On the use of metal purine derivatives (M=Ir, Rh) for the selective labeling of nucleosides and nucleotides

Article abstract of DOI:10.1002/chem.201304091

The reactions of neutral or cationic Ir^III and Rh^III derivatives of phenyl purine nucleobases with unsymmetrical alkynes produce new metallacycles in a predictable manner, which allows for the incorporation of either photoactive (anthracene or pyrene) or electroactive (ferrocene) labels in the nucleotide or nucleoside moiety. The reported methodology (metalation of the purine derivative and subsequent marker insertion) could be used for the postfunctionalization and unambiguous labeling of oligonucleotides.

Full text of DOI:10.1002/chem.201304091

DOI: 10.1002/chem.201304091
Full Paper
&
Nucleobases
On the Use of Metal Purine Derivatives (M=Ir, Rh) for the
Selective Labeling of Nucleosides and Nucleotides
Marta Valencia,[a] Mamen Martꢀn-Ortiz,[a] Mar Gꢁmez-Gallego,*[a]
Carmen Ramꢀrez de Arellano,[b] and Miguel A. Sierra*[a]
Abstract: The reactions of neutral or cationic IrIII and RhIII de-
rivatives of phenyl purine nucleobases with unsymmetrical
alkynes produce new metallacycles in a predictable manner,
which allows for the incorporation of either photoactive (an-
thracene or pyrene) or electroactive (ferrocene) labels in the
nucleotide or nucleoside moiety. The reported methodology
(metalation of the purine derivative and subsequent marker
insertion) could be used for the postfunctionalization and
unambiguous labeling of oligonucleotides.
Introduction
The design and synthesis of unnatural nucleic acids is a fasci-
nating area of research.[1] The idea that pairing of purine and
pyrimidine derivatives through hydrogen bonding is not an ab-
solute requirement for duplex formation or replication of DNA
has opened new ways of approaching the design of bases and
base pairs that in many cases have scarce similarity with the
natural ones.[1–3] These new bases are not just valuable to be
used as part of an expanded genetic alphabet,[4,5] or as the
origin of novel interactions in RNA translation,[6] but are also
suitable places to introduce specific labeling points, which
would occur in applications ranging from fluorescent probes
to the design of new nanomaterials.[7,8] Additionally, unnatural
nucleic acid bases could also be interesting for their biological
activity.[9]
We have reported a general approach to metal arylpurine
nucleosides 1, nucleotides 2, and dinucleotides 3 having MÀC
bonds (M=Ir, Rh; Scheme 1).[10] These compounds are suscep-
tible to selective postfunctionalization by means of alkyne in-
sertion reactions. Our preliminary work with iridanucleoside 4
and dimethylacetylene dicarboxylate showed the clean and
complete incorporation of a single alkyne unit to form cyclo-
metalated complex 5 in excellent yield (Scheme 1). This simple
experiment opens doors to the possibility of postfunctionaliza-
[a] M. Valencia, M. Martꢀn-Ortiz, Prof. Dr. M. Gꢁmez-Gallego,
Prof. Dr. M. A. Sierra
Departamento de Quꢀmica Orgꢂnica I, Facultad de Ciencias Quꢀmicas
Universidad Complutense, 28040 Madrid (Spain)
E-mail: margg@quim.ucm.es
sierraor@quim.ucm.es
Scheme 1. Some representative examples of metal (MÀC) purine derivatives.
[b] Prof. Dr. C. Ramꢀrez de Arellano
Departamento de Quꢀmica Orgꢂnica
Universidad de Valencia, 46100 Valencia (Spain)
tion of metalated nucleic acid derivatives in a complete site-se-
lective fashion (the site selectivity will be provided by the ini-
tial metal complex). We report herein the regioselective post-
functionalization of metallanucleosides and metallanucleotides
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201304091, and contains general proce-
dures, full synthetic details, spectroscopic data, electrochemical studies, and
crystallographic data of compounds 9Db, 10Pc, 25, and 28.
Chem. Eur. J. 2014, 20, 3831 – 3838
3831
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
and explore it as tool to incorpo-
rate fluorescent or redox labels
in their structures. The successful
application of the reported
methodology to the selective
postfunctionalization of oligonu-
cleotides (metalation of specific
modified bases within the chain
and subsequent incorporation of
the label) could be considered
an alternative approach[11] to the
incorporation of labeling points
in modified DNA fragments.
Results and Discussion
The possibility of effecting inser-
tion reactions on nucleotides
with unsymmetrically substitut-
Scheme 2. Insertion reactions of model purine derivatives with unsymmetrical alkynes.
ed alkynes 6 was initially tested
on cyclometalated N-9-methyl-6-
phenylpurines 7 and 8 as the
model compounds (Scheme 2). Compounds 7 and 8 were pre-
pared in 72 and 69% yields, respectively, from N-9-methyl-6-
phenylpurine, [MCl2Cp*]2 (M=Ir, Rh; Cp*=pentamethylcyclo-
pentadienyl), and AcONa in CH2Cl2 at room temperature.[12] The
reactions with alkynes 6 were carried out in MeOH at room
temperature and the products were isolated by chromatogra-
phy on silica gel.[13] As two orientations were in principle possi-
ble in the insertion process, throughout the discussion we will
define P (proximal) as the regioisomer in which the carboxylate
group is placed next to the metal fragment in the final prod-
uct, and D (distal) as that in which the carboxylate is placed
next to the 6-phenyl ring (Scheme 2).
the analysis of the NMR spectrum of the reaction crude
showed extensive decomposition and only the signals of the
proximal regioisomer 10Pb could be clearly observed. The
product was isolated in 37% yield.
With these results in hand, the reaction of metallacycles 7
and 8 with alkyne 6c having the bulky 9-anthranyl group was
assayed, and led exclusively to the formation of the proximal
regioisomers 9Pc and 10Pc in good yields of isolated prod-
ucts. The structure and regiochemistry of compounds 9Pc and
10Pc were established by analytical and spectroscopic tech-
niques and, for rhodium complex 10Pc, confirmed by X-ray dif-
fraction analysis of a suitable crystal obtained by slow diffusion
Terminal alkynes were tested first against the insertion reac-
tion. The reaction of phenylacetylene with 7 and 8 only led to
the recovery of unreacted starting material, together with de-
composition byproducts. However, reaction of methyl propio-
late 6a with cyclometalated Ir and Rh complexes 7 and 8 led
respectively to the monoinsertion products 9Pa and 10Pa
bearing the carboxylate group next to the metal fragment,
with complete regioselectivity and in excellent yields of isolat-
ed products (Scheme 2). The structure and regiochemistry of
the reaction products were unequivocally established by
mono- and bidimensional spectroscopic analysis, and mass
spectrometry.[13]
To determine the influence of steric effects on the regiose-
lectivity of the insertion reaction, ethyl phenyl propiolate 6b
was tested next. Reaction of iridium metallacycle 7 with 6b led
to a mixture of regioisomers 9Pb and 9Db (1:1 proximal/distal
1
ratio in the H NMR spectrum of the crude product). The prod-
ucts were separated by chromatography on silica gel and their
structures established on spectroscopic and analytical grounds.
Additional structural confirmation was obtained by X-ray dif-
fraction analysis of a suitable crystal of complex 9Db grown
by slow diffusion in 1,2-dichloroethane/hexane (Figure 1). In
the case of the reaction of alkyne 6b with rhodium complex 8,
Figure 1. X-ray molecular diagram of complex 9Db. Selected bond lengths
[ꢃ] and angles [8]: IrÀCl(1) 2.415(2), IrÀN(1) 2.153(5), IrÀC(14) 2.153(5); Cl(1)-
Ir-N(1) 88.5(1), C(14)-Ir-N(1) 84.8(2), C(14)-Ir-(Cl) 89.6(2). CCDC-952794 (9Db)
contain the supplementary crystallographic data for this paper. These data
can be obtained free of charge from The Cambridge Crystallographic Data
Centre via www.ccdc.cam.ac.uk/data_request/cif.
Chem. Eur. J. 2014, 20, 3831 – 3838
www.chemeurj.org
3832
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
The excellent results obtained in the alkyne insertions using
model purine metallacycles 7 and 8 prompted us to test the
reactivity and regioselectivity of the process in more complex
metallanucleosides and metallanucleotides 11 and 12 (1:1 dia-
stereomeric mixtures; Scheme 3).[10] In this manner, insertion of
Figure 2. X-ray molecular diagram of complex 10Pc. Selected bond lengths
[ꢃ] and angles [8]: Rh(1)ÀCl(1) 2.415(2), Rh(1)ÀN(51) 2.134(5), Rh(1)ÀC(2)
2.076(6); Cl(1)-Rh(1)-N(51) 88.9(1), C(2)-Rh(1)-N(51) 87.2(2), C(2)-Rh(1)-Cl(1)
87.9(2). CCDC-952795 (10Pc) contain the supplementary crystallographic
data for this paper. These data can be obtained free of charge from The
Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_re-
quest/cif.
in 1,2-dichloroethane/hexane (Figure 2). In both cases, the
presence of the 9-anthranyl group exerts a noticeable influ-
ence on the signals of the neighboring carboxylate group in
their NMR spectra. For example, the OCH3 protons of 10Pc
1
appear in the H NMR spectrum (258C) as two broad signals
(d=3.12 ppm, 2H, and 2.52 ppm, 1H) that coalesced to
a single one at d=3.09 ppm at 708C in [D6]DMSO. Additionally,
the signal of the CO ester group is hardly visible in the
13C NMR spectrum, in contrast to the strong IR absorption
band observed at 1693 cmÀ1 [14]
.
It is remarkable how the insertion of propiolates 6a and 6c
that have substituents so different in size (R1 =H in 6a to R1 =
9-anthranyl in 6c) takes place with total selectivity, leading in
both cases to the proximal regioisomers. Steric and electronic
effects,[15] among others,[16,17] have been claimed to determine
the regioselectivity of unsymmetrical alkyne insertions in met-
allacycles. Although the data reported suggest that the bulkier
the substituent in the alkyne the higher the preference to
place it next to the metal fragment,[15a] it is also true that this
conclusion has been drawn from studies just limited to inser-
tions of ethyl phenyl propiolate and ethyl 4,4,4-trifluoro-2-bu-
tynoate, with Ph and CF3 as the bulky groups. None of these
reactions was totally regioselective. This behavior is opposite
to that reported here for compounds 9Pc and 10Pc, which, in
addition, are fully regioselective processes. It is our feeling that
the control of the regioselectivity in this type of insertion is
a more complex matter than the evaluation of electronic/steric
effects of the substituents in the alkyne, and would require an
in-depth evaluation of the different steps of the process (kinet-
ic and thermodynamic effects).
Scheme 3. Insertion of alkynes in arylpurine nucleosides.
alkynes 6a–c under the above conditions led in all cases to
the expected monoinsertion products, with excellent yields
and regioselectivities identical to those found in the model re-
action (see above). Reactions with alkynes 6a and 6c pro-
duced exclusively the proximal regioisomers 13, 14, 17, and
18, whereas alkyne 6b led to proximal/distal (1:1) regioisomer-
ic mixtures 15P/15D for the iridium metallacycle 11 and (7:3)
16P/16D for the rhodium metallacycle 12 (1H NMR ratios in
the crude reaction). All the insertion products were purified by
chromatography on silica gel and their structures established
by spectroscopic and analytical data.
Insertion reactions of alkyne 6a were next attempted with
metallanucleotides 19 and 20 (1:1 diastereomeric mixtures),[10]
which afforded the expected insertion products 21 and 22
with total regioselectivity and in nearly quantitative yields. Fi-
Chem. Eur. J. 2014, 20, 3831 – 3838
www.chemeurj.org
3833
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
nally, the methodology was assayed with dinucleotide 23,
which formed the bis-insertion iridametallacycle 24 in 64%
yield of isolated product (Scheme 4). These results demon-
strate the compatibility of the reaction with the phosphate
bond.
Due to their photophysical and photochemical properties,
anthracene-derived compounds have found diverse applica-
tions as fluorescent probes.[18] In the context of our research in-
terest in the postfunctionalization of metalla-DNA fragments,
the good yields and total selectivities observed in the insertion
of bulky alkyne 6c (R1 =9-anthranyl) in both Ir and Rh metalla-
cycles 7 and 8 let us consider applying this reaction as a regio-
selective method to incorporate other fluorescent or redox
probes in a metallanucleotide. Unfortunately, attempts to ach-
ieve the reaction of model purine metallacycles 7 and 8 with
alkynes 6d (R1 =1-pyrenyl) and 6e (R1 =ferrocenyl) under the
conditions shown in Scheme 2 were fruitless, leading to the re-
covery of unreacted reagents. Heating at reflux in MeOH over-
night or carrying out the reaction in a sealed tube did not lead
either to the insertion products. It is known that alkyne inser-
tions into cyclometalated complexes are facilitated by substi-
tuting the chloride in the starting compounds with acetonitrile
to provide cationic complexes.[19] In this regard we prepared
cationic MeCN complexes 25 and 26 in excellent yields by stir-
ring metallacycles 7 and 8 with NaPF6 overnight at room tem-
perature (Scheme 5). The structure of the cationic complexes
was established by spectroscopic and analytical data and for
IrIII complex 25 by X-ray diffraction analysis (Figure 3).
The efficiency of the insertions of terminal alkynes into the
model purine cationic complexes 25 and 26 was tested first.
Stoichiometric amounts of 25 and phenylacetylene (an alkyne
that proved to be unreactive towards model complexes 7 and
Scheme 4. Insertion of alkynes in arylpurine nucleotides and dinucleotides.
8
in the preliminary assays)
yielded monoinsertion product
27 as a single regioisomer, in
90% yield of isolated prod-
uct.[20,21] Excess phenylacetylene
cleanly afforded the double in-
sertion product 28, the structure
of which was established by
spectroscopic and analytical data
and by X-ray diffraction analysis
of a suitable crystal obtained by
slow diffusion in ether (Figure 3).
The structure shows a six-mem-
bered metallacycle formed by in-
sertion of two molecules of
PhCꢀCH into the MÀC bond.[22]
The organic fragment is bonded
through the purine nitrogen
atom N(1) and has an asymmet-
rical h3 interaction with three of
the alkyne carbon atoms C(1), Figure 3. X-ray molecular diagram of compounds 25 (left) and 28 (right). Selected bond lengths [ꢃ] and angles [8]:
25: Ir(1)ÀN(6) 2.018(14), Ir(1)ÀN(7) 2.104(11), Ir(1)ÀC(27) 1.994(16); N(6)-Ir(1)-N(7) 86.4(4), C(27)-Ir(1)-N(6) 89.0(5),
C(14), and C(21). The IrÀC(21)
C(27)-Ir(1)-N(7) 77.2(5); 28: Ir(1)ÀN(1) 2.118(6), Ir(1)ÀC(1) 2.156(8), Ir(1)ÀC(14) 2.261(8), Ir(1)ÀC(21) 2.108(8), C(1)À
has the shortest bond length
C(14) 1.419(11), C(14)ÀC(21) 1.445(11), C(21)ÀC(22) 1.303(11); N(1)-Ir(1)-C(1) 87.9(3). CCDC-952794 (25), and 952797
(2.108(8) ꢃ) and the IrÀC(14) the
(28) contain the supplementary crystallographic data for this paper. These data can be obtained free of charge
longest (2.261(8) ꢃ). The C(21)À from The Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
Chem. Eur. J. 2014, 20, 3831 – 3838
www.chemeurj.org
3834
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
absorption properties of the
metallanucleobases 9Pc, 10Pc,
17, 18, and 31 relative to those
of the corresponding structurally
related complexes. UV/Vis spec-
tra of all iridium metallacycles
discussed throughout this work
are represented in Figures 4–6
and data are collected in
Tables 1 and 2.[23] As shown in
Figure 4, the insertion of the al-
kynes 6 in the model neutral
metallacycle 7 provokes a dra-
matic effect in the UV/Vis spec-
trum, with the almost complete
disappearance of the 390 nm ab-
sorption band (assignable to
metal-to-ligand charge transfer,
MLCT) of the starting compound
(see the spectra of 9Pa, 9Pb,
and 9Db). In this scenario, the
incorporation of the 9-anthranyl
moiety in the structure of the
metallacycle is clearly noticed by
the appearance of new bands in
the 360–400 nm region (see the
spectrum of the anthranyl deriv-
ative 9Pc compared with those
of 9Pa or 9Pb in Figure 4). A
similar discussion could be ap-
Scheme 5. Reactivity of cationic IrIII arylpurine derivatives with alkynes.
C(22) distance of 1.303(11) ꢃ is typical of a double bond and is
shorter than the C(21)ÀC(14) bond length (1.445(11) ꢃ) within
the h3-allyl group. The phenyl substituent on the central
carbon atom of the allyl moiety is placed anti with respect to
the Cp* on Ir.
Stoichiometric reaction of methyl propiolate 6a with cation-
ic iridametallacycle 25 led to the expected monoinsertion
product 29 as a single proximal regioisomer (54% yield of iso-
lated product; Scheme 5). Monoinserted cationic complexes
could also be obtained from the corresponding neutral ones.
As an example, 9Pa was treated with NaPF6 in acetonitrile at
room temperature to form 29 in almost quantitative yield.
Once the reaction conditions were tuned, the reactivity of
cationic complex 25 towards alkynes 6d (R1 =1-pyrenyl) and
6e (R1 =ferrocenyl) was tested. In both cases the insertion
products 30 and 31 were obtained in good yields and as
single proximal isomers (Scheme 6). Pure products were sepa-
rated by chromatography on silica gel and their structures es-
tablished on spectroscopic and analytical grounds.
Scheme 6. Synthesis of cationic IrIII arylpurine derivatives 30 and 31.
plied to the UV/Vis spectra in Figure 5A and B and data in
Table 2. Thus, relative to parent compound 11, nucleosides 13
and 15 show the almost complete disappearance of the MLCT
band (Figure 5A). The same trend is observed for nucleotides
21 and dinucletotides 24 compared to 19 and 23, respectively
(Figure 5B). Again, the incorporation of the 9-anthranyl moiety
in the structure of the metallacycle is shown by the appear-
ance of new absorbances in the 360–400 nm region (see the
UV/Vis spectrum of 17 in Figure 5A).
The usefulness of the above methodology when applied to
the labeling of unnatural oligonucleotides is related to the ex-
istence of a simple technique to efficiently detect the success-
ful incorporation of the label after the reaction. In this regard
a study of the UV/Vis spectra was carried out to establish the
effect of the presence of anthracene and pyrene labels on the
A similar behavior is observed for the cationic complexes 25,
29, and 31, in which the insertion of the alkyne unit also pro-
Chem. Eur. J. 2014, 20, 3831 – 3838
www.chemeurj.org
3835
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 4. UV/Vis spectra of metallacycles 7 and 9a–c in CH2Cl2 (ca. 10À5 m).
Table 1. UV/Vis main absorptions of metallacycles 7 and 9a–c.
Complex[a]
l [nm]
392, 320, 307
10À3 e [mÀ1 cmÀ1
]
7
8, 19, 21
26, 30
5
18
7, 8, 6, 20
9Pa
9Pb
9Db
9Pc
293, 268
273
274
396, 376, 356, 293
[a] In CH2Cl2 (ca. 10À5 m).
Figure 5. A) UV/Vis spectra of IrIII nucleosides 11, 13, 15, and 17. B) UV/Vis
spectra of IrIII nucleotides 19 and 21 and dinucleotides 23 and 24. Spectra
registered in CH2Cl2 (ca. 10À5 m).
Table 2. UV/Vis main absorptions of IrIII nucleosides, nucleotides and cat-
ionic arylpurines.
Complex[a]
l [nm]
400, 321, 307
10À3 e [mÀ1 cmÀ1
]
11
13
15D
15P
17
6, 18, 19
18, 19
292, 268
273
18
21
279
396, 375, 356, 339
396, 320, 307
291, 264
11, 12, 8, 6, 20
8, 20, 23
18, 20
19
21
23
24
394, 320, 307
291, 266
14, 35, 40
32, 35
25
29
350, 318, 306
442, 340
11, 22, 19
2, 6
Figure 6. UV/Vis spectra of metallacycles 25, 29, and 31 in CH2Cl2 (ca.
10À5 m).
31
354, 342, 280
27, 24, 41
[a] In CH2Cl2 (ca. 10À5 m).
tion potentials between 1.2 and 1.7 V, is irreversible and may
be attributable to a metal-centered IrIV/IrV process or to
a ligand-centered process. The model rhodium complexes
studied exhibit similar redox behavior but, as expected, the ox-
idation takes place at higher potentials than for the iridium
counterparts. Also, the alkyne insertion products generally
show oxidation waves at higher potential values than the cy-
clometalated starting materials.
duces a dramatic decrease of all the absorption bands (see, for
example, the UV/Vis spectra of 25 and 29 in Figure 6, Table 2).
Now, the incorporation of the pyrenyl group in the structure
can be easily detected by the appearance of three new intense
bands at 280, 342, and 354 nm (see the spectrum of 29 relative
to the pyrenyl derivative 31 in Figure 6, Table 2).
Alternatively, the easiest way to determine the effect of the
incorporation of a redox label in the prepared metallacycles is
the study of their redox properties. In this regard, the electro-
chemistry of the neutral and cationic cyclometalated products
made was carried out. Full data and cyclic voltammograms are
shown in the Supporting Information. All iridium complexes
studied show two different redox events. The first one is a re-
versible oxidation at potentials between 0.9 and 1.3 V (assigna-
ble to a metal-centered electron-transfer process) and in the
range of IrIII/IrIV reported values.[24] The second step, at oxida-
The effect of the incorporation of a ferrocenyl substituent
was evaluated by studying the redox properties of cationic
complex 30 (Figure 7, Table 3). Model cationic complex 25
shows an oxidation wave at 1.59 V, hardly visible in the prod-
uct of insertion of methyl propiolate 29. The incorporation of
the ferrocenyl moiety in 30 is clearly detectable by the appear-
ance of a reversible oxidation wave assignable to the ferro-
+
1
cene/ferrocenium couple (Fc/Fc ; E = =0.47 V) together with
2
1
a new quasireversible wave at (E = =0.89 V), very likely due to
2
Chem. Eur. J. 2014, 20, 3831 – 3838
www.chemeurj.org
3836
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
cleotides or arylpurine/pyrimidine-containing DNA fragments
will be attempted in the near future in our laboratories.
Acknowledgements
Financial support from the Spanish MICINN Projects CTQ-2010-
20414-C02-01/BQU and Consolider Ingenio 2010 (CSD2007-
00006) and from the Comunidad de Madrid (grant S2009/PPQ-
1634, AVANCAT) is acknowledged. A generous loan of MCl3
(M=Ir, Rh) from Johnson-Matthey is gratefully acknowledged.
Keywords: iridium · labeling · metallacycles · nucleobases ·
rhodium
[1] a) D. A. Megger, J. Mꢄller, in Nucleic Acids with Purine- and Pyrimidine-
Derived Nucleosides, Vol. 10 (Eds.: A. Sigel, H. Sigel), RSC, Cambridge,
2012, pp. 295–317; b) A. A. Henry, F. E. Romesberg, Curr. Opin. Chem.
Biol. 2003, 7, 727–733.
Figure 7. Cyclic voltammetry of compounds 25, 29, and 30 in CH2Cl2 (ca.
10À3 m) with Ag/AgCl (3m) as reference.
[2] See for example: a) S. Matsuda, F. E. Romesberg, J. Am. Chem. Soc.
2004, 126, 14419–14427; b) A. M. Leconte, S. Matsuda, G. T. Hwang, F. E.
Romesberg, Angew. Chem. 2006, 118, 4432–4435; Angew. Chem. Int. Ed.
2006, 45, 4326–4329; c) A. M. Leconte, S. Matsuda, F. E. Romesberg, J.
Am. Chem. Soc. 2006, 128, 6780–6781; d) G. T. Hwang, Y. Hari, F. E. Ro-
mesberg, Nucleic Acids Res. 2009, 37, 4757–4763; e) T. Lavergne, D. A.
Malyshev, F. E. Romesberg, Chem. Eur. J. 2012, 18, 1231–1239.
[3] For examples of metal-mediated base pairs, see: a) D. A. Megger, N.
Megger, J. Mꢄller, in Interplay Between Metal ions and Nucleic Acids,
Vol. 10 (Eds: A. Sigel, H. Sigel, R. K. O. Sigel), Springer, Netherlands,
2012, pp. 295–317; b) K. Petrovec, B. J. Ravoo, J. Mꢄller, Chem.
Commun. 2012, 48, 11844–11846; c) D. A. Megger, C. F. Guerra, F. M.
Bickelhaupt, J. Mꢄller, J. Inorg. Biochem. 2011, 105, 1398–1404; d) S. Jo-
hannsen, N. Megger, D. Bçhme, R. K. O. Sigel, J. Mꢄller, Nat. Chem.
2010, 2, 229–234; e) G. H. Clever, M. Shionoya, Coord. Chem. Rev. 2010,
254, 2391–2402; f) G. H. Clever, C. Kaul, T. Carell, Angew. Chem. 2007,
119, 6340–6350; Angew. Chem. Int. Ed. 2007, 46, 6226–6236; g) K.
Tanaka, M. Shionoya, Coord. Chem. Rev. 2007, 251, 2732–2742.
[4] a) A. A. Henry, A. G. Olsen, S. Matsuda, C. Yu, B. H. Geierstanger, F. E. Ro-
mesberg, J. Am. Chem. Soc. 2004, 126, 6923–6931; b) Y. J. Seo, S. Matsu-
da, F. E. Romesberg, J. Am. Chem. Soc. 2009, 131, 5046–5047; c) D. Ma-
lyshev, Y. J. Seo, P. Ordoukhanian, F. E. Romesberg, J. Am. Chem. Soc.
2009, 131, 14620–14621; d) D. A. Malyshev, K. Dhami, H. T. Quach, T.
Lavergne, P. Ordoukhanian, A. Torkamani, F. E. Romesberg, Proc. Natl.
Acad. Sci. USA 2012, 109, 12005–12010; e) T. Lavergne, M. Degardin,
D. A. Malyshev, H. T. Quach, K. Dhami, P. Ordoukhanian, F. E. Romesberg,
J. Am. Chem. Soc. 2013, 135, 5408–5419.
[5] a) A. A. Henry, C. Yu, F. E. Romesberg, J. Am. Chem. Soc. 2003, 125,
9638–9646; b) S. Matsuda, J. D. Fillo, A. A. Henry, P. Rai, S. J. Wilkens,
T. J. Dwyer, B. H. Geierstanger, D. E. Wemmer, P. G. Schultz, G. Spraggon,
F. E. Romesberg, J. Am. Chem. Soc. 2007, 129, 10466–10473; c) D. A. Ma-
lyshev, D. A. Pfaff, S. I. Ippoliti, G. T. Hwang, T. J. Dwyer, F. E. Romesberg,
Chem. Eur. J. 2010, 16, 12650–12659; d) K. Betz, D. A. Malyshev, T. Lav-
ergne, W. Welte, K. Diederichs, T. J. Dwyer, P. Ordoukhanian, E. Romes-
berg, A. Marx, Nat. Chem. Biol. 2012, 8, 612–614.
Table 3. Electrochemical data for ferrocenyl complexes, 25, 29 and 30 (in
CH2Cl2, with 3m Ag/AgCl as reference).[a]
1
1
E =
2
Complex
Epa1
Epa
2
E =
Epa2
Epc2
2
25
29
30
1.59
1.64
0.57
–
–
0.38
–
–
0.47
–
–
0.95
–
–
0.84
–
–
0.89
[a] Values are given in volts.
the oxidation of the IrIII. The presence of the ferrocenyl group
shifts the oxidation potential of the metal to lower values rela-
tive to the model cationic complex 25. This behavior is similar
to that of the neutral complexes, which show a reversible oxi-
dation wave for the metal (see the Supporting Information).
Conclusion
The reactions of neutral or cationic IrIII and RhIII derivatives of
phenyl purine nucleobases with unsymmetrical alkynes pro-
duce new metallacycles with high proximal regioselectivity. Al-
kynes combining electronic and steric effects show a general
tendency to place the bulkier group away from the metal
center. The predictability of these reactions allows for the in-
corporation of either photoactive (anthracene or pyrene)
or electroactive (ferrocene) labels in the nucleotide or nucleo-
side moiety. The detection of these labels can be easily ach-
ieved by UV/Vis spectroscopy or electrochemistry, as has
been demonstrated by comparison with the parent (unlabeled)
compounds.
[6] a) S. G. Srivatsan, Y. Tor, J. Am. Chem. Soc. 2007, 129, 2044–2053; b) T.
Mitsui, M. Kimoto, Y. Harada, S. Yokoyama, I. Hirao, J. Am. Chem. Soc.
2005, 127, 8652–8658; c) R. Kawai, M. Kimoto, S. Ikeda, T. Mitsui, M.
Endo, S. Yokoyama, I. Hirao, J. Am. Chem. Soc. 2005, 127, 17286–17295;
d) T. Ohtsuki, M. Kimoto, M. Ishikawa, T. Mitsui, I. Hirao, S. Yokoyama,
Proc. Natl. Acad. Sci. USA 2001, 98, 4922–4925; e) Y. Tor, P. B. Dervan, J.
Am. Chem. Soc. 1993, 115, 4461–4467; f) J. D. Bain, C. Switzer, A. R.
Chamberlin, S. A. Benner, Nature 1992, 356, 537–539.
[7] for recent articles, see: a) R. W. Sinkeldam, N. J. Greco, Y. Tor, Chem. Rev.
2010, 110, 2579–2619; b) M. Kimoto, T. Mitsui, R. Yamashige, A. Sato, S.
Yokoyama, I. Hirao, J. Am. Chem. Soc. 2010, 132, 15418–15426; c) B.
Man, D. Chan, H. Yang, S. Ang, F. Yang, S. Yan, C. Ho, P. Wu, C. Che, C.
Leung, D. Ma, Chem. Commun. 2010, 46, 8534–8536; d) D. S. Chan, H.
The methodology reported could have applications as a site-
specific labeling technique for oligonucleotides with easily de-
tectable orthogonal markers. The strategy, based on the selec-
tive postfunctionalization of oligonucleotides (metalation of
specific modified bases within the chain and subsequent incor-
poration of the label), differs from those reported for modified
DNA fragments that rely on the assembly of labeled units.
Translation of these results to the functionalization of oligonu-
Chem. Eur. J. 2014, 20, 3831 – 3838
www.chemeurj.org
3837
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Lee, C. Che, C. Leung, D. Ma, Chem. Commun. 2009, 7479–7481; e) C.
Chiang, C. Huang, C. Liu, H. Chang, Anal. Chem. 2008, 80, 3716–3721;
f) A. Ono, H. Togashi, Angew. Chem. 2004, 116, 4400–4402; Angew.
Chem. Int. Ed. 2004, 43, 4300–4302; g) L. H. Thoresen, G.-S. Jiao, W. C.
Haaland, M. L. Metzker, K. Burgess, Chem. Eur. J. 2003, 9, 4603–46104.
[8] See, for example: a) Y. J. Seo, D. A. Malyshev, T. Lavergne, P. Ordoukhani-
an, F. E. Romesberg, J. Am. Chem. Soc. 2011, 133, 19878–19888; b) S. P.
Liu, S. H. Weisbrod, Z. Tang, A. Marx, E. Scheer, A. Erbe, Angew. Chem.
2010, 122, 3385–3388; Angew. Chem. Int. Ed. 2010, 49, 3313–3316;
c) A. M. Leconte, M. P. Patel, L. E. Sass, P. McInerney, M. Jarosz, L. Kung,
J. L. Bowers, P. R. Buzby, J. W. Efcavitch, F. E. Romesberg, Angew. Chem.
2010, 122, 6057–6060; Angew. Chem. Int. Ed. 2010, 49, 5921–5924;
d) J. W. Liu, Z. H. Cao, Y. Lu, Chem. Rev. 2009, 109, 1948–1998; e) E. M.
Nolan, S. J. Lippard, Chem. Rev. 2008, 108, 3443–3480; f) Y. J. Seo, I. A.
Berlatzky, A. Rouvinski, S. Ben-Yehuda, Proc. Natl. Acad. Sci. U.S.A 2008,
105, 14136–14140; g) A. A. Voityuk, J. Phys. Chem. B 2006, 110, 21010–
21013.
[18] a) Z. Fei, d.-R. Zhu, X. Yang, L. Meng, Q. Lu, W. H. Ang, R. Scopelliti, C. G.
Hartinger, P. J. Dyson, Chem. Eur. J. 2010, 16, 6473–6481; b) M. J. Plater,
I. Greig, M. H. Helfrich, S. H. Ralston, J. Chem. Soc. Perkin Trans. 1 2001,
2553–2559; c) Q.-X. Liu, F.-B. Xu, Q.-S. Li, H.-B. Song, Z.-Z. Zhang, Orga-
nometallics 2004, 23, 610–614; d) B. R. Willeford, Jr., E. O. Fischer, J. Or-
ganomet. Chem. 1965, 4, 109–113; e) P. D. Edirisinghe, J. F. Moore, W. F.
Calaway, I. V. Veryovkin, M. J. Pellin, L. Hanley, Anal. Chem. 2006, 78,
5876–5883.
[19] a) F. Maassarani, M. Pfeffer, G. Leborgne, Organometallics 1987, 6, 2029–
2043; b) D. L. Davies, O. Al-Duaij, J. Fawcett, K. Singh, Organometallics
2010, 29, 1413–1420.
[20] The regioselectivity is similar to that observed in insertions of phenyl-
acetylene in related cationic complexes. See ref. [19b].
[21] Other reactions with terminal alkynes: a) R. P. Korivi, Y.-C. Wu, C.-H.
Cheng, Chem. Eur. J. 2009, 15, 10727–10731; b) K. Cheng, B. Yao, J.
Zhao, Y. Zhang, Org. Lett. 2008, 10, 5309–5312; c) C.-C. Liu, R. P. Korivi,
C.-H Cheng, Chem. Eur. J. 2008, 14, 9503–9506; d) R. P. Korivi, C.-H.
Cheng, Org. Lett. 2005, 7, 5179–5182; e) M. R. Meneghetti, M. Grellier,
M. Pfeffer, J. Fischer, Organometallics 2000, 19, 1935–1939.
[9] a) Modified Nucleosides in Biochemistry Biotechnology and Medicine (Ed.:
P. Herdewijn), Wiley-VCH, Weinheim, 2008.
[10] M. Martꢀn-Ortꢀz, M. Gꢁmez-Gallego, C. Ramirez de Arellano, M. A. Sierra,
Chem. Eur. J. 2012, 18, 12603–12608.
[22] This result is not surprising as insertions of phenylacetylene in half-
sandwich cyclometalated complexes are generally very fast processes
that lead frequently to di- and even polyinsertion products together
with other byproducts. See ref. [19].
[23] For a similar study of Rh complexes, see the Supporting Information.
[24] a) J. D. Blakemore, N. D. Schley, D. Balcells, J. F. Hull, G. W. Olack, C. D. In-
carvito, O. Eisenstein, G. W. Brudvig, R. H. Crabtree, J. Am. Chem. Soc.
2010, 132, 16017–16029; b) A. B. Tamayo, B. D. Alleyne, P. I. Djurovich,
S. Lamansky, I. Tsyba, N. Ho, R. Bau, M. E. Thompson, J. Am. Chem. Soc.
2003, 125, 7377–7387; c) S. Jung, Y. Kang, H.-S. Kim, Y.-S. Kim, C.-L. Lee,
J.-J. Kim, S.-K. Lee, S.-K. Kwon, Eur. J. Inorg. Chem. 2004, 3415–3423;
d) V. V. Grushin, N. Herron, D. LeCloux, W. J. Marshall, V. A. Petrov, Y.
Wang, Chem. Commun. 2001, 1494–1495 and references therein.
[11] The established strategies to introduce labels in DNA fragments are
usually based on the assembly of already modified units, see: a) Y. N.
Teo, E. T. Kool, Chem. Rev. 2012, 112, 4221–4245; b) R. Varghese, H.-A.
Wagenknecht, Chem. Commun. 2009, 2615–2624; c) V. L. Malinovskii, D.
Wenger, R. Hꢅner, Chem. Soc. Rev. 2010, 39, 410–422.
[12] Y. Boutadla, D. L. Davies, R. C. Jones, K. Singh, Chem. Eur. J. 2011, 17,
3438–3448, and references therein.
[13] See the Supporting Information for details.
[14] In the case of complex 9Pc the OCH3 appears as three broad signals at
1
d=3.87, 3.10, and 2.48 ppm in the H NMR spectrum at 258C.
[15] a) L. Li, Y. Jiao, W. W. Brennessel, W. D. Jones, Organometallics 2010, 29,
4593–4605; b) Y. Boutadla, D. L. Davies, O. Al-Duaij, J. Fawcett, R. C.
Jones, K. Singh, Dalton Trans. 2010, 39, 10447–10457.
[16] a) N. Guimond, K. Fagnou, J. Am. Chem. Soc. 2009, 131, 12050–12051;
b) D. R. Stuart, M. Bertrand-Laperle, K. M. N. Burgess, K. Fagnou, J. Am.
Chem. Soc. 2008, 130, 16474–16475.
Received: October 18, 2013
Revised: December 11, 2013
[17] J. M. Neely, T. Rovis, J. Am. Chem. Soc. 2013, 135, 66–69.
Published online on February 23, 2014
Chem. Eur. J. 2014, 20, 3831 – 3838
www.chemeurj.org
3838
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim