Combinatorial design of multimeric chelating peptoids for selective metal coordination

Article abstract of DOI:10.1039/c9sc01068h

Current methods for metal chelation are generally based on multidentate organic ligands, which are generated through cumbersome multistep synthetic processes that lack flexibility for systematically varying metal-binding motifs. Octadentate ligands incorporating hydroxypyridinone or catecholamide binding moieties onto a spermine scaffold are known to display some of the highest affinities towards f-elements. Enhancing binding affinity for specific lanthanide or actinide ions however, necessitates ligand architectures that allow for modular and high throughput synthesis. Here we introduce a high-throughput combinatorial library of 16 tetrameric N-substituted glycine oligomers (peptoids) containing hydroxypyridinone or catecholamide chelating units linked via an ethylenediamine bridge and, for comparison, we also synthesized the corresponding mixed ligands derived from the spermine scaffold: 3,4,3-LI(1,2-HOPO)₂(CAM)₂ and 3,4,3-LI(CAM)₂(1,2-HOPO)₂. Coordination-based luminescence studies were carried out with Eu³⁺ and Tb³⁺ to begin probing the properties of the new ligand architecture and revealed higher sensitization efficiency with the spermine scaffold as well as different spectroscopic features among the structural peptoid isomers. Solution thermodynamic properties of selected ligands revealed different coordination properties between the spermine and peptoid analogues with Eu³⁺ stability constants log β₁₁₀ ranging from 28.88 ± 3.45 to 43.97 ± 0.49. The general synthetic strategy presented here paves the way for precision design of new specific and versatile ligands, with a variety of applications tailored towards the use of f-elements, including separations, optical device optimization, and pharmaceutical development.

Full text of DOI:10.1039/c9sc01068h

Chemical
Science
View Article Online
View Journal
EDGE ARTICLE
Combinatorial design of multimeric chelating
peptoids for selective metal coordination†
Cite this: DOI: 10.1039/c9sc01068h
a
a
Abel Ricano,‡a Ilya Captain,‡a Korey P. Carter,
Gauthier J.-P. Deblonde
Bryan P. Nell,
All publication charges for this article
have been paid for by the Royal Society
of Chemistry
a
ab
*
and Rebecca J. Abergel
Current methods for metal chelation are generally based on multidentate organic ligands, which are
generated through cumbersome multistep synthetic processes that lack flexibility for systematically
varying metal-binding motifs. Octadentate ligands incorporating hydroxypyridinone or catecholamide
binding moieties onto a spermine scaffold are known to display some of the highest affinities towards f-
elements. Enhancing binding affinity for specific lanthanide or actinide ions however, necessitates ligand
architectures that allow for modular and high throughput synthesis. Here we introduce a high-
throughput combinatorial library of 16 tetrameric N-substituted glycine oligomers (peptoids) containing
hydroxypyridinone or catecholamide chelating units linked via an ethylenediamine bridge and, for
comparison, we also synthesized the corresponding mixed ligands derived from the spermine scaffold:
3,4,3-LI(1,2-HOPO)2(CAM)2 and 3,4,3-LI(CAM)2(1,2-HOPO)2. Coordination-based luminescence studies
were carried out with Eu3+ and Tb3+ to begin probing the properties of the new ligand architecture and
revealed higher sensitization efficiency with the spermine scaffold as well as different spectroscopic
features among the structural peptoid isomers. Solution thermodynamic properties of selected ligands
revealed different coordination properties between the spermine and peptoid analogues with Eu3+
stability constants log b110 ranging from 28.88 ꢀ 3.45 to 43.97 ꢀ 0.49. The general synthetic strategy
presented here paves the way for precision design of new specific and versatile ligands, with a variety of
applications tailored towards the use of f-elements, including separations, optical device optimization,
and pharmaceutical development.
Received 4th March 2019
Accepted 5th June 2019
DOI: 10.1039/c9sc01068h
rsc.li/chemical-science
participate in bonding, thus local coordination environments
do not overtly inuence their respective solution or solid-state
Introduction
electronic and photophysical properties.
Interest in f-block elements has increased substantially in
recent years as these elements have been used in a growing
number of applications such as luminescent phosphors1,2 and
diagnostic imaging and therapy.3–6 While the radioactive 5f
actinides (An) have garnered signicant attention due to the
legacy of their use in energy and weapons,7–11 lanthanides (Ln),
aside from promethium (Pm), consist wholly of stable isotopes.
Many of the applications that utilize Ln metals are based on
transitions within the 4f orbitals, which are shielded from
external perturbations by lled 5s and 5p orbitals. This means
that the 4f orbitals of Ln metal centers do not appreciably
Harnessing the unique capabilities of lanthanide metal
cations so that one may selective tune the resulting properties
involves exercising control over the rst coordination sphere
of Ln(III) ions. Using high-affinity, chelating ligands is one
pathway for controlling Ln local coordination geometry, and
currently the highest affinity f-block binding is achieved by
organic, bio-inspired ligands.12–14 Raymond and coworkers15
have demonstrated the most striking example of f-block metal
cation specicity by preparing a macrocyclic ligand that
chelates Th4+ with a formation constant (Kf) of 1054, yet similar
examples are limited as synthesis of such molecular structures
is time consuming, expensive, and oen not modular. As
a result, new synthetic methods are needed to develop ligands
aChemical Sciences Division, Lawrence Berkeley National Laboratory, Berkeley, CA
94720, USA. E-mail: abergel@berkeley.edu
in
a high-throughput manner, which can also provide
bDepartment of Nuclear Engineering, University of California, Berkeley, CA 94720, USA
† Electronic supplementary information (ESI) available: Detailed experimental
procedures, LC-MS, high resolution MS, TOF-MS/MS, and NMR spectra for
peptoid ligands, UV-vis and luminescence spectra for Eu and Tb peptoid and
343-LI complexes, luminescence lifetime values, example data sets from
solution thermodynamic titrations, and quantum yield data are provided. See
DOI: 10.1039/c9sc01068h
improved understanding of the subtleties of 4f-element coor-
dination chemistry.
Most ligands with high-affinity for the f-block are multi-
dentate and are composed of small metal-binding moieties
placed on a molecular scaffold.7 Spermine (also called 3,4,3-LI)
is described as a privileged scaffold due to the large number of
‡ Equal contributions.
This journal is © The Royal Society of Chemistry 2019
Chem. Sci.

View Article Online
Chemical Science
Edge Article
ligands that have been derived from it and their relatively facile,
low-cost, synthesis. Ligands of form X (Fig. 1) are typically
synthesized by reacting spermine with activated esters of
bidentate moieties and oen display impressive binding char-
acteristics. Mixed ligands of form Y (Fig. 1) that contain two
different types of binding groups can be prepared by exploiting
reactivity differences between the 3,4,3-LI primary and
secondary amines, but are not very common due to low
synthetic yields. Further variation of binding units along the
3,4,3-LI backbone has not been done due to impractical and low
yielding syntheses of form Z ligands (Fig. 1). Access to ligands of
the latter type is important, as it would allow for systematic
binding unit variation along the backbone, leading to an
enhanced understanding of Ln–ligand interactions.
To circumvent synthetic hurdles in the 3,4,3-LI system, we
developed a new strategy by incorporating high-affinity metal-
binding units onto peptoid scaffolds. Peptoids are structural
analogs of peptides composed of N-substituted glycine oligo-
mers. They are composed of primary amine sub-monomers, can
be synthesized on solid support, and may be prepared using
traditional automated peptide synthesizers.16 The combinato-
rial nature of peptoid synthesis and asymmetric scaffold
provides a unique opportunity to control binding moiety
sequence. This method can yield thousands of ligands in
a matter of days, yet most studies have been limited to using
commercially available amine sub-monomers. Peptoids have
been examined as potential binders for a number of metals
using combinatorial libraries,17–22 and here we aimed to
improve upon the process of peptoid ligand synthesis by using
custom-made sub-monomers known for their f-element
binding properties. In this pilot study, we incorporated two
well-characterized binding moieties into the peptoid scaffold
using standard synthetic methods. The modularity of the scaf-
fold allowed for the rapid preparation of all 16 possible tetra-
meric peptoid, octadentate ligands. In parallel, we also
prepared two spermine-based ligands of form Y (Fig. 1) to serve
as baseline comparisons. We then carried out luminescence-
based coordination studies on all resulting ligands using Eu3+
and Tb3+, and stability constants for selected chelators with
Eu3+ were determined. These systematic studies revealed that
slight variations in ligand structure could result in signicant
differences in spectroscopic properties and complex stability,
even for chemically similar Ln elements.
Results and discussion
Synthesis of peptoid library
Bidentate metal-binding units 1-hydroxy-pyridin-2-one (1,2-
HOPO) and catecholamide (CAM) are well known moieties in f-
element binding (Fig. 1).7 Extensive studies have established the
octadentate 3,4,3-LI(1,2-HOPO) as a agship chelator for An and
Ln decorporation,13,23 and 3,4,3-LI(CAM) was recently evaluated
for Eu3+, Th4+, and Zr4+ binding (Fig. 1).14 To accommodate the
larger ionic radii of f-elements upon chelation, ethylenediamine
bridges were inserted within the sub-monomer units to link 1,2-
HOPO and CAM moieties to the peptoid scaffold (Scheme 1).
Primary amine sub-monomers that we term H and C (corre-
sponding to ethylenediamine-substituted 1,2-HOPO and CAM,
respectively) were synthesized by reacting known precursor acyl
chlorides with excess ethylenediamine to ensure single addition
of chelating moieties (Scheme 1). The benzyl group was used to
protect the N-hydroxyl of H while a diphenylacetal was used for
the 1,2-diol on the C sub-monomer, and both sub-monomers
were characterized by 1H and 13C NMR spectroscopy (ESI†).
Acid labile protecting groups were chosen so they would be
removed during cleavage of peptoids from solid support using
triuoroacetic acid (TFA).
H and C sub-monomers were then used to prepare the
peptoids on Rink Amide resin employing standard coupling
chemistry. In lieu of automated peptoids synthesis, we used
fritted syringes as reaction vessels, which allowed for recovery of
unreacted sub-monomers for future use. We were able to
successfully reuse the solution of sub-monomer up to three
times without any decrease in performance. Spent solutions
were concentrated and re-puried by column chromatography.
Subsequent test cleavages on small samples of resin revealed
that benzyl groups were not completely removed from H sub-
Fig. 1 Spermine-based octadentate ligands. The 3,4,3-LI scaffold
(blue) is connected to four metal-binding units (R1, R2, R3, and R4). With Scheme 1 Synthesis of tetrameric peptoid ligands. (a) Oxalyl chloride,
1,2-HOPO and CAM (metal-binding atoms highlighted in red), only then ethylenediamine. (b) 0.1 M BBr3 in CH2Cl2. (c) 95/2.5/2.5 – TFA/
four ligand combinations are easily accessible through classical water/triisopropylsilane. One peptoid structure, CHHC, shown, out of
synthetic methods: 3,4,3-LI(1,2-HOPO), 3,4,3-LI(HCCH), 3,4,3- 16 possible tetrameric combinations with the two different types of
LI(CHHC), and 3,4,3-LI(CAM).
sub-monomers C and H. Metal-binding oxygens are highlighted in red.
Chem. Sci.
This journal is © The Royal Society of Chemistry 2019

View Article Online
Edge Article
Chemical Science
monomers during TFA treatment, even with prolonged reaction crude ligands, which were puried using reverse phase
1
times. Aer screening a number of conditions, we found that preparative HPLC and characterized by H and 13C NMR spec-
addition of 5% concentrated aqueous HCl to the TFA cleavage troscopy and high-resolution mass spectrometry (ESI†).
cocktail afforded the deprotected product but also led to
signicant peptoid backbone cleavage, which greatly compli-
cated purication. Post-cleavage hydrogenolysis yielded the
UV-visible spectroscopic properties
The electronic absorption spectra (ESI, Fig. S24†) of all sixteen
peptoid ligands revealed the characteristic p–p* transitions of
each aromatic system. (16) HHHH (lmax ¼ 348 nm) and (15)
CCCC (lmax ¼ 320 nm) served as upper and lower bounds,
respectively, for the maximum absorption of the peptoid
ligands. Absorption maxima of peptoid ligands with more H
units were red-shied when compared with the multi-C unit
peptoids (Fig. 2A), but featured similar extinction coefficients
(Fig. 2B). Both 3,4,3-LI(CAM) and 3,4,3-LI(HCCH) free ligands
contained a small shoulder around 285 nm in addition to their
main absorption band centered at 320 nm and 350 nm,
respectively. The absorption band for free 3,4,3-LI(CHHC) was
centered at approximately 322 nm and changed in intensity and
width upon complexation. All the complexes (of both Eu3+ and
Tb3+) exhibited similar absorbance maxima at ca. 330 nm, with
desired products but was impractical on milligram scales at
which the peptoids were prepared. Alternatively, we found that
the benzyl groups were removed by a BBr3 treatment without
cleaving the peptoid from solid support. This allowed us to
deprotect the peptoids and easily remove byproducts using
a solvent wash. We nally settled on treating the resin with
0.1 M BBr3 in dichloromethane (CH2Cl2) followed by TFA
cleavage to obtain the deprotected crude peptoid ligands. The
ligands were puried using reverse phase preparative HPLC and
1
were characterized by H and 13C NMR spectroscopy as well as
high-resolution mass spectrometry, with TOF-MS/MS used to
verify the peptoid sequence (ESI, Fig. S1–S23†). Ligands con-
taining H moieties were generally more difficult to work with
and to purify due to their propensity to bind iron and other
trace metals present in instrument components even in acidic
media. Due to the asymmetric nature of the peptoid scaffold
(Scheme 1), the formation of 2n ligands is possible, where 2
denotes the number of sub-monomers (H and C) used and “n”
the oligomer length. Sufficient amounts of all 16 possible
tetrameric peptoid ligands were obtained in pure form. The
synthesized peptoid chelators are listed in Table 1.
Synthesis of 3,4,3-LI mixed ligands
In order to compare our new peptoid scaffold to the 3,4,3-LI
backbone, we prepared two mixed derivatives, each either
bearing two 1,2-HOPO units on the two internal tertiary amines
and two CAM subunits on the terminal secondary amines of the
spermine backbone (3,4,3-LI(CAM)2(1,2-HOPO)2, adopting the
same binding group sequence as peptoid CHHC), or vice versa
(3,4,3-LI(1,2-HOPO)2(CAM)2, adopting the same binding group
sequence as peptoid HCCH). This synthesis takes advantage of
the higher reactivity of the primary amines of spermine. One of
the acid derivatives of 1,2-HOPO or CAM (with hydroxyl func-
tionalities protected by benzyl or methyl groups, respectively) is
rst added onto the terminal amines through reaction with
spermine in a controlled, dilute manner, and in the presence of
amide coupling reagent carbonyldiimidazole (CDI). The acid
chloride of the other subunit is then made in situ and directly
added to the internal amines. Both methyl and benzyl protect-
ing groups could be removed by BBr3 treatment, providing the
Fig.
2 Ligand and complex UV-vis spectroscopic parameters
dependence as a function of the number of 1,2-HOPO metal-binding
units (0 for CCCC and 3,4,3-LI(CAM) to 4 for HHHH and 3,4,3-LI(1,2-
HOPO)) on the peptoid (black full circles) and spermine-based scaf-
folds (red open symbols; circles for 3,4,3-LI(CAM) and 3,4,3-LI(1,2-
HOPO); upward and downward triangles for 3,4,3-LI(HCCH) and
3,4,3-LI(CHHC), respectively). (A) Ligand maximum absorption wave-
length. (B) Ligand extinction coefficient. (C) Wavelength shift upon Eu
complexation. (D) Change in extinction coefficient upon Eu
complexation. (E) Wavelength shift upon Tb complexation. (F) Change
in extinction coefficient upon Tb complexation.
Table 1 Synthesized Peptoid Librarya
(1) HHHC
(5) CHHC
(9) HCCH
(13) CCHC
(2) CHHH
(6) HHCC
(10) CHCH
(14) CCCH
(3) HCHH
(7) CCHH
(11) HCCC
(15) CCCC
(4) HHCH
(8) HCHC
(12) CHCC
(16) HHHH
a
Synthesized tetrameric peptoids with H and C binding moieties. The
lemost designation in the ligand name denotes the rst added
moiety on the peptoid scaffold (i.e. nearest the terminal amide).
This journal is © The Royal Society of Chemistry 2019
Chem. Sci.

View Article Online
Chemical Science
Edge Article
Ln–peptoid complexes with multiple H units blue-shied with these observed weak emission properties provide evidence of
respect to the free ligands and vice versa for those with multiple poor energy transfer to the metal ([a] in Fig. 3), signicant non-
C moieties (Fig. 2C and E for Eu3+ and Tb3+ complexes, radiative de-excitation ([b] in Fig. 3), or a combination thereof.
respectively). Extinction coefficients decreased signicantly
One of the critical components in Ln(III) sensitization is the
upon Eu or Tb complexation for peptoid ligands containing energy transfer efficiency between the rst triplet-state of the
more than 2 C sub-monomers, whereas the opposite trend was ligand (T1) and the accepting state of the metal (Fig. 3).28 Since
observed for the spermine-based ligands (Fig. 2D and F for Eu3+ Gd3+ exhibits a size and atomic weight similar to Eu3+ but lacks
and Tb3+ complexes, respectively).
an appropriately positioned electronic acceptor level, the triplet-
state of the ligand can be estimated by measuring the phos-
phorescence spectra of the respective Gd(III) complex. Energy
transfer is most efficient when the respective resonance levels
are similar enough in energy such that thermal back-transfer
can be minimized. For Eu3+ and Tb3+, this occurs when the
triplet-state donor level is 2500–3500 cmꢁ1 and 2500–
4000 cmꢁ1, respectively, higher than the metal accepting state.29
Triplet state energies were determined for all investigated
compounds, according to described methods.30,31 The T1 ener-
gies were found close to the Eu3+ 5D2 level, similar to what has
been reported with polyaminocarboxylates28 but contrasting
with ligands such as Schiff bases32 (5D0) or b-diketonates33 (5D1).
Ligand-sensitization of Eu3+ and Tb3+ luminescence
The ligand of form X (Fig. 1) that bears 1,2-HOPO units (i.e.
3,4,3-LI(1,2-HOPO)) is known to sensitize the luminescence
emission of selected Ln(III) cations through the formation of
stable stoichiometric complexes and the so-called “antenna
effect”,23,24 as depicted through a simplied Jablonski diagram
in Fig. 3. To this end, we sought to determine if the peptoid
ligands could also sensitize Ln(III) ions in a similar fashion,
using Eu3+ and Tb3+ as representative systems. The emission
spectra of the Eu3+ and Tb3+ peptoid complexes revealed rela-
tively weak signals (ESI, Fig. S25–S30†). Quantum yields of
Eu:HHCH (4) (f ¼ 0.0043 ꢀ 0.0002) and Tb:3,4,3-LI(CHHC) (f ¼
0.0011 ꢀ 0.0001) were determined by the optical dilution
method. These values were then used as benchmarks to esti-
mate the values for the remaining complexes (ESI, Table S1†) by
taking the ratio of the integrated emission spectra to that of the
corresponding metal complex, assuming identical extinction
coefficients. Here we can use Eu:3,4,3-LI(1,2-HOPO) as a refer-
ence with an estimated value of 0.14, consistent with the pub-
lished quantum yield values of 15.6 ꢀ 0.6% (isolated complex)
and 14.0 ꢀ 0.3% (complex formed in situ).25 This comparative
method is not meant to be interpreted as rigorous for quantum
yield determination, rather we aimed to probe relevant varia-
tions within the large number of ligands and the small scale of
the pilot study. The largest quantum yields for Eu3+ and Tb3+,
not including 3,4,3-LI(1,2-HOPO), are complexes formed with
(16) HHHH and 3,4,3-LI(CHHC), respectively. Nonetheless,
5
Structures with triplet state energies values nearest to the D2
level resulted, overall, in higher Eu emission intensities
(Fig. 4A). To further probe the role of each HOPO or CAM
binding unit on luminescence sensitization, phosphorescence
spectral shapes were examined in more details for each Gd
complex. Deconvolution of the main triplet-state peak of (15)
CCCC and (16) HHHH with two Gaussian curves each revealed
Fig. 4 (A) Eu luminescence emission intensity as a function of ligand
triplet-state energy for the peptoid (black full circles) and spermine-
based scaffolds (red open symbols; circles for 3,4,3-LI(CAM) and 3,4,3-
LI(1,2-HOPO); upward and downward triangles for 3,4,3-LI(HCCH)
and 3,4,3-LI(CHHC), respectively). (B) % contribution of the 1,2-HOPO
(full black squares) and CAM (open red squares) metal-binding units to
the phosphorescence from the triplet excited state as a function of the
number of 1,2-HOPO metal-binding units for peptoid ligands. (C) Eu
luminescence emission intensity and (D) ligand triplet-state energy as
Fig. 3 Jablonski diagram representing the major energy transfer a function of the number of 1,2-HOPO metal-binding units (0 for
pathways of sensitization. (a) Represents energy transfer from the CCCC and 3,4,3-LI(CAM) to 4 for HHHH and 3,4,3-LI(1,2-HOPO)) on
ligand to the metal; (b) indicates non-radiative de-excitation. Energy the peptoid and spermine-based scaffolds; symbols are as described
levels for free ions were adapted from previous publications.26,27
for panel A.
Chem. Sci.
This journal is © The Royal Society of Chemistry 2019

View Article Online
Edge Article
Chemical Science
bands centered around 20 000 cmꢁ1 (ꢂ498 nm) for both, along with some Stark splitting of the 7F2 transition for 4 that is
18 100 cmꢁ1 (ꢂ552 nm) for (16) HHHH, and a higher energy not observed with the analogous Eu:HCHH (3) complex.
contribution from (15) CCCC centered around 21 400 cmꢁ1
The Tb3+ spectra revealed much less efficient sensitization
(ꢂ467 nm). There was an additional small band in the (16) than with Eu3+ congeners as all spectra had signicant ligand
HHHH phosphorescence spectra centered around 25 000 cmꢁ1 luminescence leakage around 450 nm. None of the triplet-states
(ꢂ395 nm) that was also observed for (5) CHHC (ꢂ391 nm). The for any of the investigated ligands had optimum energy to
extent of 1,2-HOPO and CAM triplet-state character in the rest of efficiently populate the Tb3+ accepting level (ꢂ20 500 cmꢁ1) and
the peptoid ligands was gauged by the relative contribution of there was a signicant difference in the sensitization efficien-
the spectral bands centered at 18 100 cmꢁ1 and 21 400 cmꢁ1
,
cies between the Eu(III) and Tb(III) systems. For Eu3+ the most
respectively, since both (15) CCCC and (16) HHHH featured luminescent peptoid was (16) HHHH, followed by (4) HHCH,
a peak centered at ca. 20 000 cmꢁ1 (Fig. 4B). Luminescence whereas with Tb3+ it was (2) CHHH.
intensity was found to increase with the number of H moieties
Luminescence lifetimes of Eu:peptoid complexes as well as
on the peptoid scaffold (Fig. 4, panels C and D, and ESI†), which 3,4,3-LI mixed ligand complexes with Eu3+ and Tb3+ revealed bi-
was notable as the 1,2-HOPO triplet-state contribution is just exponential decays with the exception of Eu:HHCH (4),
under 1000 cmꢁ1 above the 5D2 manifold of Eu(III), much lower Eu:CHHC (5), both mono-exponential decays, and Eu:HCCC
than the 2500 cmꢁ1 energy gap suggested by Latva et al.28 for (11) and Tb:3,4,3-LI(CHHC), which were t as tri-exponential.
optimal energy transfer.
The longest lifetimes came from Eu complexes of (16) HHHH
The triplet-state data of the spermine based ligands was (827 ms) and 3,4,3-LI(CHHC) (733 ms), both corresponding to
similarly treated (Fig. 4). 3,4,3-LI(CAM) and 3,4,3-LI(1,2- one coordinated water molecule according to Kimura's empir-
HOPO) had deconvoluted bands centered around ical equation (ESI, Table S2†).34 Comparing (16) HHHH (827 ms)
22 000 cmꢁ1 (ꢂ455 nm) and 18 200 cmꢁ1 (ꢂ550 nm), respec- with 3,4,3-LI(1,2-HOPO) (805 ms),24 reveals that when all binding
tively, and an additional component around 19 800 cmꢁ1 units are 1,2-HOPO moieties there is little variation between
(ꢂ506 nm). As in the peptoid analogue, 3,4,3-LI(1,2-HOPO) 3,4,3-LI and peptoid scaffolds. With peptoid ((5) CHHC and (9)
had an additional small peak centered around 24 400 cmꢁ1 HCCH) and 3,4,3-LI mixed ligands each featuring two C and two
(ꢂ410 nm). A larger triplet-state character from 3,4,3-LI(CAM) H units the results get more interesting as we note a small (64
was observed in 3,4,3-LI(CHHC) than in the 3,4,3-LI(HCCH). ms) increase in Eu(III) lifetime when going from (9) to 3,4,3-
As with the peptoid scaffold ((15) CCCC), no uorescence was LI(HCCH) and a more substantial change (180 ms) when going
observed for 3,4,3-LI(CAM) complexed with Eu3+. There was from (5) to 3,4,3-LI(CHHC) (ESI, Table S2†).
about a ꢂ30% weaker quantum yield for the Eu complex with
The structural peptoid isomers bearing three H and one C
3,4,3-LI(HCCH) than 3,4,3-LI(CHHC), but this could very well units have lifetimes ranging from 432–679 ms (ESI, Table S2†)
be attributed to the low molar absorptivity (vide infra) of the with the most noteworthy difference between (3) HCHH and (4)
3,4,3-LI(HCCH) complex.
Further analysis of the Eu emission spectra revealed Stark observe a ca. 250 ms change in Eu3+ lifetime, which corresponds
HHCH. By exchanging C units in the 2nd and 3rd positions we
5
7
splitting and small shis in the D0 / F2 transition for most to one and two bounds water molecules, respectively.34 This is
complexes (ESI, Fig. S25–S30†). Two comparisons are presented indicative of a possible change in coordination of Eu(III) by (3)
in Fig. 5. A 2 nm shi between the peaks of the Eu:1,2-HOPO and (4) as a result of the difference in the order of H and C
complexes, 3,4,3-LI(1,2-HOPO) and (16) HHHH, as well as binding units, which suggests the location of H and C moieties
a shoulder peak at 618.5 nm for Eu:3,4,3-LI(1,2-HOPO) was may enhance/decrease recognition of Eu3+. The structural pep-
observed (Fig. 5A). Panel B of Fig. 5 shows the directional toid isomers bearing two H and two C units have lifetimes
isomers (3) HCHH and (4) HHCH and highlights the substantial ranging from 425–553 ms (ESI, Table S2†) and an analogous
differences that can result from small changes in the placement comparison of Eu:peptoid complexes where C units are
of H and C binding moieties. We note an approximate 4-fold exchanged in the 2nd and 3rd positions ((7) CCHH and (10)
intensity increase when comparing the Eu complexes of 3 and 4, CHCH) yields a difference of only ca. 25 ms, which is a much
smaller change than what was noted above. For peptoid isomers
with two H and two C units the more interesting results come
from comparing complexes with H units in the 3rd and 4th
positions. We observe decreases of ca. 75–125 ms in Eu(III) life-
times when H moieties are switched from the 3rd position to the
4th position (comparing (5) CHHC with (10) CHCH and (8)
HCHC with (9) HCCH), which is also possibly indicative of
changes in ligand conformation about the Eu(III) metal center as
a result of changing the order of binding moieties. The struc-
tural peptoid isomers bearing one H and three C units have
lifetimes ranging from 569–682 ms (ESI, Table S2†) and display
the least amount of variance (in lifetime values) with changes in
the order of the binding units, suggesting that once a ligand
includes a certain number of C units (three) only one binding
Fig. 5 The 5D0 / 7F2 emissive transition of the Eu complexes formed
with (A) 3,4,3-LI(1,2-HOPO) (dotted line) and (16) HHHH (solid line); (B)
(3) HCHH (dotted line) and (4) HHCH (solid line).
This journal is © The Royal Society of Chemistry 2019
Chem. Sci.

View Article Online
Chemical Science
Edge Article
conformation, about the Eu3+ metal center, is observed. Note- 3,4,3-LI(CAM) is bound around pH 11. The protonation constants
worthy, Gd coordination properties will be investigated in the for 3,4,3-LI(HCCH) were also investigated via spectrophotometric
future for those four ligand structures with which Eu3+ lumi- titrations, but the similarity of the Eu complex absorption signal
nescence lifetimes suggested two inner-sphere water molecules, with the signal from the free ligand (3,4,3-LI(HCCH)) prevented
(3) HCHH, (7) CCHH, (9) HCCH, and (10) CHCH. The ability to renement with HypSpec. Therefore, competition batch titra-
form Gd:peptoid complexes with q values of 2 may result in tions were employed against 3,4,3-LI(1,2-HOPO) (ESI†). The
enhanced relaxivity properties, which could prove attractive for log b111 value obtained for 3,4,3-LI(HCCH) was 43.34 ꢀ 0.49. For
magnetic resonance imaging applications when associated with comparison, we selected a peptoid with two H and two C moieties
high thermodynamic stability constants, as detailed below.
and the Eu complex formed with (7) CCHH displayed a stability
constant (log b110) of 36.01 ꢀ 4.13, and is also fully formed at pH
7.4. As previously mentioned, having access to ligands with
HOPO and CAM units opens up avenues for manipulating the
properties of the chelator, as seen with the mixed ligands and
when comparing the CAM analogues.
Solution thermodynamics
Inspired by previous efforts of our group to characterize 3,4,3-
LI(CAM),14 we sought to investigate the solution thermodynamic
properties of the peptoid analogue (15) CCCC. The last 3 pKa's of
(15) CCCC were not resolvable by spectrophotometric titrations
and were held constant at values comparable to the spermine
analogue. The proton-independent stability constant (log b110) of
the (15) CCCC complex with Eu3+ was rened to 28.88 ꢀ 3.45 (ESI,
Table S3†) and this result is quite similar to Eu3+ complexed with
3,4,3-LI(CAM) (log b110 ¼ 29.65).14 Speciation modeling for equal
concentrations of 3,4,3-LI(CAM) and (15) CCCC in the presence of
Eu3+ reveals that 3,4,3-LI(CAM) predominately binds Eu3+ above
a pH of 6 and further, the speciation diagram demonstrates that
(15) CCCC is also fully binding Eu3+ under the conditions
employed for uorescence measurements (ESI†). Incorporating
a mixture of CAM and HOPO groups onto a scaffold was expected
to increase the affinity of the chelator to the metal while
increasing the pH range at which it binds, in comparison to the
parent ligands. The stability constant (log b110) of 3,4,3-LI(CHHC)
complexed with Eu3+ is 35.97 ꢀ 0.06 (ESI, Table S3†). Thermo-
dynamic modeling (Fig. 6) of 3,4,3-LI(1,2-HOPO), 3,4,3-LI(CAM),
and 3,4,3-LI(CHHC) in the presence of Eu3+ at equal concentra-
tions revealed that the predominant complexing ligand above
a pH of 6 is the mixed 343-LI species 3,4,3-LI(CHHC). At lower pH
values, 3,4,3-LI(1,2-HOPO) dominated while a small fraction of
Conclusions
We report herein the synthesis of a new library of tetrameric
peptoid ligands that can be used for chelation of f-block metals.
We prepared and characterized a library of octadentate peptoids
based upon 1,2-HOPO- and CAM-based sub-monomers, inves-
tigated their thermodynamic and photophysical properties with
Eu3+ and Tb3+, and compared them with their spermine
analogues. Both ligand systems show high affinity for Ln(III)
metals with stability constants greater than 1029 for Eu3+ with
the investigated chelators. Overall, the peptoid scaffold
provides
a platform to conduct a more comprehensive,
systematic approach to study Ln–ligand interactions, and one
can envision ne-tuning the peptoid manifold to incorporate
the desired cavity size, binding moieties, and energetics of the
chelator. As synthesis is done on solid-support, creativity and
modularity can be imparted onto the ligand, depending on the
desired function. We are now working on acquiring structural
parameters of metal complexation, and further characterizing
peptoid complexes with other f-block metals, by probing their
ability to efficiently sensitize near-IR emitters as well as
exploring the fundamental differences in energy transfer
processes between 4f and 5f species. On-going work also focuses
on expanding synthetic modications of the scaffold for the
development of targeted therapeutic applications.
Materials and methods
Chemicals were obtained from commercial suppliers and were
used as received unless stated otherwise. The ligands 3,4,3-LI(1,2-
HOPO) and 3,4,3-LI(CAM) were prepared and characterized as
previously described.14,35 Synthesis precursors 2,2-diphenylbenzo
[d][1,3]dioxole-4-carboxylic acid and 1,6-carboxy-1-(benzyl)hydroxy-
2-pyridinone were prepared according to literature procedures.14,36
While the mixed ligand 3,4,3-LI(CHHC) has been made before,7,37
herein we followed a slightly different approach and expanded
efforts beyond what had been previously reported to produce both
3,4,3-LI(CHHC) and 3,4,3-LI(HCCH). Detailed syntheses and
characterization of the two compounds can be found in the ESI.†
All NMR spectra were recorded at ambient temperature on Bruker
FT-NMR spectrometers, using tetramethylsilane as an internal
reference. SilicaFlash G60 (particle size 60–200 mm) was used for
Fig. 6 Speciation diagram for a competition between 3,4,3-LI(1,2-
HOPO), 3,4,3-LI(CAM), and 3,4,3-LI(CHHC) abbreviated HOPO (gray),
CAM (green), and CHHC (blue), respectively, in the presence of Eu3+
.
[HOPO] ¼ [CAM] ¼ [CHHC] ¼ [metal] ¼ 25 mM, I ¼ 0.1 M, T ¼ 25 ꢃC.
Only the major species were labeled on the plots for clarity.
Chem. Sci.
This journal is © The Royal Society of Chemistry 2019

View Article Online
Edge Article
Chemical Science
ash column chromatography. LC-MS was performed on an Agi- it became clear and the evolution of gas has ceased (ꢂ1 hour).
lent LC/MS system consisting of an Agilent 1200 binary LC pump, The volatiles were then removed using the manifold vacuum
a temperature-controlled autosampler, a PDA UV detector, and and the resulting white solid was dissolved in dry dichloro-
a 6530 Accurate Mass Q-TOF mass spectrometer (Wilmington, DE, methane. A separate 1 L roundbottom ask outtted with an
USA). The mass spectrometer was equipped with a JetStream® ESI addition funnel was charged with ethylenediamine (8 ml, 120
probe operating at atmospheric pressure. The ESI source param- mmol) and 50 ml dry dichloromethane; the resulting solution
eter settings were: mass range m/z 100–1200, gas temperature was cooled to 0 ꢃC using an ice bath. The aforementioned
350 ꢃC, gas ow 10 L minꢁ1, nebulizer 50 psi, sheath gas solution of acyl chloride was transferred into the addition
temperature 400 ꢃC, sheath gas ow 12 L minꢁ1, capillary voltage funnel and was diluted with dichloromethane to a total volume
(Vcap) 3500 V, nozzle voltage 500 V, fragmentor 200 V, skimmer of 700 ml. The acyl chloride solution was then added into the
65 V, octopole RF (OCT 1 RF Vpp) 750 V. Reverse phase preparatory vigorously stirred ethylenediamine over 1.5 hours at 0 ꢃC.
HPLC was performed on a Varian ProStar system with a Vydac C18 Following the addition, the reaction solution was transferred
column. HRMS and MS-MS were obtained on a Waters Xevo G2 into a separatory funnel and was washed with 0.5 M NaOH in
Qtof mass spectrometer, and leucine encephalin lockspray with 50% saturated aq. NaCl (50 ml ꢄ 2). The organic phase was
mass correction was used for HR-MS.
dried over MgSO4 and was concentrated on a rotary evaporator
yielding the crude. The crude was puried using silica column
chromatography (5 / 10% MeOH in DCM with 1% Et3N, Rf ¼
0.35 in 10% MeOH in DCM). The desired fractions were
combined, concentrated under reduced pressure, and dried
under vacuum yielding the HOPO submonomer as a yellow oil
(3.49 g, 9.68 mmol, 80% yield). 1H NMR (300 MHz, CDCl3)
d 7.42–7.49 (3H, m, ArH and NH), 7.31–7.40 (3H, m, ArH), 7.26
(1H, dd, J ¼ 9.2, 6.8 Hz, CHCHCH), 6.67 (1H, dd, J ¼ 9.2, 1.6 Hz,
CHCHCH), 6.44 (1H, dd, J ¼ 6.7, 1.6 Hz, CHCHCH), 5.28 (2H, s,
CH2Ph), 3.36 (2H, t, J ¼ 6.0 Hz, NHCH2), 3.80 (2H, q, J ¼ 6.1 Hz,
NH2CH2). 13C NMR (75 MHz, CDCl3) d 160.3, 158.5, 142.5, 138.0,
133.2, 130.1, 129.4, 128.6, 124.0, 106.4, 79.3, 42.7, 40.7. Calcd for
Ethylenediamine substituted CAM submonomer, “C”
2,2-Diphenylbenzo[d][1,3]dioxole-4-carboxylic acid (3.84 g,
12.06 mmol) was suspended in 30 ml of toluene under an argon
atmosphere. Oxalyl chloride (1.14 ml, 13.3 mmol) was then
added followed by a catalytic amount of N,N-dimethylforma-
ꢃ
mide. The suspension was heated to 40 C and stirred until it
became clear and the evolution of gas had ceased (ꢂ1 hour).
The volatiles were then removed using the manifold vacuum
and the resulting white solid was dissolved in dry CH2Cl2. A
separate 1 L roundbottom ask outtted with an addition
funnel was charged with ethylenediamine (8 ml, 120 mmol) and
50 ml dry CH2Cl2; the resulting solution was cooled to 0 ꢃC
using an ice bath. The aforementioned solution of acyl chloride
was transferred into the addition funnel and was diluted with
C
15H17N3O3 + H, 288.1328; found, 288.1361.
Synthesis of peptoids
CH2Cl2 to a total volume of 700 ml. The acyl chloride solution Unless noted otherwise, all steps were carried out in fritted
was then added into the vigorously stirred ethylenediamine over polypropylene syringes, which allowed for recovery of sub-
1.5 hours at 0 ꢃC. Following the addition, the reaction solution monomer for re-use. Automatic peptoid synthesis was not an
was transferred into a separatory funnel and was washed with option for this work due to difficulty of sub-monomer prepara-
0.5 M NaOH in 50% saturated aq. NaCl (50 ml ꢄ 2). The organic tion. 100–150 mg of Rink amide resin was added to a fritted
phase was dried over MgSO4 and was concentrated on a rotary syringe. The resin was swollen by adding 2 ml of DMF and rocking
evaporator yielding the crude. The crude was puried using for 30 minutes, and then solution was ejected to isolate the
silica column chromatography (5 / 10% MeOH in CH2Cl2 with swollen resin. 1 ml of 20% 4-methylpiperidine in DMF (v/v) was
1% Et3N, Rf ¼ 0.35 in 10% MeOH in CH2Cl2). The desired added to deprotect the Fmoc group followed by agitation for two
fractions were combined, concentrated under reduced pressure, minutes, draining of the solution, and repeating for 12 minutes in
and dried under vacuum yielding the CAM submonomer as total. Immediately following, the resin was rinsed with DMF (2 ml,
a sticky yellow oil (3.49 g, 9.68 mmol, 80% yield). 1H NMR (300 5 times for 1 minute). To bromoacetylate, premix 0.8 M bromo-
MHz, CDCl3) d 7.64 (1H, br s, NH), 7.56–7.61 (5H, m, ArH), 7.37– acetic acid in DMF with 0.8 M N,N-diisopropylcarbodiimide (DIC)
7.42 (6H, m, ArH), 7.01 (1H, dd, J ¼ 7.7, 1.4 Hz, ArH), 6.94 (1H, t, (2 ml total solution with 0.4 M of each reagent). Draw the solution
J ¼ 7.9 Hz, ArH), 3.56 (2H, q, J ¼ 6.0 Hz, NHCH2), 2.97 (2H, t, J ¼ into the syringe, agitate for ve minutes, and rinse again with 2 ml
6.0 Hz, NH2CH2), 2.75 (2H, s, NH2). 13C NMR (75 MHz, CDCl3) DMF (5 ꢄ 1 minute). In order for displacement to occur, draw in
d 163.7, 147.2, 144.7, 139.4, 129.5, 128.4, 126.2, 122.4, 122.0, 1.5 ml of submonomer solution (0.2 M in DMF), agitate for one1
ꢃ
118.0, 116.0, 111.6, 42.3, 41.5. HRMS-ESI (m/z) [M + H] calcd For hour at 45 C, and then rinse with DMF (5 ꢄ 1 minute). Repeat
C
22H20N2O3 + H, 361.1563; found, 361.1581.
bromoacetylation and displacement steps until synthesis is
nished and then wash with DCM (3 ꢄ 1 minute) aer the last
DMF wash. Finally, dry resin by pulling the plunger out and
applying a gentle vacuum onto the syringe needle.
Ethylenediamine substituted 1,2-HOPO submonomer, “H”
1,6-Carboxy-1-(benzyl)hydroxy-2-pyridinone36 (3.84 g, 12.06
mmol) was suspended in 30 ml of toluene under an argon
atmosphere. Oxalyl chloride (1.14 ml, 13.3 mmol) was then
Test cleavage
added followed by a catalytic amount of N,N-dimethylforma- A small amount of resin was placed into an LC-MS vial and was
mide. The suspension was heated to 40 ꢃC and was stirred until treated with ꢂ200 ml of cleavage cocktail (95/2.5/2.5 TFA/water/
This journal is © The Royal Society of Chemistry 2019
Chem. Sci.

View Article Online
Chemical Science
Edge Article
38
Samples of quinine sulfate at various absorbance (0.02–0.1)
triisopropylsilane). The suspension was agitated for one hour at
room temperature, aer which it was diluted to 1 ml with
methanol and the beads were ltered off. The solution was
analyzed by LC-MS for the desired mass. This procedure was
only used analytically as benzyls from 1,2-HOPO containing
peptoids are not cleaved under these conditions (keep this in
mind when looking for the desired mass).
were prepared in 50 mM sulfuric acid. Ln complexes (complexes
¼ Tb:3,4,3-LI(CHHC) or Eu:HHCH (4)) at various absorbance
(0.02–1) were prepared in 0.1 M KCl, buffered in 50 mM HEPES
at pH 7.4. All samples were background corrected for solvent
auto uorescence. The ligand luminescence leakage for Tb3+
spectra was background corrected with QTIPlot. At least ve
samples were used per trial for both quinine sulfate and metal
complexes. The reported quantum yields and uncertainty were
based off three independent trials. Additionally, the triplet-state
energies of the ligands were determined by preparing Gd3+
complexes in situ (91% EtOH, 4.5% DMSO, and 4.5% H2O) and
measuring the phosphorescence of the ligand under cryogenic
conditions. The main peak in 3,4,3-LI(CAM), 3,4,3-LI(1,2-
HOPO), CCCC (15), and HHHH (16) phosphorescence spectra
was modeled by two Gaussian curves each and further used to
model the triplet-state for the mixed peptoid and spermine
ligands. It was assumed that the triple-state character of the
mixed ligands was a linear combination of the pure CAM and
1,2-HOPO ligands.
Deprotection, cleavage, and purication
Dry resin (100–150 mg) was placed into a scintillation vial and
was swollen in 9 ml of DCM by shaking for 30 minutes. 1 ml of
1.0 M BBr3 in hexanes was added via a syringe and the vial was
capped and shaken for 60 minutes ensuring that all of the resin
was thoroughly submerged, which removes benzyl protecting
groups from HOPO units. The solvent was carefully removed
with a glass pipette and the resin was washed with DCM (2 ml)
and methanol (2 ml ꢄ 2), followed by DCM (2 ml ꢄ 2). The
peptoid was then cleaved from resin by treatment with cleavage
cocktail for 60 minutes (the treatment also deprotects CAM
units). The cleavage cocktail was ltered from resin and a small
aliquot was removed and diluted with methanol for LC-MS
analysis (1 / 30% MeCN in H2O over 20 minutes, both with
0.1% formic acid) while the resin was washed of TFA traces and
discarded. Most LC-MS analyses showed a relatively clean
desired compound. Iron complexes were sometimes observed,
which we believe came from stainless steel components of the
instrument. Volatiles were then removed from the ltrate using
a vacuum pump. The resulting residue was dissolved in 90/10
acetic acid/water (0.5–1 ml) and the resulting clear solution
was stirred at 42 ꢃC and treated with water in 0.5 ml increments.
The solution turned turbid upon addition of water and slowly
claried with continued stirring (5–15 min between additions).
A total of ꢂ2.5 ml of water was added, at which point the
solution remained turbid even with prolonged stirring. The
turbid solution was taken up into a syringe and injected onto
reverse-phase prep-HPLC through a 0.45 mm lter in no more
than 2.0 ml batches (ꢂ2 injections per peptoid).
Solution thermodynamics
Absorbance spectra for spectrophotometric titrations were
recorded on an Ocean Optics USB 4000 spectrophotometer (slit
50 nm, grating 600 grooves per mm, blaze 400 nm) equipped
with a PX-3 pulsed xenon or HPX-2000 high-powered xenon
light source. Spectral data were acquired with a 1 cm path
length dip probe (Ocean Optics, Inc.). A microCombi Electrode
(Metrohm) H+-response electrode was calibrated prior to each
titration and used in conjunction with a Metrohm Titrando 907
to measure the pH of the experimental solutions. Absorption
and pH measurements were taken aer each incremental
addition of carbonate-free 0.1 M KOH (prepared from Baker
Dilut-It concentrate) by the Metrohm autoburet. The instru-
ment set-up was completely automated by in-house titration
soware. All spectrophotometric titration samples were
prepared under acidic conditions (ꢂpH 1) with 25 mM ligand
buffered in HEPES, CHES, MES, and acetic acid (3 mM each) in
0.1 M ionic strength (KCl supporting electrolyte) at 25ꢃ under
positive argon pressure. Solutions used for stability constants
had 1 equivalent of metal (M ¼ Eu3+). Data from each titration
were imported and rened by the nonlinear least squares tting
program HypSpec (see ESI†). The stability constant of Eu:3,4,3-
LI(HCCH) complex was determined by uorimetric batch
titrations. Solutions of 100 nM Eu:3,4,3-LI(HCCH), with varying
equivalents of 3,4,3-LI(1,2-HOPO) (ꢂ0.6–32) buffered in 50 mM
HEPES at pH 7.4 at 0.1 M ionic strength (KCl supporting elec-
trolyte) were prepared, allowed to equilibrate at 25ꢃ for at least 1
day (until no spectral change could be noted) and emission
spectra recorded. The analytical signal used was from the
Eu:3,4,3-LI(1,2-HOPO) complex, as previously described.24
Photophysical measurements
UV-visible absorption spectra were recorded on a Varian Cary
5G double-beam absorption spectrophotometer, using either
1.00 cm or 2.0 mm path length quartz cells. Each of the
measured peptoid solutions was 45.4 mM (100 mM for 3,4,3-LI
ligands) with 1.1 equivalents of metal (M ¼ Eu3+ or Tb3+), to
ensure the absence of free ligand, and buffered in 50 mM
HEPES at pH 7.4. Fluorescence and time-resolved measure-
ments were acquired on a HORIBA Jobin Yvon IBH Fluorolog-3
spectrophotometer. The excitation source was a xenon ash
lamp. A 400 nm long-pass lter was used when the second
harmonic of the excitation beam overlapped with the emission
spectra. For luminescence measurements, the ligands (5.0 mM)
were mixed with 0.9 equivalents of metal (M ¼ Eu3+ or Tb3+) to
ensure full metal complexation in 50 mM HEPES buffer (pH
7.4), and excited at the absorption maximum for each of the
Author contributions
complex. Quantum yield measurements were obtained using IC, AR, and RJA designed the research. IC developed the
the optical dilution method with quinine sulfate as a reference. synthetic pathway for peptoid ligands and IC, AR, and BN
Chem. Sci.
This journal is © The Royal Society of Chemistry 2019

View Article Online
Edge Article
Chemical Science
prepared all ligands. AR and KPC performed the spectroscopic 10 G. J.-P. Deblonde, M. P. Kelley, J. Su, E. R. Batista, P. Yang,
and thermodynamic analyses of metal complexes. GJPD assis-
ted with experimental design of solution thermodynamic
C. H. Booth and R. J. Abergel, Angew. Chem., Int. Ed., 2018,
57, 4521–4526.
measurements and interpretation of thermodynamic and 11 K. P. Carter, R. G. Surbella III, M. Kalaj and C. L. Cahill,
spectroscopic data. All authors discussed the experimental Chem.–Eur. J., 2018, 24, 12747–12756.
results and contributed to the manuscript. All authors have 12 G. J. P. Deblonde, M. Sturzbecher-Hoehne and R. J. Abergel,
given approval to the nal version of the manuscript.
Inorg. Chem., 2013, 52, 8805–8811.
13 M. Sturzbecher-Hoehne, T. A. Choi and R. J. Abergel, Inorg.
Chem., 2015, 54, 3462–3468.
14 I. Captain, G. J. P. Deblonde, P. B. Rupert, D. D. An, M.-C. Illy,
E. Rostan, C. Y. Ralston, R. K. Strong and R. J. Abergel, Inorg.
Chem., 2016, 55, 11930–11936.
15 T. A. Pham, A. B. Altman, S. C. E. Stieber, C. H. Booth,
S. A. Kozimor, W. W. Lukens, D. T. Olive, T. Tyliszczak,
J. Wang, S. G. Minasian and K. N. Raymond, Inorg. Chem.,
2016, 55, 9989–10002.
Conflicts of interest
IC and RJA are listed as inventors on a patent application led
by the Lawrence Berkeley National Laboratory and describing
inventions related to the research results presented here. The
authors declare no other competing nancial interest.
Acknowledgements
16 G. M. Figliozzi, R. Goldsmith, S. C. Ng, S. C. Banville and
R. N. Zuckermann, in Methods in Enzymology, Academic
Press, 1996, vol. 267, pp. 437–447.
17 A. S. Knight, E. Y. Zhou, J. G. Pelton and M. B. Francis, J. Am.
Chem. Soc., 2013, 135, 17488–17493.
Early peptoid library development was supported by the Fuel
Cycle Research and Development Campaign (FCRD)/Fuel
Resources Program, Office of Nuclear Energy, the U.S. Depart-
ment of Energy (DOE) under Contract No. DE-AC02-05CH11231
at the Lawrence Berkeley National Laboratory (LBNL).
Spermine-based ligand development and all spectroscopic
characterization were supported by the DOE, Office of Science
Early Career Research Program and Office of Science, Office of
Basic Energy Sciences, Chemical Sciences, Geosciences, and
Biosciences Division at LBNL under Contract DE-AC02-
05CH11231. Peptoid ligand purication and analysis were per-
formed at the Molecular Foundry, a User Facility supported by
the Office of Science, Office of Basic Energy Sciences, of the DOE
under Contract No. DE-AC02-05CH11231. The authors thank
Drs Ronald Zuckermann and Michael Connolly from the LBNL
Molecular Foundry for invaluable help with peptoid
preparation.
18 A. S. Knight, E. Y. Zhou and M. B. Francis, Chem. Sci., 2015, 6,
4042–4048.
19 M. Baskin and G. Maayan, Chem. Sci., 2016, 7, 2809–2820.
20 B. F. Parker, A. S. Knight, S. Vukovic, J. Arnold and
M. B. Francis, Ind. Eng. Chem. Res., 2016, 55, 4187–4194.
21 A. S. Knight, R. U. Kulkarni, E. Y. Zhou, J. M. Franke,
E. W. Miller and M. B. Francis, Chem. Commun., 2017, 53,
3477–3480.
22 M. Baskin and G. Maayan, Dalton Trans., 2018, 47, 10767–
10774.
23 M. Sturzbecher-Hoehne, C. Ng Pak Leung, A. D'Aleo,
B. Kullgren, A.-L. Prigent, D. K. Shuh, K. N. Raymond and
R. J. Abergel, Dalton Trans., 2011, 40, 8340–8346.
´
24 R. J. Abergel, A. D'Aleo, C. Ng Pak Leung, D. K. Shuh and
K. N. Raymond, Inorg. Chem., 2009, 48, 10868–10870.
25 L. J. Daumann, D. S. Tatum, B. E. R. Snyder, C. Ni, G.-l. Law,
E. I. Solomon and K. N. Raymond, J. Am. Chem. Soc., 2015,
137, 2816–2819.
References
¨
1 J.-C. G. Bunzli, Coord. Chem. Rev., 2015, 293–294, 19–47.
¨
2 J.-C. G. Bunzli, in Handbook on the Physics and Chemistry of 26 W. T. Carnall, P. R. Fields and K. Rajnak, J. Chem. Phys.,
¨
Rare Earths, ed. J.-C. G. Bunzli and V. K. Pecharsky,
1968, 49, 4424–4442.
Elsevier, 2016, ch. 287, vol. 50, pp. 141–176.
27 K. Binnemans, Coord. Chem. Rev., 2015, 295, 1–45.
3 C. S. Cutler, H. M. Hennkens, N. Sisay, S. Huclier-Markai and 28 M. Latva, H. Takalo, V.-M. Mukkala, C. Matachescu,
´
S. S. Jurisson, Chem. Rev., 2013, 113, 858–883.
4 R. D. Teo, J. Termini and H. B. Gray, J. Med. Chem., 2016, 59,
6012–6024.
5 A. J. Amoroso, I. A. Fallis and S. J. A. Pope, Coord. Chem. Rev.,
2017, 340, 198–219.
J. C. Rodrıguez-Ubis and J. Kankare, J. Lumin., 1997, 75,
149–169.
29 V. V. Utochnikova and N. P. Kuzmina, Russ. J. Coord. Chem.,
2016, 42, 679–694.
30 M. Congiu, M. Alamiry, O. Moudam, S. Ciorba,
P. R. Richardson, L. Maron, A. C. Jones, B. S. Richards and
N. Robertson, Dalton Trans., 2013, 42, 13537–13545.
6 J. A. Rees, G. J. P. Deblonde, D. D. An, C. Ansoborlo,
S. S. Gauny and R. J. Abergel, Sci. Rep., 2018, 8, 4419.
7 A. E. V. Gorden, J. Xu, K. N. Raymond and P. Durbin, Chem. 31 X. Q. Song, Q. F. Zheng, L. Wang and W. S. Liu, Luminescence,
Rev., 2003, 103, 4207–4282.
2012, 27, 459–465.
8 L. S. Natrajan, A. N. Swinburne, M. B. Andrews, S. Randall 32 R. D. Archer, H. Chen and L. C. Thompson, Inorg. Chem.,
and S. L. Heath, Coord. Chem. Rev., 2014, 266–267, 171–193. 1998, 37, 2089–2095.
9 G. J. P. Deblonde, M. Sturzbecher-Hoehne, P. B. Rupert, 33 S. Sato and M. Wada, Bull. Chem. Soc. Jpn., 1970, 43, 1955–
D. D. An, M.-C. Illy, C. Y. Ralston, J. Brabec, W. A. de Jong,
1962.
R. K. Strong and R. J. Abergel, Nat. Chem., 2017, 9, 843–849.
This journal is © The Royal Society of Chemistry 2019
Chem. Sci.

View Article Online
Chemical Science
Edge Article
34 T. Kimura, R. Nagaishi, Y. Kato and Z. Yoshida, J. Alloys 36 M. A. Deri, S. Ponnala, P. Kozlowski, B. P. Burton-Pye,
Compd., 2001, 323–324, 164–168.
H. T. Cicek, C. Hu, J. S. Lewis and L. C. Francesconi,
35 R. J. Abergel, P. W. Durbin, B. Kullgren, S. N. Ebbe, J. Xu,
Bioconjugate Chem., 2015, 26, 2579–2591.
P. Y. Chang, D. I. Bunin, E. A. Blakely, K. A. Bjornstad, 37 L. C. Uhlir, P. W. Durbin, N. Jeung and K. N. Raymond, J.
C. J. Rosen, D. K. Shuh and K. N. Raymond, Health Phys.,
2010, 99, 401–407.
Med. Chem., 1993, 36, 504–509.
¨
38 C. Wurth, M. Grabolle, J. Pauli, M. Spieles and U. Resch-
Genger, Nat. Protoc., 2013, 8, 1535.
Chem. Sci.
This journal is © The Royal Society of Chemistry 2019