On-Demand Degradation of Metal-Organic Framework Based on Photocleavable Dianthracene-Based Ligand

Article abstract of DOI:10.1021/jacs.8b05047

We have designed a rigid photocleavable dianthracene-based ligand that reacts with ytterbium as coordination metal ion for the creation of a class of tridimensional light-degradable metal-organic framework (MOF). We demonstrated that we can obtain a high level of control on the disassembly of the MOF formed with this ligand which can be triggered either through light irradiation or temperature increase. The reversible 4π-4π photodimerization is the intrinsic chemical mechanism ruling the ligand and MOF cleavage. In the fields of biology and medicine, MOFs have sparked a strong interest as highly porous vehicles for drug release but have only been explored so far through the passive leakage of their payloads. The designed light-degradable MOFs can potentially overcome this limitation and serve as prototypes for drug delivery and corresponding therapeutic applications.

Full text of DOI:10.1021/jacs.8b05047

Article
Cite This: J. Am. Chem. Soc. XXXX, XXX, XXX−XXX
pubs.acs.org/JACS
On-Demand Degradation of Metal−Organic Framework Based on
Photocleavable Dianthracene-Based Ligand
Guillaume Collet,* Timothee Lathion, Celine Besnard, Claude Piguet, and Step
́ ́ ́
hane Petoud*,†,§
,
†
†
‡
†
†
‡
Department of Inorganic and Analytical Chemistry and Laboratory of Crystallography, University of Geneva, 30 Quai E Ansermet,
CH-1211 Geneva 4, Switzerland
§
Centre de Biophysique Moleculaire (CBM), UPR CNRS 4301, Rue Charles Sadron, F-45071 Orleans 2, France
́ ́
*
S Supporting Information
ABSTRACT: We have designed a rigid photocleavable
dianthracene-based ligand that reacts with ytterbium as
coordination metal ion for the creation of a class of
tridimensional light-degradable metal−organic framework
(
MOF). We demonstrated that we can obtain a high level
of control on the disassembly of the MOF formed with this
ligand which can be triggered either through light irradiation
or temperature increase. The reversible 4π-4π photodimeriza-
tion is the intrinsic chemical mechanism ruling the ligand and
MOF cleavage. In the fields of biology and medicine, MOFs
have sparked a strong interest as highly porous vehicles for drug release but have only been explored so far through the passive
leakage of their payloads. The designed light-degradable MOFs can potentially overcome this limitation and serve as prototypes
for drug delivery and corresponding therapeutic applications.
INTRODUCTION
couple of works report strategies to entrap the payload inside
the MOFs pores such as using hydrophobic molecules or
taking advantage of specific interactions within the cavities of
the MOF without providing an effective control over the time
■
Facing side effects related to detrimental interactions of active
drugs (i.e., anti-inflammatory, chemotherapies) with healthy
tissues or organs, active research is performed to design
vehicles able to safely carry the therapeutic compounds inside
the body. Properly addressed at the place to cure by targeting
or EPR effect (enhanced permeability and retention), such
materials aim to restrict the delivery of drugs exclusively to the
pathological tissue/organ. In such context, the loading capacity
plays a central role for the efficiency of the delivery.
19
and location of the releases of payloads. We envisage as the
major next breakthrough for the use of MOF in the fields of
biology and/or medicine a specific mechanism to control the
delivery of payloads at the right place and at the right moment.
The controlled disassembly of materials has been extensively
studied since the last decades to fulfill this demand. It has been
2
0−22
successfully applied to polymeric nanoparticles,
mi-
Initially described for gas storage, separation, catalysis, and
23,24
25,26
27−29
30
2
−5
celles,
liposomes,
or hydrogels
among others,
sensing,
metal−organic frameworks (MOFs) are crystalline
reporting such smart materials as carriers for drugs delivery or
materials that were recently introduced in the field of biology
and medicine for their high porosity offering unprecedented
31,32
as activatable probes for imaging.
These studies report
6
,7
interesting mechanisms with materials designed to specifically
respond to endogenous triggers such as acidosis or oxidative
stress linked to inflammation and tumors but also to external
loading capacity of therapeutic payloads. As an example, an
impressive demonstration has been reported by Horcajada et
al. for MIL-101 containing up to 1.376 g of ibuprofen per gram
22,30,33
9
triggers such as light irradiation.
External triggers such as
of dehydrated MOF. Several types of such materials are
light have the advantage of providing an “on-demand” release
biocompatible and can be miniaturized to reach a nanometer
range size compatible with applications in living organ-
30
with a high level of control over time and location. Moreover,
the degradability of such materials is a great advantage for their
administration in living organisms because particles will fall
apart as small size fragments that can be cleared out of the
body (i.e., urine), thus avoiding or reducing their massive
1
0−13
isms.
Obtained by the self-assembly between organic
ligands and metals, MOFs can be produced in large amounts
and are easily purified and characterized due to their crystalline
nature.
Despite all these advantages as carriers, they have so far only
a limited development in therapeutic applications. Indeed,
MOF have so far mainly been reported as “passive sponges”
releasing their payloads out of the pores by slow diffusion.
New classes of responsive MOF that undergo changes resulting
from external stimulation have emerged recently.
34,35
accumulation in filtering organs such as the liver.
In this report, we bridge the gap between light-degradable
materials and metal−organic frameworks by synthesizing a
1
4,15
Received: May 14, 2018
1
6−18
A
©
XXXX American Chemical Society
A
DOI: 10.1021/jacs.8b05047
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 1. Overview of the light-degradable MOF behavior upon light irradiation. In this schematic representation, CD-MOF-161, which is a
photocleavable metal−organic framework, is exposed to a UV light beam (wavelength below 300 nm). The light reaching the CD-MOF-161 crystal
induces cleavage of the ligands located within the crystal structure which falls apart.
4
,41,42
MOF that is made photocleavable on-demand due to the light-
responsive nature of its organic ligand. This ligand is
photocleaved upon light exposure at a specific wavelength
yield the corresponding anthracene dimers.
Due to the
Cs-symmetry point group adopted by 9-anthracenecarboxylic
acid (9AC), two geometrical isomers of the dimeric 9-
anthracenecarboxylic acid (Di-9AC) could be formed: the
head-to-head (hh) dimer, and the head-to-tail (ht) dimer
resulting from the stacked (hh) or opposite (ht) arrangements
of the carboxylic groups. According to previous studies, the
formation of the hh dimer in solution is prevented by the steric
hindrance occurring between the two carboxylic acid side
chains, and, as a result, only the ht dimer is obtained (Scheme
(
<300 nm) or a rise in temperature, as illustrated in Figure 1. If
some works describe photoactive behaviors of MOFs such as
the already described UiO-AZB, MIL-100, and UiO-66
or with the Ph-An-COF presenting a photoinduced structural
36
37
38,39
42
4
0
change, to the best of our knowledge, no such system has
been reported so far. The challenge was to achieve such
control with a ligand possessing the rigidity required for the
formation of MOFs. Among molecular mechanisms reported
for controlled degradation, anthracene and its derivatives are
interesting because they undergo a light-driven reversible 4π-
1
1). The H NMR chemical shift of the bridgehead proton
a
Scheme 1. Reversible Photodimerization Mechanism
4
π photodimerization leading to the formation of rigid
photodimers upon light irradiation above 350 nm whereas
wavelengths below 300 nm induce their photocleavages.
To establish a proof-of-principle, 9-anthracenecarboxylic
acid (9AC) was photodimerized by irradiation at 365 nm to
yield the 9-anthracenecarboxylic acid dimer (Di-9AC). Its
light-induced cleavage was extensively tested using two
irradiation sources: a continuous UV lamp at 254 nm and a
pulsed laser at 266 nm. The temperature-induced cleavage of
anthracene dimers was also tested with Di-9AC. After full
characterization and validation of its ability to be cleaved in a
triggered way, the synthesized Di-9AC was used as a ligand for
MOF synthesis. As a result, a novel tridimensional crystalline
network was formed with the two carboxylic groups of the Di-
a
Overview showing the photodimerization of 9AC into Di-9AC under
a wavelength superior to 350 nm UV light exposure and the
photocleavage when exposed either to UV light below 300 nm or to a
rise in temperature.
allows for an unambiguous differentiation in solution between
the two isomers due to distinct chemical environments. Based
on works published by Ito et al., the chemical shift (DMSO-d6)
of the bridgehead proton in the ht dimer is δ = 5.63 ppm.
These results match with our own, as we observe a chemical
shift for the bridgehead proton at 5.62 ppm in DMSO-d6
(
ht Di-9AC dimer in solution. This ht dimer is preferred
because diametrically opposed carboxylic acid groups con-
stitute a favorable shape for the formation of MOFs. The
formation of the dimeric 9-anthracenecarboxylic acid (Di-
9
AC ligands, bridging a dinuclear clusters of ytterbium used as
coordination metal ions. This 3D lanthanoid metal−organic
framework (Ln-MOF) was named CD-MOF-161, for con-
trolled degradable metal−organic framework, and its structure
was characterized by single-crystal X-ray diffraction. To
investigate the controlled disassembly of CD-MOF-161, the
same light sources mentioned for the Di-9AC ligand studies
41
Figure S1), indicating the exclusive formation of the desired
(
254 and 266 nm) and a rise in temperature (160 °C) were
used as triggers for the degradation. In both cases, the triggers
led to the photocleavage of Di-9AC in the CD-MOF-161
structure that fell apart.
9
AC) in the desired ht orientation was confirmed by X-ray
diffraction on collected crystals (Tables S1−3 and Figures S5−
RESULTS AND DISCUSSION
1
13
■
7) and by H and C NMR spectra in solution (Figures S1−
4). The Di-9AC formation was also studied by UV−vis and
FTIR absorbance spectroscopies. The comparison of the FTIR
spectra for 9AC (Figure S9) and Di-9AC (Figure S8) reveals
Synthesis and Characterization of the Di-9AC Ligand.
Anthracenes and their derivatives are known to undergo a [4 +
] photodimerization upon light irradiation (λ > 350 nm) to
4
B
DOI: 10.1021/jacs.8b05047
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
1
Figure 2. Controlled degradability of the Di-9AC ligand. (a) H NMR spectra of Di-9AC collected after increasing irradiation times with a pulsed
1
laser at 266 nm wavelength. (b) Quantification by H NMR of the photocleavage of Di-9AC into 9AC and byproducts obtained at different
irradiation times at 266 nm. (c) UV−vis absorbance spectra showing the appearance over time of 9AC upon the irradiation of Di-9AC at 266 nm.
(d) Quantification by UV−vis absorbance of 9AC at different irradiation times at 266 nm.
Figure 3. Structure of the controlled degradable MOF. Pictures of the crystal structure of (a) the asymmetric unit of CD-MOF-161, and (b) the 3D
framework of CD-MOF-161 with cell (blue). (a, b) Hydrogen atoms are omitted for clarity.
the typical signals reported by More et al. with an intense
vibration at 1682 cm−1 assigned to the typical CO stretching of
This observation is due to the photodimerization process that
disrupts the conjugation of anthracene, generating four o-
44
9
AC and with a series of broad vibrational signals in the 2800
xylene chromophores. Similarly, in the solid-state diffuse
reflectance spectrum, no signal can be detected from 300 to
400 nm for Di-9AC (Figure S11), indicating that the behaviors
are the same in presence and in absence of solvents.
−
1
to 3650 cm range assigned to CH and OH stretching
vibrations. The formation of Di-9AC is revealed by
disappearance of the signal at 1682 cm−1 corresponding to
43
the CO stretching vibration observed on the 9AC monomer
Figures S8 and S9). Moreover, the presence of two well-
Di-9AC Cleavage upon Light Exposure or Rise in
(
Temperature. The cleavability of synthesized Di-9AC was
−1
1
defined vibrations at 1450 and 1470 cm (splitting of C−O
bonding) is a good indication of the presence of an anthracene
first validated by thermocleavage experiments. H NMR
experiments showed that the incubation of the Di-9AC
dimer at 130 °C leads to its gradual degradation over time
into the 9AC monomer (Figure S12). This degradation profile
was confirmed by UV−vis absorbance spectroscopy experi-
4
4
photodimer. The UV−vis absorption spectrum of Di-9AC
does not show the characteristic absorption bands between
3
00 and 400 nm, which can be observed for 9AC (Figure S10).
C
DOI: 10.1021/jacs.8b05047
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 4. Characterization of the controlled degradable MOF. (a) Brightfield microscopy image of the as-synthesized CD-MOF-161. Scale bar is
00 μm. (b) CD-MOF-161 size distribution histogram (black) obtained from brightfield images analysis and Gaussian fitting (red). (c) A 0.803 μm
thickness optical section of a fluorescein-loaded CD-MOF-161 obtained by confocal fluorescence microscopy. Scale bar is 10 μm. (d) Experimental
1
1
(
continuous line) and calculated (dashed line) powder diffraction patterns. The calculated pattern was calculated with Mercury software using the
single-crystal structure recorded at 180 K and correcting the unit-cell parameters for the temperature difference. The room-temperature unit-cell
8
parameters were obtained doing a Le Bail profile fitting in the TOPAS software of the experimental diagram and are a = 16.84 Å, b = 18.33 Å, c =
19.52 Å, α = 85.671°, β = 79.06°, γ = 65.63°.
ments, quantifying the appearance of the characteristic bands
of 9AC between 300 and 400 nm as shown in Figure S13. The
photocleavability of Di-9AC was demonstrated with two
different irradiation sources, respectively, a 266 nm pulsed
laser and a continuous 254 nm UV-Pen (spectra provided in
Figure S14). Di-9AC presents a high energy π → π*
absorption at these two wavelengths that allows its photo-
induced cleavage of Di-9AC over the light-irradiation, no
byproducts were observed (Figure 2 and Figures S15 and S16).
Synthesis of the Di-9AC-Based Metal−Organic
Framework and Analysis of Its Structure. Having
confirmed the formation of the pure ht Di-9AC isomer and
its controlled degradability, a MOF was successfully assembled
3
+
with ytterbium (Yb ) as coordination metal ion to give a
4
5
1
m a t e r i a l w i t h t h e f o r m u l a [ Y b ( D i -
cleavage. In both cases, this process was followed by H
NMR (Figure 2a and Figure S15a) to determine the speciation
curves presented in the Figure 2b (Figure S15b). The
irradiation-driven cleavage of Di-9AC is clearly demonstrated
with the concomitant appearance of 9AC and some
byproducts. The photocleavage was also followed by UV−vis
absorbance spectroscopy (Figure 2c,d and Figure S15c,d)
showing the gradual appearance of the typical bands of 9AC
with increasing irradiation time. The quantification of 9AC on
the basis of UV−vis spectra is perfectly correlated with the
NMR data (Figure 2b,d and Figure S15b,d). A faster
degradation kinetic was observed upon laser irradiation
compared to UV-Pen in direct correlation with the irradiation
power. Because the temperature is known to rule the cleavage
2
9
AC) (DMA) (acetate)0.76(formate)1.24] (Figure 3b). This
2 2
novel MOF assembled from this Di-9AC ligand was named
CD-MOF-161 (controlled degradable metal−organic frame-
work). CD-MOF-161 crystallizes in the triclinic space group
P1. The asymmetric unit is shown in Figure S17. Interestingly,
̅
the CD-MOF-161 contains Yb dinuclear clusters as secondary
3+
building units, as shown in Figure 3a. The two Yb ions of the
cluster are bridged by one acetate and one carboxylate group
belonging to one Di-9AC ligand. A second acetate shares one
oxygen atom located between the two ytterbium atoms while
his second oxygen atom is only bound to one of the ytterbium
atom (Yb1). Yb2 is therefore coordinated to one Di-9AC
(
O11, O12), one bridging Di-9AC (O10), one DMA molecule
(O14), one bridging acetate (O9), one shared acetate (O8,
O15), and a water molecule (O13). On its side, Yb1 is
coordinated to two Di-9AC (O1 O2 and O3 O4), the bridging
Di-9AC (O6), the bridging acetate (O5), the shared acetate
45
1
of the dimeric anthracene, similar studies by H NMR and by
UV−vis absorbance spectroscopy were realized upon heating
at 130 °C (Figure S16). As an advantage of the temperature-
D
DOI: 10.1021/jacs.8b05047
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 5. Controlled degradation of the light degradable MOF. (a) Brightfield microscopy images of a CD-MOF-161 suspension irradiated during
various times with a pulsed laser at 266 nm wavelength. Scale bar is 200 μm. (b) Quantification of the number of CD-MOF-161 particles along the
irradiation time with a 266 nm pulsed laser obtained from the analysis of brightfield microcopy images (black) and exponential decay fitting (red).
(
c) UV−vis absorbance spectra of 9AC resulting from the thermocleavage of remaining CD-MOF-161 collected after various times of irradiation
with a pulsed laser at 266 nm wavelength. (d) Quantification by UV−vis absorbance spectroscopy of the ligand of the remaining CD-MOF-161
after the irradiation times with a 266 nm pulsed laser (black) and exponential decay fitting (red).
−
1
(
O15), and one DMA molecule (O7). This analysis reveals a
carboxylate anion appear in the MOF around 1406 cm and
−
1
total coordination number of 8 for both Yb1 and Yb2 with a
Yb−Yb distance of 4.2584(5) Å. The Yb−O distances span
from 2.242(3) to 2.451(6) Å (Table S5). Crystal data for CD-
MOF-161 are summarized in Table S4.
around 1558−1597 cm . These data reveal changes of the Di-
3+
9AC ligands in the MOF when coordinated with Yb
47
cations.
The dinuclear clusters are interconnected to form a 3D
network. A topological analysis was carried out in order to gain
a deeper level of understanding of the nature of this
framework. Clusters forming four-connected vertexes are
connected via the Di-9AC ligands forming two connected
edges (Figure S20). This organization results in a diamond net
with RCSR (Reticular Chemistry Structure Resource database)
symbol dia. The packing of the clusters in the MOF and the
underlying net are depicted in Figure S21. The Di-9AC ligand
maintains its original shape in the structure without any
distortion or π−π stacking effect. From the perspectives of the
crystallographic directions, three types of 1D microporous
channels can be found in CD-MOF-161, as shown in Figure
S22. These channels are interconnected. Using PLATON
46
The SHAPE software was used to analyze the shape of the
coordination polyhedron around lanthanide cations by
calculating continuous shape measures (CShM) between the
observed coordination polyhedrons and idealized ones. The
shape is closely related to several ideal polyhedrons (see table
in Figure S18), but one can conclude that the Yb atom is in a
distorted dodecahedral environment. The two polyhedrons
share a corner to form the dinuclear cluster, with an Yb−Yb
distance of 4.2584(5) Å. The Yb−O distances span from
48
2
.242(3) to 2.451(6) Å (Table S5).
Experimental data showing the coordination of Yb ions
3
+
with the Di-9AC ligand were obtained by comparing results of
FTIR absorption spectra of the free Di-9AC and of the MOF
formed with this dimer (Figure S19). The carboxylic groups
are the main groups in the structure of Di-9AC which undergo
a transformation during the deprotonation and metal
complexes formation. The broad band at 3032 cm assigned
to the ν(OH) stretching vibrations of the carboxylic acid
groups and the bands around 1700 cm−1 assigned to the
stretching vibrations ν(C = O) of carboxylic groups of Di-9AC
disappear in the spectrum collected on the MOF due to the
complete deprotonation of carboxylic groups and coordination
49
software, the potential solvent volume was estimated to fill
about 47% of the unit-cell volume on the basis of a probe size
of 1.2 Å and a grid of 0.2 Å.
−
1
Crystals collected after synthesis were studied by powder X-
ray diffraction (PXRD) to compare the patterns obtained
experimentally with the one computed from the single crystal
structure. The theoretical peaks are clearly identified in the
experimental sample as shown in Figure 4d, and no additional
diffraction peaks were found, indicating the synthesis of a
unique crystalline entity. This result, in combination with
results of elemental analysis fitting well with the molecular
3
+
with Yb metal ions. Strong bands assigned to the asymmetric
ν (COO ) and symmetric ν (COO ) stretching vibrations of
−
−
as
s
E
DOI: 10.1021/jacs.8b05047
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
formula of the measured structure for CD-MOF-161, allows us
to conclude that our synthesis method leads to a unique and
pure MOF.
the individualization of some aggregated CD-MOF-161
crystallites present after the synthesis.
Compared to light-induced degradation, the rise in temper-
ature up to 160 °C induces the thermocleavage of the ligand
and the resulting MOF disassembly with the same decreasing
exponential profile. However, before its exponential decrease,
an increase of a factor 15 of the initial number of particles is
observed (Figure S25a). This important increase can be due to
the mechanism of degradation involved and the type of
employed trigger (either light irradiation or rise in temper-
ature). The light-induced cleavage of CD-MOF-161 results
mainly from an erosion process by exposing the surface of
crystals to light irradiation whereas the temperature is
transmitted to all the CD-MOF-161 crystals and is not
restricted to their surfaces. The thermocleavage of ligand inside
the CD-MOF-161 structure compromises its structural
integrity, resulting in fragmentation processes of the crystals.
Molecular evidence for the controlled degradation were
obtained through the quantification by UV−vis absorbance
spectroscopy of the Di-9AC ligand from remaining undegraded
CD-MOF-161 (see methodology in Figure S27). To do so, at
each time point of the kinetic experiment, the remaining CD-
MOF-161 was collected and thermocleaved at 160 °C for 48 h,
resulting in (i) the complete disassembly of CD-MOF-161
crystals, and (ii) the complete cleavage of Di-9AC ligands. The
resulting 9AC and its typical bands were quantified and plotted
as function of time to obtain the profile represented in Figure
Thermogravimetric analysis (TGA) of the dried MOF
showed a first weight loss from 20 °C to 200 °C which is
attributed to the loss of a molecule of water, DMA, and
formate (two molecules each) (−14% experimental weight
loss, −18% calculated). A second weight loss from 200 °C to
5
50 °C is attributed to the loss of the remaining organic
content of the MOF which correspond to the two Di-9AC
ligands (−60% experimental weight loss, −58% calculated).
Above 550 °C and up to 1000 °C, no weight loss was
observed, with an experimental remaining weight of 26%,
corresponding to the inorganic content of the MOF (i.e., two
Yb(III) as Yb O , calculated remaining weight: 26%) (Figure
2
3
S23).
Brightfield optical microscopy analysis was used to evaluate
the size distribution of CD-MOF-161. Quantifications on
experimental pictures show an average size of 11.97 μm with a
polydispersity of 58.84% (Figure 4a and 4b).
To visualize the microscopic structure of crystals, an optical
sectioning was obtained by confocal microscopy of a CD-
MOF-161 incorporating fluorescein. A picture is presented in
Figure 4c for a 0.803 μm thickness section. To do so,
fluorescein molecules were incorporated in the MOF during
the synthesis. This confocal imaging method allows us to
observe only a thin section of the crystal without the
contributions of the planes located above and below the
imaging plane. Such observation of the MOF through the
fluorescence of the fluorescein it contains reveals the shape of
the crystal in the plane of the imaging. This result indicates
that the fluorescent dye is located in the core of the MOF
therefore demonstrating that its pores can be loaded with such
molecules. No change in the crystalline structure is observed
when pores contain fluorescein molecules (Figure S28).
The potential toxicity of the CD-MOF-161 particles and the
5
c,d (and Figures S24c,d and S25c,d). These data combined
with results of the particles counting show that CD-MOF-161
degrades upon trigger addition, either with light irradiation
(
<300 nm) or rise in temperature.
CONCLUSION
■
By designing a photocleavable dianthracene-based ligand, we
created a system for which we obtained a high level of control
for the disassembly of the corresponding MOF which can be
triggered either through light irradiation or temperature
increase. The Di-9AC ligand has been extensively characterized
using different experimental techniques. We have demon-
strated that the cavity of this MOF can be used to incorporate
molecules such as fluorescein.
Yb(OAc) was assessed by a series of cytotoxicity experiments
3
on a model cell line (HeLa cells). Results show that the
presence of ytterbium acetate up to 1 mM does not induce any
significant cell toxicity after 24 h and 48 h of incubation. A
toxicity associated with CD-MOF-161 was observed when a
large number of MOFs were present with cells. Interestingly,
MOFs that have been washed with water appear to be less
toxic. This observation could be associated with the release of
DMA (synthesis solvent) from the pores through this washing
process (Figure S29−31).
Controlled Disassembly of the MOF upon Light
Irradiation or Temperature. The ability of CD-MOF-161
to degrade upon trigger addition, either light irradiation or rise
in temperature, was determined through a microscopy-based
experiment, quantifying the number of microparticles present
per field of view over time (Figure 5a and Figures S24a and
S25a). A suspension of CD-MOF-161 in DMA was exposed to
a 266 nm pulsed laser irradiation. Microscopy-based
quantifications reveal that the number of CD-MOF-161
particles exponentially decreases along the irradiation time as
presented in Figure 5b. Similar observations could be obtained
upon exposition to UV-Pen irradiation except that a longer
time is required to get all particles photocleaved due to the
lower power of the irradiation source (Figure S24a).
Interestingly, an increase by a factor 2 of the counted particles
was observed after a few minutes of irradiation resulting from
We successfully established in this work the first proof-of-
principle of a light-degradable metal−organic framework that is
obtained photocleavable by design and operating on-demand.
The use of light presents the advantage to be an external
stimulus that acts in a remote-control manner, turning ON and
OFF the degradation on-demand at desired times and
29
locations. Future developments of this MOF family will
open exciting perspectives toward advanced medicine and, in
particular, would allow researchers to remotely induce release
of molecules in living organisms, or, eventually, physicians to
spatiotemporally control drugs release in patients.
EXPERIMENTAL SECTION
■
Materials. Anthracene-9-carboxylic acid (9AC, 99%, Acros
Organics), tetrahydrofuran (THF, Fisher Scientific), deuterated
dimethyl sulfoxide (DMSO-d , D, 99.9%, Cambridge Isotope
6
Laboratories Inc.), ytterbium(III) acetate hydrate (Yb(OAc)3,
Aldrich), ammonium formate (Fluka), ammonia solution (25%,
VWR chemicals), fluorescein sodium salt (Fluka), N,N-dimethylace-
tamide (DMA, 99+%, Acros Organics), and acetonitrile (99.95%,
VWR chemicals) were used as received.
F
DOI: 10.1021/jacs.8b05047
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
General Procedures and Instrumentation. H and 13C NMR
spectra were recorded at room temperature on a Bruker Avance 400
MHz spectrometer. Chemical shifts are given in ppm with respect to
TMS. Pneumatically assisted electrospray (ESI-MS) mass spectra
were recorded on an Applied Biosystems API 150EX LC/MS system
equipped with a Turbo Ionspray source. Elemental analyses were
performed by the Microchemical Laboratory of the University of
Geneva (Unige) and the Supramolecular Chemistry Laboratory at the
1
X - r a y S t r u c t u r e D e t e r m i n a t i o n f o r
2 2 4 9 2 2 3 2 2
[Yb (C30H18O) (C H NO) (C H O )0.76(CHO )1.24]. A single-crystal
of CD-MOF-161 was deposited into a Mitegen cryoloop in fomblin
oil. Data were collected at 180 K on an Agilent Supernova
diffractometer equipped with an Atlas CCD detector. Details of the
53
52
crystallographic refinement, carried out in SHELXL using Olex2,
can be found in Supporting Information Table 4. Two disordered
DMA molecules are coordinated to the metals. They were refined
using two components. Geometrical restraints (SADI for the Yb−O
distances) were applied, as well as restraints (RIGU)/constraints
(EADP on close atoms) on anisotropic displacement parameters.
Acetate/formate molecules are present in the coordination sphere of
the Yb atoms. A mixed acetate/formate model was therefore used.
DFIX was applied on the C−C acetate distance for one (C52−C53)
acetate. Restraints were applied on anisotropic displacement
parameters (RIGU). An oxygen atom is observed that can be either
attributed to a water molecule present in the coordination sphere of
Ecole Polytechnique Federale de Lausanne (EPFL). Thermogravi-
́ ́
metric analyses (TGA) were performed on a Mettler Toledo TGA/
DSC 1 STARe System under a nitrogen flow. Several milligrams of
sample were put in an aluminum oxide cell, and the cell was heated to
1
000 °C with a heating rate of 5 °C/min. The mathematical analyses
were performed using OriginPro8 (OriginLab) and Excel (Microsoft)
softwares.
Synthesis and Characterization of Di-9AC. The head-to-tail 9-
anthracene carboxylic acid dimer (Di-9AC) was prepared according
to the procedure published by Nofen et al. Briefly, 10.10 g of 9AC
45.45 mmol) was dissolved in 200 mL of degassed THF under
nitrogen atmosphere and vigorous magnetic stirring. The reaction was
maintained in darkness, and the solution was photoirradiated for 4
days using two UV bench lamps (λ = 365 nm, VL-6-LC, Vilber-
Lourmat, 6 W each, with a light density of approximately 610 μW/
42
3+
Yb or to a highly disordered DMA molecule for which only the
(
position of the oxygen atom is well-defined. No satisfactory model
could be found so the water molecule was chosen. Disordered solvent
molecules present in the channels of the MOF were taken into
54
account by using the squeeze/bypass method as implemented in
Olex2 software. Before this squeeze/bypass procedure, the R factors
2
3
cm located at a distance of 15 cm from the reaction, spectra in Figure
were R1 = 0.1090, wR2 = 0.2898. A void of about 2500 Å was found
S26). A white precipitate appeared which was collected by filtration
as a cavity of the MOF containing about 885 electrons. This value
corresponds to the presence of about 18 molecules of DMA per unit-
cell (nine per formula unit). The “squeezed” molecules were not
added to the formula unit so that the density and the absorption
coefficient are somehow lower than expected. A view of the
asymmetric-unit depicted at 50% probability is shown in Figure S17.
Light Irradiation for Photodegradation. A Quantel Nd:YAG
Q-smart 850 laser that is pulsed at 10 Hz was used to obtain photons
at 266 nm (selection of the fourth harmonic of the fundamental 1064
nm). The beam of this laser at 266 nm is around 7 mm in diameter
with a measured power of 1.30 W. As a continuous light source, a
short-wave UV quartz pencil lamp (Pen-Ray mercury light source,
Edmunds Optics) was used, emitting a light at 253.7 nm with an
using a Buchner funnel and washed repeatedly with cold THF (2 × 5
̈
mL) and cold water (2 × 5 mL). The collected solid was then dried
for 24 h under high vacuum to yield 5.59 g (12.59 mmol, η = 55%) of
1
Di-9AC as a white powder. H NMR (DMSO-d , 400 MHz, 298 K) δ
6
ppm: 13.60 (2H, bs), 6.90−6.85 (4H, m), 6.84−6.82 (8H, m), 6.78−
1
3
6
.74 (4H, m), 5.62 (2H, s). C NMR (DMSO-d , 100 MHz, 298 K) δ
6
ppm: 174.40, 142.60, 141.90, 127.46, 126.53, 125.77, 125.36, 65.52,
4.40. ESI-MS (soft neg, MeOH/H O/NEt 49.5/49.5/1.0, m/z):
5
2
3
−
−
[
Di-9AC-H] : 443.25 (exp), 443.13 (calcd); [(Di-9AC) -H] :
2
8
87.55 (exp), 887.27 (calcd). Elemental analysis calculated for
C H O : C 81.07, H 4.54; found: C 81.23, H 4.46.
30
20
4
X-ray Crystallography of Di-9AC. A slow evaporation of
2
concentrated DMA solutions (Table S3 and Figure S7) and DMSO
solution (Table S1 and Figure S5) of Di-9AC led to the formation of
X-ray quality prisms suitable for single crystal analysis and structure
determination. Other crystals of Di-9AC were obtained in DMA by
warming up the solution to 130 °C and then by cooling to room
temperature (Table S2 and Figure 5).
Optical Spectroscopy. Diffuse reflectance spectra were collected
on solid samples on a PerkinElmer Lambda 900 equipped with a 60
mm Labsphere integration sphere specifically designed for this
output power density of approximately 4.5 mW/cm .
MOF Sizing and Counting. The as-synthesized CD-MOF-161
size distribution was determined by a microscopy-based method using
a Tali Image-Based Cytometer (Invitrogen). Twenty-five microliters
of the MOF suspension were introduced to the measurement
chamber of an analysis slide. The Tali was set to acquire 20 fields
of view per sample. The analysis of the 20 acquired pictures were
performed by ImageJ software using the “analyze particles” plug-in on
previously binarized and thresholded pictures. The size distribution
was fitted by a Gaussian curve with OriginPro8 software. The mean
particle size and the polydispersity were determined from this fit.
Polydispersity (%) = standard deviation/average particles size × 100.
For degradation analysis of CD-MOF-161 over time upon trigger
addition (light or temperature), the same analysis was performed,
extracting the total count of particles per time point.
MOF Optical Imaging. Optical microscopy was realized at the
Bioimaging Center of the University of Geneva, Switzerland.
Brightfield microscopy images of MOF crystals were acquired with
an Axio Imager Z1m from Zeiss equipped with a high resolution
sCMOS Hamamatsu camera Flash4.0. Confocal fluorescence imaging
of fluorescein-loaded MOF crystals was performed with an inverted
confocal microscope Leica SP8. The fluorescein dye located inside of
the MOF pores was excited with a 488 nm laser while the emitted
fluorescence was selectively collected with the HyD Leica detector
from 515 to 545 nm. The pinhole was set to obtain optical sections of
0.803 μm thicknesses.
instrument. Absorbance spectra were recorded in solution (in H O
2
with 0.5% of NH ) with a PerkinElmer Lambda 1050 spectrometer
3
using quartz suprasil cells of 1 cm optical path length. For
quantifications (Figure 2d and Figures S15d, S16d), a post-acquisition
signal processing was applied to subtract the contribution of the UV-
absorbing byproducts generated during light irradiations. FTIR
spectra were measured using a Bruker Tensor 27 spectrometer.
Synthesis of CD-MOF-161. The reaction of ytterbium acetate
and Di-9AC was prepared in a 4:1 (metal:ligand) ratio in 2.3 mL of
DMA by mixing 68 μmol of Yb(OAc).H O (23.81 mg) with 17 μmol
2
of Di-9AC (7.56 mg). The mixture was sealed in a Pyrex tube, heated
at 60 °C for 2 days, and then cooled to room temperature at a rate of
2
°C/h. The resulting colorless block crystals were obtained and
washed with DMA to remove the unreacted materials. Larger crystals
were obtained by the addition of 138 μmol of ammonium formate
50,51
(8.70 mg) into the mixture prior heating.
Fluorescein-loaded
crystals of CD-MOF-161 were obtained by the addition of 5 μmol of
fluorescein sodium salt (1.88 mg) into the mixture prior heating. To
perform the elemental analysis, the as-synthesized CD-MOF-161
crystals were extensively washed with DMA and then with acetonitrile
Cell Cytotoxicity. HeLa cells (ATCC) were cultured in EMEM
medium (ATCC) supplemented with 10% (vol:vol) fetal bovine
serum (FBS, Sigma), 100 U/mL penicillin (Sigma), and 100 μg/mL
streptomycin (Sigma). Cells were routinely cultured at 37 °C in a
−
6
to be further dried for 48 h under high vacuum (8 × 10 mbar) to
yield between 12.5 and 12.9 mg of MOF. η = 96−99% (based on Di-
humidified incubator in a 95% air/5% CO atm and passaged by
2
9
AC). Elemental analysis calculated for C H N O Yb : C, 55.83;
detaching cells with 0.05% trypsin−EDTA solution (Gibco). HeLa
cells were seeded in a 96-well plate at 5000 cells/well in a final volume
7
1
60
2
15
2
H, 3.96; N, 1.83. Found: C, 56.04; H, 4.14; N, 1.81.
G
DOI: 10.1021/jacs.8b05047
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
of 200 μL of complete medium. After 24 h, cell culture medium was
removed and compounds to be tested were added at various
(2) Kreno, L. E.; Leong, K.; Farha, O. K.; Allendorf, M.; Van Duyne,
R. P.; Hupp, J. T. Chem. Rev. 2012, 112, 1105.
(3) Furukawa, H.; Cordova, K. E.; O’Keeffe, M.; Yaghi, O. M. Science
2013, 341, 974.
(4) Li, J. R.; Kuppler, R. J.; Zhou, H. C. Chem. Soc. Rev. 2009, 38,
1477.
(5) Murray, L. J.; Dinca, M.; Long, J. R. Chem. Soc. Rev. 2009, 38,
1294.
(6) Horcajada, P.; Chalati, T.; Serre, C.; Gillet, B.; Sebrie, C.; Baati,
T.; Eubank, J. F.; Heurtaux, D.; Clayette, P.; Kreuz, C.; Chang, J. S.;
Hwang, Y. K.; Marsaud, V.; Bories, P. N.; Cynober, L.; Gil, S.; Ferey,
G.; Couvreur, P.; Gref, R. Nat. Mater. 2010, 9, 172.
concentrations, either CD-MOF-161 or Yb(OAc) . The toxicity of
3
these compounds was evaluated after 24 h and 48 h of incubation by
the AlamarBlue Cell Viability Reagent (Invitrogen). Briefly,
compounds were removed and cells were washed once with PBS
2
+
2+
(
with Ca /Mg ) before the incubation at 37 °C with the
AlamarBlue reagent diluted to 1:10 in PBS, final volume of 200
μL/well. After 2−3 h at 37 °C, the fluorescence of the 96-well plates
3
TM
were read with a Victor V plate reader (PerkinElmer, excitation
filter 560/10 nm, emission filter 605/10 nm). Cell viability was
expressed in %, considering the untreated cells as the reference for
100% of viability.
(
7) Horcajada, P.; Gref, R.; Baati, T.; Allan, P. K.; Maurin, G.;
Couvreur, P.; Ferey, G.; Morris, R. E.; Serre, C. Chem. Rev. 2012, 112,
232.
8) Coelho, A. Topas Academic; Coelho Software: Brisbane,
Australia, 2017.
9) Horcajada, P.; Serre, C.; Vallet-Regi, M.; Sebban, M.; Taulelle,
F.; Ferey, G. Angew. Chem., Int. Ed. 2006, 45, 5974.
10) Foucault-Collet, A.; Gogick, K. A.; White, K. A.; Villette, S.;
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/jacs.8b05047.
1
(
■
S
(
Crystal structure data for the ligand Di-9AC (mono-
clinic) crystallized in DMSO (CCDC 1835667) (CIF)
Crystal structure data for the ligand Di-9AC (mono-
clinic) crytallized in DMA (CCDC 1835666) (CIF)
Crystal structure data for the ligand Di-9AC (triclinic)
crystalliized in DMA (CCDC 1835665) (CIF)
(
Pallier, A.; Collet, G.; Kieda, C.; Li, T.; Geib, S. J.; Rosi, N. L.; Petoud,
S. Proc. Natl. Acad. Sci. U. S. A. 2013, 110, 17199.
(11) McKinlay, A. C.; Morris, R. E.; Horcajada, P.; Ferey, G.; Gref,
R.; Couvreur, P.; Serre, C. Angew. Chem., Int. Ed. 2010, 49, 6260.
(
(
12) Lu, K.; He, C.; Lin, W. J. Am. Chem. Soc. 2014, 136, 16712.
13) Carne, A.; Carbonell, C.; Imaz, I.; Maspoch, D. Chem. Soc. Rev.
Crystal structure data for the CD-MOF-161 (CCDC
2
011, 40, 291.
(14) Horcajada, P.; Serre, C.; Maurin, G.; Ramsahye, N. A.; Balas,
F.; Vallet-Regí, M.; Sebban, M.; Taulelle, F.; Ferey, G. J. Am. Chem.
Soc. 2008, 130, 6774.
15) Cunha, D.; Ben Yahia, M.; Hall, S.; Miller, S. R.; Chevreau, H.;
Elkaim, E.; Maurin, G.; Horcajada, P.; Serre, C. Chem. Mater. 2013,
5, 2767.
1
835668) (CIF)
Additional figures about absorption spectroscopy
́
1
13
measurements, H and C NMR spectra, thermogravi-
metric analysis, crystallographic data and tables, FTIR
spectroscopy measurements, and particles counting
(
2
(
(PDF)
16) Coudert, F.-X. Chem. Mater. 2015, 27, 1905.
(17) Park, J.; Yuan, D.; Pham, K. T.; Li, J.-R.; Yakovenko, A.; Zhou,
AUTHOR INFORMATION
Corresponding Authors
guillaume.collet@cnrs-orleans.fr
stephane.petoud@inserm.fr
H.-C. J. Am. Chem. Soc.2012, 134, 99.
18) Lyndon, R.; Konstas, K.; Ladewig, B. P.; Southon, P. D.;
Kepert, C. J.; Hill, M. R. Angew. Chem., Int. Ed. 2013, 52, 3695.
19) Della Rocca, J.; Liu, D. M.; Lin, W. B. Acc. Chem. Res. 2011, 44,
57.
(20) Bachelder, E. M.; Beaudette, T. T.; Broaders, K. E.; Dashe, J.;
Frechet, J. M. J. J. Am. Chem. Soc. 2008, 130, 10494.
21) Broaders, K. E.; Grandhe, S.; Frechet, J. M. J. J. Am. Chem. Soc.
011, 133, 756.
■
(
*
(
9
*
ORCID
ine Besnard: 0000-0001-5699-9675
hane Petoud: 0000-0003-4659-4505
Cel
Step
́
́
(
2
́
́
Author Contributions
(
(
22) Colson, Y. L.; Grinstaff, M. W. Adv. Mater. 2012, 24, 3878.
23) Nguyen, M. M.; Carlini, A. S.; Chien, M. P.; Sonnenberg, S.;
The manuscript was written through contributions of all
authors. All authors have given approval to the final version of
the manuscript.
Luo, C. L.; Braden, R. L.; Osborn, K. G.; Li, Y. W.; Gianneschi, N. C.;
Christman, K. L. Adv. Mater. 2015, 27, 5547.
(
24) Callmann, C. E.; Barback, C. V.; Thompson, M. P.; Hall, D. J.;
Mattrey, R. F.; Gianneschi, N. C. Adv. Mater. 2015, 27, 4611.
25) Shum, P.; Kim, J. M.; Thompson, D. H. Adv. Drug Delivery Rev.
001, 53, 273.
26) Yavlovich, A.; Smith, B.; Gupta, K.; Blumenthal, R.; Puri, A.
Notes
The authors declare no competing financial interest.
(
2
(
ACKNOWLEDGMENTS
The authors gratefully acknowledge the Swiss National Science
Foundation for financial support (grant numbers
■
Mol. Membr. Biol. 2010, 27, 364.
(27) Lux, C. d. G.; Lux, J.; Collet, G.; He, S.; Chan, M. N.;
Olejniczak, J.; Foucault-Collet, A.; Almutairi, A. Biomacromolecules
2015, 16, 3286.
2
00020_159881 and 206021_164019) as well as la Ligue
Nationale Contre le Cancer, la Region Centre, l’Agence
Nationale de la Recherche (NIRA − ANR-13-BS08-0011).
S.P. acknowledges support from the Institut National de la
́
(28) Purcell, B. P.; Lobb, D.; Charati, M. B.; Dorsey, S. M.; Wade, R.
J.; Zellars, K. N.; Doviak, H.; Pettaway, S.; Logdon, C. B.; Shuman, J.
A.; Freels, P. D.; Gorman, J. H., III; Gorman, R. C.; Spinale, F. G.;
Burdick, J. A. Nat. Mater. 2014, 13, 653.
Sante
also grateful to Kerry-Lee Buchwalder and to Dr. Euro Solari
for elemental analyses and to Jer me Bosset for the
microscopy.
et de la Recherche Medicale (INSERM). Authors are
́ ́
(29) Kharkar, P. M.; Kiick, K. L.; Kloxin, A. M. Polym. Chem. 2015,
o
́ ̂
6
(
, 5565.
30) Mura, S.; Nicolas, J.; Couvreur, P. Nat. Mater. 2013, 12, 991.
(
31) Viger, M. L.; Collet, G.; Lux, J.; Nguyen Huu, V. A.; Guma, M.;
REFERENCES
■
Foucault-Collet, A.; Olejniczak, J.; Joshi-Barr, S.; Firestein, G. S.;
Almutairi, A. Biomaterials 2017, 133, 119.
(32) Viger, M. L.; Sankaranarayanan, J.; de Gracia Lux, C.; Chan,
M.; Almutairi, A. J. Am. Chem. Soc. 2013, 135, 7847.
(
1) Macrae, C. F.; Bruno, I. J.; Chisholm, J. A.; Edgington, P. R.;
McCabe, P.; Pidcock, E.; Rodriguez-Monge, L.; Taylor, R.; van de
Streek, J.; Wood, P. A. J. Appl. Crystallogr. 2008, 41, 466.
H
DOI: 10.1021/jacs.8b05047
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
(
33) Ganta, S.; Devalapally, H.; Shahiwala, A.; Amiji, M. J. Controlled
Release 2008, 126, 187.
34) Longmire, M.; Choyke, P. L.; Kobayashi, H. Nanomedicine
008, 3, 703.
35) Balasubramanian, S. K.; Jittiwat, J.; Manikandan, J.; Ong, C. N.;
Yu, L. E.; Ong, W. Y. Biomaterials 2010, 31, 2034.
36) Epley, C. C.; Roth, K. L.; Lin, S. Y.; Ahrenholtz, S. R.; Grove, T.
Z.; Morris, A. J. Dalton Trans. 2017, 46, 4917.
37) di Nunzio, M. R.; Agostoni, V.; Cohen, B.; Gref, R.; Douhal, A.
J. Med. Chem. 2014, 57, 411.
38) Nazari, M.; Rubio-Martinez, M.; Tobias, G.; Per
et al. Adv. Funct. Mater. 2016, 26, 3244.
39) Xu, X.-Y.; Chu, C.; Fu, H.; Du, X.-D.; Wang, P.; Zheng, W.;
Wang, C.-C. Chem. Eng. J. 2018, 350, 436.
40) Huang, N.; Ding, X. S.; Kim, J.; Ihee, H.; Jiang, D. L. Angew.
Chem., Int. Ed. 2015, 54, 8704.
(
2
(
(
(
(
́
ez Barrio, J.;
(
(
(
(
41) Ito, Y.; Fujita, H. J. Org. Chem. 1996, 61, 5677.
42) Nofen, E. M.; Wickham, J.; Koo, B.; Chattopadhyay, A.; Dai, L.
L. Mater. Res. Express 2016, 3.035701
43) More, R.; Busse, G.; Hallmann, J.; Paulmann, C.; Scholz, M.;
Techert, S. J. Phys. Chem. C 2010, 114, 4142.
44) Bouas-Laurent, H.; Castellan, A.; Desvergne, J. P.; Lapouyade,
R. Chem. Soc. Rev. 2000, 29, 43.
45) Bouas-Laurent, H.; Castellan, A.; Desvergne, J. P.; Lapouyade,
R. Chem. Soc. Rev. 2001, 30, 248.
46) Casanova, D.; Llunell, M.; Alemany, P.; Alvarez, S. Chem. - Eur.
J. 2005, 11, 1479.
47) Sienkiewicz-Gromiuk, J.; Rusinek, I.; Kurach, Ł.; Rzączyns
J. Therm. Anal. Calorim. 2016, 126, 327.
48) Friedrichs, O. D.; O’Keeffe, M.; Yaghi, O. M. Acta Crystallogr.,
Sect. A: Found. Crystallogr. 2003, 59, 22.
49) Spek, A. L. Acta Crystallogr., Sect. D: Biol. Crystallogr. 2009, 65,
48.
(
(
(
(
(
́
ka, Z.
(
(
1
(
50) Hu, Z. G.; Zhao, D. Dalton Trans. 2015, 44, 19018.
(51) Zahn, G.; Zerner, P.; Lippke, J.; Kempf, F. L.; Lilienthal, S.;
Schroder, C. A.; Schneider, A. M.; Behrens, P. CrystEngComm 2014,
6, 9198.
52) Sheldrick, G. M. Acta Crystallogr., Sect. C: Struct. Chem. 2015,
1, 3.
53) Dolomanov, O. V.; Bourhis, L. J.; Gildea, R. J.; Howard, J. A.
K.; Puschmann, H. J. Appl. Crystallogr. 2009, 42, 339.
54) Vandersluis, P.; Spek, A. L. Acta Crystallogr., Sect. A: Found.
Crystallogr. 1990, 46, 194.
1
(
7
(
(
I
DOI: 10.1021/jacs.8b05047
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX