1.0) The therapeutic area: Cancer. In the UK cancer is one of the three major causes of death.1 It costs the English economy £18.33 billion annually when healthcare costs, loss in productivity, and the costs to patients and their families are taken into account.2 These costs are set to rise in the future.3 Cancer research takes 25% of the UK Government's medical sciences research budget.4 Cancers are typified by a breakdown in the cellular controls that regulate the cell cycle resulting in uncontrolled cell division to produce tumours.5
Tumour cells are characterised by their enhanced and relentless progression through the cell cycle. This means the usual cellular processes taking place at a greater rate. This is what provides the targets for many cancer drugs currently in use allowing their action on tumour cells over and above normal healthy cells. For example Pemetrexed (folate-dependent enzymes), and Cytarabine (purine/pyrimidine pathway) all inhibit enzymes that play a role in the metabolic pathways.6 Other essential processes that are overly enhanced in diseased cells make attractive targets. In this work we look at targeting the proteasome, a protein complex that plays an essential role in the regulation of proteins at the post-translational level and the removal of damaged or mis-folded proteins.
1.1) Development of a new Proteasome inhibitor
The inhibition of the proteasome is a validated therapeutic target in many cancers.7 Carcinoma cell lines show increased sensitivity to apoptosis when the ubiquitin-proteasome system is inhibited.8 The proteasome is responsible for the controlled degradation of proteins in cells. Proteins that are damaged through heat shock or oxidative attack or simply mis-folded require removal from the cytostol and destruction before they cause problems. The protease breaks proteins down into shorter peptides of 4 or more amino acids in length that can then be recycled or, as in the case of certain transcription factors, be activated through being small enough to translocate. Protein destruction also acts as a regulatory pathway as when a proteins function is no longer required it will be tagged for specific destruction. In the case of the proteasome, proteins are labelled with the small molecule ubiquitin. This triggers further ubiquitination to generate the polyubiquitin tag that the proteasome recognises and brings the tagged protein into a position to undergo proteolysis. The other main cellular pathway to remove unwanted or damaged proteins is that of lysosomal degradation. The lysosome contains many cysteine, aspartate and a zinc protease that are able to destroy proteins. The lysosome is characterised by its low internal pH, an environment that aids protein destruction.
1.1.1) Structure of the proteasome
The proteolytic process as carried out by the proteasome is a highly controlled reaction with a series of steps to ensure that only the intended proteins are broken down. A key feature of the proteasome itself is its structure that represents a hollow barrel with the active sites being located well inside the central core.9 Two outer rings act as gates that only allow unfolded proteins to enter the catalytic chamber and only proteins exhibiting polyubiquitin tags are bound and positioned to enter the structure. These specific polyubiquitin binding sites are found on the regulatory caps (the 19S subunits). These subunits consist of the ubiquitin binding nine-protein lid joined to a ten-protein base that in turn binds to the core catalytic structure of the proteasome. The 19S subunit contains isopeptidase activity that allows the removal of the polyubiquitin tag from the target protein prior to its entrance to the catalytic core.10 This allows the recycling of the ubiquitin tags.
The central core of the proteasome, the 20S unit, consists of four stacked protein rings of which the inner two contain three protease active sites. The 20S subunit contains outer structural alpha subunits and inner catalytic beta units. The outer alpha structural units act as docking domains, recognising the regulatory 19S unit. The N-termini of these alpha units form a gate that blocks the access to the catalytic core meaning that only partially unfolded, de-ubiquitinated proteins can enter. Of the beta subunits only the beta 1, 2 and 5 subunits show proteolytic activity and these show varying substrate specificity, ranging from chymotrypsin-like, trypsin-like and caspase-like.11 Each of the three sites mimics those found in the individual proteases i.e. trypsin like means that the peptide is cut adjacent to a basic residue, and chymotrypsin-like means it is cut next to a hydrophobic amino acid. The substrate specificity pocket that binds and positions the substrates peptide bond for cutting dictates this.
C:\Users\default.default-PC\Documents\Brief 346050\Yeast 20S proteasome.bmp
Figure 1.1 The x-ray crystal structure of the yeast 20S subunit of the proteasome (PDB file).11 The protein is depicted in ribbon view showing its barrel type shape. The 19S regulatory unit would sit on the top and bottom of the barrel restricting entrance to the active site that is located inside the core.
1.1.2) The role of ATP
The cofactor ATP plays several key regulatory roles in the function of the protease. Substrates are unfolded in an ATP dependent process as the energy required is released through ATP hydrolysis. The exact process involves interactions with the 19S subunit. The binding of the 19S to the 20S is dependent on the binding of ATP and this complexation of the two subunits in an ATP dependent manner acts as another control point in the activity of the proteasome. As the 19S regulatory ATPase binds by the docking of its C-terminus into specific sites on the top of the 20S unit this interaction opens the gate to the proteolytic core.
1.1.3) Polyubiquitination
Polyubiquitination is a vital part of the proteolytic process. The ubiquitin molecule is a small protein of 76 amino acids in length. It is activated by adenylation and it is this activated molecule that is used to covalently modify a target proteins lysine residue. A number of enzymes are involved in the process; a ubiquitin activating enzyme, a ubiquitin conjugating enzyme and lastly the ubiquitin ligase that recognises the protein to be destroyed. The process is known as the ubiquitylation cascade. The ubiquitin activating enzyme forms a ubiquityl-AMP complex and it is this modified protein that recognises specific protein partners. The activated ubiquitin molecule is passed to the partner conjugating enzyme via transthiolation. Whilst some ubiquitin conjugating enzymes are able to interact directly with their target protein others form a complex with an adapter ligase. Polyubiquitination (a minimum of four ubiquitin molecules) is required for the protein to be recognised by the 19S regulatory particle of the lid.
1.1.4) Catalytic function of the Proteasome
The catalytic step in proteolysis occurs via nucleophilic attack of the proteasome's catalytic threonine residue (Thr1) on the substrate's peptide bond to generate short peptides.12 A lysine residue has been found to be crucial to activity as, being charged at physiological pH, it lowers the pKa of the backbone amino group of the threonine allowing it to function as a proton acceptor. The hydroxyl oxygen of the threonine attacks the electrophilic group of the peptide bond (or inhibitor). A water molecule is necessary to mediate proton transfer between the hydroxyl oxygen and the amino group of the threonine during substrate binding to produce the acyl-ester intermediate. Water is also involved in the subsequent hydrolysis of the acyl-ester bond to regenerate the free threonine residue.
The protease has been increasingly studied as a target for anti-cancer therapies. Inhibition of the proteasome can result in cellular stress through protein aggregate accumulation13, endoplasmic reticular stress and unfolded protein response. Of increasing interest is the effect of proteasome inhibition on the levels of other proteins that play crucial roles in the control and regulation of the cell cycle. Proteins such as the cyclins and their associated kinases are necessary regulators of the cell cycle and, as such fundamentally important proteins, they are highly controlled and regulated.
1.2) Cyclin Kinase Inhibitors (CKIs)
One such family of regulatory proteins are the cyclin kinase inhibitors. These important proteins in the mitotic process have short life-times and this is (in part) regulated post translationally by their proteasome mediated destruction.
1.2.1) p27 Tumour Suppressing Protein
The cyclin kinase inhibitor protein p27 has been identified as a key protein in the progression of cancer cells. p27 has a tumour suppressing function due to its inhibition of specific cyclin-cyclin dependent kinase (CDK) complexes [20]. CDKs are only functional when in complex with their partner cyclin and these cyclin-CDK complexes are essential for progression through the cell cycle. These complexes carry out the phosphorylation (and thereby regulation) of key proteins appropriate for specific stages in the cell cycle of a particular cell. As a cyclin kinase inhibitor p27 acts in the nucleus of cells enforcing the cell cycle checkpoints through binding to specific cyclin-CDK complexes. It has two important functions; the first is promotion of G1 progression (the gap phase after mitosis that allows cell growth).15 Its second role is to inhibit the function of cyclin E and cyclin A-bound CDK2 complexes that prevents entry into S-phase. As a result p27 is influential in sustaining the G1 phase of cell growth and preventing cell cycle progression from phase G1 to the S phase. Its continued action causes cell division to arrest at this point.
In many human cancers the levels of p27 is found to be reduced due to its proteolysis by the ubiquitin-proteasome pathway. In almost all tumours its expression and/or functions are altered. In the normal cell the expression level of the p27 will determine its activity. Growth factors and cytokines can stimulate or repress p27 expression16, and its destruction via the proteasome, controls its presence post translationally.17 A further mechanism by which its activity is controlled is that of cellular location as its actions are directed on proteins found in the nucleus. Many anti-cancer agents have been found to up regulate the expression of p27.18 This up regulation can occur at the transcriptional, translational or post-translational level. Studies using prostate cancer cells have looked at small molecules that can increase the levels of p27 in the cell nucleus.19 Compounds were identified as increasing the nuclear levels of p27 when dosed in the micomolar range and induced G1 delay and stalled growth.19 p27 also affects cell motility, a function that is regulated by its phosphorylation.20 This phosphorylation plays a role in the stability of the protein, reducing the proteins degradation.
The removal of p27 is through the ubiquitin-proteasome pathway. In this process p27 is phosphorylated and targeted by a protein called Skp2, a nuclear F-box protein. The Skp2 forms a complex with SCF that functions as the ubiquitin ligase. This complex polyubiquitinates p27.21 Chemicals and proteins that interfere with either Skp2 or SCF or disrupt any of the protein interactions involved will have an effect on the levels of p27 in the cell. For example, the tumour suppressing protein tuberin binds directly to p27 and prevents its Skp2 mediated degradation.22 Rapamycin mediated reduction of Skp2 has been shown to cause cell growth arrest at the G1 checkpoint due to the maintenance of intra-nuclear p27 levels.23 Phosphorylation of p27 by viruses results in p27 being targeted for destruction.24
As mentioned in above section, the proteasome is influential in many apoptotic pathways. This is therefore a validated anti-tumour target as its inhibition has been shown to be key in pro-apoptotic changes to the cell either through the accumulation of tumour suppressing proteins such as p27 that signal for cell death, or via the accumulation of toxic junk protein. As such the proteasome is increasingly being used as a target for chemotherapy agents.
To aid the development of a new proteasome inhibitor, information regarding current proteasome drugs were sought, in particular their mode of inhibition and their interaction with the proteasome itself.
1.3) Current Proteasome Inhibitors
Probably the most widely used proteasome inhibitor in cancer therapy regimes is bortezomib with unmatched activity in multiple myeloma25, gastric cancer26 and hepatocellular carcinoma.27 Bortezomib was the first proteasome inhibitor to enter clinical trials and is a dipeptide boronic acid.28 Bortezomib inhibits the 20S proteasome in a time and dose dependent manner.29 By its inhibition of the proteasome it up regulates p27.30 Norvir has been shown to inhibit the chymotrypsin-like activity of the proteasome.31 Various metal binding compounds have also been shown to inhibit the proteasome.32 Bortezomib has been shown to be an effective proteasome inhibitor, but it has been associated with a number of adverse events in clinical trials33,34. It has been observed that some multiple myeloma cells have shown resistance to the proteasome inhibitor bortezomib.35
1.4) Development of a new Proteasome inhibitor
As the proteasome is proving to be an attractive and validated target in cancer therapy new inhibitors are being produced. These include the small molecule proteasome inhibitor SC68896 [49]. Other compounds are being identified and worked up in pre-clinical studies with the aim of identifying efficacious and safe cancer therapies. Lead compounds that can be developed into active drug candidates are identified in screens of multiple compounds in cell based arrays that typically look for cell death or inhibition of growth.
1.5) The Argyrins
Cyclic peptides have also been found to inhibit the proteasome by a route that does not involve modification of the active site threonine. For example, the inhibitor TMC-95 from Apiospora montagnei37, Argyrin A from the myxobacteria sp. and TP-110 from the Kitasatospora sp.38 was identified as a proteasome inhibitor with anti-tumour activities in a screen of natural products.39 Administration of argyrin A in pre-clinical models showed that it was able to inhibit the proteasome in vivo and decrease tumour growth. It was shown to be equally effective in bortezomib resistant cells and this led to mechanistic studies that have shown argyrin to function through a pathway involving raised p27 levels.39
A large number of secondary metabolites with antifungal, antibacterial and antitumour activities have been derived from the soil dwelling gram negative bacteria of the myxobacterium species. Argyrins are cyclic octa-peptides and have been found to be produced by members of the myxobacterium, Archangium gephyra and Cystobacter fuscus.40 Isolated from culture broths, the structures of argyrins A to H have been elucidated. All of the argyrins have been shown to elicit antibiotic activity against species such as Psuedomonas and varying degrees of growth inhibition in mammalian cell lines.41 For example, argyrin B has been shown to be a potent inhibitor of T cell independent antibody formation in murine B cells.
argyrin A
Figure 2 Structure of Argyrin A. The cyclic octa-peptide is made up units derived from the amino acids alanine, tryptophan and glycine. Structure taken from Stauch et al.42
Argyrin A is thought to exert its tumour suppressing effect through a mechanism that involves specific inhibition of the proteasome. Gene expression experiments indicated that the actions of argyrin A could be mimicked by genetic knock-down of the proteasome 20S subunit. Studies using mutant forms of p27 showed that its cyclin kinase inhibitor function is not involved as inactive forms produced a different cell pathology to argyrin treated cells. This is contrary to earlier thoughts drawn from the idea that bortezomib increased the activity of p27 through preventing its proteasome degradation. It is now speculated that the effects of bortezomib may be as a result of off-target effects of the drug.43
1.5.1) Argyrins as a basis for novel proteasome inhibitors
The interaction of argyrin A with the 20S unit of the proteasome has been studied by Stauch and co-workers.42 They determined the solution structure of argyrin A by NMR and then docked it to a model of the human proteasome based on the crystal structure of the yeast 20S subunit. The study gave useful indications of the possible interactions and how the binding and selectivity of the argyrins for the human proteasome may be enhanced.
solution structure arbyrinA
Figure 3 The solution structure of argyrin A as determined by NMR spectroscopy. The residues De-Ala5 and Sarc6 are bent out of the plane of the macrocyclic ring. Coventional CPK colouring is used. Picture taken from Stauch et al.42
Their model showed that argyrin A binds in the specificity pocket adjacent to the active site thereby blocking access to the catalytic threonine residue. They found an excellent steric fit of the compound in the binding pocket. The glycine and alanine residues of the octapeptide were buried within the protein whilst the more hydrophilic thiazole group interacted with the pocket wall. The backbone carbonyls of argyrin A's glycine and alanine formed hydrogen bonds with the amino group of Gly47 and the hydroxyl of Thr1 of the proteasome. Further hydrogen bonds between the protein backbone and the amino and carbonyl groups of argyrin A's tryptophan moiety strengthen the interaction. Lastly there is a polar interaction between the carbonyl of the sarcosine group in argyrin A and the positively charged N-termini of the protein. There are other interactions between argyrin A and the protein depending on which of the three active subunits (beta 1, 2 or 5) are examined, but the above interactions are common to all.
Argyrin A binding
Figure 4 Key contacts between the argyrins and the three different substrate specificity pockets of the human proteasome as deduced from the docking of the solution structure of argyrin A to the model of the human proteasome using GOLD. Shown are the caspase-like pocket (a), the trypsin-like pocket (b) and the chymotrypsin-like pocket (c). Picture taken from Stauch et al.42
Other members of the argyrin family (argyrins A to F44) were also docked to the modelled proteasome. The following observations were in good agreement with experimentally measured affinity of the different compounds for the proteasome. Argyrin B has an elongated methyl side chain on the alanine moiety by a single carbon. This was found to be well tolerated in the model as the longer side chain extended into the deep substrate binding cleft. Therefore this additional hydrophobic interaction would enhance binding. In argyrin D there is an additional methyl group on Trp2 but again this is accommodated in the large substrate binding cleft. Argyrin F has an additional hydroxyl group on the thiazole derivative of alanine (Ala-Thiaz) moiety. This was found to form an additional hydrogen bond with the backbone carbonyl of the proteasome's Gly23 residue thus strengthening its interaction. Argyrin E lacks the methoxy group on the tryptophan side-chain. This group has been shown to be essential for activity in cell assays and the docking illustrated this groups importance in several interactions with the proteins side chain. Besides the many hydrophobic interactions and hydrogen bonds the authors also noted the versatile yet specific contacts between the two tryptophan moieties of the argyrins and the variable regions of the proteasome's binding pocket.
1.5.2) Total synthesis of the Argyrin family
The total synthesis of several members of the argyrin family of cyclic peptides has been described. Notably work by the Ley group45 described the total synthesis of argyrin B using a flexible step-wise route that conserved the stereochemistry. Their route could potentially allow the incorporation of subunits featuring different substituents thus allowing the development of a library of argyrin based compounds. They used an enzymatic resolution step to generate the unusual 4-methoxy tryptophan with good effect and their final cyclization utilised a phenylseleno cysteine that underwent oxidative elimination to yield the dehdyroalanine.
1.6) Aim
This project aims to design and produce a potent inhibitor of the proteasome. This is a validated anti-tumour target as its inhibition ultimately causes apoptosis. Our new inhibitor will use argyrin A as a lead compound and we will make various analogues to build up a structure-activity profile. By using the information published in the literature about the interaction of argyrin with the proteasome we will synthesise new molecules with the aim of building a structure-activity profile for the drug. This project will build on and develop the previous work that has been published in the literature by adapting the published chemistry to produce cyclic peptides via solid phase synthesis. The synthesis of two alternate subunits of argyrin A is described here, both of the new compounds involve changes to the Trp2 moiety of argyrin A. Firstly the residue is substituted with a 4-methyl-DL-tryptophan group to clarify the role of the methoxy substituent, and secondly the position of the methoxy substituent is evaluated by changing it from C4 to C5 of Trp2. The rationale behind the choice of the modifications and the synthetic pathways followed will be discussed in the next chapter.
