaa₂-Xaa₃

Where Xaai is Cys, Lys, Orn, diamino-butyric acid or diamino-propionic acid as an attachment point for a diol-containing moiety; where Xaa₂ may be Cys, Lys, Orn, diamino- butyric acid or diamino-propionic acid as an attachment point for a diol-containing moiety; and where Xaa₃ is Ala or Thr. Optionally either Xaai or Xaa₃ (but not both) may a non-basic side/non-thiol-related chain. As a further option, Xaa₃ may be amidated or absent.

[0031] SEP ID NO: 31

Gly-Phe-Phe-Tyr-Thr-Pro-Xaai-Xaa₂-Xaa₃-Xaa₄

Where Xaai is Cys, Lys, Orn, diamino-butyric acid or diamino-propionic acid as an attachment point for a diol-containing moiety; where Xaa₂ may be Thr or Ala; where Xaa₃ is Cys, Lys, Orn, diamino-butyric acid or diamino-propionic acid as an attachment point for a diol-containing moiety; and where Xaa₄ is Cys, Lys, Orn, diamino-butyric acid or diamino- propionic acid as an attachment point for a diol-containing moiety. As a further option, either Xaa₃ or Xaa₄ (but not both) may be either Gly or Glu. As yet another option, Xaa₄ may be absent.

[0032] SEP ID NO: 32

Gly-Phe-Phe-Tyr-Xaai-Pro-Glu-Xaa₂-Xaa₃-Xaa₄

Where Xaai is Cys, Lys, Prn, diamino-butyric acid or diamino-propionic acid as an attachment point for a diol-containing moiety; where Xaa₂ may be Thr or Ala; where Xaa₃ is Cys, Lys, Prn, diamino-butyric acid or diamino-propionic acid as an attachment point for a diol-containing moiety; and where Xaa₄ is Cys, Lys, Prn, diamino-butyric acid or diamino- propionic acid as an attachment point for a diol-containing moiety. As a further option, either Xaa₃ or Xaa₄ (but not both) may be either Gly or Glu. As yet another option, Xaa₄ may be absent.

[0033] SEP ID NP: 33

Gly-Phe-Phe-Tyr-Thr-Xaa₁-Xaa₂-Xaa₃-Xaa₄-Xaa₅

Where Xaai is Cys, Lys, Prn, diamino-butyric acid or diamino-propionic acid as an attachment point for a diol-containing moiety; where Xaa₂ may be Glu, Ala or Pro; where Xaa₃ may be Thr or Ala; where Xaa₄ is Cys, Lys, Orn, diamino-butyric acid or diamino- propionic acid as an attachment point for a diol-containing moiety; and where Xaa_¾ is Cys, Lys, Orn, diamino-butyric acid or diamino-propionic acid as an attachment point for a diol- containing moiety. As a further option, either Xaa₄ or Xaa_¾ (but not both) may be either Gly or Glu. As yet another option, Xaa_¾ may be absent.

[0034] SEQ ID NO: 34

Gly-Phe-Phe-Tyr-Thr-Xaai-Xaa₂-Xaa₃-Xaa₄-Xaa₅

Where Xaai is Ala, Asn, Asp, Gin, Glu or Thr; where Xaa₂ may be Cys, Lys, Orn, diamino-butyric acid or diamino-propionic acid as an attachment point for a diol-containing moiety; where Xaa₃ may be Thr or Ala; where Xaa₄ is Cys, Lys, Orn, diamino-butyric acid or diamino-propionic acid as an attachment point for a diol-containing moiety; and where Xaa_¾ is Cys, Lys, Orn, diamino-butyric acid or diamino-propionic acid as an attachment point for a diol-containing moiety. As a further option, either Xaa₄ or Xaa_¾ (but not both) may be either Gly or Glu. As yet another option, Xaa_¾ may be absent.

[0035] It is important to note that, prior to use of the above peptides in SEQ ID 29-36 in trypsin-mediated semi-synthesis, the £fes-octapeptide[B23-B30] insulin fragment will first be modified at or near the N-terminus of the A chain with a glucose-binding element as exemplified by phenylboronic acid and its halogenic derivatives.

[0036] Analogues of insulin containing diol-containing adducts of cysteine or basic side chains, as substituted within the C-terminal region of the B chain, will be prepared by trypsin- mediated semisynthesis using synthetic peptides of length 7, 8, 9 or 10 residues. The peptides were be derivatized with diol-containing adducts prior to the semisynthetic reaction. The protocol for semi-synthesis employed a <i<?s-octapeptide[B23-B30] fragment of human insulin or insulin analogue together with a synthetic peptide containing an N-terminal Glycine (octapeptide, nonapeptide, or decapeptide) and a diol adduct at one, two or three sites within the synthetic peptide as specified by the substituted thiol-containing or basic side chain. The £fes-octapeptide[B23-B30] fragment contains the three native disulfide bridges of wild-type insulin; the protocol including purification of the fragment, peptide, and product by high- performance liquid chromatography was a modification of that described (Mirmira, R.G., and Tager, H.S., 1989. /. Biol. Chem. 264: 6349-6354.) This protocol employs (i) a synthetic peptide containing a monosaccaride pyranoside adduct (SEQ ID NO: 53-65) and (ii) truncated analogue <ie5-tripeptide[B l-B3]-<ie5-octapeptide[B23-B30]-insulin, or in the case of [HisA4, His^A8, Gly^A21]-insulin analogues, [HisA4, His^A8, Gly^A21]-d<?s-tripeptide[Bl-B3]-d<?s-

B13 B 13

octapeptide[B23-B30]-insulin, or in the case of Gin -insulin analogues, Gin -des-

A8

tripeptide[Bl-B3]-<i<?s-octapeptide[B23-B30]-insulin, or in the case of His -insulin analogues, The following literature is cited to demonstrate that the testing and assay methods described herein would be understood by one of ordinary skill in the art.

[0037] Diol-containing reagents appropriate to modify thiol-containing side chains (such as cysteine or homocysteine) in a variant peptide corresponding to residues B23-B29, B23- B30, B23-B31 or B23-B32 of an insulin analogue are exemplified but not restricted to 1- thioglycerol, the 4-mercapto-derivative of 1,2,3-butanetriol, and the 1-thio-derivative of β-D- glucose. Diverse diol-containing reagents appropriate to modify (via an N- hydroxysuccinamide (NHS) activation step; see below) the basic side chains of Lysine, Ornithine, diamino-butyric acid or diamino-propionic acid in a variant peptide corresponding to residues B23-B29, B23-B30, B23-B31 or B23-B32 of an insulin analogue are given in Table 1. The insulin analogues of the present invention may also contain one or more diol- containing monosaccharides, disaccharides or oligosaccharides linked to the side chains of Ser, Thr, Asn or Gin. The insulin analogues of the present invention are not restricted to the above examples of diol-containing elements. For example, diol-containing elements may contain nitrogen functional group s (such as in (±)-3-amino-l,2-propanediol and glucosamine hydrochloride) amenable to activation and linkage to the amino groups of side chains in the above-cited synthetic peptides. Indeed, we envision a large molecular diversity of potential diol-containing adducts and linkage chemistries that may share the shared attribute of intramolecular interaction with a phenylboronic acid moiety linked at or near the N-terminus of the A chain such that this interaction is competable by high concentrations of exogenous glucose.

[0038] Table 1. Diol-containing Elements for NHS-Activated Modification of Side-Chain Amoni Groups at or near the C-terminus of the B chain of Insulin.

The following additional biochemical procedures have been demonstrated.

[0039] NHS activation of Phenylboronic Acid: The activation of 4-carboxy-3-fluoro- phenylboronic acid pinacol ester (PBA) from Combi-Blocks (San Diego, CA) was achieved with N-hydroxysuccinamide (NHS) (Sigma Aldrich, St. Louis, MO). PBA (300 mg, 1.126 mmol) was dissolved in 4.4 mL ethyl acetate and incubated at 4°C for 20 minutes. To this, 131 mg (1.14 mmol) NHS and 247 mg (1.86 mmol) dicyclohexyl carbodiimide (DCC) (Sigma Aldrich, St. Louis, MO) were added. The solution was stirred at room temperature overnight. N, N' -dicyclohexyl urea byproduct was removed via centrifugation for 5 minutes at 13,500 rpm in an Eppendorf® microfuge. The solvent was removed ex vacuo. PBA-N- hydroxysuccinimide ester was purified by recrystallization from acetone and n-hexane.

Product was made 30 mg/ml in acetonitrile.

[0040] Coupling of PBA to N-termini of Insulin Polypeptides: Des-octapeptide insulin

(DOI) (5 mg in 100 μΐ of 0.1 M sodium carbonate pH 7.6) was combined with 100 μΐ of NHS- ester of PBA (30 mg/ml) in acetonitrile. The solution was agitated at 25 °C for 2 hours.

Single- and double-coupled DOI molecules were purified by reverse-phase high performance liquid chromatography (HPLC). A Waters® 2535 quaternary gradient chromatography system was used with a Higgins Analytical® Proto 300 C4 column (ΙΟμιη, 250x20mm). A two-buffer mobile phase was used for purification: aqueous 0.1% trifluoroacetic acid (TFA) (buffer A) and 0.1% TFA in acetonitirile (buffer B) with a gradient of 5-95% buffer B over 40 minutes. Protein elution time was monitored by UV absorbance at 215 and 280 nm using a Waters® 2489 UV/Vis detector. Single- and double-coupled molecules were eluted at 18 minutes and 19.5 minutes, respectively. Identity of molecules was confirmed by MALDI-TOF mass spectrometry using an Applied Biosystems® 4700 Proteomics Analyzer. A saturated solution of a-cyano-4-hydroxycinnamic acid (a-CHCA) (Sigma- Aldrich, St. Louis, MO) in 50% acetonitrile 0.1% TFA was used as matrix. Masses showed loss of two hydroxyl groups associated with anhydride formation upon laser desorption reported in previous literature (Hoeg-Jensen, et al). (Mass of Single-Coupled DOI: 4997, Mass of Double-Coupled DOI: 5127) The reaction was found to sequentially couple PBA to the N-terminus of the A-chain before coupling to the B -chain. Identity of the single-coupled analog was confirmed after reduction with 50 mM dithiothreitol (DTT) in lx PBS pH 7.4 for 1 hour at room temperature. Protein was desalted using Millipore® CI 8 ZipTip® pipette tips and eluted into 5 iL a-CHCA matrix. Masses of individual polypeptide chains were confirmed with MALDI-TOF mass spectrometry as described above (mass PBA-coupled A-chain: 2515, mass B-chain: 2488). Samples were lyophilized using a Labconco® Freezezone 6® lyophilizer. [0041] Diol Coupling: Double cysteine containing peptides, octapeptide (Gly-Phe-

Phe-Tyr-Thr-Cys-Orn-Cys), nonapeptide (Gly-Phe-Phe-Tyr-Thr-Pro-Cys-Thr-Cys), and decapeptide (Gly-Phe-Phe-Tyr-Thr-Pro-Orn-Cys-Gly-Cys), (where "Orn" designates the nonstandard amino acid, ornithine) were obtained from the molecular biotechnology core at the Cleveland Clinic Foundation (Cleveland, OH). Peptides were made 1 mM in 10 mM sodium phosphate buffer pH 7.3. 1-thioglycerol from Sigma- Aldrich® (St. Louis, MO) was used as a diol adduct. 1-thioglycerol was added to the solution to a final concentration of 20 mM.

Reaction was incubated at 12° C with agitation and exposure to room air for 48 hours.

Reaction was quenched with the addition of crystallized iodoacetic acid from the Aldrich Chemical Company (Milwaukee, WI) to a final concentration of 400 mM and incubated overnight under low-light conditions. Diol-coupled peptides were isolated by reverse-phase HPLC utilizing the system described above with a Higgins Analytical® TARGA C8 column (250x20mm 5 μιη) and a 5-95% buffer B gradient over 50 minutes. Identity of peptides was confirmed via MALDI-TOF mass spectrometry and lyophilized using the system described above. In addition, gluconic acid will be coupled to appropriate variant octapeptides via the side chains of ornithine, diaminobutyric acid, and/or diamino-propionic acid (variants of the naturally-occurring basic amino-acid lysine). NHS crosslinker chemistry, as described above, will be used to accomplish this reaction.

Hexamer disassembly: Cobalt insulin hexamer were utilized to measure hexamer disassembly rates by optical absorption spectroscopy using an EDTA sequestration assay. An analogue containing Lys B28, Pro B29 substitutions ("lispro" or "KP" insulin; SEQ ID NOs: 4 and 2) was additionally modified with a fluorinated phenylboronic acid adduct (F-PBA) at position Al and a mannose adduct at B30. This analogue is designated T-2020. KP insulin was used as a control. Disassembly of phenol- stabilized R₆ Co²⁺-substituted insulin hexamers was measured as follows. The insulin analogues were made 0.6 mM in a buffer containing 50 mM Tris-HCl (pH 7.4), 50 mM phenol, 0.2 mM CoCl₂ and 1 mM NaSCN (Roy et al., /. Biol. Chem. 264, 19081-19085). Samples were incubated overnight at room temperature prior to the studies to ensure that a conformational equilibrium was reached. Spectra (450-700 nm) were obtained to monitor tetrahedral Co²⁺ coordination with its signature peak absorption band at 574 nm (Roy et al., /. Biol. Chem. 264, 19081-19085). To determine the rate of Co²⁺ release from the hexamers, metal-ion sequestration was initiated at 25 °C by addition of an aliquot of ethylene-diamine-tetra-acetic acid (EDTA; 50 mM at pH 7.4) to a final concentration of 2 mM. Attenuation of the 574 nm absorption band was monitored on a time scale of seconds to hours. As shown in Table 2 and Figs. 8A and 8B, the insulin analogue containing a PBA-diol tether provides greater stabilization of the insulin hexamer. It also provides increased hexamer disassociation in the presence of glucose, providing faster action during periods of hyperglycemia.

TABLE 2

T_{1 2} (minutes)

Glucose Level KP T-2020

None 1.35 5.58

25 mM 1.49 3.12

Claims

CLAIMS What is claimed is:

1. An insulin analogue containing a modification of the A chain at or within 4 residues of its N-terminus by a monomeric glucose-binding moiety, with a spacer element, and a variant B chain containing one or more diol-containing modifications at or within six residues of its C-terminus.

2. An insulin analogue of Claim 1 wherein the monomeric glucose-binding moiety is selected from the group consisting of a phenylboronic acid, a halogen-modified phenylboronic acid derivative, and the spacer element is an acyl group containing 3-16 carbon atoms.

3. An insulin analogue of Claim 1 or Claim 2 wherein the monomeric glucose- binding moiety is linked to the alpha-amino functional group of Gly^A1, is linked to the epsilon-amino functional group of a D-Lysine or L- Lysine substituted at position Al, is linked to the delta-amino functional group of a D-Ornithine or L-Ornithine substituted at position Al, is linked to the gamma-amino functional group of a D-a,y-diaminobutyric acid or L-α,γ- diaminobutyric acid substituted at position Al, is linked to the beta-amino function of a D- α,β-diaminopropionic acid or L-α,β- diaminopropionic acid substituted at position Al, is linked to the epsilon-amino functional group of an L-Lysine substituted at position A4, is linked to the side-chain amino functional group of a variant A chain substituted at position A4 by a non-standard basic amino acid selected from the group consisting of L-α,γ- diaminobutyric acid, and L-a, -diaminopropionic acid, or is linked simultaneously to the side-chain amino functional group of an A4 substituent as drawn from the group consisting of L-Lysine, L-Ornithine, L-a,y-diaminobutyric acid, and L- α, β -diaminopropionic acid.

4. An insulin analogue of Claims 3, wherein the monomeric glucose-binding moiety linked to the alpha-amino functional group of the Al substituent.

5. An insulin analogue of Claim 3, wherein the monomeric glucose -binding moiety linked simultaneously to the alpha-amino functional group of the Al substituent.

6. An insulin analogue of claim 1 or claim 2 additionally comprising a monomeric glucose-binding moiety linked simultaneously to (a) the alpha-amino functional group of an Al substituent drawn from the group consisting of L- or D-Lysine, L- or D-Ornithine, L- or D-a,y-diaminobutyric acid, and L-or D-a, -diaminopropionic acid, (b) the side-chain amino functional group of the said Al substituent and (c) the side-chain side-chain amino functional group of an A4 substituent as drawn from the group consisting of L-Lysine, L-Ornithine, L- α,γ-diaminobutyric acid, and L-oc, -diaminopropionic acid.

7. An insulin analogue of claim 6 additionally comprising a diol-containing modification of the epsilon-amino functional group of Lys^B29, a diol-containing moiety linked simultaneously to the side-chain amino functional group of a non-standard amino-acid substituent at position B29 drawn from the group L-Ornithine, L-a,y-diaminobutyric acid, and L-a, -diaminopropionic acid, a diol-containing moiety linked simultaneously to the epsilon- amino functional group of Lysine as substituted at position B28, a diol-containing moiety linked simultaneously to the side-chain amino functional group of a non-standard amino-acid substituent at position B29 drawn from the group L-Ornithine, L-a,y-diaminobutyric acid, and L-a, -diaminopropionic acid, or an S-linked diol-containing moiety at one or more cysteines (or homocysteine) as substituted at positions within the wild-type B27-B30 segment and/or within a B-chain extended segments (B27-B31 or B27-B32) such that the derivatized amino acid is cysteine or homocysteine.

8. An insulin analogue of of claim 1 or claim 2 additionally comprising an amino- linked diol-containing moiety at one or more positions within the wild-type B27-B30 segment and/or within a variant B27-B30 segment such that the derivatized amino acids are drawn from the set of basic side chains consisting of Lys, Orn, diamino-butyric acid or diamoni- propinic acid.

9. An insulin analogue of of claim 1 or claim 2 wherein the B chain is modified by two or three diol-containing elements such that at least one such element is a monosaccharide, disaccharide or oligosaccharide attached to the side chain of Ser, Thr, Asn or Gin as naturally occurring residues (Thr^B27 or Thr^B3°) or as substituted at or near the C-terminus of the B chain or as optionally contained within a one- or two-residue extension of the B chain (residues B31 and B32).

10. An insulin analogue of of claim 1 or claim 2 additionally comprising a substitution at position A8 drawn from the group consisting of Glutamic Acid, Glutamine, Histidine, Lysine or Arginine.

11. An insulin analogue of of claim 1 or claim 2 additionally comprising a Tryptophan substitution at position A13.

12. An insulin analogue of of claim 1 or claim 2 additionally comprising a nonstandard substitution at position B24 drawn from the group consisting of Cyclohexanylalanine, penta-fluoro-Phenylalanine or a mono-halogenic derivative of Phenylalanine at the ortho, meta or para positions where the halogen atom is fluorine, chlorine, bromine or iodine.

13. An insulin analogue of of claim 1 or claim 2 additionally comprising a Glutamic Acid substitution at position B29.

14. An insulin analogue of of claim 1 or claim 2 additionally comprising a Proline substitution at position B29.

15. An insulin analogue of of claim 1 or claim 2 additionally comprising an Aspartic Acid substitution at position B28.

16. An insulin analogue of of claim 1 or claim 2 additionally comprising a Lysine substitution at position B28 and a Proline substitution at position B29.

17. An insulin analogue of of claim 1 or claim 2 additionally comprising a non-proline natural amino acid substitution at position B28 and a Proline substitution at position B29.

18. An insulin analogue of of claim 1 or claim 2 wherein the B chain comprises at least one site that contains a monosaccharide selected from the group consisting of glucose, mannose, or N-acetyl-galactose.

19. An insulin analogue of of claim 1 or claim 2 wherein the B chain additionally comprises at least one site that contains a disaccharide drawn from the group consisting of glucose-glucose, mannose-mannose, glucose-mannose or mannose-glucose.

20. An insulin analogue of of claim 1 or claim 2 wherein the B chain additionally comprises at least one site contains a an oligosaccharide that is branched.

21. An insulin analogue whose B chain conforms to the polypeptide sequences provided in any one of SEQ ID NOs: 7-26.

22. A peptide sequence of length 7, 8, 9 or 10 containing an N-terminal glycine conforming to SEQ ID NOs: 27-34.

23. A method of treating a patient comprising administering a physiologically effective amount of an insulin analogue or a physiologically acceptable salt thereof to the patient, wherein the insulin analogue is as provided in claiml or claim 2.