Determination of orientational isomerism in rhodium(II) metallopeptides by pyrene fluorescence

Article abstract of DOI:10.1039/c2ob26667a

Rhodium(II) metallopeptides display useful secondary structure, self-assembly, and catalytic activity. The bis-peptide complexes exhibit subtle orientational isomerism that affects function, but is challenging to characterize. We report that pyrene excime

Full text of DOI:10.1039/c2ob26667a

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Organic &
Biomolecular
Chemistry
Cite this: Org. Biomol. Chem., 2012, 10, 8203
www.rsc.org/obc
COMMUNICATION
Determination of orientational isomerism in rhodium(II) metallopeptides by
pyrene fluorescence†
Ramya Sambasivan and Zachary T. Ball*
Received 24th August 2012, Accepted 10th September 2012
DOI: 10.1039/c2ob26667a
Rhodium(II) metallopeptides display useful secondary struc-
ture, self-assembly, and catalytic activity. The bis-peptide
complexes exhibit subtle orientational isomerism that affects
function, but is challenging to characterize. We report that
pyrene excimer fluorescence measurements provide a con-
clusive proof of isomeric structure.
Dirhodium centers coordinate to carboxylate side chains to
produce kinetically inert rhodium(II) metallopeptides.1 When
Rh2(tfa)4 reacts with bis-carboxylate peptides, bis-peptide com-
plexes are formed which provide a simple method for assem-
bling peptides with defined orientation.2 The bis-peptide
complexes can generate a chiral environment that allows for
enantioselective reactions of diazo compounds.3 The synthesis of
bis-peptide complexes results in the formation of isomeric paral-
Fig. 1 Metallopeptide catalysts for asymmetric diazo reactions. L16 =
KZTDAALDLK, L21 = KZNDAAIDAKZ. The term “isoX” is an arbi-
trary isomer designation based on HPLC run time. For a complete depic-
tion of catalyst structure, see Fig. 2. Synthesis of 23 and 3b4 has been
reported previously.
lel and antiparallel orientations of the peptide ligands (Fig. 2).
These isomers are expected to have differing catalytic, biological
and supramolecular properties, yet we struggled to determine the
orientational structure. In this report, we present a method, based
on pyrene excimer fluorescence, to conclusively determine the
orientational structure of rhodium(II) metallopeptides.
Rhodium(II) metallopeptides of the general structure Rh2(pep-
tide)2 have proven to be useful and general platforms for the
development of asymmetric catalysts, yet important structural
questions remained that limit future design efforts. We reported3
that a metallopeptide, previously named Rh2(L21)2-isoB, cata-
lyzes enantioselective silane insertion reactions of diazoacetate
derivatives (Fig. 1, 2).3 An alternative peptide sequence (L16)
was found to affect enantioselective cyclopropanation with
phenyl diazodoacetate derivatives.4 The success of metallopep-
tides based on peptide L16 extends to styrenyl diazoacetate
derivatives as well, allowing access to useful vinylcyclopropane
products (Fig. 1, 5), indicating the generality of peptide-based
ligands. In all cases studied, the orientational isomerism is
important for enantioselectivity. For a given peptide, one metal-
lopeptide isomer was typically far more selective. However, we
were unable to establish the orientational structure and named
the metallopeptides arbitrarily (A and B) based on HPLC reten-
tion time, severely restricting our ability to design peptide
ligands for asymmetric transition-metal catalysis.5–9
On a fundamental level, the origins of selectivity in the two
isomers are likely quite different. Although both metallopeptides
have C2 symmetry, the parallel isomer has an axis of symmetry
that bisects the Rh–Rh bond, rendering the two rhodium atoms
chemically equivalent (Fig. 2a). The antiparallel isomer, on the
other hand, has a C2 axis through the Rh atoms, rendering the
rhodium atoms non-equivalent (Fig. 2b). We have modeled the
two isomers, assuming an α-helical backbone based on CD spec-
troscopy (see ESI†) and previous studies with metalation of car-
boxylate residues with i, i + 4 spacing.2 In this model, the
parallel isomer contains two identical rhodium sites with a clear
steric differentiation of approaching reagents. On the other hand,
the antiparallel isomer model has two very different rhodium
sites: one (top) surrounded by significant steric bulk and a
second (bottom) with a binding site that is quite open and
unimpeded.
Department of Chemistry MS60, Rice University, 6100 Main St,
Houston, TX 77005, USA. E-mail: zb1@rice.edu; Tel: +1 713 348 6159
†Electronic supplementary information (ESI) available: Experi-
mental procedures and complete characterization data. See DOI:
10.1039/c2ob26667a
We examined several methods that might shed light on the
issue of orientational isomerism. 2D NMR spectra of the two
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 8203–8206 | 8203

View Article Online
Fig. 3 31P NMR spectra of 3,3′,3′′-phosphinidynetris(benzenesulfonic
acid) trisodium salt (TPPTS, 0.9 equiv) in the presence of Rh2(L21)2-para
(top) and Rh2(L21)2-anti (bottom). The observed peaks correspond to,
monoadduct Rh2(L21)2(tppts)1 (−41.6 ppm), and bis-adduct
Rh2(L21)2(tppts)2 (−17.8 ppm). The peak at +33.6 ppm corresponds to
the free phosphine oxide. Spectra measured in CD3OD relative to an
external standard (85% H3PO4).
Fig. 2 Parallel (a) and antiparallel (b) structures of the bis-metallopep-
tides Rh2(L21)2. (top) Line drawings illustrating the axis of symmetry
and (bottom) models of catalyst structure (rendered using PyMol).
Amino acids are color-coded. Peptide sequence L21 = KZNDAAIDAKZ.
isomers are very similar (see ESI†). We hoped that the two
different chemical environments of the antiparallel isomer could
be visualized by NMR analysis of axial ligands. Unfortunately,
the 31P NMR spectra of isomeric metallopeptides in the presence
of a phosphine (TPPTS) were also indistinguishable, containing
three peaks consistent with a mono-axial adduct, a bis-axial
adduct, and free TPPTS phosphine oxide (Fig. 3).10 We could
not discern peaks corresponding to different ligand environments
even at low temperature. We have been unable to crystallize
peptide-rhodium adducts.
The pyrene excimer fluorescence is an established tool to
study peptide assembly.11–15 For supramolecular peptide assem-
blies, orientational questions are difficult to address conclusively
due to orientational interconversion and the possibility for
pyrene groups to alter thermodynamic preferences; extending the
use of pyrene excimer fluorescence to rhodium coordination
assemblies removes many of these concerns. These rhodium
metallopeptide isomers do not interconvert; under forcing
Fig. 4 Fluorescence emission spectra (λex = 340 nm) of metallopep-
tides (1.0 μM) in aq methanol. Parallel isomers (blue) and antiparallel
isomers (red). (a) Rh2(L21*)2-para and Rh2(L21*)2-anti in 70% aq
methanol. (b) Rh2(L16*)2-para and Rh2(L16*)2-anti in 90% aq metha-
nol. (c) Rh2(L13*)2-para and Rh2(L13*)2-anti in 90% aq methanol.
L13
=
KZFDGALDAKZ, L16
=
KZTDAALDLK, L21
=
KZNDAAIDAKZ.
conditions (pH > 8 or T > 70 °C), only decomposition is
observed. Pyrene dimerization results in a unique excimer fluor-
escence above 450 nm. Attaching pyrene groups to the peptide
N-terminus would form pyrene-containing metallopeptides as
parallel and antiparallel isomers, of which only the parallel
8204 | Org. Biomol. Chem., 2012, 10, 8203–8206
This journal is © The Royal Society of Chemistry 2012

View Article Online
Table 1 Comparison of ee values for catalytic diazo decomposition
isomer could achieve a conformation allowing pyrene dimeriza-
tion (see Fig. 4). Alternative approaches involving chemical
functionalization of a parent metallopeptide were discarded,
because chemical modification of rhodium(II) metallopeptides,
though possible,16 is challenging. Cysteine-based cross-
linking17,18 is another approach to study orientational questions,
but the potential for disruptive rhodium–thiol reactivity pre-
vented its use here. Importantly, asymmetric catalysis provides a
means of correlating the structures of pyrene-containing variants
with that of the parent metallopeptides.
2
3a
2
3a
We synthesized pyrene-containing versions of the optimized
peptide (L21) and two other sequences (L13, L16) from our
initial catalyst library by standard solid phase methods, capping
the N-terminus with 1-pyrenebutyric acid to afford pyrene-
containing variants L13*, L16*, and L21*. Metalation using
previously developed procedures3 afforded the orientational
isomers Rh2(L21*)2-para and Rh2(L21*)2-anti, which were then
separated by preparative HPLC. Analogous complexes were syn-
thesized with the ligands L13* and L16*. As with the parent
peptides, the formation of pyrene-containing metallopeptides
was non-selective with respect to orientational isomer formation.
This observation is consistent with our general understanding of a
kinetic metalation reaction that does not allow product interconver-
sion; presumably, pyrene dimerization does not template the
second metalation event in the reaction solvent, trifluorothanol.
When the fluorescence spectra of the metallopeptides derived
from L21* were obtained in 70% aqueous methanol, one isomer
exhibits a strong excimer fluorescence band near 470 nm, allow-
ing its assignment as the parallel isomer, Rh2(L21*)2-para
(Fig. 4a). The other (antiparallel) isomer lacked the excimer
band and exhibited a fluorescence signature similar to 1-pyrene-
butyric acid. The intensity of the excimer emission in this case is
striking, because organic cosolvents such as methanol
(we employ 70% aqueous methanol for solubility) typically
result in large diminution of excimer fluorescence. The excimer
fluorescence observed serves as evidence for a strongly pre-orga-
nized pyrene dimerization.
The metallopeptides derived from the other ligands, L13* and
L16*, had even more limited solubility, and we were only able
to obtain fluorescence spectra in 90% aqueous methanol.
Increasing the organic co-solvent predictably led to a decrease in
the intensity of the excimer band observed,‡ but sufficient signal
was observed to assign the orientational structure of these pep-
tides as well (Fig. 4b and c).
To establish the orientational structure of the actual, pyrene-
free catalysts, we compared enantioselectivity in two diazo
decomposition reactions catalysed by the various catalysts
(Table 1).§ Some variation in selectivity was expected, even
though the pyrene group is far from the rhodium center. The ee
values for reactions with the L21 ligand were largely invariant
upon pyrene incorporation, allowing us confidently to dis-
tinguish the parallel isomer (previously labelled “isoA” based on
HPLC elution order) and the antiparallel isomer. In the case of
L16, ee values similarly identify “isoA” as the parallel structure.
In the case of L13, the ee values do change meaningfully upon
incorporation of pyrene. We tentatively assign “isoA” as the par-
allel structure in this case as well, based on available data. The
change in ee observed with L13 is a potential limitation of this
method.
L21
L21*
L16
L16*
L13
L13*
para (isoA)
para
para (isoA)
para
para (isoA)
para
54
55
60
60
71
56
−12
−23
76
65
74
anti (isoB)
anti
anti (isoB)
anti
anti (isoB)
anti
92
92
71
66
73
42
52
64
46
41
45
12
57
a Comparison of enantioselectivity for reactions catalysed by isomeric
metallopeptides, Rh2(Lxx)2-para and Rh2(Lxx)2-anti. See Fig. 4 for
peptide sequences.
Most surprising to us, the optimal catalyst for silane insertion
turned out to be an antiparallel isomer, Rh2(L21)2-anti (92%
ee); the parallel isomer, Rh2(L21)2-para, was significantly less
selective (52% ee). Given the symmetry and modelling consider-
ations, it is striking that the most selective catalyst found in our
screen of >70 catalysts (a) contains two different types of catalyst
sites and (b) appears from modeling studies to contain an easily
accessible catalyst site with no apparent means of controlling
enantioinduction. The structure proof afforded by pyrene fluor-
escence lends credence to the value of high-throughput screening
methods for the discovery of asymmetric catalysts, which
remains a largely empirical exercise. In this case, our intuition
had led us to assume a parallel structure for the most selective
catalysts, and our efforts to design new catalysts have been based
on these (now clearly incorrect) assumptions. This study does
provide a structural foundation for new ligand design, and pro-
vides a tool to study metal-based peptide assembly.
Transition-metal-mediated assembly of peptides is a well-
established field.19–26 The size and stoichiometry of these assem-
blies are often effectively addressed by known analytical
methods, but the precise orientation of peptide chains is often
established only indirectly by inference.27,28 We believe this
work further demonstrates the value of pyrene excimer fluor-
escence to probe metal-mediated assembly of peptides and
proteins.
This work was supported by the NSF CAREER program
(CAREER, CHE-1055569) and the Robert A. Welch Foundation
research grant C-1680. We acknowledge Angel Marti and
Nathan Cook for assistance with the fluorescence measurements.
Notes and references
‡Pyrene excimer fluorescence is not completely general: we have found
pyrene-peptide-rhodium complexes which are unsuitable for analysis
because they are either insoluble or inseparable by preparative HPLC.
§Order of elution in RP-HPLC is not a reliable means of correlating the
structure of parent metallopeptides with their pyrene-containing variants.
In some cases, order of elution for the isomers changes upon pyrene
incorporation.
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 8203–8206 | 8205

View Article Online
1 A. N. Zaykov, K. R. MacKenzie and Z. T. Ball, Chem.–Eur. J., 2009, 15,
8961–8965.
2 A. N. Zaykov, B. V. Popp and Z. T. Ball, Chem.–Eur. J., 2010, 16, 6651–
6659.
3 R. Sambasivan and Z. T. Ball, J. Am. Chem. Soc., 2010, 132, 9289–9291.
4 R. Sambasivan and Z. T. Ball, Angew. Chem., Int. Ed., 2012, 51, 8568–
8572.
5 A. Agarkov, S. J. Greenfield, T. Ohishi, S. E. Collibee and
S. R. Gilbertson, J. Org. Chem., 2004, 69, 8077–8085.
6 S. R. Gilbertson, G. Chen and M. McLoughlin, J. Am. Chem. Soc., 1994,
116, 4481–4482.
16 A. N. Zaykov and Z. T. Ball, Chem. Commun., 2011, 47, 10927–10929.
17 B. Bilgiçer and K. Kumar, Tetrahedron, 2002, 58, 4105–4112.
18 J. D. Steinkruger, G. J. Bartlett, E. B. Hadley, L. Fay, D. N. Woolfson
and S. H. Gellman, J. Am. Chem. Soc., 2012, 134, 2626–2633.
19 O. Torres, D. Yüksel, M. Bernardina, K. Kumar and D. Bong, ChemBio-
Chem, 2008, 9, 1701–1705.
20 J. Hong, O. A. Kharenko, M. V. Tsurkan and M. Y. Ogawa, Bioinorganic
Chemistry, American Chemical Society, 2009, vol. 1012, pp. 167–182.
21 M. M. Pires and J. Chmielewski, J. Am. Chem. Soc., 2009, 131, 2706–
2712.
22 A. K. Sharma, S. T. Pavlova, J. Kim, D. Finkelstein, N. J. Hawco,
N. P. Rath, J. Kim and L. M. Mirica, J. Am. Chem. Soc., 2012, 134,
6625–6636.
7 S. R. Gilbertson, S. E. Collibee and A. Agarkov, J. Am. Chem. Soc.,
2000, 122, 6522–6523.
8 D. Coquière, J. Bos, J. Beld and G. Roelfes, Angew. Chem., Int. Ed.,
2009, 48, 5159–5162.
23 M. R. Ghadiri and C. Choi, J. Am. Chem. Soc., 1990, 112, 1630–
1632.
9 R. P. Megens and G. Roelfes, Chem.–Eur. J., 2011, 17, 8514–8523.
10 F. P. Pruchnik, R. Starosta, P. Smoleński, E. Shestakova and P. Lahuerta,
Organometallics, 1998, 17, 3684–3689.
11 H. Zhu and F. D. Lewis, Bioconjugate Chem., 2007, 18, 1213–1217.
12 S. Thurley, L. Röglin and O. Seitz, J. Am. Chem. Soc., 2007, 129,
12693–12695.
24 J. Dong, J. M. Canfield, A. K. Mehta, J. E. Shokes, B. Tian,
W. S. Childers, J. A. Simmons, Z. Mao, R. A. Scott, K. Warncke and
D. G. Lynn, Proc. Natl. Acad. Sci. U. S. A., 2007, 104, 13313–13318.
25 M. V. Tsurkan and M. Y. Ogawa, Inorg. Chem., 2007, 46, 6849–6851.
26 K. Suzuki, H. Hiroaki, D. Kohda, H. Nakamura and T. Tanaka, J. Am.
Chem. Soc., 1998, 120, 13008–13015.
13 H. Mihara, J. Hayashida, H. Hasegawa, H. I. Ogawa, T. Fujimoto and
N. Nishino, J. Chem. Soc., Perkin Trans. 2, 1997, 517–522.
14 M. Goedeweeck, M. Van der Auweraer and F. C. De Schryver, J. Am.
Chem. Soc., 1985, 107, 2334–2341.
27 Z. Chen, F. Vohidov, J. M. Coughlin, L. J. Stagg, S. T. Arold,
J. E. Ladbury and Z. T. Ball, J. Am. Chem. Soc., 2012, 134, 10138–
10145.
28 K. Maloney, D. Shnek, D. Sasaki and F. Arnold, Chem. Biol., 1996, 3,
185–192.
15 C. Garcia-Echeverria, J. Am. Chem. Soc., 1994, 116, 6031–6032.
8206 | Org. Biomol. Chem., 2012, 10, 8203–8206
This journal is © The Royal Society of Chemistry 2012