exploring biologically relevant chemical space with … para estudo...exploring a different aspect...

Exploring biologically relevant chemical space with metalcomplexesEric Meggers

Altering biological processes with small synthetic molecules is

a general approach for the design of drugs and molecular

probes. Medicinal chemistry and chemical biology are focused

predominately on the design of organic molecules, whereas

inorganic compounds find applications mainly for their

reactivity (e.g. cisplatin as a DNA-reactive therapeutic) or

imaging properties (e.g. gadolinium complexes as MRI

diagnostics). In such inorganic pharmaceuticals or probes,

coordination chemistry in the biological environment or at the

target site lies at the heart of their modes of action. However,

past and very recent results suggest that it is also worth

exploring a different aspect of metal complexes: their ability to

form structures with unique and defined shapes for the design

of ‘organic-like’ small-molecule probes and drugs. In such

metal–organic compounds, the metal has the main purpose to

organize the organic ligands in three-dimensional space. It is

likely that such an approach will complement the molecular

diversity of organic chemistry in the quest for the discovery of

compounds with superior biological activities.

Addresses

Department of Chemistry, University of Pennsylvania,

231 South 34th Street, Philadelphia, PA 19104, USA

Corresponding author: Meggers, Eric ([email protected])

Current Opinion in Chemical Biology 2007, 11:287–292

This review comes from a themed issue on

Combinatorial chemistry and molecular diversity

Edited by Gregory A Weiss and Richard Roberts

Available online 4th June 2007

1367-5931/$ – see front matter

# 2007 Elsevier Ltd. All rights reserved.

DOI 10.1016/j.cbpa.2007.05.013

IntroductionThe identification of compounds with novel and defined

biological functions is of high importance for research in

medicinal chemistry and chemical biology. The total num-

ber of theoretically accessible compounds with biological

activity span the ‘biologically relevant chemical space’ [1].

Charting this subset of the chemical space is focused

primarily on small organic molecules. In this respect one

might wonder whether organic-based scaffolds are capable

of covering all areas of the biologically relevant chemical

space in an efficient fashion. To address this question, it is

interesting to explore the opportunities of inorganic

elements to help build small compounds with defined

three-dimensional structures. Transition metals appear

www.sciencedirect.com

especially appealing for this purpose because they can

support a multitude of coordination numbers and geome-

tries that go far beyond the linear (sp-hybridization),

trigonal planar (sp2-hybridization) and tetrahedral (sp3-

hybridization) binding geometries of carbon (Figure 1).

For example, it is intriguing that an octahedral center with

six different substituents is capable of forming 30 stereo-

isomers compared with just two for an asymmetric tetra-

hedral carbon [2]. Thus, by increasing the number of

substituents from four (tetrahedral center) to six (octa-

hedral center), the ability of the center to organize sub-

stituents in the three-dimensional space increases

substantially. In addition, using a hexavalent center could

provide new synthetic opportunities for accessing globular

shapes by building structures from a single center in six

different directions.

Inorganic pharmaceuticals play an important role in

clinical therapy (e.g. cisplatin) and diagnostics (e.g.

MRI contrast agents) [3]. For these classes of inorganic

compounds the coordination chemistry itself is at the

heart of the mode of action. This short review is limited

to past and current research activities that aim to design

metal complexes with biological activities in which the

metal center mainly serves as a structural center for

organizing the presentation of organic ligands at the

binding site of protein targets.

Pioneering work by DwyerMore than half a century ago, the Australian chemist

Francis Dwyer started to investigate the biological activi-

ties of simple coordination complexes such as ruthenium

complexes with 2,20-bipyridine and 1,10-phenanthroline

(phen) ligands [4–8]. He discovered that some very

hydrophobic complexes (e.g. Figure 2a) displayed bacter-

iostatic and bacteriocidal activities and were capable of

inhibiting tumor growth in a mouse xenograph model

[5,6]. Interestingly, fluorescence microscopy studies

demonstrated that these dicationic ruthenium complexes

were able to pass the cellular membrane and were found

to localize in the mitochondria [6]. The ruthenium com-

plexes also caused paralysis and respiratory failure after

intraperitoneal (IP) injection into mice at high concen-

trations, apparently owing to their potent direct inhibition

of acetylcholinesterase [7]. Investigations into the bio-

logical stability of [Ru(phen)3](ClO4)2] confirmed that the

ruthenium complex after IP injection into mice was not

metabolized and excreted mainly unchanged in the urine

[8]. Because of the coordinative saturation of the coordi-

nation sphere, their chemical and biological stabilities,

Dwyer drew the following important conclusion: ‘Such


mailto:[email protected]

http://dx.doi.org/10.1016/j.cbpa.2007.05.013

288 Combinatorial chemistry and molecular diversity

Figure 1

Transition metals provide an expanded set of coordination geometries for the generation of molecular diversity.

doubly charged pre-formed Ru(II) chelates are stable in

boiling concentrated acids or alkali and in animal tissues.

Hence their biological effects depend solely on the phy-

sico-chemical properties of the Ru(II) complex cation as a

whole since no ruthenium ion or ligand is liberated’ [6].

Ruthenium complexes as protein kinaseinhibitorsFollowing Dwyer’s spirit, Meggers and coworkers

developed organoruthenium compounds as protein kinase

inhibitors [9,10,11�,12,13�,14]. For this, the class of ATP-

competitive indolocarbazole alkaloids (e.g. staurosporine,

Figure 3a) was used as a lead structure. The indolocarba-

zole alkaloid scaffold was replaced with simple metal

complexes in which the main features of the indolocarba-

zole aglycon are retained in the metal-chelating pyrido-

carbazole ligand, whereas the carbohydrate is replaced by a

ruthenium fragment. Following this design strategy, nano-

molar and even subnanomolar ATP-competitive ruthe-

nium-based inhibitors for different protein kinases were

discovered (e.g. Figures 2b and 3b), some of them by

combinatorial chemistry [12,14]. These compounds are

air-stable, stable in water, and can even withstand milli-

molar concentrations of thiols. This stability is due to the

inert character of typical coordinative bonds to ruthenium.

This, together with the modest price of its starting material

RuCl3, its low toxicity, and its predictable and established

synthetic chemistry, makes ruthenium possibly the most

attractive metal for establishing octahedral or pseudo-octa-

hedral coordination geometries. It was furthermore demo-

nstrated that such ruthenium compounds can function

within mammalian cells [10,13�], Xenopus embryos [10]

and zebrafish embryos [13�]. Recent cocrystal structures

confirm that these compounds bind to the ATP-binding

site of protein kinases [11�]. For example, Figure 3b shows

an organoruthenium half-sandwich complex bound to the

ATP-binding site of Pim-1. Importantly, the ruthenium

center is not involved in any direct interactions and has

solely a structural role. Furthermore, superposition of the

binding positions of cocrystalized ruthenium complex and

staurosporine (PDB code 1YHS) with Pim-1 demonstrates


how closely the ruthenium complex mimics the binding

mode of staurosporine (Figure 3c). The pyridocarbazole

ligand perfectly mimics the position of the indolocarbazole

moiety of staurosporine, whereas the cyclopentadienyl ring

and the CO group occupy the binding position of the

carbohydrate moiety of staurosporine.

Mimicking the structure of natural productswith metal complexesWhereas the work on metal-based kinase inhibitors used

the class of indolocarbazole alkaloids primarily as an

inspiration, it is also desirable to mimic the structure

and function of natural products more closely to gain

more efficient access to a complex three-dimensional

structure and to add desired physicochemical properties.

The latter is what Katzenellenbogen and coworkers had

in mind when they designed rhenium complexes as

mimics of the structure of steroid hormones [15–17].

Steroid receptors are often overexpressed in cancer cells

and are therefore a target in the design of radiolabeled

small-molecule diagnostics (e.g. Tc-99m) and thera-

peutics (e.g. Re-186 and Re-188) for cancer detection

and treatment, respectively [16]. Katzenellenbogen’s

group reported the bis-bidentate oxorhenium(V) complex

shown in Figure 2d, which is a remarkable structural and

stereochemical mimic of the androgen 5a-dihydrotestos-

terone (DHT) as can also be seen from the comparison of

the two space-filling models in Figure 2d [17]. An exciting

aspect of this study is that the metal center fulfils a dual

purpose. Unfortunately, the instability of such rhenium

compounds with bidentate ligands precludes in vivoapplications.

Bioorganometallic chemistry with ferroceneThe organometallic sandwich compound ferrocene is an

interesting building block for the design of biologically

active molecules because of its unique structure, its

robustness in aqueous solutions, and its favorable redox

properties [18]. For example, Metzler-Nolte and cow-

orkers incorporated the organometallic amino acid 10-aminoferrocene-1-carboxylic acid into peptides to induce


Biologically relevant metal complexes Meggers 289

Figure 2

Towards chemical biology with metal complexes. (a) Ruthenium complex with antibacterial and antitumor activity. (b) Nanomolar inhibitors

for the protein kinases GSK-3a (left) and MSK1 (right). (c) 10-Aminoferrocene-1-carboxylic acid induces turns in peptides. (d) A bis(bidentate)

rhenium(V) complex mimics the structure of dihydrotestosterone. (e) Organometallic estrogen receptor modulator ferrocifen derived from

the drug tamoxifen. (f) [Fe(EDTA)(H2O)]� binds to the periplasmic nickel transporter NikA. (g) Copper complex as HIV-1 protease inhibitor.

(h) AMD3100 binds with high affinity to the CXCR4 coreceptor and this binding is enhanced in the presence of zinc ions.

turns (Figure 2c) [19�]. In this system, the peptides are

arranged in an antiparallel b-sheet-like arrangement

stabilized by intramolecular hydrogen bonds. Because

turn structures and b-sheets are important recognition

motifs in protein–protein interactions, it can by expected

that ferrocene serves as an interesting building block for

turn-containing peptidomimetics.

In another application of ferrocene in bio-organometallic

chemistry, Jaouen and coworkers developed simple


organometallic analogs of tamoxifen [20]. Tamoxifen

(Figure 2e) is an antagonist of the estrogen receptor

(ER) and widely used in the clinic for the treatment of

hormone-dependent breast cancer. The organometallic

analog ferrocifen (Figure 2e), which bears an additional

hydroxyl group and has one phenyl group replaced by a

bulky and hydrophobic ferrocene moiety, exhibits strong

antiproliferative effects [21]. It is thought that ferrocifen,

in analogy to tamoxifen, suppresses estradiol-mediated

DNA transcription by binding to ERa. However, it was



Figure 3

Ruthenium complexes as ATP-competitive protein kinase inhibitors. (a) Shape mimicry: staurosporine as a lead structure. (b) Cocrystal structure

of a pseudo-octahedral ruthenium half-sandwich compound with protein kinase Pim-1 (PDB code 2BZI). (c) The superimposed cocrystal

structure of Pim-1 with staurosporine (gray color, PDB code 1YHS) demonstrates the close match in binding mode between the natural product

and the organometallic complex.

recently discovered that ferrocifen displays strong anti-

proliferative activities both in hormone-dependent (con-

tain ER) and hormone-independent (lack ER) breast

cancer cells, indicating a distinguished mode of action

[20]. Because the ruthenium derivative does not show the

same cytotoxicity profile, it is assumed that the prominent

redox properties of ferrocene play an important role. A

sequence of oxidation, deprotonation, and further oxi-

dation could generate an electrophile that is capable of

damaging DNA or reacting with proteins [22�]. Thus, in

ferrocifene, the organometallic unit fulfils both a struc-

tural and reactive role.

Related uses of metal complexesIt is apparent that for in vitro and in vivo applications, metal

complexes with kinetically inert bonds are highly desir-

able. However, multidentate ligands such as the hexaden-

tate ligand ethylenediaminetetraacetate (EDTA) can

form highly stable complexes even with kinetically very


labile metals. Fontecilla-Camps and coworkers found ser-

endipitously that the heptacoordinated pentagonal bipyr-

amidal iron complex [Fe(EDTA)(H2O)]� (Figure 2f)

binds tightly to the periplasmic nickel transport protein

NikA [23�]. A 1.8 A cocrystal structure is shown in Figure 4

and demonstrates that the complex is bound through a

combination of hydrogen bonds, electrostatic, and hydro-

phobic contacts. Accordingly, carboxylate groups of EDTA

hydrogen bond with Arg97 and Arg137 and form three

additional water-mediated contacts. This is supplemented

by a set of hydrophobic contacts between methylene

groups of EDTA and Met27, Tyr22, Trp100, and

Trp398. Especially the two tryptophans serve as wedges

that complement the shape of the iron EDTA complex.

Importantly, the iron does not form any direct coordinative

bond with NikA residues and thus has mainly the purpose

to organize the structure of the EDTA ligand. However,

one can postulate an unshielded cation-p interaction be-

tween the iron ion and Trp398, with an indole-to-metal


Biologically relevant metal complexes Meggers 291

Figure 4

Binding of the iron complex [Fe(EDTA)(H2O)]� to the periplasmic

nickel transporter NikA (PDB code 1ZLQ). Important amino acids are

highlighted. W, water.

distance of 5.5 A [23�]. Interestingly, NikA has the natural

function to bind to Ni(II) ions; however, the affinity to

[Fe(EDTA)(H2O)]� exceeds substantially the affinity to

Ni(II) ions, and thus the authors suggest that a similar

natural metallophore is involved in periplasmic Ni(II)

transport by NikA.

A promising strategy for the recognition of proteins by

metal complexes makes use of a combination of recog-

nition through the ligand sphere with direct coordination

to the target site. Only two examples will be discussed

here. Reboud-Ravaux and coworkers developed copper

complexes as substrate competitive inhibitors for HIV-1

protease [24,25]. For example, [bis-(2-pyridylcarbonyl)-

amido] copper(II) nitrate dihydrate (Figure 2g) binds

with an inhibition constant of 480 mM [24]. Molecular

modeling suggests that the catalytic water between Asp25

and Asp125 of HIV-1 protease is directly coordinated to

the Cu(II) ion. A current drawback of this class of com-

pounds is the hydrolytical instability of the Cu(II) com-

plexes. One could potentially overcome this problem by

designing suitable multidentate ligands or by choosing

more kinetically inert metal ions.

Sadler and coworkers investigated the binding of metal

complexes of 1,4,8,11-tetraazacyclotetradecane (cyclam)

macrocycles to the CXCR4 coreceptor and lysozyme as a

model protein (Figure 2h) [26�,27]. In such metallocyclam

complexes, the metal is supposed to function by controlling

the conformation and configuration of the macrocycle.

Additionaldirect coordinativebonds with the targetprotein

can be formed with the vacant axial coordination sites. One


of the most potent members of this family is the bicyclam

AMD3100 (Figure 2), which is in clinical trials for the

treatment of AIDS. The anti-HIV activity correlates with

its binding to the coreceptor protein CXCR4. Interestingly,

the complexation of Zn(II) enhances the binding strength

to CXCR4 and also its anti-HIV activity [28].

ConclusionsMetal compounds provide new opportunities for building

structures with unique and defined three-dimensional

globular shapes in an economical fashion, and thus comp-

lement molecular diversity created by purely organic

scaffolds. This access to unexplored chemical space could

lead to the discovery of molecules with unprecedented

properties. A key aspect for using metal-containing com-

pounds as structural scaffolds is the kinetic stability of the

coordination sphere in the biological environment. This

can be accomplished with multidentate ligands or, more

generally, by employing kinetically inert metals such as

ruthenium. Future work will increasingly rely on combi-

natorial chemistry of metal complex libraries followed by

screening against selected targets or even phenotypic

assays. Furthermore, it is also appealing to complement

the structural role of metals with their special physico-

chemical, reactive, and/or catalytic properties in one

molecule to yield compounds with new abilities to probe

and modulate biological processes.

AcknowledgementsEM thanks the University of Pennsylvania and the US National Institutesof Health (R01 GM071695) for financial support.

References and recommended readingPapers of particular interest, published within the annual period ofreview, have been highlighted as:

� of special interest�� of outstanding interest

1. Koch MA, Schuffenhauer A, Scheck M, Wetzel S, Casaulta M,Odermatt A, Ertl P, Waldmann H: Charting biologically relevantchemical space: a structural classification of natural products(SCONP). Proc Natl Acad Sci USA 2005, 102:17272-17277.

2. Miessler GL, Tarr DA: Coordination Chemistry I: Structures andIsomers. In Inorganic Chemistry. Prentice Hall, 3rd edition,2003:299-336.

3. Orvig C, Abrams MJ, (Eds.): Medicinal inorganic chemistry.Chem Rev 1999, 99:2201–2204.

4. Dwyer FP, Gyarfas EC, Rogers WP, Koch JH: Biological activityof complex ions. Nature 1952, 170:190-191.

5. Dwyer FP, Mayhew E, Roe EMF, Shulman A: Inhibition ofLandschutz ascites tumour growth by metal chelatesderived from 3,4,7,8-tetramethyl-1,10-phenanthroline.Br J Cancer 1965, 19:195-199.

6. Dwyer FP, Reid IK, Shulman A, Laycock GM, Dixson S: Thebiological actions of 1,10-phenanthroline and 2,20-bipyridinehydrochlorides, quaternary salts and metal chelates andrelated compounds. 1. Bacteriostatic action on selectedGram-positive, Gram-negative and acid-fast bacteria.Aust J Exp Biol Med Sci 1969, 47:203-218.

7. Dwyer FP, Gyarfas EC, Wright RD, Shulman A: Effect of inorganiccomplex ions on transmission at a neuromuscular junction.Nature 1957, 179:425-426.



8. Koch JH, Rogers WP, Dwyer FP, Gyarfas EC: The metabolic fateof tris-1,10-phenanthroline ruthenium-106 (II) perchlorate, acompound with anticholinesterase and curare-like activity.Aust J Biol Sci 1957, 10:342-350.

9. Bregman H, Williams DS, Atilla GE, Carroll PJ, Meggers E: AnOrganometallic inhibitor for glycogen synthase kinase 3.J Am Chem Soc 2004, 126:13594-13595.

10. Williams DS, Atilla GE, Bregman H, Arzoumanian A, Klein PS,Meggers E: Switching on a signaling pathway with anorganometallic ruthenium complex. Angew Chem Int Ed Engl2005, 44:1984-1987.

11.�

Debreczeni JE, Bullock AN, Atilla GE, Williams DS, Bregman H,Knapp S, Meggers E: Ruthenium half-sandwich complexesbound to protein kinase Pim-1. Angew Chem Int Ed Engl 2006,45:1580-1585.

This article demonstrates nicely how organometallic compounds canmake use of their unique structural opportunities to fill an enzyme activesite.

12. Bregman H, Carroll PJ, Meggers E: Rapid access to unexploredchemical space by ligand scanning around a rutheniumcenter: discovery of potent and selective protein kinaseinhibitors. J Am Chem Soc 2006, 128:877-884.

13.�

Atilla-Gokcumen GE, Williams DS, Bregman H, Pagano N,Meggers E: Organometallic compounds with biologicalactivity: a very selective and highly potent cellular inhibitor forglycogen synthase kinase 3. ChemBioChem 2006, 7:1443-1450.

The presented results demonstrate the usefulness of organometalliccompounds as molecular probes in cultured cells and whole organisms.

14. Bregman H, Meggers E: Ruthenium half-sandwich complexesas protein kinase inhibitors: an N-succinimidyl ester for rapidderivatizations of the cyclopentadienyl moiety. Org Lett 2006,8:5465-5468.

15. Chi DY, Katzenellenbogen JA: Selective formation ofheterodimeric bis-bidentate aminothiol-oxometal complexesof rhenium(V). J Am Chem Soc 1993, 115:7045-7046.

16. Hom RK, Katzenellenbogen JA: Technetium-99m-labeledreceptor-specific small-molecule radiopharmaceuticals:recent developments and encouraging results. Nucl Med Biol1997, 24:485-498.

17. Hom RK, Chi DY, Katzenellenbogen JA: Heterodimeric bis(aminothiol) complexes of oxorhenium(V) that mimic the structureof steroid hormones. Synthesis and stereochemical issues.J Org Chem 1996, 61:2624-2631.

18. Van Staveren DR, Metzler-Nolte N: Bioorganometallic chemistryof ferrocene. Chem Rev 2004, 104:5931-5985.

19.�

Barisic L, Cakic M, Mahmoud KA, Liu Y, Kraatz H-B, Pritzkow H,Kirin SI, Metzler-Nolte N, Rapic V: Helically chiral ferrocenepeptides containing 10-aminoferrocene-1-carboxylic acidsubunits as turn inducers. Chemistry 2006, 12:4965-4980.


The authors present a detailed structural study of peptides containing theorganometallic amino acid 10-aminoferrocene-1-carboxylic acid.

20. Jaouen G, Top S, Vessieres A, Leclercq G, McGlinchey MJ: Thefirst organometallic selective estrogen receptor modulators(SERMs) and their relevance to breast cancer. Curr Med Chem2004, 11:2505-2517.

21. Top S, Tang J, Vessieres A, Carrez D, Provot C, Jaouen G:Ferrocenyl hydroxytamoxifen: a prototype for a new rangeof oestradiol receptor site-directed cytotoxics.Chem Commun 1996:955-956.

22.�

Hillard E, Vessieres A, Thouin L, Jaouen G, Amatore C: Ferrocene-mediated proton-coupled electron transfer in a series offerrocifen-type breast-cancer drug candidates. Angew ChemInt Ed Engl 2006, 45:285-290.

This work proposes a mechanism for the oxidative activation offerrocifen-type organometallics.

23.�

Cherrier MV, Martin L, Cavazza C, Jacquamet L, Lemaire D,Gaillard J, Fontecilla-Camps JC: Crystallographic andspectroscopic evidence for high affinity binding ofFeEDTA(H2O)S to the periplasmic nickel transporter NikA.J Am Chem Soc 2005, 127:10075-10082.

The crystal structure is a wonderful example of how coordination com-plexes can fit perfectly into protein binding sites.

24. Lebon F, de Rosny E, Reboud-Ravaux M, Durant F:De novo design of a new copper chelate molecule actingas HIV-1 protease inhibitor. Eur J Med Chem 1998,33:733-737.

25. Lebon F, Boggetto N, Ledecq M, Durant F, Benatallah Z, Sicsic S,Lapouyade R, Kahn O, Mouithys-Mickalad A, Deby-Dupont G,Reboud-Ravaux M: Metal-organic compounds: a newapproach for drug discovery N1-(4-methyl-2-pyridyl)-2,3,6-trimethoxybenzamide copper(II) complex as an inhibitorof human immunodeficiency virus 1 protease.Biochem Pharmacol 2002, 63:1863-1873.

26.�

Hunter TM, McNae IW, Liang X, Bella J, Parsons S,Walkinshaw MD, Sadler PJ: Protein recognition of macrocycles:binding of anti-HIV metallocyclams to lysozyme.Proc. Natl. Acad. Sci. USA 2005, 102:2288-2292.

This work provides a structural basis for the design of macrocycles andmetallo-macrocycles that bind with high affinity to G-coupled receptorsand other proteins.

27. Paisey SJ, Sadler PJ: Anti-viral cyclam macrocycles: rapidzinc uptake at physiological pH. Chem Commun (Camb)2004:306-307.

28. Este JA, Cabrera C, de Clercq E, Struyf S, van Damme J,Bridger G, Skerlj RT, Abrams MJ, Henson G, Gutierrez A et al.:Activity of different bicyclam derivatives against humanimmunodeficiency virus depends on their interaction withthe CXCR4 chemokine receptor. Mol Pharmacol 1999,55:67-73.


exploring biologically relevant chemical space with … para estudo...exploring a different aspect...

Documents