INSULIN ANALOGUES
FIELD OF THE INVENTION
The present application broadly relates to insulin analogues, and methods for the
synthesis of these analogues, including dicarba-disulfide bridge-containing insulin analogues.
BACKGROUND TO THE INVENTION
Diabetes is a chronic disease characterised by the onset of hyperglycaemia. This
metabolic disorder arises either upon failure of the pancreas to secrete effective
concentrations of insulin in the case of type 1 diabetes, or as a result of a build up of
resistance by cells towards any bio-available insulin (type 2). Insulin is a peptidic hormone
required by the body to transport glucose from the bloodstream to cells for energy. Failure of
its production or effective use results in impaired glucose metabolism, and thus allows
abnormally high quantities of this sugar to accumulate in the blood. In 2003, international
surveys showed 194 million people have diabetes, and it is widely considered to be the
world's fastest growing disease. Over one million Australians suffer from this disorder and it
is Australia's sixth leading cause of death. These figures are cause for concern and
unfortunately at present there is no cure. Diabetes is however manageable, but treatments
such as insulin injections must be carried out everyday over a patient's lifetime after
diagnosis, to allow a healthy and fulfilling life.
Discovery and isolation of the insulin molecule by Fredrick Banting and Charles Best
in 1921 provided a significant breakthrough for those suffering diabetes, whose treatments
until this time included only starvation diets. Medical benefits aside, scientists found insulin
to be a fascinating molecule. It was the first protein to have its primary structure elucidated,
and was the focal point of many Nobel Prize winning research projects. Knowledge of the
key molecular determinants of insulin function is important not only for examining the
downstream pathways leading to its physiological effects but also the development of new
clinical compounds and biochem ical/pharmacological probes. Despite the fact that nearly 85
years have passed since insulin was first isolated and 35 years since its structure was
determined by X-ray crystallography, a high resolution, three dimensional structure of the
insulin-receptor complex is still unavailable. Consequently, several key aspects of insulin's
biochemistry remain to be discovered and understood. Recent studies using synthetic insulin
analogues support the concept that insulin binds asymmetrically to two discrete sites within
the receptor dimer, however, the biologically active conformation of the insulin molecule,
complete structural features that constitute the receptor-binding domain, and the mechanism
of insulin receptor activation, remain unknown.
The insulin receptor is a tetrameric integral membrane glycoprotein consisting of two
735 amino acid a-chains and two 620 amino acid b-chains. The a-chain resides on the
extracellular side of the plasma membrane and contains the cysteine-rich insulin binding
domain. Covalent insulin-receptor complexes have been isolated, however, the exact nature
of this molecular interaction has yet to be fully elucidated. Insulin is firstly synthesised as a
pre-prohormone precursor containing signal peptide-B-C-A domains. Mature insulin,
however, consists of a 5 1 residue A-B heterodimer that is covalently linked by two interchain
disulfide bonds. An intrachain disulfide bridge is also located within the insulin A-chain. The
circulating and biologically active form of insulin is monomeric and its primary structure has
been determined for at least 100 vertebrate species (Figure 1) . Of these, only six cysteine
residues and ten other amino acids are fully conserved during evolution. The primary
structure of insulin from several species is shown in Figure 1. Invariance of these residues
may be indicative of their key roles in receptor binding and/or maintaining a biologically
active conformation.
Disulfide bonds serve structural and functional roles in peptides. In some peptides,
the disulfide is involved in disulfide exchange chemistry at the receptor resulting in activation
of receptors and downstream signaling. Yet in other peptides, the S-S motif serves only to
preserve and/or create a bioactive conformation of the peptide. In such cases, chemistry can
be used to replace the native disulfide with other bridging amino acid residues. Structural
alteration will be well tolerated if key receptor interactions can be preserved and this will
depend on the surrogate motif's ability to replicate native peptide tertiary and secondary
structure. The role of each of insulin's disulfide bridges is currently unknown but it is highly
likely that they play a key role in regulating insulin's function at its cell surface tyrosine kinase
receptors. Invariance of the cysteine residues across the species highlights the importance
of the S-S bridges to insulin's structure and function. The cystine framework found in insulin
is also found in other so called 'insulin superfamily' molecules, e.g. relaxin. Despite their
framework similarity to insulin, however, the relaxins bind and activate a different receptor,
the G-protein coupled receptor, and are responsible for remarkably distinctive biological
roles. Hence, nature has capitalized on a generic disulfide template to perform diverse
neuroendocrine through to homeostasis roles.
Problems with insulin as a drug still exist and improved analogues are continually
being sought. For example, insulin can only be taken by injection, as oral delivery results in
protein cleavage by digestive enzymes before it can be absorbed into the bloodstream . In
addition, insulin is generally refrigerated before use so that it does not degrade before it is
injected. There is therefore a need for insulin analogues which display enhanced stability to
proteolyic enzymes and those which can be stored at room temperature to facilitate transport
and simplify storage.
Several studies have been conducted to determine the role of disulfide bonds in
insulin and all have concluded that each cystine bridge, to varying degrees, is required to
maintain biological function. In previous studies, however, cysteine residues have been
replaced by non-bridging amino acids such as alanine and serine via site-directed
mutagenesis. Previous research is difficult to analyse, since loss of biological activity cannot
be unam biguously attributed to either unfolded protein structure and hence loss of a binding
domain, or loss of a reactive disulfide motif. There is therefore a need to further investigate
the role of the disulfide bonds in insulin.
Although insulin therapy provides a better quality of life for those afflicted with
diabetes, it also remains a difficult drug to use. A narrow therapeutic index, inconsistent
magnitude of effect, poor oral availability, weight gain, poor stability and lim ited potency are
just of some of insulin's many limitations. Each of these make it very difficult for a diabetic
patient to treat their disease and maintain the tightly regulated glucose concentrations of a
healthy individual. The achievement of a "normal" 24 hour physiological insulin profile is
nearly impossible to mimic. The nature of meals, exercise regimes, sleeping patterns,
development of infections and endogenous production of glucose of the liver, are just some
of the many causes of fluctuation in blood glucose levels. It is essential that basal insulin
levels are maintained in the body 24 hours a day, but additionally extra supplies (bolus) are
needed for the management of glucose ingested at meal times. Unfortunately, however, the
ideal therapeutic window for insulin dosing is very narrow (4-6 mmol), and one of the major
drawbacks of intensive insulin treatments is the potential for hyperglycaemic events (low
blood sugar levels). The advent of recombinant access to insulin in the early 1980s, led to
the development of analogues that are specifically engineered to accomplish slower, faster
and more predictable activity profiles that help with tim ing, dosing and the maintenance of
more stable physiological insulin levels (Figure 2A and 2B). Existing insulin analogues still,
however, suffer from storage and stability problems.
Although there are a number of commerically available analogues of insulin, there is
still a considerable desire to improve basal insulin treatments. This is partly because some
of these existing analgoues have significant drawbacks. For example, some show enhanced
IGF-1 (insulin-like growth factor 1) receptor binding activity which can result in enhanced
mitogenic potency. Other analogues possess reduced potency in vivo. Accordingly, there is
a need for insulin analogues with superior therapeutic profiles, as well as analogues which
are able to retain all of the beneficial properties of the unmodified native insulin sequence
while addressing the inherent physicochemical instability.
SUMMARY OF THE INVENTION
This application relates to a range of new insulin derivatives and methods for the
synthesis of these analogues, including dicarba-disulfide bridge-containing insulin analogues.
The dicarba insulin analogues described herein display unrivalled biological activity
and formulation stability. Several of the analogues show equipotent binding in competitive
insulin binding studies at the insulin receptor (e.g. Figure 4A) and excellent activity in
phosphorylation studies at the insulin receptor (e.g. Figure 4B). The analogues are also
equipotent in stimulating glucose uptake in differentiated L6 GLUT4-myc cells (Figure 7) and
are also active in in vivo glucose uptake tests where some dicarba analogues lowered blood
glucose levels over a 2 hour time period with a profile equivalent to Actrapid (Figure 8).
Significantly however, many of the dicarba analogues showed significantly less binding
affinity for the IGF receptor compared to IGF and insulin (Figure 5A). A study evaluating
DNA stimulation by the dicarba analogues (Figure 5B), a measure of mitogenic activity,
showed significantly reduced activity as compared to insulin itself. It has been suggested
that insulin analogues causing excess stimulation of the IGF-1 receptor (e.g. glargine) have
the potential for promoting mitogenic effects and hence may increase the risk of cancer in
insulin users. More significantly, the dicarba insulin analogues also display remarkable
stability at room temperature. Cystine replacement with a dicarba bridge provides molecules
which do not need to be refrigerated to preserve biological activity and potency. Thermally
stable insulins are in great demand for the effective management of diabetic patients in
remote and third world countries.
According to one embodiment, there is provided a dicarba analogue of insulin
comprising an A-chain and a B-chain or fragments, salts, solvates, derivatives, isomers or
tautomers of the A-chain, the B-chain or both, provided that the dicarba analogue is not A7,
B7-(2,7-diaminosuberoyl]-des-(B26-B30)-insulin B25-amide.
Preferably, the dicarba analogue of insulin includes an intrachain or an interchain
dicarba bridge. It is also preferred that the dicarba bridge is unsaturated.
The only previous study in relation to a dicarba analogues of insulin attempted to
replace the A7-B7-disulfide bridge with a dicarba isostere, (S, S)-2,7-diaminosuberic acid, by
solution phase convergent synthesis of four separate peptide fragments: A 1-6, A8-21 , B9-25
and a B1-8 fragment containing a preformed, orthogonally protected, 2,7-diamino-suberic
acid unit at the B7 position. This study produced a truncated insulin molecule (46 residues,
native insulin has 5 1 residues) as an unresolvable product which was cited to contain the
target analogue 'in only a small amount'. Although the analogue mixture exhibited some
biological activity, it was not possible to predict the intrinsic activity of this dicarba insulin.
The fraction containing the postulated truncated and saturated dicarba-A7-B7-insulin
molecule was subjected to receptor binding on cultured human lymphocytes. Relative
binding affinity was determined to be 0 .1% of human insulin. The crude, truncated and
saturated dicarba-A7-B7-insulin preparation was also tested for insulin activity via
lipogenesis in isolated rat adipocytes where it was found to possess an ED50 of 4.8x1 0 9 M:
The relative potency was quoted as 0.66% on a molar, and 0.72% on a weight basis,
compared to human insulin.
A "dicarba analogue" refers to a peptide which contains an amino acid sequence
corresponding to a naturally occurring, native or synthetic insulin, but containing at least one
dicarba bridge either as an addition to the peptide, or as a substitution for one or more of the
bridged cystine-amino acid residue pairs.
"Dicarba-substituted" analogues, which are analogues of naturally-occurring, native
or synthetic insulins, but with one or more of the disulfide bridge form ing cystine amino acid
residue pairs, are a subclass of particular interest. A notable subclass of the dicarba
analogues are the mono-dicarba analogues (which retain one or more of the disulfide
bridges), and the bis- and higher dicarba analogues of insulin.
The present invention also relates to methods for the synthesis of these dicarba
insulin analogues.
According to one embodiment, there is provided a method for the synthesis of a
dicarba analogue of insulin comprising an A-chain and a B-chain or fragments, salts,
solvates, derivatives, isomers or tautomers of the A-chain, the B-chain or both, the method
comprising:
(i) providing the A-chain having at least one pair of complementary metathesisable
groups;
(ii) subjecting the A-chain to metathesis to form at least one dicarba bridge between the
pair of complementary metathesisable groups; and
(iii) adding the B-chain.
The phrase "adding the B-chain" is taken to refer to any means by which the B-chain
of insulin or a fragment, salt, solvate, derivative, isomer or tautomer thereof may be
combined with the A-chain.
According to another embodiment, there is provided a method for the synthesis of a
dicarba analogue of insulin comprising an A-chain and a B-chain or fragments, salts,
solvates, derivatives, isomers or tautomers of the A-chain, the B-chain or both, the method
comprising:
(i) providing part of the A-chain having at least two complementary metathesisable
groups;
(ii) subjecting the A-chain to metathesis to form at least one dicarba bridge; and
(iii) adding one or more further amino acids to one or both ends of the A-chain; and
(iv) adding the B-chain.
According to a further embodiment, there is provided a method for the synthesis of a
dicarba analogue of insulin comprising an A-chain and a B-chain or fragments, salts,
solvates, derivatives, isomers or tautomers of the A-chain, the B-chain or both, the method
comprising:
(i) providing a part of the A-chain and/or a part of the B-chain having at least two
complementary metathesisable groups between them ;
(ii) subjecting the A-chain and the B-chain to metathesis to form at least one dicarba
bridge; and
(iii) adding one or more further amino acids to one or both ends of the A-chain and/or Bchain.
According to yet a further embodiment, there is provided a method for the synthesis
of a dicarba analogue of insulin comprising an A-chain and a B-chain or fragments, salts,
solvates, derivatives, isomers or tautomers of the A-chain, the B-chain or both, the method
comprising:
(i) providing a readable peptide comprising a removable tether between the A-chain
and the B-chain, the A-chain and the B-chain each having at least one
complementary metathesisable group;
(ii) subjecting the readable peptide to metathesis to form at least one dicarba bridge
between the complementary metathesisable groups; and
(iii) removing the removeable tether to produce a dicarba bridge linking the A-chain and
the B-chain of insulin.
The method may also be used to include a second dicarba bridge. This may be
achieved by any of the methods described above, wherein further complementary
metathesisable groups are provided in the A-chain and/or the B-chain, and the method
comprises subjecting the peptide to a further metathesis step to form a further dicarba bridge
between the complementary metathesisable groups.
The present invention further provides an anti-hyperglycemic agent comprising a
dicarba analogue of insulin or a fragment, salt, solvate, derivative, isomer or tautomer
thereof, as defined above.
The dicarba analogue of insulin may be provided as a pharmaceutical composition
together with a pharmaceutically acceptable carrier.
The dicarba analogue of insulin may also be used to reduce hypoglycemia, or for
the treatment of diabetes mellitus or metabolic syndrome.
BRIEF DESCRIPTION OF THE FIGURES
Figure 1 is a schematic diagram showing the primary structure of insulin from
several species. Invariant residues are shaded grey with the cysteine residues coloured dark
grey.
Figure 2A is a schematic diagram showing the A-chain and the B-chain of human
insulin and the disulfide bridges present in human insulin compared to rapid acting insulin
analogues, lispro, apart and glulisine.
Figure 2 B is a schematic diagram showing the A-chain and the B-chain of human
insulin and the disulfide bridges present in human insulin com pared to long acting insulin
analogues, glargine and detemir.
Figure 3 is a schematic diagram showing an RCM approach using tethers to
introduce interchain dicarba bridges.
Figure 4A is a graph showing competition binding curves for Eu-labelled insulin
(ActRapid®) binding to the immunopunfied human IR-B in the presence or absence of dicarba
analogues 13(1) and 13(11).
Figure 4B is a graph showing competition binding curves for Eu-labelled insulin
(ActRapid®) binding to the immunopunfied human IR-B in the presence or absence of dicarba
analogues 52(1) and 52(11).
Figure 5A is a graph showing activation of the immunopunfied human IR isoforms
by native insulin (ActRapid®) and intrachain dicarba analogues 13(1) and 13(11) to IR-B.
Figure 5B is a graph showing activation of the immunopunfied human IR isoforms
by native insulin (ActRapid®) and intrachain dicarba analogues 13(1) and 13(11) to IR-A.
Figure 6A is a graph showing competition binding curves of Eu-labelled IGF-I
binding to the immunopunfied human IGF-IR in the presence of each of 13(I) A, 13(I) B, 13(II) A
and 13(II)B and native insulin (ActRapid®) , with results expressed as the percentage of Eu-
IGF-I bound in the absence of competing ligand (B/B0) .
Figure 6B is a graph showing stimulation of DNA synthesis in response to IGF-I,
insulin and dicarba insulins 13(I) A, 13(I) B, 13(II) A and 13(II) B using L6 myoblast cells over
expressing IR-B.
Figure 7 is a graph showing a dose-response study of dicarba insulin (1 3(1) and
13(11)) stimulated 2-deoxy-D-glucose uptake in differentiated L6 GLUT4-myc cells.
Figure 8 is a graph showing the blood glucose content in anesthetised 7 week old
male C57BL/6 mice following intraperitoneal bolus injection with dicarba analogues 13(1) and
13(11) or ActRapid ®.
DETAILED DESCRIPTION OF THE INVENTION
As described above, this application relates to insulin analogues, and methods for
the synthesis of these analogues, including dicarba-disulfide bridge-containing peptides.
Insulin and dicarba analogues of insulin
The term "insulin" as used herein is used in it broadest sense, and encompasses: (i)
the peptide or peptide fragments that are produced in the islets of Langerhans in the
pancreas, (ii) peptide or peptide fragments that are produced in the islets of Langerhans in
the pancreas or analogues thereof which can be used in the treatment of diabetes mellitus,
(iii) naturally-occurring or native human insulin, (iv) naturally-occurring or native insulin
derived from non-human organisms, such as for example bovine or porcine insulin, (v) any
insulin that is homologous to human, bovine or porcine insulin, (vi) insulin extracted from
bovine and/or porcine sources, (vii) synthetically produced insulin, (viii) recombinantly
produced insulin, and mixtures of any of these insulin products. The term insulin is also
intended to encompass a fragment thereof, which may be one of the A-chain or the B-chain,
both the A-chain and the B-chain or any truncation of the A-chain and/or the B-chain. The
term insulin also includes an insulin which may include additional amino acid residues on the
N- or C-terminus of either the A-chain or the B-chain. The term insulin is also intended to
encompass the polypeptide normally used in the treatment of diabetes in a substantially
purified form but encompasses the use of the polypeptide in its commercially available
pharmaceutical form , which includes additional excipients. The term insulin also includes
insulin analogues or derivatives of insulin. Examples of insulin analogues include rapidacting
and slow-acting insulins.
The term "dicarba analogue of insulin" or "dicarba insulin" or "dicarba insulin
analogue" as used herein is intended to encompass any form of "insulin" as defined above,
which shares a common functional activity with insulin itself and typically shares common
structural features as well and which contains a dicarba bridge either as an addition to the
peptide or as a substitution for one or more of the bridged cystine-amino acid residue pairs.
The "dicarba analogue of insulin" encompasses any form of "insulin" as defined above, in
which two or more amino acids are replaced by residues which form a dicarba bridge.
Preferably, the dicarba bridge or bridges replace structural features present in the native
insilun peptide. "Dicarba-substituted" analogues of insulin are analogues of insulin in which
with one or more of the disulfide bridge forming cystine amino acid residue pairs are replaced
by a dicarba bridge. Such analogues are a subclass of particular interest.
The dicarba bridge may replace one or more of the disulfide bridges located on the
A- and/or B-chain of insulin or salt bridges or non-covalent interactions which contribute to
the structure of native insulin. The dicarba bridge may be an intrachain dicarba bridge
located in the A-chain of insulin and/or an interchain dicarba bridge linking the A-chain and
the B-chain of insulin. A notable subclass of the dicarba analogues are the mono-dicarba
analogues, and the bis- and higher dicarba analogues.
The molecular nature of insulin
Without synthetic modifications, the insulin monomer exists as a small globular
protein comprising two separate chains, the A and the B chain. In humans, these
polypeptides are composed of 2 1 and 30 amino acids, respectively. The main structural
features of this molecule include three a-helices over residues A2-A8, A 13-A1 9 and B9-B 19 ,
a hydrophobic core and a primarily polar outer surface with two hydrophobic faces. At
micromolar concentrations insulin dimerises, and further associates into hexamers in the
presence of zinc ions, to bury the hydrophobic faces. Studies have shown that these
hydrophobic chain portions are essential for receptor binding, hence monomeric insulin is the
circulating and biologically active form of the molecule.
Additional key features of the insulin molecule include its three disulfide (or cystine)
bridges (These are shown in the schematic diagram of the insulin A and B chains shown in
Figures 1, 2A and 2B). Two of the three disulfide bridges are interchain bridges joining the A
and B polypeptides together at the A7-B7 and A20-B 19 positions, and one intrachain link
creating a monocycle between the A6 and A 11 residues (as shown in Figures 1, 2A and 2B).
Disulfide bridges form upon oxidation of thiol side chains between cysteine residues within a
peptide, and help to stabilise the secondary and tertiary structure of a protein. Although they
do exist under physiological conditions, these bridges can be both chemically and
metabolically labile, showing a tendency to undergo facile decomposition to reactive thiol
groups under reducing, nucleophilic and basic conditions.
Rapid and long acting insulins
The dicarba analogues of insulin of the present invention may be rapid or long
acting insulins. The activity of the insulin may be modulated by mutating the native amino
sequence to disrupt the self association between monomers of insulin, or by modification of
one or more amino acids of the native insulin sequence. The activity of the insulin may be
modulated by mutating the native amino sequence or by modification of one or more amino
acids of the native insulin sequence to disrupt the solubility of the insulin. These
modifications are encompassed by the dicarba analogues of insulin of the present invention.
Rapid-acting insulins, as the name suggests, have an almost immediate onset of
action. At a molecular level, the primary sequence of these analogues have been modified
to prevent the formation of dimers and hexamers through self-association, which limit the
rate that monomers are absorbed into the blood stream . Through disruption of b-sheet
interactions in the B-chain by charge repulsion, or via alterations in the hydrophobic
interactions responsible for self-association, three commercially available rapid-acting insulin
formulations have been developed: Insulin lispro (Humalog®, Eli Lilly), insulin aspart
(NovoRapid®, Novo Nordisk) and insulin glulisine (Apidra®, Sanofi Aventis). The structures of
these rapid-acting insulin analogues, lispro, aspart and glulisine, compared with native
human insulin are shown in Figure 2A. These modifications to the native insulin sequence,
and the corresponding dicarba analogues of insulin are encompassed by the present
invention.
Insulin lispro utilises the latter strategy to disrupt the self-association surfaces of the
insulin dimer through a simple ProB28 ® LysB29 inversion. This sequence change leads to an
analogue with more rapid pharmokinetic properties; it possesses an enhanced speed of in
vivo action and clearance after subcutaneous injection. Insulin aspart and glulisine, on the
other hand, both utilise charge repulsion strategies to disrupt association via a single amino
acid substitution of aspartic acid for ProB28 in the case of aspart, and glutamic acid for LysB29
in glulisine. Both of these analogues possess similar pharmacokinetic and
pharmacodynamic properties to insulin lispro. In their storage vials, insulin lispro and aspart
exist as stabilised Zn(l l) hexamers that dissociate into monomeric forms more readily than
native insulin formulations. Glulisine, however, is not administered as a Zn(l l) hexamer and
has overcome inherent storage-stability issues via formulation with detergent. These
modifications to the native insulin sequence, and the corresponding dicarba analogues of
insulin are encompassed by the present invention.
In direct contrast, intermediate- and long-acting insulin preparations display slower
onsets of action and can often last up to 24 hours. Initial preparations involved altering the
solubility profile of native insulin via complexation with zinc, or using additives such as
protamine or phenol-like derivatives. This produced the original isophane insulin
suspensions (NPH), Humulin N/L and Novolin N/L, which are absorbed slowly from the
subcutaneous injection site. Unfortunately, use of these intermediate-acting insulin
suspensions produces highly variable results, leading to inconsistent onset and duration of
effect. More recently, however, sequence modification has been used to extend the duration
of insulin action in a safe and reproducible manner. Two analogues, insulin glargine and
detemir, are now available for commercial application (Figure 2B). Insulin glargine exploits
the reduced solubility of insulin at physiological pH to control release. Addition of two arginine
residues to the C-terminus of the B-chain, in addition to a glycine substitution at A21 ,
increases the isoelectric point of the native molecule and increases formulation stability of
the peptide in acidic media. These structural alterations generate an analogue that
precipitates in the presence of zinc at the site of subcutaneous injection, and then slowly and
reproducibly solubilises to produce a steady supply of insulin at target tissues. Insulin detemir
on the other hand, exerts its activity through the presence of a long fatty acid chain joined to
the e-amino group of LysB29- This allows reversible, non-covalent binding to albumin in
serum, which produces a human insulin analogue that displays a flatter and significantly
longer time-action profile when compared to the native molecule. Although many diabetic
patients are now enjoying the convenience of these long-acting analogues, there is still a
considerable desire to improve basal insulin treatments. Unfortunately, both insulin variants
have significant draw-backs, with glargine showing enhanced IGF-I receptor binding affinity,
and detemir possessing reduced in vivo potency. In addition, both these analogues fail to
produce a 24 hour duration of action, and hyperglycemia is still a common occurrence with
these basal treatments. These modifications to the native insulin sequence, and the
corresponding dicarba analogues of insulin are encompassed by the present invention.
Dicarba analogues of rapid, medium and slow acting insulins are all potential
synthetic targets. Replacement of the native cystine framework of insulin, either partially or
completely, with dicarba bridges (either saturated or unsaturated) can be performed in
conjunction with other sequence modifications (such as those described above) which
moderate onset of action. Hence, generation of rapid, medium and slow acting dicarba
analogues of existing therapeutic insulins, including insulin glargine, lispro, insulin apart,
insulin glulisine, isophane insulin suspensions and determir, would be desirable.
Other terminology used in the context of the compounds and components of
the peptides
The term "amino acid" as used herein is used in its broadest sense and refers to Land
D- amino acids including the 20 common amino acids such as alanine, arginine,
asparagine, aspartic acid, cysteine, glutamic acid, glutamine, glycine, histidine, isoleucine,
leucine, lysine, methionine, phenylalanine, proline, serine, threonine, tryptophan, tyrosine
and valine; and the less common amino acid derivatives such as homo-amino acids (e.g. b-
amino acids), W-alkyl amino acids, dehydroamino acids, aromatic amino acids and a,a-
disubstituted amino acids, for example, cystine, 5-hydroxylysine, 4-hydroxyproline, aaminoadipic
acid, o-amino-n-butyric acid, 3,4 -dihydroxyphenylalanine, homoserine, amethylserine,
ornithine, pipecolic acid, ortho, meta or para-aminobenzoic acid, citrulline,
canavanine, norleucine, d-glutamic acid, aminobutyric acid, L-fluorenylalanine, L-3-
benzothienylalanine and thyroxine; b-amino acids (as compared with the typical a-amino
acids) and any amino acid having a molecular weight less than about 500. The term amino
acid can also include non-natural amino acids such as those described in U.S. Pat. No.
6,559, 126, which are incorporated herein by reference. The term also encompasses amino
acids in which the side chain of the amino acid comprises a metathesisable group, as
described herein. Further, the amino acid may be a pseudoproline residue (YRGO) .
The term "side chain" as used herein is used in the usual sense and refers to the
side chain on the amino acid, and the backbone to the H2N-(C) -C02H (where x = 1, 2 or 3)
component, in which the carbon in bold text bears the side chain (the side chain being
possibly linked to the amino nitrogen, as in the case of proline).
The amino acids may be optionally protected. The term "optionally protected" is
used herein in its broadest sense and refers to an introduced functionality which renders a
particular functional group, such as a hydroxyl, amino, carbonyl or carboxyl group, unreactive
under selected conditions and which may later be optionally removed to unmask the
functional group. A protected amino acid is one in which the reactive substituents of the
amino acid, the amino group, carboxyl group or side chain of the amino acid are protected.
Suitable protecting groups are known in the art and include those disclosed in Greene, T.W.,
"Protective Groups in Organic Synthesis" John Wiley & Sons, New York 1999 (the contents
of which are incorporated herein by reference) as are methods for their installation and
removal.
The amino group of the amino acid may be optionally protected. Preferably the Nprotecting
group is a carbamate such as, 9-fluorenylmethyl carbamate (Fmoc), 2,2,2-
trichloroethyl carbamate (Troc), f-butyl carbamate (Boc), allyl carbamate (Alloc), 2-
trimethylsilylethyl (Teoc) and benzyl carbamate (Cbz), more preferably Fmoc.
The carboxyl group of the amino acid may be optionally protected. The carboxyl
protecting group is preferably an ester such as an alkyl ester, for example, methyl ester, ethyl
ester, f-Bu ester or a benzyl ester.
The side chain of the amino acid may be optionally protected. For example, the
carboxyl groups of aspartic acid, glutamic acid and a-aminoadipic acid may be esterified (for
example as a C C6 alkyl ester), the amino groups of lysine, ornithine and 5-hydroxylysine,
may be converted to carbamates (for example as a C(=0)OC -C alkyl or C(=0)OCH 2Ar
aromatic carbamates) or imides such as pthalimide or succinimide, the hydroxyl groups of 5-
hydroxylysine, 4-hydroxyproline, serine, threonine, tyrosine, 3,4-dihydroxyphenylalanine,
homoserine, a-methylserine and thyroxine may be converted to ethers (for example a C
alkyl or a (CrC 6 alkyl)arylether) or esters (for example a C(=0)C C6 alkyl ester) and the thiol
group of cysteine may be converted to thioethers (for example a C C alkyl thioether) or
thioesters (for example a C(=0)CrC 6 alkyl thioester).
The term "peptide" as used herein refers to any sequence of two or more amino
acids, regardless of length, post-translation modification, or function. "Polypeptide", "peptide"
and "protein" are used interchangeably herein. The peptides or mimetics thereof of the
invention are typically, though not universally, of any length and therefore includes all
truncations of the complete amino acid sequence of insulin.
The term "polypeptide" as used herein refers to an oligopeptide, peptide or protein.
Where "polypeptide" is recited herein to refer to an amino acid sequence of a naturallyoccurring
protein molecule, "polypeptide" and like terms are not intended to limit the amino
acid sequence to the complete, native amino acid sequence associated with insulin, but
instead is intended to also encompass biologically active variants or fragments, including
polypeptides having substantial sequence similarity or sequence identity relative to the amino
acid sequences provided herein.
The dicarba analogues of insulin can be described as peptidomimetics - that is, a
peptide that has a series of amino acids that mimics identically or closely, a naturally
occurring peptide, but with the inclusion of at least one dicarba bridge. For example, the
dicarba bridge may replace a naturally occurring cystine disulfide bridge, a salt bridge or a
non-covalent interaction (such as a hydrogen bond, an ionic bond, van der Waals forces or
hydrophobic interactions), which may be involved in formation of the 3-dimensional structural
features of insulin. The dicarba analogues of insulin may also optionally include one or more
further differences, such as the removal of a cystine bridge, a change by up to 20% of the
amino acids in the sequence, a modification to the amino acid sequence to remove a
protease cleavage site, a modification to the amino acid sequence to disrupt the selfassociation
between monomers of the dicarba insulin analogue, or a modification to the
amino acid sequence to alter the solubility profile of the dicarba insulin analogue, as nonlimiting
examples. These may also be classified as pseudo-peptides. As noted above, of
particular interest are dicarba analogues of insulin, in which one or more of the disulfide
bonds are replaced with dicarba bridges.
The dicarba analogues of insulin may be in the form of the free peptides, or in the
form of a salt, solvate, derivative, isomer or tautomer thereof.
The salts of the peptides are preferably pharmaceutically acceptable, but it will be
appreciated that non-pharmaceutically acceptable salts also fall within the scope of the
present invention, since these are useful as intermediates in the preparation of
pharmaceutically acceptable salts. Examples of pharmaceutically acceptable salts include
salts of pharmaceutically acceptable cations such as sodium , potassium, lithium , calcium,
magnesium, ammonium and alkylammonium; acid addition salts of pharmaceutically
acceptable inorganic acids such as hydrochloric, orthophosphoric, sulphuric, phosphoric,
nitric, carbonic, boric, sulfamic and hydrobromic acids; salts of pharmaceutically acceptable
organic acids such as acetic, propionic, butyric, tartaric, maleic, hydroxymaleic, fumaric,
citric, lactic, mucic, gluconic, benzoic, succinic, oxalic, phenylacetic, methanesulphonic,
trihalomethanesulphonic, toluenesulphonic, benzenesulphonic, salicylic, sulphanilic, aspartic,
glutamic, edetic, stearic, palmitic, oleic, lauric, pantothenic, tannic, ascorbic and valeric acids;
or salts or complexes with pharmaceutically acceptable metal ions, including non-toxic alkali
metal salts such as sodium and potassium salts, or non-toxic transition metal complexes
such as zinc.
In addition, some of the peptides may form solvates with water or common organic
solvents. Such solvates are encompassed within the scope of the invention.
The term "derivative" as used herein refers to any salt, hydrate, protected form,
ester, amide, active metabolite, analogue, residue or any other compound which is not
biologically or otherwise undesirable and induces the desired pharmacological and/or
physiological effect. Preferably the derivative is pharmaceutically acceptable. The term
derivative does not encompass the natural insulin.
The term "tautomer" as used herein is used in its broadest sense and includes
dicarba analogues of insulin which are capable of existing in a state of equilibrium between
two isomeric forms. Such compounds may differ in the bond connecting two atoms or
groups and the position of these atoms or groups in the compound.
The term "isomer" as used herein is used in its broadest sense and includes
structural isomers, geometric isomers and stereoisomers. As the dicarba analogues of
insulin that may be synthesised may have one or more stereogenic centres, they are capable
of existing in enantiomeric forms.
The dicarba analogue of insulin may be synthesised, used or both synthesised and
used as a purified enantiomer, as an enriched enantiomer or diastereomer, or as a mixture of
any ratio of stereoisomers. Where the dicarba bridge of the dicarba analogues of insulin is
an alkene-containing dicarba bridge the alkene-containing group of the bridge may be
present as a mixture of any ratio of geometric isomers (e.g. as an £- or Z-configured alkene),
or as an enriched geometric isomer. It is however preferred that where the dicarba analogue
of insulin is present as a mixture of stereoisomers, the mixture is enriched in the preferred
isomer.
In its broadest sense "enriched" means that the mixture contains more of the
preferred isomer than of the other isomer. Preferably, an enriched mixture comprises greater
than 50% of the preferred isomer, where the preferred isomer gives the desired level of
potency and selectivity. More preferably, an enriched mixture comprises at least 60%, 70%,
80%, 90%, 95%, 97.5% or 99% of the preferred isomer. The dicarba analogue of insulin
which is enriched in the preferred isomer can either be obtained via a stereospecific reaction,
stereoselective reaction, isomeric enrichment via separation processes, or a combination of
all three approaches.
In this specification, including the claims, except where the context requires
otherwise due to express language or necessary implication, the word "comprising" or
variations such as "comprise" or "comprises" is used in the inclusive sense, to specify the
presence of the stated features or steps but not to preclude the presence or addition of
further features or steps.
It must be noted that, as used in the specification, the singular forms "a", "an" and
"the" include plural aspects unless the context clearly dictates otherwise. Thus, for example,
reference to "a dicarba analogue of insulin" or "a dicarba bridge" includes a single dicarba
analogue of insulin or a single dicarba bridge, as well as two or more dicarba analogues of
insulin or two or more dicarba bridges, respectively; and so forth.
Dicarba bridges
Replacement of the disulfide bridge in insulin with biostable structural mimics may
lead to enhanced chemical stability and a greater potential to treat diabetes. For this to be
feasible, however, the role of each disulfide bond in the structure and activity of insulin must
be determined. In some cystine containing peptides, the bridges have a functional purpose
and represent segments of a binding domain or active site, which release chelating metal
ions upon reduction or undergo disulfide exchange. In many other disulfide-containing
peptides the role of the disulfide bridge is purely structural, being present to maintain
secondary and tertiary structure.
The dicarba bridge may form a bridge between the two separate peptide chains of
insulin (e.g. the A-chain and the B-chain of insulin), to form an interchain dicarba bridge, or it
may form a bridge between two points in a single peptide chain (for example as found in
insulin's A-chain, or a fragment thereof), so as to form an intrachain dicarba bridge,
otherwise known as a ring. Preferably, the dicarba analogue of insulin at least contains an
intrachain dicarba bridge.
In some instances it may be difficult to form intrachain dicarba bridges due to steric
hindrance, aggregation and/or the need to bring the readable (metathesisable) groups
together. We have found that the use of one or more of (i) microwave radiation in the crossmetathesis
step, (ii) turn-inducing groups, and/or (iii) alternating solid phase peptide
synthesis (SPPS) and catalysis enables such dicarba bridges to be formed within insulin
analogues, often in an efficient manner. This is discussed below.
In some instances it may be difficult to form interchain dicarba bridges due to the
peptide sequence and the positioning of the reacting motifs. We have also found that the
use of a removeable tether between two amino acids or peptides to be connected may
enhance the metathesis. This is discussed below.
The term "dicarba bridge" as used herein is used broadly, unless the context
indicates otherwise, to refer to a bridging group that includes at least one of the groups
selected from -C-C-, -C=C- and - CºC- This means that the dicarba bridge could be
wholly or partly composed of the groups -C-C-, -C=C- and - CºC- or could for example be
one of the dicarba bridges shown in formula (I) to (VI) below. In a preferred embodiment, the
atoms directly attached to the carbon atoms of the dicarba bridge are C or H. Further or
alternative reactions may be performed to introduce substituents other than hydrogen onto
the carbon atoms of the dicarba sequence of the dicarba bridge.
Preferably, the dicarba analogue of insulin contains at least one unsaturated dicarba
bridge. The term "unsaturated dicarba bridge" as used herein refers to dicarba bridges which
contain the group -C=C- (an alkene-containing dicarba bridge) or dicarba bridges which
contain the group - CºC- (an alkyne-containing dicarba bridge). The term "unsaturated
hydrogen dicarba bridge" as used herein refers to dicarba bridges which contain the group -
CH=CH-.
The term "alkyne-containing dicarba bridge" as used herein is used broadly, unless
the context indicates otherwise, to refer to a bridging group that includes at least an alkyne
group (-C ºC-). This means that the alkyne-containing dicarba bridge could be wholly or
partly composed of the group - CºC- or could for example be one of the dicarba bridges
shown in formula (I) or ( I I) below.
In addition to the alkyne group, the alkyne-containing dicarba bridge may include
any other series of atoms, typically selected from C, N, O, S and P. The atoms directly
attached to the carbon atoms of the alkyne-containing dicarba bridge are preferably carbon.
However, any of the other atoms listed above may also be present, with the proviso that the
nitrogen atoms present in the compound during metathesis are not free amines (protected
amines, such as carbamates and salts, are acceptable). The alkyne-containing dicarba
bridge encompasses the following possible bridges, as illustrative examples:
wherein R to R6 are each independently absent or selected from a divalent linking group. R
to R6 may be the same (for example R = R2) or different (for example R ¹ R2) , or one or
more or all of R to R may be absent. Such divalent linking groups should not be groups
that poison the metathesis catalyst. Preferably, the divalent linking groups R to R are
substituted or unsubstituted alkylene or substituted or unsubstituted alkoxyl groups.
The term "alkylene" as used herein refers to divalent alkyl groups including straight
chain and branched alkylene groups having from 1 to about 20 carbon atoms. Typically, the
alkylene groups have from 1 to 15 carbons or, in some embodiments, from 1 to 8, 1 to 6 , or 1
to 4 carbon atoms. Examples of straight chain alkylene groups include methyl, ethyl, npropyl,
r?-butyl, n-pentyl, n-hexyl, n-heptyl, and n-octyl groups. Examples of branched alkyl
groups include, but are not limited to, /so-propyl, /so-butyl, sec-butyl, ferf-butyl, /ieo-pentyl,
/so-pentyl, and 2,2-dimethylpropyl groups. The alkylene groups may also be substituted and
may include one or more substituents.
The term "alkoxyl" as used herein refers to the divalent group -OR- where R is an
alkylene group as defined above. Examples of straight chain alkoxy groups include
methoxyl, ethoxyl, propoxyl and longer chain variants. Examples of branched alkoxyl groups
include, but are not limited to, a-methylmethyoxyl and a-methylethoxyl groups. The alkoxyl
groups may also be substituted and may include one or more substituents, which are as
defined below.
A "substituted" alkylene or alkoxyl group has one or more of its hydrogen atoms
replaced by non-hydrogen or non-carbon atoms. Substituted groups also include groups in
which one or more bonds to a carbon(s) or hydrogen(s) atom are replaced by one or more
bonds, including double or triple bonds which may optionally be blocked via an adjacent
heteroatom . Thus, a substituted group will be substituted with one or more substituents,
unless otherwise specified. In some embodiments, a substituted group is substituted (in
protected or unprotected form) with 1, 2 , 3 , 4 , 5 , or 6 substituents. Examples of substituent
groups include halogens (i.e. , F, CI, Br and I), hydroxyls, alkoxyl, alkenoxyl, alkynoxyl,
aryloxyl, aralkyloxyl, heterocyclyloxyl, and heterocyclylalkoxyl groups; carbonyls (oxo);
carboxyls; esters; ethers, urethanes; oximes; hydroxylamines; alkoxyamines;
aralkoxyamines; thiols; sufides; sulfoxides; sulfones; sulfonyls; sulfonamides; amines; Noxides;
hydrazines; hydrazides; hydrazones; azides; amides; ureas; amidines; guanidines;
enam ines; imides; isocyanates; isothiocyanates; cyanates; thiocyanates; imines; nitriles (i.e.
CN); and the like. Such substituents should not be groups that poison the metathesis
catalyst or affect its selectivity.
The term "alkene-containing dicarba bridge" as used herein is used broadly, unless
the context indicates otherwise, to refer to a bridging group that includes at least an
unsaturated alkene (-C=C-). This means that the dicarba bridge could be wholly or partly
composed of the group -C=C- or could for example be one of the dicarba bridges shown in
formula ( III) or (IV) below. The alkene-containing dicarba bridge (-C=C-) may possess cisor
trans-geometry.
In a preferred embodiment, the atoms directly attached to the carbon atoms of the
alkene-containing dicarba bridge are typically H or C. Further or alternative reactions can be
performed to introduce substituents other than hydrogen onto the carbon atoms of the
dicarba sequence or at other positions on the dicarba bridge.
In addition to the -C=C- dicarba sequence, the dicarba bridge may include any
other series of atoms, typically selected from C, N, O, S and P, although the atoms to either
side of the dicarba sequence are preferably carbon, and with the proviso that the nitrogen
atoms present in the compound during metathesis are not free amines protected amines,
such as carbamates, and salts are acceptable). Thus, the dicarba bridge encompasses the
following possible bridges, as illustrative examples:
wherein R to R6 are each independently absent or selected from a divalent linking group. R
to R6 may be the same (for example R = R2) or different (for example R ¹ R2) , or one or
more or all of R to R may be absent. Such divalent linking groups should not be groups
that poison the metathesis catalyst. Preferably, the divalent linking groups R and R2 are
substituted or unsubstituted alkylene or substituted or unsubstituted alkoxyl group, as defined
above.
The term "saturated dicarba bridge" or "alkane-containing dicarba bridge" is used
broadly, unless the context indicates otherwise, to refer to a bridging group that includes at
least a saturated alkane containing dicarba bridge (-C-C-). This means that the dicarba
bridge could be wholly or partly composed of the groups -C-C-, or could for example be one
of the dicarba bridges shown in formula (V) or (VI) below.
The atoms directly attached to the carbon atoms of the saturated dicarba bridge are
typically H or C, although further or alternative reactions can be performed to introduce
substituents other than hydrogen onto the carbon atoms of the dicarba sequence of the
dicarba bridge. The term "saturated hydrogen dicarba bridge" refers to the specific case
where the dicarba bridge is -CH 2-CH 2- . Typically, saturated dicarba bridges are prepared by
hydrogenation or other reduction of unsaturated dicarba bridges.
In addition to the dicarba sequence, the dicarba bridge may include any other series
of atoms, typically selected from C, N, O, S and P, although the atoms to either side of the
dicarba sequence are preferably carbon, and with the proviso that the nitrogen atoms
present in the compound during metathesis are not free amines (protected amines, such as
carbamates, and salts are acceptable). Thus, the dicarba bridge encompasses the following
possible bridges, as illust
wherein R-i to are each independently absent or selected from a divalent linking group. Rto
R6 may be the same (for example = R2) or different (for example R ¹ R2) or one or
more or all of R to R may be absent. Such divalent linking groups should not be groups that
poison the metathesis catalyst. Preferably, the divalent linking groups R and R2 are
substituted or unsubstituted alkylene or substituted or unsubstituted alkoxyl group, as defined
above.
Where the terms "alkyne-containing", "alkene-containing" or "alkane-containing" or
"saturated" are not specified, the term "dicarba bridge" is taken to refer to a bridging group
that includes at least one of the groups selected from a saturated dicarba bridge (-C-C-), an
unsaturated alkene-containing dicarba bridge (-C=C-) and an alkyne-containing dicarba
bridge (-C ºC-) as described above. This means that the dicarba bridge could be wholly or
partly composed of the groups -C-C-, -C=C- or - CºC- or could for example be any one of
the dicarba bridges shown in formula (I) to (VI) above.
Methods for the preparation of dicarba insulin analogues
Metathesis
Metathesis is a powerful synthetic tool that enables the synthesis of carbon-carbon
bonds via a transition metal-catalysed transformation of alkyl-unsaturated reactants. The
construction of dicarba analogues of m / /-cystine containing peptides, however, presents
more of a synthetic challenge.
The use of uniform metathesis substrates leads to a statistical product distribution
and therefore metathesis selectivity is severely compromised. For example,
homodimerisation of equivalent olefins A and B in the absence of selectivity results in a
statistical mixture of three products (as shown below). The yield of desired products (A-A
and B-B) is not more than 50% in the absence of selection. In order to exclusively form the
target A-B product, selective metathesis strategies must be employed to avoid the formation
of the A-A and B-B homodimers.
J cross metathesis 25% 25% 50%
+ / — \ +
A A B B A B
Cross-metathesis (CM) is a type of metathesis reaction involving the formation of a
new bond across two unblocked, reactive metathesisable groups, to form a new bridge
between the two reactive metathesisable groups. For example, cross-metathesis can be
used to generate a dicarba analogue of insulin having an intermolecular bridge via formation
of a dicarba bridge between two reactive metathesisable groups each located in different
chains of insulin.
Ring-closing metathesis (RCM) is a type of metathesis reaction where the two
reactive metathesisable groups are located within one peptide chain so as to form an
intramolecular bridge, or ring. For example, ring-closing metathesis involves the formation of
a dicarba bridge between two reactable metathesisable groups located on a single chain of
insulin to produce a dicarba analogue of insulin having an intrachain bridge.
The dicarba analogues of insulin or fragments, salts, solvates, derivatives, isomers
or tautomers as described above, may be synthesised by a method comprising:
(i) providing the A-chain having at least one pair of complementary metathesisable
groups;
(ii) subjecting the A-chain to metathesis to form at least one dicarba bridge between the
pair of complementary metathesisable groups; and
(iii) adding the B-chain.
In some instances, the dicarba analogue of insulin may need to assume a particular
conformation in order to serve as a suitable peptidomimetic of insulin. In particular, when the
dicarba bridge replaces one or more of the disulfide bonds present in insulin, it may be
advantageous for the dicarba bridge to adopt a particular geometry. The ability of the
dicarba bridge to mimic the conformation of the disulfide bridge formed in the active form of
insulin may effect the activity of the dicarba analogue of insulin compared to the activity of
the native form of insulin.
The initial product of the metathesis reaction is a compound with an unsaturated
dicarba bridge (-C ºC- or -C=C-). That is an alkyne-containing dicarba bridge formed by
alkyne metathesis or an alkene-containing dicarba bridge formed by alkene metathesis.
If the target insulin analogue is to contain an alkene-containing dicarba bridge (-
C=C-) or a saturated alkane-containing dicarba bridge (-C-C-), the process may further
comprise the step of subjecting the unsaturated dicarba bridge to hydrogenation.
In the preparation of the dicarba analogues of insulin it is preferred that at least one
chain of insulin is provided on a solid support. The types of solid supports that may be used
are described below. Tethering a peptide sequence to a solid support can also promote
RCM: A pseudo-dilution effect operates on resin to promote RCM over otherwise competing
CM reaction. Hence high dilution is not required for the promotion of RCM conversion.
The preparation of the dicarba analogues of the present invention provides a
number of advantages when at least one chain of insulin is provided on a solid support. The
combination of SPPS and catalysis using a single solid support is highly efficient.
Homogeneous catalysts, such as those used to affect metathesis and hydrogenation, can be
exposed to a resin bound peptide and simply separated from the product peptide via filtration
of the resin-peptide from the reaction solution. This eliminates and/or minimises metalcontamination
of the product and aids the separation of the product peptide from solution
phase by-products and/or impurities. Protecting groups for reactive sidechains which are
commonly employed in SPPS protocols are also tolerated by organotransition metal catalysts
and hence catalysis can conveniently be performed immediately after SPPS.
The dicarba analogues of insulin or fragments, salts, solvates, derivatives, isomers
or tautomers as described above, may be synthesised by a method comprising:
(i) providing a part of the A-chain having at least two complementary metathesisable
groups;
(ii) subjecting the A-chain to metathesis to form at least one dicarba bridge; and
(iii) adding one or more further amino acids to one or both ends of the A-chain; and
(iv) adding the B-chain.
The dicarba analogues of insulin or fragments, salts, solvates, derivatives, isomers
or tautomers as described above, may also be synthesised by a method comprising:
(i) providing a part of the A-chain and/or a part of the B-chain having at least two
complementary metathesisable groups between them ;
(ii) subjecting the A-chain and B-chain to metathesis to form at least one dicarba
bridge; and
(iii) adding one or more further amino acids to one or both ends of the A-chain and/or Bchain.
Additionally, alternating SPPS-Catalysis-SPPS methods can be used to introduce
metathesisable groups into a sequence, stepwise. Such a technique removes and/or
eliminates the need for orthogonal protecting group strategies. The technique also assists
the catalysis of 'difficult' sequences (e.g. residues which promote deleterious aggregation,
unfavourable conformations, and poor peptide solubility and reagent penetration) by allowing
troublesome sections of the sequence to be omitted until after the scheduled catalysis has
been performed. The omitted residues are added to the truncated sequence via SPPS.
The dicarba analogues of insulin or fragments, salts, solvates, derivatives, isomers
or tautomers as described above, may be synthesised by a method comprising:
(i) providing or synthesising a reactable peptide comprising a removable tether
between the A-chain and the B-chain, the A-chain and the B-chain each having at
least one complementary metathesisable group; and
(ii) subjecting the reactable peptide to metathesis to form at least one dicarba bridge
between the complementary metathesisable groups; and
(ii) removing the removeable tether to produce a dicarba bridge linking the A-chain and
the B-chain of insulin.
This approach enhances metathesis to form interchain dicarba bridges, which may
be difficult due to the nature of the peptide sequence and the positioning of the reacting
motifs. This approach provides an alternative to performing direct cross metathesis, and can
assist in formation of an interchain dicarba bridge. The approach forms an intramolecular
dicarba bridge between two amino acids or peptides to be joined, which are provided in a
contiguous sequence, connected by a removeable tether. Once the dicarba bridge has been
formed, the removeable tether is removed to produce the interchain dicarba bridge.
During the step of metathesis may be performed with the peptide attached to a
resin. Preferably, when the peptide substrate is attached to a resin, the metathesis step uses
a homogeneous catalyst.
Alkyne metathesis
Alkyne metathesis can be used to install one or more alkyne-containing dicarba
bridge to form a dicarba analogue of insulin. The alkyne containing dicarba bridges may be
intramolecular or intermolecular.
Alkyne metathesis involves the formation of a new alkyne bond from two unblocked
or reactive alkynes. The new alkynyl bridge covalently joins the two reactive starting
alkynes. As shown below, the alkyne-containing dicarba bridge may be formed between two
reactable metathesisable groups provided within the scaffold of insulin. The two reactable
metathesisable groups may each be residues of separate peptides, to form an intermolecular
bridge (a cross metathesis reaction). Cross-metathesis occurs when the alkyne-containing
dicarba bridge is formed between two or more peptides having between them a pair of
complementary alkyne-containing metathesisable groups. In this case, the alkyne-containing
dicarba bridge is an intermolecular dicarba bridge (shown as (A) below). Alternatively, ring
closing metathesis occurs when the alkyne-containing dicarba bridge is formed between two
amino acids within a single readable peptide. In this case, the alkyne-containing dicarba
bridge is an intramolecular dicarba bridge (shown as B below).
(A) Alkyne Cross Metathesis (CAM)
reactive amino acid
reactive amino acid
intermolecular
dicarba bridge
(B) Ring Closing Alkyne Metathesis (RCAM)
reactive alkynes
intramolecular dicarba
reactive peptide bridge
In some sequences, it can be difficult to form intramolecular bridges due to
aggregation and deleterious hydrogen bonding, and the need to bring the reactable
(metathesisable) groups together. We have found that the use of microwave radiation, turninducing
groups in the metathesis step and/or alternating solid phase peptide synthesis and
catalysis strategies (as discussed below), facilitates the metathesis reaction to occur, or
occur more efficiently.
In one embodiment of the present invention, the dicarba analogue contains an
alkyne-containing dicarba bridge. This alkyne-containing dicarba bridge may subsequently
be subjected to stereoselective reduction to preferentially generate either the cis- or the trans
isomer of the alkene-containing dicarba bridge. This approach produces a dicarba analogue
of insulin which is enriched in one of the geometric isomers of the alkene-containing dicarba
bridge. In this manner, the most/only active isomer may be targeted. Advantageously, the
need for time-consuming chromatographic separation of the unwanted geometric isomer can
be avoided using this approach.
Alkyne metathesis is facilitated by a catalyst. Some metal-complexes are highly
active alkyne metathesis catalysts and some are capable of highly selective alkyne
metathesis in the presence of other potentially reactable groups, e.g. alkenes.
Catalysts which may be used to perform alkyne metathesis in the method of the
present invention are those catalysts which are selective for alkyne-containing
metathesisable groups, while not interfering with the functional groups present in the amino
acids and peptides between which the alkyne-containing dicarba bridge is formed. Examples
of suitable catalysts include those described in Fijrstner, A.; Davies, P. W. Chem. Commun.
2005, 2307-2320; Zhang, W.; Moore, Jeffrey S. Adv. Synth. Catal. 2007, 349, 93-1 20; Grela,
K., Ignatowska, J. , Organic Letters, 1992, 4(21 ) , 3747; and Mortreux, A.; Coutelier, O. J.
Mol. Catal. A: Chem. 2006, 254, 96-1 04, incorporated herein by reference. There are many
catalysts available to achieve this transformation, which vary in their ability to catalyse the
metathesis reaction, in their ability to tolerate other functional groups, and their stability
towards water and other functional groups. Preferably the catalyst used for alkyne
metathesis is a homogenous catalyst. More preferably, the catalyst is a tungsten containing
catalyst, or a molybdenum-containing catalyst such as those based on W(IV) and
Mo(CO)6/phenol systems. Still more preferably, the catalyst is a tungsten-containing
catalyst. An example of a suitable tungsten-containing catalysts is tris(ferf-butoxy)(2,2-
dimethylpropylidyne)tungsten.
A preferred tungsten-containing alkyne-metathesis catalyst is a tungsten-alkylidyne
complex commonly known as Schrock's catalyst, tris(ferf-butoxy)(2,2-dimethylpropylidyne)
tungsten(VI), which is shown below. This catalyst is a highly air and moisture sensitive
molecule which necessitates the rigorous use of inert and anhydrous reaction atmosphere
and solvents respectively. It is however, highly tolerant of a wide range of functionality which
is often found in peptide sequences.
_
o-w —o Schrock's catalyst
Schrock's catalyst
The formation of an alkyne-containing dicarba bridge involves the use of
complementary pairs of alkyne-containing metathesisable groups which may be connected to
an amino acid in a peptide of insulin. A metathesisable group is a functional group that can
undergo metathesis when unblocked or in an activated state. The alkyne containing
metathesisable group may be connected to an amino acid via the amino acid side chain or
via the amino group of the amino acid. As an example, a side chain of the amino acid may
include at least an alkyne-containing metathesisable group, and the side chain may be
wholly or partly composed of the group - CºCThe
term "alkyne-containing metathesisable group" as used herein is used broadly,
unless the context indicates otherwise, to refer to a group that includes at least an alkyne
moiety. The alkyne-containing metathesisable group could for example be an alkynecontaining
metathesisable group of the general formula drawn below:
The integer n may be 0 , 1, 2 , 3 , 4 , 5 , 6 , 7, 8 , 9 or 10 . The R7 and R8 groups should
not be a group that poisons the metathesis catalyst. Preferably, the R7 group is substituted
or unsubstituted alkyl. The R8 group is either H or substituted or unsubstituted alkyl.
Preferably, the R8 group either H or methyl.
The term "alkyl" as used herein refers to a monovalent alkyl group including straight
chain and branched alkyl groups having from 1 to about 20 carbon atoms. Typically, the
alkyl group has from 1 to 15 carbons or, in some embodiments, from 1 to 8 , 1 to 6 , or 1 to 4
carbon atoms. Examples of straight chain alkyl groups include methyl, ethyl, n-propyl, nbutyl,
n-pentyl, n-hexyl, n-heptyl, and n-octyl groups. Examples of branched alkyl groups
include, but are not limited to, /so-propyl, /so-butyl, sec-butyl, ferf-butyl, neo-pentyl, isopentyl,
and 2,2-dimethylpropyl groups. The alkyl group may also be substituted and may
include one or more substituents. The term "substituted" is as defined above in relation to
alkylene groups.
The alkyne-containing metathesisable group may be connected to an amino acid of
insulin. The metathesisable group is preferably located on the amino group or on the side
chain of the amino acid.
During the metathesis reaction, a by-product is produced, which comprises an
alkyne bond which is substituted with the R group. Preferably, the R7 groups are such that
the resulting by-product is gaseous, and is eliminated from the reaction mixture. For
example, when R is methyl, the by-product is 2-butyne, which evaporates from the reaction
mixture to leave the reaction product. Other alkyne by-products, such as 2-pentyne and 3-
hexyne (having boiling poimts of 56°C and 8 1°C respectively), could also be generated from
the combination of butynylglycine and pentynylglycine residues. These low boiling point
liquids are readily removed from the metathesis reaction mixture. It will however be
appreciated that techniques for the separation of a non-gaseous by-product from the reaction
mixture would also be known by a person skilled in the art.
It is noted that a pair of complementary alkyne-containing metathesisable groups
need not be identical. For example, an alkyne-containing metathesisable group in which R7
is methyl can react with an alkyne-containing metathesisable group in which R7 is ethyl to
form an alkyne-containing dicarba bridge. The term "complementary" is used to indicate that
the pair of unblocked alkyne-containing metathesisable groups are not necessarily identical,
but are merely complementary in the sense that metathesis can take place between the two
alkyne-containing groups.
Alkene Metathesis
Alkene metathesis provides a versatile method for the cleavage and formation of C=C bonds,
and involves a mutual intermolecular exchange of alkylidene fragments between two alkene
groups.
In alkene metathesis reactions, the redistribution can result in three main outcomes
shown below: (A) ring-opening metathesis (ROM) which is sometimes followed by
polymerization of the diene (ROMP); (B) ring-closing metathesis (RCM); and (C) cross
metathesis. Of particular interest to the present invention are the latter two. ADMET, acyclic
diene metathesis, is also an important process.
In the synthesis of a peptide having an alkyne-containing dicarba bridge and at least
one additional alkene-containing dicarba bridge, the formation of each bridge may occur in
any order. For example, alkyne metathesis to form the alkyne-containing dicarba bridge may
occur before or after formation of any alkene-containing dicarba bridges.
When a pair of alkene-containing metathesisable groups are incorporated into the
primary sequence of a single peptide and subjected to metathesis conditions, an
intramolecular reaction will result in the formation of a cyclic peptide (RCM). If however, the
pair of alkene-containing metathesisable groups are present within two separate peptide
chains, an intermolecular CM reaction will result in the formation of a link between the two
peptides. This is shown below.
(A) Alkene Cross Metathesis (CM)
reactive amino acid
intermolecular
reactive amino acid dicarba bridge
(B) Ring Closing Alkene Metathesis (RCM)
reactive alkenes
intramolecular dicarba
reactive peptide bridge
In some peptide sequences, such as insulin, it may be difficult to form intramolecular
bridges, due to aggregation and deleterious hydrogen bonding, and the need to bring the
reactable (metathesisable) groups together. The use of microwave radiation, turn-inducing
groups and/or an alternating SPPS-catalysis strategy to promote metathesis conversion (as
discussed below) facilitates dicarba bridge formation.
In some peptide sequences, such as insulin, it may be difficult to form interchain
dicarba bridges due to the nature of the peptide sequence and the positioning of the reacting
motifs. The use of a removeable tether between two amino acids or peptides to be
connected may enhance the metathesis. This is discussed below.
In one embodiment of the present invention, the alkene-containing dicarba bridge
that is formed can subsequently be subjected to reduction to generate the corresponding
saturated alkane-containing dicarba bridge.
Alkene metathesis is also facilitated by a catalyst. Catalysts which may be used to
perform alkene metathesis in the method of the present invention are those catalysts which
are selective for the alkene-containing metathesisable groups, while not interfering with the
functional groups present in the amino acids and peptides between which the alkenecontaining
dicarba bridge is formed. Examples of suitable catalysts include those described
in Grubbs, R.H., Vougioukalakis, G.C. C em. Rev., 201 0 , 110, 1746-1 787, Tiede, S., Berger,
A., Schlesiger, D., Rost, D., Liihl, A., Blechert S., Angew. Chem. Int. Ed., 201 0 , 49, 1-5, and
Samojlowicz, C , Bieniek, M., Grela, K. Chem. Rev., 2009, 109, 3708-3742. Preferably, the
catalyst used for alkene metathesis is a homogeneous catalyst, such as a ruthenium-based
alkylidene catalyst.
Many alkene metathesis catalysts are now commercially available or easily
synthesised in the laboratory. While early catalysts were poorly defined, lacked functional
group tolerance and were highly moisture and oxygen sensitive, later generation catalysts
have largely overcome these initial problems. Currently used Ru-based catalysts, for
example Grubbs' first and second generation catalysts, and the Hoveyda-Grubbs analogues,
are robust, display high functional group tolerance and have tuneable reactivity under mild
experimental conditions. Despite their differing substitution around the core Ru centre, all of
the catalysts cycle through an active ruthenium alkylidene species. The variation around the
reactive core however, plays an important role in mediating initiation, propagation and
substrate specificity.
Where a dicarba analogue of insulin is to contain an alkene-containing dicarba
bridge and an alkyne-containing dicarba bridge, it is useful to tailor the catalyst used in the
metathesis reaction to the substrate in order to achieve regioselective dicarba bridge
formation. For example, on exposure to second generation Grubbs' catalyst (a ruthenium
alkylidene catalyst bearing phosphine, W-heterocyclic carbene and chloride ligands), a
peptide sequence possessing an alkyne and alkene functional group will undergo en-yne
metathesis. The same peptide however, exposed to first generation Grubbs' catalyst (a
ruthenium alkylidene catalyst bearing only phosphine and chloride ligands), will only undergo
alkene cross metathesis.
Preferably, the catalyst is a metal carbene complex such as those shown below.
More preferably, where a peptide is to contain at least one alkene-containing dicarba bridge
and at least one alkyne-containing dicarba bridge the catalyst used for alkene metathesis is a
first generation catalyst.
1 generation 2 generation 1 generation 2 generation
Grubbs' catalyst Grubbs' catalyst Hoveyda-Grubbs' Hoveyda-Grubbs'
catalyst catalyst
Cy = cyclohexyl
Mes = mesitylene
The formation of an alkene-containing dicarba bridge involves the use of
complementary pairs of alkene-containing metathesisable groups connected to an amino
acid or connected to an amino acid in a peptide of insulin. A metathesisable group is a
functional group that can undergo metathesis when unblocked or in an activated state. The
alkene-containing metathesisable group may be connected to an amino acid via the amino
acid side chain or via the amino group of the amino acid. As an example, a side chain of the
amino acid may include at least an alkene-containing metathesisable group, and the side
chain may be wholly or partly com posed of the group -C=CThe
term "alkene-containing metathesisable group" as used herein is used broadly,
unless the context indicates otherwise, to refer to a group that includes at least an alkene
moiety. The alkene-containing metathesisable group could, for example, be an alkene
containing metathesisable group of the general formula drawn below:
The integer n may be 0 , 1, 2 , 3, 4 , 5 , 6 , 7, 8 , 9 or 10 . The R , R 0 and groups
should not be a group which poisons the metathesis catalyst. Preferably the R and R 0
groups are each independently H or a substituted or unsubstituted alkyl as defined above.
The R group is either H or a substituted or unsubstituted alkyl. Preferably the R1 group is
either H or methyl.
The alkene-containing metathesisable group may be connected to an amino acid of
insulin. The alkene-containing metathesisable group is preferably located on the amino
group or on the side chain of the amino acid.
The alkene-containing metathesisable group could for example include the alkenecontaining
metathesisable groups:- allylglycine (A), crotylglycine (B), prenylglycine (C) and
the extended acrylate (D) (as drawn below).
The reactivity of alkenes towards homodimerisation during metathesis, has been
categorised into four classes - Type I through IV. Type I alkenes are the most reactive, and
are characterised by sterically unhindered and electron-rich olefins such as allyl- (A) and
crotyl-glycine (B). Increasing steric hindrance and decreasing electron density about the
olefin, results in slow to non-existent homodimerisation, and sees these alkenes categorised
in Types II through IV. These include residues such as prenylglycine (C) and the extended
acrylate (D). These glycine derivatives are shown below.
A B C D
During the metathesis reaction, a by-product is produced. Preferably, the by¬
product is gaseous, and evaporates from the reaction mixture. It will however be appreciated
that techniques for the separation of non-gaseous by-products from the reaction mixture
would also be known by a person skilled in the art.
It is noted that a pair of complementary alkene-containing metathesisable groups
need not be identical. For example, an allylglycine residue can be metathesised with a
crotylglycine residue to generate a new dicarba bridge. The term "complementary" is used to
indicate that the pair of unblocked alkene-containing metathesisable groups are not
necessarily identical, but are merely com plementary in the sense that metathesis can take
place between the two alkene-containing groups.
Microwave reaction conditions
It has been found that when the metathesis reaction is performed under microwave
reaction conditions, the reaction may take place in situations where the reaction would not
otherwise take place - for instance, when the metathesisable groups are unblocked, but the
arrangement, length or spatial orientation of the readable organic compound prevents the
metathesisable groups from being close enough to one another to enable the reaction to
proceed. An alternative strategy is described below (see the description of "Turn-inducing
groups" below).
The microwave reaction conditions involve applying microwave radiation to the
reactable peptide (preferably attached to a solid support) in the presence of the metathesis
catalyst for at least part of the reaction, usually for the duration of the reaction. The
microwave or microwave reactor may be of any type known in the art, operated at any
suitable frequency. Typical frequencies in commercially available microwave reactors are
2.45 GHz, at a power of up to 500W, usually of up to 300W. The temperature of the reaction
is preferably at elevated as a consequence of the microwave radiation. Preferably, the
temperature of the reaction is at reflux, or around 100°C, as appropriate in each case. The
reaction is preferably performed in a period of not more than 5 hours, suitably for up to about
2 hours.
Turn-inducing groups
Another strategy devised for improving the performance of a metathesis (particularly
ring closing metathesis) reaction between two complementary metathesisable groups
(alkenes or alkynes) is the use of turn-inducing groups. This strategy is particularly useful for
ring-closing metathesis where the metathesisable groups are located within a single peptide.
As described above, this strategy can also be used in combination with microwave
irradiation.
According to this embodiment, there is provided a method for the synthesis of a
dicarba analogue of insulin comprising an A-chain and a B-chain or fragments, salts,
solvates, derivatives, isomers or tautomers of the A-chain, the B-chain or both, the method
comprising:
(i) providing the A-chain, the B-chain or a peptide comprising both the A-chain and the
B-chain, the A-chain, the B-chain or the peptide having a pair of unblocked
complementary metathesisable groups, and a turn-inducing group located between
the pair of complementary metathesisable groups, and
(ii) subjecting the peptide to metathesis to form a dicarba analogue of insulin
comprising a dicarba bridge.
This method can be used to facilitate the production of alkyne dicarba bridge
formation and/or alkene dicarba bridge formation. Where the method is performed on a
peptide comprising the A-chain or the B-chain, the method may also comprise the step of
adding the other chain of insulin. Where the method is performed on a peptide comprising
the A-chain and the B-chain of insulin, the peptide may further comprise a removeable tether
and the method may further comprise the step of removing the removable tether from the
peptide, to produce a dicarba insulin analogue containing an interchain dicarba bridge.
In order to facilitate the production of an alkyne dicarba bridge, the at least two
complementary metathesisable groups in step (i) will be complementary alkyne-containing
metathesisable groups. In order to facilitate the production of an alkene dicarba bridge, the
at least two complementary metathesisable groups in step (i) will be complementary alkenecontaining
metathesisable groups.
The peptide backbone in a-peptides is generally linear as the component amino
acids (especially when these are the 20 common amino acids, with the exception of proline)
form frafls-configured peptide bonds. Proline, a pyrrolidine analogue, can induce a turn in an
otherwise linear peptide. This is a naturally-occurring turn-inducing group. This embodiment
is particularly suited to those peptides that do not contain a naturally-occurring turn-inducing
amino acid, such as proline. In this case, a synthetic (non-naturally occurring) turn-inducing
group is located in the amino acid sequence of insulin.
The linear - CºC- geometry of the alkyne metathesisable groups can disfavour
productive metathesis. The new alkyne bond (-C ºC-) that is formed will be linear, and the
alkyne-containing metathesisable groups need to be brought in close proximity in order to
react. Particularly, in ring-closing alkyne metathesis (RCAM), this may require that the
backbone of the peptide curve around to bring the two alkyne-containing metathesisable
groups into such a conformation. In some instances this can be sterically disfavoured and
therefore alkyne metathesis may be relatively low yielding or not occur. Accordingly, it may
be preferable to include a turn-inducing group between a pair of complementary alkynecontaining
metathesisable groups or in a position adjacent to an alkyne-containing
metathesisable group so that the conformation of the peptide backbone can be 'unnaturally'
altered to allow a pair of complementary alkyne-containing metathesisable groups to be
brought into an improved conformation for alkyne metathesis.
Preferably the turn-inducing group is a turn-inducing amino acid, dipeptide or
protein, and is preferably synthetic (non-naturally occurring). Examples of suitable synthetic
turn-inducing amino acids are the pseudoprolines, including derivatives of serine, threonine
and cysteine (shown below). The pseudoprolines have been derivatised to contain a cyclic
group between the amino acid sidechain (via the -OH or -SH group), and the amino nitrogen
atom. A typical derivatising agent is CH3-C(=0)-CH 3, such that the turn-inducing amino acids
are:
R = H, Me
R' = Me (threonine-derived)
R' = H (serine-derived)
These turn-inducing residues are often prepared as dipeptide units to aid
incorporation into peptides. An example of a suitable turn-inducing residue is 5,5-
dimethylproline which is stable and may stay in the peptide permanently. However, after
metathesis, some pseudoproline(s) may be converted back to the underivatised amino acid
(serine, threonine or cysteine) by removal of the derivatising agent usually on treatment with
acid. The conditions for cleavage of the peptide from a solid support will usually achieve this.
If, for example, the turn-inducing amino acid is one of pseudo-serine, pseudothreonine
or pseudo-cysteine, then the method may further comprise the step of converting
the pseudo-serine, pseudo-threonine or pseudo-cysteine to serine, threonine or cysteine,
respectively.
The use of pseudoproline residues can be combined with the other preferred
features described herein. As one example, pseudoproline residues can be used in
combination with microwave conditions.
Alternating solid phase peptide synthesis and metathesis
As described above, ring closing metathesis, whether driven via alkene or alkyne
metathesis, can be difficult and low yielding with some sequences. Another strategy which
may be used to improve the performance of ring closing metathesis between two
complementary metathesisable groups (alkenes or alkynes) which are within the one peptide
is by alternating peptide synthesis and catalysis steps. This approach may also be combined
with the other preferred features described herein. For example, the peptide may be provided
on a solid phase, and catalysis can be combined with the use of turn-inducing groups (e.g.
pseudo-proline residues) and/or in combination with the use of microwave conditions.
According to this embodiment, there is provided a method for the synthesis of a
dicarba analogue of insulin comprising an A-chain and a B-chain or fragments, salts,
solvates, derivatives, isomers or tautomers of the A-chain, the B-chain or both, the method
comprising:
(i) providing a part of the A-chain having at least two complementary metathesisable
groups;
(ii) subjecting the A-chain to metathesis to form at least one dicarba bridge;
(iii) adding one or more further amino acids to one or both ends of the A-chain; and
(iv) adding the B-chain.
According to this embodiment, there is further provided a method for the synthesis
of a dicarba analogue of insulin comprising an A-chain and a B-chain or fragments, salts,
solvates, derivatives, isomers or tautomers of the A-chain, the B-chain or both, the method
comprising:
(i) providing a part of the A-chain and/or a part of the B-chain having at least two
complementary metathesisable groups between them ;
(ii) subjecting the A-chain and B-chain to metathesis to form at least one dicarba
bridge; and
(iii) adding one or more further amino acids to one or both ends of the A-chain and/or Bchain.
In this approach, the sequence is grown in a stepwise fashion until both
metathesisable residues have been incorporated.
This method can be used to facilitate the production of alkyne dicarba bridge
formation and/or alkene dicarba bridge formation. In order to facilitate the production of an
alkyne dicarba bridge, the at least two complementary metathesisable groups in step (i) will
be complementary alkyne-containing metathesisable groups. In order to facilitate the
production of an alkene dicarba bridge, the at least two complementary metathesisable
groups in step (i) will be complementary alkene-containing metathesisable groups.
One of the fragments of insulin may be provided on a solid support. Preferably, the
second metathesisable group of the pair is located at or near the W-terminus of the peptide.
The resin-supported, incomplete sequence is then exposed to the metathesis catalyst to form
the dicarba bridge. Following the metathesis step, the resin can either be subjected to
secondary catalysis (i.e. hydrogenation or metathesis), or followed immediately with solid
phase peptide synthesis (SPPS) to add the one or more further amino acids. The one or
more further amino acids may be added to complete the sequence of the A-chain or the Bchain
of insulin or to add one or more further amino acid residues to the fragments of insulin.
Amino acid residues may be added to either end of the A-chain or the B-chain of insulin.
Preferably, amino acid residues are added to the W-terminus of the desired target peptide.
It will be appreciated that this process can be conducted iteratively in order to
introduce more than one dicarba bridge. This interrupted approach, shown below, can be
highly successful with sequences which are difficult to metathesise.
Tethers between peptide sequences
In some instances, cross-metathesis between peptide sequences can be difficult
and low yielding. Success is often sequence dependent and relies on favourable positioning
of reacting motifs which can be hampered by peptide size, aggregation, deleterious hydrogen
bonding/salt bridges and steric constraints imposed by the primary sequence.
One approach by which we can enhance the metathesis between two
complementary metathesisable groups (alkenes or alkynes) is to utilise a contiguous peptide
sequence, containing the two amino acids or peptides to be connected by a dicarba bridge,
joined together via a removeable tether. Such an approach capitalises on the improved
positioning of the reactive motifs imposed by the tether and hence exploits the enhanced
reactivity via an intramolecular reaction ( CM) compared to an intermolecular reaction (CM)
to produce superior ligation yields. Such an approach is illustrated below:
Removal
of the tether
PeptideA
Peptide B
PeptideA
Peptide B
ep e
In this example, SPPS is used to generate a single peptide sequence where a
transient/ removeable tether is positioned between the two metathesisable groups. Catalysis
is then performed on the resin-bound peptide (RCM, RCAM and/or H) and the resultant
cyclic peptide is then cleaved open at the tether to result in the target acyclic peptide. The
final peptide is analogous to that produced via a direct CM reaction between two peptide
sequences. The resin-appended sequence can then be further elaborated via SPPS in a
number of positions as shown below.
Groups which may function as a removeable tether are structurally diverse. Any
motif which can be chemoselectively incorporated and removed from the sequence, either
chemically or enzymatically. The removeable tether may be a motif which can be added by
reductive amination. The removeable tether may be a motif which can be removed by
photolysis. Preferably, the removeable tether is a motif which also promotes a turn in the
backbone of the primary sequence (similarly to the turn inducing residues described above).
In this situation, the metathesis reaction may be enhanced by suitable positioning of the
reactive motifs could be used. As one example, the removeable tether may be hydroxy-6-
nitrobenzaldehyde.
Solvents
The metathesis reaction may be performed in any suitable solvent which provides
good catalytic turnover and good resin swell.
Particularly for reactions conducted with the readable peptide or the first reactable
peptide attached to a solid support such as a resin, metathesis is preferably performed in a
solvent combination comprising a resin-swelling solvent with a co-ordinating solvent for the
catalyst is preferably also used. In resin-supported reactions, swelling of the resin is required
to avoid aggregation and to promote catalyst access to reactive functional groups. Some
solvents which are suitable for swelling resins are not compatible with metathesis and/or
hydrogenation catalysts, and hence careful selection must be made. For example,
polystyrene-based resins show optimal swelling in chlorinated solvents such as
dichloromethane, however these solvents are not compatible with hydrogenation catalysts.
The solvents react with such catalysts to compromise catalyst function - which in turn
reduces the catalytic cycle (or turn-over number - TON), resulting in incomplete conversion.
It was found that the addition of a small amount of a co-ordinating solvent for the catalyst,
such as an alcohol (e.g. methanol, isopropanol) which can co-ordinate into a vacant site of
the catalyst to facilitate stability, overcame this problem. The co-ordinating solvent is suitably
used in an amount of about 1-30%, for example constituting 10% of the solvent, by volume.
The resin swelling agent may be any polar solvent known to swell the resin, such as
dichloromethane. Other suitable solvents for a range of resins are as set out in Santini, R.,
Griffith, M. C. and Qi, M., Tet. Lett. , 1998, 39, 8951 -8954, the entirety of which is
incorporated herein by reference.
Solid supports
The peptide or peptides used in the preparation of dicarba analogues of insulin are
preferably attached to a solid support. The A-chain or B-chain of insulin may be attached to
a solid support, or both the A-chain and the B-chain may be attached to a solid support.
A plethora of solid supports are known and available in the art, and include pins,
crowns, lanterns and resins. Examples are polystyrene-based resins (sometimes referred to
as solid supports), including cross-linked polystyrene resins {via 1% divinylbenzene)
functionalised with linkers (or handles) to provide reversible linkages between the readable
organic compound (which may be a peptide sequence containing side-chains with crossmetathesisable
groups) and the resin. Examples of polystyrene-based resins include Wang
resin, Rink amide resin, BHA-Gly-Gly-HMBA resin and 2-chlorotrityl chloride resin. Other
forms of solid supports that may not necessarily be characterised as resins can also be used.
Under microwave reaction conditions it is possible to have a higher solid support
loading than is conventionally used in peptide synthesis on solid supports. Typical solid
support loadings are at the 0 .1 mmol/g level, but microwave radiation (optionally combined
with solvent choice, as described above) overcomes the aggregation problems at higher
solid support loadings, so that solid support loading at around 0.9 mmol/g (nine times higher)
is achievable. The solid support loadings may also be at around 0.2 mmol/g and above,
such as 0.5 mmol/g and above.
Formation of one or more disulfide bridges to prepare dicarba analogues of
insulin
The following refers to the use of dicarba bridges in dicarba analogues to replace
the disulfide bridges of insulin. It will be appreciated that the following decription could also
be applied to the preparation of other dicarba analogues of insulin in which the dicarba
bridges are used to replace other structural motifs, such as salt bridges, or other noncovalent
interactions.
Insulin has three disulfide bonds: CysA 6 and CysA oxidise to form the
intramolecular disulfide bridge. CysA 7 and CysB7 oxidise to form an intermolecular disulfide
bridge and CysA 2o and CysB oxidise to form the second intermolecular bridge. This
connectivity needs to be preserved to maintain biological activity. Hence, disulfide formation,
and hence chain combination, must be performed via a regioselective approach.
Significantly, undirected ligation of two or more peptide chains is usually only
partially successful in native molecules. When sequence modifications are introduced
however, unnatural folding can become more significant and exert a deleterious effect on
chain combination. The resultant topoisomers contaminate the required product and lower
the yield of native isomer. These challenges have been reported extensively in the literature,
particularly in the unsuccessful syntheses of insulin analogues containing modified A- and Bchains.
The dicarba analogues of insulin may have one or more disulfide bonds as well as
at least one dicarba bridge. The disulfide bonds may be introduced at any location, however,
it is preferred that they are introduced into a peptide or peptides at locations where disulfide
bonds are present in the native insulin.
Where more than one disulfide bridge is to be introduced into the peptide or
peptides, disulfide formation, and hence chain combination, must be performed via a
regioselective approach. One approach is to use orthogonally-protected cysteine residues to
sequentially construct the disulfide bridges. An example of the regioselective bond forming
strategy was used in the regioselective synthesis in Wade et a/. , J. Biol. Chem., 2006, 281,
34942-34954, which is incorporated herein by reference. Preferably, a combination of
complementary thiol protecting groups (e.g. Acm , Bu, Trt) is used to regioselectively install
disulfide bridges.
Reduction of unsaturated dicarba bridges in dicarba analogues of insulin
In some instances, the dicarba bridge of the dicarba analogue of insulin may have
improved activity where the dicarba bridge has a particular conformation in order to serve as
a peptidominetic of insulin. It may therefore be advantageous for the dicarba bridge to adopt
a particular geometry.
The product of alkyne or alkene metathesis is a dicarba analogue of insulin with a
new unsaturated alkyne or alkene-containing dicarba bridge (-C ºC- or -C=C-). If the target
compound is to contain an alkane-containing dicarba bridge (-CH 2-CH 2-), the preparation of
the dicarba analogue may also involve subjecting the alkyne/alkene-containing dicarba
bridge to complete reduction. If the target compound is to contain an alkene-containing
dicarba bridge (-CH=CH-), the preparation may involve semi-reduction of the alkynecontaining
dicarba bridge.
Hydrogenation of an alkyne- or alkene-containing dicarba bridge
The product of the alkyne or alkene metathesis is a dicarba analogue of insulin with
a new unsaturated alkyne- or alkene-containing dicarba bridge (-C ºC- or -C=C-). If the
target compound is to contain an alkene-containing dicarba bridge (-C=C-) or an alkanecontaining
dicarba bridge (-C-C-) the process may further comprise the step of
hydrogenating the alkyne or alkene bond.
The hydrogenation can be conducted at any temperature, such as room
temperature or at an elevated temperature. The reaction is typically conducted at elevated
pressure, although if slower reaction times can be tolerated, the reaction can be performed at
atmospheric pressure. The hydrogenation reaction can be performed on substrates which
are attached or unattached to a solid support.
Hydrogenation of the unsaturated dicarba bridge can be performed with any known
hydrogenation catalyst. Examples of suitable catalysts include those described in March, J.
Advanced Organic Chemistry: Reactions, Mechanisms and Structure. 1992, pages 771 to
780 and in Ojima, I . Catalytic Asymmetric Synthesis; Wiley-VCH: New York, 2000; Second
Edition, Chapter 1, 1- 110 , incorporated herein by reference. Suitable hydrogenation
catalysts are chemoselective for unblocked, non-conjugated carbon-carbon double or triple
bonds.
Suitable hydrogenation catalysts may be either insoluble in the reaction medium
(heterogeneous catalysts) or soluble in the reaction medium (homogeneous catalysts).
Examples of suitable heterogeneous catalysts include Raney nickel, palladium-on-charcoal
(Pd/C) and platinum oxide. Examples of suitable homogeneous catalysts include Wilkinson's
catalyst, other Rh(l) phosphine complexes, and Ru(ll) phosphine complexes.
The particular hydrogenation catalyst that is used will depend on the dicarba insulin
analogue (the target compound) to be prepared. For example, if the target compound is to
include a saturated alkane-containing dicarba bridge, a hydrogenation catalyst capable of
reducing an alkyne bond (possibly via an alkene intermediate) or an alkene bond to an
alkane bond will be selected ("complete hydrogenation"). As another example, if the target
compound is to include an unsaturated alkene-containing dicarba bridge, a hydrogenation
catalyst capable of reducing an alkyne bond to an alkene bond ("semi-hydrogenation") will be
selected.
If the target compound is to include a saturated alkane-containing dicarba bridge,
the hydrogenation is performed with a catalyst that is chemoselective for unblocked, nonconjugated
carbon-carbon triple and carbon-carbon double bonds as distinct from other
double bonds such as carbon-oxygen double bonds in carbonyl groups, carboxylic acids and
blocked conjugated double bonds. The hydrogenation of an alkyne (-C ºC-) bridge to an
alkane (-CH 2-CH 2- ) bridge involves the initial step of producing an alkene (-C=C-) bridge.
The alkene bridge then becomes a substrate for further reduction to finally produce the
required -CH 2-CH 2- bridge.
Any catalyst which is chemoselective for unblocked, non-conjugated carbon-carbon
triple and double bonds may be used. Examples of hydrogenation catalysts capable of
reducing an alkyne bond to an alkane bond include palladium-on-charcoal (Pd/C), platinum
oxide, and Raney nickel. Hydrogenation catalysts which are suitable for reducing an alkyne
or alkene bond to an alkane bond also include asymmetric hydrogenation catalysts.
Although the use of an asymmetric hydrogenation catalyst is not necessary for the
hydrogenation of the alkyne- or alkene-containing dicarba bridges, asymmetric
hydrogenation catalysts can nevertheless be used. Any asymmetric hydrogenation catalyst
which is chemoselective for unblocked non-conjugated carbon-carbon double or triple bonds
may be used. Catalysts in this class are described in US 5,856,525 which is incorporated
herein by reference. Such homogenous hydrogenation catalysts are tolerant of sulfide, and
disulfide bonds, so that the presence of disulfide bonds and the like will not interfere with the
synthetic strategy. Examples of suitable asymmetric hydrogenation catalysts are the chiral
phosphine catalysts, including chiral phospholane Rh(l) catalysts.
Some hydrogenation catalysts are chemoselective for alkyne groups as distinct from
alkene groups. Such catalysts are thus capable of hydrogenating an alkyne and stopping the
reaction at an alkene. The hydrogenation of an alkyne-containing dicarba bridge (-C ºC-) to
form an alkene containing dicarba bridge (-C=C-) is therefore performed with a catalyst that
is chemoselective for unblocked non-conjugated carbon-carbon triple bonds as distinct from
other double bonds such as carbon-carbon double bonds, carbon-oxygen double bonds in
carbonyl groups, carboxylic acids and blocked conjugated double bonds. Any catalyst which
is chemoselective for unblocked non-conjugated carbon-carbon triple bonds may be used.
Within the group of hydrogenation catalysts are catalysts that are chemoselective for
alkyne groups as distinct from alkene groups. These catalysts are also able to
stereoselectively reduce an alkyne-containing dicarba bridge to form an alkene-containing
dicarba bridge that is enriched in either the cis- or the frans-isomer. This method allows
biased generation of either the cis- or frans-isomer by selecting a catalyst which produces a
product enriched in the desired isomer.
The controlled reduction of C=C and CºC in organic compounds is an important
synthetic transformation and many catalysts are available to achieve this end. Partial
conversion of alkynes into alkenes provides a particularly useful route to geometrically well
defined alkenes. A large number of well defined homogeneous transition metal complexes
can be used to affect stereoselective semi-hydrogenation of the CºC bond, and many of
these catalysts are also tolerant of a wide range of organic functionality. Organochromium,
iron, ruthenium , osmium, rhodium, iridium and palladiaum complexes, inter alia, have all
been used in semi-hydrogenation reactions of alkynes. Reaction conditions (e.g. solvent,
temperature) play a large role in influencing reaction selectivity. Towards this end, catalysts
and reaction conditions can be coupled to selectivity and stereoselectivity perform CºC ®
C=C transformations in the presence of existing C=C bonds without resulting in over¬
reduction. For example, zerovalent palladium catalysts bearing bidentate nitrogen ligands
are able to homogeneously hydrogenate alkynes to Z-alkenes and do not reduce existing
alkene functionality.
Any catalyst which is stereoselective and chemoselective for unblocked nonconjugated
carbon-carbon triple bonds may be used. Catalysts in this class include those
described in Kluwer, A.M. , Elsevier, C.J. "The Handbook of Homogeneous Hydrogenation",
2007, Wiley-VCH (de Vries, J .G. , Elsevier, C.J. (Editors)), Ch 14 : Homogeneous
Hydrogenation of Alkynes and Dienes, pp 375-41 1, incorporated herein by reference.
Examples of suitable chemoselective hydrogenation catalysts include poisoned Lindlar's
catalyst, Pd(0), Ru(ll), ruthenium carbonyl clusters, Pt(0), P2-Ni, chromium tricarbonyl
compounds of the generic formula [Cr(CO) 3(arene)] , Fe(l l) catalyst presursors such as
[(PR3)FeH(N2)]BPh4, [(PR3)FeH(H2)]BPh4 and (PR3=P(CH2CH2PPh2)3) , osmium catalysts
such as [OsH(CI)(CO)(PR 3)2) and rhodium catalysts such as the Schrock/Osborn cationic
Rh-catalyst.
In one example of chemoselective and stereoselective hydrogenation, the alkynecontaining
dicarba bridge is hydrogenated in the presence of Pd(0)-catalyst. The resulting
dicarba analogue of insulin has an alkyne-containing dicarba bridge that is enriched in the
c/s-isomer. In another example performing the hydrogenation in the presence of a Ru(l l)-
catalyst results in a dicarba analogue of insulin having an alkene-containing dicarba bridge
that is enriched in the trans-isomer. This is shown schematically below.
Where the dicarba bridge of the dicarba analogue of insulin is an alkene-containing
dicarba bridge, the alkene containing group of the bridge may be present as a mixture of any
ratio of geometric isomers (e.g. E- or Z-configured alkenes), or as an enriched geometric
isomer. As defined above, "enriched" means that the mixture contains more of the preferred
isomer than of the other isomer. Preferably, an enriched mixture comprises greater than
50% of the preferred isomer, where the preferred isomer gives the desired level of potency
and selectivity. More preferably, an enriched mixture comprises at least 60%, 70%, 80%,
90%, 95%, 97.5% or 99% of the preferred isomer.
When the product produced by the method of the present invention is a peptide
having an alkyne-containing dicarba bridge, the step of hydrogenating the alkyne-containing
dicarba bridge can be performed with the peptide attached to a resin. Preferably, when the
peptide substrate is attached to a resin, the hydrogenation step uses a homogeneous
hydrogenation catalyst.
Reduction of an alkyne-containing dicarba bridge
As described above, alkyne metathesis produces an alkyne-containing dicarba
bridge formed between two amino acids. This alkyne-containing dicarba bridge may be
converted to the corresponding alkene-containing or alkane-containing dicarba bridge by
reduction methods other than hydrogenation of the alkyne-containing dicarba bridge.
Reduction of the alkyne-containing dicarba bridge may be stereo-selective to reduce
an alkyne-containing dicarba bridge to an alkene-containing dicarba bridge that is enriched in
either the cis- or the frans-isomer.
In one example, the alkyne-containing dicarba bridge may be stereo-selectively
reduced via hydrosilylation and protodesilylation to produce an alkene-containing dicarba
bridge. The alkene that results is enriched in the frans-isomer. This type of reduction is
described in Fijrstner, A., Radkowski, K. Chem Commun. 2002, 2 182 and Lacombe, F.,
Radkowski, K., Seidel, G. and Fijrstner, A., Tetrahedron, 2004, 60, 731 5 , and incorporated
herein by reference. In this approach, the alkyne-containing dicarba bridge can be
selectively reduced to the fra/is-isomer by frans-selective hydrosilylation followed by
protodesilylation. The two steps involved in this selective reduction are shown below.
hydrosilylation
The hydrosilylation step may be performed using (EtO)3SiH in the presence of the
cationic ruthenium complex [Cp*Ru(MeCN) 3]PF . In this reaction the HSi(OEt)3 reagent is
added with c/s-selectively across the alkyne bond of the alkyne-containing dicarba bridge.
After protodesilylation, a product that is enriched in the frans-isomer across the alkenecontaining
dicarba bridge is produced.
When the product being produced is a peptide having an alkyne-containing dicarba
bridge, the step of reducing the alkyne-containing dicarba bridge can be performed with the
peptide attached to a resin.
Formation (regioselectively) of multiple dicarba bridges in the peptide or
peptides
The strategy for the formation of a dicarba bridge as described above is suitable to
form dicarba analogues of insulin with multiple dicarba bridges, optionally with disulfide
bridges, or to form dicarba analogues of insulin with at least one dicarba bridge and at least
one disulfide bridge. The dicarba analogues of insulin may include more than one alkynecontaining
dicarba bridge, more than one alkene-containing dicarba bridge or more than one
alkane-containing dicarba bridge, or combinations thereof, with or without one or more
disulfide bridges.
To form dicarba analogues of insulin having multiple dicarba bridges, it may be
necessary to include at appropriate locations in the insulin peptides, pairs of complementary
metathesisable groups which are blocked or deactivated for the times when different pairs of
metathesisable groups are being linked together, and unblocked or "activated" to enable
reaction to occur between those pairs. Accordingly, for each bridge-forming pair, there
should to be an unblocking reaction available that will selectively unblock the required pairs.
The first pair of complementary metathesisable groups to be subjected to
metathesis (alkene metathesis or alkyne metathesis) need not be blocked. The pair of
unblocked complementary metathesisable groups is then subjected to the metathesis
reaction, as described above, to form a dicarba bridge. Optionally, the newly formed
unsaturated dicarba bridge may be subjected to reduction. The step of reduction may occur
after the first dicarba bridge has been formed, or after the further dicarba bridges or disulfide
bridges are formed.
When the dicarba analogue of insulin contains more than one dicarba bridge. The
dicarba bridges may be selected from an alkyne-containing dicarba bridge, an alkenecontaining
dicarba bridge and an alkane-containing bridge. The two or more dicarba bridges
may be the same or different.
As described above, metathesis is used to introduce dicarba bridges into the dicarba
analogue of insulin.
Tandem alkyne metathesis
When the dicarba analogue of insulin is to contain two or more alkyne-containing
dicarba bridges, it is important to avoid the formation of an intractable mixture of different
products from random metathesis between pairs of alkyne-containing metathesisable groups
or between an alkyne-containing metathesisable group or groups and a formed alkynecontaining
dicarba bridge.
In one approach, a pair of complementary alkyne-containing metathesisable groups
may be introduced during synthesis of a peptide and alkyne metathesis conducted to form a
first alkyne-containing dicarba bridge before subsequent pairs of complementary alkynecontaining
metathesisable groups are introduced into the peptide. This approach is the
refered to as the Alternating-SPPS-catalysis approach and is as described above. In another
approach, blocking groups may be used to allow regioselective formation of the particular
alkyne containing dicarba bridges. It will be appreciated that any combination of these
approaches may also be used to prepare the desired product.
An example of a typical route for the introduction of two alkyne-containing dicarba
bridges into a dicarba analogue of insulin is shown below. In this example, alkyne-containing
metathesisable groups are used to form dicarba analogues containing one intra-molecular
dicarba bridge and one inter-molecular dicarba bridge, in which the peptides correspond to
the A-chain or the B-chain of insulin or fragments, salts, solvates, derivatives, isomers or
tautomers thereof.
In example (A) above, two metathesisable groups in a single peptide are subjected
to ring-closing alkyne-metathesis ( CAM) to produce an alkyne-containing dicarba bridge.
Following this, the alkyne-containing dicarba bridge is blocked and peptide synthesis is
conducted to introduce a further alkyne metathesisable group which is complementary to an
alkyne-containing metathesisable group on a separate peptide. These two separate
peptides represent the A-chain and/or the B-chain of insulin. The unblocked alkynecontaining
metathesisable groups are then subjected to alkyne-metathesis (CAM) to produce
a second alkyne-containing dicarba bridge. The blocked alkyne-containing dicarba bridge
may then be unblocked to produce a peptide containing one intramolecular and one
intermolecular alkyne-containing dicarba bridge.
In example (B) above, a peptide is synthesised having two unblocked
metathesisable groups and a blocked metathesisable group between the two. The
unblocked metathesisable groups are subjected to ring-closing alkyne-metathesis ( CAM) to
produce an alkyne-containing dicarba bridge. Following this, the alkyne-containing dicarba
bridge is reduced to an alkene-containing dicarba bridge and the blocked alkyne-containing
metathesisable group is unblocked. A second peptide having an alkyne-containing
metathesisable group is then introduced, and the two peptides are subjected to alkynemetathesis
(CAM) to produce a second alkyne-containing dicarba bridge.
Blocking and activation for alkyne metathesis
For metathesis to occur between two alkynes, the alkynes must not be blocked or
protected. A blocking group is any group that prevents metathesis from taking place in the
presence of a metathesis catalyst. Preferably, a blocking group is used to prevent reaction
of the alkyne metathesisable group during olefin metathesis, where the dicarba insulin
analogue is to include both an alkyne-containing dicarba bridge and an alkene-containing
dicarba bridge. B locking groups may also be provided on a formed alkyne-containing bridge.
In a preferred embodiment, the blocking group is provided on either the unreacted alkynecontaining
metathesisable group or the alkyne-containing dicarba bridge, during olefin
metathesis.
Examples of blocking groups for an alkyne include dicobalt hexacarbonyl groups.
Removal of one or both of the blocking groups unblocks the alkyne-containingmetathesisable
group to enable alkyne metathesis to take place or unblocks the alkynecontaining
dicarba bridge. It is noted that for subsequent alkyne metathesis the pair of
alkyne-containing metathesisable groups that remain after unblocking need not be identical.
For example, after deblocking, but-4-ynylglycine and pent-4-ynylglycine may be
metathesised to form a new alkyne bridge. The term "complementary" is used to indicate that
the pair of unblocked alkyne-containing metathesisable groups are not necessarily identical,
but are merely complementary in the sense that cross-metathesis can take place across the
two alkyne groups.
As described above, using a combination of blocking and unblocking mechanisms
allows regioselective synthesis of multiple dicarba bridges (intra- and/or interchain) in dicarba
analogues of insulin.
Tandem alkene metathesis
When the dicarba analogue of insulin is to contain two or more alkene-containing
dicarba bridges, it is also important to avoid random metathesis occurring between pairs of
alkene-containing metathesisable groups or between an alkene-containing metathesisable
group and a formed alkene-containing dicarba bridge.
For alkene metathesis, suitable groups for forming the first pair of complementary
methathesisable groups which are not blocked are -CH=CH 2 and -CH=CH-CH 3. These
groups may be included in insulin by peptide synthesis, and may be provided via an amino
acid connected to -CH=CH 2 or having -CH=CH 2 in its side chain optionally with any divalent
linking group linking the carbon at the "open" end (the -CH= carbon atom) to the amino acid
backbone, such as an -alkylene-, -alkylene-carbonyl-, and so forth. Examples of -CH=CH 2-
containing amino acids and -CH=CH-CH 3-containing amino acids are allylglycine and
crotylglycine, respectively. Each of these amino acids contains the divalent linking group -
CH2- between the alkylene and the amino acid (peptide) backbone.
At the completion of that reaction (and optionally after hydrogenation of the first
dicarba bridge), the blocked second pair of complementary metathesisable groups, can be
subjected to an unblocking reaction.
When the first pair of complementary metathesisable groups are olefins, suitable
functional groups for forming the second pair of complementary metathesisable groups are
di-blocked alkylenes, such as the group in which R 2 and R 3 are each
independently selected from blocking groups, such as alkyi. R 2 and R 3 are preferably alkyi
of C 1 to C 15 . More preferably, R 2 and R 3 are small chain alkyls, for example methyl.
Again, there may be a divalent linking group between the -CH= carbon atom , and the amino
acid backbone, such as an alkylene group, for instance -CH 2- An example of an amino acid
containing this group is prenylglycine, or protected prenylglycine.
The unblocking reaction, or activation reaction, to convert the pair of di-blocked
alkylenes into an unblocked alkylenes involves subjecting the blocked second pair of
complementary metathesisable groups to cross-metathesis with a disposable olefin, to effect
removal of the blocking groups (such as R 2 and R 3 in the example shown above).
It will be understood that in this case, cross-metathesis is used to replace the group
=CR 2R 3 with another unblocked group =CH2 or =CHR 4 , (in which R 4 may be -H, a
functionalised alkyi or alkyi for instance) which is then "activated" or "unblocked" and ready
for being subjected to cross-metathesis for the formation of a dicarba bridge, using the same
techniques described above.
The conditions for this activation-type of cross-metathesis are the same as
described above for the dicarba bridge forming metathesis. It can be performed under
microwave conditions, although it need not be, as the disposable olefin is a smaller molecule
and less subject to the spatial constraints as larger reactable peptides and single reactable
peptides in which intramolecular bridges are to be formed.
The "disposable olefin" is suitably a mono-substituted ethylene (such as
monoalkylated ethylene - such as propene, which is mono-methylated ethylene), or a 1,2-
disubstituted ethylene (such as high purity 2-butene, and optionally of cis or trans geometry,
or a mixture thereof). Previously, commercial 2-butene has been attempted to be used as
the disposable olefin in this unblocking reaction, and the reaction is thus sometimes referred
to as "butenolysis". However, commercially available 2-butene (which is a mixture of cis- and
frans-2-butene) can inhibit olefin metathesis due to low level butadiene contaminants.
The substituents on the substituted ethylene disposable olefin are substituents that
do not participate in the reaction. Examples are alkyl or a functionalised (substituted) alkyl.
The functional group of the functionalised alkyl is suitably a polar functional group, to assist
with swelling of the solid support, and solubility. Examples are hydroxy, alkoxy, halo, nitrile
and carboxylic acids/esters. One specific example is the di-ester functionalised disposable
olefin 1,4-diacetoxy-2-butene.
Thus the disposable olefin is suitably a 1,3-butadiene-free disposable olefin, or a
1,3-butadiene-free mixture of disposable olefin and is preferably 1,3-butadiene-free olefin or
olefin mixture of one or more of the following olefins:
wherein X and Y are each independently selected from the group consisting of -H, alkyl and
alkyl substituted with one or more substituents selected from halo, hydroxy, alkoxy, nitrile,
acid and ester.
Preferably, at least one of X and Y is not H.
Preferably, in the case of the alkyl substituents, the substituent is preferably on the
carbon atom . Preferably the substituted alkyl is a substituted methyl. According to one
embodiment, at least one of X and Y is a substituted alkyl, such as a substituted methyl. X
and Y may be the same or different. The olefins may be cis or trans, or mixtures of both.
Blocking and activation for alkene metathesis
For metathesis to occur between two alkene groups (olefins), the alkenes must not
be blocked by any steric or electronic blocking groups. A steric blocking group is any bulky
group that sterically prevents the metathesis from taking place in the presence of a crossmetathesis
catalyst. An example of a steric blocking group on an olefin is an alkyl group.
Prenylglycine is an example of an amino acid containing a dialkyl-blocked olefin side chain
(specifically, dimethyl-blocked). Removal of one or both of the blocking groups unblocks the
olefin, and enables the cross-metathesis to take place.
It is noted that the pair of metathesisable groups that remain after unblocking need
not be identical - a mono-substituted olefin (such as a mono-methylated olefin, e.g.
crotylglycine) and an unsubstituted olefin (being unsubstituted at the open olefinic end, e.g.
allylglycine) can form a suitable pair of cross-metathesisable groups. The term
"complementary" is used to indicate that the pair of unblocked metathesisable groups are not
necessarily identical, but are merely complementary in the sense that cross-metathesis can
take place across the two olefinic groups.
Electronic blocking refers to the presence of a group on the organic compound or
reactable peptide that modifies the electronic nature of the olefin (alkene) group of the
reactable peptide (which would otherwise undergo cross-metathesis), so as to prevent that
olefin group from undergoing cross-metathesis. An example of an electronic blocking group
is when the double bond is in conjugation with a C=0 group - that is, a double bond adjacent
to an a,b- unsaturated carbonyl containing group (e.g. C=C-C=C-C=0 where the C=C
portion is the otherwise reactable group). The a-b-unsaturation withdraws electrons away
from the olefinic (C=C) cross-metathesisable group causing electronic blocking and
prevention of cross-metathesis. By using a combination of blocking mechanisms, a series of
pairs of metathesisable groups in the reactable peptide or peptides of insulin can be
designed, with different reaction conditions to effect selective unblocking of particular pairs.
In this way, it becomes possible to regioselectively synthesise multiple dicarba bridges (inter
and/or intramolecular) in insulin.
Tandem alkene and alkyne metathesis
The use of tandem alkene and alkyne metathesis to facilitate regioselective
synthesis of multiple dicarba bridges is a viable strategy for the synthesis of a peptide or
peptides containing at least one alkyne containing dicarba bridge and at least one alkenecontaining
dicarba bridge. Modern metathesis catalysts are highly chemoselective and are
readily tuned to unsaturated substrates to achieve maximum selectivity.
When the target peptide is to contain at least one alkyne-containing dicarba bridge
and at least one alkene-containing dicarba bridge, it is important to avoid the formation of an
intractable mixture of different products from random metathesis between pairs of
metathesisable groups or between metathesisable groups and formed dicarba bridges.
It will be appreciated that the step of alkyne metathesis may occur before any
number of steps involving alkene metathesis or at the conclusion of any num ber of steps
involving alkene metathesis. However, it is preferred that either the alkyne containing
complementary metathesisable groups or the dicarba bridge formed by alkyne metathesis is
blocked during any alkene metathesis steps.
In the formation of a dicarba analogue of insulin having one alkyne-containing
dicarba bridge and at least one alkene-containing dicarba bridge, it is possible to introduce a
pair of complementary alkyne-containing or alkene-containing metathesisable groups during
synthesis of a peptide and conduct metathesis to form the first dicarba bridge before
subsequent pairs of complementary metathesisable groups are introduced into the peptide.
It may also be necessary to use blocking groups to allow regioselective formation of the
particular alkyne- or alkene-containing dicarba bridges.
Examples of synthetic routes for the introduction of one alkyne-containing dicarba
bridge and one alkene dicarba bridge are shown below. These examples show the use of
alkene-containing and alkyne-containing metathesisable groups to form dicarba analogues
containing intrachain dicarba bridges in a peptide which may correspond to the A-chain, the
B-chain or both of insulin.
In the catalytic pathways A and B, both combine alkyne and alkene metathesis for
the regioselective formation of two dicarba bridges. Both routes involve alkene cross
metathesis of a pair of alkene-containing metathesisable groups, alkyne cross metathesis of
a pair of alkyne-containing metathesisable groups, and the optional reduction of the newly
formed bridges to the corresponding alkanes. The difference between the two pathways is
the order of the catalysis. In pathway A, alkyne metathesis (RCAM) preceeds alkene (RCM)
metathesis, and in pathway B, alkene metathesis preceeds alkyne metathesis.
Transition metal metathesis catalysts show varying degrees of chemo-specificity in
their activity toward potential substrates. Transition metal alkylidene bearing catalysts, such
as those employed in alkene metathesis, show an affinity for coordination and subsequent
metathesis of readable olefins. Some of these catalysts are able to coordinate alkynes and
undergo enyne metathesis, the bond reorganisation of an alkyne to form a 1,3-diene (Chem.
Rev., 2004, 104(3), pp 13 17-1 382). These species may react with other alkenes to produce
further products. Some alkene metathesis catalysts, such as the first generation Grubbs'
catalyst, show limited propensity toward this side reaction in the presence of available
alkynes. To avoid unwanted reaction of alkyne moieties the group may be blocked by an
appropriate blocking group as described herein.
Transition metal alkylidyne bearing catalysts, such as Schrock's alkyne metathesis
catalyst, show high activity toward alkyne metathesis but do not participate in olefin or enyne
metathesis ( . R. Schrock, Chem. Commun. , 2005, 2773-2777). Hence, protection of olefins
during alkyne metathesis is not necessary with catalysts of this kind. Preferably, alkyne
metathesis in the presence of alkene-containing metathesisable groups or alkene-containing
dicarba bridges is performed with Schrock's catalyst.
Blocking and activation of alkyne-containing metathesisable groups and alkenecontaining
metathesisable groups may be achieved as described herein when the dicarba
analogue of insulin contains both an alkyne-containing dicarba bridge and an alkenecontaining
dicarba bridge.
Reduction of peptides having additional dicarba bridges
The reduction of alkyne-containing dicarba bridges and alkene-containing dicarba
bridges may be performed as described hereinabove. It will be appreciated by a person
skilled in the art that when multiple dicarba bridges are to be present in a dicarba analogue of
insulin, the step of reducing the initially installed alkyne- or alkene-containing dicarba bridge
may take place before or after the metathesis reaction to additional dicarba bridges.
Peptide synthesis
The method for the synthesis of a dicarba bridge in a peptide such as insulin is
described above.
Generally, the peptide will be a protected peptide (such as Fmoc-protected), and will
comprise a sequence corresponding to the A-chain or the B-chain or both of insulin or a
fragment, salt, solvate, derivative, isomer or tautomer thereof. The amino acids which make
up the sequence corresponding to insulin can be any of the amino acids described earlier,
but it is convenient for the synthesis of peptidomimetics for the amino acids to be selected
from the 20 naturally-occurring amino acids, g - and b- amino acids and from any crossmetathesisable
group-bearing analogues or alkyne metathesisable group bearing analogues
thereof. An example of metathesisable group-bearing analogues are allylglycine and
butynylglycine.
It will be appreciated that if a peptide sequence is added later through an
intermolecular bridge, the corresponding metathesisable groups on that peptide need not be
blocked - as they can be added to the reaction at the time of cross-metathesis, after the
unblocking of the groups on the solid-supported peptide.
Uses involving the dicarba insulin analogues
In the field of insulin delivery, where multiple repeated administrations are required
on a daily basis throughout the patient's life, it is desirable to create compositions of insulin
that do not alter physiological clinical activity and that do not require injections. Oral delivery
of insulin is a particularly desirable route of administration, for safety and convenience
considerations. In addition to minimizing or eliminating the discomfort that often attends
repeated hypodermic injections, it removes the needs for delivery devices (e.g. injecting
devices) and allows the unit doses to be formulated into convenient forms (e.g. tablets) which
are typically easier to handled, transport and stored. It has been a significant unmet goal in
the art to imitate normal insulin levels in the portal and systemic circulation via oral
administration of insulin.
Oral delivery of insulin may have advantages beyond convenience, acceptance and
compliance issues. Insulin absorbed in the gastrointestinal tract is thought to mimic the
physiologic route of insulin secreted by the pancreas because both are released into the
portal vein and carried directly to the liver before being delivered into the peripheral
circulation. Absorption into the portal vein maintains a peripheral-portal insulin gradient that
regulates insulin secretion. In its first passage through the liver, roughly 60% of the insulin is
retained and metabolized, thereby reducing the incidence of peripheral hyperinsulinemia, a
factor linked to complications in diabetes.
Insulin exemplifies the problems confronted in the art in designing an effective oral
drug delivery system for biological macromolecules. Insulin absorption in the gastrointestinal
tract is presumably hindered by, amongst other things, its susceptibility for enzymatic
degradation. The physicochemical properties of insulin and its susceptibility to enzymatic
digestion have precluded the design of a commercially viable oral or alternate delivery
system .
Insulin currently cannot be taken orally because it is broken down in the
gastrointestinal tract to peptide fragments (even single amino acid components). This
digestion of insulin results in a loss of activity.
The dicarba analogues of insulin have been found to be very stable upon storage
over long periods of time, even in acidic conditions. It is envisaged that such insulin
derivatives have the potential to be delivered orally whilst still providing a sustained profile of
action. There are several ways to assess the stability of insulin. One way is an HPLC
stability-indicating assay. This method determines the amount of intact insulin molecules
present in a sample, but does not determine whether these molecules are in a bioactive
conformation, which is necessary in order to have an effective product. Other methods are
measurement of related substances (impurities) by HPLC and assessing the bioactivity of the
product, which could be an in vivo assay or an in vitro predictor of in vivo performance. The
biological activity of a dicarba insulin may be measured in an assay as known by a person
skilled in the art as e.g. described in WO 2005/01 2347.
In one embodiment, a pharmaceutical composition comprising the therapeutic
dicarba insulin is stable for more than 6 weeks of usage and for more than 3 years of
storage. In another embodiment, the pharmaceutical composition comprising the therapeutic
dicarba insulin is stable for more than 4 weeks of usage and for more than 3 years of
storage. In another embodiment, the pharmaceutical composition comprising the therapeutic
dicarba insulin is stable for more than 4 weeks of usage and for more than 2 years of
storage. In another embodiment, the pharmaceutical composition comprising the therapeutic
dicarba insulin is stable for more than than 2 weeks of usage and for more than two years of
storage. In another embodiment, the pharmaceutical composition comprising the therapeutic
dicarba insulin is stable for more than 1 weeks of usage and for more than one year of
storage.
In another embodiment, the pharmaceutical composition is in the form of a clear
solution and is stable for more than 6 weeks of usage and for more than 3 years of storage.
In another embodiment, the pharmaceutical composition is in the form of a clear solution and
is stable for more than 4 weeks of usage and for more than 3 years of storage. In another
embodiment, the pharmaceutical composition is in the form of a clear solution and is stable
for more than 4 weeks of usage and for more than two years of storage. In another
embodiment, the pharmaceutical composition is in the form of a clear solution and is stable
for more than 2 weeks of usage and for more than two years of storage. In another
embodiment, the pharmaceutical composition is in the form of a clear solution and is stable
for more than 1 week of usage and for more than one year of storage.
In another embodiment, the pharmaceutical composition is in the form of a tablet
and is stable for more than 3 years of storage. In another embodiment, the pharmaceutical
composition is in the form of a tablet and is stable for more than two years of storage. In
another embodiment, the pharmaceutical composition is in the form of a tablet and is stable
for more than one year of storage.
Reference to "storage" above equates to "shelf-stability" or being "shelf-stable".
Shelf-stability includes chemical stability as well as physical stability. Chemical instability
involves degradation of covalent bonds, such as hydrolysis, racemization, oxidation or
crosslinking. Chemical stability of the formulations is evaluated by means of reverse phase
(RP-HPLC) and size exclusion chromatography SE-HPLC). In one aspect of the invention,
the formation of peptide related impurities during shelf-life is less than 20 % of the total
peptide content. In a further aspect of the invention, the formation of peptide related during
impurities during shelf-life is less than 10 %. In a further aspect of the invention, the
formation of peptide related during impurities during shelf-life is less than 5 %. The RP-HPLC
analysis is typically conducted in water-acetonitrile or water- ethanol mixtures. In one aspect,
the solvent in the RP-HPLC step will com prise a salt such as Na2S0 4, (NH )2S0 , NaCI, KCI,
and buffer systems such as phosphate, and citrate and maleic acid. The required
concentration of salt in the solvent may be from about 0 .1 M to about 1 M, preferable
between 0.2 M to 0.5 M, most preferable between 0.3 to 0.4 M. Increase of the concentration
of salt requires an increase in the concentration of organic solvent in order to achieve elution
from the column within a suitable time. Physical instability involves conformational changes
relative to the native structure, which includes loss of higher order structure, aggregation,
fibrillation, precipitation or adsorption to surfaces. Peptides such as insulin peptides, GLP-1
compounds and amylin compounds are known to be prone to instability due to fibrillation.
Physical stability of the formulations may be evaluated by conventional means of e.g. visual
inspection and nephelometry after storage of the formulation at different temperatures for
various time periods. Conformational stability may be evaluated by circular dichroism and
NMR as described by e.g. Hudson and Andersen, Peptide Science, vol 76 (4), pp. 298-308
(2004).
A pharmaceutical composition of the present invention is preferably shelf-stable for
at least the period which is required by regulatory agencies in connection with therapeutic
proteins. Preferably, a shelf-stable pharmaceutical composition is stable for at least one
year, more preferably at least two years, at 5 °C. Preferably, a shelf-stable pharmaceutical
composition is stable for at least one year, more preferably at least two years, at 25 °C.
Methods of Treatment
The dicarba analogues of insulin of the present invention provide methods of
treating subjects with the following diseases or conditions: hyperglycemia; impaired glucose
tolerance; early stage diabetes; late stage diabetes; diabetes mellitus including diabetes type
I and diabetes type I I ; and metabolic syndrome; or diseases or conditions that directly or
indirectly result therefrom . For example, a hyperglycem ia associated disease refers to a
disorder or disorders that directly or indirectly result from elevated levels of glucose in the
blood plasma.
The dicarba analogues of insulin of the present invention also provide methods of
treating subjects in order to achieve glucose homeostasis or in order to reduce the incidence
and/or severity of systemic hyperinsulinemia associated with chronic dosing of insulin. It is
believed that the present invention also provides methods for reducing the incidence and/or
severity of one or more disease states associated with chronic dosing of insulin; for
prophylactically sparing b-cell function or for preventing b-cell death or dysfunction in a
subject which has impaired glucose tolerance or early stage diabetes mellitus; and for longterm
protection from developing overt or insulin dependent diabetes, or for delaying the onset
of overt or insulin dependent diabetes, in a mammal which has impaired glucose tolerance or
early stage diabetes.
The term "diabetes mellitus" refers to a group of metabolic diseases characterised
by the onset of chronic hyperglycem ia. Diabetes mellitus has been classified into three main
groups:
1. Type I diabetes (previously referred to as insulin dependent or juvenile-onset
diabetes) results from an absolute insulin deficiency arising from autoimmune
destruction of insulin-secreting b-cells in the pancreas.
2 . Type II diabetes is also known as non-insulin dependent diabetes, and manifests as
a result of insulin resistance and a relative insulin deficiency. It is usually associated
with obesity and onset is more common later in life. Although insulin therapy is not
essential for those suffering type II diabetes in its initial stages, patients may
eventually require treatment as their disease progresses.
3 . Gestational diabetes is usually temporary, and is defined as any degree of glucose
intolerance arising during pregnancy.
The term "metabolic syndrome" refers to a group of risk factors that raise your risk
for heart disease and other health problems, such as diabetes and stroke. These risk factors
include abdominal obesity, a high triglyceride level, a low HDL cholesterol level, high blood
pressure and high fasting blood sugar.
Preferred diseases for which the dicarba insulin analogues are used include
diabetes mellitus and metabolic syndrome.
The term "subject" refers to any animal having a disease which requires treatment
with the dicarba analogues of insulin of the present invention. In a preferred embodiment,
subjects are mammals, preferably humans. Examples of mammals which can be treated
using the dicarba analogues of insulin, compositions and methods of the present invention
include cows, sheep, goats, horses, dogs, cats, guinea pigs, rats or other bovine, ovine,
equine, canine, feline, rodent or murine species. However, the invention can also be
practiced in other species, such as avian species (e.g. , chickens).
The term "therapeutically active dicarba analogue of insulin" or "therapeutic dicarba
analogue of insulin" or "therapeutic dicarba insulin" as used herein refers to a dicarba
analogue of insulin able to treat (including cure, alleviate or partially arrest) the clinical
manifestations of diabetes and/or hyperglycemia and the complications therefrom .
The "treatment" or "treating" of a disease or condition as used herein means the
management and care of a patient having developed the disease, condition or disorder. The
purpose of treatment is to combat the disease, condition or disorder. Treatment includes the
administration of a therapeutically active dicarba insulin to eliminate or control (e.g. inhibit or
arrest development of) the disease, condition or disorder as well as to alleviate, relieve or
ameliorate the symptoms, complications or effects associated with the disease, condition or
disorder, and prevention of the disease, condition or disorder. The term "prevention of a
disease" as used herein is defined as the management and care of an individual at risk of
developing the disease prior to the clinical onset of the disease. The purpose of prevention is
to combat the development of the disease, condition or disorder, and includes the
administration of the active compounds to prevent or delay the onset of the symptoms or
complications and to prevent or delay the development of related diseases, conditions or
disorders.
The term "administering" should be understood to mean providing a therapeutic
dicarba insulin of the invention to a subject in need of treatment.
Dose, Dose Regimes and Mode of Administration
The dicarba analogues of insulin of the invention may be administered by any
suitable means. Examples of suitable routes of administration include parenterally, such as
by subcutaneous, percutaneous, intravenous, intra-arterial, intramuscular, intra(trans)demnal,
or intracisternal injection or infusion techniques (e.g., as sterile injectable aqueous or nonaqueous
solutions or suspensions); nasally such as by inhalation spray or insufflation;
topically, such as in the form of a cream or ointment ocularly in the form of a solution or
suspension; and orally, in dosage unit formulations containing non-toxic, pharmaceutically
acceptable vehicles or diluents. The compounds may, for example, be administered in a form
suitable for immediate release or extended release. Immediate release or extended release
may be achieved by the use of suitable pharmaceutical compositions comprising the certain
dicarba analogues of insulin, or, by the use of devices such as subcutaneous implants or
osmotic pumps.
In the treatment or prevention of conditions which require the administration of
insulin, an appropriate dosage level will generally be about 0.01 to 500 mg per kg patient
body weight per day which can be administered in single or multiple doses. Preferably, the
dosage level will be about 0 .1 to about 250 mg/kg per day; more preferably about 0.5 to
about 100 mg/kg per day. A suitable dosage level may be about 0.01 to 250 mg/kg per day,
about 0.05 to 100 mg/kg per day, or about 0 .1 to 50 mg/kg per day. Within this range the
dosage may be 0.05 to 0.5, 0.5 to 5 or 5 to 50 mg/kg per day. The dosage may be selected,
for example to any dose within any of these ranges, for therapeutic efficacy and/or
symptomatic adjustment of the dosage to the patient to be treated. The compounds will
preferably be adm inistered as necessary, and preferably on a regimen of 1 to 5 times per
day, preferably once or twice per day.
Each unit dosage will suitably contain from 0 .1 mg to 300 mg therapeutic dicarba
insulin. In one embodiment each unit dosage contains from 0.5 mg to 100 mg of therapeutic
dicarba insulin. In a further embodiment a unit dosage contains from 1 mg to 50 mg of
dicarba insulin. In a further embodiment a unit dosage contains from 1.5 mg to 30 mg of
dicarba insulin. Such unit dosage forms are suitable for administration 1-5 times daily
depending upon the particular purpose of therapy.
It will be understood that the specific dose level and frequency of dosage for any
particular subject may be varied and will depend upon a variety of factors including the
activity of the specific compound employed, the metabolic stability and length of action of that
compound, the age, body weight, general health, sex, diet, mode and time of administration,
rate of excretion, drug combination, the severity of the particular condition, and the host
undergoing therapy.
In one embodiment, the dicarba analogues of insulin are administered by
subcutaneous injection. The pharmaceutical compositions containing the dicarba analogues
of the present invention may be in a form suitable for any of the routes of administration
described above, by inclusion of suitable pharmaceutically acceptable excipients.
The pharmaceutical compositions may be in the form of a sterile injectable aqueous
or oleagenous suspension. This suspension may be formulated according to the known art
using those suitable dispersing or wetting agents and suspending agents which have been
mentioned above. The sterile injectable preparation may also be a sterile injectable solution
or suspension in a non-toxic parenterally-acceptable diluent or solvent, for example as a
solution in 1,3-butane diol. Among the acceptable vehicles and solvents that may be
employed are water, Ringer's solution and isotonic sodium chloride solution. In addition,
sterile, fixed oils are conventionally employed as a solvent or suspending medium . For this
purpose any bland fixed oil may be employed including synthetic mono- or diglycerides. In
addition, fatty acids such as oleic acid find use in the preparation of injectable formulations.
For topical use, creams, ointments, jellies, solutions or suspensions, etc. , containing
the compounds of the present invention are employed. (For purposes of this application,
topical application shall include mouthwashes and gargles.)
For application to the eye, the active compound may be in the form of a solution or
suspension in a suitable sterile aqueous or non-aqueous vehicle. Additives, for instance
buffers, preservatives including bactericidal and fungicidal agents, such as phenyl mercuric
acetate or nitrate, benzalkonium chloride, or chlorohexidine and thickening agents such as
hypromellose may also be included.
The dicarba insulin analogues of the present invention can also be administered in
the form of liposomes. As is known in the art, liposomes are generally derived from
phospholipids or other lipid substances. Liposomes are formed by mono- or multilamellar
hydrated liquid crystals that are dispersed in an aqueous medium. Any non-toxic,
physiologically acceptable and metabolisable lipid capable of forming liposomes can be
used. The present compositions in liposome form can contain, in addition to the dicarba
insulin analogue of the present invention, stabilisers, preservatives, excipients and the like.
The preferred lipids are the phospholipids and phosphatidyl cholines, both natural and
synthetic. Methods to form liposomes are known in the art.
A particularly preferred mode of administration is oral administration.
The dicarba analogues of the present invention may also be presented for use in the
form of veterinary compositions, which may be prepared, for example, by methods that are
conventional in the art. Examples of such veterinary compositions include those adapted for:
• parenteral administration for example by subcutaneous, intramuscular or
intravenous injection, e.g. as a sterile solution or suspension; or (when appropriate)
by intramammary injection where a suspension or solution is introduced in the udder
via the teat;
• topical applications, e.g. as a cream, ointment or spray applied to the skin.
• oral
The term "therapeutically effective amount" refers to the amount of the dicarba
insulin that will elicit the biological or medical response of a tissue, system, animal or human
that is being sought by the researcher, veterinarian, medical doctor or other clinician.
In a preferred embodiment, administration of the pharmaceutical composition takes
place multiple times daily, preferably at bedtime and preprandially during the day time, e.g.,
preprandially for breakfast, lunch and dinner. More preferably, administration of the
pharmaceutical formulation is at or shortly prior to bedtime and concurrently with or shortly
prior to ingestion of a meal, i.e., within about 15 minutes or less of ingestion of the meal.
In another preferred embodiment of the invention, the oral pharmaceutical
composition formulation will be administered about once daily to about four times daily,
preprandially and/or at bedtime, depending upon the extent of the patient's impaired glucose
tolerance and need for exogenous glycemic control. If the patient has a need for fasting
glycemic control, the oral pharmaceutical formulation will be adm inistered only at or shortly
prior to bedtime. If the subject has a need for post-prandial glycemic control, the oral
pharmaceutical formulation will be administered preprandially for all meals. If the subject has
a need for comprehensive glycemic control, the oral pharmaceutical formulation will be
administered preprandially for all meals and at or shortly prior to bedtime.
Preferably, the dosage form of the present invention will be administered for at least
one day, more preferably on a chronic basis, and can be administered for the life of the
subject. Most preferably, the dosage form of the present invention will be administered on a
chronic basis, e.g., for at least about two weeks.
Preferably, the therapeutic dicarba insulin treatment of the present invention will be
administered to subjects having some form of impaired glucose tolerance. This can range
from insulin resistance seen in pre-diabetics and early stage Type 2 diabetics to failure of
insulin production by the pancreas seen in Type 1 diabetes and late stage Type 2 Diabetes.
In certain embodiments, the resulting improved insulin utilization or insulin sensitivity of the
subject's body is measured by HOMA (Homeostasis Model Assessment). In certain
embodiments, the resulting improved insulin secretion capacity of the subject's body is
measured by Stumvoll first-phase insulin secretion capacity index.
Further, the therapeutic dicarba insulin treatment of the present invention can be
administered to a mammal with an HbAi C ranging from normal to elevated levels. More
particularly, the treatment can be administered to anyone in the range of normal glycemic
control to impaired glycemic control to late stage type 2 diabetes or type 1 diabetes. In
certain embodiments, the resulting improved glycemic control in the subject's body is
measured by a reduced serum fructosamine concentration. Preferably the average decline
will be about 8.8% after at least two weeks of treatment with the present invention.
In preferred embodiments of the oral dosage forms of the invention described
above, the oral dosage form is solid, and is preferably provided incorporated within a gelatin
capsule or is contained in a tablet.
In certain preferred embodiments, the dose of the therapeutic dicarba insulin
contained in one or more dosage forms is from about 50 Units to about 600 Units (from about
2 to about 23 mg), preferably from about 100 Units (3.8 mg) to about 450 Units ( 15.3 mg)
insulin, more preferably from about 200 Units (7.66 mg) to about 350 Units ( 13.4 mg), and
still more preferably about 300 Units (11.5 mg), based on the accepted conversion of factor
of 26. 11 Units per mg.
In certain preferred embodiments of the invention, the dosage forms begin delivering
the dicarba insulin into the systemic circulation via the portal vein (via absorption through the
mucosa of the gastrointestinal tract) to achieve peak levels within about 30 minutes or less.
In certain preferred embodiments, the dosage forms of the invention provide a tmax
for insulin at from about 5 minutes to about 30 minutes, and more preferably at from about 10
minutes to about 25 minutes after oral administration to diabetic subjects. In certain preferred
embodiments of the invention, the dosage forms begin delivering the dicarba insulin into the
systemic circulation to produce a peak plasma insulin concentration at about 10 to about 20
minutes post oral administration and in further preferred embodiments, a peak plasma insulin
concentration at about 10 minutes to about 15 minutes post oral administration to subjects
who ingested the dosage at about 0 or about 10 minutes prior to ingestion of a meal.
The invention is also directed in part to an oral dosage form comprising a dose of
therapeutic analogue of insulin that achieves a therapeutically effective control of post
prandial blood glucose after oral administration to human diabetic subjects in tablet form at or
shortly before mealtime, the oral solid dosage form providing an insulin tmax at a time point
from about 10 minutes to about 15 minutes after oral administration to said subjects, at least
about 30% of the blood glucose concentration reduction caused by said dose of insulin
occurring within about less than 1 hour after oral administration of said dosage form. In
preferred embodiments of this invention, the oral dosage form is a tablet.
In certain preferred embodiments, the composition provides a tmax for maximum
control of glucose excursion at about 0.25 to about 1.5 hours, more preferably at about 0.75
to about 1.25 hours, after oral administration. In certain preferred embodiments, the tmax for
post-prandial glucose control occurs preferably at less than about 120 minutes, more
preferably at less than about 80 minutes, and still more preferably at about 45 minutes to
about 60 minutes, after oral administration of the composition.
Because insulin entry into the bloodstream produces a decrease in blood glucose
levels, oral absorption of the therapeutic dicarba insulin may be verified by observing the
effect on a subject's blood glucose following oral administration of the composition. In a
preferred embodiment of the invention, the oral dosage forms of the invention facilitate the
oral delivery of therapeutic dicarba insulin, and after the therapeutic dicarba insulin is
absorbed into the bloodstream , the composition produces a maximal decrease in blood
glucose in treated type 2 diabetic subjects from about 5 to about 60 minutes after oral
administration. In another embodiment of the present invention, the pharmaceutical
composition produces a maximal decrease in blood glucose in treated type 2 diabetic
subjects from about 10 to about 50 minutes post oral administration. More particularly, the
pharmaceutical composition produces a maximal decrease in blood glucose in treated type 2
diabetic subjects within about 20 to about 40 minutes after oral administration.
The magnitude of the decrease in blood glucose produced by the therapeutic
dicarba insulin absorbed into the bloodstream following entry into the gastrointestinal tract
varies with the dose of therapeutic dicarba insulin. In certain embodiments of the invention,
type 2 diabetic diabetic patients show a maximal decrease in blood glucose by at least 10%
within one hour post oral administration. In another embodiment, type 2 diabetic diabetic
patients show a maximal decrease in blood glucose by at least 20% within one hour post oral
administration, alternatively, at least 30% within one hour post oral administration.

CLAIMS:
1. A dicarba analogue of insulin comprising an A-chain and a B-chain or fragments,
salts, solvates, derivatives, isomers or tautomers of the A-chain, the B-chain or both,
provided that the dicarba analogue is not [A7, B7-(2,7-diaminosuberoyl]-des-(B26-
B30)-insulin B25-amide.
2 . The dicarba analogue according to claim 1, comprising one or more intrachain
dicarba bridges located on the A-chain, the B-chain or both.
3 . The dicarba analogue according to claim 1, comprising one or more interchain
dicarba bridges located between the A-chain and the B-chain.
4 . The dicarba analogue according to claim 1, comprising one or more unsaturated
dicarba bridges.
5 . The dicarba analogue according to claim 1, wherein one or more of the disulfide
bridge forming cysteine amino acid residue pairs of native insulin are replaced by a
dicarba bridge.
6 . The dicarba analogue according to claim 2 , wherein the one or more intrachain
dicarba bridges are located in the A-chain.
7 . The dicarba analogue of insulin according to claim 1, comprising a dicarba bridge
which includes at least one of the groups selected from -C-C-, -C=C- and - CºC-
8 . The dicarba analogue according to claim 7 , wherein the dicarba bridge is selected
from the following:
wherein R to R6 are each independently absent or selected from a divalent linking
group.
9 . The dicarba analogue according to claim 8 , wherein the dicarba bridge has the
formula ( I II) or (IV) and the -C=C- group is in a c/s-conformation or a transconformation.
10 . The dicarba analogue according to claim 1, wherein the dicarba bridge is formed via
metathesis of two complementary metathesisable groups.
11. The dicarba analogue according to claim 10, wherein the metathesisable groups are
each connected to an amino acid in insulin.
12 . The dicarba analogue according to claim 11, wherein the metathesisable groups are
located on the amino group or on the side chain of the amino acid.
13 . The dicarba analogue according to claim 11 or 12 , wherein the amino acid is
selected from the group consisting of glycine or alanine.
14 . The dicarba analogue according to claim 13 , wherein the amino acid having the
metathesisable group is selected from the group consisting of allylglycine,
crotylglycine, prenylglycine and butynylglycine.
15 . The dicarba analogue of insulin according to claim 1, wherein the dicarba analogue
comprises one or more dicarba bridges which replace at least one of the disulfide
bridges located between cysteine residue 6 and cysteine residue 11 of the A-chain
of native human insulin, cysteine residue 7 of the A-chain and cysteine residue 7 of
the B-chain of native human insulin, or cysteine residue 20 of the A-chain and
cysteine residue 19 of the B-chain of native human insulin.
The dicarba analogue according to claim 1, having a stability in human blood
plasma greater than that of the corresponding insulin not containing at least one
dicarba bridge after 6 hours of contact with human blood plasma.
The dicarba analogue according to claim 1, having a stability at room temperature
that is greater than that of the corresponding insulin not containing at least one
dicarba bridge.
A method for the synthesis of a dicarba analogue of insulin comprising an A-chain
and a B-chain or fragments, salts, solvates, derivatives, isomers or tautomers of the
A-chain, the B-chain or both, the method comprising:
(i) providing the A-chain having at least one pair of complementary
metathesisable groups;
(ii) subjecting the A-chain to metathesis to form at least one dicarba bridge
between the pair of complementary metathesisable groups; and
(iii) adding the B-chain.
The method of claim 18 , which further comprises the step of hydrogenating the
dicarba bridge(s) to form at least one alkane-containing dicarba bridge either before
or after the B-chain is added.
The method of claim 18 , which further comprises the step of sem i-hydrogenating the
dicarba bridge(s) to form at least one alkene-containing dicarba bridge either before
or after the B-chain is added.
The method of claim 20 wherein the semi-hydrogenation step forms an alkenecontaining
dicarba bridge which is enriched in the cis- or frans-isomer.
The method of claim 18 , wherein the A-chain comprises a turn inducing residue
located between the at least two complementary metathesisable groups.
A method for the synthesis of a dicarba analogue of insulin comprising an A-chain
and a B-chain or a fragments, salts, solvates, derivatives, isomers or tautomers
thereof the A-chain, the B-chain or both, the method comprising:
(i) providing a part of the A-chain having at least two complementary
metathesisable groups;
(ii) subjecting the A-chain to metathesis to form at least one dicarba bridge;
(iii) adding one or more further amino acids to one or both ends of the A-chain;
and
(iv) adding the B-chain.
A method for the synthesis of a dicarba analogue of insulin comprising an A-chain
and a B-chain or fragments, salts, solvates, derivatives, isomers or tautomers of the
A-chain, the B-chain or both, the method comprising:
(i) providing a part of the A-chain and/or a part of the B-chain having at least
two complementary metathesisable groups between them;
(ii) subjecting the A-chain and B-chain to metathesis to form at least one
dicarba bridge; and
(iii) adding one or more further amino acids to one or both ends of the A-chain
and/or B-chain.
The method of claim 23 or 24, wherein the A-chain or B-chain or both comprise
further complementary metathesisable groups, and the method includes a further
metathesis step to form at least one further intrachain dicarba bridge, at least one
interchain dicarba bridge or both between the complementary metathesisable
groups.
A method for the synthesis of a dicarba analogue of insulin comprising an A-chain
and a B-chain or fragments, salts, solvates, derivatives, isomers or tautomers of the
A-chain, the B-chain or both, the method comprising:
(i) providing a readable peptide comprising a removable tether between the
A-chain and the B-chain, the A-chain and the B-chain each having at least
one complementary metathesisable group; and
(ii) subjecting the reactable peptide to metathesis to form at least one dicarba
bridge between the complementary metathesisable groups; and
(iii) removing the removeable tether to produce a dicarba bridge linking the Achain
and the B-chain of insulin.
The method of claim 26, wherein the peptide comprises further complementary
metathesisable groups located on the A-chain or B-chain or both, and the reactable
peptide is subjected to a second metathesis step either before or after the step of
removing the removeable tether, to form at least one further intrachain dicarba
bridge, at least one interchain dicarba bridge or both between the complementary
metathesisable groups.
The method of claim 23, 24 or 26, which further comprises forming at least one
disulfide bridge between the cysteine amino acid residues in the A-chain and/or Bchain.
The method of claim 23, 24 or 26, which further comprises the step of hydrogenating
the dicarba bridge(s) to form at least one alkane-containing dicarba bridge either
before or after the step of removing the removeable tether.
The method of claim 23, 24 or 26, which further comprises the step of semihydrogenating
the dicarba bridge(s) to form at least one alkene-containing dicarba
bridge either before or after the step of removing the removeable tether.
The method of claim 28, wherein the semi-hydrogenation step forms an alkenecontaining
dicarba bridge which is enriched in the cis- or frans-isomer.
The method of any one of claims 18 , 23, 24 or 26, wherein the cross-metathesis of
the complementary metathesisable groups is conducted under microwave radiation
conditions.
The method of any one of claims 18, 23, 24 or 26, wherein the cross-metathesis is
performed using a cross-metathesis catalyst in the presence of a solvent
combination comprising a resin-swelling solvent and a co-ordinating solvent for the
catalyst.
The method of claim 33, wherein the co-ordinating solvent is an alcohol.
35. A dicarba analogue of an A-chain of insulin or a fragment, salt, solvate, derivative,
isomer or tautomer thereof.
A method for the synthesis of a dicarba analogue of an A-chain of insulin or a
fragment, salt, solvate, derivative, isomer or tautomer thereof, the method
comprising:
(i) providing the A-chain having at least one pair of complementary
metathesisable groups; and
(ii) subjecting the A-chain to metathesis to form at least one dicarba bridge
between the pair of complementary metathesisable groups.
An anti-hyperglycemic agent comprising a dicarba analogue of insulin or a fragment,
salt, solvate, derivative, isomer or tautomer thereof according to claim 1.
A pharmaceutical composition comprising a dicarba analogue of insulin or a
fragment, salt, solvate, derivative, isomer or tautomer thereof according to claim 1
and a pharmaceutically acceptable carrier.
A dicarba analogue of insulin or a fragment, salt, solvate, derivative, isomer or
tautomer thereof according to claim 1, for use in treatment of hyperglycemia,
diabetes mellitus or metabolic syndrome.
Use of a dicarba analogue of insulin according to claim 1, for the treatment of
hyperglycemia, diabetes mellitus or metabolic syndrome.
Use of a dicarba analogue of insulin or a fragment, salt, solvate, derivative, isomer
or tautomer thereof according to claim 1, in the manufacture of a medicament for the
treatment of hyperglycemia, diabetes mellitus or metabolic syndrome.
A method for reducing hyperglycemia comprising, administering a dicarba analogue
of insulin or a fragment, salt, solvate, derivative, isomer or tautomer thereof
according to claim 1 to a subject in need thereof.
A method for the treatment of diabetes mellitus or metabolic syndrome comprising,
administering a dicarba analogue of insulin or a fragment, salt, solvate, derivative,
isomer or tautomer thereof according to claim 1 to a subject in need thereof.