Current Topics in Medicinal Chemistry, 2013, 13, 821 836...

Send Orders of Reprints at [email protected] Current Topics in Medicinal Chemistry, 2013, 13, 821-836 821

Form and Function in Cyclic Peptide Natural Products: A Pharmacoki-netic Perspective

Andrew T. Bockus, Cayla M. McEwen and R. Scott Lokey*

Department of Chemistry and Biochemistry, University of California Santa Cruz, Santa Cruz, CA 95064, USA

Abstract: The structural complexity of many natural products sets them apart from common synthetic drugs, allowing them to access a biological target space that lies beyond the enzyme active sites and receptors targeted by conventional small molecule drugs. Naturally occurring cyclic peptides, in particular, exhibit a wide variety of unusual and potent bio-logical activities. Many of these compounds penetrate cells by passive diffusion and some, like the clinically important drug cyclosporine A, are orally bioavailable. These natural products tend to have molecular weights and polar group counts that put them outside the norm based on classic predictors of “drug-likeness”. Because of their size and complex-ity, cyclic peptides occupy a chemical “middle space” in drug discovery that may provide useful scaffolds for modulating more challenging biological targets such as protein-protein interactions and allosteric binding sites. However, the relation-ship between structure and pharmacokinetic (PK) behavior, especially cell permeability and metabolic clearance, in cyclic peptides has not been studied systematically, and the generality of cyclic peptides as orally bioavailable scaffolds remains an open question. This review focuses on cyclic peptide natural products from a “structure-PK” perspective, outlining what we know and don’t know about their properties in the hope of uncovering trends that might be useful in the design of novel “rule-breaking” molecules.

Keywords: Cyclic peptides, Pharmacokinetics, Cell permeability, Lipinski’s rules, Cyclosporine A.

INTRODUCTION High-throughput –omics technologies (e.g., shRNA and

expression profiling, deep sequencing, proteomics, me-tabolomics) have delivered vast datasets for biologists to mine for new clues into the molecular basis of disease. One result of these efforts has been the emergence of a new gen-seration of candidate drug targets. Unlike typical enzymes and receptors, many of these potential targets are signaling proteins and transcription factors whose functions are carried out through protein-protein or protein-DNA interactions[1]. Since few of these biomolecules have endogenous low-molecular weight ligands, finding traditional, drug-like small molecules that can modulate their functions represents a ma-jor challenge ahead. Medicinal chemists are therefore charged with the task of identifying scaffolds that are large and complex enough to bind macromolecular interfaces and non-canonical allosteric binding sites, but yet retain drug-like membrane permeability and metabolic stability.

It is our view that cyclic peptides, especially those of natural origin, offer strategic insights into this design prob-lem. Many cyclic peptide natural products are biosynthesized by non-ribosomal peptide synthetases (NRPSs), which, un-like the ribosome, often utilize as substrates both regular and non-proteinogenic amino acids. NRPSs also often incorpo-rate “tailoring” activities such as epimerization, N-methylation, and heterocyclization[2], further increasing the diversity of these natural products. And like other natural

*Address correspondence to this author at the Department of Chemistry and Biochemistry, University of California Santa Cruz, Santa Cruz, CA 95064, USA; Tel: 831-459-1307; Email: [email protected]

products, especially those in the polyketide family, cyclic peptides tend to have properties (e.g., MW, number of polar atoms, total polar surface area) that put them outside conven-tional predictors of “drug-likeness” such as Lipinski’s “Rule of Five”[3, 4]. Yet many of these compounds exhibit drug-like properties, including the ability to penetrate biological membranes. Some cyclic peptides, such as the important immunosuppressive agent cyclosporine A (CSA), are in clinical use as orally administered drugs.

Drug discovery often involves the iterative (or simulta-neous) optimization of pharmacodynamics (PD) on the one hand, related to target binding and specificity, and pharma-cokinetics (PK) on the other, related to factors that impinge on a molecule’s disposition and fate within the body, e.g., absorption, distribution, metabolism, and excretion (ADME). Often the same structural changes that improve PD have a negative impact on PK, and vice versa, driving the need for reliable predictors in both arenas. For natural products, the focus has historically centered on potency and specificity, while PK is often taken for granted since most natural prod-ucts were “designed” by natural selection to avoid metabolic degradation and cross biological membranes, the same fac-tors that impact PK in modern drug discovery.

In addition to membrane permeability, for a compound to be orally bioavailable it must survive metabolism by prote-ases in the gut and first-pass gut and hepatic biliary excretion and metabolism, especially by cytochrome P450 (CYP) en-zymes[5]. The selectivity profiles of metabolic CYPs are biased toward more hydrophobic compounds[6, 7], creating another trade-off for the medicinal chemist to negotiate: more hydrophobic compounds, while they may be more

��/13 $58.00+.00 © 2013 Bentham Science Publishers

822 Current Topics in Medicinal Chemistry, 2013, Vol. 13, No. 7 Bockus et al.

membrane permeable, tend to be cleared more quickly by CYPs in the liver, whereas compounds that are hydrophilic enough to survive clearance may not be permeable enough to reach circulation in the first place. As with permeability, QSAR models of clearance are generally based on training sets comprised of typical small molecules[7], and so predict-ing the clearance of large, “non-Lipinski” molecules remains a challenge.

Cyclization in peptides eliminates charged termini, which can favor both membrane permeability[8] and metabolic stability[9]. Cyclization is also thought to minimize confor-mational entropy losses upon target binding, although some studies have shown the impact of cyclization on binding en-tropy to be more complex than conventional wisdom sug-gests[10, 11]. Nevertheless, while cyclization may improve the PK properties of peptides in general, the systematic de-sign of orally bioavailable cyclic peptide therapeutics (e.g. CSA) will require a deeper understanding of the impact of structure on physicochemical properties in these non-“drug-like” molecules. This in turn will require more direct and systematic investigations into the PK properties of these re-markable compounds, including cell-based permeability and absolute oral bioavailability studies. This review examines cyclic peptide natural products from a PK perspective, with an emphasis on those molecular properties and structural motifs that might be emulated in the design of novel syn-thetic inhibitors against the next generation of protein tar-gets.

Cyclic Peptide Natural Products Classified by Physical Properties

Cyclic peptide natural products can be grouped into the following categories based on a broad survey of their struc-tures and known biological functions: 1) Highly charged molecules (polycationic and polyanionic) that function mostly as antimicrobial agents by disrupting bacterial mem-branes; 2) Non-polar cyclic peptides that contain mostly lipophilic side chains and modifications to the amide back-bone (e.g., N-methylation) and are likely to penetrate eu-karyotic cells by passive diffusion; 3) Cyclic peptides of mixed polarity, most of which are amphiphilic but are not limited to microbial targets; 4) Cyclotides and cysteine-knot proteins, which are small (2-8 kD) proteins with unusual topologies that can show remarkable oral activity.

Within each of these categories, cyclic peptides can be organized into three main sub-groups based on struc-ture/topology: 1) homodetic, cyclized from head-to-tail; 2) heterodetic, cyclized between side chains or from a side chain to one of the termini; and 3) complex, comprised of a mixture of homodetic and heterodetic linkages (Fig. 1). Many heterodetic cyclic peptides have a “lariat” structure formed by cyclization of the C-terminal carboxylic acid onto a side chain, typically via a lactam linkage through a lysine or ornithine residue, or a lactone linkage through serine or threonine. Heterodetic cyclic peptides can also be capped on their amino termini with a lipid tail of varying length and composition. Compounds in the “complex” category include bicyclic peptides and cyclic peptides with knot topologies.

Fig. (1). The three main sub-groups of cyclic peptides based on structure: a) homodetic, b) heterodetic, c) complex.

1. Cationic Antimicrobial Peptides The cationic antimicrobial peptides (AMPs) comprise a

diverse group of compounds whose activities are associated primarily with their ability to interact with and, in many cases, form pores in bacterial membranes (Fig. 2). Most AMPs are aphiphilic peptides that disrupt negatively charged microbial membranes with varying selectivity depending on the structure of the AMP. While some natural AMPs are cyclic, many are linear peptides that adopt various secondary structures upon their interaction with membranes. Despite much evidence that AMPs exert their antimicrobial activity by disrupting membranes, some cationic AMPs have potent antimicrobial activity without disrupting membranes, and some have activities that are associated with inhibition of intracellular targets[12]. Nevertheless, these compounds are not orally bioavailable and tend to show selectivity toward bacterial membranes. Their structures and activities have been reviewed elsewhere[12-14] and will not be taken up further here.

Fig. (2). Antimicrobial Peptides a) gramicidin S, b) polymyxin B, c) tachyplysin, d) bactenecin.

2. Non-Polar Cyclic PeptidesCyclic peptide natural products that are comprised of

mostly neutral, lipophilic side chains and few if any polar residues can be classified as “non-polar” cyclic peptides. These cyclic peptides and depsipeptides exhibit a wide range of biological activities, many of which are associated with intracellular targets in eukaryotes (Table 1). Although the

OH

OH2N

OX

H2N

ONH

homodetic heterodetic

X = N, O

ONH

complex

Arg Trp Cys Phe Arg Val

GlyTyrArgArgCysArg

Cys

Cys

Tyr

Ile

ArgSS

SS

H N2

Arg Leu Cys Arg Ile ValVal

IleArgValCysArg

SS

H N2

HO C2

HO C2

Pro Val Orn Leu

ValOrnLeud-Phe

d-Phe

Pro

a

b

cO

HN

NH2

O

NH

OH

O

HN

NH2

O

NHNH

OHN

O

HN O

HNO

NHO

NH

O

NHO

H2N

NH2

NH2

HO

d

gramicidin S

polymyxin B

tachyplysin

bactenecin

Form and Function in Cyclic Peptide Natural Products Current Topics in Medicinal Chemistry, 2013, Vol. 13, No. 7 823

cell permeabilities of few of these compounds have been investigated directly (e.g., by quantifying diffusion across Caco-2 or other epithelial cell lines), many are assumed to be cell permeable based on an observed correlation between direct in vitro binding and/or activity data and activity in cell-based assays. Because they do not interact with mem-branes electrostatically, these compounds are thought to penetrate membranes via the same classical transcellular route ascribed to most small molecule drugs. However, until detailed in vitro studies are performed to evaluate their mechanism(s) of cell penetration, such hypotheses remain purely speculative. The archetype of the nonpolar cyclic pep-tides is the immunosuppressive natural product cyclosporine A, a cyclic undecapeptide with a molecular weight of ~1200 that has an oral bioavailability in humans of 29% and is pas-sively permeable[15]. The high molecular weights of these compounds compared with traditional small molecule drugs make them excellent model systems for studying the role of structure in passive membrane permeability. Our lab has investigated synthetic cyclic peptides inspired by these natu-ral products to gain a better understanding of the chemical features driving their unusual PK properties. We have shown using simple model systems that the ability to form in-tramolecular hydrogen bonds in low dielectric media corre-lates with passive membrane diffusion[8, 16]. We have also observed that N-methylation patterns that facilitate the for-mation of transannular hydrogen bonds are associated with good passive permeability and oral bioavailability[17]. This result is corroborated by many examples of cyclic peptide natural products whose structures have been solved, either by NMR in low dielectric solvents or in the solid state, or in some cases both. Many of these peptides contain multiple N-methyl and/or Pro residues, and most have extensive in-tramolecular hydrogen bond networks that sequester polar amide groups in the low dielectric of the membrane. While

the total hydrogen bond donor and acceptor counts put most compounds in this class out of compliance with Lipinski’s Rule (<5 donors and <10 acceptors), the number of free, non-hydrogen bonded donors and acceptors puts them well within the acceptable range (Table 2). Although they contain many rotatable bonds and have molecular weights in the ~800-1500 range, solution NMR data suggest that these compounds are relatively rigid, especially in low dielectric media. In this sense, their membrane permeability is consis-tent with the observation by Veber, et al., that rigidity is a better predictor of bioavailability than molecular weight[18]. In that retrospective study, the number of rotatable bonds (which has a roughly linear relationship with MW) was a better predictor of bioavailability than MW. Although the number of rotatable bonds in natural macrocycles may be high, they remain relatively rigid due to constraints imposed by cyclization and local nonbonded interactions.

Non-polar cyclic peptides are characterized by a pre-dominance of aliphatic residues, the presence of D-amino acids, and N-methylation of backbone amides. If they are present, polar residues often form hydrogen bonds between the side chain and backbone, for example in the cases of ar-gyrin B, aureobasidin, and guangomide A (Fig. 3). Few if any cyclic peptides in this category have ionizable side chains, and they rarely contain highly polar neutral residues such as Gln and Asn.

The cyclic peptides in this category whose crystal struc-tures have been solved often contain one or more !-turns that template various patterns of intramolecular hydrogen bonds. These !-turns are defined by a transannular hydrogen bond between the i and i + 3 residues, while the stereochemistry, N-methylation status, and side chain functionality of the i + 1 and i + 2 residues provide the phi and psi restraints that template the turn. Detailed reviews of various subtypes of !

Table 1. Cyclic Peptide Structural Classes.

Classification Structural features Mechanisms of membrane interactions Biological activities

Charged

•cationic More than one cationic residue, e.g., Arg orLys

Destabilize and/or form pores in anionic (usu-ally bacterial) membranes

antimicrobial

•anionic More than one anionic residue, usually Asp orGlu

Interact with membranes by chelation of diva-lent cations

antimicrobial

Polar Up to one charged residue and/or a number ofhighly polar residues (e.g., Asn, Gln). Manyare heterodetic and contain a lipid appended tothe macrocycle

Direct binding to specific phospholipids. Formany in this category the mechanisms of mem-brane permeability are poorly defined

Antimicrobial; various intracellu-lar mammalian targets

Nonpolar Primarily comprised of aliphatic residues;often contain multiple sites of N-methylation;networks of intramolecular hydrogen bonds inlow dielectric media generate defined, lipo-philic conformations; beta-hydroxyl-substituted residues common

Passive (transcellular) diffusion. Active trans-port observed in some cases. Few systems havebeen studied in detail with respect to membranepermeability.

Diverse activities in mammaliancells. Most known targets areintracellular.


Table 2. Selected Cyclic Peptides, their Biological Activities, and Key Structural Features.

Total Free Cyclic Peptide Mw Target in vitroIC50 (nM)

EC50 in cells

(nM)

# Resi-dues

# N-methyls

Donors Acceptors Donors Acceptors

cyclosporine A [37, 38] 1203 Cyclophilin/ calcineurin

30 7-10 11 7 5 12 1 8

cyclomarin A[39, 40] 1043 ClpC1 Subunit of the Caseinolytic

Protease

100 300

2500

7 2 7 11 4 9

argyrin[41, 42] 825 26S protea-some

100 2.8 8 1 8 9 2 3

HUN7293/

CT8[43-47]

893 Sec61 30 10 7 3 3 8 0 5

luzopeptin[48, 49] 1427 dsDNA 53 7 10 4 8 22 0 14

didemnin B[50, 51] 1112 EF-1alpha; inhibits pro-

tein synthesis

90 2.5 9 2 5 12 2 9

aureobasidin A[52] 1114 phosphatidy-linositol: ceramide

phosphoinosi-tol transferase

0.2 90 9 4 4 10 0 6

YM-254890[53, 54] 959 Hetero-trimeric Gaq;inhibits GDP-

GTP ex-change

50 0.1 8 3 5 12 0 8

turns found in proteins[19, 20] and peptides[21], and the residues that favor their formation, have been published. Many of these turns contain Pro and/or Gly residues and D-amino acids, all of which favor !-turns when present at both the i + 1 or i + 2 positions. Both Pro and N-methyl residues can adopt the cis-amide conformation, which, when located at i + 2, induces the type VI !-turns found in many neutral cyclic peptide natural products. Homodetic Non-Polar Cyclic Peptides

The homodetic non-polar cyclic peptides can be further classified into four major categories defined by the types of residues and backbone modifications that they contain: regu-lar (all-amide) peptides, depsipeptides, pattelamide-like pep-tides, and proline-rich peptides. Examples of regular non-polar cyclic peptides are argyrin B, cyclomarin, and CSA (Figs. 3 and 4). Most of the cyclic peptides in this category contain multiple N-methyl residues, yet very few are per-methylated. This trend is consistent with work in our lab

showing that permethylation can decrease membrane perme-ability relative to specific patterns of N-methylation that fa-vor intramolecular hydrogen bonds[17]. Indeed, for those peptides in this class whose structures are known, the vast majority of non-N-methylated amides are involved in tran-sannular hydrogen bonds. While this simple hypothesis is consistent with many cyclic peptide structures (vide infra), other work using similar model systems suggests a more complex role of N-methylation in the active transport of cy-clic peptides[22-25]. Recent reviews treat N-methylation in peptides from synthetic and conformational perspectives[22, 26].

CSA is an 11-amino acid, homoetic non-polar cyclic pep-tide that contains seven N-methyl groups, a single D-amino acid, and a non-proteinogenic prenylated Thr residue. CSA is a clinically used immunosuppressive agent that binds the prolyl cis-trans isomerase cyclophilin. The resulting CSA/cyclophilin complex is a potent inhibitor of calcineurin,


Fig. (3). Selected nonpolar cyclic peptide natural products showing intramolecular hydrogen bonding observed in X-ray crystal structures. a) selenamide, b) argyrin B35, 36, c) callynormine D d) aureobasidin A46, e) didemnin B44, 45, f) destruxin B, g) HUN-7293, h) cyclomarin33, 34, i)

Fig. (4). Structures of cyclic peptides in free (a, c, e) and target-bound (b, d, f) states. a,b) cyclosporine A31, 32; c,d) luzopeptin42, 43; e) FR900359 (R = CH3); f) YM254890 [47, 48] (R = H); Intramolecular hydrogen bonds indicated in orange dotted lines and intermolecular hydrogen bonds indicated in blue dotted lines.

NO

O

O

O

O

HN ONH

O

N

Ph

!eO

O

NH

OHN

OO

Cl

HN

O

O

HOH

H

HO

NH

HN

HN

O!e

O

NH

NH

O

NH O

HN

ON!e

HN ON

SO

H

N!e

ON

NH

O

O

NO

NH

O

!e

N!e

HN

O

OO

ON!e

OHO

ONH

NHO O

OH O

OOO

N!eO

N

O

"

N!e

ON

HNO

O

O!e

!eN

O O

R2

HNO

ON

ONH

O!e

ON

!eNHR1

O NHO

HN

NO

O

HNO

NH

O

O

N!e

NH

OOR

RO

N

O

O!e

!e

O

ON

NH

O

ON

O

O

!e

HN

O

N !e

O

NMe

OO

OO

NH O

O

ON Me

O

NH

H

a b c

d ef

g h i

N

O OHN

OO

NO

N

ONHNH

NHO

O

NO

O

NH

NHO

O

OH

selenamide argyrin B

callynormine D

aureobasidin A

Y = OH: didemnin BY = O: aplidin

destruxin E

HUN-7293cyclomarin guagomide A

N!eO

N!e

ONH

ONH O

N

O

N

OH

H

!e

ON!e O

N!e

O

N

ONH

ONH

O

!e !eN!e

O

N!eO

NH

ONH O

N

O

N

HO

!e

ON!e

O

N!e

O

NO

NH

O

NH

O!e

!e

H

OO

HNO

N!e

O

O

N!e

O!eO

O R'R'

O

O NH

!eNHN

O

NH

OR

OH

O

O

OO

NHO

N!e

O

ON!e

O!eO

O R'R'

OO NH

HNO

HNO

R

OH

O

O!eN

N NH

!eN

O

NOAcO

NH

O

O

NHO

O!e

O

N!eHO

O

O

NHNN!e O

NAcO O

HN

O

O

N OH

O!e

O

!eN OH

O

O

NHN !e

NO

NOAcO

NH

O

O

NHO

O!e

O O

O

!eN OH

NNH

N!e O

NAcO O

HN

O

O

N OH

!eO

OO

O

N!eHO

a b

c d

e f


a central phosphatase involved in T-cell activation[27]. Thus, CSA has the unusual function of inducing an inhibi-tory protein-protein interaction. The structure of CSA has been determined both in the free state, by X-ray crystallog-raphy and NMR[28, 29], and in the ternary complex with cyclophilin and calcineurin[30]. In the free state, in both the crystal structure and the NMR solution structure in CDCl3,CSA adopts a rectangular conformation characterized by three transannular hydrogen bonds and an external, "-turn forming hydrogen bond (Fig. 4a). In contrast, in the crystal structure of the ternary complex with cyclophilin and cal-cineurin, CSA adopts a more open conformation in which many of the polar groups point outward, forming intermo-lecular hydrogen bonds with both protein partners (Fig. 4b)[30]. Interestingly, the non-polar conformation of CSA observed in CDCl3 was also significantly populated in 50% aqueous methanol[31]. This type of conformational flexibil-ity may be common to nonpolar cyclic peptides and has been identified in at least two other cases (see below). Whether this behavior contributes to the unexpected passive perme-ability of this class has not been examined directly.

The homodetic non-polar cyclic depsipeptides have at least one depsi- (i.e., lactone) linkage and include com-pounds such as destruxin B, aureobasidin A, HUN-7293/CT08, and guangomide A (Fig. 3). Whether these lac-tone linkages serve a critical role in the pharmacokinetic properties of these molecules has not been studied systemati-cally, although their prevalence among natural cyclic pep-tides of disparate origins suggests that the lactone, like N-methylation, may serve an important role in enhancing the permeability and/or stability of these compounds.

A well-known cyclic peptide natural product is the K+

ionophore valinomycin, a cell-permeable and cytotoxic cy-clic depsipeptide that acts as a mitochondrial uncoupler by facilitating the passive diffusion of potassium ions across the mitochondrial inner membrane (Fig. 5). The apo-form of valinomycin crystallized from octane shows a rectangular geometry in which all of the amide NH groups are involved in trans-annular hydrogen bonds. These hydrogen bonds de-fine #-turns between the lactic acid C=O groups and the valine NH groups at i and i + 3, respectively (Fig. 5b)[32]. In complex with K+, the molecule rearranges and adopts a toroidal geometry using a different arrangement of # turns in which six of the C=O groups point toward the K+ in the cen-ter of the torus (Fig. 5c)[33]. In both the apo-form and the K+

complex, all six of the NH groups are involved in transannu-lar hydrogen bonds, but the two structures have very differ-ent H-bonding arrangements. In yet another conformation (crystallized from DMSO in the absence of K+) that might be considered a model for its aqueous, apo state, valinomycin adopts a flat, triangular conformation in which three of the amide N-H groups are solvent-exposed, while the remaining three N-H groups point inward forming type II #-turns (Fig. 5d).

The KD between valinomycin and K+ is 1-3 mM in aque-ous solution[34], significantly higher than environmental levels but 10-fold lower than the concentration of K+ in the mitochondrion, suggesting that in order to function as an uncoupler, valinomycin must be cell permeable in both its apo- and K+-complexed forms. That it is able to adopt lipo-

philic conformations in both complexed and uncomplexed forms points toward a mechanism of passive diffusion, while the “open” conformer may account for its water solubility in the apo form. Indeed, valinomycin dissolves in solvents ranging in polarity from water to hexane[34]. Heterodetic Non-Polar Cyclic Peptides

Many non-polar cyclic peptides have heterodetic linkages and include compounds such as coibamide A, didemnin B, YM254890, selenamide, luzopeptin, and callynormine A. Didemnin B (Fig. 3e) is a non-polar lariat cyclic peptide that potently inhibits protein synthesis by binding to the eukary-otic translation elongation factor EF-1! [35-37]. It has low-nM activity in cells and was the first marine natural product to enter human clinical trials. Although clinical trials of didemnin B were discontinued due to toxicity and low effi-cacy, the nearly identical compound aplidin (Fig. 3, also called plitidepsin) has shown promise in phase II/III clinical trials against a variety of cancers[38, 39]. Aplidin has been the subject of in vitro metabolism[40] and in vivo i.v. phar-macokinetics studies and has lower plasma half-life[41] (44 h vs. 0.14 h) and lower cardiotoxicity[42] than didemnin B. No direct cell permeability or oral bioavailability data have been reported for didemnin B or aplidin.

Fig. (5). a) Chemical structure of valinomycin; b-d) schematic dia-grams of X-ray structures showing hydrogen bonding patterns. Valinomycin crystallized from non-polar solvent in the absence (b) and presence (c) of K+ and (d) crystallized from DMSO in the ab-sence of K+.

Another heterodetic lariat peptide, callynormine A (MW 1188), consists of 11 amino acids with an unusual !-# dehy-

O

OO

HN

HNONH

O

ONHK+

NH

HNO

HNO

NH O

NHO HN

ONH

OHN

O

O

NH

HN

HN

HN

O

NH

HN

SO

SO

SO

a

b c

d

O

NHO

O

O

HN

O

O

O

HN

O

O

O NH

O

O

O

HN

O

OO

HN

O Ovalinomycin

#


drodiaminopropionic acid group linking the cyclic octapep-tide portion with the short tripeptide tail (Fig. 3c). While callynormine contains no N-methyl residues, the crystal structure shows multiple hydrogen bonds both within the ring and between the ring and the tail[43]. The lariat dep-sipeptide coibamide (MW 1288) also contains 11 amino ac-ids, with 7 residues in the ring and 4 in the tail, and shows low nM cytotoxicity in various human cancer cell lines. Coi-bamide exhibited an unusual COMPARE fingerprint in the NCI60 cell line screen, suggesting that it has a unique bio-logical target or MOA. In contrast to callynormine A, coi-bamide is extensively N-methylated both in the tail and in the cyclic portion. Although its three-dimensional structure is unknown, its bioactivity and N-methylation pattern are consistent with cell permeability by the same mechanism(s) associated with other members of this family.

Luzopeptin is a heterodetic cyclic peptide that contains two pendant aromatic 3-hydroxy-6-methoxyquinaldic acid groups. It has C2 symmetry and binds to DNA by intercalat-ing both chromophores in a staple-like fashion with the mac-rocycle sitting in the major groove. In both the X-ray crystal structure[44] and the NMR structure in CDCl3[45], the free compound has a collapsed, rectangular structure with in-tramolecular hydrogen bonds between the Gly NH and the sarcosine carbonyl groups, and between the OH groups of the hydroxyvalines to their own carbonyls. In the DNA-bound form, luzopeptin adopts a square conformation in which the Gly NH groups and the hydroxyvaline carbonyls form hydrogen bonds with bases in the major groove[46, 47].

YM-254890 is another potently bioactive heterodetic cy-clic peptide that was first identified as an inhibitor of ADP-induced platelet aggregation[48]. This compound was subse-quently shown to inhibit signaling through G!q, a subfamily of the heterotrimeric guanine nucleotide-binding proteins (G proteins) that transduce signals from ligand-activated G-protein coupled receptors (GPCRs). An activated GPCR stimulates the GDP-to-GTP exchange in a G! protein inside the cell, causing the dissociation of G! from its associated G#" subunit. The free G! and G#" subunits each bind and activate various downstream effectors, which, for the G!qsubtype, is phospholipase C. The cyclic peptide YM-254890 binds to G!q and prevents the exchange of GDP for GTP, locking the G!q subunit in its “off” state[49]. The X-ray crystal structure (of FR900359, a close structural analog of YM-254890)[50] reveals a complex network of intramolecu-lar hydrogen bonds involving all four free amide NH groups and the –OH group of the #-hydroxyleucine residue (Fig. 4e). The molecule undergoes a major conformational rear-rangement upon target binding such that, in the bound con-formation, most of the polar groups point away from the ring forming hydrogen bonds with the protein, with a new in-tramolecular hydrogen bond formed within the macrocycle (Fig. 4f)[49]. Both luzopeptin and YM-254890 are presumed to be cell permeable based on their nM potencies in both invitro and in cell-based assays and the congruence between their cellular phenotypes and in vitro activities, although their cell permeabilities have not been measured directly.

Could the favorable PK properties such as good perme-ability and low clearance of the non-polar cyclic peptides

relate in general to their ability to equilibrate between open, hydrophilic conformers in water and closed, hydrophobic conformers in the membrane? For example, if CSA could be locked into its low-dielectric conformation so that it could not equilibrate into its “open” conformer in aqueous solu-tion, what impact would it have on clearance, permeability, and solubility? How general and/or important is conforma-tional flexibility in determining the ADME properties of neutral cyclic peptides? The impact of such conformational dynamics on permeability and solubility has not been inves-tigated using well-controled model systems, suggesting that more systematic studies are needed in this area.

One study that may shed light on the relationship be-tween structure and cell permeability found that the perme-ability of CSA in Caco-2 increased by two orders of magni-tude over the temperature range from 5 to 37˚C[51]. Equilib-rium sedimentation studies showed a similarly dramatic in-crease in the partial specific volume of CSA over the same range, and a corresponding decrease in its aqueous solubility. Importantly, these authors found that the equilibrium distri-bution of conformers in aqueous solution remained constant over the same temperature range, suggesting that the ob-served increase in permeability was not due to a correspond-ing increase in the relative abundance of the internally hy-drogen bonded conformer at the higher temperature. Like CSA, the permeabilities of salicylic acid and propranolol (two other drugs that penetrate membranes by passive diffu-sion) also increased with increasing temperature, although for these drugs the effect was an order of magnitude less pronounced than that of CSA. These data suggest that the classical, entropy-driven hydrophobic effect drives the ener-getics of membrane permeability for CSA. Could the ability to adopt open, hydrophilic conformers in aqueous solution serve to prevent aggregation in large, non-polar cyclic pep-tides like CSA? Since non-specific, hydrophobic aggregation is presumably also entropically driven, any advantage gained with an increase in non-polar surface area would be counter-acted by a tendency to aggregate. Equilibration among “open” hydrophilic conformers would therefore disfavor aggregation in large, non-polar solutes by decreasing the relative concentration of the “lipophilic” conformer com-pared to more water-soluble ones. Furthermore, how does the size-dependent diffusion coefficient, a factor that predicts decreasing permeability with increasing cross-sectional area[52, 53], factor into the permeability equation in opposi-tion to the classical hydrophobic effect? Applying similar experimental techniques to other cyclic peptides could help tease apart the contributions of each of these factors in de-termining the membrane permeability of large macrocycles. Patellamide-Like Cyclic Peptides

While most nonpolar cyclic peptides are biosynthesized by NRPSs, the patellamide-like cyclic peptides are derived from ribosomally synthesized sequences of alternating lipo-philic and Ser, Thr, or Cys residues. These hydroxyl- and sulfhydryl-containing residues are precursors of the oxa-zoline and thiazoline rings that characterize this class of natural products, formal products of the cyclodehydration between the side chain -OH or -SH groups and the neighbor-ing i - 1 backbone carbonyl group. Oxidation of the oxa-zoline and thiazoline rings leads to the corresponding aro-


matic oxazoles and thiazoles. Examples of the pattelamide-like cyclic peptides include lissoclinamide 7, patellamide D, and ascidiacyclamide (Fig. 6). Some of these, such as the patellamides and ascidiacyclamide, are comprised exclu-sively of these backbone-cyclized residues, while others, such as trunkamide, are hybrids and contain one or more unmodified residues. While the pure patellamide-like pep-tides rarely contain N-methyl groups, consistent with their ribosomal origin, they generally exhibit multiple in-tramolecular hydrogen bonds, either between the 5-membered ring nitrogen lone pair and the neighboring N-H group, or, in some cases, between NH groups and C=O groups within the molecule. Patellamides and ascidiacycla-mide can adopt a square conformation in which the NH groups point into the center of the ring, or a twisted “Figure eight” conformation in which the thiazole rings are stacked and the structure is stabilized by transannular hydrogen bonds. Which of these conformations a particular patella-mide adopts appears to be determined by the nature of the R-groups on the ! carbons, with the square conformation being preferred by C2-symmetric arrangements of R-groups and the twisted conformation favored by asymmetric arrange-ments of R-groups. Although there was one report that lisso-clinamide 7 exhibits nM cytotoxicity in some cell lines[54], this degree of toxicity was not recapitulated in a screen by the NCI[55]. Other peptides in this family have failed to show appreciable cytotoxicity or any other significant bio-logical activity, leading to the speculation that they act as multivalent cation transporters or play some other role in the metabolic maintenance of the producing organism.

Fig. (6). Patellamide like cyclic peptides, a) lissoclinamide 7, b) patellamide D, c) ascidiacyclmide, d) trunkamide

Proline-Rich Cyclic PeptidesThe phakellistatins[56-58] (Fig. 7) are examples of ho-

modetic proline-rich cyclic peptides. These rarely contain N-

methyl residues, although many of them have extensive net-works of internal hydrogen bonds. These peptides are asso-ciated with various bioactivities, although, similar to the pa-tellamide-like peptides, their potencies are moderate, at least in the assays in which they have been screened to date. Hybrid Polyketide-Polypeptide Macrocycles

Conjoined NRPS and polyketide synthase (PKS) mod-ules yield a diverse class of macrocycles comprised of both polyketide and polypeptide fragments[59-61]. These com-pounds display impressive backbone variety and complexity, pushing beyond the limits of the 20 proteinogenic amino acids (and modifications thereof) by incorporating polyketide fragments of varying lengths and functionality. Such scaffolds are biosynthesized in modular fashion by hybrid PK-NRPS Claisen-like mechanisms forming back-bone alkene and #-hydroxy functionality. These molecules commonly possess a polypeptide hemisphere and a polyketide hemisphere, the latter consisting of a carbon chain with multiple stereocenters defined by pendant methyl and hydroxyl groups and endocyclic double bonds.

Fig. (7). Proline-rich cyclic peptide phakellistatin 8.

Apratoxin A (Fig. 8a) is one such 25-member hybrid polyketide-polypeptide macrocycle. Isolated from the marine cyanobacterium Lyngbya majuscula, this secondary metabo-lite exhibits potent activity against human solid tumor cell lines (KB IC50 = 0.52 nM, LoVo IC50 = 0.36 nM) and has shown marginal activity in vivo against colon and mammary cancer[62]. More recent studies have shown that apratoxin A modulates cotranslational translocation[63], although other evidence points toward a mechanism involving association with Hsp/Hsc70 and subsequent stimulation of chaperone-mediated autophagy of specific client proteins[64]. No in-tramolecular hydrogen bonding was observed via CDCl3NMR and subsequent molecular dynamics minimization[62].

Jasplakinolide (Fig. 8b) is a 19-member polyketide cy-clodepsipeptide hybrid isolated from the marine sponge Jaspis. The backbone is composed of a #-Tyr, N-Me bromo-Trp, and Ala tripeptide coupled to a 12-carbon alkene-containing polyketide fragment cyclized through an ester linkage. The natural product exhibits anti-proliferative activ-ity in human cell lines[65, 66] and is one of the few cell permeable small molecules known to stabilize F-actin[67], binding competitively at the phalloidin site with a Kd of 15 nM[68]. The solution and crystal structures display no in-tramolecular hydrogen bonding[69], although the skeleton is quite flexible, capable of adopting several conformations as

N

O

NO

HN PhO

N

SHN

PhO

NS

NH

OON

R4

NH

O R$

NS

HNO

R1NO

HN

OR2

NS

NHO

R%

a

HN

OHNSN

NH

O

NO

O

HN

SN

ON

O

c

b

lissoclinamide 7 patellamide D

ascidiacyclmide

d

O

HN

O NO

HN

N

SO

NH

O NH

O

NHO

O

trunkamide

NH

ON

N

O

O

NH O

HN

ONHNH

O

O

HN

ON

O

N

O

OHphakellistatin 8


a result of the long carbon polyketide fragment[70]. Jas-plakinolide has long been presumed to be cell permeable based on the correspondance between its actin-stabilizing activity in vitro and its dramatic effect on the cytoskeleton in living cells. Its cell permeabilty was confirmed directly, however, in a recent report that used synthetic fluorescent probes of jasplakinolide to image the actin cytoskeleton in live cells[71].

The complex skeletons of hybrid polyketide-polypeptide macrocycles permit the molecules to adopt conformations that are not available to strictly peptide-based scaffolds. The lack of intramolecular hydrogen bonding in the few mole-cules of this class whose 3-D structures have been character-ized suggests that perhaps the hydrophobic nature of the polyketide segment is able to compensate for the polarity imparted by the non-N-methylated amides. However, it should be noted that the mechanism of membrane penetra-tion of these compounds has not been investigated explicitly.

Fig. (8). Examples of cell-permeable hybrid polypeptide-polyketide macrocycles a) apratoxin A, b) jaspakinolide.

3. Cyclic Peptides of Mixed Polarity The various membrane-disrupting and membrane-

penetrating mechanisms attributed to the amphiphilic antimi-crobial peptides are primarily physical in nature, that is, their MOAs do not depend on inhibition of specific macromole-cules. In contrast, there is a class of cyclic peptide natural products that are amphiphilic and yet exhibit activities in mammalian cells that point toward intracellular targets and PK profiles that suggest passive membrane diffusion. These cyclic peptides of mixed polarity encompass a variety of structural types and biological activities. For example, ka-halilide F[72](KF; Fig. 9a), is a 13-amino acid cyclic lariat peptide (C-terminal-to-Thr linked) that is potently cytotoxic in human cells (EC50 ~100 nM) with good selectivity for tumor cells vs. primary human cells[73]. Although its bio-logical activity has been the subject of intense study[73-76], the target and mechanism of action (MOA) remain obscure.

The cyclic portion of KF contains 6 residues and is neu-tral except for a single Orn residue, with a relatively short, 6-carbon aliphatic cap at the N-terminus of the 7-residue tail. While the mechanism of internalization of kahalalide F is unknown, its rapid effect on endosome morphology suggests that its target may be associated with intracellular membrane compartments[76]. Fluorescent[77] and gold nanoparti-cle[78] conjugates of KF were localized to endosomal com-partments, and KF itself causes dramatic swelling of

lysosomes[76] followed by severe disruption of organelle architecture and cell death via a non-apoptotic mecha-nism[73]. KF was also stable to proteolysis and oxidative metabolism in vitro in the presence of pooled human plasma and liver microsomes[79], consistent with its in vivo stabil-ity.

Human PK studies[80-83] showed a rapid clearance from plasma following i.v. infusion, with a steady state volume of distribution (Vss) of ~7 L/kg, indicative of low distribution in tissues. Whether this Vss value is a function of poor cell per-meability or high plasma protein binding was not deter-mined.

Perhaps reflecting the deleterious effect of highly polar substituents on passive membrane diffusion, many polar cy-clic peptides have a lipid appendage. Examples in this group include largamide, pseudodesmin A, and the papuamides (Fig. 9b-f). Pseudodesmin A is a member of a family of “cy-clolipeptides” (CLPs) derived from Pseudomonas spp. char-acterized by a lipid tail appended to an oligopeptide macro-cycle cyclized through a depsi-linkage to a Thr side chain[84]. The crystal structures of several members of this class have been solved, including pseudodesmin A[85], WLIP[86], pseudophomin[87], amphisin[88], and tensin[89], and all adopt a similar conformation in which one half of the macrocycle is folded into a left-handed !-helix stabilized by a network of intramolecular hydrogen bonds (Fig. 9d). The lipid tail is extended along one side of the macrocycle with the lipophilic residues, while the hydrophilic residues project from the opposite side of the macrocycle, creating an am-phiphilic geometry (Fig. 9d, e). Overall, the biological activ-ity of these compounds in various antimicrobial screens is mild to moderate[84], and little has been reported on their effects, if any, on mammalian cells. Further, the mecha-nism(s) by which these molecules interact with and/or pene-trate membranes has not been studied directly.

Papuamide A protects cells against HIV-induced toxicity by interfering with viral entry, with an EC50 of ~100 nM[90]. The very similar papuamide B (Fig. 9f) was shown to interact with membranes in S. cerevisiae by specific bind-ing to phosphatidylserine (PS)[91], although PS binding could not be linked to the anti-viral activity of either papuamide A or B[90].

Thiostrepton (Fig. 9g), and the related molecule siomycin A, are antibiotics that inhibit prokaryotic translation[92], but are also selectively cytotoxic to transformed eukaryotic cells. Siomycin A induced apoptosis in transformed, but not in normal lung fibroblasts, and it selectively killed breast can-cer cells through inhibition of FOXM1 signaling[93]. Sio-mycin A also inhibited tumor growth from glioblastoma stem cells through down-regulation of maternal embryonic leucine zipper kinase (MELK)[94]. More recent studies showed that thiostrepton and siomycin A are both protea-some inhibitors and that this activity is responsible for their ability to decrease FOXM1 levels and selectively kill certain cancer cells[95, 96]. However, another group reported re-cently that the down-regulation of FOXM1 in malignant mesothelioma cells is not related to proteasome inhibition, but rather correlates with the ability of thiostrepton to disable the mitochondrial anti-oxidant pathway by covalently dimer-

a

HNO

NS

OH

OO

NO

NO

N

OO

HN O

O

OH

O

NO

NHBr

NHO

b

apratoxin A

jaspakinolide


izing the protein peroxiredoxin 3 (PRX3) through thiostrep-ton’s multiple dehydroalanine residues[97].

Despite the conflicting hypotheses regarding the MOA of thiostrepton, its target(s) is/are likely intracellular. The X-ray[98] and solution NMR[99] structures of thiostrepton show some intramolecular hydrogen bonding, but also a sig-nificant number of solvent-exposed polar groups, suggesting that thiostrepton is unlikely to traverse membranes by a sim-ple passive diffusion-type mechanism. However, the lack of cationic residues suggests its mechanism of membrane pene-tration is distinct from the pore-forming and membrane-disrupting mechanisms shared by the classical amphiphilic antimicrobial peptides.

Microcin J25 is a lariat “protoknot” peptide with a lactam linkage formed between the N-terminal Gly and an internal Glu side chain (Fig. 9h). It inhibits bacterial transcription by blocking the nucleotide uptake channel of bacterial RNA polymerase[100, 101], and has also been shown to increase

superoxide production in bacteria. Microcin J25 also acts as a mitochondrial uncoupler and stimulates cytochrome C re-lease from isolated, intact mammalian mitochondria[102, 103], although it has minimal apoptotic activity in human cells in culture. In this remarkable structure, the “tail” por-tion of microcin J25 is threaded through and sterically trapped within the ring, forming a lasso topology [104, 105].

The 21-residue compound is mostly hydrophobic and has numerous intramolecular hydrogen bonds, with a single His residue that forms an intramolecular side-chain-to-backbone hydrogen bond. It interacts with artificial liposome prepara-tions, showing selectivity for zwitterionic over negatively charged phospholipids, and transports small ions across the membrane without causing membrane disruption[106, 107]. The interaction of microcin J25 with membranes has also been studied using Langmuir monolayers and was shown to be entropically favorable, suggesting that the classic hydro-phobic effect may drive the thermodynamics of its partition-ing into membranes[108].

Fig. (9). Heterodetic polar cyclic peptides a) kahalilide F (KF), b) largamide, c) pseudodesmin A d) X-ray crystal structure of pseudodesmin A shown as stick diagram e) front and back views of molecular surface based on pseudodesmin A crystal structure; color coded by hydro-philicity: purple = hydrophilic; green = lipophilic; white = neutral; f) papuamide B, g) thiostrepton, h) X-ray crystal structure of microcin J25; side chains have been ommitted for clarity.

h

O

HN

O

NH O

NH2

O NHO

ONH

ONHO

HN

ONH

O

NHO

NH

O

OH

OH

OH

O

HN

O

HN

OHNO

NHO

O

O NH

HN

ONH

OH2N

HNO

N

OHN

ONH

O

OH

HN

ONH

O

ONH

ONH

OH

O

HN

H2N

O

NH

O NH2

O

NHO

O

N O

NH

OHN

ONH O

HN

ONHO

O

OHOH

O

OHHO

O

HN

OHN

O

O

O NHO

HN

NH

O

OH

HN

ONH

O

O

OH

ONH

ONH

O

HN

OHN

N SN

O

O

N

OHNH

OH

NH

OS

NHN O

SN

HNOH

N

O

SN

HO

HO

OH

NS

O

HNO

NHO

H2N

a b

c d

gf

e

front back

kahalilide F largamide

pseudodesmin A

X-ray crystal structure of pseudodesmin A

molecular surface of pseudodesmin A

papuamide Bthiostrepton

X-ray crystal structure of microsin J25 (backbone atoms only)


4. Cyclotides and Cysteine-Knot Proteins Cyclotides[109] are backbone-cyclized “microproteins”

that contain multiple disulfide bonds and a conserved multi-loop structure[110, 111]. Kalata B1, a plant-derived cy-clotide with 29 amino acids and three disulfide bonds (Fig. 10a), was discovered as the active constituent of a tea made from Oldenlandia affinis, used by midwives in Central Af-rica to stimulate uterine contractions[112]. Remarkably, kalata B1 is orally active, although its in vivo PK in animals has not been reported. It is therefore difficult to discern whether the oral activity of kalata B1 is due to good absorp-tion, or rather is achieved despite poor absorption, for exam-ple, because of very high affinity for its pharmacological target. A classic example of this phenomenon is the clinical drug desmopressin, an orally delivered derivative of the cy-clic peptide vasopressin that is used to treat diabetes in-sipidus, nocturia, and various coagulation disorders. Desmo-pressin has an absolute oral bioavailability of only 0.1%[113] and yet shows oral efficacy at doses as low as 100 !g in adults. Such potent oral activity in the context of poor oral bioavailability can occur when a ligand has a very slow dissociation rate from its receptor, allowing the receptor to become fully occupied even though steady-state levels of the drug in serum remain very low[114].

In studies using mammalian cell cultures, kalata B1 formed aggregates on the cell exterior. In the same study, a structurally similar cyclotide, MCoTI-II, was shown to pene-trate cells by macropinocytosis[115], an energy-dependent but receptor-independent mechanism of internalization known to operate with poly-Arg and other cell penetrating peptides such as Tat[116]. Nonetheless, solution NMR stud-ies investigating the interaction of kalata B1 and other cy-clotides with micelles have revealed that the cyclic peptides interact with membranes in a specific geometry that depends on the 3-dimensional arrangement of hydrophobic residues among the three loops[117, 118]. The cell permeability of cyclotides has been successfully applied to the design of orally active cyclotides with novel functions by grafting bio-active peptide sequences into cyclotide loops, suggesting that these proteins may be general scaffolds with which to design orally active molecules based on any bioactive peptide of interest[119-121]. Despite the intriguing oral activity of these compounds, however, their absolute oral bioavailabil-ity has not been reported.

Another disulfide-rich protein that has interesting phar-macokinetic properties is the soybean-derived 8 kD Bow-man-Birk inhibitor (BBI; Fig. 10b)[122]. BBI is not back-bone-cyclized, although like the cyclotides it has a knot-like topology[123, 124]. BBI suppressed radiation-induced trans-formation of pluripotent stem cells at sub-nM concentra-tions[125], and when fed to mice prevented chemical-induced adenocarcinoma[126]. Indeed, BBI was widely dis-tributed in tissues within 3 hours after oral administra-tion[127]. It retained full protease inhibitory activity after oral delivery and purification from serum and internal or-gans, indicating that in addition to its permeability it sur-vived first pass metabolism intact and had both low meta-bolic clearance and minimal proteolytic degradation. How-ever, whether BBI penetrates membranes by passive diffu-sion or some other mechanism (e.g. by receptor-mediated

endocytosis) remains unknown, and its absolute oral biovail-ability has not been reported.

While the mechanism(s) of membrane permeation of the cysteine knot microproteins has not been firmly established, at first glance it would seem unlikely that these molecules penetrate cells by passive diffusion. If, however, cysteine knot microproteins such as kalata B1 and BBI prove to ex-hibit passive membrane permeability in a cell-free membrane permeability system, it will force us to re-evaluate the mo-lecular weight barrier typically associated with passive membrane diffusion. These proteins are highly rigid, stabi-lized by C-to-N cyclization and multiple internal disulfide bonds, which allows them to adopt unusual folds in which the polar groups are buried (forming complex networks of hydrogen bonds) and multiple hydrophobic residues are ex-posed to solvent. Could the rigidity of these molecules cou-pled with their relatively large hydrophobic surfaces allow them to pass through membranes by a mechanism more akin to passive diffusion than the sort of direct, electrostatic inter-action ascribed to the cationic antimicrobial peptides? NMR evidence showing the specific interaction of kalata B1 and other cyclotides with micelles[117, 118] points toward the possibility of passive diffusion for these molecules, although more studies are needed to establish a detailed mechanism of penetration.

Fig. (10). a) cyclotide Kalata B1 and b) Bowman-Birk trypsin in-hibitor from soybean.

DISCUSSION The cell permeability associated with many relatively

large cyclic peptide natural products runs counter to “drug-likeness” predictors applied to typical small molecules. These natural products thus provide excellent model systems with which to study ADME properties of large macrocycles, and perhaps for large, “non-Lipinski” molecules in general.

b

a kalata B1

Bowman-Birk trypsin Inhibitor


The observation that molecules like CSA and others can be cell permeable begs the question: What is the relationship between molecular weight and membrane permeability? If the number of rotatable bonds is indeed a better predictor of permeability than MW, this supports the notion of an en-tropic barrier to permeability that increases with increasing size. For example, transit through the ordered phospholipid bilayer may involve the freezing out of bond rotations with an associated entropic liability, while structurally rigid mole-cules are preorganized enough to avoid such an entropic cost. While the retrospective study by Veber et al.[18], using a database of known drugs and their associated permeability supports this hypothesis, thermodynamic measurements us-ing controlled model systems would provide an orthogonal means of testing the impact of entropy on permeability and its connection to molecular weight and/or rigidity. Current predictive models of permeability that take size into ac-count[52] might benefit from measurements obtained on such “outlier” systems as these large cyclic peptides.

CSA has been by far the most intensively studied “non-polar” cyclic peptide and is the only such molecule for which we have extensive pharmacokinetic data in humans. Al-though it is tempting to infer general principles based on CSA, its PK properties may not be generalizable to other cyclic peptides. For example, although there is strong evi-dence that CSA penetrates epithelial cells by passive diffu-sion[15], it is also a substrate for the P-glycoprotein (Pgp) efflux pump and is actively transported from epithelial cells in the basolateral-to-apical direction[128]. Other non-polar cyclic peptides are also known P-glycoprotein substrates (and, like CSA, capable of reversing the multi-drug resis-tance due to Pgp overexpression)[129], although it is unclear to what extent this observation can be further generalized. Further, the oral bioavailability and volume of distribution of CSA are highly variable in the population [130], a function of patient-to-patient variability in activities and expression levels of Pgp as well as CYP3a, the major CYP involved in CSA metabolism[131, 132]. Whether this degree of variabil-ity holds for other non-polar cyclic peptides remains to be seen.

While we are beginning to learn about the connections between structure, molecular weight, and membrane perme-ability in cyclic peptides, we have only begun to scratch the surface in terms of the interplay of factors that determine PK in these large, functionally rich molecules. Indeed, few cy-clic peptide natural products have been subjected to direct cell permeability measurements, much less in vivo PK stud-ies. These studies would benefit not only from the traditional techniques used in this area (e.g. PAMPA and Caco-2 per-meability), but also from the biophysical techniques that have been applied to the study of cationic antimicrobial pep-tides using defined, cell-free membrane systems. These in-clude solid state NMR using 19F and 15N probes[133], at-tenuated total reflection-Fourier transform infrared spectros-copy (ATR-FTIR)[134] and surface plasmon resonance (SPR)[135]. While these studies have shed light on the de-tailed interactions between pore-forming peptides and bio-logical membranes, their application to the study of passive membrane diffusion by cyclic peptides has been limited. A more detailed biophysical model of passive membrane diffu-sion would likely benefit from an investigation into the

structure-permeability relationship in these larger, “non-Lipinski” molecules.

Chemists tend to assume that natural products were de-signed by evolution to carry out functions related primarily to defense or predation, although evidence is scant either in support of or against this hypothesis. Nonetheless, it would explain why natural products tend to target essential and highly conserved biological processes and structures such as protein synthesis, the cytoskeleton, and critical organelles, and similarly why natural products continue to be the logical source of leads in antimicrobial and anticancer drug discov-ery. This hypothesis has also led to the view that natural products may not be the best choice for leads against most other human disease targets, since humans did not co-evolve with the organisms that produce them. Indeed, a systems-level study of the protein targets of natural products revealed that they tend to be more highly connected pathway “hubs” than human disease-related targets in general[136].

Yet, while evolution may have selected natural products that inhibit a relatively narrow range of human disease tar-gets, the realm of PK may represent a nexus where the con-straints driving both natural product evolution and drug de-sign converge. In terms of natural selection, the ability to cross biological membranes and avoid metabolic degrada-tion, both oxidative and proteolytic, are likely to be just as important as binding affinity in the evolution of natural products. Large cyclic peptides, in particular, seem to have evolved tricks that allow them to push the PK envelope, maximizing size and complexity (toward greater potency) while maintaining good stability and membrane permeabil-ity. Thus, the design of combinatorial libraries of “natural product-like” molecules, which attempt to recapitulate the diversity and complexity of natural products without the tar-get limitations set by natural selection, can perhaps be in-formed by the structural features shared by natural products which favor permeability and stability. For cyclic peptides, these features include general trends in side chain functional-ity, and also specific patterns of backbone stereochemistry and N-methylation, which permit or even stabilize membrane permeable conformations. Indeed, it is likely that cyclic pep-tide natural products will continue to inform our understand-ing of the relationship between structure and pharmacokinet-ics, especially as we attempt to design new molecules in the outer reaches of “drug-like” chemical space, for some time to come.

CONFLICT OF INTEREST The author(s) confirm that this article content has no con-

flicts of interest.

ACKNOWLEDGEMENTS Declared none.

REFERENCES [1] Luo, B.; Cheung, H. W.; Subramanian, A.; Sharifnia, T.; Okamoto,

M.; Yang, X.; Hinkle, G.; Boehm, J. S.; Beroukhim, R.; Weir, B. A.; Mermel, C.; Barbie, D. A.; Awad, T.; Zhou, X.; Nguyen, T.; Piqani, B.; Li, C.; Golub, T. R.; Meyerson, M.; Hacohen, N.; Hahn, W. C.; Lander, E. S.; Sabatini, D. M.; Root, D. E., Highly parallel


identification of essential genes in cancer cells. Proc. Natl. Acad. Sci. U. S. A., 2008, 105 (51), 20380-5.

[2] Hur, G. H.; Vickery, C. R.; Burkart, M. D., Explorations of cata-lytic domains in non-ribosomal peptide synthetase enzymology. Nat. Prod. Rep., 2012.

[3] Lipinski, C. A.; Lombardo, F.; Dominy, B. W.; Feeney, P. J., Ex-perimental and computational approaches to estimate solubility and permeability in drug discovery and development settings. Adv. Drug Deliv. Rev., 2001, 46 (1-3), 3-26.

[4] Lipinski, C. A., Drug-like properties and the causes of poor solubil-ity and poor permeability. J. Pharmacol. Toxicol. Methods, 2000,44 (1), 235-49.

[5] Kwan, K. C., Oral bioavailability and first-pass effects. Drug Met. Disp., 1997, 25 (12), 1329-36.

[6] Lewis, D. F.; Dickins, M., Factors influencing rates and clearance in P450-mediated reactions: QSARs for substrates of the xenobi-otic-metabolizing hepatic microsomal P450s. Toxicology, 2002,170 (1-2), 45-53.

[7] Lewis, D. F., Structural characteristics of human P450s involved in drug metabolism: QSARs and lipophilicity profiles. Toxicology, 2000, 144 (1-3), 197-203.

[8] Rezai, T.; Yu, B.; Millhauser, G. L.; Jacobson, M. P.; Lokey, R. S., Testing the Conformational Hypothesis of Passive Membrane Per-meability Using Synthetic Cyclic Peptide Diastereomers. J. Am. Chem. Soc., 2006, 128, 2510-2511.

[9] Gentilucci, L.; De Marco, R.; Cerisoli, L., Chemical modifications designed to improve peptide stability: incorporation of non-natural amino acids, pseudo-peptide bonds, and cyclization. Curr. Pharm. Des., 2010, 16 (28), 3185-203.

[10] Delorbe, J. E.; Clements, J. H.; Whiddon, B. B.; Martin, S. F., Thermodynamic and Structural Effects of Macrocyclization as a Constraining Method in Protein-Ligand Interactions. ACS Med. Chem. Lett., 2010, 1 (8), 448-452.

[11] DeLorbe, J. E.; Clements, J. H.; Teresk, M. G.; Benfield, A. P.; Plake, H. R.; Millspaugh, L. E.; Martin, S. F., Thermodynamic and structural effects of conformational constraints in protein-ligand in-teractions. Entropic paradoxy associated with ligand preorganiza-tion. J. Am. Chem. Soc., 2009, 131 (46), 16758-70.

[12] Hancock, R. E.; Scott, M. G., The role of antimicrobial peptides in animal defenses. Proc. Natl. Acad. Sci. U. S. A., 2000, 97 (16), 8856-61.

[13] Yeaman, M. R.; Yount, N. Y., Mechanisms of antimicrobial pep-tide action and resistance. Pharmacol. Rev., 2003, 55 (1), 27-55.

[14] Zasloff, M., Antimicrobial peptides of multicellular organisms. Nature, 2002, 415 (6870), 389-95.

[15] Augustijns, P. F.; Bradshaw, T. P.; Gan, L. S. L.; Hendren, R. W.; Thakker, D. R., Evidence for a Polarized Efflux System in Caco-2 Cells Capable of Modulating Cyclosporine-a Transport. Biochem. Biophys. Res. Commun., 1993, 197 (2), 360-365.

[16] Rezai, T.; Bock, J. E.; Zhou, M. V.; Kalyanaraman, C.; Lokey, R. S.; Jacobson, M. P., Conformational flexibility, internal hydrogen bonding, and passive membrane permeability: successful in silico prediction of the relative permeabilities of cyclic peptides. J. Am. Chem. Soc., 2006, 128 (43), 14073-80.

[17] White, T. R.; Renzelman, C. M.; Rand, A. C.; Rezai, T.; McEwen, C. M.; Gelev, V. M.; Turner, R. A.; Linington, R. G.; Leung, S. S.; Kalgutkar, A. S.; Bauman, J. N.; Zhang, Y.; Liras, S.; Price, D. A.; Mathiowetz, A. M.; Jacobson, M. P.; Lokey, R. S., On-resin N-methylation of cyclic peptides for discovery of orally bioavailable scaffolds. Nat. Chem. Biol., 2011, 7 (11), 810-7.

[18] Veber, D. F.; Johnson, S. R.; Cheng, H. Y.; Smith, B. R.; Ward, K. W.; Kopple, K. D., Molecular properties that influence the oral bioavailability of drug candidates. J. Med. Chem., 2002, 45 (12), 2615-23.

[19] Wilmot, C. M.; Thornton, J. M., Analysis and Prediction of the Different Types of b-Turn in Proteins. J. Mol. Biol., 1988, 203,221-232.

[20] Wilmot, C. M.; Thornton, J. M., Beta-turns and their distortions: a proposed new nomenclature. Protein Eng., 1990, 3 (6), 479-93.

[21] Rose, G. D.; Gierasch, L. M.; Smith, J. A., Turns in Peptides and Proteins. In Adv. Protein Chem., Anfinsen, C. B., Ed. Academic Press: Orlando, FL, 1985; Vol. 37, pp 1-97.

[22] Chatterjee, J.; Rechenmacher, F.; Kessler, H., N-methylation of peptides and proteins: an important element for modulating bio-logical functions. Angew. Chem., 2013, 52 (1), 254-69.

[23] Ovadia, O.; Greenberg, S.; Chatterjee, J.; Laufer, B.; Opperer, F.; Kessler, H.; Gilon, C.; Hoffman, A., The effect of multiple N-methylation on intestinal permeability of cyclic hexapeptides. Mo-lecular pharmaceutics, 2011, 8 (2), 479-87.

[24] Biron, E.; Chatterjee, J.; Ovadia, O.; Langenegger, D.; Brueggen, J.; Hoyer, D.; Schmid, H. A.; Jelinek, R.; Gilon, C.; Hoffman, A.; Kessler, H., Improving oral bioavailability of peptides by multiple N-methylation: somatostatin analogues. Angew. Chem., 2008, 47(14), 2595-9.

[25] Beck, J. G.; Chatterjee, J.; Laufer, B.; Kiran, M. U.; Frank, A. O.; Neubauer, S.; Ovadia, O.; Greenberg, S.; Gilon, C.; Hoffman, A.; Kessler, H., Intestinal permeability of cyclic peptides: common key backbone motifs identified. J. Am. Chem. Soc., 2012, 134 (29), 12125-33.

[26] Bock, J. E.; Gavenonis, J.; Kritzer, J. A., Getting in Shape: Control-ling Peptide Bioactivity and Bioavailability Using Conformational Constraints. ACS Chem. Biol., 2012.

[27] Schreiber, S. L.; Crabtree, G. R., The mechanism of action of cy-closporin A and FK506. Immunol. Today, 1992, 13 (4), 136-42.

[28] Loosli, H. R.; Kessler, H.; Oschkinat, H.; Weber, H. P.; Petcher, T. J.; Widmer, A., Peptide conformations. Part 31. The conformation of cyclosporin A in the crystal and in solution. Helv. Chim. Acta, 1985, 68 (3), 682-704.

[29] Kessler, H.; Kock, M.; Wein, T.; Gehrke, M., Reinvestigation of the Conformation of Cyclosporin A in Chloroform. Helv. Chim. Acta, 1990, 73, 1818-1832.

[30] Jin, L.; Harrison, S. C., Crystal structure of human calcineurin complexed with cyclosporin A and human cyclophilin. Proc. Natl. Acad. Sci. U. S. A., 2002, 99 (21), 13522-6.

[31] Ko, S. Y.; Dalvit, C., Conformation of cyclosporin A in polar sol-vents. Int. J. Pept. Protein Res., 1992, 40 (5), 380-2.

[32] Smith, G. D.; Duax, W. L.; Langs, D. A.; DeTitta, G. T.; Edmonds, J. W.; Rohrer, D. C.; Weeks, C. M., The crystal and molecular structure of the triclinic and monoclinic forms of valinomycin, C54H90N6O18. J. Am. Chem. Soc., 1975, 97 (25), 7242-7.

[33] Hamilton, J. A.; Sabesan, M. N.; Steinrauf, L. K., Crystal structure of valinomycin potassium picrate: anion effects on valinomycin cation complexes. J. Am. Chem. Soc., 1981, 103 (19), 5880-5885.

[34] Rose, M. C.; Henkens, R. W., Stability of Sodium and Potassium Complexes of Valinomycin. Biochim. Biophys. Acta, 1974, 372 (2), 426-435.

[35] Crews, C. M.; Collins, J. L.; Lane, W. S.; Snapper, M. L.; Schrei-ber, S. L., GTP-dependent binding of the antiproliferative agent didemnin to elongation factor 1 alpha. J. Biol. Chem., 1994, 269(22), 15411-4.

[36] Ahuja, D.; Vera, M. D.; SirDeshpande, B. V.; Morimoto, H.; Wil-liams, P. G.; Joullie, M. M.; Toogood, P. L., Inhibition of protein synthesis by didemnin B: how EF-1alpha mediates inhibition of translocation. Biochemistry, 2000, 39 (15), 4339-46.

[37] SirDeshpande, B. V.; Toogood, P. L., Mechanism of protein syn-thesis inhibition by didemnin B in vitro. Biochemistry, 1995, 34(28), 9177-84.

[38] Le Tourneau, C.; Faivre, S.; Ciruelos, E.; Dominguez, M. J.; Lo-pez-Martin, J. A.; Izquierdo, M. A.; Jimeno, J.; Raymond, E., Re-ports of clinical benefit of plitidepsin (Aplidine), a new marine-derived anticancer agent, in patients with advanced medullary thy-roid carcinoma. Am. J. Clin. Oncol., 2010, 33 (2), 132-6.

[39] Mateos, M. V.; Cibeira, M. T.; Richardson, P. G.; Prosper, F.; Oriol, A.; de la Rubia, J.; Lahuerta, J. J.; Garcia-Sanz, R.; Extrem-era, S.; Szyldergemajn, S.; Corrado, C.; Singer, H.; Mitsiades, C. S.; Anderson, K. C.; Blade, J.; San Miguel, J., Phase II clinical and pharmacokinetic study of plitidepsin 3-hour infusion every two weeks alone or with dexamethasone in relapsed and refractory mul-tiple myeloma. Clin. Cancer Res., 2010, 16 (12), 3260-9.

[40] Brandon, E. F.; Sparidans, R. W.; van Ooijen, R. D.; Meijerman, I.; Lazaro, L. L.; Manzanares, I.; Beijnen, J. H.; Schellens, J. H., In vi-tro characterization of the human biotransformation pathways of aplidine, a novel marine anti-cancer drug. Invest. New Drugs, 2007,25 (1), 9-19.

[41] Faivre, S.; Chieze, S.; Delbaldo, C.; Ady-Vago, N.; Guzman, C.; Lopez-Lazaro, L.; Lozahic, S.; Jimeno, J.; Pico, F.; Armand, J. P.; Martin, J. A.; Raymond, E., Phase I and pharmacokinetic study of aplidine, a new marine cyclodepsipeptide in patients with advanced malignancies. J. Clin. Oncol., 2005, 23 (31), 7871-80.

[42] Depenbrock, H.; Peter, R.; Faircloth, G. T.; Manzanares, I.; Jimeno, J.; Hanauske, A. R., In vitro activity of aplidine, a new ma-


rine-derived anti-cancer compound, on freshly explanted clono-genic human tumour cells and haematopoietic precursor cells. Br. J. Cancer, 1998, 78 (6), 739-44.

[43] Berer, N.; Rudi, A.; Goldberg, I.; Benayahu, Y.; Kashman, Y., Callynormine A, a new marine cyclic peptide of a novel class. Org. Lett., 2004, 6 (15), 2543-5.

[44] Arnold, E.; Clardy, J., Crystal and molecular structure of BBM-928 A, a novel antitumor antibiotic from Actinomadura luzonensis. J. Am. Chem. Soc., 1981, 103 (5), 1243-1244.

[45] Searle, M. S.; Hall, J. G.; Wakelin, P. G., 1H- and 13C-n.m.r. stud-ies of the antitumour antibiotic luzopeptin. Resonance assignments, conformation and flexibility in solution. Biochem. J., 1988, 256 (1), 271-8.

[46] Searle, M. S.; Hall, J. G.; Denny, W. A.; Wakelin, L. P., NMR studies of the interaction of the antibiotic nogalamycin with the hexadeoxyribonucleotide duplex d(5'-GCATGC)2. Biochemistry, 1988, 27 (12), 4340-9.

[47] Zhang, X. L.; Patel, D. J., Solution structure of the luzopeptin-DNA complex. Biochemistry, 1991, 30 (16), 4026-41.

[48] Taniguchi, M.; Nagai, K.; Arao, N.; Kawasaki, T.; Saito, T.; Mori-tani, Y.; Takasaki, J.; Hayashi, K.; Fujita, S.; Suzuki, K.; Tsuka-moto, S., YM-254890, a novel platelet aggregation inhibitor pro-duced by Chromobacterium sp. QS3666. J. Antibiotics, 2003, 56(4), 358-63.

[49] Nishimura, A.; Kitano, K.; Takasaki, J.; Taniguchi, M.; Mizuno, N.; Tago, K.; Hakoshima, T.; Itoh, H., Structural basis for the spe-cific inhibition of heterotrimeric Gq protein by a small molecule. Proc. Natl. Acad. Sci. U. S. A., 2010, 107 (31), 13666-71.

[50] Miyamae, A.; Fujioka, M.; Koda, S.; Morimoto, Y., Structural studies of FR900359, a novel cyclic depsipeptide from Ardisia cre-nata sims(Myrsinaceae). J. Chem. Soc., Perkin Trans. 1, 1989, (5).

[51] Augustijns, P. F.; Brown, S. C.; Willard, D. H.; Consler, T. G.; Annaert, P. P.; Hendren, R. W.; Bradshaw, T. P., Hydration changes implicated in the remarkable temperature-dependent mem-brane permeation of cyclosporin A. Biochemistry, 2000, 39 (25), 7621-30.

[52] Leung, S. S.; Mijalkovic, J.; Borrelli, K.; Jacobson, M. P., Testing physical models of passive membrane permeation. J. Chem. Inf. Mod., 2012, 52 (6), 1621-36.

[53] Seelig, A., The role of size and charge for blood-brain barrier per-meation of drugs and fatty acids. J. Mol. Neurosci., 2007, 33 (1), 32-41.

[54] Hawkins, C. J.; Lavin, M. F.; Marshall, K. A.; van den Brenk, A. L.; Watters, D. J., Structure-activity relationships of the lissoclina-mides: cytotoxic cyclic peptides from the ascidian Lissoclinum pa-tella. J. Med. Chem., 1990, 33 (6), 1634-8.

[55] Wipf, P.; Fritch, P. C.; Geib, S. J.; Sefler, A. M., Conformational studies and structure-activity analysis of lissoclinamide 7 and re-lated cyclopeptide alkaloids. J. Am. Chem. Soc., 1998, 120 (17), 4105-4112.

[56] Pettit, G. R.; Xu, J.-p.; Dorsaz, A.-C.; Williams, M. D.; Boyd, M. R.; Cerny, R. L., Isolation and structure of the human cancer cell growth inhibitory cyclic decapeptides phakellistatins 7, 8 and 91,2. Bioorg. Med. Chem. Lett., 1995, 5 (13), 1339-1344.

[57] Pettit, G. R.; Rhodes, M. R.; Tan, R., Antineoplastic agents. 400. Synthesis of the Indian Ocean marine sponge cyclic heptapeptide phakellistatin 2. J. Nat. Prod., 1999, 62 (3), 409-14.

[58] Pettit, G. R.; Toki, B. E.; Xu, J. P.; Brune, D. C., Synthesis of the marine sponge cycloheptapeptide phakellistatin 5(1). J. Nat. Prod., 2000, 63 (1), 22-8.

[59] Cane, D. E.; Walsh, C. T., The parallel and convergent universes of polyketide synthases and nonribosomal peptide synthetases. Chem. Biol., 1999, 6 (12), R319-25.

[60] Du, L.; Shen, B., Biosynthesis of hybrid peptide-polyketide natural products. Curr. Op. Drug Disc. Dev., 2001, 4 (2), 215-28.

[61] Du, L.; Sanchez, C.; Shen, B., Hybrid peptide-polyketide natural products: biosynthesis and prospects toward engineering novel molecules. Metab. Eng., 2001, 3 (1), 78-95.

[62] Luesch, H.; Yoshida, W. Y.; Moore, R. E.; Paul, V. J.; Corbett, T. H., Total structure determination of apratoxin A, a potent novel cy-totoxin from the marine cyanobacterium Lyngbya majuscula. J. Am. Chem. Soc., 2001, 123 (23), 5418-23.

[63] Liu, Y.; Law, B. K.; Luesch, H., Apratoxin a reversibly inhibits the secretory pathway by preventing cotranslational translocation. Mol. Pharmacol., 2009, 76 (1), 91-104.

[64] Shen, S.; Zhang, P.; Lovchik, M. A.; Li, Y.; Tang, L.; Chen, Z.; Zeng, R.; Ma, D.; Yuan, J.; Yu, Q., Cyclodepsipeptide toxin pro-motes the degradation of Hsp90 client proteins through chaperone-mediated autophagy. J. Cell Biol., 2009, 185 (4), 629-39.

[65] Senderowicz, A. M.; Kaur, G.; Sainz, E.; Laing, C.; Inman, W. D.; Rodriguez, J.; Crews, P.; Malspeis, L.; Grever, M. R.; Sausville, E. A.; et al., Jasplakinolide's inhibition of the growth of prostate car-cinoma cells in vitro with disruption of the actin cytoskeleton. J. Natl. Cancer Inst., 1995, 87 (1), 46-51.

[66] Odaka, C.; Sanders, M. L.; Crews, P., Jasplakinolide induces apop-tosis in various transformed cell lines by a caspase-3-like protease-dependent pathway. Clin. Diagn. Lab. Immunol., 2000, 7 (6), 947-52.

[67] Bubb, M. R.; Spector, I.; Beyer, B. B.; Fosen, K. M., Effects of jasplakinolide on the kinetics of actin polymerization. An explana-tion for certain in vivo observations. J. Biol. Chem., 2000, 275 (7), 5163-70.

[68] Bubb, M. R.; Senderowicz, A. M.; Sausville, E. A.; Duncan, K. L.; Korn, E. D., Jasplakinolide, a cytotoxic natural product, induces ac-tin polymerization and competitively inhibits the binding of phal-loidin to F-actin. J. Biol. Chem., 1994, 269 (21), 14869-71.

[69] Inman, W.; Crews, P., Novel Marine Sponge Derived Amino-Acids .8. Conformational-Analysis of Jasplakinolide. J. Am. Chem. Soc., 1989, 111 (8), 2822-2829.

[70] Tabudravu, J. N.; Morris, L. A.; Milne, B. F.; Jaspars, M., Confor-mational studies of free and Li+ complexed jasplakinolide, a cyclic depsipeptide from the Fijian marine sponge Jaspis splendens. Org. Biomol. Chem., 2005, 3 (5), 745-749.

[71] Milroy, L. G.; Rizzo, S.; Calderon, A.; Ellinger, B.; Erdmann, S.; Mondry, J.; Verveer, P.; Bastiaens, P.; Waldmann, H.; Dehmelt, L.; Arndt, H. D., Selective chemical imaging of static actin in live cells. J. Am. Chem. Soc., 2012, 134 (20), 8480-6.

[72] Gao, J.; Hamann, M. T., Chemistry and biology of kahalalides. Chem. Rev., 2011, 111 (5), 3208-35.

[73] Suarez, Y.; Gonzalez, L.; Cuadrado, A.; Berciano, M.; Lafarga, M.; Munoz, A., Kahalalide F, a new marine-derived compound, in-duces oncosis in human prostate and breast cancer cells. Mol. Can-cer Ther., 2003, 2 (9), 863-72.

[74] Sewell, J. M.; Mayer, I.; Langdon, S. P.; Smyth, J. F.; Jodrell, D. I.; Guichard, S. M., The mechanism of action of Kahalalide F: vari-able cell permeability in human hepatoma cell lines. Eur. J. Can-cer, 2005, 41 (11), 1637-44.

[75] Janmaat, M. L.; Rodriguez, J. A.; Jimeno, J.; Kruyt, F. A.; Giac-cone, G., Kahalalide F induces necrosis-like cell death that involves depletion of ErbB3 and inhibition of Akt signaling. Mol. Pharma-col., 2005, 68 (2), 502-10.

[76] Garcia-Rocha, M.; Bonay, P.; Avila, J., The antitumoral compound Kahalalide F acts on cell lysosomes. Cancer Lett., 1996, 99 (1), 43-50.

[77] Piggott, A. M.; Karuso, P., Rapid identification of a protein binding partner for the marine natural product kahalalide F by using reverse chemical proteomics. Chembiochem, 2008, 9 (4), 524-30.

[78] Hosta, L.; Pla-Roca, M.; Arbiol, J.; Lopez-Iglesias, C.; Samitier, J.; Cruz, L. J.; Kogan, M. J.; Albericio, F., Conjugation of Kahalalide F with gold nanoparticles to enhance in vitro antitumoral activity. Bioconjugate Chem., 2009, 20 (1), 138-46.

[79] Sparidans, R. W.; Stokvis, E.; Jimeno, J. M.; Lopez-Lazaro, L.; Schellens, J. H.; Beijnen, J. H., Chemical and enzymatic stability of a cyclic depsipeptide, the novel, marine-derived, anti-cancer agent kahalalide F. Anticancer. Drugs, 2001, 12 (7), 575-82.

[80] Rademaker-Lakhai, J. M.; Horenblas, S.; Meinhardt, W.; Stokvis, E.; de Reijke, T. M.; Jimeno, J. M.; Lopez-Lazaro, L.; Lopez Mar-tin, J. A.; Beijnen, J. H.; Schellens, J. H., Phase I clinical and pharmacokinetic study of kahalalide F in patients with advanced androgen refractory prostate cancer. Clin. Cancer Res., 2005, 11(5), 1854-62.

[81] Beijnen, J. H.; Rademaker-Lakhai, J. M.; Horenblas, S.; Meinhardt, W.; Stokvis, E.; De Rijke, T. M.; Jimeno, J. M.; Lazaro, L. L.; Martin, J. A. L.; Schellens, J., Phase I and pharmacokinetic study of Kahalalide F in patients with advanced androgen refractory pros-tate cancer. J. Clin. Oncol., 2004, 22 (14), 146S-146S.

[82] Schellens, J. H.; Rademaker, J. L.; Horenblas, S.; Meinhardt, W.; Stokvis, E.; De Reijke, T. M.; Jimeno, J. M.; Lazaro, L. L.; Martin, J. A. L.; Beijnen, J. H., Phase I and pharmacokinetic study of kaha-lalide F in patients with advanced androgen refractory prostate can-cer. Clin. Cancer Res., 2001, 7 (11), 3694S-3694S.


[83] Pardo, B.; Paz-Ares, L.; Tabernero, J.; Ciruelos, E.; Garcia, M.; Salazar, R.; Lopez, A.; Blanco, M.; Nieto, A.; Jimeno, J.; Izquierdo, M. A.; Trigo, J. M., Phase I clinical and pharmacoki-netic study of kahalalide F administered weekly as a 1-hour infu-sion to patients with advanced solid tumors. Clin. Cancer Res., 2008, 14 (4), 1116-1123.

[84] de Bruijn, I.; de Kock, M. J.; Yang, M.; de Waard, P.; van Beek, T. A.; Raaijmakers, J. M., Genome-based discovery, structure predic-tion and functional analysis of cyclic lipopeptide antibiotics in Pseudomonas species. Mol. Microbiol., 2007, 63 (2), 417-28.

[85] Sinnaeve, D.; Hendrickx, P. M.; Van Hemel, J.; Peys, E.; Kieffer, B.; Martins, J. C., The solution structure and self-association prop-erties of the cyclic lipodepsipeptide pseudodesmin A support its pore-forming potential. Chemistry, 2009, 15 (46), 12653-62.

[86] Han, F.; Mortishire-Smith, R. J.; Rainey, P. B.; Williams, D. H., Structure of the white-line-inducing principle isolated from Pseu-domonas reactans. Acta Crystallogr. Sect. C: Cryst. Struct. Com-mun., 1992, 48 (11), 1965-1968.

[87] Quail, J. W.; Ismail, N.; Pedras, M. S. C.; Boyetchko, S. M., Pseu-dophomins A and B, a class of cyclic lipodepsipeptides isolated from a Pseudomonas species. Acta Crystallogr. Sect. C: Cryst. Struct. Commun., 2002, 58 (5), o268-o271.

[88] Sorensen, D.; Nielsen, T. H.; Christophersen, C.; Sorensen, J.; Gajhede, M., Cyclic lipoundecapeptide amphisin from Pseudo-monas sp. strain DSS73. Acta Crystallogr. Sect. C: Cryst. Struct. Commun., 2001, 57 (9), 1123-1124.

[89] Henriksen, A.; Anthoni, U.; Nielsen, T. H.; Sorensen, J.; Christo-phersen, C.; Gajhede, M., Cyclic lipoundecapeptide tensin from Pseudomonas fluorescens strain 96.578. Acta Crystallogr. Sect. C: Cryst. Struct. Commun., 2000, 56 (1), 113-115.

[90] Andjelic, C. D.; Planelles, V.; Barrows, L. R., Characterizing the anti-HIV activity of papuamide A. Marine drugs, 2008, 6 (4), 528-49.

[91] Parsons, A. B.; Lopez, A.; Givoni, I. E.; Williams, D. E.; Gray, C. A.; Porter, J.; Chua, G.; Sopko, R.; Brost, R. L.; Ho, C. H.; Wang, J.; Ketela, T.; Brenner, C.; Brill, J. A.; Fernandez, G. E.; Lorenz, T. C.; Payne, G. S.; Ishihara, S.; Ohya, Y.; Andrews, B.; Hughes, T. R.; Frey, B. J.; Graham, T. R.; Andersen, R. J.; Boone, C., Explor-ing the mode-of-action of bioactive compounds by chemical-genetic profiling in yeast. Cell, 2006, 126 (3), 611-25.

[92] Harms, J. M.; Wilson, D. N.; Schluenzen, F.; Connell, S. R.; Stachelhaus, T.; Zaborowska, Z.; Spahn, C. M.; Fucini, P., Transla-tional regulation via L11: molecular switches on the ribosome turned on and off by thiostrepton and micrococcin. Mol. Cell, 2008,30 (1), 26-38.

[93] Radhakrishnan, S. K.; Bhat, U. G.; Hughes, D. E.; Wang, I. C.; Costa, R. H.; Gartel, A. L., Identification of a chemical inhibitor of the oncogenic transcription factor forkhead box M1. Cancer Res., 2006, 66 (19), 9731-5.

[94] Nakano, I.; Joshi, K.; Visnyei, K.; Hu, B.; Watanabe, M.; Lam, D.; Wexler, E.; Saigusa, K.; Nakamura, Y.; Laks, D. R.; Mischel, P. S.; Viapiano, M.; Kornblum, H. I., Siomycin A targets brain tumor stem cells partially through a MELK-mediated pathway. Neuro-oncology, 2011, 13 (6), 622-34.

[95] Pandit, B.; Bhat, U. G.; Gartel, A. L., Proteasome inhibitory activ-ity of thiazole antibiotics. Cancer biology & therapy, 2011, 11 (1), 43-7.

[96] Bhat, U. G.; Halasi, M.; Gartel, A. L., FoxM1 is a general target for proteasome inhibitors. PloS one, 2009, 4 (8), e6593.

[97] Newick, K.; Cunniff, B.; Preston, K.; Held, P.; Arbiser, J.; Pass, H.; Mossman, B.; Shukla, A.; Heintz, N., Peroxiredoxin 3 is a redox-dependent target of thiostrepton in malignant mesothelioma cells. PloS one, 2012, 7 (6), e39404.

[98] Anderson, B.; Hodgkin, D. C.; Viswamitra, M. A., The structure of thiostrepton. Nature, 1970, 225 (5229), 233-5.

[99] Hensens, O. D.; Albers-Schonberg, G.; Anderson, B. F., The solu-tion conformation of the peptide antibiotic thiostrepton: a 1H NMR study. J. Antibiotics, 1983, 36 (7), 799-813.

[100] Yuzenkova, J.; Delgado, M.; Nechaev, S.; Savalia, D.; Epshtein, V.; Artsimovitch, I.; Mooney, R. A.; Landick, R.; Farias, R. N.; Salomon, R.; Severinov, K., Mutations of bacterial RNA polym-erase leading to resistance to microcin j25. J. Biol. Chem., 2002,277 (52), 50867-75.

[101] Mukhopadhyay, J.; Sineva, E.; Knight, J.; Levy, R. M.; Ebright, R. H., Antibacterial peptide microcin J25 inhibits transcription by

binding within and obstructing the RNA polymerase secondary channel. Mol. Cell, 2004, 14 (6), 739-51.

[102] Niklison-Chirou, M. V.; Dupuy, F.; Pena, L. B.; Gallego, S. M.; Barreiro-Arcos, M. L.; Avila, C.; Torres-Bugeau, C.; Arcuri, B. E.; Bellomio, A.; Minahk, C.; Morero, R. D., Microcin J25 triggers cy-tochrome c release through irreversible damage of mitochondrial proteins and lipids. Int. J. Biochem. Cell Biol., 2010, 42 (2), 273-81.

[103] Niklison Chirou, M.; Bellomio, A.; Dupuy, F.; Arcuri, B.; Minahk, C.; Morero, R., Microcin J25 induces the opening of the mitochon-drial transition pore and cytochrome c release through superoxide generation. The FEBS journal, 2008, 275 (16), 4088-96.

[104] Wilson, K. A.; Kalkum, M.; Ottesen, J.; Yuzenkova, J.; Chait, B. T.; Landick, R.; Muir, T.; Severinov, K.; Darst, S. A., Structure of microcin J25, a peptide inhibitor of bacterial RNA polymerase, is a lassoed tail. J. Am. Chem. Soc., 2003, 125 (41), 12475-83.

[105] Bayro, M. J.; Mukhopadhyay, J.; Swapna, G. V.; Huang, J. Y.; Ma, L. C.; Sineva, E.; Dawson, P. E.; Montelione, G. T.; Ebright, R. H., Structure of antibacterial peptide microcin J25: a 21-residue lariat protoknot. J. Am. Chem. Soc., 2003, 125 (41), 12382-3.

[106] Rintoul, M. R.; de Arcuri, B. F.; Morero, R. D., Effects of the anti-biotic peptide microcin J25 on liposomes: role of acyl chain length and negatively charged phospholipid. Biochim. Biophys. Acta, 2000, 1509 (1-2), 65-72.

[107] Dupuy, F.; Morero, R., Microcin J25 membrane interaction: selec-tivity toward gel phase. Biochim. Biophys. Acta, 2011, 1808 (6), 1764-71.

[108] Bellomio, A.; Oliveira, R. G.; Maggio, B.; Morero, R. D., Penetra-tion and interactions of the antimicrobial peptide, microcin J25, into uncharged phospholipid monolayers. J. Colloid Interface Sci., 2005, 285 (1), 118-24.

[109] Daly, N. L.; Rosengren, K. J.; Craik, D. J., Discovery, structure and biological activities of cyclotides. Adv. Drug Deliv. Rev., 2009, 61(11), 918-30.

[110] Skjeldal, L.; Gran, L.; Sletten, K.; Volkman, B. F., Refined struc-ture and metal binding site of the kalata B1 peptide. Arch. Biochem. Biophys., 2002, 399 (2), 142-8.

[111] Saether, O.; Craik, D. J.; Campbell, I. D.; Sletten, K.; Juul, J.; Norman, D. G., Elucidation of the primary and three-dimensional structure of the uterotonic polypeptide kalata B1. Biochemistry, 1995, 34 (13), 4147-58.

[112] Gran, L.; Sandberg, F.; Sletten, K., Oldenlandia affinis (R&S) DC. A plant containing uteroactive peptides used in African traditional medicine. J. Ethnopharmacol., 2000, 70 (3), 197-203.

[113] Fjellestad-Paulsen, A.; Hoglund, P.; Lundin, S.; Paulsen, O., Phar-macokinetics of 1-deamino-8-D-arginine vasopressin after various routes of administration in healthy volunteers. Clin. Endocrinol. (Oxf). 1993, 38 (2), 177-82.

[114] Kenakin, T., Quantifying Biological Activity in Chemical Terms: A Pharmacology Primer to Describe Drug Effect. ACS Chem. Biol., 2009.

[115] Greenwood, K. P.; Daly, N. L.; Brown, D. L.; Stow, J. L.; Craik, D. J., The cyclic cystine knot miniprotein MCoTI-II is internalized into cells by macropinocytosis. Int. J. Biochem. Cell Biol., 2007, 39(12), 2252-64.

[116] Wadia, J. S.; Stan, R. V.; Dowdy, S. F., Transducible TAT-HA fusogenic peptide enhances escape of TAT-fusion proteins after lipid raft macropinocytosis. Nat. Med., 2004, 10 (3), 310-5.

[117] Shenkarev, Z. O.; Nadezhdin, K. D.; Sobol, V. A.; Sobol, A. G.; Skjeldal, L.; Arseniev, A. S., Conformation and mode of membrane interaction in cyclotides. Spatial structure of kalata B1 bound to a dodecylphosphocholine micelle. The FEBS journal, 2006, 273 (12), 2658-72.

[118] Wang, C. K.; Colgrave, M. L.; Ireland, D. C.; Kaas, Q.; Craik, D. J., Despite a conserved cystine knot motif, different cyclotides have different membrane binding modes. Biophys. J., 2009, 97 (5), 1471-81.

[119] Chan, L. Y.; Gunasekera, S.; Henriques, S. T.; Worth, N. F.; Le, S. J.; Clark, R. J.; Campbell, J. H.; Craik, D. J.; Daly, N. L., Engineer-ing pro-angiogenic peptides using stable, disulfide-rich cyclic scaf-folds. Blood, 2011, 118 (25), 6709-17.

[120] Gunasekera, S.; Foley, F. M.; Clark, R. J.; Sando, L.; Fabri, L. J.; Craik, D. J.; Daly, N. L., Engineering stabilized vascular endothe-lial growth factor-A antagonists: synthesis, structural characteriza-tion, and bioactivity of grafted analogues of cyclotides. J. Med. Chem., 2008, 51 (24), 7697-704.


[121] Wong, C. T.; Rowlands, D. K.; Wong, C. H.; Lo, T. W.; Nguyen, G. K.; Li, H. Y.; Tam, J. P., Orally active peptidic bradykinin B1 receptor antagonists engineered from a cyclotide scaffold for in-flammatory pain treatment. Angew. Chem., 2012, 51 (23), 5620-4.

[122] Kennedy, A. R., The Bowman-Birk inhibitor from soybeans as an anticarcinogenic agent. Am. J. Clin. Nutr., 1998, 68 (6 Suppl), 1406S-1412S.

[123] Voss, R. H.; Ermler, U.; Essen, L. O.; Wenzl, G.; Kim, Y. M.; Flecker, P., Crystal structure of the bifunctional soybean Bowman-Birk inhibitor at 0.28-nm resolution. Structural peculiarities in a folded protein conformation. Eur. J. Biochem., 1996, 242 (1), 122-31.

[124] Park, E. Y.; Kim, J. A.; Kim, H. W.; Kim, Y. S.; Song, H. K., Crys-tal structure of the Bowman-Birk inhibitor from barley seeds in ter-nary complex with porcine trypsin. J. Mol. Biol., 2004, 343 (1), 173-86.

[125] Yavelow, J.; Collins, M.; Birk, Y.; Troll, W.; Kennedy, A. R., Nanomolar concentrations of Bowman-Birk soybean protease in-hibitor suppress x-ray-induced transformation in vitro. Proc. Natl. Acad. Sci. U. S. A., 1985, 82 (16), 5395-9.

[126] Weed, H. G.; McGandy, R. B.; Kennedy, A. R., Protection against dimethylhydrazine-induced adenomatous tumors of the mouse co-lon by the dietary addition of an extract of soybeans containing the Bowman-Birk protease inhibitor. Carcinogenesis, 1985, 6 (8), 1239-41.

[127] Billings, P. C.; St Clair, W. H.; Maki, P. A.; Kennedy, A. R., Dis-tribution of the Bowman Birk protease inhibitor in mice following oral administration. Cancer Lett., 1992, 62 (3), 191-7.

[128] Saeki, T.; Ueda, K.; Tanigawara, Y.; Hori, R.; Komano, T., Human P-glycoprotein transports cyclosporin A and FK506. J. Biol. Chem., 1993, 268 (9), 6077-80.

[129] Sharom, F. J.; Lu, P.; Liu, R.; Yu, X., Linear and cyclic peptides as substrates and modulators of P-glycoprotein: peptide binding and effects on drug transport and accumulation. Biochem. J., 1998, 333 ( Pt 3), 621-30.

[130] Ptachcinski, R. J.; Venkataramanan, R.; Burckart, G. J., Clinical pharmacokinetics of cyclosporin. Clin. Pharmacokinet., 1986, 11(2), 107-32.

[131] Lown, K. S.; Mayo, R. R.; Leichtman, A. B.; Hsiao, H. L.; Tur-geon, D. K.; Schmiedlin-Ren, P.; Brown, M. B.; Guo, W.; Rossi, S. J.; Benet, L. Z.; Watkins, P. B., Role of intestinal P-glycoprotein (mdr1) in interpatient variation in the oral bioavailability of cy-closporine. Clin. Pharmacol. Ther., 1997, 62 (3), 248-60.

[132] Wacher, V. J.; Silverman, J. A.; Zhang, Y.; Benet, L. Z., Role of P-glycoprotein and cytochrome P450 3A in limiting oral absorption of peptides and peptidomimetics. J. Pharm. Sci., 1998, 87 (11), 1322-30.

[133] Koch, K.; Afonin, S.; Ieronimo, M.; Berditsch, M.; Ulrich, A. S., Solid-state (19)F-NMR of peptides in native membranes. Top. Curr. Chem., 2012, 306, 89-118.

[134] Oren, Z.; Shai, Y., Cyclization of a cytolytic amphipathic alpha-helical peptide and its diastereomer: effect on structure, interaction with model membranes, and biological function. Biochemistry, 2000, 39 (20), 6103-14.

[135] Tomasinsig, L.; Skerlavaj, B.; Papo, N.; Giabbai, B.; Shai, Y.; Zanetti, M., Mechanistic and functional studies of the interaction of a proline-rich antimicrobial peptide with mammalian cells. J. Biol. Chem., 2006, 281 (1), 383-91.

[136] Dancik, V.; Seiler, K. P.; Young, D. W.; Schreiber, S. L.; Clemons, P. A., Distinct biological network properties between the targets of natural products and disease genes. J. Am. Chem. Soc., 2010, 132(27), 9259-61.

Received: January 08, 2013 Revised: February 13, 2013 Accepted: February 13, 2013

Current Topics in Medicinal Chemistry, 2013, 13, 821 836...

Documents

Transcript of Current Topics in Medicinal Chemistry, 2013, 13, 821 836...