peptide helix drug development

Peplix has developed new technology for making short stable peptide helix mimetics.  The alpha helix is the most common regular structure in proteins and is a key part of many biological interactions.  Critically these interactions are involved in diseases including cancer, diabetes, arthritis, osteoporosis and viral infections.  A helix mimetic is a modified version of a helix and is potentially useful as a drug.

Why make helix mimetics?

There are many important drug targets that involve single helical domains – several are illustrated in the discussion below.  These helical targets have not been effectively addressed with traditional small molecule chemistry or with antibody technology.  As a result a significant opportunity currently exists to develop new treatments for a range of diseases by using helix mimetic technology.

Compared to traditional drug targets helix interactions cover a larger physical area.  This makes them difficult to develop drugs for using traditional small molecule methods.  One way of developing drugs for larger targets is to make antibodies (or modified versions of them).  While this can work for extracellular targets it is not feasible for intracellular targets as the antibodies cannot effectively cross the cell membrane.  Many of the most important targets for helix mimetics are found within the cell and these interactions therefore represent a class of targets that are difficult for both traditional small molecule drug development and for antibody biotechnology.

The image to the left is a model of a helical peptide hormone.  The ribbon is not part of the molecule but follows the backbone of the molecule illustrating the helical (or spiral) shape.

About the Technology

The peplix technology has a number of attractive features:

-  based on new chemistry with a strong intellectual property position
-  highly flexible placement in the peptide sequence
-  lower impact than competing technologies on the existing helix

Below we describe some targets for helix mimetics followed by a brief discussion of methods for making mimetics.

Examples of areas where helix mimetics can be useful are listed in the table below.

The illustrations below show helices in a range of significant biological interactions.  All the structures are of real (rather than modeled) interactions derived from protein data bank structures as referenced.  The main proteins are displayed with low detail in order to highlight the interacting helix.

A fragment of the tumor suppressor protein p53 (white) bound to protein mdm2.  Most cancers suppress the action of p53, reactivating it could be useful in cancer treatment.

The same p53/mdm2 interaction viewed down the helical axis.  The key interacting parts have been highlighted in yellow. pdb: 1ycr

The peptide hormone Glp-1 (white) bound to the extracellular part of its receptor (grey).  Modified versions of this hormone are approved for the treatment of diabetes (liraglutide, byetta).  The right hand end of the hormone binds to the transmembrane part of the receptor (not present in the crystal structure - transmembrane domains are difficult to crystallise). pdb: 3iol

The peptide Bim (white) bound to Bcl-2 (grey).  This interaction is involved in the control of programmed cell death (apoptosis).  Blocking the action of Bcl family proteins has been shown to be useful in cancer therapy, this is an emerging field with good potential.    pdb: 3fdl

A 12 residue peptide inhibitor of HIV capsid assembly bound to a part of the HIV capsid protein.
pdb: 2buo

A helical peptide (Rev-ARM, yellow) bound to an RNA molecule (stem loop IIB), the highest affinity binding site on Rev-responsive element (RRE).  This interaction is a target for HIV(AIDS). pdb:1etg.

About the Development of Helix Mimetics

Approaches to making helix mimetics

The above images are 2D renderings of a short section of alpha helix.  Structure 1 shows a normal helix with the three stabilising hydrogen bonds highlighted in red.  The pattern of hydrogen bonds is what stabilises (and defines) an alpha helix.  There are a number of approaches to making helix mimetics.  One popular approach is to join two of the peptide sidechains together - this is illustrated in structure 2 above with the highlighted link between the R and R4 positions (these positions are closest together on consecutive loops of the helix).  This approach has been successfully used by the company Aileron Therapeutics producing what they call stapled peptides.

One attraction of the approach is that normal amino acid side-chains can be used to form the link - the most successful method of doing this is joining a lysine at the R position by amide bond to an aspartic acid at the R4 position.  This "K->D" approach has been known since at least 1996 (review) and has more recently shown some quite impressive results (pnas 2010).  Drawbacks are that the constraint is indirect, two sidechains are needed and the patent position is weak (since the general approach is in the public domain).

Structure 3 above illustrates another approach to helix stabilisation where one of the hydrogen bonds (which are very weak) is replaced by a chemical link.  The link is called a covalent hydrogen bond mimic and the approach was first proposed in 1987.

In practice making covalent hydrogen bond mimics (or hydrogen bond surrogates, "HBS") is difficult.  Structure 4 shows an attempt at creating an HBS mimetic based on structure 3.  The difficulty of synthesis has meant that the lower part of the helix has been left out, all of the sidechains except R4 have been left out, and a methyl group (highlighted in blue) had to be added to aid cyclisation.  This hydrazone mimic (structures 3 and 4) has not seen much practical application.  The HBS 5 is similar to 4 as it can only be placed at the end of the helix and does not include the R, R1 or R2 groups.  Despite these limitations it has seen some successful application.

Structure 6 is the first (and currently only) example of an HBS that can and has been applied within a peptide rather than at the end of one.  As illustrated only the R1 position is missing.  The development of 6 was carried out by the founder of peplix and collaborators.  The difficulty of synthesis is high and although this HBS showed some promise in model systems the difficulty of making it means it is not likely to be useful in drug development.

There are other methods of making helix mimetics - notably non-peptide scaffolds such as terphenyl systems and related, and also peptoid and beta peptide approaches.  These have been generally less successful than the more peptide based methods described above - the main reason is that the jump from the natural helix to non-peptides/peptoids is greater and tends to suffer from a correspondingly greater loss of activity which has proven difficult to recover by optimisation.