Somatostatin subtype-2 receptor-targeted metal-based anticancer complexes

Article

pubs.acs.org/bc

Somatostatin Subtype‑2 Receptor-Targeted Metal-Based Anticancer

Complexes

Flavia Barragan,^†,‡,#Dolors Carrion-Salip,^§,#Irene Gomez-Pinto,^⊥Alejandro Gonzalez-Canto,^†,‡

́

Peter J. Sadler,^∥Rafael de Llorens,^§Virtudes Moreno,^‡Carlos Gonzalez,^⊥Anna Massaguer,*^,§

́

and Vicente Marchan*^,†

^†Departament de Química Organ

́

̀

ica and IBUB and ^‡Departament de Química Inorgan

̀

ica, Universitat de Barcelona, Martí i Franques

̀

1-11, E-08028 Barcelona, Spain

^§Departament de Biologia, Universitat de Girona, Campus Montilivi, E-17071 Girona, Spain

^⊥Instituto de Química-Física “Rocasolano”, CSIC, Serrano 119, E-28006, Madrid, Spain

^∥Department of Chemistry, University of Warwick, Coventry, CV4 7AL, United Kingdom

S

* Supporting Information

ABSTRACT: Conjugates of a dicarba analogue of octreotide,

a potent somatostatin agonist whose receptors are overex-

pressed on tumor cells, with [PtCl₂(dap)] (dap = 1-(carboxylic

acid)-1,2-diaminoethane) (3), [(η⁶-bip)Os(4-CO₂-pico)Cl]

(bip = biphenyl, pico = picolinate) (4), [(η⁶-p-cym)RuCl-

(dap)]⁺(p-cym = p-cymene) (5), and [(η⁶-p-cym)RuCl-

(imidazole-CO₂H)(PPh₃)]⁺(6), were synthesized by using a

solid-phase approach. Conjugates 3−5 readily underwent

hydrolysis and DNA binding, whereas conjugate 6 was inert

to ligand substitution. NMR spectroscopy and molecular

dynamics calculations showed that conjugate formation does not perturb the overall peptide structure. Only 6 exhibited

antiproliferative activity in human tumor cells (IC₅₀= 63 2 μM in MCF-7 cells and IC₅₀= 26 3 μM in DU-145 cells) with

active participation of somatostatin receptors in cellular uptake. Similar cytotoxic activity was found in a normal cell line (IC₅₀

=

45 2.6 μM in CHO cells), which can be attributed to a similar level of expression of somatostatin subtype-2 receptor. These

studies provide new insights into the eﬀect of receptor-binding peptide conjugation on the activity of metal-based anticancer

drugs, and demonstrate the potential of such hybrid compounds to target tumor cells speciﬁcally.

with better receptor aﬃnity and higher stability under

physiological conditions.⁶

INTRODUCTION

■

The overexpression of the receptors for many regulatory

peptides in human tumor cells in comparison to their

expression in normal cells has prompted research on the use

of such peptides in tumor targeting for both diagnostic and

therapeutic purposes.¹⁻³From the therapeutic point of view, a

promising approach for the treatment of cancer consists of the

attachment of a cytotoxic drug to a peptide moiety with the aim

of improving its activity and bioavailability. This targeted

anticancer strategy will result in therapeutic agents with

increased tumor selectivity and decreased toxicity in normal

tissues.

Among natural receptor-binding peptides, the neuroendo-

crine hormone somatostatin and its analogues have received

much attention because of their high aﬃnity against its ﬁve

human receptors (sst₁−sst₅).^1,3,4These receptors are overex-

pressed in various major tumor types, including lung, breast,

prostate, adrenal, and neuroendocrine tumors in comparison

with normal tissues.⁵The successful use of radiolabeled

somatostatin for imaging and radionuclide therapy has

prompted the development of new cyclic and acyclic analogues

Octreotide,⁷1 (Chart 1), a metabolically stabilized cyclo-

ctapeptide analogue of somatostatin that includes ^D-amino acids

to enhance resistance to enzymatic degradation and a cystine

bridge to stabilize the pharmacophore β-turn, is particularly

interesting because of its high aﬃnity and selectivity for the

receptor sst₂,⁶which is the receptor subtype most frequently

found on tumor cells. This selectivity has allowed the

development of derivatives such as [¹¹¹In-DTPA]-octreotide

and [⁹⁰Y-DOTA-Tyr³]-octreotide which are used in the clinics

for molecular imaging and therapy of neuroendocrine tumors,

respectively.⁸Octreotide has also been conjugated to cytotoxic

organic compounds such as doxorubicin, camptothecin, and

paclitaxel, with promising results in some cases because of the

reduced toxicity and the increased selectivity of the conjugates

compared with that of the free drug.^4,9

Received: March 30, 2012

Revised: August 6, 2012

Published: August 8, 2012

1838

dx.doi.org/10.1021/bc300173h | Bioconjugate Chem. 2012, 23, 1838−1855

Bioconjugate Chemistry

Article

a

ligand exchange in a similar way to cisplatin, which makes them

particularly attractive for developing new anticancer drugs.

Unlike cisplatin, which has two reactive Pt−Cl coordination

bonds, some active organometallic piano-stool complexes have

only one reactive site (Ru/Os−Cl). The aquation of the

chlorido complexes, mostly suppressed in extracellular ﬂuids

(chloride concentration ∼100 mM), is triggered inside the cell

nucleus by the low chloride concentration (4 mM), allowing

the interaction of the active aqua species with biomolecules

such as DNA.²⁰⁻²²

Chart 1. Structure of the Conjugates Synthesized

On the basis of these precedents, a promising approach in

the search of new metal-based anticancer drugs would consist

of the conjugation of platinum and classical nonplatinum metal

complexes (e.g., ruthenium and osmium) to carrier molecules

such as receptor-binding peptides. Indeed, given the over-

expression of selective receptors for these peptides on the

membrane of tumoral cells, it is expected that the peptide

sequences will have a positive eﬀect on the therapeutic

properties of the metal-based drug (e.g., reduction of the

toxic side-eﬀects) since they will act as a “tumour-targeting

devices”. With this idea in mind, we have synthesized

conjugates between a dichloridoplatinum(II) complex and

octreotide derivatives in which the disulﬁde bond had been

replaced by CH₂CH₂or CHCH linkages.²³This chemical

modiﬁcation on the peptide does not substantially alter the

binding aﬃnity of the dicarba-octreotide analogues for the sst₂

receptor but increases the stability of the carrier biomolecule in

the reductive cellular environment.²⁴In addition, we have

recently described the synthesis and DNA ruthenation under

visible light irradiation of a photoactivated ruthenium(II) arene

complex conjugated to a dicarba analogue of octreotide.²⁵In

recent years, several coordination or organometallic complexes

have also been conjugated to carrier molecules such as

receptor-binding peptides or known delivery peptides to

improve both cellular uptake to cancer cells and cytotoxic

activity of the metal-based drug.²⁶⁻³³

Herein, we report on the use of conjugates between

octreotide and several metal complexes as new chemo-

therapeutic agents. In particular, the synthesis of hybrid

compounds between the dicarba analogue of octreotide (2)

and platinum(II), ruthenium(II), and osmium(II) metal

complexes has been accomplished by eﬃcient solid-phase

procedures (compounds 3−6 in Chart 1). The inﬂuence of the

attachment of the metal complex on the structure of octreotide

has been studied by NMR spectroscopy and molecular

dynamics, as well as the capacity of these conjugates to bind

DNA. Finally, cellular uptake and cytotoxicity studies have been

carried out in a normal cell line and two human tumoral cell

lines.

a

Numbering of the amino acid residues follows that of native

somatostatin.

Despite the successful history of cisplatin and its second-

generation derivatives (carboplatin and oxaliplatin) in the

treatment of some types of cancer,¹⁰the severe toxic side-

eﬀects and the acquired or intrinsic resistance of certain tumors

has focused research on other metal-based complexes in an

eﬀort to develop improved chemotherapeutic agents.^11,12

Indeed, in addition to several platinum compounds, two

ruthenium(III) coordination complexes, trans-[RuCl₄(DMSO)-

(Im)]ImH (NAMI-A) and trans-[RuCl₄(Ind)₂]IndH (NKP-

1339), are currently on clinical trials.^13,14

In recent years, research on bioinorganic chemistry has also

been focused on organometallic ruthenium(II) and osmium(II)

arene complexes, since these compounds have shown

promising in vitro and in vivo anticancer activities, including

cell lines that have become resistant to cisplatin.¹⁵⁻¹⁹

Ruthenium and its heavier congener osmium oﬀer several

advantages in comparison with platinum compounds, including

the ability to bind to a wide variety of types of ligands, the

availability of additional coordination sites in octahedral

complexes, the possibility of controlling the shape of the

complex, and the changes in the oxidation state. It is also

particularly relevant that they have the capacity to undergo

EXPERIMENTAL PROCEDURES

■

Materials and Methods. Unless otherwise stated, common

chemicals and solvents (HPLC-grade or reagent-grade quality)

were purchased from commercial sources and used without

further puriﬁcation. Peptide-grade DMF was from Scharlau.

Fmoc-protected amino acids, resins, and coupling reagents for

solid-phase synthesis were obtained from Novabiochem,

Bachem, or Iris Biotech. Fmoc-Hag-OH was purchased from

Bachem. Fmoc-^L-threoninol p-carboxyacetal was synthesized

following previously reported procedures.³⁴Solid-phase syn-

theses were performed manually in a polypropylene syringe

ﬁtted with a polyethylene disk. Second-generation Grubbs

catalyst and Wilkinson’s catalyst were purchased from Aldrich.

1839

dx.doi.org/10.1021/bc300173h | Bioconjugate Chem. 2012, 23, 1838−1855

Bioconjugate Chemistry

Article

Milli-Q water was directly obtained from a Milli-Q system

equipped with a 5000 Da ultraﬁltration cartridge. Metal

complexes ( )-[PtCl₂(dap)] (7),²³( )-[PtCl₂(Etdap)]

(15),³⁵[(η⁶-bip)Os(4-CO₂-pico)Cl] (8),³⁶[{(η⁶-p-cym)Ru-

(μ-Cl)Cl}₂] (9), [(η⁶-p-cym)RuCl₂(PPh₃)] (10),³⁷and [(η⁶-p-

cym)RuCl(PPh₃)₂][PF₆] (11)³⁸were synthesized and charac-

terized as previously described. (^L)-2,3-Diaminopropionic acid

methyl ester dihydrochloride³⁹and N^α,N^β-bis-(9-ﬂuorenylme-

thyloxycarbonyl)-L-2,3-diaminopropionic acid⁴⁰were synthe-

sized as previously described. 5(6)-Carboxyﬂuorescein was

purchased from Aldrich and octreotide acetate from Bachem.

All the assayed compounds displayed a purity of ≥95%,

determined by HPLC analysis.

The human breast cancer cell line MCF-7 and the human

prostate carcinoma cell line DU-145 were obtained from the

American Tissue Culture Collection (ATTC, Rockville, MD,

USA). Chinese hamster ovary (CHO) cells were used as a

nontumor cell line. The cells were maintained in Dulbecco’s

modiﬁed Eagle’s medium supplemented with 10% fetal bovine

serum and 1% penicillin−streptomycin (GIBCO BRL, Grand

Island, NY) at 37 °C in a humidiﬁed atmosphere containing 5%

CO₂. The cells were passaged two times per week.

NMR spectra were recorded at 25 °C on Varian Gemini 300

MHz and Varian Mercury 400 MHz spectrometers using

deuterated solvents. Tetramethylsilane (TMS) was used as an

internal reference (δ 0 ppm) for ¹H spectra recorded in CDCl₃

and the residual signal of the solvent (δ 77.16 ppm) for ¹³C

spectra. For CD₃OD, acetone-d₆, DMSO-d₆, or D₂O, the

residual signal of the solvent was used as a reference.

High-resolution MALDI-TOF mass spectra were recorded

on a 4800 Plus MALDI-TOF/TOF spectrometer (Applied

Biosystems) in the positive-ion mode using 2,4-dihidroxyben-

zoic acid as a matrix. ESI mass spectra (ESI-MS) were recorded

on a Micromass ZQ instrument with single quadrupole

detector coupled to an HPLC. High-resolution electrospray

mass spectra (HR ESI MS) were obtained on an Agilent 1100

LC/MS-TOF instrument.

as indicated for each metal. The reactor was protected from

light by covering it with aluminum foil.

Side-chain deprotection and cleavage from the resin was

performed simultaneously with TFA/TIS/H₂O 95:2.5:2.5

(peptides, ﬂuorescein-labeled dicarba octreotide 13 and

conjugate 4), TFA/phenol/H₂O 95:2.5:2.5 (conjugates 3, 5,

6, and 12), or TFA/H₂O/TIS/EDT 94:2.5:2.5:1 (ﬂuorescein-

labeled octreotide 14) for 1 h at room temperature. Most of the

TFA was removed by bubbling N₂into the solution, and the

resulting residue was poured onto cold ether to precipitate the

target compound.

Analytical reversed-phase HPLC analyses were carried out on

a GraceSmart RP C₁₈column (250 × 4 mm, 5 μm, ﬂow rate: 1

mL/min), using linear gradients of 0.045% TFA in H₂O

(solvent A) and 0.036% TFA in ACN (solvent B). In some

cases, small-scale puriﬁcation was carried out using the same

column. Large-scale puriﬁcation was carried out on a Jupiter

Proteo semipreparative column (250 × 10 mm, 10 μm, ﬂow

rate: 3−4 mL/min), using linear gradients of 0.1% TFA in H₂O

(solvent A) and 0.1% TFA in ACN (solvent B). After several

runs, pure fractions were combined and lyophilized.

Platinum−Octreotide Conjugate 3. [PtCl₂(dap)] (5 mol

equiv) was coupled with DIPC (5 mol equiv) and NHS (10

mol equiv) in anhydrous DMF for 15 h at rt. Overall yield

(synthesis + puriﬁcation): 14%. Characterization: R_t= 14.2 min

(analytical gradient: 20% to 60% in 30 min); HR ESI MS,

positive mode: m/z 1479.5574 (calcd mass for

C₆₀H₈₈Cl₂N₁₃O₁₄Pt [M+H]⁺: 1479.5598); HR MALDI-TOF

MS, positive mode: m/z 1479.85 (calcd mass for

C₆₀H₈₈Cl₂N₁₃O₁₄Pt [M+H]⁺: 1479.5598), m/z 1501.87

(calcd mass for C₆₀H₈₇Cl₂N₁₃NaO₁₄Pt [M+Na]⁺: 1501.5418).

Osmium−Octreotide Conjugate 4. [(η⁶-bip)Os(4-CO₂-

pico)Cl] (5 mol equiv) was coupled with DIPC (5 mol

equiv) and HOAt (5 mol equiv) in anhydrous DMF for 3 h at

rt. Overall yield (synthesis + puriﬁcation): 16%. Character-

ization: R_t= 20.2 min (analytical gradient: 20% to 60% in 30

min); HR ESI MS, positive mode: m/z 1657.6621 (calcd mass

for C₇₆H₉₄ClN₁₂O₁₆Os [M+H]⁺: 1657.6214); HR MALDI-

TOF MS, positive mode: m/z 1679.60 (calcd mass for

C₇₆H₉₂ClN₁₂NaO₁₆Os [M+Na]⁺: 1679.6034).

Synthesis of Metal−Octreotide Conjugates (3−6 and

12) and Fluorescein-Labeled Peptides (13−14). Solid-

phase peptide synthesis was performed on a Rink amide resin-

p-MBHA (f = 0.34 mmol/g, 100−200 mesh) using standard

Fmoc/tBu chemistry with the following side-chain protecting

groups: Boc (Nⁱ-tert-butoxicarbonyl, tryptophan, and N^ε-tert-

Ruthenium(dap)−Octreotide Conjugate 5. N^α,N^β-Bis-(9-

ﬂuorenylmethyloxycarbonyl)-L-2,3-diaminopropionic acid (4

mol equiv) was coupled with HATU (3.9 mol equiv) in

anhydrous DMF in the presence of DIPEA (8 mol equiv) for 1

h at rt. After removal of the Fmoc protecting groups,

octreotide-bound resin was reacted with 9 (3 mol equiv) in

the presence of LiCl (6.6 mol equiv) and NEt₃(7.2 mol equiv)

in a DMF/EtOH 9:1 mixture at 100 °C for 1 h under

microwave irradiation (40 W). Overall yield (synthesis +

puriﬁcation): 10%. Characterization: R_t= 14.1 min (analytical

gradient: 20% to 60% in 30 min); HR MALDI-TOF MS,

positive mode: m/z 1448.40 (calcd mass for C₇₀H₁₀₀N₁₃O₁₄Ru

[M−Cl−H]⁺: 1448.6556).

Ruthenium−Octreotide Conjugate 6. 4-(1H-Imidazol-1-

yl)benzoic acid (10 mol equiv) was ﬁrst coupled with DIPC

(10 mol equiv) and HOBt (10 mol equiv) in anhydrous DMF

for 1 h at rt. Then, imidazole-derivatized peptide-bound resin

was reacted with 11 (3 mol equiv) in the presence of LiCl (3.3

mol equiv) and NEt₃(20 mol equiv) in a DMF/EtOH 9:1

mixture for 8 h at 50 °C. Overall yield (synthesis +

puriﬁcation): 12%. Characterization: R_t= 14.6 min (analytical

gradient: 30% to 100% in 30 min); HR ESI MS, positive mode:

m/z 1830.7240 (calcd mass for C₉₅H₁₁₆ClN₁₃O₁₄PRu [M]⁺:

t

butoxicarbonyl, lysine), Bu (O-tert-butyl, threonine), and Trt

(S-trityl, cysteine). The assembly of the dicarba analogue of

octreotide was performed as previously described.^23,24Brieﬂy,

Fmoc-^L-threoninol p-carboxyacetal (1.8 mol equiv) was ﬁrst

coupled using DIPC (1.8 mol equiv) and HOBt (1.8 mol

equiv) in anhydrous DMF for 1 h. The following Fmoc-

protected amino acids (2.4 mol equiv) were incorporated with

HATU (2.3 mol equiv) and DIPEA (4.8 mol equiv) in

anhydrous DMF for 1 h. Microwave-assisted ring-closing

methatesis using second-generation Grubbs catalyst and

hydrogenation reaction with Wilkinson’s catalyst were carried

out as previously described.²³Finally, 8-(9-ﬂuorenylmethylox-

ycarbonyl-amino)-3,6-dioxaoctanoic acid (3 mol equiv) or γ-

aminoisobutyric acid (3 mol equiv) were coupled onto the

cyclic dicarba analogue of octreotide 2 bound to resin with

DIPC (3 mol equiv) and HOAt (3 mol equiv) in anhydrous

DMF for 1 h. After removal of the Fmoc group (20% piperidine

in DMF) from the Fmoc-protected linker-derivatized dicarba

analogue of octreotide bound to resin, coupling or assembly of

the metal complexes on the free N-terminal end was carried out

1840

dx.doi.org/10.1021/bc300173h | Bioconjugate Chem. 2012, 23, 1838−1855

Bioconjugate Chemistry

Article

1830.7234), m/z 915.8672 (calcd mass for C₇₀H₁₀₂ClN₁₃O₁₄Ru

[M+H]²⁺: 915.8656); HR MALDI-TOF MS, positive mode:

m/z 1795.44 (calcd mass for C₉₅H₁₁₅N₁₃O₁₄PRu [M−Cl−H]⁺:

1794.7468).

and stirred at room temperature. Slowly, the suspension

dissolved, and after 4 h, diethyl ether was added with stirring to

the resulting orange solution until it became turbid. The

separated red microcrystalline product was isolated by ﬁltration,

washed several times with ether, and dried under vacuo (165

mg, 50%). ¹H NMR (400 MHz, CDCl₃) δ (ppm): 1.31 (6H, d,

J = 6.8 Hz), 2.24 (3H, s), 3.03 (1H, sp, J = 6.8 Hz), 3.95 (3H,

s), 5.32 (2H. d, J = 5.4 Hz), 5.51 (2H. d, J = 5.4 Hz), 7.32 (1H,

s), 7.43 (2H, d, J = 8.4 Hz), 7.53 (1H, s), 8.14 (2H, d, J = 8.4

Hz), 8.38 (s, 1H). ¹³C{¹H} NMR (100 MHz, CDCl₃) δ (ppm):

20.1, 23.8, 32.2, 54.0, 83.0, 84.0, 100.0, 104.4, 119.9, 122.4,

131.6, 133.1, 135.0, 139.9, 140.9, 167.2.

Ruthenium−Octreotide Conjugate 12. Characterization: R_t

= 17.7 min (analytical gradient: 30% to 100% in 30 min); HR

ESI MS, positive mode: m/z 1768.6863 (calcd mass for

C₉₃H₁₁₂ClN₁₃O₁₂PRu [M]⁺: 1768.6867), m/z 884.8481 (calcd

mass for C₇₀H₁₀₂ClN₁₃O₁₄Ru [M+H]²⁺: 884.8478). The MS

data correspond to the dicarba analogue of octreotide with

CHCH linkage.

Fluorescein-Labeled Dicarba Analogue of Octreotide 13.

Resin-bound linker-derivatized dicarba analogue of octreotide

was allowed to react with 5(6)-carboxyﬂuorescein (5 mol

equiv), DIPC (5 mol equiv), and HOAt (5 mol equiv) in

anhydrous DMF for 16 h, protected from light. After ﬁltration

and washing, peptide-bound resin was treated with 20%

piperidine/DMF (2 × 45 min). Overall yield (synthesis +

puriﬁcation): 40%. Characterization: R_t= 23.3 min (analytical

gradient: 20% to 60% in 30 min); HR ESI MS, positive mode:

m/z 1486.6572 (calcd mass for C₇₈H₉₂N₁₁O₁₉[M+H]⁺:

1486.6571).

[(η⁶-p-cym)RuCl(Im-BzCOOMe)(PPh₃)][PF₆] (17). Complex

[(η⁶-p-cym)RuCl₂(Im-BzCOOMe)] (81 mg, 0.16 mmol) was

reacted with PPh₃(1 mol equiv) in methanol (5 mL) for 1 h

under reﬂux. The resulting yellow solution was cooled to room

temperature and ammonium hexaﬂuorophosphate (1.1 mol

equiv) dissolved in methanol (1 mL) was added. A yellow

crystalline solid was ﬁltered oﬀ and washed several times with

1

methanol and diethyl ether (106 mg, 90%). H NMR (400

MHz, acetone-d₆) δ (ppm): 1.19 (6H, dd, J = 7.2 Hz), 1.83

(3H, s), 2.60 (1H, sp, J = 7.2 Hz), 3.93 (3H, s), 5.36 (1H, dt, J

= 6 Hz), 5.87 (1H, d, J = 6 Hz), 6.07 (1H, d, J = 6 Hz), 6.20

(1H, d, J = 6 Hz), 7.44−7.54 (17H, m), 7.65 (2H, dt, J = 10.4

Hz), 8.15 (2H, d, J = 8.8 Hz), 8.41 (1H, s). ³¹P{¹H} NMR

(acetone-d₆) δ (ppm): 36.51 (s). ¹³C{¹H} NMR (100 MHz,

acetone-d₆) δ (ppm): 19.1, 22.3, 24.2, 32.7, 53.9, 87.9, 90.2,

90.8, 95.7, 95.8, 105.0, 115.2, 115.3, 121.8, 123.1, 130.2, 130.3,

132.0, 132.1, 132.5, 132.9, 132.8, 133.0, 136.1, 136.2, 141.1,

141.3, 167.1. HR-ESI MS, positive mode: m/z 735.1488 (calcd

mass for C₃₉H₃₉ClN₂O₂PRu [M]⁺: 735,1481), m/z 533.0729

(calcd mass for C₂₈H₂₉ClPRu [M-(Im-BzCOOMe)]⁺:

533.0739), m/z 497.0959 (calcd mass for C₂₈H₂₈PRu [M-

(Im-BzCOOMe)-Cl-H]⁺: 497.0972).

DNA Binding Studies. Complexation reactions were

carried out in H₂O (4 mM NaCl) at 37 °C. The required

volume of an aqueous solution of the corresponding metal−

octreotide conjugate was mixed with the required volume of an

aqueous solution of 5′dCATGGCT (1−2 mol equiv).

Regarding conjugate 5, DNA binding studies were performed

with the analogue with the CHCH linkage isoster of

octreotide (denoted as 5a). In all cases, the solutions were 100

μM in the oligonucleotide. The evolution of the reactions was

monitored by reversed-phase HPLC on Kromasil C₁₈columns

(250 × 4.6 mm, 10 μm, ﬂow rate: 1 mL/min), using linear

gradients of aqueous triethylammonium acetate (0.05 M)

(solvent A) and ACN (solvent B) or ACN/H₂O 1:1 (solvent

B′). Metal adducts were isolated after several HPLC runs by

using analytical separation conditions. HR MALDI-TOF MS

analysis was carried out in the negative mode using 2,4,6-

trihidroxyacetophenone matrix with ammonium citrate as an

additive. Enzymatic digestions with 5′- and 3′-exonucleases

(bovine spleen and snake venom phosphodiesterases, respec-

tively) were performed as previously described.⁴¹⁻⁴³

Fluorescein-Labeled Octreotide 14. Linear ﬂuorescein-

linker-derivatized octreotide was assembled on a Rink amide

resin as described above with minor modiﬁcations: Fmoc-Hag-

OH was replaced by Fmoc-Cys(Trt)-OH and all Fmoc-

protected amino acids, Fmoc-linker-OH and 5(6)-carboxy-

ﬂuorescein (5 mol equiv) were coupled with DIPC (5 mol

equiv) and HOAt (5 mol equiv) in anhydrous DMF. After

treatment with 20% piperidine/DMF (2 × 45 min) and acidic

cleavage and deprotection, the linear peptide was dissolved

(ﬁnal concentration ca. 1 mM) in an aqueous ammonium

hydrogen carbonate solution (5%, pH 7.7) and stirred at room

temperature for 72 h (some drops of DMSO and MeOH were

added after 24 h to improve solubility as well as to increase the

oxidation rate). Overall yield (synthesis + puriﬁcation): 15%.

Characterization: R_t= 23.5 min (analytical gradient: 20% to

60% in 30 min); HR-ESI MS, positive mode: m/z 1522.5673

(calcd mass for C₇₆H₈₈N₁₁O₁₉S₂[M+H]⁺: 1522.5699).

Synthesis of Metal Complexes. [(η⁶-p-cym)RuCl(dap)]-

[PF₆] (16). (L)-2,3-Diaminopropionic acid methyl ester

dihydrochloride (0.068 g; 0.359 mmol) was reacted with

complex 9 (0.1 g; 0.163 mmol) and NEt₃(513 μL; 1.63 mmol)

in MeOH (15 mL). The mixture was stirred at room

temperature for 3 h. The resulting solution was ﬁltered and

the volume of the ﬁltrate reduced (to ca. 7 mL) in a rotary

evaporator. Solid NH₄PF₆(10 mol equiv) was added to the

solution and stirred at room temperature for 4 h. The resulting

orange microcrystalline solid was ﬁltered, washed with cold

1

MeOH (10 mL), and dried in vacuo (68 mg, 40%). H NMR

(400 MHz, DMSO-d₆) δ (ppm): 1.22 (6H, d, J = 6.8 Hz), 2.15

(3H, s), 2.16−2.31 (2H, m); 2.7 (1H, sp, J = 6.8 Hz), 3.02 (1H,

br s); 5.04−5.14 (2H, br m), 5.32 (1H, d, J = 5.6 Hz), 5.46

(1H, d, J = 6 Hz), 5.63 (1H, d, J = 6 Hz), 6.67 (1H, d, J = 5.6

Hz), 5.88−5.97 (2H, br m). ¹³C{¹H} NMR (100 MHz,

DMSO-d₆) δ (ppm): 18.0, 22.5, 30.4, 58.9, 78.6, 79.4, 80.8,

81.2, 95.7, 101.6, 176.8. HR ESI MS, positive mode: m/z

339.0650 (calcd mass for C₁₃H₂₁N₂O₂Ru [M−Cl−H]⁺:

339.0646). HR ESI MS, negative mode: m/z 373.0228 (calcd

mass for C₁₃H₂₀ClN₂O₂Ru [M−2H]⁻: 373.0255).

Adduct 5′dCATGGCT-conjugate 3. R_t= 11.0 min (gradient:

15% to 50% B in 30 min); HR MALDI-TOF MS, negative

mode: m/z 3501.0 (calcd mass for C₁₂₈H₁₇₁N₃₈O₅₅P₆Pt [M-

3H]⁻: 3500.95). MALDI-TOF MS after digestion with snake

venom phosphodiesterase: m/z 2907.9 (-pCpT) (calcd mass

for C₁₀₉H₁₄₆N₃₃O₄₂P₄Pt [M-3H]⁻: 2907.89). MALDI-TOF MS

after digestion with bovine spleen phosphodiesterase: m/z

2594.8 (-CpApTp) (calcd mass for C₉₉H₁₃₄N₂₈O₃₇P₃Pt [M-

[(η⁶-p-cym)RuCl₂(Im-BzCOOMe)]. A suspension of 9 (200

mg, 0.33 mmol) in 10 mL of DCM was reacted with the methyl

ester of the 4-(1H-Imidazol-1-yl)benzoic acid (1 mol equiv)

1841

dx.doi.org/10.1021/bc300173h | Bioconjugate Chem. 2012, 23, 1838−1855

Bioconjugate Chemistry

Article

3H]⁻: 2594.83), m/z 2898.9 (-CpAp) (calcd mass for

C₁₀₉H₁₄₇N₃₀O₄₄P₄Pt [M-3H]⁻: 2898.88).

medium (Dako, Carpinteria, CA, USA) and examined using a

Leica TCS-SP5 multiphoton and high-velocity spectral confocal

microscope (Leica Microsystems, Nussloch, Germany).

Adduct 5′dCATGGCT-conjugate 4. R_t= 21.2 min (gradient:

5% to 90% B′ in 30 min); HR MALDI-TOF MS, negative

mode: m/z 3715.8 (calcd mass for C₁₄₄H₁₇₈N₃₇O₅₇OsP₆[M-

2H]⁻: 3715.02). MALDI-TOF MS after digestion with snake

venom phosphodiesterase: m/z 2793.0 (-pGpCpT) (calcd mass

for C₁₁₅H₁₄₁N₂₇O₃₈OsP₃[M-2H]⁻: 2792.88). MALDI-TOF

MS after digestion with bovine spleen phosphodiesterase: m/z

2808.9 (-CpApTp) (calcd mass for C₁₁₅H₁₄₁N₂₇O₃₉OsP₃[M-

2H]⁻: 2808.87), m/z 3113.9 (-CpAp) (calcd mass for

Internalization Experiments with 13 and 14. The

uptake eﬃciency of 13 and 14 by the cells was quantiﬁed by

ﬂow cytometry. The cells were seeded onto 12-well plates and

allowed to attach for 24 h. Next, the cells were treated at 37 °C

with 50 μM of either 13 or 14 during 1, 3, and 6 h. After rinsing

the cells three times with cold PBS, the cells were harvested by

trypsinization and the ﬂuorescence of the cells, corresponding

to the uptake of the ﬂuorescein-modiﬁed peptides, was analyzed

using a FACSCalibur (Becton Dickinson Immunocytometry

Systems, San Jose, CA) equipped with the CellQuest software

(Becton Dickinson). Fluorescence intensity was represented on

a 4 orders of magnitude log scale (1−10 000). Ten thousand

cells were analyzed in each experiment.

C

₁₂₅H₁₅₄N₂₉O₄₆OsP₄[M-2H]⁻: 3112.92).

Adduct 5′dCATGGCT-conjugate 5a. R_t= 20.5 min

(gradient: 5% to 90% B′ in 30 min); HR MALDI-TOF MS,

negative mode: m/z 3540.8 (calcd mass for

C

₁₃₈H₁₈₃N₃₈O₅₅P₆Ru [M-3H]⁻: 3540.02). MALDI-TOF MS

after digestion with snake venom phosphodiesterase: m/z

2617.6 (-pGpCpT) (calcd mass for C₁₀₉H₁₄₆N₂₈O₃₆P₃Ru [M-

3H]⁻: 2617.87). MALDI-TOF MS after digestion with bovine

spleen phosphodiesterase: m/z 2633.7 (-CpApTp) (calcd mass

for C₁₀₉H₁₄₆N₂₈O₃₇P₃Ru [M-3H]⁻: 2633.87), m/z 2937.7

(-CpAp) (calcd mass for C₁₁₉H₁₅₉N₃₀O₄₄P₄Ru [M-3H]⁻:

2937.91).

Cytotoxicity Assays. The cytotoxicity of the conjugates

and of the control complexes in MCF-7, DU-145, and CHO

cells was determined by the MTT assay. Aliquots of 4000 DU-

145 cells, 3500 MCF-7, or 2000 CHO cells were seeded onto

ﬂat-bottomed 96-well plates. Twenty-four hours later, the cells

were treated for 72 h with the compounds at concentrations

ranging from 0 μM to 250 μM. After removal of the treatment,

the cells were washed with PBS and incubated for 3 additional

hours with 100 μL of fresh culture medium together with 10 μL

of MTT (Sigma-Aldrich). The medium was discarded and

DMSO (Sigma-Aldrich) was added to each well to dissolve the

purple formazan crystalls. Plates were agitated at room

temperature for 10 min, and the absorbance of each well was

determined on a Multiscan Plate Reader (ELX800, Biotek,

Winooski, USA) at a wavelength of 570 nm. Three replicates

were used in each experiment. For each treatment, the cell

viability was determined as a percentage of the control

untreated cells, by dividing the mean absorbance of each

treatment by the mean absorbance of the untreated cells. The

concentration that reduces by 50% the cell viability (IC₅₀) was

established for each compound.

Ruthenium Accumulation in Cancer Cells. For

ruthenium cellular uptake studies, 1.5 × 10⁶MCF-7 cells

were plated in 100 mm Petri dishes and allowed to attach for 48

h. Next, the plates were exposed to the ruthenium−octreotide

conjugate 6 or to the unconjugated ruthenium complex 17 at a

concentration corresponding to a ﬁfth of their IC₅₀(12.5 μM

and 0.66 μM, respectively). Additional plates were incubated

with medium alone as negative control. After 24 h of

incubation, the cells were rinsed three times with cold PBS

and harvested by trypsinization. The number of cells in each

sample was counted manually in a hemocytometer using the

Trypan blue dye exclusion test. Then, the cells were centrifuged

to obtain the whole cell pellet for ICP-MS analysis. All

experiments were conducted in triplicate.

NMR Spectroscopy. The samples of the conjugates were

prepared by dissolving the appropriate amount of compound in

500 μL of H₂O/DMSO-d₆8:2 to give 0.8 mM solutions. NMR

spectra were acquired on Bruker Avance spectrometers

operating at 600 or 800 MHz, and processed with Topspin

software. NOESY experiments were acquired with mixing times

of 150 and 300 ms, and TOCSY spectra were recorded with

standard MLEV-17 spin-lock sequence, and 80 ms mixing time.

Water suppression in H₂O experiments was achieved by

including a WATERGATE module in the pulse sequence prior

to acquisition. NMR experiments were carried out at

temperatures ranging from 0 to 25 °C. The spectral analysis

program Sparky⁴⁴was used for semiautomatic assignment of

the NOESY cross-peaks and quantitative evaluation of the

NOE intensities. Sequential assignment of the peptide moiety

1

was carried out using standard H NMR techniques.

Structural Modeling. Molecular dynamics calculations

were carried out with the program Sybyl X-1.3 (Tripos Inc.).

Ten initial models for each conjugate were built from the

dicarba analogue of octreotide 2 coordinates.²⁴Metal complex

coordinates were built from crystallographic structures of

similar compounds.,^36,45Each of the initial structures was

minimized and then submitted to 300 ps of molecular dynamics

simulations. Explicit solvent and periodic boundary conditions

were included in the calculations. The last 50 ps of each run

was averaged and energy-minimized. Backbone geometry of

those peptide residues that do not exhibit signiﬁcant chemical

shift variations was loosely constrained by imposing dihedral

angle restraints. Analysis of the representative structures was

carried out with the programs Sybyl and MOLMOL.⁴⁶

Immunocytochemistry. MCF-7 cells were seeded on

ﬁbronectin-coated coverslips and allowed to attach overnight.

Then, the cells were incubated at 37 °C with medium

containing 50 μM of 13 during 1, 3, and 6 h, or with medium

alone as control. After that, the cells were washed with cold

PBS (Gibco BRL) and ﬁxed with 4% paraformaldehyde in PBS

for 15 min at 4 °C. The cells nucleus were stained with 2 μg/

mL Hoescht 33258 (excitation/emission: 352 nm/461 nm)

during 15 min. After washing twice with cold PBS, the cells

coverslips were mounted using a ﬂuorescence mounting

ICP-MS Analysis. The whole cell pellets were dissolved in

300 μL of concentrated 72% v/v nitric acid, and the samples

were then transferred into Wheaton v-vials (Sigma-Aldrich) and

heated in an oven at 373 K for 18 h. The vials were then

allowed to cool, and each cellular sample solution was

transferred into a volumetric tube and combined with washings

with Milli-Q water (1.7 mL). Digested samples were diluted 10

times with Milli-Q to obtain a ﬁnal HNO₃concentration of

approximately 2.5% v/v. Ruthenium content was analyzed on

an ICP-MS Perkin-Elmer Elan 6000 series machine at the

́

̀

Centres Cientıﬁcs I Tecnologics of the Universitat de

Barcelona. The solvent used for all ICP-MS experiments was

1842

dx.doi.org/10.1021/bc300173h | Bioconjugate Chem. 2012, 23, 1838−1855

Bioconjugate Chemistry

Article

Scheme 1. Schematic Representation of the Solid-Phase Approach Used for the Synthesis of Conjugates 3 and 4

Milli-Q water with 1% HNO₃. The ruthenium standard (High-

Purity Standards, 1000 μg/mL 5 μg/mL in 2% HCl) was

diluted with Milli-Q water to 100 ppb. Ruthenium standards

were freshly prepared in Milli-Q water with 1% HNO₃before

each experiment. The concentrations used for the calibration

curve were in all cases 0, 1, 2, 5, and 10 ppb. The isotope

detected was ¹⁰¹Ru and readings were made in triplicate.

Rhodium was added as an internal standard at a concentration

of 10 ppb in all samples.

SSTR2 Expression Analysis. SSTR2 expression on MCF-

7, DU-145, and CHO cells surface was determined by double

immunoﬂuorescence. The cells were ﬁrst ﬁxed with PBS-1.5%

formaldehyde and permeabilized with PBS-0.2%Tween 20

(Biorad). Then, the cells were incubated with a monoclonal

antibody against human sst₂receptor (R&D systems) during 30

min at 4 °C. After rinsing the cells with PBS, the cells were

incubated 30 min at 4 °C in the presence of the Alexa-Fluor

488 conjugated goat antimouse IgG antibody (Invitrogen).

Cells were washed again, and the ﬂuorescence was analyzed

using a ﬂow cytometer (FACSCalibur, Becton Dickinson

Immunocytometry Systems, San Jose, CA) equipped with the

CellQuest software (Becton Dickinson). The ﬂuorescence

intensity of the cells was represented on a 4 orders of

magnitude log scale (1−10 000). Ten thousand cells were

analyzed in each experiment.

tested using the Kolmogorov−Smirnov test. The diﬀerences

between data with normal distribution and homogeneous

variances were analyzed using the parametric Student’s t test. A

value of p < 0.05 was considered signiﬁcant.

RESULTS AND DISCUSSION

■

Synthesis and Characterization of Conjugates be-

tween a Dicarba Analogue of Octreotide and

Dichloridoplatinum(II) and Osmium(II) Complexes. In

our previous work,²³cis-dichlorido-(1-(carboxylic acid)-1,2-

diaminoethane)platinum(II) complex, ([PtCl₂(dap)]), was

chosen as a cisplatin analogue since it can form the expected

intrachain Pt-N7/5′G,N7/3′G chelates upon reaction with

DNA. The attachment of the platinum complex to the γ-

aminoisobutyryl-derivatized dicarba analogue of octreotide (2)

was carried out on a solid phase through the formation of an

amide bond between its carboxylic function and the amine

group of the spacer. Although the conjugate was able to bind to

DNA in the same way as cisplatin, its low solubility in water

precluded the evaluation of its biological activity. In order to

improve the aqueous solubility for all the target conjugates, 3−

6 (Chart 1), a poly(ethylene glycol) spacer has been introduced

between the peptide sequence and the metal complex. This

spacer is expected to be suﬃciently long to keep the metal

complex away from the pharmacophore sequence.

Statistical Analysis. The statistical analysis was performed

with the SPSS statistical software for Windows (v 15.0; SPSS

Inc., Chicago, IL, USA). Quantitative variables were expressed

as mean and standard error (SE). The normality of the data was

Following our previous results on the synthesis of platinum-

(II)−octreotide conjugates, a stepwise solid-phase strategy has

been used for the preparation of conjugates 3−6. This

approach is particularly useful in this case, since it allows the

1843

dx.doi.org/10.1021/bc300173h | Bioconjugate Chem. 2012, 23, 1838−1855

Bioconjugate Chemistry

Article

Scheme 2. Schematic Representation of the Solid-Phase Approach Used for the Synthesis of Conjugates 5, 6, and 12

regioselective introduction of the metal complex at the N-

terminal end of the cyclic peptide. Otherwise, in a solution-

phase approach the ε-NH₂of the Lys residue would compete

with the N-terminal amino group of octreotide. In order to

increase the stability of the metal−octreotide conjugates, the

CH₂−CH₂linkage has been chosen as a disulﬁde isoster, since

this modiﬁcation keeps a higher selectivity for sst₂somatostatin

receptor than the CHCH linkage.²⁴

from the resin and removal of the protecting groups were

carried out by treatment with a TFA/phenol/H₂O cocktail

(95:2.5:2.5) for 1 h at rt. Reversed-phase HPLC analysis of the

crude revealed a main double peak. Both peaks were isolated by

semipreparative HPLC and characterized by HR MALDI-TOF

and ESI MS and NMR as the expected diastereomeric

dichloridoplatinum(II)−octreotide conjugates, 3. The gener-

ation of both isomers is accounted from the use of a racemic

platinum(II) complex.

First, the linear peptide was synthesized manually on a Rink

amide resin-p-MBHA using standard Fmoc-tBu methodology

(Scheme 1). All suitably protected Fmoc-amino acids were

incorporated with HATU in the presence of DIPEA in

anhydrous DMF as a solvent. As previously described,^23,24

Fmoc-protected threoninol functionalized as the p-carboxyben-

zaldehyde acetal was ﬁrst anchored to the solid support. In

addition, the two cysteine amino acids in octreotide were

replaced by allyl glycine residues to perform the cyclization

reaction. Once the linear peptide precursor had been

assembled, on-resin microwave-assisted ring-closing metathesis

was carried out with second-generation Grubbs catalyst for 1 h

at 100 °C. Then, hydrogenation of the double bond using

Wilkinson’s catalyst aﬀorded the protected cyclic saturated

dicarba analogue of octreotide (2) bound to the resin. Finally,

an Fmoc-protected poly(ethylene glycol) linker, 8-(9-ﬂuore-

nylmethyloxycarbonyl-amino)-3,6-dioxaoctanoic acid, was in-

corporated at the N-terminal end using HATU as a coupling

reagent.

Our next objective was focused on the conjugation of an

organometallic osmium(II) complex to the dicarba analogue of

octreotide. Among osmium complexes with anticancer activity,

[(η⁶-bip)Os(pico)Cl] is particularly interesting, since it exhibits

promising activity in human ovarian cancer cells.^47,48For this

reason, we choose an analogue bearing a carboxylic function on

the picolinate ligand, [(η⁶-bip)Os(4-CO₂-pico)Cl] (8), which

would allow the attachment to octreotide. This complex has

been very recently conjugated to polyarginines with diﬀerent

chain lengths using solid-phase procedures.³⁰Hence, for the

synthesis of conjugate 4, [(η⁶-bip)Os(4-CO₂-pico)Cl] was

coupled with either PyBOP or the more reactive HATU in the

presence of DIPEA in anhydrous DMF (Scheme 1). After

cleavage and deprotection with a TFA/TIS/H₂O cocktail

(95:2.5:2.5; 1 h at rt, protected from light), reversed-phase

HPLC analysis revealed a main peak (30−40%), which was

isolated and characterized by HR MS and NMR as the target

conjugate 4. The conversion yield was substantially increased

(90%) when coupling was performed with DIPC and HOAt in

anhydrous DMF for 3 h. It is important to mention that, during

HPLC analysis of conjugate 4, a less retained peak was always

observed, which was characterized by MS as the aqua conjugate

(substitution of Cl by H₂O). This result reveals fast hydrolysis

The ﬁnal step involved the removal of the Fmoc protecting

group of the spacer (20% piperidine in DMF) and the

incorporation of the required metal complex onto the free

amino function of the linker. For the synthesis of conjugate 3,

[PtCl₂(dap)] (7) was coupled using DIPC and NHS in

anhydrous DMF (15 h at rt, protected from light).²³Cleavage

1844

dx.doi.org/10.1021/bc300173h | Bioconjugate Chem. 2012, 23, 1838−1855

Bioconjugate Chemistry

Article

kinetics for this osmium(II) complex in comparison with that

of the platinum(II) complex (see below).

focused on compounds of the type [(η⁶-p-cym)RuCl(Im-

Ph)(PPh₃)]⁺, where Im-Ph is 1-phenyl-imidazole. The

conjugation of such triphenylphosphine-containing complexes

to octreotide was planned through the phenyl ring linked to the

imidazole ligand, which would allow the coordination of the

ruthenium moiety. As in previous metal complexes, the

hydrolysis of the Ru−Cl bond was expected to generate

corresponding aqua species reactive toward DNA binding.

First, 4-(1H-imidazol-1-yl)benzoic acid was coupled onto

two batches of linker-octreotide resin with DIPC and HOBt in

anhydrous DMF for 1 h at rt, where linker refers to γ-

aminoisobutyryl group or the poly(ethylene glycol) spacer

(Scheme 2). The assembly of the metal complex was ﬁrst

explored with the γ-Abu linker-containing batch by using two

ruthenium compounds, [(η⁶-p-cym)RuCl₂(PPh₃)] (10) and

[(η⁶-p-cym)RuCl(PPh₃)₂]⁺(11), since we hypothesized that

the imidazole ligand could replace the more labile chlorido

ligands. 10 was easily synthesized by reﬂuxing dimer 9 with

triphenylphosphine (2.3 mol equiv) in hexane for 15 h, whereas

complex 11 was obtained from 10 by reaction with a slight

excess of triphenylphosphine (2 mol equiv) in methanol at 35

°C for 2.5 h.^37,38

Synthesis and Characterization of Conjugates be-

tween a Dicarba Analogue of Octreotide and

Ruthenium(II) Arene Complexes. Ruthenium−octreotide

conjugates, 5 and 6, were synthesized on a solid-phase

following a diﬀerent strategy than that used for platinum and

osmium conjugates (Scheme 2). Thus, instead of synthesizing

the required ruthenium complexes bearing a carboxylic

function, we decided to incorporate the appropriate ligand at

the N-terminal end of the linker-derivatized peptide, which

would allow the assembly of the organometallic ruthenium(II)

complex around it. This approach avoids some of the diﬃculties

found during the synthesis of carboxylic acid containing

ruthenium complexes, and it could be applied to the parallel

synthesis of libraries of ruthenium−peptide conjugates.

First, we focused on mononuclear ruthenium(II) arene

complexes of the type [(η⁶-arene)RuCl(XY)]⁺, where XY is a

N,N-chelating ligand such as ethylenediamine. These com-

pounds show high cytotoxicities both in vitro and in vivo, with

IC₅₀values comparable to carboplatin in human ovarian cancer

cell lines (e.g., 9 μM for [(η⁶-p-cym)RuCl(en)][PF₆]).⁴⁹In

addition, they form monofunctional adducts at guanine

nucleobases on DNA.

Reaction of the imidazole-derivatized peptide-bound resin

with complex 10 (4 mol equiv) was performed in the presence

of NEt₃(16 mol equiv) in a DMF/EtOH mixture (3:1) for 8 h

at 55 °C. After treatment of the resin with a TFA/phenol/H₂O

cocktail (95:2.5:2.5), the desired conjugate 12 was detected in

the crude (18%) together with the imidazole-derivatized

octreotide (20%), as inferred by HPLC and MS analysis

(Figure 1A).

Attachment of [(η⁶-p-cym)RuCl(en)]⁺to octreotide was

planned through the ethylenediamine ligand, as perfomed for

the dichloridoplatinum(II) conjugate, which would lead to

conjugate 5 (Chart 1). In order to incorporate 2,3-

diaminopropionic acid into the N-terminal amino function of

the linker-derivatized dicarba analogue of octreotide bound to

the resin, the required N^α,N^β-bis-(9-ﬂuorenylmethyloxycarbon-

yl)-L-2,3-diaminopropionic acid was synthesized by reaction

between ^L-2,3-diaminopropionic acid hydrochloride and 9-

ﬂuorenylmethyloxycarbonyl chloride. This Fmoc-protected

chelating ligand was coupled with HATU in DMF in the

presence of DIPEA for 1 h at rt (Scheme 2). After removal of

the Fmoc protecting groups, the binding of the ruthenium

arene to the chelating ligand was studied by reaction with the

ruthenium dimer [{(η⁶-p-cym)Ru(μ-Cl)Cl}₂] (9). Optimized

conditions required microwave irradiation at 100 °C for 1 h and

the use of a slight excess of 9 (3 mol equiv) in the presence of

LiCl (6.6 mol equiv) and NEt₃(7.2 mol equiv) in a DCM/

EtOH 9:1 mixture. The use of microwave irradiation in

combination with this solvent was proven to be particularly

important to achieve a good conversion yield. However, 30−

40% conversion yields where obtained when DMF replaced

DCM, and no irradiation was used (15 h heating at 60 °C).

These results might be attributed to the instability of 9 at high

temperatures during a prolonged time. Cleavage from the resin

and deprotection was performed by treatment with a TFA/

phenol/H₂O cocktail (95:2.5:2.5) for 1 h at rt. Reversed-phase

HPLC analysis of the crude revealed a main peak (65%) which

was isolated by semipreparative HPLC and characterized by

HR MALDI-TOF and ESI MS as the expected ruthenium−

octreotide conjugate, 5.

Figure 1. Reversed-phase HPLC traces of the reaction crudes for the

synthesis of conjugate 12 by using complex 10 (A) or 11 (B) in the

absence of LiCl or 11 in the presence of LiCl (C). Reaction crude for

the synthesis of conjugate 6 when 11 was used in the presence of LiCl

(D).

We next investigated the reaction of octreotide-bound resin

with complex 11 under the same conditions. Unfortunately,

reversed-HPLC analysis only revealed the presence of the

imidazole-derivatized peptide, which indicates that imidazole

was not able to substitute the chloride ligand in 11 (Figure 1B).

This negative result might be attributed to the high steric

hindrance around the metal because of the two bulky PPh₃

ligands. However, to our surprise, when the reaction of 11 (4

mol equiv) with the peptide was repeated in the presence of

LiCl (4.4 mol equiv) and NEt₃(16 mol equiv) in a DMF/

EtOH 9:1 mixture for 8 h at 55 °C, HPLC analysis of the

Once we had synthesized conjugate 5, we explored the

possibility of assembling mononuclear ruthenium(II) arene

complexes through a single monodentate ligand incorporated

into the peptide fragment instead of using chelating ligands. On

the basis of the excellent antitumor activities of some

ruthenium(II) three-legged “piano stool” complexes containing

planar N-heteroaromatic σ-bonded ligands and phosphine

ligands,⁵⁰with IC₅₀values lower than that of cisplatin, we

1845

dx.doi.org/10.1021/bc300173h | Bioconjugate Chem. 2012, 23, 1838−1855

Bioconjugate Chemistry

Article

cleaved, deprotected crude (TFA/phenol/H₂O 95:2.5:2.5, 1 h

rt) showed the presence of a main peak (57%) which was

isolated and characterized by MS as conjugate 12 (Figure 1C).

Hence, imidazole is able to replace one of the two PPh₃ligands

in 11 instead of the Cl ligand, thereby allowing the conversion

yield of conjugate 12 to be improved in comparison with that

obtained when complex 10 is used. As shown in Figure 1D, the

assembly of the ruthenium complex by using these optimized

conditions was almost quantitative onto the imidazole-

derivatized octreotide resin that contains the poly(ethylene

glycol) linker, which allowed the target conjugate 6 to obtain

with high yield (Scheme 2). This compound was isolated by

semipreparative HPLC and fully characterized by HR MS and

NMR.

through hydrolysis. As previously stated, the activation of

piano-stool organometallic ruthenium(II) complexes of the

type [(η⁶-arene)RuCl(XY)]⁺, where XY is a neutral chelating

ligand, proceeds through the release of chloride ligand. The

resulting aqua species is responsible for ruthenation of DNA by

generating monofunctional adducts on guanine nucleobases. In

a similar manner to cisplatin and related platinum(II)

compounds, the high extracellular chloride concentration

(100 mM) in comparison with that present in the nucleus

and cytoplasm (4 and 22.7 mM, respectively)⁵²ensures that the

intact drug reaches its biological target before activation takes

place. In addition, a close correlation between cancer cell

cytotoxic activity and the capacity of some ruthenium(II) arene

complexes to undergo hydrolysis of the Ru-halide bond has

been established.⁵³Osmium(II) arene complexes containing

N,N-chelating ligands, such as ethylenediamine, experience

much slower ligand exchange kinetics than ruthenium

analogues.⁵⁴The higher acidity of the water molecule bound

to the osmium at physiological pH makes these compounds

exist largely in their less reactive hydroxido form. These

problems have been solved by using picolinate as chelating

ligand since it increases both the hydrolysis rate and the basicity

of the water molecule in the aqua species (e.g., [(η⁶-

arene)Os(pico)Cl]) and, for instance, generates compounds

with similar anticancer activity as ruthenium-arene ana-

logues.^36,47

In order to gain insight into the mechanism of this reaction,

complex 11 was mixed with LiCl (1.4 mol equiv) and NEt₃(1.9

mol equiv) in anhydrous DMF and heated at 55 °C. As shown

in the ³¹P NMR spectra (Figure 2), the single chemical shift

On the basis to these precedents, we wanted to assess

whether hydrolysis of the M-Cl bond occurs when Pt, Ru, and

Os complexes are conjugated to octreotide. The stability of

conjugates 3−6 (Chart 1) was investigated in aqueous solution

at diﬀerent chloride concentrations mimicking typical blood

plasma, cell cytoplasm, and cell nucleus (100, 22.7, and 4 mM,

respectively). In all cases, metal−octreotide conjugates (10

μM) were incubated at 37 °C for 24 h. In general, HPLC

analysis showed that the peak of the parent chloride compound

(e.g., R_t= 20.3 min for conjugate 4) evolved into a new less

retained peak (R_t= 16.9 min), which was isolated and

characterized by ESI MS as the aqua conjugate (substitution of

Cl by H₂O).

Figure 2. Schematic representation of the transformation of complex

11 to 10 (left) and ³¹P NMR spectra of 11 (A) and of the reaction

crude after heating 11 in the presence of LiCl and NEt₃for 4 h at 55

°C (B).

As previously found in some osmium(II) complexes,⁴⁸the

hydrolysis extent of conjugate 4 was highly dependent on the

NaCl concentration. Almost 95% of the conjugate was

hydrolyzed in the 4 mM NaCl solution, but the hydrolysis

was substantially decreased in the more concentrated solutions

of NaCl (Table 1). These results indicate that the covalent

attachment of the peptide does not interfere with the hydrolysis

properties of this osmium complex and,⁴⁸more importantly,

that the aqua species is not deactivated by reaction with amino

corresponding to the two PPh₃ligands in 11 (δ: 20.0 ppm)

disappeared after 4 h. The two new signals (δ: 23.2 and −6.4

ppm) were assigned to the PPh₃ligand in complex 10 and to

the released PPh₃, respectively. These results demonstrate the

in situ generation of complex 10 from 11 because of the

presence of LiCl. This reaction might have future applications

in the synthesis of new ruthenium complexes for catalysis or

biomedical applications.

Activation of Metal−Octreotide Conjugates by Aque-

ous Hydrolysis. In general, the cytotoxic activity of metal-

based anticancer drugs is intimately related with the processes

that mediate their activation in aqueous media; these processes

are known to facilitate the interaction of the metal, its ligands or

the complex, or a fragment with the biological target (e.g.,

DNA, RNA, or proteins). In addition, aqueous stability is

another fundamental prerequisite, since it might condition their

success in clinical development. Cisplatin and its analogues can

be considered prodrugs since their activation is produced

through the dissociation of anionic ligands.^22,51The resulting

positively charged aqua species coordinate strongly to nuclear

DNA while neutral complexes do not, which demonstrates that

in these family of compounds the metal plays a functional

role.²²Organometallic complexes containing a labile group

prone to substitution can also be activated inside the cells

Table 1. Percentage of Aqua Adduct Formation in Water at

Chloride Levels Typical of Blood Plasma (100 mM), Cell

Cytoplasm (22.7 mM), and Cell Nucleus (4 mM), and

Percentage of the Conjugate−DNA Adduct Formation in

Aqueous 4 mM NaCl

% aqua adduct

% DNA adduct

4 mM NaCl

conjugate 4 mM NaCl 22.7 mM NaCl 100 mM NaCl

3

16

95

97

0

7

3

42

49

37

0

4

35

83

0

12

74

0

a

5a

6

a

See the experimental section.

1846

dx.doi.org/10.1021/bc300173h | Bioconjugate Chem. 2012, 23, 1838−1855

Bioconjugate Chemistry

Article

acids from the carrier peptide. Similar results were obtained for

the ruthenium conjugate 5 at the 4 mM chloride concentration.

However, the extent of hydrolysis of the Ru−Cl bond was

substantially decreased at the higher NaCl concentrations

(Table 1).

Scheme 3. Formation of Adducts between Conjugates 3−5

and the Oligonucleotide

Regarding conjugate 3, the hydrolysis of the Pt−Cl bond

occurred to a lower extent than for conjugates 4 and 5 (only

16% after 24 h in 4 mM NaCl solution), and it was almost

abolished at higher chloride concentration.

On basis to these results, it is expected that octreotide in

conjugates 3 and 4 will have a positive eﬀect on the delivery of

the less reactive intact prodrug while outside the cell, but once

in the cell cytoplasm and, particularly, in the cell nucleus, Pt

and Os complexes will be selectively activated through

hydrolysis to generate a species reactive toward DNA binding.

However, hydrolysis of conjugate 5 in the blood plasma and in

the cell cytoplasm might have important consequences for its

biological activity since the metal complex could be deactivated

by reaction with other potential ligands such as peptides or

proteins.

To our surprise, no hydrolysis occurred in the case of

conjugate 6 upon incubation for 24 h at 37 °C, as inferred by

HPLC analysis together with MS (see below).

DNA Binding Studies. In general, the chlorido species of

Pt(II), Ru(II), and Os(II) complexes are relatively less reactive

toward DNA nucleobases in comparison with the aqua adducts.

The chloride concentration in the nucleus of cells (ca. 4 mM) is

much less than that in the cytoplasm (ca. 22.7 mM) or outside

cells (ca. 100 mM), which favors the formation of reactive aqua

adducts in the vicinity of DNA. We incubated conjugates 3−6

in 4 mM NaCl at 37 °C with the short DNA oligonucleotide,

5′dCATGGCT. This oligonucleotide contains the target GG

sequence of the anticancer drug cisplatin and is known to give

the expected platinum 1,2-intrastrand GG chelates upon

reaction with cisplatin, [Pt(en)Cl₂], or a dichloridoplatinum(II)

conjugate of a dicarba analogue of octreotide.^23,41

In the case of the platinum conjugate 3, HPLC analysis

showed the formation of a new peak with higher retention time

than the parent oligonucleotide (42% after 24 h; see Table 1),

attributable to the hydrophobicity of the platinum−peptide

fragment. This new product was isolated and characterized by

HR MALDI-TOF MS as an adduct between 3 and

5′dCATGGCT in which both chloride ligands had been lost.

As expected from previous results,²³MS analysis after

enzymatic digestion with 5′- and 3′-exonucleases (bovine

spleen and snake venom phosphodiesterases, respectively)^42,43

revealed the formation of a chelate between the platinum

fragment of the conjugate and both guanines of the DNA chain

(Scheme 3).

When osmium− and ruthenium(dap)−octreotide conjugates

were incubated with 5′dCATGGCT, a major monofunctional

adduct was formed between the metal fragment of the

conjugate and the DNA chain. In both cases, the high

metalation yield (49% for 4 and 37% for 5) is in good

agreement with the high degree of hydrolysis of the Os/Ru−Cl

bonds. Enzymatic digestion in combination with MS revealed

that in both cases metalation occurred at the guanine in the 5′

position (G4, see Scheme 3). This is a similar trend to that

previously found with monofunctional platinum(II) complexes,

such as {Pt(dien)}²⁺, since they react exclusively with the 5′G

nucleobase in the 5′dCATGGCT sequence.⁴¹The high aﬃnity

of similar ruthenium(II) arene complexes, such as [(η⁶-p-

cym)RuCl(en)]⁺or [(η⁶-benzene)Ru(en)(OH₂)]²⁺, for the N7

position of guanine over the other DNA nucleobases has been

observed previously.⁵⁵⁻⁵⁷Regarding organometallic osmium-

(II) complexes with picolinate ligands, such as [(η⁶-biphenyl)-

Os(picolinate)Cl]⁺, rapid binding to calf thymus DNA has also

been found.⁵⁸

Finally, the reaction of the ruthenium−octreotide conjugate

6 with DNA was studied. In good agreement with the absence

of hydrolysis of the Ru−Cl bond in this compound, ruthenation

of 5′dCATGGCT was not observed even after 48 h.

The overall results demonstrate that the activation of the M-

Cl bond through hydrolysis in the Ru and Os conjugates seems

to be very important for the binding to DNA, and there is a

strong preference for the guanine nucleobases, especially those

located at the 5′ end in GG sequences. However, in the case of

the Pt(dap)-octreotide conjugate, platination is observed

despite the slow kinetics for the hydrolysis of the Pt−Cl

bond in the low chloride medium as found in the nucleus. The

high thermodynamic stability of the N7,N7-GG intrastrand

adduct might account for this result.

Eﬃciency of Intracellular Delivery of Dicarba Ana-

logue of Octreotide (2). Compared to octreotide (1), the

binding aﬃnity of the dicarba analogue of octreotide (2) (Chart

1) is slightly reduced for sst₂, sst₄, and sst₅receptors, although

this reduction is smaller in the case of sst₂receptor (about 23-

fold for sst₂vs 36- and 48-fold for sst₄and sst₅, respectively).^6,24

However, as previously stated, this high aﬃnity toward sst₂

receptor (44 nM) still makes 2 a suitable stabilized analogue of

somatostatin to deliver cytotoxic metal complexes into cancer

cells. In order to check the capacity of internalization of

analogue 2 in our cell lines by confocal microscopy and ﬂow

cytometry, we labeled it with ﬂuorescein. For the synthesis of

the ﬂuorescein-labeled dicarba analogue of octreotide (13),

5(6)-carboxyﬂuorescein was coupled on the linker derivatized

peptide-bound resin by using DIPC and HOAt as coupling

reagents. Washings with 20% piperidine/DMF before the ﬁnal

acidic treatment yielded a highly pure peptide.⁵⁹As a control,

ﬂuorescein-labeled octreotide (14) was also synthesized using

solid-phase procedures.

Next, the internalization capacity of analogue 2 was

determined by confocal microscopy, after the incubation of

1847

dx.doi.org/10.1021/bc300173h | Bioconjugate Chem. 2012, 23, 1838−1855

Bioconjugate Chemistry

Article

Figure 3. Confocal microscopic imaging of the internalization of the ﬂuorescein-labeled dicarba analogue of octreotide (13) in MCF-7 cells. Left:

MCF-7 cells incubated with 50 μM of 13 for 1, 3, and 6 h at 37 °C or medium alone as a control (CTL). The cell nuclei were stained with Hoechst

(blue). The localization of 13 is indicated by the green ﬂuorescence. Right: Higher-magniﬁcation image (1000×) of a MCF-7 cell incubated with 50

μM of 13 for 3 h at 37 °C. Arrows indicate the nuclear localization of 13.

Figure 4. Intracellular delivery eﬃciencies of ﬂuorescein-labeled dicarba analogue of octreotide (13) and of ﬂuorescein-labeled octreotide (14) in

MCF-7 cells. MCF-7 cells were incubated with the peptides for 1, 3, and 6 h at 37 °C or medium alone (CTL). The ﬂuorescence intensity of the

cells, corresponding to the intracellular uptake of the peptides, was determined by ﬂow cytometry. (A) Flow cytometry histograms representing the

ﬂuorescence intensity of 10 000 cells on a four-decade log scale. Peptides 13 and 14 are depicted by solid and dashed lines, respectively. (B) Kinetics

of the cellular uptake of 13 and 14. Each point in the graphs represents the mean intracellular ﬂuorescence intensity of three independent

experiments SE.

sstr₂-expressing human breast cancer MCF-7 cells with 50 μM

of the corresponding ﬂuorescein-labeled peptide (13) for

diﬀerent times (1, 3, and 6 h). After 1 h of treatment,

ﬂuorescent vesicles, most likely endosomes, were visible in the

cytoplasm of most of the examined cells, conﬁrming the cellular

uptake of the peptide (Figure 3). After 3 and 6 h, the number

and the ﬂuorescence density of the vesicles markedly increased,

revealing an intense internalization of 13 during this period of

time. At these time points, the vesicles were mainly clustered at

the periphery of the cell nucleus. Higher-magniﬁcation

(1000×) images from cells exposed to 13 for 3 h (Figure 3)

revealed that, although the ﬂuorescence vesicles were located

mainly in the cell cytoplasm, a small proportion of the

ﬂuorescence was localized within the cell nucleus. Some authors

have reported that somatostatin ligands may accumulate in the

cell nucleus,^60,61although the pathway for their nuclear

localization remains to be elucidated.

The qualitative observations in confocal microscopy studies

were conﬁrmed with quantitative data obtained by ﬂow

cytometry. As represented in Figure 4A, the mean intracellular

ﬂuorescence intensity, corresponding to the internalization of

13, increased over time. The intracellular ﬂuorescence intensity

1848

dx.doi.org/10.1021/bc300173h | Bioconjugate Chem. 2012, 23, 1838−1855

Bioconjugate Chemistry

Article

increased from 78 16 at 1 h to 111 17 at 3 h and 182 39

at 6 h conﬁrming an active cellular uptake of the peptide at

these time points. In order to evaluate if 13 internalizes to a

similar extent as native octreotide, MCF-7 cells were incubated

in parallel with 14. As expected from the higher aﬃnity of

octreotide for the somatostatin receptors,^6,24the intracellular

uptake of 14 was more elevated, by 40−57%, compared with

that obtained in cells incubated with 13. As shown in Figure 4B,

the internalization kinetics was parallel for both peptides and it

is also in good agreement with previous reports.^62,63Despite

the diﬀerences in the amount of internalized peptide, these

results demonstrate that the dicarba analogue of octreotide (2)

displays the internalization properties required to deliver the

cytotoxic metal complexes into cancer cells.

Cancer Cell Cytotoxicity. To estimate the in vitro

antitumor potential of the conjugates 3, 4, 5, and 6, their

antiproliferative activity was determined in the MCF-7 cell line.

The compounds were screened at a wide range of

concentrations (from 0 to 250 μM) to determine the

concentration that inhibits the cell growth by 50% (IC₅₀).

The conjugates that did not inhibit cell growth by more than

50% at 250 μM were considered to be inactive. As shown in

Table 2, only conjugate 6 displayed a measurable antiprolifer-

cell lines (IC₅₀= 44 and 61 μM, respectively) has been recently

reported in comparison with that of [(η⁶-bip)OsCl(4-Me-

pico)Cl)]⁺(IC₅₀= 4.4 and 7.6 μM in the A2780 and A2780cis

cell lines, respectively).³⁶Cytoxicity assays of the control

complexes revealed that 17 was highly eﬀective in MCF-7 cells,

displaying an IC₅₀value (3.32 0.54) comparable to that of

cisplatin in this cell line (IC₅₀= 3.04 0.25). In contrast, a

moderate antiproliferative activity was obtained for complex 15

(IC₅₀= 67.00

19.01) while complex 16 was noncytotoxic

(Table 2).

Two important conclusions can be drawn from these results.

First, high cytotoxic activity of the metal complex itself seems

to be crucial for the eﬃcacy of the conjugates as antitumoral

agents. Second, conjugation of the complexes to the peptide

moiety diminishes their cytotoxic capacity. For example, the

cytotoxic activity of the ruthenium(II) arene complex [(η⁶-p-

cym)RuCl(Im-BzCOOMe)(PPh₃)]⁺(17) is reduced about 19-

fold when conjugated to the octreotide analogue (6), whereas

complexes with moderate activity, such as [PtCl₂(Etdap)] (15)

or [(η⁶-bip)OsCl(4-CO₂Et-pico)Cl)]⁺(16),³⁶aﬀord inactive

peptide conjugates (3 and 4, respectively). On the other hand,

the lack of activity for complex [(η⁶-p-cym)RuCl(dap)]⁺(16)

or its octreotide conjugate (5) might be attributable to the high

hydrolysis rate of the Ru−Cl bond, which would cause

deactivation of the activated species by reaction with other

biomolecules before it reaches nuclear DNA.

Table 2. IC₅₀Values of Conjugates 3−6 and Control

a

Complexes 15−17 in MCF-7 Cells

It is also interesting that the only conjugate that displays

cytotoxic activity (6) does not interact with DNA, which can be

attributed to the absence of hydrolysis of the Ru−Cl bond,

even at a low chloride concentration (4 mM). This is also

consistent with the fact that control complex 17 does not

interact with double-stranded plasmid DNA after 48 h of

incubation, as determined by electrophoretic mobility experi-

ments (see Figure S-18 in the Supporting Information).

Similarly to conjugate 6, the aqua adduct of complex 17 was

only formed after overnight incubation with a large excess of

IC₅₀(μM)

Conjugates

Complexes

3

4

5

6

>250

16

63.00 1.53

17

15

-

67.00 19.01

>250

3.32 0.54

a

IC₅₀: Inhibitory concentration that reduce by 50% the cell viability.

The cytotoxicity was determined by the MTT assay after 72 h of drug

exposure. Data represents the mean SE of at least three independent

experiments.

1

AgNO₃, as inferred by H and ³¹P NMR (see the Supporting

Information). The overall results suggest that the cytotoxic

activity of 6 and 17 might not be a consequence of their

coordination to N-donor ligands from nucleic acids (DNA or

RNA) or proteins. However, it cannot be rule out that their

activation under physiological conditions occurs through

coordination to sulfur-containing amino acids in peptides

such as glutathione or in proteins.

In order to gain insight into the implication of somatostatin

subtype-2 receptor on the cytotoxic activity of the ruthenium−

octreotide conjugate 6, the cellular uptake of this compound

was compared with that of the unconjugated ruthenium

complex 17. Ruthenium accumulation (determined here as

the net eﬀect of inﬂux and eﬄux of Ru) in the MCF-7 cancer

cell line was determined by ICP-MS after a 24 h exposure to

the compounds. Interestingly, the intracellular level of

ative activity, and even that was only moderate (IC₅₀= 63.00

1.53). On the basis of these results, we wondered whether the

cytotoxic activity might be a consequence of the metal complex

itself or of the conjugation to the peptide moiety. Hence, we

tested the cytotoxicity of the parent metal complex alone,

including the ligand attached to the peptide. The following

complexes were synthesized and used as controls (Chart 2):

[PtCl₂(Etdap)] (15), [(η⁶-p-cym)RuCl(dap)]⁺(16), and [(η⁶-

p-cym)RuCl(Im-BzCOOMe)(PPh₃)]⁺(17). Although we

planned to mask the carboxylic function in all the complexes

as a methyl or ethyl ester to avoid a negative charge that might

hinder cell uptake and lower their biological activity, the

complex [(η⁶-p-cym)RuCl(Etdap)]⁺spontaneously hydrolyzed

to 16 during the workup of the reaction. Regarding the osmium

complex [(η⁶-bip)OsCl(4-CO₂Et-pico)Cl)]⁺(18), a moderate

cytotoxic activity in the human ovarian A2780 and A2780cis

ruthenium after exposure to conjugate 6 (143.14

16.6

Chart 2. Structures of the Control Metal Complexes

1849

dx.doi.org/10.1021/bc300173h | Bioconjugate Chem. 2012, 23, 1838−1855

Bioconjugate Chemistry

Article

Figure 5. Chemical shift diﬀerences for NH, Hα, and Hβ protons between conjugates 3 (A), 4 (B), and 6 (C) and the dicarba analogue of octreotide

(2).²⁴

pmol Ru/10⁶cells) was higher than that of the parent complex

17 (68.01

2.2 pmol Ru/10⁶cells). Such a signiﬁcant

poly(ethylene glycol) linker might alter the active conformation

of the peptide and, in particular, the structure around D-Phe²

and Thr-ol¹⁵.

diﬀerence between these compounds reveals that conjugation

to the peptide moiety has a positive eﬀect on the internalization

of the ruthenium complex, thereby suggesting that the

attachment of the metal complex to the peptide does not

disturb its active conformation and hence its aﬃnity toward

somatostatin receptors is maintained (see below). Although the

covalent attachment of a potential metal-based drug to a

receptor-binding peptide such as octreotide is expected to

increase the bioavailability, delivery, and accumulation of the

complex to cancer cells, it must also be borne in mind that the

peptide moiety might also inhibit or weaken the interaction of

the metal fragment with the ﬁnal biological target (e.g., DNA,

RNA, or proteins), which would compromise its cytotoxic

activity.²⁵Release of the ruthenium fragment from the

conjugate might be necessary to allow it to reach its target

site. This might explain the apparently contradictory results

from cytotoxicity and cellular uptake studies, since in this case,

the covalent attachment of the ruthenium complex to the

dicarba analogue of octreotide 2 increases uptake and

accumulation of the drug in the cancer cell while the cytotoxic

activity of the ﬁnal conjugate is reduced.

Conformational Analysis by NMR Spectroscopy. It is

well-known from NMR studies in water and in DMSO-d₆

solution that octreotide (1) adopts a predominant type II′ β-

turn conformation across residues Phe⁷-D-Trp⁸-Lys⁹-Thr¹⁰.^64,65

The stability of this conformation together with the orientation

of exocyclic residues (D-Phe²and Thr-ol¹⁵) plays an important

role in the binding of octreotide and its analogues to

somatostatin receptors.^66,67Regarding the dicarba analogue of

octreotide used in this work (2), D′Addona et al. have recently

described that replacement of cystine by a CH₂−CH₂linkage

does not substantially alter the secondary structure in the active

region.²⁴These NMR studies also suggest that there exists a

conformational equilibrium between the antiparallel β-sheet

structural cluster and a second conformational ensemble

involving folded structures for the C-terminal residues. This

hypothesis is in good agreement with previously reported

ﬁndings for 1 that indicate that C-terminal amino acids fold into

a 3₁₀-helix-like array or in a similar helical ensemble.⁶⁶

Furthermore, it is important to consider that the structure of

2 in the region of the N-terminal D-Phe²and C-terminal

Thr(ol)¹⁵is signiﬁcantly diﬀerent from that previously reported

for 1,⁶⁶which might explain the diﬀerences in aﬃnity for

somatostatin receptors.²⁴Given the implication of these amino

acids both in the aﬃnity and in the speciﬁcity for somatostatin

receptors, we investigated whether the covalent attachment of

metal complexes to the N-terminal end of 2 through a

In order to gain some insight into peptide conformation,

conjugates 3, 4, and 6 were studied by 1D and 2D H NMR

1

spectroscopy. In all conjugates, signiﬁcant line-broadening was

observed at conjugate concentrations above 1 mM, suggesting

that oligomerization takes place under these conditions. This

eﬀect is specially pronounced in the case of the osmium−

octreotide conjugate (see the Supporting Information). To

avoid oligomerization, 2D NMR spectra were recorded at a

peptide concentration of 0.8 mM. Signals of the peptide were

completely assigned using standard techniques (see Tables S1−

3 in the Supporting Information). Resonances of the linker and

of ligands in the metal fragments were also indentiﬁed, although

not all of them could be unambiguously assigned to speciﬁc

protons.

A comparison between the chemical shifts of the peptide

fragment in the conjugates and the dicarba-analogue of

octreotide (2)²⁴indicates that residues D-Trp⁸-Lys⁹-Thr¹⁰-

DsaC¹⁴-Thr(ol)¹⁵have very similar chemical shifts, whereas

signiﬁcant diﬀerences are observed for protons of the amino

acids D-Phe²-DsaN³-Phe⁷(Figure 5). This suggests that the

peptide structure is only aﬀected in the latter region. The

chemical shift of Hα of D-Phe²is notably shifted downﬁeld.

This shift is attributed to the covalent attachment of the linker

to the α-amino group of this amino acid through an amide

bond. Interestingly, peaks for the NH protons of DsaN³and

Phe⁷amino acids in all conjugates (and to a lesser extent for the

Hβ signals) were shifted upﬁeld.

Molecular dynamics calculations were carried out to gain

further insight in the structural alterations induced in the

peptide structure by the covalent attachment of the metal

complex. The results are illustrated in Figure 6. The peptide

backbone is not signiﬁcantly altered with respect to the

structure of the dicarba analogue of octreotide (2). In

particular, the β-turn backbone geometry around residues D-

Trp⁸-Lys⁹-Thr¹⁰-DsaC¹⁴-Thr(ol)¹⁵is totally preserved. The

main changes occur in residues D-Phe²and DsaN³. Side

chain conformations are less deﬁned than in 2. Linker and

ligand fragments are mainly disordered. In some of the

calculated conformers, the ligand or the linker exhibits contacts

with aromatic residues D-Phe²and Phe⁷as well as with DsaN³

(see Figure 6). These residues correspond to those that have

1

changes in their H NMR chemical shifts with respect to the

unmodiﬁed peptide. This is also in good agreement with the

fact that some NOEs were observed between the linker and D-

Phe²(see Figure S-20 in the Supporting Information).

1850

dx.doi.org/10.1021/bc300173h | Bioconjugate Chem. 2012, 23, 1838−1855

Bioconjugate Chemistry

Article

relevant receptor for octreotide internalization.⁶⁸To this end,

we further analyzed the cytotoxic activity of conjugate 6 in

human prostate tumor DU-145 cells as well as and in the

Chinese hamster ovary (CHO) cell line, in order to assess the

selectivity of the ruthenium−octreotide conjugate for cancer

cells with respect to normal cells (Table 3). As shown in Figure

Table 3. IC₅₀Values of the Ruthenium−Octreotide

Conjugate 6 and Control Complex 17 in Human Cancer Cell

a

Lines (MCF-7 and DU-145) and in Normal Cells (CHO)

cell line, IC₅₀(μM)

compound

MCF-7

DU-145

CHO

6

63.00 1.53

3.32 0.54

26.00 2.00

6.80 0.99

45.17 2.61

5.8 0.01

17

a

IC₅₀: Inhibitory concentration that reduce by 50% the cell viability.

The cytotoxicity was determined by the MTT assay after 72 h of drug

exposure. Data represents the mean SE of at least three independent

experiments.

Figure 6. Model structures of the dicarba analogue of octreotide (2)

and of conjugates 3, 4, and 6 (see Chart 1). Left: representative

conformer. Right: ensemble with 10 conformers resulting from

molecular dynamics calculations. Superposition was done by ﬁtting

the backbone peptide heavy atoms.

The NMR data and molecular dynamics calculations suggest

that the covalent attachment of the metal complex does not

substantially alter the global structure of the peptide. However,

the aromatic ligands of the ruthenium and osmium complex in

conjugates 4 and 6, as well as the linker appear to have a

tendency to interact with the aromatic residues located to their

near side of the peptide sequence, which could be favored by

the high ﬂexibility of the poly(ethylene glycol) linker. This is

particularly important in the case of the Phe⁷residue, since this

amino acid is included in the so-called pharmacophore

sequence of octreotide. These results support the hypothesis

that the covalent attachment of both fragments might generate

a signiﬁcant population of peptide conformations with a

decreased aﬃnity for somatostatin receptors in comparison

with that of the dicarba analogue of octreotide 2. However, the

dicarba analogue of octreotide 2 still retains the ability to

internalize metal complexes upon conjugation, as demonstrated

cellular uptake experiments with 6 and 17.

Selectivity of the Ruthenium−Octreotide Conjugate

(6). The main objective of a targeted strategy is the selective

delivery of a cytotoxic drug to cancer cells in order to reduce

toxicity to normal cells. Consequently, the reduction of the

antiproliferative activity of a metal complex when conjugated

with octreotide (Table 2) could be compensated by a selective

and a more eﬃcient delivery of the compound to the targeted

cells. In order to corroborate this hypothesis, we determined

whether the conjugation of the ruthenium(II) arene complex

17 to the octreotide analogue 2 (conjugate 6) is correlated with

its vehiculization to tumor cells overexpressing sstr₂, the most

Figure 7. Cytotoxic eﬀect of conjugate 6 (A) and complex 17 (B) in

the MCF-7, DU-145, and CHO cell lines. Cells were treated for 72 h

with the indicated concentrations of each compound. The cell viability

was determined using the MTT assay. Each point in the graphs

represents the mean of three independent experiments SE.

7A, DU-145 cells were markedly more responsive to 6 than the

MCF-7 and CHO cells, as the IC₅₀value (26.00 3.46) was

2.4-fold lower than in MCF-7 cells and about 1.7-fold lower

than in CHO cells (IC₅₀: 45.17

2.61) (Table 3). This

diﬀerence in the cytotoxic activity of conjugate 6 between the

cell lines cannot be attributed to a diﬀerent activity of the metal

fragment in the diﬀerent cells, as the control complex 17

cytotoxicity was higher in MCF-7 (IC₅₀= 3.32

0.54) and

1851

dx.doi.org/10.1021/bc300173h | Bioconjugate Chem. 2012, 23, 1838−1855

Bioconjugate Chemistry

Article

Figure 8. Expression of the sst₂receptor on MCF-7, DU-145, and CHO cell lines. (A) Representative ﬂow cytometry histograms obtained after the

indirect immunoﬂuorescence staining of the cells. Solid lines represent the ﬂuorescence intensity of the cells after the incubation with an antibody

against sstr₂together with a control secondary antibody. Dotted lines indicate the background staining with the secondary antibody alone. (B) Each

column in the graphs represents the mean ﬂuorescence intensity of three independent experiments SE in MCF-7, DU-145, and CHO cells. *P <

0.05 vs MCF-7 cells.

more attenuated in DU-145 cells (IC₅₀= 6.80

CHO cells (5.8 0,01) (Figure 7B) (Table 3).

0.99) and

CONCLUSIONS

■

In summary, in this work we have described a straightforward

solid-phase methodology for the conjugation of a

dichloridoplatinum(II) complex as well as three “piano-stool”

organometallic ruthenium(II) and osmium(II) complexes to a

dicarba analogue of octreotide, a potent somatostatin agonist

whose receptors, particularly subtype 2, are overexpressed on

the membrane of tumor cells. At nuclear Cl⁻concentration (4

mM), platinum (3), osmium (4), and Ru(dap) (5) conjugates

are activated through hydrolysis of the M-Cl, which correlates

well with their DNA binding capacity, revealing in all cases a

strong preference for guanine nucleobases. However, neither

hydrolysis of the Ru−Cl nor interaction with DNA was

observed in the case of the triphenylphosphine-containing

ruthenium conjugate (6). Surprisingly, the latter conjugate

displays antiproliferative activity, but the rest of the conjugates

were noncytotoxic. Additional cytotoxic experiments with

control metal complexes and cellular uptake experiments by

ICP-MS, as well as conformational analysis of metal−octreotide

conjugates by 2D NMR spectroscopy and molecular dynamics

calculations, have allowed us to establish three general

conclusions. First, a signiﬁcant loss in the cytotoxic activity of

the metal complex moiety always occurs when conjugated to

the peptide sequence (e.g., IC₅₀= 3.32 0.54 in MCF-7 cells

and IC₅₀= 6.80 0.99 in DU-145 cells for complex 17 vs IC₅₀

= 63.00 1.53 in MCF-7 cells and IC₅₀= 26.00 2.00 in DU-

145 cells for conjugate 6); hence, it seems to be important to

use metal complexes with high cytotoxic activity in order to

develop peptide-conjugates with active biological properties.

Second, despite the presence of some altered amino acids in the

pharmacophore sequence of octreotide, particularly at the N-

terminal end of the peptide, the overall structure is essentially

maintained in the conjugate, which suggests that the aﬃnity for

somatostatin receptors would not be compromised upon

conjugation to a metal complex; this observation is in good

agreement with the higher accumulation of ruthenium in cancer

cells upon exposure to the ruthenium−octreotide conjugate 6

in comparison with that of the parent complex 17, which

In order to determine if the cytotoxic activity of 6 was related

to the receptor expression on the cell surface, we next

characterized the sstr₂levels by ﬂow cytometry. As shown in

Figure 8A, sstr₂expression was detected on the three cell lines.

Notably, sstr₂expression was signiﬁcantly higher in the DU-145

and CHO cells than in MCF-7 cells (mean cell ﬂuorescence

intensity 347.5

54.6, 344.0

50.56, and 206.7

14.27,

respectively) (Figure 8B), which demonstrates that the

cytotoxic activity of 6 is strongly correlated with sstr₂

expression levels, especially in the DU-145 and MCF7 cells.

However, although the CHO cells displayed similar sstr₂

expression levels to DU-145, the toxicity of 6 was attenuated

in these cells, suggesting some selective cytotoxic activity of the

ruthenium−octreotide conjugate toward cancer cells with

respect to normal cells.

As a very weak expression of sstr₂has been previously

reported in CHO cells,⁶⁹the sstr₂expression in this cell line

was additionally examined according to their mRNA expression

levels. PCR analyses conﬁrmed sstr₂gene expression in CHO

cells, with similar mRNA levels than in MCF-7 cells (see the

Supporting Information). Our results are in keeping with

diﬀerent studies describing that somatostatin receptors can also

be identiﬁed in a broad panel of normal tissues, as well as in the

human immune system.⁷⁰As a consequence, new clinical

applications for somatostatin analogues in several benign

pathological diseases are emerging in addition to their role in

malignant diseases.

All together, our results demonstrate a good correlation

between the antitumor activity of the ruthenium−octreotide

conjugate 6 and the expression of the sst₂receptor on the cell

surface and reveal the potential of conjugation with receptor-

binding peptides such as octreotide to deliver metal-based

anticancer drugs speciﬁcally to cells overexpressing somatosta-

tin subtype-2 receptor.

1852

dx.doi.org/10.1021/bc300173h | Bioconjugate Chem. 2012, 23, 1838−1855

Bioconjugate Chemistry

Article

demonstrates that the peptide has a positive eﬀect on the

internalization of the ruthenium complex. The choice of the

spacer is also important, not only to improve aqueous solubility

and to keep the metal complex away from the peptide, but also

to reduce conformational freedom between both fragments in

the conjugate. Third, the signiﬁcantly higher expression of the

sst₂receptor on DU-145 cells compared to MCF-7 cells is in

good agreement with IC₅₀values of conjugate 6 in both tumor

cell lines. The fact that control complex 17 shows an attenuated

cytotoxicity in DU-145 cells compared with MCF-7 cells

supports that the antitumoral activity of the ruthenium−

octreotide conjugate 6 is correlated with the overexpression of

the sst₂receptor on the cell surface. The cytotoxic activity of 6

in a nontumor cell line (CHO) was similar to that found in

cancer cells, a result which can be attributed to a similar level of

expression of somatostatin subtype-2 receptors.

The overall results of this work demonstrate the potential of

conjugation between metal-based anticancer drugs and

receptor-binding peptides such as octreotide to target tumoral

cells through binding to speciﬁc receptors, including

somatostatin subtype 2 receptor, overexpressed in their

membranes. In the near future, this receptor-targeted anticancer

strategy is expected to generate new antitumor drugs with

reduced side-eﬀects and reduced toxicity in normal cells. The

potential of this strategy will be increased by using novel

approaches to circumvent some of the limitations found in this

work, particularly new biological vectors in combination with

photoactivated metal complexes²⁵or cleavable linkers³²in the

biological media. Work is in progress in this direction to

generate more speciﬁc metal-based anticancer drugs.

complex 8, and Dr. Isolda Romero (University of Warwick)

́

́ ̀

and Dr. Antoni Padro (Centres Cientıﬁcs i Tecnologics,

Universitat de Barcelona) for helpful assistance in the

determination of ruthenium content by ICP-MS.

ABBREVIATIONS USED

■

Abu, aminoisobutyryl; ACN, acetonitrile; bip, biphenyl; cym,

cymene (1-methyl-4-(1-methylethyl)benzene); dap, 1-(carbox-

ylic acid)-1,2-diaminoethane; DCM, dichloromethane; dhDsaC,

dehydrodiaminosuberic acid C-terminus; dhDsaN, dehydrodia-

minosuberic acid N-terminus; DIPC, N,N′-diisopropylcarbodii-

mide; DIPEA, N,N-diisopropylethylamine; DMF, N,N-dime-

thylformamide; DMSO, dimethylsulfoxide; DTPA, diethylene-

t r i a m i n o - p e n t a a c e t i c a c i d ; D O T A , 1 , 4 , 7 , 1 0 -

tetraazacyclododecane-1,4,7,10-tetraacetic acid; DsaC, diamino-

suberic acid C-terminus; DsaN, diaminosuberic acid N-

terminus; ESI, electrospray ionization; Fluo, 5(6)-carboxyﬂuor-

escein; Fmoc, 9-ﬂuorenylmethyloxycarbonyl; Hag, ^L-2-allyl-

glycine; HATU, hexaﬂuorophosphate salt of the (2-(7-aza-1H-

benzotriazol-1-yl)-1,1,3,3-tetramethyluronium; HOAt, 1-hy-

droxy-7-azabenzotriazole; HOBt, 1-hydroxybenzotriazole; HR-

ESI, high-resolution electrospray ionization; ICP-MS, induc-

tively coupled plasma mass spectrometry; Im, imidazole; Ind,

indazole; MALDI-TOF, matrix-assisted laser desorption

ionization time-of-ﬂight; MS, mass spectrometry or mass

spectrum; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetra-

zolium bromide; NHS, N-hydroxysuccinimide; NOE, nuclear

Overhauser eﬀect; NOESY, nuclear Overhauser enhanced

spectroscopy; PBS, phosphate buﬀered saline; PEG, poly-

(ethylene glycol); pico, picolinate; sst, somatostatin; sstr,

somatostatin receptor; TIS, triisopropylsilane; TMP, trimethyl-

phosphite; TFA, triﬂuoroacetic acid

ASSOCIATED CONTENT

* Supporting Information

■

S

Selected MS spectra and HPLC traces of metal complex−

octreotide conjugates and of DNA−conjugate adducts. NMR

data of conjugates 3, 4, and 6, and of control metal complexes.

sstr2 mRNA expression in CHO and MCF-7 cells. This

material is available free of charge via the Internet at http://

pubs.acs.org.

REFERENCES

■

(1) Reubi, J. C. (2003) Peptide receptors as molecular targets for

cancer diagnosis and therapy. Endocr. Rev. 24, 389−427.

(2) Zaccaro, L., del Gatto, A., Pedone, C., and Saviano, M. (2009)

Peptides for tumour therapy and diagnosis: Current status and future

directions. Curr. Med. Chem. 16, 780−795.

(3) Okarvi, S. M. (2008) A comprehensive overview of peptide-based

radiopharmaceuticals and cytotoxic conjugates, as well as their

potential applications in cancer diagnosis and therapy. Cancer Treat.

Rev. 34, 13−26.

(4) Mezo, G., and Manea, M. (2010) Receptor-mediated tumor

targeting based on peptide hormones. Expert Opin. Drug Delivery 7,

79−96.

(5) Schaer, J. C., Waser, B., Mengod, G., and Reubi, J. C. (1997)

Somatostatin receptor subtype expression in human pituitary,

gastroentero-pancreatic and mammary tumors: comparison of

mRNA analysis with receptor autoradiography. Int. J. Cancer 70,

530−537.

(6) Janecka, A., Zubrzycka, M., and Janecki, T. (2001) Somatostatin

analogs. J. Pept. Res. 58, 91−107.

(7) Bauer, W., Briner, U., Dopfener, W., Haller, R., Huguenin, R.,

Marbach, P., Petcher, T. J., and Pless, J. (1982) SMS 201−995: a very

potent and selective analogue of somatostatin with prolonged action.

Life Sci. 31, 1133−1140.

(8) de Jong, M., Breeman, W. A. P., Kwekkeboom, D. J., Valkema, R.,

and Krenning, E. P. (2009) Tumor imaging and therapy using

radiolabeled somatostatin analogues. Acc. Chem. Res. 42, 873−880.

(9) Sun, L.-C., and Coy, D. H. (2008) Cytotoxic conjugates of

peptide hormones for cancer chemotherapy. Drugs Future 33, 217−

223.

AUTHOR INFORMATION

Corresponding Author

■

*Vicente Marchan

́

: Phone +34 934021249, Fax +34

933397878, E-mail vmarchan@ub.edu. Anna Massaguer:

Phone +34 972418370, Fax +34 972418150, E-mail anna.

massaguer@udg.edu.

Author Contributions

^#Both authors contributed equally to this work.

Notes

The authors declare no competing ﬁnancial interest.

ACKNOWLEDGMENTS

This work was supported by funding from the Ministerio de

■

Educacion

́

y Ciencia (CTQ2005-01834, CTQ2007-68014,

CTQ2008-02064, and CTQ2010-21567), the Generalitat de

Catalunya (2009SGR208 and Xarxa de Referencia de

Biotecnologia), Instituto de Salud Carlos III (grants RD06/

0020/0041), the Programa d’Intensiﬁcacio de la Recerca (UB)

̀

́

and the ERC (grant no 247450, to PJS). We thank Prof.

Alfonso Carotenuto (University of Napoli) for providing the

coordinates of the dicarba analogue of octreotide 2, Dr. Sabine

H. van Rijt (University of Warwick) for supplying the osmium

1853

dx.doi.org/10.1021/bc300173h | Bioconjugate Chem. 2012, 23, 1838−1855

Bioconjugate Chemistry

Article

(10) Wheate, N. J., Walker, S., Craig, G. E., and Oun, R. (2010) The

status of platinum anticancer drugs in the clinic and in clinical trials.

Dalton Trans. 39, 8113−8127.

(11) Jakupec, M. A., Galanski, M., Arion, V. B., Hartinger, C. G., and

Keppler, B. K. (2008) Antitumour metal compounds: more than

theme and variations. Dalton Trans., 183−194.

(12) van Rijt, S. H., and Sadler, P. J. (2009) Current applications and

future potential for bioinorganic chemistry in the development of

anticancer drugs. Drug Discovery Today 14, 1089−1097.

(13) Rademaker-Lakhai, J. M., Van den Bongard, D., Pluim, D.,

Beijnen, J. H., and Schellens, J. H. M. (2004) A phase I and

pharmacological study with imidazolium-trans-DMSO-imidazole-tetra-

chlororuthenate, a novel ruthenium anticancer agent. Clin. Cancer Res.

10, 3717−3727.

(14) Jakupec, M. A., Arion, V. B., Kapitza, S., Reisner, E., Eichinger,

A., Pongratz, M., Marian, B., Graf von Keyserlingk, N., and Keppler, B.

K. (2005) KP1019 (FFC14A) from bench to bedside: preclinical and

early clinical development - an overview. Int. J. Clin. Pharmacol. Ther.

43, 595−596.

(Poly)arginine: effect on cellular uptake, internalization, and

cytotoxicity. Bioconjugate Chem. 22, 218−226.

(31) Puckett, C. A., and Barton, J. K. (2010) Targeting a ruthenium

complex to the nucleus with short peptides. Bioorg. Med. Chem. 18,

3564−3659.

(32) Splith, K., Hu, W., Schatzschneider, U., Gust, R., Ott, I.,

Onambele, L. A., Prokop, A., and Neundorf, I. (2010) Protease-

activatable organometal-peptide bioconjugates with enhanced cytotox-

icity on cancer cells. Bioconjugate Chem. 21, 1288−1296.

(33) Mukhopadhyay, S., Barnes, C. M., Haskel, A., Short, S. M.,

Barnes, K. R., and Lippard, S. J. (2008) Conjugated platinum(IV)-

peptide complexes for targeting angiogenic tumor vasculature.

Bioconjugate Chem. 19, 39−49.

(34) Hsieh, H.-P., Wu, Y.-T., Chen, S.-T., and Wang, K.-T. (1999)

Direct solid-phase synthesis of octreotide conjugates: precursors for

use as tumor-targeted radiopharmaceuticals. Bioorg. Med. Chem. 7,

1797−1803.

(35) Moradell, S., Lorenzo, J., Rovira, A., Robillard, M. S., Aviles, F.

X., Moreno, V., de Llorens, R., Martinez, M. A., Reedijk, J., and Llobet,

A. (2003) Platinum complexes of diaminocarboxylic acids and their

ethyl ester derivatives: the effect of the chelate ring size on antitumor

activity and interactions with GMP and DNA. J. Inorg. Biochem. 96,

493−502.

(36) van Rijt, S. H., Peacock, A. F. A., Johnstone, R. D. L., Parsons, S.,

and Sadler, P. J. (2009) Organometallic osmium(II) arene anticancer

complexes containing picolinate derivatives. Inorg. Chem. 48, 1753−

1762.

(15) Peacock, A. F. A., and Sadler, P. J. (2008) Medicinal

organometallic chemistry: Designing metal arene complexes as

anticancer agents. Chem. Asian J. 3, 1890−1899.

(16) Levina, A., Mitra, A., and Lay, P. A. (2009) Recent

developments in ruthenium anticancer drugs. Metallomics 1, 458−470.

(17) Casini, A., Hartinger, C. G., Nazarov, A., and Dyson, P. J. (2010)

Organometallic antitumour agents with alternative modes of action.

Top. Organomet. Chem. 32, 57−80.

(18) Gasser, G., Ott, I., and Metzler-Nolte, N. (2011) Organometallic

anticancer compounds. J. Med. Chem. 54, 3−25.

(37) Hodson, E., and Simpson, S. J. (2004) Synthesis and

characterization of [(η6-cymene)Ru(L)X₂] compounds: single crystal

X-ray structure of [(η6-cymene)Ru(P{OPh}₃)Cl₂] at 203 K.

Polyhedron 23, 2695−2707.

(19) Suss-Fink, G. (2010) Arene ruthenium complexes as anticancer

̈

agents. Dalton Trans. 39, 1673−1688.

(20) Wang, F., Chen, H., Parsons, S., Oswald, I. D. H., Davidson, J.

E., and Sadler, P. J. (2003) Kinetics of aquation and anation of

ruthenium(II) arene anticancer complexes, acidity and X-ray structures

of aqua adducts. Chem.Eur. J. 9, 5810−5820.

(38) Chaplin, A. B., and Dyson, P. J. (2007) Catalytic activity of bis-

phosphine ruthenium(II)-arene compounds: Structure-activity corre-

lations. Organometallics 26, 2447−2455.

(39) Charvat, T. T., Lee, D. J., Robinson, W. E., and Chamberlin, A.

R. (2006) Design, synthesis, and biological evaluation of chicoric acid

analogs as inhibitors of HIV-1 integrase. Bioorg. Med. Chem. 14, 4552−

4567.

(40) Gariepy, J., Remy, S., Zhang, X., Ballinger, J. R., Bolewska-

Pedyczak, E., Rauth, M., and Bisland, K. (2002) A simple two-step

approach for introducing a protected diaminedithiol chelator during

solid-phase assembly of peptides. Bioconjugate Chem. 13, 679−684.

́

(41) Marchan, V., Moreno, V., Pedroso, E., and Grandas, A. (2001)

Towards a better understanding of the cisplatin mode of action.

(21) Kostrhunova, H., Florian, J., Novakova, O., Peacock, A. F. A.,

Sadler, P. J., and Brabec, V. (2008) DNA interactions of monofunc-

tional organometallic osmium(II) antitumor complexes in cell-free

media. J. Med. Chem. 51, 3635−3643.

(22) Pizarro, A. M., Habtemariam, A., and Sadler, P. J. (2010)

Activation mechanisms for organometallic anticancer complexes. Top.

Organomet. Chem. 32, 21−56.

́ ́

(23) Barragan, F., Moreno, V., and Marchan, V. (2009) Solid-phase

synthesis and DNA binding studies of dichloroplatinum(II) conjugates

of dicarba analogues of octreotide as new anticancer drugs. Chem.

Commun., 4705−4707.

Chem.Eur. J. 7, 808−815.

(42) Marchan, V., Pedroso, E., and Grandas, A. (2004) Insights into

́

(24) D’Addona, D., Carotenuto, A., Novellino, E., Piccand, V., Reubi,

J. C., Di Cianni, A., Gori, F., Papini, A. M., and Ginanneschi, M.

(2008) Novel sst5-selective somatostatin dicarba-analogs: Synthesis

and conformation-affinity relationships. J. Med. Chem. 51, 512−520.

the reaction of transplatin with DNA and proteins: Methionine-

mediated formation of histidine-guanine trans-Pt(NH₃)₂cross-links.

Chem.Eur. J. 10, 5369−5375.

́ ́ ́

B., Lopez de la Osa, J., Gonzalez, C., Pedroso, E.,

(43) Alguero,

Marchan

(25) Barragan

́

, F., Lop

́

ez-Senín, P., Salassa, S., Betanzos-Lara, S.,

, V. (2011)

́

, V., and Grandas, A. (2006) Selective platination of modified

Habtemariam, A., Moreno, V., Sadler, P. J., and Marchan

́

oligonucleotides and duplex cross-links. Angew. Chem., Int. Ed. 45,

8194−8197.

(44) Goddard, D. T., and Kneller, G. SPARKY, University of

California, San Francisco.

(45) Singh, S. K., Trivedi, M., Chandra, M., Sahay, A. N., and Pandey,

D. S. (2004) Luminescent piano-stool complexes incorporating 1-(4-

cyanophenyl)imidazole: Synthesis, spectral, and sructural studies.

Inorg. Chem. 43, 8600−8608.

(46) Koradi, R., Billeter, M., and Wuthrich, K. (1996) MOLMOL: a

program for display and analysis of macromolecular structures. J. Mol.

Graphics 14, 29−32.

(47) Peacock, A. F. A., Parsons, S., and Sadler, P. J. (2007) Tuning

the hydrolytic aqueous chemistry of osmium arene complexes with

N,O-chelating ligands to achieve cancer cell cytotoxicity. J. Am. Chem.

Soc. 129, 3348−3357.

(48) van Rijt, S. H., Mukherjee, A., Pizarro, A. M., and Sadler, P. J.

(2010) Cytotoxicity, hydrophobicity, uptake, and distribution of

Photocontrolled DNA binding of a receptor-targeted organometallic

ruthenium(II) complex. J. Am. Chem. Soc. 133, 14098−14108.

(26) Dirscherl, G., and Konig, B. (2008) The use of solid-phase

̈

synthesis techniques for the preparation of peptide-metal complex

conjugates. Eur. J. Org. Chem., 597−634.

(27) Metzler-Nolte, N. (2010) Biomedical applications of organo-

metal-peptide conjugates. Top. Organomet. Chem. 32, 195−217.

(28) Peindy N’Dongo, H. W., Ott, I., Gust, R., and Schatzschneider,

U. (2009) Microwave-assisted solid-phase synthesis, cellular uptake,

and cytotoxicity studies of cymantrene-peptide bioconjugates. J.

Organomet. Chem. 694, 823−827.

(29) Gross, A., and Metzler-Nolte, N. (2009) Synthesis and

characterization of a ruthenocenoyl bioconjugate with the cyclic

octapeptide octreotate. J. Organomet. Chem. 694, 1185−1188.

(30) van Rijt, S. H., Kostrhunova, H., Brabec, V., and Sadler, P. J.

(2011) Functionalization of osmium arene anticancer complexes with

1854

dx.doi.org/10.1021/bc300173h | Bioconjugate Chem. 2012, 23, 1838−1855

Bioconjugate Chemistry

Article

osmium(II) anticancer complexes in ovarian cancer cells. J. Med. Chem.

53, 840−849.

(66) Melacini, G., Zhu, Q., and Goodman, M. (1997) Multi-

conformational NMR Analysis of sandostatin (octreotide): equilibrium

between β-sheet and helical structures. Biochemistry 36, 1233−1241.

(67) Deshmukh, M. V., Voll, G., Kuehlewein, A., Maecke, H.,

Schmitt, J., Kessler, H., and Gemmecker, G. (2005) NMR studies

reveal structural differences between the Gallium and Ytrium

complexes of DOTA-D-Phe1-Tyr3-octreotide. J. Med. Chem. 48,

1506−1514.

(49) Morris, R. E., Aird, R. E., Murdoch, P., del, S., Chen, H.,

Cummings, J., Hughes, N. D., Parsons, S., Parkin, A., Boyd, G., Jodrell,

D. I., and Sadler, P. J. (2001) Inhibition of cancer cell growth by

ruthenium(II) arene complexes. J. Med. Chem. 44, 3616−3621.

(50) Moreno, V., Font-Bardia, M., Calvet, T., Lorenzo, J., Aviles, F.

X., Garcia, M. H., Morais, T. S., Valente, A., and Robalo, M. P. (2011)

DNA interaction and cytotoxicity studies of new ruthenium(II)

cyclopentadienyl derivative complexes containing heteroaromatic

ligands. J. Inorg. Biochem. 105, 241−249.

(68) Reubi, J. C., Schar, J. C., Waser, B., Wenger, S., Heppeter, A.,

Schmitt, J. S., and Macke, H. R. (2000) Affinity profiles for human

̈

somatostatin receptor subtypes sstr1-sstr5 of somatostatin radiotracers

selected for scintigraphic and radiotherapeutic use. Eur. J. Nucl. Med.

27, 273−282.

(51) Gianferrara, T., Bratsos, I., and Alessio, E. (2009) A

categorization of metal anticancer compounds based on their mode

of action. Dalton Trans. 37, 7588−7598.

(69) Zhang, J., Jin, W., Wang, X., Wang, J., Zhang, X., and Zhang, Q.

(2010) A novel octreotide modified lipid vesicle improved the

anticancer efficacy of doxorubicin in somatostatin receptor 2 positive

tumor models. Mol. Pharm. 7, 1159−1168.

(52) Jennerwein, M., and Andrews, P. A. (1995) Effect of intracellular

chloride on the cellular pharmacodynamics of cis-

diamminedichloroplatinum(II). Drug Metab. Dispos. 23, 178−184.

(53) Wang, F., Habtemariam, A., van der Geer, E. P. L., Fernandez,

R., Melchart, M., Deeth, R. J., Aird, R., Guichard, S., Fabbiani, F. P. A.,

Lozano-Casal, P., Oswald, I. D. H., Jodrell, D. I., Parsons, S., and

Sadler, P. J. (2005) Controlling ligand substitution reactions of

organometallic complexes: Tuning cancer cell cytotoxicity. Proc. Natl.

Acad. Sci. U.S.A. 102, 18269−18274.

(70) Cascini, G. L., Cuccurullo, V., and Mansi, L. (2010) The non

tumour uptake of (111)In-octreotide creates new clinical indications in

benign diseases, but also in oncology. Q. J. Nucl. Med. Mol. Imaging 54,

24−36.

(54) Peacock, A. F. A., Habtemariam, A., Moggach, S. A., Prescimone,

A., Parsons, S., and Sadler, P. J. (2007) Chloro half-sandwich

osmium(II) complexes: influence of chelated N,N-ligands on

hydrolysis, guanine binding, and cytotoxicity. Inorg. Chem. 46,

4049−4059.

(55) Novakova, O., Chen, H., Vrana, O., Rodger, A., Sadler, P. J., and

Brabec, V. (2003) DNA Interactions of monofunctional organo-

metallic ruthenium(II) antitumor complexes in cell-free media.

Biochemistry 42, 11544−11554.

(56) Liu, H.-K., Wang, F., Parkinson, J. A., Bella, J., and Sadler, P. J.

(2006) Ruthenation of duplex and single-stranded d(CGGCCG) by

organometallic anticancer complexes. Chem.Eur. J. 12, 6151−6165.

(57) Gossens, C., Tavernelli, I., and Rothlisberger, U. (2009) Binding

of organometallic ruthenium(II) anticancer compounds to nucleo-

bases: A computational study. J. Phys. Chem. A 113, 11888−11897.

(58) Kostrhunova, H., Florian, J., Novakova, O., Peacock, A. F. A.,

Sadler, P. J., and Brabec, V. (2008) DNA Interactions of monofunc-

tional organometallic osmium(II) antitumor complexes in cell-free

media. J. Med. Chem. 51, 3635−3643.

(59) Fischer, R., Mader, O., Jung, G., and Brock, R. (2003) Extending

the applicability of carboxyfluorescein in solid-phase synthesis.

Bioconjugate Chem. 14, 653−660.

(60) Watson, J. C., Balster, D. A., Gebhardt, B. M., O’Dorisio, T. M.,

O’Dorisio, M. S., Espenan, G. D., Drouant, G. J., and Woltering, E. A.

(2001) Growing vascular endothelial cells express somatostatin

subtype 2 receptors. Br. J. Cancer 85, 266−272.

(61) Hornick, C. A., Anthony, C. T., Hughey, S., Gebhardt, B. M.,

Espenan, G. D., and Woltering, E. A. (2000) Progressive nuclear

translocation of somatostatin analogs. J. Nucl. Med. 41, 1256−1263.

(62) Zhang, Y., Wang, X., Wang, J., Zhang, X., and Zhang, Q. (2011)

Octreotide-modified polymeric micelles as potential carriers for

targeted docetaxel delivery to somatostatin receptor overexpressing

tumor cells. Pharm. Res. 28, 1167−1178.

(63) Huang, C. M., Wu, Y. T., and Chen, S. T. (2000) Targeting

delivery of paclitaxel into tumor cells via somatostatin receptor

endocytosis. Chem. Biol. 7, 453−461.

(64) Wynants, C., Van Binst, G., and Loosli, H. R. (1985) SMS 201−

995, a very potent analogue of somatostatin. Assignment of the 1H

500 MHz n.m.r. spectra and conformational analysis in aqueous

solution. Int. J. Pept. Protein Res. 25, 608−614.

(65) Wynants, C., Van Binst, G., and Loosli, H. R. (1985) SMS 201−

995, an octapeptide somatostatin analogue. Assignment of the 1H 500

MHz n.m.r. spectra and conformational analysis of SMS 201−995 in

dimethylsulfoxide. Int. J. Pept. Protein Res. 25, 615−621.

1855

dx.doi.org/10.1021/bc300173h | Bioconjugate Chem. 2012, 23, 1838−1855

Article Doi

Somatostatin subtype-2 receptor-targeted metal-based anticancer complexes

DOI: 10.1021/bc300173h

Source and publish data:

Authors:

Article abstract of DOI:10.1021/bc300173h

Full text of DOI:10.1021/bc300173h

Products guided by the article

R&D Labs maybe for 1394898-48-6

Relevant to this article

Hot Product