Drug Discovery

More than genes and cells: drug discovery in the ECM

Drug discovery in the last few decades has focused on the cellular and genetic mechanisms of disease. This has been very successful in cancer, which is a disease of somatic genetics, and moderately successful elsewhere. But the declining productivity of pharmaceutical and biotechnology investment in drug discovery and development1 suggests that we should be alert to other approaches. One is to look outside the cell, at the extracellular superstructure of the body. Once viewed as an inert structure that is just the biological equivalent of a petri dish, the extracellular milieu is now being seen as a therapeutic target, especially for diseases of old age. Importantly, targeting the scaffold of the body might be a much faster route to treatment for some conditions than attempting to find, and fix, underlying cellular or genetic aetiology of disease.

t is a truism that we become less flexible as we bodies by dry mass. It is a complex meshwork of By Dr William Bains age. This is literally true of many tissues, which proteins, proteoglycans and other molecules con- I become more mechanically rigid and less structed on a scaffold of collagen. There are two amenable to tissue remodelling with age. Loss of broad types of ECM: basement membrane and flexibility is due at least in part to chemical interstitial ECM. The basement membrane directly changes in the Extracellular Matrix (ECM) in the underlies endothelial and epithelial cells, and is tissue, converting a flexible network of molecules composed of primarily of type IV collagen, into a rigid mesh. Stiffness increases have been laminins, entactin, nidogen and heparan sulfate recorded in lung, major arteries, bone, muscle, ten- proteoglycans such as perlecan. The interstitial don and lens of the eye among other tissues in ECM, which makes up most of the extracellular older compared to younger people. However, this mass of the body, consists of many different types remodelling with age is more important to disabil- of collagen, tenascin, proteoglycans and elastin in ity and disease than just piling on the wrinkles or elastic tissues such as skin or tendon. The ECM reducing our score in tennis. In a range of contexts can also bind a variety of soluble proteins such as it has been linked to disease, and research in the cytokines and growth factors2. last decade has suggested how targeting ECM pro- Chemical cross-linking of the proteins in the teins could be blocked or reversed to open a new ECM is central to its construction: human muta- route to treat those diseases. tions or mouse knock-outs of cross-linking enzymes The ECM makes up around one-third of our are usually severely disabling or lethal. During

Drug Discovery World Winter 2013/14 65 Drug Discovery

Figure 1 Cross-links and their A chemistries. Square boxes represent the rest of the protein chain A Lysyl Oxidase. Loysyl Oxidase (LO) acts on lysine (R=H) and hydroxylysine (R=OH) to form allysyl peptides which cross-link to form (hydroxy)lysino-D- norleucine (Compound A) and (hydroxy)lysinoketonorleucine (Compound B). These then can add to further cross-links to form Lysyl Pyrrole (compound C) and Lysyl Pyridinoline (Compound D) end products B Transglutaminase. Glutamine and Lysine are cross-linked by transglutaminases (TG) to form an e-(g-glutamyl)-lysine (EGGL) cross-link B (Compound E) C Glycation. Amine side-chains (here just illustrated with lysine) react with reducing sugars (a polyol aldehyde, typically glucose or fructose, where R is a polyol of the form (CHOH)nCH2OH ) to form a Schiffs base (Compound F), which undergoes Amadori rearrangement to form a stable fuctosamine (Compound G). Oxidation and additional cross-links and rearrangement result in a complex mix of products, of which only Glucosepane (Compound H), Pentosidine (Compound I), Carboxymethyl Lysine (CML – compound J), Carboxyethyl Lysine (CEL – Compound K) and one of the C Methylglyoxal hydroimidazolones (MGH-3 – Compounds L) are shown

66 Drug Discovery World Winter 2013/14 Drug Discovery

growth and development the principal cross-linking Although the inflammation was no less severe in References chemistry is catalysed by Lysyl Oxidase (LO) the treated mice, they lived longer than control ani- 1 Scannell, JW et al. (Figure 1A), a family of at least four enzymes in mals. By targeting the fibrotic damage effectively, Diagnosing the decline in pharmaceutical R&D efficiency. man that oxidise lysine to an aldehyde which can this approach modifies disease without delving Nature Reviews Drug then react with other nitrogen-containing amino into the complexities either of inflammation or of Discovery, 2012. 11(3): acid side-chains to form cross-links that are very hepatic metabolism. p. 191 – 200. stable under physiological conditions. The amount The body does turn over collagen through 2 Karsdal, MA et al. and chemical sites of LO-derived cross-links vary wholesale degradation of the ECM, probably by Extracellular Matrix Remodeling: The Common substantially between tissues and depend, at least in metalloproteases and subsequent excretion of Denominator in Connective part, on the proteins being cross-linked. cross-links. An ideal treatment for chronic fibrotic Tissue Diseases. Assay and Collagen is the most widespread LO substrate. disease might therefore be to supplement this Drug Development There are at least 27 different collagen types, turnover with therapeutic enzymes that break just Technologies, 2013. 11(2): arranged in many different architectures3. the cross-links, allowing less radical remodelling of p. 70 – 92. 3 Birk, DE and Bruckner, P. Different collagens are cross-linked to different fibrotic tissue in situ. This would be a specific Collagen superstructures. degrees; for example, collagen III is present in bone example of an approach termed ‘Medical Topics in Current Chemistry, fibrils but not detectable in extracts, probably Bioremediation’7 (by analogy with environmental 2005. 247: p. 185 – 205. because high levels of cross-linking render it com- bioremediation) more usually discussed in terms of 4 Eyre, DR and Wu, J-J. pletely insoluble4. LO also links elastin into the clearing accumulated intracellular material using Collagen cross-links. Topics in Current Chemistry, 2005. 247: ECM, although it forms different crosslinks exogenously supplied enzymes. Applying such an p. 207 – 229. between elastin chains. approach to fibrotic disease is precedented with the 5 van der Slot, AJ et al. LO is not only essential for the construction of approval of injectable Clostridium collagenase as a Increased formation of ECM during growth and development, but also treatment for Dupuytren’s Contracture8, and is pyridinoline cross-links due to during wound healing. After mechanical damage being pursued at a preclinical level as a therapy for higher telopeptide lysyl hydroxylase levels is a general or inflammation new ECM is laid down and cross- macular degeneration and atherosclerosis among fibrotic phenomenon. Matrix linked with LO family members and transglutami- other diseases. However, no suitable enzyme for Biology, 2004. 23: p. 251 – 257. nases (discussed below). With chronic damage, cross-link clearance is known, and targeting would 6 Barry-Hamilton, V et al. especially chronic inflammation, this wound heal- have to be quite specific to avoid large-scale Allosteric inhibition of lysyl ing response becomes dysregulated and excessive unlinking of ECM collagen. oxidase-like-2 impedes the development of a pathologic masses of poorly organised ECM are laid down, While LO is a major cross-link of collagen, a sec- microenvironment. Nat Med, resulting in fibrosis. Many fibrotic conditions are ond class of enzyme cross-links a wider variety of 2010. 16(9): p. 1009-1017. associated with increase in collagen cross-linking ECM proteins during development, inflammation 7 De Grey, ADNJ et al. Medical density as well as increase in collagen mass5. and wound-healing. Transglutaminases (TGs) are a bioremediation: Prospects for Changed mechanical tension in the ECM itself in group of enzymes that catalyse the formation of an the application of microbial catabolic diversity to aging and such fibrotic tissue enhances fibroblast’s secretion isopeptide bond between the side-chains of gluta- several major age-related of inflammatory cytokines, driving further inflam- mine and lysine residues (Figure 1B). Unlike LOs, diseases. Ageing Research mation and fibrosis. TGs have been widely discussed as a drug target. In Reviews, 2005. 4: p. 315 – 338. Collagen is turned over very slowly: some meas- particular, the ubiquitously expressed TG2 has 8 Hurst, LC et al. Injectable ures of ECM turnover in tendons and bone suggest been researched as a drug target for fibrosis and Collagenase Clostridium Histolyticum for Dupuytren’s protein half-lives of years to decades in man. Partly for cancer9. Contracture. New England for this reason, fibrosis has primarily been seen as TG2 is a multifunctional protein; as well as its Journal of Medicine, 2009. a pathology of excess ECM production, and espe- transglutaminase activity, it is active as a G protein to 361(10): p. 968-979. cially production of collagen fibres that are poorly GPCRs, a GTPase, a disulfide isomerase, a protein 9 Bains, W. Transglutaminse 2 ordered compared to healthy tissue, because once kinase and it binds directly to fibronectin and inte- and EGGL, the Protein Cross- Link Formed by in place the collagen is seen as ‘there for life’. grin, thus mediating cell:ECM interactions. It is Transglutaminse 2, As However, cross-linking is key to this, both because found in the cell nucleus, cytoplasm and plasma Therapeutic Targets for it locks collagen into disordered networks and membrane as well as in the ECM, has at least two Disabilities of Old Age. because it renders it harder to turn over and spice variants, is modulated by Ca2+, NO, GTP and Rejuvenation Research, 2013. remodel into healthier material. Thus LO redox state, and has several roles in inflammation 16(6): p. 495 – 517. 10 Fesus, L and Piacentini, M. inhibitors are being seen as an important therapeu- and apoptosis as well as ECM structural modifica- Transglutaminase 2: an 10 tic approach to a wide range of fibrotic disease. In tion . Its biology is therefore complex, but its trans- enigmatic enzyme with diverse particular, Barry-Hamilton et al6 recently showed glutamination role can be well elucidated by looking functions. Trends in that inhibiting Lysyl Oxidase homologue LOXL2 at the levels of the cross-link it forms (EGGL – see Biochemical Sciences, 2002. with an enzyme-specific monoclonal antibody Figure 1B) in the ECM. Alas, this is rarely done, as 27(10): p. 534-539. reduces fibrosis and cross-linking in mice with liver protein and mRNA levels are easier to measure, so fibrosis induced with carbon tetrachloride. those are what most research examines. Continued on page 68

Drug Discovery World Winter 2013/14 67 Drug Discovery

Continued from page 67

11 Badarau, E, Collighan, RJ, and Griffin, M. Recent advances in the development of tissue transglutaminase (TG2) inhibitors. Amino Acids, 2011. DOI 10.1007/s00726- 011-1188-4. 12 Iismaa, SE et al.,Transglutaminases and disease: lessons from genetically engineered mouse Figure 2: AGE-breaker chemistry. Structure of compounds ALT711, TRC4186 models and inherited disorders. Physiol. Rev., 2009. 89: p. 991 – 1023. 13 Friedrich, T et al. An Like LO, TG2 is relatively promiscuous in its Amadori products. The Schiff’s bases are readily isopeptide bond splitting substrate, acting on amino acids on the surface of hydrolysed back to the free amines and sugars, enzyme from Hirudo proteins that are in inherently disordered regions of but the Amadori products are relatively stable medicinalis similar to ␥-glutamyl structure and which approach near enough in the and can undergo further, complex oxidative transpeptidase. European ECM to be cross-linked. Unlike LO, knocking out chemistry to form a complex soup of compounds Journal of Biochemistry, 1998. 256(2): p. 297-302. TG2 in mice results in animals that are nearly nor- collectively called Advanced Glycation 14 Simpson, RM and Christeller, mal, a bizarre result for a multifunctional, ubiqui- Endproducts (AGEs) (Figure 1C). AGEs are stable JT. Identification of isopeptidase tously expressed enzyme. It is likely that other under physiological conditions, and so accumu- activity in the midgut of insects: enzymes, especially blood clotting factor XIIIa late in long-lived proteins. Some are fluorescent, Purification, properties and (which is also a transglutaminase) can replace and so tissue fluorescence is sometimes used as a nutritional ecology of a Hofmannophila pseudospretella TG2’s function in some roles in development. measure of AGEs. The most common AGEs are (Lepidoptera: Oecophoridae) TG2’s transglutaminase activity is seen in CML and CEL, and many studies examine these larval enzyme. Insect Science, inflammatory disease, and especially fibrotic dis- (they are the immunodominant target of the few 2010. 17(4): p. 325-334. ease, so this may be another target for fibrosis. commercial anti-AGE antibodies that show any 15 Sell, DR et al. Glucosepane There has been a lot of work on inhibitors of TG2, specificity for an AGE). However, recent research is a major protein cross-link of 11 the senescent human primarily aimed at its role in cancer biology . shows that other, non-fluorescent AGEs are more extracellular matrix. Journal of Small molecule TG2 inhibitors reduce fibrotic scar- physiologically important, and particularly the Biological Chemistry, 2005. ring in rodent models of diabetic nephropathy, and cross-link glucosepane15. 280(13): p. 12310 – 12316. a TG2 knock-out mouse has reduced fibrosis in a Glucosepane is strongly correlated with diabetic 16 Fan, X et al. Anaerobic vs lung fibrosis model12. However, these have not yet complications in man16, and has been linked to aerobic pathways of carbonyl and oxidant stress in human had quite as dramatic a therapeutic effect as the degradation of the mechanical properties of skin, 17 lens and skin during aging and LO inhibitor described above. lung, bone and the arterial wall . The main focus in diabetes: A comparative As with LO cross-links, it would be useful not of research in glycation has been the arterial wall, analysis. Free Radical Biology just to block the formation of EGGL in the ECM where AGE cross-links have been specifically asso- and Medicine, 2010. 49: but to clear EGGL embedded in long-lived proteins ciated with hypertension18. It is believed that this p. 847 – 856. 17 Sjoberg, JS and Bulterijs, S. such as tendon and bone collagen. Here we have is because AGE cross-links between all the mole- Characteristics, Formation, and more reason to be optimistic, because enzymes that cules in the elastic ECM of the arterial wall reduce Pathophysiology of Glucosepane: cleave the EGGL isopeptide bond specifically are its ability to expand and contract as the systolic A Major Protein Cross-Link. known, for example from leeches13 and the brown pressure wave passes, thus increasing the systolic Rejuvenation Research, 2009. moth14. The technical barrier to getting even a pre- pressure peak, although other mechanisms are pos- 12(2): p. 137 – 148. 18 Sell, DR and Monnier, VM. clinical candidate therapeutic will be to find an sible. AGEs are also correlated with nephrosis in Molecular Basis of Arterial enzyme with minimum activity against the peptide diabetics. Major causes of morbidity in diabetes Stiffening: Role of Glycation – bonds in proteins, and hence with an acceptable are hypertension, kidney damage and associated A Mini-Review. Gerontology, safety profile. circulatory damage that leads to ulcers and 2012. 58(3): p. 227-237. There are many other chemical changes in the retinopathy, so AGEs are a good potential target 19 Nobécourt, E et al. Effects of cross-link breakers, glycation ECM with disease and age, but the last we will for these diseases. inhibitors and insulin sensitisers consider here is glycation. Glycation is an entire- Alas, AGE formation is purely chemical, and so on HDL function and the non- ly spontaneous and non-enzymatic chemical cannot be blocked by an enzyme inhibitor or a enzymatic glycation of change to proteins (as opposed to glycosylation, mouse knock-out. AGE formation could be apolipoprotein A-I. Diabetologia, which is a different, enzyme-controlled process). blocked by agents that trap the ␣-dicarbonyl inter- 2008. 51: p. 1008 – 1017. Reducing sugars react spontaneously with amines mediates on the path to AGE. Aminoguanidine and Continued on page 69 to form Schiffs bases, which can rearrange to pyridoxamine have been tested as AGE blockers19.

68 Drug Discovery World Winter 2013/14 Drug Discovery

Results have not been that impressive in man, how- little physiological importance unless they happen Continued from page 68 ever, possibly in part because you have to take the to modify a key functional residue, but the MGH compounds continuously for 10 years to block the AGEs are being seen as physiologically impor- 20 Bakris, GL et al. Advanced 27 glycation end-product cross- accumulation of AGEs in a protein with a 10-year tant , and may also be a target for future AGE- link breakers. A novel approach half-life. It is likely that one of the benefits of calo- clearing strategies. to cardiovascular pathologies rie restriction (among many others) is reduced I have focused on fibrotic and inflammatory dis- related to the ageing process. blood sugar levels and hence reduced glycation, eases above because they provide the most obvious AM. J. Hypertension, 2004. but this is an even harder regime to adhere to. conventional target for ECM-targeting therapies. 17(12): p. 23S – 30S. 21 Hartog, JWL et al. Effects A better route would be to clear glucosepane But there is substantial evidence that repairing the of alagebrium, an advanced from the body, but here again we are stymied for ECM could be a valuable approach to cancer ther- glycation end product breaker, lack of an enzyme or, until very recently, even a apeutics and will be a central plank of future on exercise tolerance and good supply of authentic glucosepane to test regenerative medicine treatments. cardiac function in patients enzymes on. However one compound – The ECM environment modulates cancer cell with chronic heart failure. 28 European Journal of Heart Alagebrium (ALT711 – see Figure 2) – has been growth and migration . Cells sense mechanical Failure, 2011. 13: p. 899 – 908. developed as a chemical AGE-breaker. It was tension as well as chemical environment and 22 Little, WC et al. The Effect designed to be a reagent that catalytically breaks ␣- respond to changes in the mechanical structure of of Alagebrium Chloride (ALT- dicarbonyls, which (at the time) were seen as a the ECM: many cancers modulate the ECM 711), a Novel Glucose Cross- major form of cross-link in vivo20. Early trials around themselves to form a more permissive envi- Link Breaker, in the Treatment of Elderly Patients With showed a beneficial effect in man, but a larger fol- ronment for their own growth. Barry-Hamilton et Diastolic Heart Failure. Journal 20-22 6 low-up trial showed no benefit . There is also al showed elevated LOXL2 in some tumours, and of Cardiac Failure, 2005. 11(3): controversy over the drug’s mode of action. Recent that blocking LOXL2 reduced tumour growth in p. 191-195. research has confirmed that Alagebrium does cat- mouse Xenograft models. The role of TG2 in can- 23 Kim, T and Spiegel, DA. The alytically cleave dicarbonyls23, resolving doubt cer is well known29. ECM-targeting strategies have Unique Reactivity of N- Phenacyl-Derived Thiazolium about this in the past, but it is now believed that generally not been high on the list of preferred Salts Toward ␣-Dicarbonyl dicarbonyls are quantitatively trivial compared to anti-cancer strategies because they can only slow Compounds. Rejuvenation the chemically stable AGE cross-links16. So the tri- cancer growth, not kill the cells. However, they Research, 2013. 16(1): als of Alagebrium have generated a drug that does work through fundamentally different mechanisms p. 43 – 50. something in rats24, may do something in man, but from more mainstream approaches, ones unrelated 24 Forbes, JM et al. The breakdown of pre-existing it is as yet not clear what or how. to intracellular mechanics of cell growth or apop- advanced glycation end Torrent Pharmaceuticals is taking the compound tosis, and so may provide valuable adjuncts to products is associated with TRC4186 into clinical trials as an AGE-breaker. Its other therapies. reduced renal fibrosis in chemical mode of action is obscure; the compound A more exciting application of ECM-modulat- experimental diabetes. The was identified by randomly reacting highly AGEd ing therapeutics is in Regenerative Medicine. FASEB Journal, 2003. 17(12): p. 1762-1764. serum albumin (made by incubating BSA with Regenerative Medicine seeks to repair disease- or 25 Pathak, P et al. TRC4149, a 1.67M glucose for 16 weeks) with collagen, a age-related damage through one of two broad novel Advanced Glycation End chemistry that bears little relationship to AGE strategies: stimulating endogenous repair mecha- Product breaker improves cross-links in vivo, and then searching for com- nisms that have been damaged or down-regulated, hemodynamic status in pounds that released the BSA. TRC4186 appears or providing exogenous cells or tissues. It is now diabetic spontaneously 25 hypertensive rats. Eur. J. Med. beneficial in animal hypertension models but no well-known that cells grown in tissue culture Res, 2008. 13: p. 388 – 398. 26 human efficacy data has been released . mimic in vivo cell behaviour better if the cells are 26 Chandra, KP et al. Phase I In the absence of reliable chemical or enzymat- grown in a 3D matrix mimicking the ECM, rather Clinical Studies of the ic methods of cleaving glucosepane, a supply of than on flat glass or plastic surfaces, and that, far Advanced Glycation End- pure material to screen, any reliable high- from being inert, many components of the ECM product (AGE)-Breaker 2 TRC4186. Clinical Drug Invest, throughput assay methods, and no realistic have powerful signalling effects on cells . Even 2009. 29(9): p. 559 – 575. prospect for stopping its formation, AGE-breaker cell growth that is dependent on growth factors 27 Wang, T, Kartika, R and strategies remain speculative at the moment. such as TGF-␤ is in fact also dependent on the Spiegel, DA. Exploring Post- Development of technology in any of these areas mechanical properties of the ECM in vivo, translational Arginine could open AGE-breaking to be a third approach because TGF-␤ is stored in the ECM as bound Modification Using Chemically Synthesized Methylglyoxal to ECM cross-link clearance. complexes, and released by cell-specific proteases Hydroimidazolones. Journal of 30 Other AGEs accumulate in tissues, but are prob- or mechanical tension . the American Chemical ably of less physiological importance. The cross- It is now understood that the quality of the ECM Society, 2012. 134(21): link pentosidine has been the subject of much in vivo can also be decisive in determining how p. 8958-8967. research because it is easy to detect, but is a minor well cells function to repair tissue after damage. component compared to glucosepane. Monovalent Several studies have suggested that the failure of AGEs such as CML and CEL are also probably of older tissue to repair itself as well as young tissue Continued on page 70

Drug Discovery World Winter 2013/14 69 Drug Discovery

Continued from page 69 is due at least as much to the ‘old’ ECM as to the subjects as diverse as drug chemistry, company law cells themselves being old. For example, the and extraterrestrial life, and four books. He is cur- 28 Butcher, DT, Alliston, T and replicative and differentiation potential of mes- rently a Staff Scientist at MIT, Senior Research Weaver, VM. A tense situation: forcing tumour progression. enchymal stem cells from old mice could be ‘res- Associate and Programme Manager of the SRF lab- Nat Rev Cancer, 2009. 9(2): cued’ by growing them on ECM generated by oratory at Cambridge University, and wants to get p. 108 – 122. young stem cells31. Kang and Lightman tracked rid of his wrinkles! 29 Lentini, A et al. regenerating peripheral nerves in the mouse and Transglutaminases: key showed that the poorer regeneration in old mice regulators of cancer metastasis. Amino Acids, 2013. over young ones was due to the lack of ability of 44(1): p. 25-32. the growing axon tips to clear debris from dam- 30 Tenney, RM and Discher, aged nerves from their path. In the absence of DE. Stem cells, debris, they grew and innervated muscle as well as microenvironment mechanics, nerves from young animals32. Loss of functional and growth factor activation. Current Opinion in Cell fibroblasts in the dermis of the skin, and hence loss Biology, 2009. 21: p. 630 – 635. of skin strength, elasticity and thickness with ‘nor- 31 Sun, Y et al. Rescuing mal’ ageing is primarily driven by degradation of replication and osteogenesis of the collagen matrix to which the cells attach, aged mesenchymal stem cells including glycation33. Thus ECM repair is likely to by exposure to a young extracellular matrix. FASEB be a key component of either of the regenerative 34 Journal, 2008. 25: medicine strategies . p. 1474 – 1485. The structural proteins of the body, especially 32 Kang, H and Lichtman, JW. collagen, have traditionally been viewed as non- Motor Axon Regeneration and starters as drug discovery targets, because they are Muscle Reinnervation in Young Adult and Aged Animals. not obviously active as catalysts or signal transduc- Journal of Neuroscience, 2013. tion molecules. The recent work summarised above 33(50): p. 19480 – 19491. shows that this is not necessarily a good argument 33 Fisher, GJ, Varani, J and for ignoring the ECM. As an alternative route to Voorhees, JJ. Looking older: identifying the role of every gene in every cell in a fibroblast collapse and therapeutic implications. complex disease such as chronic inflammation or in Archives of Dermatology, cancer, targeting damage to the ECM may prove a 2008. 144(5): p. 666 – 672. valuable approach to new treatments for a range of 34 Kurtz, A and Oh, S-J. Age diseases and disabilities of old age. DDW related changes of the extracellular matrix and stem cell maintenance. Preventive Medicine, 2012. 54, Dr William Bains is an academic, scientist and Supplement(0): p. S50-S56. entrepreneur. After an academic career in the UK and the USA, William worked in technical consul- tancy and venture capital before founding his first start-up in 1998. He teaches company creation and financing at Cambridge and Warwick Universities, and is author of more than 90 research papers on

ADVERTISEMENT INDEX

BioTek Instruments, Inc 17 Electrospinning Company Ltd 48 Seahorse Bioscience, Inc 8,64 Cellular Dynamics International 34 ForteBio, Inc 13 Singulex 59 CISBIO International SA 33 HTStec Ltd 24 TAP Biosystems 55 Charles River Laboratories International, Inc 3 IntelliCyt Corporation 6 Tecan Schweiz AG IFC Corning, Inc 47 InvivoSciences, Inc 40 TTP LabTech Ltd IBC eBioscience, Inc 4 Scivax USA, Inc 53 Waters Corporation OBC

70 Drug Discovery World Winter 2013/14