← Back
Cytotoxic half-sandwich rhodium(III) complexes: Polypyridyl ligand influence on their DNA binding properties and cellular uptake
JournalofOrganometallicChemistry693(2008)2299–2309
ContentslistsavailableatScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
Cytotoxic half-sandwich rhodium(III) complexes: Polypyridyl ligand influence
on their DNA binding properties and cellular uptake
Michael A. Scharwitza, Ingo Ottb, Yvonne Geldmachera, Ronald Gustb, William S. Sheldricka,*
aLehrstuhlfürAnalytischeChemie,Ruhr-UniversitätBochum,Universitätsstrasse150,D-44780Bochum,Germany
bInstitutfürPharmazie,FreieUniversitätBerlin,Königin-Luise-Straße2-4,D-14195Berlin,Germany
a r t i c l e i n f o a b s t r a c t
Articlehistory: The DNA binding of polypyridyl (pp) (g5-C Me )RhIII complexes of the types [(g5-C Me )RhCl(pp)]-
5 5 5 5
Received12February2008 (CF SO ) (2–6) (pp=bpy, phen, dpq, dppz, dppn), [(g5-C Me )Rh{(Me N) CS}(pp)](CF SO ) (7–9)
3 3 5 5 2 2 3 32
Receivedinrevisedform1April2008 (pp=dpq,dppz,dppn)and[(g5-C Me )Rh(L)(pp)](CF SO )(10)(L=C H S(cid:2))and(11)(L=C H S(cid:2))has
5 5 3 3 6 5 10 7
Accepted2April2008
been studied by UV/Vis spectroscopy, circular dichroismus and viscosity measurements. Complexes
Availableonline6April2008
3–11 are cytotoxic towards the human MCF-7 breast and HT-29 colon cancer cell lines and exhibit
IC values in the range 0.56–10.7lM. Stable intercalative binding into CT-DNA is indicated for the
50
Keywords:
Bioorganometallicchemistry
dpqanddppzcomplexesbylargeincreasesDT
m
of6–12(cid:2)CintheDNAthermaldenaturationtemperature
forr=[complex]/[DNA]=0.1andbyinducedCDbandsandlargeviscosityincreases.Incontrast,signif-
Rhodium
icantDNAlengtheningisnotobservedafterincubationofthebiopolymerwiththedppncomplexes2and
Polypyridylligands
DNAbinding
9atmolarratiosofr<0.08.Pronouncedhypochromicshiftsforthep–p*transitionsofthedppnligandsin
Cytotoxicity therange320–425nmindicatethepossiblepresenceofsurfacestacking.Theinvitrocytotoxicitiesofthe
Cellularuptake chlorocomplexes4–6andthe(Me 2 N) 2 CScomplexes7–9aredependentonthesizeofthepolypyridyl
ligandwithIC valuesdecreasingintheorderdpq>dppz>dppn.Forinstance,IC valuesof5.3,1.5
50 50
and0.91lMweredeterminedfor7–9againstMCF-7cells.RapidCl(cid:2)/H Oexchangeleadstheformation
2
ofaquadicationsfor4–6,whoselevelsofcellularuptakeandcytotoxicitiesaresimilartothoseestab-
lishedfor7–9. Intramolecularinteractionsbetweenthearomatic thiolateanddppzligandsof10and
11preventsignificantDNAintercalation.X-raystructuraldeterminationshavebeenperformedfor2–7
and11.
(cid:3)2008ElsevierB.V.Allrightsreserved.
1.Introduction whereas those with polar substituents on the arene or with aro-
matic diimine ligands such as 2,20-bipyridine (bpy) or 1,10-
Organometallic compounds containing the Group 8 elements phenanthroline(phen),exhibiteitherpoorornoactivity[2].Further
iron and ruthenium are attracting considerable current interest increasesinthesizeofthepolypyridylligand(pp)lead,however,
aspotentialanticanceragents[1–3].Forinstance,ferrocifen,aferr- toadramaticreversalofthelattertrendandtheinvitrocytotox-
ocenyl derivative of tamoxifen, exhibits promising antitumour icities of the complexes [(g6-C Me )RuCl(pp)](CF SO ) towards
6 6 3 3
activityandisapossiblecandidateforclinicaltrials[4].Half-sand- the human cell lines HT-29 (colon cancer) and MCF-7 (breast
wich(g6-arene)RuIIcomplexeswithimidazole[5],sulfoxide[6–8], cancer)arestronglydependentonthesurfaceareaofthearomatic
phosphane[3,9],chelatingaminoacidato[10],anddiamineordii- system[15].Forinstance,theIC valuesdecreasesfrom11.1over
50
mineligands[2,11]havealsobeenevaluatedforcytotoxicactivity. 2.12to0.13lMforMCF-7cellsasthesizeofthepolypyridylligand
Such studies have recently been extended to analogous (g6-are- increases in the order dpq<dppz<dppn (dpq=dipyrido-
ne)OsIIcompounds[12–14]. [3,2-f:20,30-h]quinoxaline; dppz=dipyrido[3,2-a:20,30-c]phenazine;
Both the size of the arene and the lability of the Ru–Cl bond dppn=benzo[i]dipyrido[3,2-a:20,30-c]phenazine). These values
have found to play a crucial role in determining the cytotoxicity correlatewellwiththecellularuptakeefficiency,whichincreases
of ruthenium(II) complexes of the type [(g6-arene)RuCl(LL0)](PF ) from1.1over146.6to906.7ng(Ru)/mg(protein)withintheseries
6
with bidentate ligands LL0. Compounds with extended polycyclic [15]. The kinetically inert complexes [(g6-C Me )Ru{(NH ) CS}-
6 6 22
arenes(e.g.tetrahydroanthracene)andLL0=ethylenediamine(en) (pp)](CF SO ) (pp=dppz, dppn) are also cytotoxic and this sug-
3 32
are most active towards A2780 human ovarian cancer cells, geststhatspecificpropertiesofthelargepolypyridylligands(e.g.
DNAintercalation[16,17]and/orcleavage[18])mayberesponsi-
blefortheirbiologicalactivity.DNAbindingstudiesindicatethat
* Correspondingauthor.Tel.:+492343224192;fax:+492343214420.
E-mailaddress:william.sheldrick@rub.de(W.S.Sheldrick). (g6-C 6 Me 6 )RuII compounds containing dpq or particularly dppz
0022-328X/$-seefrontmatter(cid:3)2008ElsevierB.V.Allrightsreserved.
doi:10.1016/j.jorganchem.2008.04.002
2300 M.A.Scharwitzetal./JournalofOrganometallicChemistry693(2008)2299–2309
ligands are good metallointercalators but that the dppn ligand is 2+(CFSO)2
too large to support stable intercalation between the base pairs 3 3
ofthedoublehelix[15].
N N
InstrikingcontrasttotheirGroup8neighbours,half-sandwich
Rh
complexesoftheGroup9transitionmetalsrhodium(III)andirid-
ium(III)haveattractedlittleinterestaspotentialanticanceragents. S N N
An in vitro evaluation of the RAPTA analogues [(g6-p-cyme- Me 2 N
ne)RhCl 2 (pta)], [(g5-C 5 Me 5 )RhCl 2 (pta)] and [(g5-C 5 Me 5 )RhCl- NMe 2
(pta) ]Cl (pta=1,3,5-triaza-7-phosphatricyclo[3.3.1.1]decane)
2 pp = dpq 7
towards HT-29, A549 (lung carcinoma) and T47D (breast carci-
dppz 8
noma)cellshasrecentlybeenappeared[14],butthereportedhigh
IC values(P380lM)areindicativeofverylimitedcytotoxicity. dppn 9
50
In contrast, much lower IC values of, respectively, 7.4 and
50 Scheme2. Structuresofcomplexes7–9.
0.41lM have recently reported [15] for the [(g5-C Me )IrIII com-
5 5
plexes[(g5-C Me )IrCl(dppz)](CF SO )and[(g5-C Me )Ir{(Me N) -
5 5 3 3 5 5 2 2
CS}(dppn)](CF SO ) [19] towards HT-29 cells. Interestingly, the
3 32 +(CFSO)
increased lability of the Ir–Cl bond in the former complex leads 3 3
to thermodynamically favoured coordinative Ir–N (nucleobase)
N N
bindingtoDNAratherthanintercalation,asobservedfortheanal-
ogous (g6-C Me )RuII complex [15]. We now report DNA binding Rh
6 6
studiesandevaluationsoftheinvitrocytotoxicityofpolypyridyl S N N
half-sandwichRhIIIcomplexesofthetypes[(g5-C Me )RhCl(pp)]-
5 5
(CF SO ) (Scheme 1), [(g5-C Me )Rh{(Me N) CS}(pp)](CF SO )
3 3 5 5 2 2 3 32
(pp=dpq, dppz, dppn) (Scheme 2) and [(g5-C Me )Rh(L)(dppz)]- 10
(CF SO )(L=C H S(cid:2),C H S(cid:2))(Scheme3).Two 5 fact 5 orsmaybeex- 11
3 3 6 5 10 7
pected to possibly modify the biological properties of such RhIII Scheme3. Structuresofcomplexes10and11.
compounds with respect to analogous half-sandwich IrIII or RuII
complexes:(a)theincreaseintherateofchloridesubstitutionin
the order (g6-C Me )RuII(cid:3)(g5-C Me )IrIII<(g5-C Me )RhIII [20] [(g5-C Me )Rh(acetone)(pp)]2+afforded[(g5-C Me )Rh{(Me N) CS}-
6 6 5 5 5 5 5 5 5 5 2 2
and(b)thepronouncedsoftnessofthethird-rowtransitionmetal (pp)](CF SO ) (7–9) (pp=dpq, dppz, dppn), after heating the
3 32
IrIIIincomparisontoRuIIorRhIII,owingtorelativisticdestabilisa- reactionmixturefor2hat75(cid:2)CinCH OH/CH Cl .Thedppzcom-
3 2 2
tionofits5dshell[21]. plexes[(g5-C Me )Rh(L)(dppz)](CF SO )(10)(L=C H S(cid:2))and(11)
5 5 3 3 6 5
(L=C H S(cid:2))werepreparedinananalogousmannerbyreactionof
10 7
2.Resultsanddiscussion [(g5-C 5 Me 5 )Rh(acetone)(dppz)]2+ with solutions of the respective
thiols C H SH and C H SH in the presence of an equivalent of
6 5 10 7
2.1.Synthesisof1–11 30% NaOH/CH 3 OH. All the complexes were characterised by 1H
and13CNMRandpositve-ionLSIMSandgavesatisfactorymicroa-
The compounds of the type [(g5-C Me )RhCl(en)](CF SO ) (1) nalyses.Themolecularstructuresof2–7and11weredetermined
5 5 3 3
and [(g5-C Me )RhCl(pp)](CF SO ) (2–6) (pp=bpy, phen, dpq, byX-raystructuralanalysis(Table1).
5 5 3 3
dppz, dppn) were synthesised by heating the solvent complex Complexes 1–3 (pp=en, bpy, phen) were synthesised for the
[(g5-C Me )RhCl(acetone) ](CF SO )withtheappropriatepolypyr- purposeofevaluatingtheirinvitrocytotoxicitiestowardsthehu-
5 5 2 3 3
idyl ligand (pp) at 75(cid:2)C in CH OH/CH Cl for 2h. [(g5-C Me )- man cell lines MCF-7 and HT-29. Their cations have been
3 2 2 5 5
RhCl(acetone) ]+waspreparedinsitubyadditionoftwoequivalents previously characterised in the presence of other counter anions
2
of Ag(CF 3 SO 3 ) to a solution of the dinuclear starting compound (1, Cl(cid:2) [22]; 2, Cl(cid:2) [23–25], ClO 4 (cid:2) [25]; 3, Cl(cid:2) [23–25], ClO 4 (cid:2)
[{(g5-C 5 Me 5 )RhCl} 2 (l-Cl) 2 ]inacetoneandsubsequentfiltrationof [25]) as has that of the dppz complex 5 (PF 6 (cid:2) [26]), which itself
precipitatedAgClafterstirringinthedarkfor0.5h.Followingre- is known [27]. Crystal structures have been reported for [(g5-
movaloftheremainingchloridebyanalogoustreatmentofcom- C 5 Me 5 )RhCl(bpy)](ClO 4 ) [20] and [(g5-C 5 Me 5 )RhCl(phen)](ClO 4 )
plexes4–6withanequivalentofAg(CF SO )in acetone,addition [25].Thebondlengthsandanglesinthecationsof2and3aresim-
3 3
oftetramethylthiourea(Me N) CStotheresultinginsitucomplex ilar to those determined for the same cations in the presence of
2 2
perchlorate anions. For instance, complex 3 (pp=phen) exhibits
Rh–N distances of 2.100 (2) and 2.121 (2)Å, Rh–C distances in
therange2.141(3)–2.171(3)andanRh–Clbondlengthof2.406
+(CFSO)
3 3 (1)Å. The analogous distances in [(g5-C Me )RhCl(phen)](ClO )
5 5 4
[25]are 2.109(3)Å and 2.128(3), 2.132 (5)–2.165(4) and 2.386
N N (1)Å.Similarbondlengthsareobservedincomplexes4–6,whose
Rh cations are depicted in Fig. 1. These exhibit C s crystallographic
Cl N N symmetry in the cases of 4 and 6. An interesting feature of the
half-sandwichcationsisthevariabilityoftheinterplanaranglebe-
tween their cyclopentadienyl and polypyridyl ring systems.
pp = bpy 2 Respective values of 76.6(2)(cid:2), 58.5(3)(cid:2) and 71.6(5)(cid:2) are found in
phen 3 complexes 4–6. The significantly larger interplanar angles for 4
dpq 4 and6resultfromtheadoptionofapronouncedenvelopeconfor-
dppz 5 mationbythecentralfive-memberedNRhNCCchelateringinthese
complexes(Fig.1aandc).Ameasureofthisdistortionisgivenby
dppn 6
therespectiveinterplanaranglesof14.7(4)(cid:2)in4and9.6(2)(cid:2) in6
Scheme1. Structuresofcomplexes2–6. between the N1–Rh1–N1A and N1–C–C–N1A planes. In contrast,
M.A.Scharwitzetal./JournalofOrganometallicChemistry693(2008)2299–2309 2301
Table1
Crystalandrefinementdatafor2–6,7and11
Compound 2 3 4 5 6 7(cid:4)H2O 11(cid:4)2H2O
M(gmol(cid:2)1) 578.8 602.9 654.9 705.0 755.0 918.8 864.8
T(K) 294 294 113 163 103 294 100
Radiation MoKa MoKa MoKa MoKa CuKa MoKa MoKa
Crystalsystem Orthorhombic Triclinic Monoclinic Triclinic Monoclinic Monoclinic Monoclinic
spacegroup Pmn21 P((cid:2)1) P21/m P((cid:2)1) P21/m P21/n C2/c
a(Å) 13.103(2) 7.963(1) 7.792(1) 8.166(2) 12.198(1) 8.787(2) 36.700(9)
b(Å) 8.461(1) 11.761(1) 12.166(1) 11.848(4) 12.396(1) 37.540(8) 8.787(2)
c(Å) 10.647(6) 13.099(2) 13.849(1) 15.698(4) 12.245(1) 12.292(3) 23.019(6)
a((cid:2)) 90 90.12(1) 90 73.42(3) 90 90 90
b((cid:2)) 90 99.56(2) 103.42(1) 76.09(2) 110.33(1) 97.63(3) 91.28(2)
c((cid:2)) 90 98.26(2) 90 74.11(2) 90 90 90
V(Å3) 1180.4(7) 1196.8(3) 1277.0(1) 1378.1(6) 1736.2(3) 4018.9(14) 7422(3)
Z 2 2 2 2 2 4 8
D(gcm(cid:2)3) 1.629 1.673 1.703 1.699 1.444 1.518 1.548
F(000) 584 608 660 712 764 1872 3536
l(mm(cid:2)1) 0.974 0.964 0.913 0.853 5.713 0.659 0.638
2hmax((cid:2)) 50 55 55 50.5 117.9 50 51.8
Collectedreflections 1561 6626 8234 11544 6631 8822 33444
Independentreflections 1258 5467 3002 4892 2372 7081 7123
Refinedparameters 160 313 211 384 245 426 500
R1[I>2r(I)] 0.048 0.032 0.047 0.045 0.056 0.122 0.054
wR2(alldata)a 0.130 0.082 0.110 0.091 0.157 0.323 0.134
S(Goodness-of-fit) 1.117 1.042 1.058 0.648 1.057 0.957 0.845
Maximum/minimumDp(eÅ(cid:2)3) 1.28/(cid:2)0.54 0.49/(cid:2)0.45 1.51/(cid:2)0.54 0.50/(cid:2)0.46 0.72/(cid:2)0.65 0.97/(cid:2)0.94 1.84/(cid:2)0.80
a wR2=[Pw(F2 o(cid:2)F2 c)2/Pw(F2 o)2]1/2.
Fig.1. Molecularstructuresofthecationsof:(a)[(g5-C5Me5)RhCl(dpq)](CF3SO3)(4),(b)[(g5-C5Me5)RhCl(dppz)](CF3SO3)(5)and(c)[(g5-C5Me5)RhCl(dppn)](CF3SO3)(6).
2302 M.A.Scharwitzetal./JournalofOrganometallicChemistry693(2008)2299–2309
thechelateringinthedppzcomplex5isalmostflatwithananal- 30(cid:2)C.Itcan,therefore,beassumedthatcompounds1–3willrap-
ogousinterplanarangleofonly1.7(5)(cid:2). idly bind to CT DNA through coordinative Rh–N (nucleobase)
bonds. The low DT values of, respectively, +2 and +1(cid:2)C for the
m
2.2.DNAbindingstudies monocations of the kinetically inert dppz complexes 10 and 11
indicate that their aromatic thiolate ligands must impair stable
2.2.1.UV/Visabsorptionandmeltingtemperaturemeasurements DNAintercalation.
ThermaldenaturationmeasurementsforcalfthymusDNA(CT Following incubation periods of 5–30min, UV/Vis spectra of
DNA)inthepresenceoftransitionmetalcomplexessuchas1–11 buffered solutions of complexes 1–11 (10mM phosphate buffer,
canprovideasimplemeansofgaugingthestrengthofpolypyridyl pH7.2)withCTDNAat25(cid:2)Candr=0.1exhibitnofurthersignif-
intercalation, provided that electrostatic and hydrogen bonding icantchanges,therebyindicatingthatachievementofequilibrium
interactions remain effectively unchanged by pp ligand variation conditions must be relatively rapid. Pronounced decreases in
[27–29]. The DT values listed in Table 2 were recorded under absorbancefor complexes 5 and 8 at about 364 and 384nm and
m
equilibrium conditions in a 10mM phosphate buffer at pH 7.2 shiftsoftheseabsorptionmaximatohigherwavelengthsareindic-
for1:10complex/CTDNAmixtures[DNAconcentration=M(nucle- ative of dppz intercalation into the DNA double helix. The hypo-
otide)]andareindicativeofstrongDNAintercalationforthecat- chromic shift DA/A of (cid:2)47% at 364nm with its associated red
ions of the dppz complexes 5 and 8. Their DT values of +12(cid:2)C shift of 5nm is depicted for complex 5 in Fig. 2. Similar hypo-
m
are even higher than those of +7 and +8(cid:2)C recorded for [(g5- and bathochromic shifts of respectively DA/A of (cid:2)41% and
C Me )Rh(HcysOMe)(dppz)]2+ (HcysOMe=cysteine methyl ester) Dk=3nmwerealsoobservedfortheanalogousp–p*transitionof
5 5
and [(g5-C
5
Me
5
)Rh(H
2
metOMe)(dppz)]3+ (H
2
metOMe=L-methio- complex8.Thespectralchangesobservedfortheinteractionof5
ninemethylester),whichexhibitstrongside-onintercalationwith and8withCTDNAarealsosimilartothosereportedfortheanal-
binding constants of, respectively, 2.8(cid:5)105 and 1.3(cid:5)106M(cid:2)1 ogous (g6-C Me )RuII complexes [(g6-C Me )RuCl(dppz)]+ and
6 6 6 6
[27].RelativelyhighDT valuesinthepositiverange6–8(cid:2)Careob- [(g6-C Me )Ru{(Me N) CS) }(dppz)]2+ [15] and the kinetically in-
m 6 6 2 2 2
servedforthedpqcomplexes4and7andthedppncomplexes6 ert (g5-C Me )IrIII complex [(g5-C Me )Ir{(Me N) CS) }(dppz)]2+
5 5 5 5 2 2 2
and9andareindicativeofpossibleintercalationorsurfacestack- [19].Instrikingcontrast,[(g5-C Me )IrCl(dppz)]+alsoexhibitsini-
5 5
ingofthepolypyridylligands.Althoughthehigherformalcharge tial large negative hypochromic shifts DA/A at 364 and 383nm
(n=2) of the kinetically inert (Me N) CS-containing dications of after 1min of mixing with CT DNA, but these are followed by a
2 2
7–9 would be expected to lead to additional stabilization of the steadyincreaseinabsorbanceatbothwavelengths.Astateofequi-
DNA helixthrough stronger electrostatic cation–phosphate inter- librium is achieved after 10min, in which the final spectrum
actions, no significant increases are apparent for the DT values exhibits small absorbance increases for the p–p* transitions in
m
ofthesecomplexes,incomparisontothecationsof4–6withthe accordancewithIr–N(nucleobase)bindingandnegligibleinterca-
same pp ligand. This finding suggests that rapid aquation [20] of lationorsurfacestacking.
the original monocations of 4–6 will afford dications of the type Kineticstudieshaveestablishedthattherateconstantsforsub-
[(g5-C Me )Rh(H O)(pp)]2+asequilibriumspeciesforDNAinterac- stitution of (g5-C Me )RhIII complexes are much higher than for
5 5 2 5 5
tion.Instrikingcontrasttothe(g5-C Me )RhIIIcomplex4,theDT analogous(g5-C Me )IrIIIor(g6-arene)RuIIspecies[20].Valuesof,
5 5 m 5 5
value of only +2(cid:2)C for the analogous (g5-C Me )IrIII species is respectively, 1.59(cid:5)103s(cid:2)1, 2.19(cid:5)102s(cid:2)1 and 10.2(cid:5)10(cid:2)2s(cid:2)1
5 5
clearlyinaccordancewithcoordinativeIr–N(nucleobase)binding. were determined, for instance, for the rate of water exchange at
Melting temperatures and UV/Vis-titrations for CT-DNA with 298K in [(g5-C Me )Rh(bpy)(H O)]2+, and its [(g5-C Me )IrIII and
5 5 2 5 5
the (g5-C Me )IrIII cations [(g5-C Me )Ir(HcysOMe)(phen)]2+ and [(g6-C Me )RuII analogues. As a change in the DNA interaction
5 5 5 5 6 6
[(g5-C Me )Ir(H metOMe)(phen)]3+ [27] have demonstrated that modefrominitialintercalationtocoordinativeM–N(nucleobase)
5 5 2
the phenanthroline ligand is too small to support intercalation bindingshouldclearlybekineticallyfavouredforM=RhIIIincom-
forthisclassofhalf-sandwichcomplexes.ThepositiveDT values parisontoM=IrIII,theretentionoftheformermodeforthe[(g5-
m
of 1–2(cid:2)C for these kinetically inert cations are, however, signifi- C Me )RhIIIcomplexes5and8mustresultfromitsrelativether-
5 5
cantly higher than those in the range (cid:2)5 to (cid:2)1(cid:2)C observed for modynamicstabilityforthe4dtransitionmetal.M–N(nucleobase)
1–3,whichcanpotentiallybindtothenucleobaseNatomsofthe bonds will be significantly weaker for rhodium(III) than for its
biopolymer. A similar DT value of (cid:2)2(cid:2)C has been reported for muchsofter5dhomologue.
m
[(g6-C H )RuCl(en)]+,whichpreferentiallycoordinatesDNAguan-
6 6
inebasesattheirendocyclicN7atoms[11].1HNMRstudiesdem-
onstrated that N7 coordination is complete within 5min for 1:2
0.5
reaction mixtures of 1–6 with guanosine 50-monophosphate at
5: pp = dppz
0.4
L = Cl
Table2
Melting temperature shifts DTm, for the interaction of complexes [(g5-
C5Me5)Rh(L)(pp)](CF3SO3)n (1–11) with CT DNA at r=0.1 in a 10mM phosphate 0.3
bufferatpH7.2afterincubationfor60min
A 5
Compound pp L n DTm((cid:2)C)
0.2
1 en Cl 1 (cid:2)4
2 bpy Cl 1 (cid:2)5
5+DNA
3 phen Cl 1 (cid:2)1 0.1
4 dpq Cl 1 +6
5 dppz Cl 1 +12
6 dppn Cl 1 +7
0
7 dpq (Me2N)2CS 2 +8 300 340 380 420 460
8 dppz (Me2N)2CS 2 +12
wavelength /nm
9 dppn (Me2N)2CS 2 +6
10 dppz C6H5S 1 +2
11 dppz C10H7S 1 +1 Fig.2. UV/Visspectraofdppzcomplex5(20lM)anditsequilibriumreactionm-
ixturewithCTDNA(200lM)ina10mMphosphatebuffer(pH7.2)after5min.
M.A.Scharwitzetal./JournalofOrganometallicChemistry693(2008)2299–2309 2303
for 9 and, in contrast to the analogous [(g5-C Me )IrIII [19] and
5 5
[(g6-C Me )RuII complexes [15], no positive changes in the DA/A
6 : pp = dppn 6 6
values are observed during this time period. The respective final
1.2 L = Cl
DA/A values for 6 and 9 at the first maximum are (cid:2)46% and
(cid:2)38%. No effective changes in absorption in the range 320–
450nmwererecordedforcomplexes1–3(pp=en,bpy,phen)on
0.8 mixing with CT DNA and the changes for complex 4 (pp=dpq)
A
were of a minor nature. It is interesting to compare the UV/Vis
spectraof1:10mixturesof10and11withCTDNAwiththosere-
cordedfor5and8.Anintramolecularp–pinteractionbetweenthe
0.4 dppzandnaphthalenearomaticsystemsisapparentfor11inthe
6
solidstate(Fig.4b),asevidencedbytheirinterplanarangleofonly
11.3(3)(cid:2).Asobservedforthedppzcomplex5,thefive-membered
6+DNA
chelatingin 11is almostflat with aninterplanar angleof 1.4(4)(cid:2)
0
320 360 400 440 480 between N1–Rh–N10 and N1–C–C–N10 planes. The dppz and
wavelength /nm cyclopentadienyl planes are inclined at 60.7(3)(cid:2). Low respective
values of (cid:2)15% and (cid:2)10% for DA/A at 364nm for complex/DNA
Fig.3. UV/Visspectraofdppncomplex6(20lM)anditsequilibriummixturewith mixturesof10and11indicatethatsuchinteractionsarestrongen-
CTDNA(200lM)ina10mMphosphatebuffer(pH7.2)after20min.
oughtopreventasignificantdegreeofintercalation.Fig.4adepicts
themolecularstructureofthe[(g5-C Me )Rh{(Me N) CS) }(dpq)]+
5 5 2 2 2
In similarity to the dppz complexes, significant decreases in cation 7, whose dppz and dppn analogues 8 and 9 exhibit large
absorbance are also observed for the dppn complexes 6 (Fig. 3) negativevaluesofDA/Aforp–p*transitionsonmixingwithDNA.
and 9 at their characteristic maxima at about 323, 398 and It is apparent that the steric bulk of tetramethyl thiourea ligand
420nm. Equilibrium is achieved within 12min for 6 and 30min willbegreaterthanthatofthebenzenethiolateligandin10,whose
a
b
Fig.4. Molecularstructuresofthecationsof:(a)[(g5-C5Me5)Rh(dpq){(Me2N)2CS}](CF3SO3)2(7)and(b)[(g5-C5Me5)Rh(C10H7S)(dppz)](CF3SO3)(11).
2304 M.A.Scharwitzetal./JournalofOrganometallicChemistry693(2008)2299–2309
volumerequirementscanbeestimatedfromthemolecularstruc-
tureoftheanalogous[(g5-C 5 Me 5 )Rh(dppz)(C 10 H 7 S)]+cationof11 8 DNA
(Fig.4b).Theobservationofintercalationfor8butnotfor10indi-
catesthattheintramolecularp–pinteractionsmayberetainedfor 7
thelattercomplexoninteractionwithCTDNA.Incontrasttodpq
complex 4, the N1–Rh1–N10 plane in 7 is titled below and not 4 4
abovetheplaneofthepolypyridylligand,presumablyduetosteric
repulsionbetween(Me N) CSmethylgroupsanddpqatoms.This
2 2
distortionleadstoamuchsmallerinterplanarangleof44.3(5)(cid:2)be-
0
tweenthecyclopentadienylanddpqplanesincomparisiontothat [θ] 240 260 280 300 330 340
of76.6(2)(cid:2)in4.TheN1–Rh1–N10planeisinclinedat10.5(9)(cid:2)to λ (nm)
theN1–Rh1–N10planeofthechelatering.
-4
2.2.2.Circulardichroismus
Characteristic changes in the observed circular dichroismus
(CD)spectraofDNAintherange220–350nmprovideaconvenient
-8
means of monitoring conformational changes for the biopolymer
[30,31].AnegativeCDbandat246nmcausedbythehelicalBcon-
formationandapositivebandat275nmduetobasestackingare
observedforCTDNA[31].Additionofthecomplexes1–3(pp=en, -12
bpy, phen) at a 1:10 complex/[DNA] molar ratio leads to only
minor spectral changes, thereby indicating little distortion of the F an ig d .6 [( . g C 5- D C5 s M pe e c 5 t ) r R a h o (d f p C q T ){ D (M NA e2 a N n ) d 2C m S} i ] x ( t C u F r 3 e S s O o 3 f )2 [( ( g 7 5 ) -C w 5 i M th e5 C ) T R D h N Cl A (d i p n q a )] 1 (C 0 F m 3S M O3 p ) h ( o 4 - )
BhelixonformationofRh–N(nucleobase)bondstothebiopoly- sphatebuffer(pH7.2)after2hincubation.
mer. Although aromatic molecules often generate CD bands be-
tween 260 and 400nm on interaction withDNA,the presenceof
suchsignals is not necessarily ofrelevance for the assignmentof (cid:2)DA/A values. When compared to the CD spectrum of CT DNA
the binding mode. The observed induced circular dichroismus is alone,abroadeningandapparentsplittingare,however,discern-
caused by the rigid orientation of the molecule with respect to able for the positive bands of the 1:10 mixture of the dpq com-
thedoublehelixandthiscanalsoresultfromcoordinative,surface plexes 4 and 7 with the biopolymer (Fig. 6). As 4 and 7 both
orgroovebindinginadditiontointercalation.AsdepictedinFig.5, exhibitabsorptionmaximaforspin-allowedMLCTp–p*transitions
thecomplexes5and8do,however,exhibitcharacteristicnegative atk=260and300nm,itisprobablethattheshapeofthebandsin
CD bands with molar ellipticities [h] of (cid:2)3.4(cid:5)10(cid:2)3 and (cid:2)4.5(cid:5) the 270–295nm range will be due to spectral overlap with a
10(cid:2)3degcm2mol(cid:2)1 at k=296nm and it has previously been negative CD contribution caused by intercalating dpq ligands. A
established[15,19,29]thattheappearanceofsuchbandsis,indeed, similarshapehasbeenobservedforthepositivebandofamixture
characteristicfordppzintercalation.Thesignificantincreasesand of the intercalating complex [(g6-C Me )RuCl(dpq)](CF SO ) with
6 6 3 3
the blue shifts of the positive CD bands at 268nm may of be CT DNA [15] and this is accompanied by an additional positive
attributedtoresultingchangesinthebasestackingofthedouble bandat315nm,asisalsothecaseforcomplexes4and7.TheCT
helix. DNAbandsintherange220–300nmexhibitonlyminorchanges
Incontrastto5and8,thedppzcomplexes10and11donotex- following incubation with the dppn complexes 6 and 9 for 2h
hibitnegativeinducedCDbandsataboutk=300nm.Thisfinding andremainunchangedoveraperiodof24h.Similarobservations
isinaccordancewithanabsenceofsignificantDNAintercalation, were also made for the analogous chlorido cations [(g5-
as was indicated for these complexes by their low DT and C Me )IrCl(dppn)]+and[(g6-C Me )RuCl(dppn)]2+[19,15].Incon-
m 5 5 6 6
trast, the interaction with [(g5-C Me )Ir{(R N) CS}(dppn)]2+
5 5 2 2
(R=H, Me) and [(g6-C Me )Ru{(H N) CS}(dppn)]2+ leads to mas-
16 6 6 2 2
sive changes in the original B DNA conformation. The negative
5 bandatabout246nmdisappearsinallthreecasesandtheresult-
ingCDspectrumforthe(g6-C Me )RuIIcomplexisinaccordance
12 6 6
withDNAdamageandcompletelossoftheB-helixconformation.
8 Alargeincreasein[H]to10.2(cid:5)103degcm2mol(cid:2)1forthepositive
8 CDbandsofthe[(g5-C 5 Me 5 )Ir{(Me 2 N) 2 CS}(dppn)]2+/CTDNAmix-
ture is indicative of the possible adoption of an A DNA
conformation.
4
[θ]
DNA 2.2.3.Viscositymeasurements
ThemostconvincingevidenceforDNAintercalationisprovided
byviscositymeasurements[32,33].Insertionofaromaticsligands
240 260 280 300 320 340 360 such as dpq or dppz between adjacent nucleobase pairs leads to
λ (nm) lengthening and stiffening of the double helixand these changes
-4
arereflectedinanincreaseinDNAviscosity.Fig.7illustratesthe
dependenceofthelogarithmicrelativereducedviscosityln(g/g )
0
on the concentration function ln(1+r) (r=[complex]/[DNA]] for
-8
the chlorido complexes 4–6. Effective intercalation is confirmed
bytheslopevaluesof2.74and2.17forthecomplexes4(pp=dpq)
and 5 (pp=dppz), in accordance with CD observations. Large in-
F an ig d .5 [ . (g C 5 D -C s 5 p M e e ct 5 r ) a Rh o ( f d C p T pz D ) N {( A M a e n 2N d ) m 2C i S x } t ] u (C re F s 3S o O f 3 [ ) ( 2 g5 ( - 8 C ) 5M w e i 5 th )Rh C C T l(d D p N p A z)] (r (C = F 0 3 . S 1 O ) 3 i ) n (5 a ) creases in ln(g/g 0 ) are also found for the likewise intercalating
10mMphosphatebuffer(pH7.2)after2hincubation. (Me 2 N) 2 CS complexes 7 (pp=dpq) and 8 (pp=dppz) with their
M.A.Scharwitzetal./JournalofOrganometallicChemistry693(2008)2299–2309 2305
0.4 surface or by intermolecular interactions between coordinating
[(η-Cp*)RhCl(dpq)] (4) metalfragments.
[(η-Cp*)RhCl(dppz)] (5)
0.3 [(η-Cp*)RhCl(dppn)] (6) 2.3.Cellularuptake
Wehavepreviouslydemonstratedthatthecytotoxicityofcom-
η
)0
0.2 plexes of the types [(g6-C Me )RuCl(pp)]+ and [(g6-C Me )Ru-
η
n
( / {(NH 2 ) 2 CS}(pp)]2+isgoverne 6 dby 6 thesizeofthepolypyrid 6 ylli 6 gand
l (pp=dpq,dppz,dppn)andtheresultingcellularuptakeefficiency
0.1
[15].Owingtotheirdecreasedlipophilicity,cellularrutheniumlev-
els are much lower for the latter dicationic thiourea complexes
than for the monocationic chlorido species and this causes a de-
0
creaseincytotoxicity.Inordertoevaluatetheinfluenceofcellular
uptakeonthecytotoxicityof(g5-C Me )RhIIIcomplexes,wequan-
0.02 0.04 0.06 0.08 0.1 0.12 5 5
ln (1+r) tifiedtherhodiumconcentrationsintumourcellsexposedto1–11
by atomic absorption spectroscopy. The similar DT values (Sec-
m
Fig.7. ViscometrictitrationsofsonicatedDNAwiththechloridocomplexes4–6 tion2.2.1)forpairsof(g5-C Me )RhIIIcomplexes4/7,5/8and6/9
5 5
(pp=dpq,dppz,dppn)ina10mMphosphatebufferatpH7.2;g=reducedvicosity withthesamepolypyridylligandindicatethat4–6mustbepres-
oftheDNAsolutioninthepresenceofacomplex,g0=reducedvicosityoftheDNA entasdications[(g5-C Me )Rh(H O)(pp)]2+followingrapidaqua-
solutionalone,r=[complex]/[DNA]. 5 5 2
tion.Asaresultofthisincreaseincationcharge,levelsofcellular
uptake(Table3)arelowforcomplexes4–6(and1–3)andcompa-
rablewiththevaluesestablishedforthekineticallyinertdicationic
respectiveslopevaluesbeing2.66and1.44.Incontrasttothedpq complexes7–9.Thesmallincreasesinintracellularrhodiumlevels
anddppzcomplexes,significantDNAlengtheningisnotobserved forthelattercomplexes(comparethepairs4/7,5/8and6/9)may
for the dppn complexes 6 and 9 in the 0.0–0.08 range for be due to the presence of hydrophobic (Me N) CS ligands rather
2 2
ln(1+r). It can, therefore, be concluded that coordinative Rh–N thancoordinatingwatermolecules.Thisresultisofspecialinterest
(nucleobase)bindingwillprobablybepredominantforthechlori- astheoppositetrendwasexpectedbasedonthefactthat7–9rep-
docomplex6andelectrostaticsurfaceinteractionsfor(Me N) CS resentdicationicspecies,whicharesupposedlylesslipophilicthan
2 2
complex9atlowvaluesofr.Interestingly,asmallbutsteadyin- themonocationiccomplexes4–6.Atrendtoincreaseduptakeeffi-
creaseisobservedfortherelativereducedvelocityln(g/g )ofboth ciency is apparent for both series, in accordance with increasing
0
complexesatln(1+r)valuesgreaterthan0.08.Thissuggeststhat lipophilicity of the polypyridyl ligands in the order of increasing
surfacestackingofadjacentdppnligandscouldcauseDNAlength- sizedpq<dppz<dppn.However,100lMconcentrationsofthese
eningathigher[complex]/[DNA]ratios.Thepresenceofsuchinter- complexesarenecessarytoachievecellularlevelssimilartothose
actions is clearly indicated by the significant decreases in obtainedfor10lMsolutionsofanalogous(g5-C Me )IrIIIand(g6-
5 5
absorbance for the p–p* transitions of complexes 6 (Fig. 3) and 9 C Me )RuII dications.For instance,similarcellular metal levels of
6 6
atr=0.1. respectively150ng Ir/mg protein(=0.78nmol Ir/mgprotein) and
ThefailureofthelargedppnligandtoexhibitsignificantDNA 85ngRu/mgprotein(=0.84nmolRu/mgprotein)weredetermined
intercalation in its (g5-C Me )RhIII and (g6-C Me )RuII [15] com-
5 5 6 6
plexesisinstrikingcontrasttofindingsforothertypesoftransition
metal complexes. For instance, a binding constant K =7.8(cid:5)
b Table3
104M(cid:2)1 and site size s of 1.4 have been recently determined for
Cellularrhodiumlevels[ng(Rh)/mg(protein)]inMCF-7andHT-29tumourcellsafter
the complex [Ir(dppn)(ppy)
2
](PF
6
) (Hppy=2-phenylpyridine) exposuretogivenconcentrationsofcomplexes[(g5-C5Me5)Rh(L)(pp)](CF3SO3)n(1–
[34]. This K value is significantly larger than that of 2.0(cid:5) 11)for4h
b
104M(cid:2)1 (site size s=1.3) obtained for the analogous dppz
Compound pp L n c(lM) MCF-7 HT-29
complex,inagreementwithsimilarmeasurementsforrhenium(I)
(i)Complexes1–11
complexes [35,36] and (tpm)RuII complexes of the type [(tpm)-
1 en Cl 1 100 4.9(2.7) 4.4(3.2)
RuCl(pp)]+ (tpm=tris(1-pyrazolyl)methane; pp=dppn, dppz) 2 bpy Cl 1 100 14.2(9.1) 10.7(3.1)
[37].ThesefindingsareinaccordancewiththeresultsofSartorius 3 phen Cl 1 100 11.9(2.9) 10.1(1.9)
and Schneider [38] for heterocyclic derivatives with positively 4 dpq Cl 1 100 11.3(0.5) 4.1(1.7)
5 dppz Cl 1 100 45.0(2.5) 42.2(4.4)
charged ammonium groups in their side chains. These authors
6 dppn Cl 1 10 1.8((cid:6)) 3.5(1.2)
foundthattheintercalationbindingstrengthisessentiallyafunc- 100 189.3(15.5) 378.1(66.0)
tionofthesizeofthearomaticsystem,independentofthehetero- 7 dpq (Me2N)2CS 2 100 18.7(1.7) 15.2(2.8)
atomsorthepresenceoflocalpositivechargeswithintheplanar 8 dppz (Me2N)2CS 2 100 90.5(6.1) 77.7(3.7)
moieties. The apparent lack of stable intercalation for complexes
9 dppn (Me2N)2CS 2 10 5.1((cid:6)) 6.7(0.9)
100 217.1(12.0) 331.5(25.2)
6and9andforanalogous(g6-C Me )RuIIcomplexesmust,there-
6 6 10 dppz C6H5S 1 1 59.2(2.8) 20.7(9)
fore, be attributed to their preferred side-on intercalation mode 10 453.3(37.9) 84.8(20.3)
[27,28]. Close non-bonding contacts to the DNA backbone would 100 2337(122.8) 1137.1(174.7)
be inevitable for the atoms of the additional six-membered ring 11 dppz C10H7S 1 1 76.1(29.2) 17.8(2.2)
10 398.9(44.2) 303.7(47.9)
ofdppnandmaypreventitssuitableside-onalignmentbetween
100 2904(132.3) 2029(36.5)
the nucleobases of B DNA. Massive conformational distortions
(ii)Analogousiridium(III)complexes[15,19]
andapossibleBtoADNAchangearenecessarytofacilitatedppn
5(Ir) dppz Cl 1 10 n.d. 70.4(1.0)
intercalation for [(g5-C 5 Me 5 )Ir{(Me 2 N) 2 CS}(dppn)]2+ [19]. Small 100 n.d. 556.1(32.3)
decreases in the reduced viscosity were recorded for mixtures of 9(Ir) dppn (Me2N)2CS 2 10 n.d. 149.6(7.8)
the complexes 1–3 with CT DNA at r values in the range 0.08– 100 n.d. 906.3(117.6)
0.12.Thesedecreasesmaybeduetokinkingorbendingofthedou- The values in brackets represent the standard errors ((cid:6), single value; n.d., not
blehelix[39,40]causedbyhydrophobicinteractionswiththeDNA determined).
2306 M.A.Scharwitzetal./JournalofOrganometallicChemistry693(2008)2299–2309
for HT-29 cells following incubation periods of 4h with 10lM theppligandsandtheir(g5-C Me )RhIIIcomplexesprovidesstrong
5 5
solutions of [(g5-C Me )Ir{(NMe ) CS}(dppn)]2+ [41] and [(g6- evidenceforthedominantinfluenceofthepolypyridylsurfacearea
5 5 22
C Me )Ru{(NH ) CS}(dppn)]2+ [15]. For the corresponding com- onthecytotoxicactivityof2–11.
6 6 22
plexes6and9at10lMconcentrationssignificantlylowervalues Inthiscontext,itisinterestingtoobservethattheIC values
50
werefound(below10ng(=0.10nmol)Rh/mgproteininallexper- for half-sandwich dppz complexes all lie within fairly narrow
iments).Thelowvalueof70.4ngIr/mgproteinobservedfor[(g5- ranges (MCF-7, 1.5–2.5lM; HT-29, 2.5–7.4lM) independent of
C Me )IrCl(dppn)]+suggeststhatthiscomplexmustalsobepres- theorganometallicfragment,theoverallcationchargeandthenat-
5 5
entasanaquadicationfollowingrapidH O/Cl(cid:2)exchange.Instrik- ure of the non-polypyridyl ligands. Values of 2.1/2.5 and 2.5/
2
ing contrast, a cell uptake of 1054.7ng Ru/mg protein was 6.7lMwerereportedforMCF-7/HT-29followingincubationwith
measured for a 10lM solution of the analogous [(g6-C Me )RuCl(dppz)]+ and [(g6-C Me )Ru{(NH ) CS}(dppz)]2+,
6 6 6 6 22
(g6-C Me )RuII complex, for which the aquation rate is much respectively[15].Incontrasttothedppzcomplexes,significantdif-
6 6
lower. ferences are, however, apparent for both the smaller and larger
Theinfluenceofcationchargeonthelevelofcellularuptakeis polypyridyl ligands. Despite their much poorer cell uptake, the
clearly apparent for the monocations of [(g5-C Me )Rh(L)- dicationsofthedpqcomplexes4and7aresome2–3timesmore
5 5
(dppz)](CF SO ) 10 (L=C H S(cid:2)) and 11 (L=C H S(cid:2)), where active than [(g6-C Me )RuCl(dpq)]+, for whichIC values of 11.1
3 3 6 5 7 10 6 6 50
intracellular levels for 1lM solutions are comparable with those and30.5lMwereestablishedforMCF-7andHT-29,respectively.
establishedfor100lMsolutionsofthedppzdicationsof5and8. Theimprovementincytotoxicityisevenmorestrikingforthephen
Increasedlipophilicityduetothepresenceofaromaticthiolateli- complex3,whichexhibitspromisingIC valuesof4.7and8.0lM
50
gandsmayalsoplayaroleinimprovingcomplexpassagethrough againstMCF-7andHT-29cells.Incontrast,previouslytestedcom-
thecellmembranefor10and11. plexes of the type [(g6-indan)RuCl(pp)]+ containing phenanthro-
line derivatives all show poor or no activity towards human
2.4.Cytotoxicitymeasurements A2780 cells (ovarian cancer) [2] and [(g5-C Me )IrCl(phen)]+ is
5 5
inactiveforMCF-7andHT-29cells.Asboththeenandbpycom-
IC valuesforthecomplexes1–11towardsMCF-7andHT-29 plexes 1 and 2 do not exhibit no significant cytotoxicity against
50
cells are listed in Table 4. As previously reported for analogous these celllines (Table4), it canbe assumedthat cellularinterac-
(g6-C Me )RuII complexes [15], the cytotoxicity of members of tions involving the phen aromatic system must be responsible
6 6
the series 4–6 and 7–9 is directly correlated to the surface area fortheactivityofcomplex3.
ofthepolypyridylligand(pp=dpq,dppz,dppn)and,therefore,to These observations for phen and dpq complexes indicate that
thelevelofcellularuptake.Howevertheincreasesincytotoxicity thenatureoftheorganometallicfragmentcansignificantlyinflu-
within the series dpq<dppz<dppn are much less pronounced encethecytotoxicityofcomplexescontainingsmallerpolypyridyl
than for the monocations [(g6-C Me )RuCl(pp)]+, where, for in- ligands.TheDNAbindingstudiesandtheresultsofthecellularup-
6 6
stance, the IC value for HT-29 cells decreases from 30.3 over takeandcytotoxicityinvestigationssuggestthatinitialDNArecog-
50
2.5to0.4lM[15].Thisweakeningofthepolypyridylligandinflu- nition [42] and intercalation rather than kinetically controlled
encefor(g5-C Me )RhIIIcomplexesismostapparentforpp=dppn coordinativeRh–N(DNA)bindingmaybeofcentralimportancein
5 5
andisclearlyaresultoftheverylowlevelsofcellularuptakefor determining the biological activity of the polypyridyl complexes
the aqua dications of 4–6 and the (Me N) CS dications of 7–9. It 4–11. A case in point is provided by the thiolate complexes 10
2 2
shouldbenotedthatwealsoobservedcytotoxiceffectsforthefree and 11, whose monocations exhibit much better cellular uptake
polypyridyl ligands, which increase in the order bpy<phen, valuesthanthoseestablishedforthedicationsof5and8.10and
dpq<dppz<dppn (Table 4). The similarity of the IC values for 11 are indeed, respectively, 2.3 and 2.7 more active towards
50
MCF-7 cells but significant improvements are not observed for
HT-29cells.Itseemsprobablethattheimpairmentofintercalation
Table4 caused by intramolecular stacking of the dppz ligands with the
IC50values(lM)aforthecomplexes[(g5-C5Me5)Rh(L)(pp)](CF3SO3)n(1–11) thiolatearomaticsystems,willrestrictpossibleincreasesincyto-
toxicity due to the much higher intracellular concentrations of
Compound pp L n MCF-7 HT-29
the complexes. An instructive example for the importance of
(i)Complexes1–11 DNAinteractionsisgivenbydppncomplex9andits(g5-C Me )IrIII
1 en Cl 1 >100 >100 5 5
2 bpy Cl 1 >100 >100 counterpart(Tables3and4),whoseIC 50 valuesforMCF-7andHT-
3 phen Cl 1 4.7(0.1) 8.0(1.6) 29cellsare,respectively,some5.4and8.8timeslowerthanforthe
4 dpq Cl 1 5.1(0.2) 8.5(0.3) (g5-C Me )RhIIIcomplex.Apossibleexplanationisprovidedbythe
5 5
5 dppz Cl 1 1.5(0.4) 4.3(0.2) fact that the (g5-C Me )IrIII complex exhibits strong intercalative
6 dppn Cl 1 0.8(0.1) 3.2(0.5) 5 5
binding,whichleadstoapossibleBtoAchangeintheDNAconfor-
7 dpq (Me2N)2CS 2 5.3(0.7) 10.7(0.5)
8 dppz (Me2N)2CS 2 1.5(0.1) 4.3(0.1) mation [19], whereas only surface stacking is observed for the
9 dppn (Me2N)2CS 2 0.91(0.06) 3.6(0.5) dppn ligand of the (g5-C
5
Me
5
)RhIII analogue. This example also
10 dppz C6H5S 1 0.64(0.08) 6.1(0.5) demonstratesthatthenatureofthemetalfragmentcanincertain
11 dppz C10H7S 1 0.56(0.25) 3.3(0.2)
cases significantly influence the cytotoxicity of its polypyridyl
(ii)Polypyridylligands complexes, even when one of the larger pp ligands (dppz, dppn)
bpy 52.7(7.8) 45.7(4.6)
ispresent.
phen 3.5(0.2) 2.7(0.5)
dpq 6.7(2.0) 7.0(2.2)
dppz 0.8(0.6) 1.8(0.2) 3.Conclusions
dppn 0.15(0.05) n.d.
(iii)Analogousiridium(III)complexes[15,19] We have demonstrated that the DNA binding of polypyridyl
5(Ir) dppz Cl 1 2.3(0.4) 7.4(0.9) organorhodium(III)complexesofthetypes[(g5-C Me )RhCl(pp)]-
5 5
9(Ir) dppn (Me2N)2CS 2 0.17(0.02) 0.41(0.16) (CF SO )1–6and[(g5-C Me )Rh{(Me N) CS}(pp)](CF SO ) 7–9is
3 3 5 5 2 2 3 32
(iv)Cisplatin 2.0(0.3) 7.0(2.0) governed by the size of the polypyridyl ligand. Whereas stable
a TowardsthehumancancercelllinesMCF-7(breastcancer)andHT-29(colon intercalative binding is observed for the dpq complexes 4 and 7
cancer);thevaluesinbracketsrepresentthestandarderrors. and the dppz complexes 5 and 8, the surface area of the dppn
M.A.Scharwitzetal./JournalofOrganometallicChemistry693(2008)2299–2309 2307
ligandof6and9isapparentlytoolargetosupportstableside-on [M(cid:2)CF SO ]+, 393 (45%) [M(cid:2)CF SO (cid:2)Cl]+. 1H NMR (d -DMSO,
3 3 3 3 6
intercalation. Intramolecular interactions between the aromatic 30(cid:2)C):d=1.67(s,15H,Cp*-CH ),7.87(dd,2H,H4/H40),8.33(dd,
3
systemsofthethiolateL(cid:2)anddppzligandsappeartopreventsig- 2H, H5/H50), 8.68 (d, 2H, H6/H60), 8.97 (d, 2H, H3/H30) ppm. 13C
nificant DNA intercalation for [(g5-C Me )Rh(L)- NMR(d -DMSO,30(cid:2)C):d=8.4(Cp*-CCH ),96.8(Cp*-CCH ),123.7,
5 5 6 3 3
(pp)](CF SO ),10(L=C H S(cid:2))and11(L=C H S(cid:2)). 128.3,140.3,152.2(bpy-CH),153.9(bpy-C)ppm.
3 3 6 5 10 7
RapidCl(cid:2)/H Oexchangeleadstoformationofaquadicationsfor
2
complexes 1–6 and similar low levels of cellular uptake are 4.1.3.[(g5-C Me )RhCl(1,10-phen)](CF SO )(3)
5 5 3 3
observed for 4–6 and their kinetically inert (Me N) CS analogues Preparation as for 1 with the ligand phen (36mg, 0.2mmol).
2 2
7–9.Thecytotoxicityofthelattercompounds4–9isgovernedby Yield: 85% (103mg). C H ClF N O RhS; M=602.9g/mol. Anal.
23 23 3 2 3
thesizeofthepolypyridylligandswithIC valuesvaryinginthe Calc.:C,45.8;H,3.8;N,4.6;S,5.3.Found:C,45.3;H,3.8;N,4.6;
50
orderdppn<dppz<dpq.Anexceptiontothisruleisprovidedby S, 5.6%. LSIMS: m/z 604 (10%) [M]+, 567 (70%) [M(cid:2)Cl]+. 453
thephencomplex3,whoseIC valuesof4.7and 8.0lMagainst (100%) [M(cid:2)CF SO ]+, 417 (42%) [M(cid:2)CF SO (cid:2)Cl]+. 1H NMR (d -
50 3 3 3 3 6
MCF-7 and HT-29 cells are comparable with those observed for DMSO, 30(cid:2)C): d=1.74 (s, 15H, Cp*-CH ), 8.20 (dd, 2H, H3/H8),
3
thedpqcomplexes4and7. 8.33 (s, 2H, H5/H6), 8.95 (d, 2H, H4/H7), 9.39 (d, 2H, H2/H9)
ppm. 13C NMR (d -DMSO, 30(cid:2)C): d=8.5 (Cp*-CCH ), 96.9 (Cp*-
6 3
4.Experimental CCH 3 ),127.0,127.6,139.0,152.4(phen-CH),130.0,144.7(phen-C)
ppm.
Solvents were dried and distilled before use. 1H and 13C NMR
spectra were recorded on a Bruker DRX 400 spectrometer and 4.1.4.[(g5-C 5 Me 5 )RhCl(dpq)](CF 3 SO 3 )(4)
LSIMS data (liquid secondary ion mass spectrometry) on a VG Preparation as for 1 with the ligand dpq (46.5mg, 0.2mmol).
Autospec instrument using 3-nitrobenzyl alcohol as the matrix. Yield: 83% (108mg). C 25 H 23 ClF 3 N 4 O 3 RhS; M=654.9g/mol; Anal.
13CNMRsignalsfortheCF SO (cid:2)anionsof1–11wereobservedin Calc.:C,45.8;H,3.5;N,8.6;S,4.9.Found:C,45.8;H,3.5;N,8.6;
3 3
thedrange121.8–122.8ppmandarenotlistedforindividualcom- S, 4.9%. LSIMS: m/z 505 (100%) [M(cid:2)CF 3 SO 3 ]+, 469 (60%)
plexes. An analytik jena SPECORD 200 was employed for UV–Vis [M(cid:2)CF 3 SO 3 (cid:2)Cl]+.1HNMR(d 6 -DMSO,30(cid:2)C):d=1.77(s,15H,Cp*-
measurements and CD spectra were registered on a Jasco J-715 CH 3 ), 8.38 (dd, 2H, H3/H8), 9.35 (s, 2H, H11/H12), 9.51 (d, 2H,
instrumentintherange220–400nmfor1:10complex/[DNA]mix- H4/H7), 9.72 (d, 2H, H2/H9) ppm. 13C NMR (d 6 -DMSO, 30(cid:2)C):
tures [DNA concentrations in M(nucleotide)] in a 10mM phos- d=8.5 (Cp*-CCH 3 ), 97.0 (Cp*-CCH 3 ), 128.1, 128.7, 135.4, 153.7
phate buffer at pH 7.2. Elemental analysis were performed on a (dpq-CH),138.8,146.3,146.8(dpq-C)ppm.
Vario El (Elementar Analysensysteme). RhCl (cid:4)3H O and Ag(CF
3 2 3-
SO 3 )wereobtainedfromChempur,2,20-bipyridine(bpy),1,10-phe- 4.1.5.[(g5-C 5 Me 5 )RhCl(dppn)](CF 3 SO 3 )(6)
nanthroline (phen), and tetramethyl thiourea from Acros, Preparationasfor1withtheliganddppn(66.5mg,0.2mmol).
ethylenediamine (en) from Riedel de Haen and 1-benzenethiol Yield:68%(107mg).C 33 H 27 ClF 3 N 4 O 3 RhS(cid:4)2H 2 O;M=791.0g/mol;
and 2-naphthalenethiol from TCI. Solvents were purchased from Anal. Calc.: C, 50.1; H, 3.9; N, 7.1; S, 4.0. Found: C, 50.0; H, 3.9;
TJBakerandDEUTEROGmbHandNaOMefromMerck.Thestarting N, 7.0; S, 4.2%. LSIMS: m/z 605 (100%) [M(cid:2)CF 3 SO 3 ]+, 569 (75%)
complex [{(g5-C 5 Me 5 )RhCl} 2 (l-Cl) 2 ] [43] and the polypyridyl li- [M(cid:2)CF 3 SO 3 (cid:2)Cl]+.1HNMR(d 6 -DMSO,30(cid:2)C):d=1.77(s,15H,Cp*-
gands dpq [44], dppz [45] and dppn [34] and [(g5- CH 3 ),7.74(dd,2H,H13/H14),8.42(dd,2H,H3/H8),8.47(dd,2H,
C Me )RhCl(dppz)](CF SO ) (5) [27] were synthesised according H12/H15), 9.18 (s, 2H, H11/H16), 9.44 (dd, 2H, H4/H7), 9.75 (dd,
5 5 3 3
toliteratureprocedures. 2H, H2/H9) ppm. 13C NMR (d 6 -DMSO, 30(cid:2)C): d=8.5 (Cp*-CCH 3 ),
97.1 (Cp*-CCH ), 127.9, 128.5, 128.6, 129.8, 129.9, 135.7 (dppn-
3
CH),134.5,137.9,140.4,148.3,153.9(dppn-C)ppm.
4.1.Preparationof1–4and6–11
4.1.6.[(g5-C Me )Rh(dpq){SC(NMe ) }](CF SO ) (7)
4.1.1.[(g5-C Me )RhCl(en)](CF SO )(1) 5 5 2 2 3 3 2
5 5 3 3 TwoequivalentsofAg(CF SO )(51.4mg,0.2mmol)wereadded
TwoequivalentsofAg(CF SO )(51.4mg,0.2mmol)wereadded 3 3
3 3 to[{(g5-C Me )RhCl} (l-Cl) ](61.8mg,0.1mmol)in10mlacetone
to[{(g5-C Me )RhCl} (l-Cl) ](61.8mg,0.1mmol)in10mlacetone 5 5 2 2
5 5 2 2 andstirredinthedarkfor0.5h.FiltrationoftheresultingAgClpre-
andstirredinthedarkfor0.5h.FiltrationoftheresultingAgClpre-
cipitate and subsequent solvent removal under vacuum yielded
cipitate and subsequent solvent removal under vacuum afforded
[(g5-C Me )RhCl(acetone) ](CF SO ),whichwasstirredwiththeli-
[(g5-C Me )RhCl(acetone) ](CF SO ),whichwasstirredwiththeli- 5 5 2 3 3
5 5 2 3 3 gand dpq (46.5mg, 0.2mmol) in CH OH/CH Cl (1:1, 10ml) at
gand ethylenediamine (en) (13.5ll, 0.2mmol) in CH OH/CH Cl 3 2 2
3 2 2 75(cid:2)C for 2h. Following cooling and solvent removal under vac-
(1:1,10ml)at75(cid:2)Cfor2h.Followingcoolingandsolventremoval
uum,theredproduct[(g5-C Me )RhCl(dpq)](CF SO )(4)wasdis-
under vacuum, the resulting solid was dissolved in 3ml CH OH. 5 5 3 3
3 solved in 10ml acetone and stirred in the dark for 0.5h with
Theproductwassubsequentlyprecipitatedbyadditionofdiethyl
Ag(CF SO ) (51.4mg, 0.2mmol). After filtration of the resulting
ether, washed and dried in vacuo. Yield: 71% (68mg). C H - 3 3
13 23 AgCl precipitate and solvent removal red [(g5-C Me )Rh(ace-
ClF N O RhS; M=482.7g/mol.Anal. Calc.:C, 32.3;H, 4.8; N, 5.8; 5 5
3 2 3 tone)(dpq)](CF SO ) was stirred with (Me N) CS (26.5mg,
S, 6.7. Found: C, 32.0; H, 4.8; N, 5.9; S, 6.7%. LSIMS: m/z 447 3 32 2 2
(65%) [M(cid:2)Cl]+, 333 (100%) [M(cid:2)CF SO ]+, 297 (35%)
0.2mmol)for2hinCH
3
OH/CH
2
Cl
2
(1:1,10ml)at75(cid:2)C.Following
3 3 coolingandsolventremovaltheremainingsolidwasdissolvedin
[M(cid:2)CF SO (cid:2)Cl]+.1HNMR(d -DMSO,30(cid:2)C):d=1.68(s,15H,Cp*-
3 3 6 3mlCH OHandthenslowlyprecipitatedbycoveringthesolution
CH ), 2.33 (mm, 2H, en-CH ), 2.53 (m, 2H, en-CH ), 4.31 (m, 2H, 3
3 2 2 withdiethylether.Theresultingfinalproductwasdriedinvacuo.
en-NH ), 5.02 (m, 2H, en-NH ) ppm. 13C NMR (d -DMSO, 30(cid:2)C):
2 2 6 Yield: 71% (133mg). C H F N O RhS (cid:4)2H O; M=936.8 g/mol;
d=8.4(Cp*-CCH ),44.2(en-CH ),93.6(Cp*-CCH )ppm. 31 35 6 6 6 3 2
3 2 3 Anal.Calc.:C,39.7;H,4.2; N, 9.0;S,10.3. Found:C,39.5;H,4.2;
N, 9.0; S, 10.3%. LSIMS: m/z 751 (22%) [M(cid:2)CF SO ]+, 619
3 3
4.1.2.[(g5-C 5 Me 5 )RhCl(2,20-bpy)](CF 3 SO 3 )(2) (44%) [M(cid:2)CF 3 SO 3 (cid:2){(Me 2 N) 2 CS}]+, 470 (100%) [M(cid:2)2CF 3 SO 3 (cid:2)
Preparation as for 1 with the ligand bpy (31.3mg, 0.2mmol). {(Me N) CS}]+. 1H NMR (d -DMSO, 30(cid:2)C): d=1.73 (s, 15H, Cp*-
2 2 6
Yield: 76% (87mg). C 21 H 23 ClF 3 N 2 O 3 RhS; M=578.8 g/mol; Anal. CH 3 ),2.31(s,12H,CH 3 –SC{N(CH 3 ) 2 } 2 ),8.41(dd,2H,H3/H8),9.40
Calc.:C,43.6;H,4.0;N,4.8;S,5.5.Found:C,43.5;H,4.0;N,4.8; (s, 2H, H11/H12), 9.47 (d, 2H, H4/H7), 9.77 (d, 2H, H2/H9) ppm.
S,5.4%.LSIMS:m/z580(6%)[M]+,543(36%)[M(cid:2)Cl]+,429(100%) 13CNMR(d -DMSO,30(cid:2)C):d=8.3(Cp*-CCH ),42.7(SC{N(CH ) } ),
6 3 32 2
2308 M.A.Scharwitzetal./JournalofOrganometallicChemistry693(2008)2299–2309
99.4(Cp*-CCH ),128.5,129.2,136.0,154.4(dpq-CH),138.9,146.0, (k=0.71073Å).AnOxfordDiffractionSapphire2CCDdiffractome-
3
147.0(dpq-C),179.4(SC{N(CH ) } )ppm. ter was employed for the intensity data collection of compounds
32 2
4–6 and 11(cid:4)2H O using 1(cid:2) x scans and Mo Ka (for 4, 5 and
2
4.1.7.[(g5-C Me )Rh(dppz){SC(NMe ) }](CF SO ) (8) 11(cid:4)2H O) or Cu Ka radiation (for 6=1.54178Å). Semi-empirical
5 5 2 2 3 3 2 2
Preparationasfor7withtheligandsdppz(56.5mg,0.2mmol) absorptioncorrectionswereappliedtotheintensitiesinallcases.
and (Me
2
N)
2
CS (26.5mg, 0.2mmol). Yield: 77% (152mg). ThestructuresweresolvedbydirectmethodswithSHELXSandre-
C 35 H 37 F 6 N 6 O 6 RhS 3 (cid:4)2H 2 O; M=986.8 g/mol; Anal. Calc.: C, 42.6; fined against F2 with SHELXL [46]. Anisotropic temperature factors
H,4.2;N,8.5;S,9.8.Found:C,42.5;H,4.3;N,8.4;S,10.7%.LSIMS: were introduced for non-hydrogen atoms and with exception of
m/z801(43%)[M(cid:2)CF SO ]+,669(38%)[M(cid:2)CF SO (cid:2){(Me N) CS}]+, watermolecules,protonswererefinedatgeometricallycalculated
3 3 3 3 2 2
520 (100%) [M(cid:2)2CF SO (cid:2){(Me N) CS}]+. 1H NMR (d -DMSO, positions as riding atoms. The high values of R (0.122) and wR
3 3 2 2 6 1 2
30(cid:2)C):d=1.75(s,15H,Cp*-CH ),2.35(s,12H,CH –SC{N(CH ) } ), (0.323)for7(cid:4)H OareduetodisorderoftheCF SO(cid:2) counterions.
3 3 32 2 2 3 3
8.22 (d, 2H, H12/H13), 8.42 (dd, 2H, H3/H8), 8.54 (dd, 2H, H11/
H14), 9.46 (d, 2H, H4/H7), 9.87 (dd, 2H, H2/H9) ppm. 13C NMR 4.3.DNAbindingstudies
(d -DMSO, 30(cid:2)C): d=8.3 (Cp*-CCH ), 42.7 (SC{N(CH ) } ), 99.5
6 3 32 2
(Cp*-CCH ), 128.8, 129.4, 129.8, 139.5, 154.5 (dppz-CH), 132.8, The thermal denaturation temperatures T of 1:10 complex/
3 m
136.3,142.1,147.2(dppz-C),179.2(SC{N(CH ) } )ppm. DNA mixtures [DNA concentration=M(nucleotide)] were deter-
32 2
minedfor1–11ina10mMphosphatebufferatpH7.2.Measure-
4.1.8.[(g5-C Me )Rh(dppn){SC(NMe ) }](CF SO ) (9) ments for 10 and 11 were performed in the presence of 1%
5 5 2 2 3 3 2
Preparationasfor7withtheligandsdppn(66.5mg,0.2mmol) DMSO. Melting curves were recorded at 1(cid:2) steps for the wave-
and(Me N) CS(26.5mg,0.2mmol).Yield:66%(139mg).C H - lengthk=260nmwithananalytikjenaSPECORD200spectrometer
2 2 39 39
F N O RhS (cid:4)3H O;M=1054.9g/mol;Anal.Calc.:C,44.4;H,4.3; equippedwithaPeltiertemperaturecontroller.T valueswerecal-
6 6 6 3 2 m
N, 8.0; S, 9.1. Found: C, 44.9; H, 4.3; N, 8.0; S, 10.4%. LSIMS: m/z culatedbydeterminingthemidpointsofmeltingcurvesfromthe
851 (47%) [M(cid:2)CF SO ]+, 719 (48%) [M(cid:2)CF SO (cid:2){(Me N) CS}]+, firstorderderivatives.TheexperimentalDT valuesareestimated
3 3 3 3 2 2 m
570 (100%) [M(cid:2)2CF SO {(Me N) CS}]+, 1H NMR (d -DMSO, tobeaccuratewithin±1(cid:2)C.ConcentrationsofCTDNAweredeter-
3 3 2 2 6
30(cid:2)C):d=1.74(s,15H,Cp*-CH ),2.41(s,12H,CH –SC{N(CH ) } ), mined spectrophotometrically using the molar extinction coeffi-
3 3 32 2
7.80 (dd, 2H, H13/H14), 8.42 (dd, 2H, H3/H8), 9.27 (s, 2H, H11/ ciente =6600M(cid:2)1cm(cid:2)1[47].
260
H16),9.43 (dd,2H,H4/H7),9.49 (m, 2H,H12/H15),9.85 (dd,2H, Viscositiesformixturesofcomplexes1–9withsonicatedDNA
H2/H9)ppm. weredeterminedusing a Cannon-Ubbelhode semi-micro dilution
viscometer(SeriesNo.75,CannonInstrumentsCo.)heldatacon-
4.1.9.[(g5-Cp*)Rh(dppz)(C H S)](CF SO )(10) stanttemperatureof25(cid:2)Cinawaterbath.Theviscometerinitially
6 5 3 3
Preparationasfor7withtheligandsdppz(56.5mg,0.2mmol) contained 2ml of 0.4mM sonicated DNA solution in a 10mM
and C H SH (20.5ll, 0.2mmol) and additional 37ll 30% NaOH/ phosphatebuffer(pH7.2).0.2mMcomplexsolutionsalsocontain-
6 5
CH OH(0.2mmol).Yield:86%(140mg).C H F N O RhS (cid:4)2H O; ing 0.4mM sonicated DNA were added in increments of 100ll
3 35 30 3 4 3 2 2
M=814.7g/mol;Anal.Calc.:C,51.6;H,4.2;N,6.9;S,7.9.Found:C, fromamicropipet.Solutionswerepassedthroughfilterstoremove
51.2;H,4.2;N,6.7;S,8.0%.LSIMS:m/z629(82%)[M(cid:2)CF SO ]+,519 particulatematerialpriortouse.Reducedviscositiesgwerecalcu-
3 3
(100%)[M(cid:2)CF SO (cid:2){C H S}]+,1HNMR(d -DMSO,30(cid:2)C):d=1.78 lated by literature methods [32] and plotted as In(g/g ) (g =re-
3 3 6 5 6 o o
(s, 15H, Cp*-CH ), 5.83 (d, 2H, H20/H60), 5.90 (t,2H, H30/H50), 5.99 duced viscosity of the DNA solution in the absence of complex)
3
(t,1H,H40),8.16(d,2H,H12/H13),8.26(dd,2H,H3/H8),8.47(d, againstIn(1+r)forrod-likeDNA(approximately600basepairs).
2H,H11/H14), 9.27 (d, 2H,H4/H7), 9.57 (d, 2H,H2/H9) ppm.13C
NMR (d -DMSO, 30(cid:2)C): d=8.0 (Cp*-CCH ), 97.0 (Cp*-CCH ), 4.4.Cellcultures
6 3 3
129.4, 132.5, 134.2, 139.1 153.6 (dppz-CH), 134.4, 138.0, 141.8,
146.5 (dppz-C), 126.27, 126,32, 127. 70, 127.87, 128.95 (CH MCF-7 breast cancer and HT-29 human colon carcinoma cells
C H S),123.92(C-C H S)ppm. weremaintainedin10%(v/v)fetalcalfserumcontainingcellcul-
6 5 6 5
ture medium (minimum essential medium eagle supplemented
4.1.10.[(g5-C Me )Rh(dppz)(C H S)](CF SO )(cid:4)H O(11) with 2.2g NaHCO , 110mgl(cid:2)1 sodium pyruvate and 50mgl(cid:2)1
5 5 10 7 3 3 2 3
Preparationasfor7withtheligandsdppz(56.5mg,0.2mmol) gentamicin sulfate adjusted to pH 7.4) at 37(cid:2)C: 5% CO and pas-
2
and C H SH (32mg, 0.2mmol) and additional 37ll 30% NaOH/ sagedtwiceaweekaccordingtostandardprocedures.
10 7
CH OH(0.2mmol).Yield:67%(113mg).C H F N O RhS (cid:4)H O;
3 39 32 3 4 3 2 2
M=846.7g/mol;Anal.Calc.:C,55.3;H,4.1;N,6.6;S,7.6.Found: 4.5.Cellularuptakestudies
C,55.7;H,4.0;N,6.6;S,7.3.LSIMS:m/z679(90%)[M(cid:2)CF SO ]+,
3 3
520 (100%) [M(cid:2)CF SO (cid:2){C H S}]+, 1H NMR (d -DMSO, 30(cid:2)C): Forcellularuptakestudies,HT-29andMCF-7cellsweregrown
3 3 10 7 6
d=1.81(s,15H,Cp*-CH ),5.81(s,1H),6.08(d,1H),6.21(m,1H), untilatleast70%confluencyin175cm2cellcultureflasks.Stock
3
6.50 (m, 3H), 6.73 (d, 1H), 8.11 (d, 2H, H12/H13), 8.23 (dd, 2H, solutionsofcomplexes1–11inDMSOwerefreshlypreparedand
H3/H8), 8.32 (d, 2H, H11/H14), 9.29 (d, 2H, H4/H7), 9.39 (d, 2H, dilutedwithcellculturemediumtothedesiredconcentrations(fi-
H2/H9) ppm. 13C NMR (d -DMSO, 30(cid:2)C): d=8.1 (Cp*-CCH ), 97.2 nal DMSO concentrations: 0.1% v/v, final complex concentration:
6 3
(Cp*-CCH ), 129.5, 131.0, 132.2, 141.5, 153.6 (dppz-CH), 134.2, 1.0–100lM). The cell culture medium of the cell culture flasks
3
135.6, 138.4, 146.1 (dppz-C), 121.1, 123.3, 123.9, 124.5, 125.2, wasreplacedwith10mlofthecellculturemediumsolutionscon-
125.7,127.4(CH-C H S),125.9,128.7,129.1(C-C H S)ppm. taining1–11andtheflaskswereincubatedfor6hat37C/5%CO
10 7 10 7 2
The cell pellets were isolated, resuspended in 1–5ml twice dis-
4.2.X-raystructuralanalyses tilledwater,lysedbyusingasonotrodeandappropriatelydiluted
using twice distilled water. The rhodium content of the samples
Crystalsofcompounds2–7and11suitableforX-raystructural wasdeterminedbyatomicabsorptionspectroscopy(AAS,seebe-
analysis were obtained by slow evaporation of CH OH/H O solu- low)andtheproteincontentofseparatealiquotsbytheBradford
3 2
tions.Their crystal and refinementdata are summarised in Table method.TocorrectformatrixeffectsinAASmeasurements,sam-
1.Intensitydatafor2,3and7(cid:4)H OwerecollectedonaSiemens plesandstandardswereadjustedtothesameproteinconcentra-
2
P4 diffractometer at 294K using x scans and Mo Ka radiation tion by dilution with twice distilled water (matrix matched
M.A.Scharwitzetal./JournalofOrganometallicChemistry693(2008)2299–2309 2309
Table5 paper. These data can be obtained free of charge from The Cam-
GraphitefurnaceprogramforAASmeasurements bridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/da-
Step Temperature((cid:2)C) Ramp((cid:2)C/s) Hold(s) ta_request/cif. Supplementary data associated with this article
can be found, in the online version, at doi:10.1016/
Drying 90 10 40
Drying 105 7 30 j.jorganchem.2008.04.002.
Drying 120 15 20
Drying 500 50 30
References
Pyrolysis 1200 200 20
AZ(zeroing) 1200 0 6
[1] E.Hillard,A.Vessières,F.LeBideau,D.Plazuk,D.Spera,M.Huche,G.Jaouen,
Atomisation 2400 Maximum 9
ChemMedChem1(2006)551.
Tubecleaning 2500 1000 6
[2] A. Habtemariam, M. Melchart, R. Fernandez, S. Parsons, I.D.H. Oswald, A.
Parkin, F.P.A. Fabbiani, J.E. Davidson, A. Dawson, R.E. Aird, D.I. Jodrell, P.J.
calibration).PriortoAASanalysis20llTritonX-100(1%)and20 ll Sadler,J.Med.Chem.49(2006)6858.
[3] P.J.Dyson,G.Sava,DaltonTrans.(2006)1929.
nitricacidacid(13%)wereaddedtoeach200llsampleofthecell
[4] A.Vessières,S.Top,W.Beck,E.Hillard,G.Jaouen,DaltonTrans.(2006)529.
suspensions.Cellularuptakewasexpressedasngrhodiumpermg [5] L.Dale,J.H.Tocher,T.M.Dyson,D.I.Edwards,D.A.Tocher,Anti-CancerDrug
cellproteinfordataobtainedfromtwoindependentexperiments. Des.7(1992)3.
[6] Y.N.VashishtGopal,D.Jayaraju,A.K.Kandapi,Biochemistry38(1999)4382.
[7] Y.N. Vashisht Gopal, N. Konura, A.K. Kandapi, Arch. Biochem. Biophys. 401
4.6.Atomicabsorptionspectroscopy (2002)53.
[8] L.A.Huxham,E.L.S.Cheu,B.O.Patrick,B.R.James,Inorg.Chim.Acta352(2003)
238.
A Vario 6 graphite furnace atomic absorption spectrometer
[9] C.S.Allardyce,P.J.Dyson,D.J.Ellis,S.L.Heath,Chem.Commun.(2001)1396.
(Analytik Jena) was employed for the Rh quantification using a [10] W.S.Sheldrick,S.Heeb,Inorg.Chim.Acta168(1990)93.
wavelengthof343.5nm withabandpassof 0.5nm.A deuterium [11] R.E.Morris,R.E.Aird,P.delS.Murdoch,H.M.Chen,J.Cummings,N.D.Hughes,S.
lampwasusedforbackgroundcorrection.Matrixcontainingstan- Parsons,A.Parkin,G.Boyd,D.I.Jodrell,P.J.Sadler,J.Med.Chem.44(2001)
3616.
dards were obtained by addition of a rhodium stock solution [12] A.F.A.Peacock,A.Habtemariam,R.Fernández,V.Walland,F.P.A.Fabbiani,S.
(1mg/mLRhin10%HCl)tountreatedcellsuspensions.Probeswere Parsons,R.E.Aird,D.J.Jodrell,P.J.Sadler,J.Am.Chem.Soc.128(2006)1739.
injectedatavolumeof20llintostandardgraphitetubes.Drying, [13] A.F.A.Peacock,A.Habtemariam,S.A.Moggach,A.Prescimone,S.Parsons,P.J.
Sadler,Inorg.Chem.46(2007)4049.
pyrolysisandatomizationinthegraphitefurnacewereperformed
[14] A.Dorcier,W.H.Ang,S.Bolanˇo,L.Gonsalvi,L.Juillerat-Jeannerat,G.Laurenczy,
accordingtotheconditionslistedinTable5.Duringthetemperature M.Peruzzini,A.D.Phillips,F.Zanobini,P.J.Dyson,Organometallics25(2006)
programthegraphitetubewaspurgedwithaconstantargongas 4090.
[15] S.Schäfer,I.Ott,R.Gust,W.S.Sheldrick,Eur.J.Inorg.Chem.(2007)3034.
flow, which was only halted during the zeroing and atomisation
[16] K.E.Erkkila,D.T.Odom,J.K.Barton,Chem.Rev.99(1999)2777.
steps.Pyrolysisandatomisationtemperatureswereoptimisedprior [17] C.Metcalfe,J.A.Thomas,Chem.Soc.Rev.32(2003)215.
to the experiments. The recovery rates of the metallodrugs were [18] E.L. Menon, R. Perera, M. Navarro, R.J. Kuhn, H. Morrison, Inorg. Chem. 43
(2004)5373.
determinedinitiallyandusedforcalculationofthefinalexperimen-
[19] S.Schäfer,W.S.Sheldrick,J.Organomet.Chem.692(2007)1300.
talvalues.Themeanintegratedabsorbancesofduplicateinjections [20] L.Dadci,H.Elias,U.Frey,A.Hörnig,U.Koelle,A.F.Merbach,H.Paulus,J.S.
wereusedthroughoutthestudy.Thecharacteristicalconcentration Schneider,Inorg.Chem.34(1995)306.
forthedescribedmethodwas0.85±0.05(lgRhl(cid:2)1)/1%A. [21] D.A.Herebian,C.S.Schmidt,W.S.Sheldrick,C.vanWüllen,Eur.J.Inorg.Chem.
(1998)1991.
[22] G. Garcia, G. Sánchez, I. Romero, I. Solano, M.D. Santana, G. Lopez, J.
4.7.Cytotoxicitymeasurements Organomet.Chem.408(1991)241.
[23] U.Kölle,M.Grätzel,Angew.Chem.,Int.Ed.Engl.26(1987)507.
[24] U.Kölle,B.-S.Kang,P.Infelta,P.Comte,M.Grätzel,Chem.Ber.122(1989)
Theantiproliferativeeffectsofthecompoundsweredetermined 1869.
followinganestablishedprocedure[48].Cellsweresuspendedin [25] M.-T.Youinou,R.Ziessel,J.Organomet.Chem.363(1989)197.
cellculture medium(HT-29:2850cells/ml,MCF-7: 105 cells/ml), [26] S.Berger,J.Fiedler,R.Reinhardt,W.Kaim,Inorg.Chem.43(2004)1530.
[27] D.A.Herebian,W.S.Sheldrick,J.Chem.Soc.,DaltonTrans.(2002)966.
and100llaliquotsthereofwereplatedin96wellplatesandincu-
[28] A.Frodl,D.Herebian,W.S.Sheldrick,J.Chem.Soc.,DaltonTrans.(2002)3664.
batedat37(cid:2)C:5%CO 2 for48h(HT-29)or72h(MCF-7).Stocksolu- [29] S.Genc(cid:2)aslan,W.S.Sheldrick,Eur.J.Inorg.Chem.Soc.(2005)3840.
tions of the compounds in DMSO were freshly prepared and [30] D.M.Gray,in:G.D.Fasman(Ed.),CircularDichroismandtheConformational
AnalysisofBiomolecules,PlenumPress,NewYork,1996,p.469.
dilutedwithcellculturemediumtothedesiredconcentrations(fi-
[31] W.C. Johnson, in: N. Berova, K. Nakanishi, R.W. Woody (Eds.), Circular
nalDMSOconcentration:0.1%v/v).Themediumintheplateswas Dichroism:PrinciplesandApplications,2nded.,VCHPublishers,NewYork,
replacedwithmediumcontainingthecompoundsingradedcon- 2000,p.523.
[32] G.Cohen,H.Eisenberg,Biopolymers4(1966)429.
centrations(sixreplicates).Afterfurtherincubationfor72h(HT-
[33] D.Suh,J.B.Chaires,Bioorg.Med.Chem.3(1995)723.
29) or 96h (MCF-7) the cell biomass was determined by crystal [34] K.K.-W.Lo,C.-K.Chung,N.Zhu,Chem.Eur.J.12(2006)1500.
violetstainingandtheIC valueswereestablishedasthosecon- [35] V.W.-W.Yam, K.K.-W. Lo, K.-K.Cheung, R.Y.-C. Kong,J. Chem.Soc., Dalton
50
Trans.(1997)2067.
centrations causing 50% inhibition of cell proliferation. Results
[36] K.K.-W.Lo,R.Y.-C.Kong,Organometallics23(2004)3062.
werecalculatedfrom2to3independentexperiments. [37] S.P. Foxon, C. Metcalfe, H. Adams, M. Webb, J.A. Thomas, Inorg. Chem. 46
(2007)409.
Acknowledgements [38] J.Sartorius,H.-J.Schneider,J.Chem.Soc.,PerkinTrans.2(1997)2319.
[39] A.Raja,V.Rajendrian,P.UmaMasheswari,R.Balamunrugam,C.A.Kilner,M.A.
Halcrow,M.Palaniandavar,J.Inorg.Biochem.99(2005)1717.
Financial support of this work in Bochum and Berlin by the
[40] B. Selvakumar, V. Rajendrian, P. Uma Masheswari, H. Stoecki-Evans, M.
Deutsche Forschungsgemeinschaft (DFG) within the research Palaniandavar,J.Inorg.Biochem.100(2006)316.
groupFOR630‘‘BiologicalfunctionofOrganometallicCompounds” [41] M. Scharwitz, I. Ott, R. Gust, A. Kromm, W.S. Sheldrick, J. Inorg. Biochem.
(2008),doi:10.1016/j.jinorgbio.2008.03.001.
isgratefullyacknowledged.WearealsogratefultoHeikeMayer-
[42] J.S.Ren,T.C.Jenkins,J.B.Chaires,Biochemistry39(2000)8439.
FiggeandHeikeSchefflerfortechnicalsupport. [43] J.W.Kang,K.Moseley,P.M.Maitlis,J.Am.Chem.Soc.91(1969)5970.
[44] J.G. Collins, A.D. Sleemann, J.R. Aldrich-Wright, I. Greguric, T.W. Hambley,
Inorg.Chem.37(1998)3133.
AppendixA.Supplementarymaterial [45] A.Delgadillo,P.Romo,A.M.Leiva,B.Loeb,Helv.Chim.Acta86(2003)2110.
[46] G.M.Sheldrick,SHELXS97andSHELXL97,Göttingen,Germany,1997.
CCDC 675527, 675528, 675529, 675530, 675531, 675532 and [47] J.Marmur,J.Mol.Biol.3(1961)208.
[48] I.Ott,K.Schmidt,B.Kircher,P.Schuhmacher,T.Wiglenda,R.Gust,J.Med.
675533 contain the supplementary crystallographic data for this Chem.48(2005)622.