structure – Page 2 – Drug Discovery Opinion

Structure of Aromatase Determined

The majority (75-80%) of breast cancers are hormone-sensitive and their growth is stimulated by the hormones estrogen and progesterone. In premenopausal women, most estrogen is produced by the ovaries, and selective estrogen receptor modulators such as tamoxifen are used to block the cancer-promoting properties of estrogen. In post-menopausal women, however, estrogens are produced largely by the action of aromatase on androgens produced by the adrenal glands, and reversible (anastrazole and letrazole) and irreversible (exemestane) inhibitors of aromatase have become widely used as treatments in these women. Now Dr Debashis Ghosh’s group at the Hauptman-Woodward Medical Research Institute have solved the structure of the aromatase cytochrome P450 enzyme from human placenta at 2.9Å resolution. The work is published in the 8 January 2009 issue of the journal Nature.

Unlike the active sites of many microsomal P450s that metabolise drugs and other xenobiotics, the aromatase, which is anchored in the membrane of the endoplasmic reticulum, has an androgen-specific cleft that forms hydrophobic and polar interactions with the substrate, androstenedione. The group hope that the new structural information will pave the way to improved aromatase inhibitors for the treatment of breast cancer. The group has previously solved the structures of two other enzymes involved in estrogen biosynthesis, estrone sulfatase (2003) and 17β-hydroxysteroid dehydrogenase type 1 (1996).

NS1 Protein ‘Hides’ Influenza Virus RNA

The non-structural (NS1) protein of influenza A viruses plays a major role in countering host immune defence by limiting production of interferon and also limiting the antiviral effects of IFN-induced proteins. NS1 also directly modulates stages of the virus replication cycle, including viral RNA replication and viral protein synthesis. Researchers at the Baylor College of Medicine have now identified a mechanism whereby NS1 protein from the highly virulent avian influenza strain, H5N1, ‘hides’ the pieces of double-stranded RNA that would otherwise trigger an antiviral response. NS1 is comprised of two domains: an RNA binding domain and an effector domain. X-ray crystallography of full-length NS1 revealed that molecules of the NS1 protein combine to form tiny tubules which sequester viral RNA. The oligomeric organisation allows the residues involved in RNA-binding to face inwards towards a 20Å wide central tunnel. Binding sites for cellular factors, which may help to evade an immune response, were also identified on the surface of the tubules. It is not yet known whether NS1 proteins from other influenza strains also form similar tubules but, if so, it is possible that the disruption of such structures could form the basis for new treatments. The findings have been reported in the November 5th online edition of the journal, Nature.

Structure of “Caffeine Receptor” Solved

Collaborating scientists at the Scripps Research Institute and The Amsterdam Center for Drug Research have determined the crystal structure of the human adenosine A_2A receptor, also known as the caffeine receptor. The receptor is a member of the hetero-trimeric G-protein coupled receptor (GPCR) superfamily and plays an important role in mediating responses to adenosine in many physiological processes. The scientists were able to obtain crystals of the protein by binding it to a potent adenosine antagonist, ZM241385, which had been developed as a potential drug to combat Parkinson’s disease. Full details have been published in the journal Science.

Despite the importance of GPCRs as drug targets, determination of their crystallographic structure has proven difficult. This new structure follows the success of the Scripps team’s publication of the β2-adrenergic receptor structure last year.

Adenosine interacts with a number of GPCRs including the A₁, A_2A, A_2B, and A₃ subtypes. Each of these plays a role in responding to adenosine in the central nervous system in pain regulation, cerebral blood flow, basal ganglia functions, respiration, and sleep. Insights obtained from the study of the A_2A structure have already suggested mechanisms for receptor subtype selectivity.

It is hoped that this new information will help in the design of new drugs that could be important in the treatment of numerous neurological disorders, including Parkinson’s and Huntington disease.

Small Molecules Lend a Hand in Protein Folding

Celastrol Protein folding is the process whereby newly synthesised linear polypeptide chains fold into the well-defined 3-dimensional shape of the functional protein. In many cases, molecular chaperones assist in correct protein folding by preventing the newly synthesised protein from aggregating into non-functional structures. A variety of diseases result from misfolded proteins; loss-of-function diseases are often caused by a point mutation in the sequence of the protein which disturbs the normal balance between protein folding and clearance. There has been recent interest in the development of ‘pharmacological chaperones’ which are small molecules that stabilise the correct protein fold.

God Vine A recent study describes two small molecules, celastrol and MG-132, that are able to enhance mutant protein folding and function in cell culture experiments. These compounds acted synergistically with known pharmacological chaperones and increased the activity of mutant proteins to 50% of wild-type activity. This study provides encouragement for the concept of developing regulators of proteostatis for the treatment of a range of loss-of-function diseases.

Structure of Telomerase Revealed

Telomeres are repetitive sequences at the 3’-end of DNA which protect the end of the chromosome from destruction during cell division. During the process, the telomeres are themselves destroyed and this mechanism normally limits cells to a fixed number of divisions. Embryonic stem cells express an enzyme, telomerase, which replaces the telomeres and allows the cells to divide repeatedly. Telomerase remains active in some rapidly dividing adult cells, but is switched off almost completely in most other cells to prevent excessive proliferation. Cancer cells often regain telomerase activity and are able to replicate indefinitely. Telomerase activity has been observed in approximately 90% of human tumours and inhibition of this enzyme is seen as a potential treatment for many cancers.

The telomerase is a reverse transcriptase that carries its own RNA primer sequence and has some similarities to the retroviral reverse transcriptases, viral RNA polymerases and B-family DNA polymerases. The first telomerase inhibitor to enter clinical trials for the treatment of cancer is GRN163L, a lipid-conjugated thiophosphoramidate. GRN163L is resistant to nuclease digestion in blood and tissues and has very high affinity and specificity for telomerase.

Small molecule inhibitors such as BIBR1532, which inhibits telomerase activity in vitro with an IC₅₀ in the low nanomolar range, have also been identified. The nucleoside analogue AZT, which is used to treat HIV by inhibiting the viral reverse transcriptase, weakly inhibits telomerase activity.
An advance online publication in the journal Nature describes a high resolution structure of the Tribolium castaneum catalytic subunit of telomerase, TERT (Telomerase Reverse Transcriptase).

It is hoped that the new structure will help in the design of small molecule telomerase inhibitors. As well as de novo design, the similarity between TERT and HIV reverse transcriptase suggests that it may be possible to modify reverse transcriptase inhibitors to inhibit telomerase. Such compounds could potentially be used to treat a wide range of cancers.

New Targets to Fight Flu

Influenza A is a major human and animal pathogen with the ability to mutate and cross species: the recent emergence of the H5N1 strain of the Avian Influenza A virus has emphasised the need for new treatments. Between 2003 and 2008 there were 385 confirmed human cases of the avian H5N1 strain (WHO data).
Existing flu treatments such as Tamiflu^® and Relenza^® target the viral neuraminidase, a highly variable protein on the surface of the virus which can mutate to give viral strains which are resistant to the drugs.

PA-PB1 Complex (pdb id: 3cm8)

Two independent reports (He et.al., Obayashi et.al.) have recently described the crystal structure of the viral RNA polymerase, a protein complex that is essential for viral transcription and replication. The complex contains three proteins, PB1 which has polymerase and endonuclease activities, PB2 which is responsible for binding capped RNA, and PA, the function of which is less clear. Both structures show large fragments of PA bound to a smaller helical fragment of PB1. If it were possible to devise a small molecule that could disrupt this binding, it would likely prevent polymerase activity and viral replication. Although protein-protein interactions are considered to be difficult targets for drug discovery, in this case the binding area is relatively small and offers a potential target for novel anti-influenza drugs.

Progress in Targeting the Sliding Clamp

There has been interest in the DNA polymerase sliding clamp as an antibacterial target for the last 15 years. Sliding clamp proteins, found in all organisms, encircle DNA (and slide along it!) and tether polymerases to enable rapid and processive DNA replication. The proteins are known as proliferating cell nuclear antigen (PCNA) in eukaryotes and as the β-clamp in prokaryotes. In PCNA the clamp is composed of three subunits of two domains each, whilst the bacterial β-clamp is assembled from two subunits of three domains. Although the overall structures of the eukaryotic and prokaryotic clamps are similar, there is no detectable sequence homology.

RU7 In a paper to be published in the August 12^th edition of PNAS, the authors disclose a small molecule inhibitor of the E.coli β-clamp, RU7, which differentially inhibits polymerases II, III and IV.

RU7 selectively inhibits Pol III in β-dependent replication assays, with no activity in the eukaryotic PCNA system. The compound, which has modest potency, was identified by screening for compounds able to displace a Pol III peptide from the β-clamp. The authors have also determined the co-crystal structure of RU7 bound to the clamp (pdb identifier 3d1g), paving the way for structure-based design.
PDB:3D1G

PDB:3D1G