G-quadruplex Nucleic Acids as a target for Pancreatic Cancer Drug Discovery

Pancreatic Cancer

Pancreatic cancer is highly lethal, with a mere 3% survival rate 5 years after diagnosis in England and Wales, and it is the fifth most common cause of cancer death (London School of Hygiene and Tropical Medicine, 2014 & Cancer Research UK, 2014). The reason for the poor prognosis of pancreatic cancer is largely due to the delay in its diagnosis. Symptoms present in a similar manner to other illnesses, and the location of the pancreas, deep inside the body, presents an obstacle for the identification of tumours. Additionally a particular difficulty in drug discovery for pancreatic cancer is that the standard drug used for its treatment, gemcitabine, develops resistance. In the past 40 years there has been no improvement on the 3% survival rate, compared to breast cancer survival rates which have increased from 50 – 80% since the 1970s (Pancreatic Cancer Research Fund, accessed 2016). Accordingly there is a great urgency that the unacceptable prognosis of pancreatic cancer be improved through identifying novel therapeutics for its treatment.

What are G-quadruplexes?

G-quadruplexes are guanine rich secondary structures of DNA. Four guanine bases can associate through Hoogsteen hydrogen bonding to form a square planar structure, known as a guanine tetrad. When 2 or more guanine tetrads stack they form a G-quadruplex, which is further stabilised by the presence of a cation, usually potassium. The cation sits in the central channel between each pair of tetrads. A schematic of the G-quadruplex structure (described above) is shown in Figure 1.


Rachael Picture 1 27-01-2016

Figure 1, G-quadruplex structure, taken from Huppert, 2006

Targeting G-quadruplexes

Stabilisation of G-quadruplex structures formed within telomeric DNA with the use of G-quadruplex-selective ligands have the capacity for inhibiting the enzyme telomerase (Haider et al., 2003). Telomerase is a highly attractive target for stop the indefinite division of cancerous cells. It is almost universal in all human cancer cell lines, and is found in over 85% of primary tumours (Kim et al., 1994). Telomerase is responsible for the rebuilding of telomeres, the capping ends of chromosomes (Collins & Mitchell, 2002). During normal cellular replication telomeres shorten. Once the critical limit of telomere length has been met cells are programmed for apoptosis. However, with upregulated telomerase, cancer cells have the ability to replicate indefinitely. Accordingly G-quadruplex stabilising ligands can potentially induce apoptosis in cancerous cells.

Liu et al. (2005) have shown that sub-cytotoxic doses of G-quadruplex targeting ligands TMPyP4 and telomestatin in MIA PaCa-2 cells, pancreatic cancer cells, are associated with an increase in senescent cells by 15 and 10% respectively, as determined by β-galactosidase staining at day 39 and 103 respectively. These results validate that growth inhibition of cells is induced by the progressive loss of telomere DNA. This consequently leads to cellular senescence and cell death.

More recently a great interest has been shown in the vast abundance of G-quadruplexes in the promotor regions of oncogenes. The Neidle group at UCL have designed a compound called MM41, a tetra-substituted naphthalene-diimide derivative, shown in Figure 2, which reduces the growth of pancreatic tumours by 80% in treated mice (Ohnmacht et al., 2015). MM41 was delivered by IV with twice-weekly dosage of 15 mg/kg. Tumour growth prevention by MM41 over a time course is displayed in Figure 3. MM41 binds strongly to G-quadruplexes encoded within the promoter sequences of BCL-2 and k-RAS genes. Blocking these genes, which are prominent in cancer and key to the survival and growth of cancer cells, induces cancer cell apoptosis.

Rachael Picture 2 27-01-2016

Final Remarks

Further refinements of MM41 are underway as it is not yet ready for human trials. A derivative compound of MM41, CM03, has shown to target a gene which is involved in cancer resistance (Neidle, accessed 2016). In gemcitabine-resistant MIA PaCa-2 cells CM03 is still potent, thus showing promise in overcoming the hurdle that drug resistance poses on pancreatic cancer drug discovery (Neidle group, data not published). A key consideration is to ensure that potential drugs for pancreatic cancer which designed to interact with G-quadruplexes are selective in targeting cancerous cells, and do not interfere with normal cellular function. From the ongoing trials undertaken by the Neidle group, progress has been made on the drug discovery front for pancreatic cancer in recent years, and advances to human trials may be on the horizon. Thus there is reason to be optimistic about the future of drug discovery for pancreatic cancer.

Blog written by Rachael Besser


Cancer Research UK Website (2014). Pancreatic Cancer. Available at: http://publications.cancerresearchuk.org/downloads/Product/CS_KF_PANCREAS.pdf (accessed 20th January 2016)

Collins, K. & Mitchell, J.R. (2002) Telomerase in the human organism. Oncogene. 21(4) 564 – 579

Haider, S.M., Parkinson, G.N. & Neidle, S. (2003) Structure of a G-guadruplex-ligand complex. Journal of Molecular Biology 326(1) 117-125

Huppert, J Website (2006) Biophysics and Bioinformatics of Nucleic Acids. Available at: http://people.bss.phy.cam.ac.uk/~jlh29/index.html (accessed 21st January 2016)

Kim, N.W., Piatyszek, M.A., Prowse, K.R., Harley, C.B., West, M.D., Ho, P.L., Coviello, G.M., Wright, W.E., Weinrich, S.L. & Shay, J.W. (1994) Specific association of human telomerase activity with immortal cells and cancer. Science. 266(5193) 2011–2015

Liu, W., Sun, D. & Hurley, L.H. (2005) Binding of G-quadrupex-ineractive Agents to Distinct G-Quadruplexes Induces Different Biological Effects in MiaPaCa Cells. Nucleosides, Nucleotides and Nucleic Acids. 24: 1801-1815

London School of Hygiene and Tropical Medicine data accessed via Cancer Research UK Website (2014) Pancreatic cancer survival statistics. Available at: http://www.cancerresearchuk.org/health-professional/cancer-statistics/statistics-by-cancer-type/pancreatic-cancer/survival#ref-0 (accessed 14th January 2016)

Neidle, S. UCL School of Pharmacy Website. Targeting cancer genes – A novel approach to the treatment of pancreatic cancer. Available at: https://www.ucl.ac.uk/pharmacy/research/drug-discovery/drug-discovery-projects/pancreatic-cancer (accessed 21st January 2016)

Ohnmacht, S.A., Marchetti, C., Gunaratnam, M.E., Besser, R.J., Haider, S.M., Di Vita, G., Lowe, H.L., Mellinas-Gomez, M., Diocou, S., Robson, M., Sponer, J., Islam, B., Pedley, R.B., Hartley, J.A. & Neidle, S. (2015) A G-quadruplex-binding compound showing anti-tumour activity in an in vivo model for pancreatic cancer. Scientific Reports. 5(11385)

Pancreatic Cancer Research Fund Website. Why We Exist. Available at: http://www.pcrf.org.uk/pages/why-we-exist.html (accessed 14th January 2016)

Rhodes, D. and Lipps, H.J. (2015) G-Quadruplexes and their regulatory roles in biology. Nucleic Acids Research. 43(18) 8627 – 8637


Heavyweight drugs

Non-radioactive isotopes, also known as stable isotopes, are used extensively in industry and academia. Compounds labelled with one or multiple stable isotopes (e.g. 13C, 15N, 2H, etc) combined with analytical techniques (NMR, GS, MS, IR) can provide a valuable information in the elucidation of reaction mechanisms and metabolic research.[i]

The increase in the use of deuterium labelling can be explained by the central position which hydrogen holds in chemistry and its role in proton transfer reactions. Initial studies of deuterated water, established its low toxicity towards mammalian cells.[ii] Safety is one of the benefits of stable isotopes, which enables their use for in vivo human studies.[iii] Deuteration may change the pathway of drug metabolism, which is called metabolic switching. This may lead to increased duration of action and lower toxicity.

The shape of the molecule depends on its atoms and the interactions of their electron clouds. Due to the fact that deuterated and non-labelled analogues have the same number of electrons, the shapes and sizes of the compounds are very similar (Figure 1).[iv]

Victor Fig 1 25-01-2016

Figure 1: An overlay of the crystal structures of CTP-347 hydrochloride hemihydrate (purple) and paroxetine hydrochloride hemihydrate (green)

Concert Pharmaceuticals developed a compound called CTP-347, a selectively deuterated analogue of paroxetine, for the treatment of hot flashes.[v] Modification was required, because paroxetine irreversibly inactivates the liver enzyme CYP2D6, leading to various side effects when used in combination with other medications. The proposed mechanism (Scheme 1) shows the covalent binding between paroxetine metabolite and the active site of CYP2D6, forming irreversible complex.[vi] On the contrary, CTP-347 demonstrated little to no CYP2D6 inactivation, when the experiments were performed in vitro. It is believed that selective deuteration prevents the formation of the undesired carbene metabolite.

Victor Fig 2 25-01-2016

Scheme 1: Proposed mechanism of CYP2D6 inactivation by the carbene metabolite of paroxetine

CTP-347 was subsequently studied in a patient clinical trial. Along with CTP-347, dextromethorphan was administered, which acts as a selective probe for CYP2D6 activity. Subjects receiving CTP-347 were able to metabolise dextromethorphan better than the patients who received paroxetine. This demonstrated the benefit of using deuterium over hydrogen in order to overcome undesired drug-drug interactions, potentially enabling the broader use of CTP-347 with other drugs.

Despite the change in the metabolic pathway, CTP-347 has the same inhibition and selectivity as paroxetine. In the enzyme assays the two drugs were essentially identical. Human assays proved that CTP-347 did not form any unwanted metabolites compared to paroxetine.

Additionally to the above, Auspex Pharmaceuticals announced that a deuterated version of venlafaxine causes fewer side effects and stays in the bloodstream longer than the non-deuterated version, which prolongs the action of the drug.[vii]

Although it is difficult to predict how incorporation of stable isotopes may affect the mode of a drug, it has been demonstrated experimentally that stable isotopes can provide potential improvement of the safety, tolerability, efficacy and dosing of drug candidates.

Blog written by Victor Zdorichenko

[i] Godin, J.-P.; Ross, A. B.; Rezzi, S.; Poussin, C.; Martin, F.-P.; Fuerholz, A.; Cléroux, M.; Mermoud, A.-F.; Tornier, L.; Arce Vera, F.; Pouteau, E.; Ramadan, Z.; Kochhar, S.; Fay, L.-B.; Anal. Chem. 2010, 82, 646–653

[ii] Kushner, D. J.; Baker, A.; Dunstall, T. G.; Can. J. Physiol. Pharmacol. 1999, 77, 79–88

[iii] Iglesias, J.; Sleno, L.; Volmer, D. A.; Curr. Drug Metab. 2012, 13, 1213–1225

[iv] Harbeson, S. L.; Tung, R. D.; Drug Discovery & Development magazine 2010, 13, 22

[v] Harbeson, S. L.; Tung, R. D.; In Annual Reports in Medicinal Chemistry; John E. Macor, Ed.; Academic Press, 2011; 46, 403–417

[vi] Fischer, J.; Ganellin, C. R.; Rotella, D. P.; Analogue-Based Drug Discovery III; John Wiley & Sons, 2013, 56–58

[vii] Sanderson, K.; Big Interest in Heavy Drugs. Nature News 2009

Structure-based drug discovery

Protein-ligand binding is important for early stage drug discovery. Structure-based drug design/discovery (SBDD), a rational and efficient strategy, aims to predict the binding mode and the affinity of protein with the ligand. For decades, researchers are on the way to improve the accuracy of SBDD and the great potentials and success of such method have been seen in the drug discovery field.

On a recent paper[1] a FEP (Free-energy perturbation) protocol was released that enabled highly accurate affinity predictions across a broad range of ligands and target classes. Researchers applied this method in several drug discovery projects and got a high level of accuracy, indicating the ability of this approach to drive decisions in lead optimization.

This method is based on an improved force field OPLS2.1, which incorporates a robust model for non-bonded interactions in conjunction with extensive training of torsional and covalent parameters against more than 10,000 representative organic compounds. OPLS2.1 displays a lower error statistics than some other force fields, such as OPLS2005. In addition, the Desmond program is employed to run FEP simulations and molecular dynamics/replica exchange capabilities are augmented in Desmond with the newly developed FEP/REST (free energy perturbation/replica exchange with solute tempering) algorithm. Desmond with FEP/REST is implemented to run on graphics processing units (GPUs).

The researchers applied this methods into several studies. First, FEP/REST methodology was used to validate eight targets and related ligands. They evaluated the data of those eight sets produced from FEP methodology and got a scatter plot of predicted versus experimental binding energies for the entire data set (Fig.1).

Tina 20-01-2016 Figure 1

Figure. 1 Correlation between FEP-predicted binding free energies and experimental data for all eight systems studied. FEP-predicted binding free energies for most of the ligands are within 1.0 kcal/mol of their experimental values, and only nine of 199 studied ligands deviate from their experimental free energies by more than 2 kcal/mol.

In order to further validate the ability of FEP methodology on lead optimization, they examined several representative compounds with the targets by both computational and wet lab measurements; the results were corresponded to each other and structure-activity relationships successfully captured by FEP (Fig 2).

Tina 20-01-2016 Figure 2

Figure 2. Representative examples of different types of interactions captured by FEP.

This technology was then employed into eight active drug discovery projects. On project one, several hundred compounds had been synthesized and the target biochemistry affinity and some other biological properties had been achieved. Next, FEP technology was used to optimize compounds, selecting compounds with better biological properties while maintaining a high affinity. 195 compounds were scored by FEP and 22 were synthesized and assayed. By compared the experimental and predicted data, the true negative rate was 93% and the true positive rate was 71%, which indicated a high accuracy of FEP technology (Fig.3).

Tina 20-01-2016 Figure 3

Figure 3. Histograms showing distribution of experimental values for compounds in project I that were predicted by FEP to have pKi > 8 (dark gray), predicted by FEP to have pKi < 8 (medium gray), and those that were not computationally predicted prior to being assayed (light gray). Numbers above the bars correspond to the actual number of compounds that were assayed. Approximately 14% of the compounds that were chosen to be synthesized and assayed without guidance from FEP had pKi > 8, while 71% of the compounds predicted by FEP to have pKi > 8 had an experimental pKi > 8. Key to labels: TP = true positive, FN = false negative, FP = false positive, TN = true negative.

Over the past decades, SBDD methods such as molecular docking, pharmacophore modeling and mapping, structure-based virtual screening have been greatly improved. On one recent study, SBDD was successfully applied to discover inhibitors of Human Helicase DDX3.[2] Based on the structural information of DDX3, a homology of a closed conformational model was built and then a hit was then docked into the predicted RNA binding pocket for pharmacophore building. Commercial compounds databases were screened by virtual screening, followed by docking experiment via GOLD software. Selected small molecules were validated by DDX3 helicase activity assay and an impressive hit rate of 40% was obtained. The success of SBDD in drug discovery field is not only dependent on the constantly changing technology, but also about the detailed structural knowledge of the target macromolecules, which are mainly obtained from crystal structures, NMR data or homology models.[3] In the future, it is still necessary to increase the accuracy and effectiveness of existing technologies in computational drug discovery field, but the most important tendency will be the integration of computational chemistry and biology together with chemoinformatics and bioinformatics.[4]

Blog written by Xiangrong CHEN


  1. Wang, L., et al., Accurate and reliable prediction of relative ligand binding potency in prospective drug discovery by way of a modern free-energy calculation protocol and force field. Journal of the American Chemical Society, 2015. 137(7): p. 2695-2703.
  2. Fazi, R., et al., Homology Model-Based Virtual Screening for the Identification of Human Helicase DDX3 Inhibitors. Journal of chemical information and modeling, 2015. 55(11): p. 2443-2454.
  3. Chen, L., et al., From laptop to benchtop to bedside: structure-based drug design on protein targets. Current pharmaceutical design, 2012. 18(9): p. 1217.
  4. Ou-Yang, S.-s., et al., Computational drug discovery. Acta Pharmacologica Sinica, 2012. 33(9): p. 1131-1140.


Abzymes: An Introduction

Antibodies (or immunoglobulins) are essential proteins for human immunological functioning. Antibodies are expressed on the membrane of B cells, a type of white blood cell essential for the adaptive immune response. After B cells are activated, they can differentiate into plasma cells that produce the same antibody in large quantities and expel it through exocytosis. Abzymes are these same molecules but with the addition of catalytic activity, a property rarely seen in antibodies naturally, and are currently seeing exciting development in various therapeutic fields.

Enzymes catalyze reactions that would otherwise be too slow for a biological time scale, and do this by having the following roles:

  • Stabilization of transition state


  • Provides reactive groups


  • Localizes the substrate


  • Exclusion of solvent


  • Provides correct dynamics of active site residues


Antibodies are most likely to be able to replicate the first three roles, but not the last two. This partial catalytic activity is reflected in the catalytic rate of abzymes, which is significantly lower than their enzyme counterparts.

How To Make An Abzyme


Enzymes catalyze reactions by stabilizing the high-energy transition state, thereby reducing the energy required and causing the energy barrier of the reaction to be lowered. The most successful way of producing abzymes to date is by raising antibodies against transition state analogs. It is thought that if they have a high affinity for these analogs, they could reduce the energy barrier of a reaction therefore acting as an enzyme.

The other, less mainstream approach is to use an idiotypic pathway to produce an abzyme. This means raising an antibody that is complementary to the active site of the enzyme you want to mimic, then raising a second antibody to the variable region of the first antibody. The theory is that the first antibody would have the complementary properties to the active site of the enzyme, therefore the secondary antibody raised against this should have the same properties of the original enzyme’s active site (Figure 1) (Padiolleau-Lefevre, 2014).

James 18-01-16 Picture 1

Figure 1. This figure illustrates two ways of producing abzymes. The chemical pathway (A) uses transition state analogs to produce catalytic antibodies. The biological pathway (B) uses an enzyme that already catalyzes the same reaction as the starting point (Padiolleau-Lefevre, 2014).


A remarkable example of the transition state analog pathway was demonstrated by G. Wayne Zhou et al 1994. They were able to crystallize a catalytic antibody (17E8) that was raised to a norleucine phosphonate transition state analog, as part of a peptide hydrolysis reaction. 17E8 contained some of the same active residues as a class of proteases knows as serine proteases, which have a reactive triad responsible for their activity (serine-histidine-aspartic acid) – 17E8 contained the serine and histidine of this triad. The antibody also has a lysine strategically placed to stabilize to stabilize oxyanion formation and a small pocket for hydrophobic for side chains (Zhou 1994).


Remarkably, when both structures of trypsin (a serine protease that also catalyses a peptide hydrolysis reaction) and 17E8 containing the analog are superimposed, the active residues are in the exact same position relative to the substrate and to each other (Figure 2). Astonishingly, the immune system has created in about a week that evolution has taken many millennia to perfect.

James 18-01-16 Picture 2

Figure 2. This figure is the superinposition of 17E8 and trypsin completed to bovine pancreatic trypsin inhibit. Trypsin in pink and white with 17E8 in red and yellow (Zhou, 1994).

Potential Clinical Applications Of Abzymes

Antibodies have exceptional specificity and high affinity, so combining this with catalytic activity seems like a perfect combination. As a large protein drug they would lack cell permeability, and therefore the focus has been on extracellular targets such as pathogens, toxins, hormones, cytokines and membrane proteins. A variety of abzyme drugs have been and are being developed and trialled.


Abzymes have been developed to hydrolyze the superantigenic region of gp102, the CD4 binding site of the HIV virus. This would prevent it from infecting its target cells, thus making the virus inert. (Planque, 2008).


Tumor necrosis factor alpha (TNFα) is a cytokine that has an important role in inflammation, and consequently dysregulation of TNFα is a common factor in autoimmune diseases such as rheumatoid arthritis and inflammatory bowel disease. Currently the most effective and useful inhibitors of TNFα are monoclonal antibodies that simply sequester TNFα, e.g. Infliximab (Remicade). The production of a catalytic antibody in this case would hopefully allow for much lower dosage to be needed as one abzyme could inactivate many more molecules of TNFα and be more cost effective in comparison. Recent research in this area has yielded ETNF-6 mAb, a catalytic antibody produced for exactly this purpose (Hifumi, 2010).


Each antibody has two variable regions, a feature that could be exploited in the development of new cancer treatments. Researchers are trying to develop antibodies where one variable region binds to a cancer associated antigen and the other has abzyme activity that actives a pro-drug (Goswami, 2009). The problem with this approach is lies in the heterogeneity of cancers, meaning cancer cells in a single patient may express different antigens.

Final Thought

Like in the development of any new therapeutic tools, the advancement of abzymes has not been without caveats and limitations. However, it is important to remember that the progress in this novel field is on-going and what has been yielded so far certainly seems promising; it does not seem far-fetched to predict that with their versatility, abzymes could be used to treat a plethora of different diseases in the future.

Blog written by James Noble


Zhou G.W, Guio J, Huang W, Fletterick R.J, Scanlan T.S (1994) Crystal Structure of a Catalytic Antibody with a Serine Protease Science, 265. 5175:1059-1064

Goswami RK, Huang ZZ, Forsyth JS, Felding-Habermann B, Sinha SC (2009) Multiple catalytic aldolase antibodies suitable for chemical programming. Bioorg Med Chem Lett 19:3821–3824

Hifumi E, Higashi K, Uda T (2010) Catalytic digestion of human tumor necrosis factor-a by antibody heavy chain. FEBS J 277:3823–3832

Planque S, Nishiyama Y, Taguchi H, Salas M, Hanson C, Paul S (2008) Catalytic antibodies to HIV: physiological role and potential clinical utility. Autoimmun Rev 7:473–479

Padiolleau-Lefevre S, Naya R. B., Shahsavarian M. A, Friboulet A, Avalle B (2014) Catalytic antibodies and their applications in biotechnology: state of the art. Biotechnol Lett 36:1369–1379

Recent advances in optogenetic technology: shedding new light onto neuropathic pain?

Over the last ten years, optogenetics has opened up a whole new world of possibilities within neuroscience research, highlighting the use of ion channels as physiological tools in the process.

Optogenetics combines the focal use of light with the targeted expression of light-sensitive proteins to manipulate and monitor the function of defined populations of cells1. These proteins are currently divided into two groups: the ‘actuator’ proteins, which can transduce a light stimulus into a neuronal signal, and the ‘indicator’ proteins, which report neuronal signals by the emission of light. Here, the focus is on the former. In a recent article in Nature Biotechnology (Dec 2015)2, Park and colleagues used the prokaryotic channelrhodopsin protein (a cation channel responsive to blue light) to reversibly stimulate peripheral and spinal pain pathways in mice which expressed this protein in all, or specific sub-populations, of sensory neurons. By using pulses of blue light, they were able to alter the membrane potential of the cells in which channelrhodopsins were expressed, resulting in neuronal activation, with observable effects on the animals’ behaviour. This in principle is nothing new; what is novel, and possibly revolutionary within optogenetics, in the method developed by this research group to deliver and control the light stimulus.

An improved, reversible pain model that enables natural behaviour.

The in vivo use of optogenetics to target discrete cell populations has several important advantages over lesion-based or genetic knockout animal models: spatial and temporal control over the stimulation or inhibition of the cell population, and more importantly, reversibility of the stimulus. In order to study pain, animals no longer have to be subject to constant pain. However, as with all developing technologies, the earlier incarnations of this technique have been far from optimal. Focal light stimulation has been achieved using fibre-optic cables, which during the course of an experiment, requires an animal to be effectively tethered by their light-source. This impedes natural behaviour and raises a number of both experimental and welfare questions, especially when monitoring behavioural changes is often one of the key read-outs of the experiment.

The research published by Park et al describes an elegant solution to this problem. They have developed a soft, stretchable wireless implant, smaller than a fingertip, which contains radio-frequency-powered LEDs capable of activation 30 cm from transmitter source. Having overcome the problems of heat generation and long-term durability, the implants have been used for reliable, reversible nerve stimulation for up to 6 months without impeding the normal movement or behaviour of animals when not in use (figure 1, used from reference 2 with permission).

Sarah's Lilley 11-01-16 Pic 1.png

Using this technology, implanted either above the sciatic nerve or lumbar spine, coupled with cre-recombinase-based transgenic expression of channelrhodopsin in the nociceptive neurones in mice, the group were able to observe robust, reproducible nocifensive responses on activation of the implanted blue LEDs, which reverted to normal behaviour on deactivation and were absent in cre-negative littermates with implants. Whole-cell voltage- and current-clamp of isolated DRG neurones from these mice confirmed frequency-matched depolarisation and neuronal firing in response to blue light stimulation.

This refined pain model appears to effectively mimic the symptoms of neuropathic pain, and could become an invaluable tool for the investigation of neuropathic pain relief, and possibly other peripheral neurological disorders. As an alternative approach, using optogenetic ion channels to stabilize membrane potential in order to reduce neuronal activity is also possible1. Indeed, the authors conclude with a reference to gene therapy approaches delivering optogenetic channels to human cells already undergoing clinical trial. With these implants, perhaps controlling chronic pain conditions may become as simple as flicking on a light switch?

Blog written by Sarah Lilley.


  • Fenno L, Yizhar O & Deisseroth K (2011). The Development and Application of Optogenetics. Annu. Rev. Neurosci. 2011. 34:389–412
  • Park S et al (2015). Soft, stretchable, fully implantable miniaturised optoelectronic systems for wireless optogenetics. Nature Biotechnology 33(12):1280-1286 (doi:10.1038/nbt for methods)

Figure used by permission from Macmillan Publishers Ltd: Nature Biotechnology (citation 2), copyright 2015.

Humanising mice for CYP modulation and DDI studies

In a study of 300,000 patients in Scotland it was estimated that the average number of drugs taken by an individual rose from 3.3 to 4.4 between the years 1995 and 2010. Of that figure 16% of patients over 65 were taking 10 or more drugs. Considering the aged population is set to double over the next 35 years (by 2050 one in four people in the UK will be 65+) polypharmacy represents a major issue for clinicians (1) and pharmaceutical research alike.

The CYP 450 superfamily of enzymes mediate the turnover and represent the major clearance route of xenobiotics, accounting for the phase 1 metabolism of 80 % of all known drugs. Whilst investigational drugs may be substrates for these enzymes they can also potentially act as inhibitors or inducers, producing a range of undesired PK effects including reduced exposure, enhanced production of reactive metabolite, altered processing of endogenous substrates and modification of the pharmacokinetics of concomitant medications (Drug- drug interactions (DDI)).

In light of increases in polypharmacy, especially in Oncology where anticancer drugs are administered close to maximum tolerated levels, DDIs present a significant safety hazard to patients requiring a thorough investigation of the metabolic/clearance modulating potential of new drugs.

A number of in vitro assays can now readily identify CYP450 inhibitors and active metabolites but identification of CYP450 inducers requires the use of more complex model systems. Immortalised hepatocyte cell lines tend to exhibit a significant reduction in CYP expression level and whilst cryo-preserved hepatocytes are the most accessible in vitro system for analysing CYP induction, they lack the host factors involved in complex DDIs and non-hepatic clearance. Animal models offer the most complex systems for characterising investigational drug PK however results must regarded with caution and only interpreted via established correlations.

Humans have 57 CYP genes compared to 102 in mice and of the 23 genes in the human CYP1-3 families (the CYPs most responsible for the metabolism of xenobiotics) there are only 5 orthologous genes in mice(2) (CYP1A1/2/B2, CYP2E1/S1) offering limited utility in accurate prediction of metabolism and CYP450 modulation by investigational drugs in humans. This difference is compounded by the CYP transcriptional activators (e.g. Pregnane X Receptor (PXR) and Constitutive Androstane Receptor (CAR), both activators of CYP3A4) exhibiting differing substrate affinities between the species.

Recently Sheer el al (3) have described the generation of a humanised mouse in which the endogenous Cyp3a, Cyp2d and Cyp2c gene-clusters, in addition to the PXR and CAR genes, were replaced by the human CYP3A4, CYP2D6, CYP2C9, PXR and CAR genes respectively (whilst not necessarily structurally orthologous, the three mice CYP clusters are responsible for the majority of xenobiotic metabolism in mice). In humans CYP3A4 accounts for 45% of CYP activity, with all three enzymes (including CYP2D6 and CYP2C9) collectively responsible for ≈75% of all phase 1 metabolism. The level of promiscuity exhibited by CYP3A4 is matched by the transcriptional activators PXR and CAR that, upon binding a broad spectrum of ligands, translocate to the nucleus and bind to response elements in the CYP3A4 promoter (as a heterodimer with the Retinoid X Receptor (RXR)) and numerous other phase 1, 2 and 3 metabolising and transport protein genes. To ensure a response to xenobiotic stimuli similar to that observed in humans the mice PXR and CAR genes were swapped with their human counterparts as whilst the human and mouse genes/proteins share moderate homology (PXR: 83% DNA/77% protein. CAR: 80% DNA/73% protein), affinity for various ligands differ significantly (4). Mouse and human RXR however are highly homologous (89%DNA/97% protein) negating the need to replace the endogenous gene.

Treatment of PXR/ CAR/CYP3A4/ CYP2D6/CYP2C9 mice with Rifampicin (RIF; a potent activator of PXR) demonstrated a significant up-regulation of CYP3A4 whilst, as expected, not significantly altering the basal levels of expression of CYP2C9 (under transcriptional control of the liver specific albumin promoter )or CYP2D6 (not responsive to CAR or PXR).

Marcus 11-01-2015 Picture 1

Figure 1. A comparison of CYP3A4, 2D6 and 2C9 expression in microsomes derived from human and RIF/vehicle (Corn oil) treated PXR/ CAR/CYP3A4/ CYP2D6/CYP2C9 mouse livers. Donor 1 and 2 represent human individuals with low and high activity respectively for both CYP3A4 and CYP2D6.

The kinetics of CYP3A4 in microsomes from RIF/vehicle treated PXR/ CAR/CYP3A4/ CYP2D6/CYP2C9 mice was verified by midazolam (MDZ; CYP3A4 specific substrate) metabolism and a series of CYP3A4 specific inhibitors: Ketoconazole (KTZ; mixed competitive/non-competitive), Erythromycin (Mechanism based inhibitor) and CYP3cide (time dependant inhibitor) with controls provided by wild type mice and human derived microsomes. A comparison to Cyp2c and Cpy3c KO mice also verified that MDZ was metabolised solely by CYP3A4 in PXR/ CAR/CYP3A4/ CYP2D6/CYP2C9 mice. A 58 fold increase in Vmax between RIF and vehicle treated PXR/ CAR/CYP3A4/ CYP2D6/CYP2C9 confirmed western blot and LC/MS-MS measured increases in protein expression of CYP3A4

Similarly, activity of CYP2D6 and 2C9 was verified with substrates Bufuralol and diclofenac and inhibitors Quinidine and Sulphenozole respectively.

A single In Vivo study was performed in which PXR/ CAR/CYP3A4/ CYP2D6/CYP2C9 mice were treated i.p. with either vehicle or RIF for three days before p.o. administration of MDZ or KTZ followed 30 min later by MDZ (figure 2). RIF treatment resulted in an 8-fold decrease in MDZ exposure over vehicle treated mice. KTZ treatment resulted in a 1.9-fold increase of the AUC over RIF only treated mice indicative of the inhibition of hepatic/intestinal CYP3A4.

Marcus 11-01-2015 Picture 2

Figure 2. A) Whole blood concentrations of MDZ vs time in vehicle, RIF and RIF + KTZ treated PXR/ CAR/CYP3A4/ CYP2D6/CYP2C9 mice. B) MDZ AUC values for graph A.

In generating a humanised mouse Scheer et al have demonstrated the utility of this model for the prediction of drug PK in humans enabling the investigation of the relationship of efficacy and exposure, CYP induction or inhibition and subsequently the effects of various drug combinations possibly resulting in DDIs. Whilst not a comprehensive assay of drug/human CYP interaction this multiply humanised mouse may be the starting point for the generation of an array of transgenic animals that might go further to model a more complete system to include (inducible, or not) human transporters and phase 2 enzymes. Further, by introducing tissue specific promoters relative contributions of hepatic and extra-hepatic clearance routes could potentially be delineated.

Viable Cyp3a and Cyp2d KO mice have previously (5) been generated with only minor phenotypic variances from w/t demonstrating a high level of redundancy of the CYP metabolic system. Considering the promiscuity of human CYPs and their corresponding inducers perhaps these models can be further humanised without jeopardising the processing of important endogenous substrates thereby enhancing the predictive power of the model.

The attrition rate of drugs in clinical development due to PK related issues fell 4-fold between the years of 1991 and 2000 as the industry better understood both the predictive (or lack of) and correlative value of in vitro and in vivo DMPK assays. Monitoring complex DDIs in a broader population falls within the remit of Phase 4 studies so perhaps the development of an array of animal models that are directly predictive (or in the least demonstrate much stronger correlation to humans than we have at present) would facilitate the earlier triage of candidates with clear DDI issues, further reducing the risk of late stage failure.

Blog written by Marcus Hanley

1 – Guthrie B et al (2012) Primary Health Care Research & Development, vol 13, supp S1: 45 2E.2.

2 – Nelson DR et al (2004) Pharmacogenetics, 14:1-18

3 – Scheer N et al (2015) Drug Metab Dispos 43:1679–1690

4 – Xie W et al. (2000) Nature 406:435–439.

5 – Scheer N et al (2012) Mol Pharmacol 82:1022–102.

The changing landscape of drug discovery and UK academia

There are currently at least 24 open-access drug discovery facilities operating in UK, 20 of which are hosted by academic institutions1. Drug research is changing, from how it is done to where and who is doing it. In the last decades the search for new medicine has been largely done by pharmaceutical companies using mostly target-centric approaches2. The pharma research and development (R&D) productivity has been far from what was expected and it is very common to read about ‘crucial challenge’, ‘plummeting’, at minimum ‘difficult’ when analysts, business heads and researchers talk about drug R&D. Pharma is consequently regrouping and simultaneously the academic drug discovery landscape is evolving.

How is pharma adjusting itself? Companies have shifted from highly diverse to a few core therapeutic areas (closing many of their research sites) and have moved away from conducting early stage preclinical research in house. In search for boosting innovation they also seem to follow a Silicon Valley model whereby they geographically cluster and this at sites in close proximity of renowned academic institutions (as is the case for Boston area in US in the vicinity of Harvard University and Massachusetts Institute of Technology and the Cambridge Science Park in UK close to the Oxford/Cambridge/London, with the so-called ‘golden triangle’ universities). More focus, outsourcing to Contract Research Organizations (CROs) and academia, knowledge and talent transfer and high tech access is what big pharma is looking after in an effort to boost its not-as-expected R&D productivity and innovation. Big pharma adapts its business model to become a ‘network integrator’ of drug discovery science rather than the ‘intellectual lead’3.

As the big companies diminish their internal research, drug discovery capabilities are currently built in academic institutions across UK. Looking at the map (see UK Drug Discovery Consortium website http://ukddc.org/members-directory/ & ref. 1) 7 drug discovery academic/not-for-profit facilities are in Scotland, condensed at Dundee/Glasgow/Edinburgh universities, one in Northern Ireland in Belfast, 9 at universities in the north of England with Universities of Leeds and Sheffield each with 3 of such facilities, 6 at London/Cambridge/Oxford and the Sussex Drug Discovery Centre in south at University of Sussex (a 45 strong team, where the author of this blog is working). Their main research focus overlaps largely the industry one in oncology, infectious diseases and neurology. Many of the groups work however on innovative targets, with almost half based on new discoveries, followed by drug targets with significant preclinical validation (35%)4. These facilities have industry standard screening platforms. Drug discovery professionals, including heads of pharmaceutical industrial research have been brought in to combine academic and industrial expertise. These groups have successfully put together the necessary infrastructure and people, being able to run multiple high-throughput screens. 19 of these facilities report being able to undertake medicinal chemistry follow-up. The source of funding is evolving with the majority being provided by UK government, charitable organizations and universities themselves. Strategic alliances between academia, government and pharma have also formed and established facilities as the National Phenotypic Screening Centre or the European Screening Centre. Traditional academic goals are also evolving with The Research Excellence Framework (REF) evaluation of the performance of UK academic institutions including now taking new compounds into the clinic and creating intellectual property.

Despite challenges4, prospects are looking upbeat for drug discovery in academia. And wonderful things happen like the encouraging results obtained by teams at University College of London in collaboration with Great Ormond Children Hospital and biotech company Cellectis in treating leukaemia (http://www.gosh.nhs.uk/news/press-releases/2015-press-release-archive/world-first-use-gene-edited-immune-cells-treat-incurable-leukaemia) or the three new drugs discovered at University of Cardiff that are now in advanced clinical trials for pancreatic and ovarian cancer, shingles and breast and colon cancer.

Blog written by Oana Popa


  1. Shanks, E., Ketteler, R. and Ebner, D. ‘Academic drug discovery within the United Kingdom: a reassessment’. Nature Reviews Drug Discovery 14: 510 (2015)
  2. Swinney, D.C. & Anthony, J. How were new medicines discovered? Nature Reviews Drug Discovery 10: 507-519 (2011)
  3. Rafolsa, I. et al. Big Pharma, little science? : A bibliometric perspective on Big Pharma’s R&D decline Technological forecasting and social change 81: 22-38 (2014)
  4. Tralau-Stewart, C., Low, C.M.R. & Marlin, N. UK academic drug discovery. Nature Reviews Drug Discovery 13:15-16 (2014)


Bargellini Reaction

I recently had the opportunity to put into practice the early 20th century work of the Italian chemist Bargellini, when I had to synthesise α-heterocyclic amino carboxylic acids.

The Bargellini reaction typically involves the addition of trichloromethide to a non hindered ketone resulting in a trichloromethyl carbinol. This rapidly forms a gem-dichloroepoxide. Phenols can add to this gem-dichloroepoxide to give α-phenoxyisobutyric acids.

I followed the work of Butcher and Hurst (Tetrahedron Letters 50, 2009, 2497–2500), which extended the range of novel nucleophiles that can add to the gem-dichloroepoxide intermediate with substituted anilines and amino heterocycles.

Using Boc-protected piperidinone, analogues were made using poorly nucleophilic amino heterocycles. Butcher found that weak, sterically hindered nucleophiles still gave products in useful yields, with the nucleophilic atom not restricted to being a substituent on an aromatic ring (pyrazole).

Kam 04-01-2015 Picture 1

Reaction mechanism for the original Bargellini reaction:Kam 04-01-2015 Picture 2

(Phenoxide can be substituted for other nucleophiles such as anilines/amino heterocycles etc)

In summary, the Bargellini reaction offers a way to introduce varying points of diversity, exploring chemical space in a one pot atom efficient approach.

Blog written by Kamlesh Bala

References: Bargellini, Guido (June 4, 1906). “Azione del cloroformio e idrato sodico sui fenoli in soluzione nell’acetone. (Action of chloroform and sodium hydroxide on phenols in acetone solution.)”. Gazzeta chimica Italiana.

Aromatic amines as nucleophiles in the Bargellini reaction; Ken J. Butcher, Jenny Hurst , Tetrahedron Letters 50 (2009) 2497–2500