Synthesis and photocleavage of a new polymerizable [2+2] hetero dimer for phototriggered drug delivery

Article abstract of DOI:10.1016/j.jphotochem.2011.02.021

7-Hydroxy-1,1-dimethylnaphtalenone is introduced as a versatile photocleavable linker system for drug attachment to polymer materials with significantly improved two-photon-absorption efficiency. The synthesis of TBDMS-protected 7-hydroxy-1,1-dimethylnaphtalenone monomer is described, which is the release group attached to the polymer moiety. The cytotoxic 5-fluoro-1-heptanoyluracil was photochemically attached via [2+2] cycloaddition to this release group. The linker-drug conjugate was photochemically polymerized into a HEMA/MMA copolymer. Photo-triggered drug release from the polymer via single-photon-absorption as well as two-photon-absorption in solution were carried out. The detachment of 5-FU from the polymer and its release from the polymer were monitored in the polymer as well as in the aqueous medium and multidose drug release was successfully demonstrated.

Full text of DOI:10.1016/j.jphotochem.2011.02.021

Journal of Photochemistry and Photobiology A: Chemistry 219 (2011) 228–234
Contents lists available at ScienceDirect
Journal of Photochemistry and Photobiology A:
Chemistry
journal homepage: www.elsevier.com/locate/jphotochem
Synthesis and photocleavage of a new polymerizable [2+2] hetero dimer for
phototriggered drug delivery
Julia Liesea, Norbert A. Hamppa,b,∗
a Department of Chemistry, University of Marburg, D-35032 Marburg, Germany
b Material Sciences Center, D-35032 Marburg, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
7-Hydroxy-1,1-dimethylnaphtalenone is introduced as a versatile photocleavable linker system for drug
attachment to polymer materials with significantly improved two-photon-absorption efficiency. The
synthesis of TBDMS-protected 7-hydroxy-1,1-dimethylnaphtalenone monomer is described, which is
the release group attached to the polymer moiety. The cytotoxic 5-fluoro-1-heptanoyluracil was pho-
tochemically attached via [2+2] cycloaddition to this release group. The linker–drug conjugate was
photochemically polymerized into a HEMA/MMA copolymer. Photo-triggered drug release from the poly-
mer via single-photon-absorption as well as two-photon-absorption in solution were carried out. The
detachment of 5-FU from the polymer and its release from the polymer were monitored in the polymer
as well as in the aqueous medium and multidose drug release was successfully demonstrated.
© 2011 Elsevier B.V. All rights reserved.
Received 3 December 2010
Received in revised form 16 February 2011
Accepted 21 February 2011
Available online 25 February 2011
Keywords:
Phototriggered drug delivery
[2+2] cycloaddition
Cycloreversion
Single photon absorption
Two photon absorption
1,1-Dimethylnaphtalenone
5-fluorouracil
1. Introduction
the age of patients, but are reported to lie in between 10% and 50%
within 3–5 years after IOL implantation [4,5].
Drug delivery systems are of interest in many medical appli-
cations. Sustained release is the most common method where the
drug is released in a predetermined time profile beginning as soon
as the system is applied to the patient. One example is the poly-
mer microcapsules carrying water soluble drugs releasing them
during decomposition of the capsules over several months [1]. But
there is an increasing demand for controllable drug delivery sys-
tems, delivering well-defined doses of drug on demand. Implanted
insulin pumps are a well-known example for this direction [2].
Our approach of controlled drug delivery is based on a photo
cleavable linker system for drug release from polymer in cases
where light access to the implant is possible. Such an example
is intraocular lenses, where drug release needs to be accom-
plished from implanted IOLs years after cataract surgery for
non-invasive treatment of posterior capsule opacification (PCO),
so-called secondary cataract. After cataract surgery epithelial cells
may proliferate on the polymer surface from the rim to the centre
of the IOL and cause progressive deterioration of the visual acuity
[3]. The incidence of PCO varies depending on the IOL materials and
The phototriggered release is achieved by selective photo-
cleavage of a cyclobutane moiety in the structure. Cleavage of
the cyclobutane ring may be accomplished by single-photon-
absorption (SPA) as well as by two-photon-absorption (TPA)
triggered reactions. The TPA process may be illustrated as the
simultaneous absorption of two photons each of half energy com-
pared to one photon required for a SPA process. The TPA tool
enables to control a photochemical reaction requiring light of
∼260 nm, inside the eye. Such wavelengths are filtered off by the
cornea. In addition, a precise 3D-spatial control of the photochemi-
cal reaction is possible. Early attempts to realize such a system were
reported by Kim et al. [6,7]. Dicoumarins formed the photocleav-
able linker between a covalently attached drug and the polymer
backbone, which causes that the drug is released with an unde-
sired helper molecule attached [6]. Another approach was to bind
the drug directly to the coumarin via [2+2]cycloaddition reaction
[7,8]. However, coumarins and in particular dicoumarins have a
very limited solubility in most solvents which makes synthesis
rather difficult and additionally the lactone ring in coumarin-dimer
is susceptible to hydrolysis [9,10] which leads to an undesired
complexity in the photochemical cleavage reaction. The approach
loading the lens polymer by mixing with derivatized oligomers [8]
is restricted to low amounts of drug loading in the polymer. To over-
come both types of problems a coumarin-analogue linker molecule,
7-hydroxy-1,1-dimethylnaphtalenone, which shows better sta-
∗
Corresponding author at: University of Marburg, Department of Chemistry,
Hans-Meerwein-Str. Bldg. H, D-35032 Marburg, Germany. Tel.: +49 6421 2825775;
fax: +49 6421 2825798.
E-mail address: hampp@staff.uni-marburg (N.A. Hampp).
1010-6030/$ – see front matter © 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.jphotochem.2011.02.021

J. Liese, N.A. Hampp / Journal of Photochemistry and Photobiology A: Chemistry 219 (2011) 228–234
229
Fig. 1. Synthesis of 7-hydroxy-1,1-dimethylnaphtalenone.
Fig. 2. Cross-dimerisation of 7-tert-butyldimethylsilyl-1,1-dimethylnaphtalenone (5) and 5-fluoro-1-heptanoyluracil (6) via [2+2] cycloaddition and attachment of the
methyl-methacrylate monomer.
bility, better solubility and an improved two-photon-absorption
coefficient [11] was synthesized and characterized for its applica-
tion in such a drug delivery system.
250 mm × 4 mm, Bischoff Chromatography) using acetonitrile or
methanol with ultra pure water (0.05 ␮S/cm), acidified with 300 ␮L
H3PO4 per litre, as an eluent. For preparative RP-HPLC a YMC-
Pack ODS-A column (C18, 5.0 ␮m, 250 mm × 4 mm, YMC) was used
with an AD 25 absorbance detector and a P680 HPLC pump (both
Dionex). LC/MS analyses were done on an ESI/MS LCQ Duo mass
spectrometer (Thermoelectron) connected to a HPLC (1050-HPLC,
Agilent Technologies) with a Nucleosil 100-3-C18 column (3 ␮m,
250 mm × 4 mm, Bischoff Chromatography). Heterodimers were
prepared in an UV-reactor equipped with 12 fluorescent tubes
Eversun 40 W/79, 25X (Osram). Photo-cleavage of heterodimers in
acetonitrile was carried out at 266 nm in a fluorescence spectropho-
tometer type RF 1502 (Shimadzu). Photon densities at 266 nm
were determined using a photodiode 1227-1010 BQ (Hamamatsu)
with a correction factor of 1.1017 determined by actinometry with
azobenzene for the fluorescence spectrophotometer at 266 nm.
Two-photon-absorption experiments were done at 532 nm using a
frequency-doubled Nd:YAG pulse-laser (Infinity 40-100, Coherent).
All UV/vis spectra were recorded on a Lambda 35 (Perkin Elmer) in
quartz cuvettes (Hellma).
Here we report the synthesis and characterisation of drug
loaded polymer containing
a new dimer composed of 1,1-
dimethylnaphtalenone and 5-fluoro-1-heptanoyluracil. 5-fluoro-
1-heptanoyluracil (H5-FU) proved to be a very suitable cytotoxic
for treating posterior capsule opacification [6,12] and was bound
directly to the 1,1-dimethylnaphtalenone monomer via [2+2]
cycloaddition. This approach has the advantage that the drug is
directly released and activated from the linker without any fur-
ther remaining linker groups attached. We characterized the drug
delivery from polymer material with respect to diffusion times and
the ability of multidose applications. Upon detachment of the H5-
FU from the polymer backbone and release from the polymer to
the aqueous medium the heptanoyl residue is split off by amide
hydrolysis [7] and the 5-FU remains.
2. Experimental
All chemicals were purchased from commercial suppliers and
used without further purification. Organic solvents were received
in technical quality and were purified by filtration and distillation
before use.
2.2. Synthesis
2.2.1. Synthesis of 7-methoxy-1,1-dimethylnaphtalenone (3)
The synthesis of 7-methoxy-1,1-dimethyltetralone (2) was
done, with some modifications, as described earlier[11]. 5.0 g
(0.028 mol) 7-methoxy-2-tetralone (1) were dissolved in 40 mL
of THF under argon atmosphere. The solution was cooled to
0 ◦C and 5.5 g (0.049 mol) potassium-tert-butoxide in 50 mL tert-
butanol were added slowly. After stirring for 20 min methyliodide
was added (3.6 mL, 0.028 mol) and the solution was stirred
at room temperature for 2 h. The light sensitive product was
purified via column chromatography on silica with 15:1 pen-
tane: ethyl acetate yielding 4.4 g (79%) of white solid. For the
introduction of the double bond (see Fig. 1) to 7-methoxy-1,1-
dimethylnaphtalenone (3) iodoxybenzoic acid (IBX) was added as
described earlier[11].
2.1. Analytical
Standard NMR spectra were recorded on a Bruker Avance 300 B
spectrometer (300 MHz) or on a Bruker DRX-400 NMR (400 MHz).
Analytical RP-HPLC measurements were done on an UltiMate 3000
System (Dionex) equipped with a Nucleosil column (RP18, 5.0 ␮m,
2.2.2. Replacement of the methoxy-protecting group by a
tert-butyldimethylsilyl-group
The methoxy group was removed with boron tribromide (BBr3).
1 g (4.96 mmol) of 7-methoxy-1,1-dimethylnaphtalenone (3) was
dissolved in 160 mL dichlormethane under argon atmosphere and
Fig. 3. Round polymer blanks of 5-FU-loaded HEMA/MMA polymer.

230
J. Liese, N.A. Hampp / Journal of Photochemistry and Photobiology A: Chemistry 219 (2011) 228–234
Fig. 4. Quantum efficiency of photochemical single-photon-absorption [2+2] cycloreversion of dimer. (a) UV/vis difference spectra monitoring the course of SPA cleavage.
The rising absorption at 331 nm indicates the energy-dependent formation of the monomer (3). (b) Time-dependence of (3) increase versus irradiation time. The number of
cleaved molecules versus the number of absorbed photons was derived to be ˚SPA = 0.010.
Fig. 5. Two-photon-absorption experiments of dimer (7) in solution. (a) UV/vis difference spectra monitoring the course of TPA cleavage. The rising absorption at 331 nm
indicates the energy-dependent formation of the monomer (3). The measurement for 55.8 mJ pulse energy is shown. (b) TPA induced cleavage at different energies. Monomer
formation measured as the increase of the absorption at 331 nm in dependence on the incident laser power. (c) Double logarithmic plot of the applied laser intensities P
versus the initial rates v0 of the linker cleavage. The TPA induced nature of the photocleavage corresponds very well with the slope of 2 when ln v0 is plotted versus ln P. (d)
Photocleavage reaction from dimer (7) to the monomers and their respective absorption maxima.

J. Liese, N.A. Hampp / Journal of Photochemistry and Photobiology A: Chemistry 219 (2011) 228–234
231
Fig. 6. Drug release experiments of the dimer in polymer. (a) UV/vis spectra of 5-FU at different irradiation times of the polymer showing the dependence of drug release
on irradiation time. (b) Amount of 5-FU in [␮g] of released drug calculated from the irradiation experiment shown in (a) and (c). Surrounding water phase of the lens before
irradiation and afterwards. No release of 5-FU was detectable without exposure to the UV-C lamp. (d) UV/vis spectra of the lens itself after irradiation. The rising of the
absorption at 331 nm of 1,1-dimethylnaphtalenon (5) is clearly shown and rises with releasing of 5-FU but remains still covalently attached to the polymer backbone. The
polymer is completely non transparent for light below 350 nm, therefore only the raise of the absorption peak at 331 nm is detectable.
cooled to −78 ◦C. 5 equivalents (eq.) of boron tribromide (25 mL,
0.025 mol) were added and the solution was stirred for another
two days. The reaction was quenched with 500 mL of water. Organic
and water phase were separated and the water phase was extracted
with dichlormethane. The solvent was removed and the obtained
7-hydroxy-1,1-dimethylnaphtalenone (4) was dried in vacuo. The
yield was quantitative.
13C NMR (75 MHz, CDCl3):
ı/ppm = 158.4, 150.4, 145.7, 131.4, 121.7, 121.4, 113.9, 47.6, 28.1
M+ calc 188.1, found 189.6
Tert-butyldimethylsilyl-chloride
was
chosen
as
a
protecting agent for the hydroxy group. 7-hydroxy-1,1-
dimethylnaphtalenone (1.0 g, 5.32 mmol) and imidazole (5 eq.,
1.81 g, 0.027 mol) were dissolved in 120 mL tetrahydrofuran under
argon atmosphere and 1.2 eq. of TBDMS-Cl (0.010 mol, 1.6 g) were
added at room temperature and stirred for 8 h. The solvent was
removed and the brown residue was dissolved in chloroform and
washed with saturated NaHCO3 solution and water. The chloroform
1H NMR (300 MHz, CDCl3):
ı/ppm = 7.34 (d, 1H, J = 9.8 Hz), 7.17 (t, 1H, J = 6.8 Hz, J = 8.3 Hz),
6.93 (d, 1H, J = 2.3 Hz), 6.73 (dd, 1H, J = 2.4 Hz, J = 8.3 Hz), 6.0 (d, 1H,
J = 9.8 Hz), 1.40 (s, 6 H)
Fig. 7. Diffusion of 5FU out of the drug loaded polymer after irradiation. (a) UV/vis spectra recorded immediately after each irradiation step. (b) Calculated amount of released
drug plotted versus irradiation times.

232
J. Liese, N.A. Hampp / Journal of Photochemistry and Photobiology A: Chemistry 219 (2011) 228–234
Fig. 8. Repeated drug release from polymer blank. (a) After irradiation the polymer was incubated in water for one week to ensure that released 5-FU diffuses completely
out of the material. The released amount of drug was measured. (b) Total of cumulative drug release from the ten identical consecutive irradiations.
13C NMR (75 MHz, CDCl3):
was removed and a brown oil was obtained. The oil was washed
with methanol. An insoluble white solid was obtained and con-
firmed to be 7-tert-butyldimethylsilyl-1,1-dimethylnaphtalenone
(1.3 g, 90%)
ı/ppm = 207.5, 173.7, 160.2, 148.8, 144.4, 132.3, 132.2, 117.3,
112.5, 112.2, 55.5, 55.18, 46.6, 41.4, 41.1, 38.9, 31.6, 29.5, 28.8, 24.5,
24.5, 22.6, 14.1, 1.1
1H NMR (300 MHz, CDCl3):
19F NMR (300 MHz, CDCl3): −155.6
M+ calc. 544.7, found 543.7
ı/ppm = 7.38 (d, 1H, J = 9.8 Hz), 7.29 (d, 1H, J = 8.3 Hz), 6.91 (d, 1H,
J = 2.3 Hz), 6.74 (dd, 1H, J = 8.2 Hz, J = 2.4 Hz), 6.04 (d, 1H, J = 9.8 Hz),
1.44 (s, 6H), 0.99 (s, 9H), 0.23 (s, 6H)
2.2.4. Substitution of the TBDMS-protecting group for
methyl-methacrylate
13C NMR (75 MHz, CDCl3)
ı/ppm = 203.5, 156.8, 148.9, 143.6, 129.9, 121.6, 121.1, 117.4,
117.2, 46.4, 27.0, 25.0, 24.7, 24.6, 17.3, 0.000, −4.0, −5.4, −6.2
M+ calc. 302.5, found 304.2
The 7-tert-butyldimethylsilyl-protecting group of the het-
erodimer (7) (40 mg) was removed with triethylamine (Et3N, 8 eq.,
83 ␮L) in dimethylformamide (1 mL) at 55 ◦C. The product was
purified via isocratic preparative RP-HPLC with acetonitrile and
acidified water (75:25) as eluent and yielded 29 mg (91%). It was
identified via LC/MS (M+ calc. 430.5, found 430.6).
2.2.3. Heterodimerisation of
7-tert-butyldimethylsilyl-1,1-dimethylnaphtalenone (5) and
5-fluoro-1-heptanoyluracil (6)
As 5-fluorouracile (5-FU) is more or less insoluble in relevant
solvents for the dimerisation, a heptanoyl group was attached to
one nitrogen to improve solubility. The heptanoyl group is quickly
removed in aqueous solutions and therefore does not affect the
cytotoxic activity [6]. H5-FU (6) was synthesized as described ear-
lier [6] (Fig. 2).
For polymerisation methyl-methacrylate was covalently
bound to the 7-hydroxy position. 160 mg dimer (0.31 mmol),
79.4 mg
ethyl-(Nꢀ,Nꢀ-dimethylamino)propylcarbodiimide
hydrochloride (EDC, 1.1 eq., 0.34 mmol) and 9.2 mg 4-(N,N-
dimethylamino)pyridine (DMAP, 0.2 eq., 0.06 mmol) were
dissolved in dichloromethane (25 mL) at 0 ◦C under argon
atmosphere. 1.1 eq. (0.34 mmol, 28.3 ␮L) of methyl-methacrylate
(MMA) were added and the mixture was stirred over night. The
solution was extracted with 5% NaHCO3 and brine. The organic
phase was dried in vacuo and yielded 198 mg of a slightly yellow
solid, the MMA-heterodimer (8) (95%). The product was analysed
via NMR and LC/MS experiments. MS/MS experiments were carried
out to verify the successful attachment of methyl-methacrylate at
the naphtalenone-site in the heterodimer.
The
dimerisation
of
7-tert-butyldimethylsilyl-1,1-
and 5-fluoro-1-heptanoyluracil
dimethylnaphtalenone
(5)
(6) was carried out in
a
Rayonett type photoreactor with
wavelengths above 300 nm. The 7-tert-butyldimethylsily-1,1-
dimethylnaphtalenone was dissolved in 10 mL of a saturated
solution of 5-fluoro-1-heptanoyluracil with 10% of benzophenone
in chloroform and degassed with argon for 15 min. Eight glass
tubes were filled and sealed and the solutions were irradiated with
UV light for 3–5 days. The reaction progress was monitored via
HPLC. When the reaction was finished, the solvent was removed.
The crude product was dissolved in acetonitrile/water (50:50) and
stirred for 24 h at 40 ◦C in order to remove the heptanoyl group
of excess 5-fluoro-1-heptanoyluracil. It should be noted that the
heptanoyl group in the heterodimer is not affected under these
conditions. The acetonitrile was removed by distillation, the 5-FU
was filtered off and the watery phase was extracted with chlo-
roform. The organic phase was dried, re-dissolved in acetonitrile
and purified via isocratic preparative RP-HPLC with acetonitrile
and acidified water (45:55) as an eluent. The heterodimer was
obtained as a yellow solid (602 mg, 1.35 mmol, 23%).
1H NMR (300 MHz, CDCl3):
ı/ppm = 7.45 (m, 1H), 7.10 (m, 2H), 6.37 (s, 1H), 5.79 (s, 1H), 5.37
(q, 1H, J = 8.7 Hz, J = 6.8 Hz), 4.29 (d, 1H, J = 8.7 Hz), 3.06 (m, 3H), 2.08
(s, 3H), 1.67 (m, 6H), 1.33 (m, 10 H), 0.87 (m, 3H)
LC/MS: M+ calc. 498.5, found 497.5
MS/MS on 497.5: M+ of fragments = (4) + MMA: calc. 256.3 found
257.6; 5-FU calc. 129.1, found 127.9.
2.2.5. Photocontrolled co-polymerisation of drug–linker
conjugate and HEMA/MMA
The polymer was prepared via photo-induced polymeri-
sation at 470 nm. 43.3 g hydroxyethyl-methacylate (HEMA),
6.0 g methyl-methacrylate (MMA), 0.5 g ethylene-glycol-
dimethacrylate (EGDMA), 0.1 g campherquinone and 0.1 g
ethyl-4-dimethylaminobenzoate were mixed and degassed
in vacuo. 90 mg of MMA-heterodimer were dissolved in 6 mL of
the polymer mixture, degassed and filled into the polymerisation
1H NMR (400 MHz, CDCl3):
ı/ppm = 8.35 (s, 1H), 7.27 (m, 1H), 6.84 (m, 2H), 5.40 (q, 1H,
J = 8.7 Hz, J = 6.5 Hz), 4.23 (d, 1H, J = 8.6 Hz), 2.98 (m, 3H), 1.64 (s,
6H, 2 × CH3), 1.30 (s, 8H), 1.0 (s, 9H), 0.87 (m, 3H, CH3), 0.23 (s, 6H)

J. Liese, N.A. Hampp / Journal of Photochemistry and Photobiology A: Chemistry 219 (2011) 228–234
233
cuvette (65 mm × 50 mm × 2 mm). The photo-polymerisation
yielded a homogenous and optically clear plate, containing 1.5%
weight percent of drug loaded heterodimer.
was plotted against the irradiation time for the different pulse ener-
gies to obtain the initial reaction rates v0. In most cases reported in
the literature the quantum yield of the TPA has been assumed to
be the same as for SPA [13–15]. Assuming this the TPA absorption
cross section was calculated to be 1.81 GM which is almost twice as
high as for dicoumarin [6]. In a double logarithmic plot of the initial
reaction rate versus the intensities P a linear slope of 2.05 results
[16,17] which confirms the two-photon nature of the process.
2.2.6. General polymer assays
The polymer was cut into round blanks of 6 mm diameter and
2 mm thickness, which were then analysed regarding 5-FU drug
release.
For the tests the blanks were manufactured into lenses of 6 mm
diameter and 1 mm thickness, weighting 20 mg each, which means
an absolute amount of 75.9 ␮g 5-FU was contained in each lens.
Every lens was incubated in water for six days for hydration at room
temperature. In parallel testing was done at 37 ◦C. UV/vis mea-
surements of the water phase in which the polymer lenses were
incubated showed no release of drug.
3.2. Characterisation of the drug loaded polymer material
A lens was irradiated with an UV-C lamp (Philipps TUV, PL-S,
9 W). The lens as well as the water phase surrounding the lens was
analysed via UV/vis.
Without any irradiation, no drug was detectable even after sev-
eral days. Only after irradiation with high intensity light a release
was detected (Fig. 6a) and the amount of 5-FU released was found
to be dependent on irradiation times (Fig. 6b). In the lens mate-
rial a rise of the absorption band of remaining naphtalenone group,
which is still covalently bound to the polymer moiety, was observed
(Fig. 6c) whereas in the water phase the characteristic absorption
at 265 nm of 5-FU was detected (Fig. 6d). After the experiments
the water phase was concentrated and additionally analysed via
HPLC, where solely 5-FU was identified as product from the photo
triggered [2+2] cycloreversion.
3. Results and discussion
The photochemical properties of the heterodimer (7) as well
as of polymer containing it in immobilized form were analysed
regarding SPA and TPA induced cycloreversion in acetonitrile solu-
tion and its release from the polymer.
3.1. Single photon and two photon absorption triggered drug
release from heterodimer (7) and polymer
3.3. Diffusion of drug out of the polymer material
For single photon absorption measurements
a sample
of heterodimer in acetonitrile was irradiated at 266 nm
(c = 0.649 mmol/L). Cycloreversion of heterodimer (7) causes
an absorption band with a maximum at ꢀ = 331 nm from 7-
hydroxy-1,1-dimethylnaphtalenon (5), which results from the
conjugated ␲-system between the carbonyl group and the double
bond formed during cycloreversion of the cyclobutane ring. The
heterodimer (7) and the released H5-FU/5-FU do not have any
absorption in this region. This enables the use of the 331 nm band
for monitoring and quantifying of the reaction.
The diffusion of released 5-FU from the polymer was measured
by irradiating a lens for a few seconds and monitoring the absorp-
tion band of 5-FU at 265 nm via UV/vis directly afterwards (Fig. 7).
The release rate form the polymeric lens was determined to
be 0.016 ␮g/s and the total drug release after 24 h incubation in
water was 9.0 ␮g. For cataract treatment we calculated a minimum
amount of 1 ␮g for a sufficient cytotoxic activity. After 20 min one
third, i.e. 3.1 ␮g were already released from the polymer mate-
rial which is a promising result for secondary cataract treatment
giving a quick and high initial drug dose. The water phase was con-
centrated and analysed via HPLC showing only 5-FU as a product
again.
Absorption spectra comprising the band at 331 nm of 1,1-
dimethylnaphtalenone as a function of irradiation time are shown
in Fig. 4a. The light energy of the lamp was measured to be
9.9331 × 1014 photons s−1
.
The efficiency of the SPA-induced photocleavage is character-
ized by the quantum yield ФSPA which is defined as the ratio of the
number of molecules nMol cleaved by the number of photons nPhot
absorbed [11]. Because the absorption at 266 nm was beyond 2.5 it
was approximated that all photons were absorbed
3.4. Multiple drug releases from drug-loaded polymer
Another requirement of the system was the possibility of a
multidose drug release. A piece of polymer of 2 mm thickness,
weighting 21.6 mg, was irradiated ten times for 10 or 20 min. After
each irradiation step the polymer piece was incubated in water for
one week and the released amount of 5-FU was measured via UV/vis
before the next irradiation step (Fig. 8).
With each irradiation step smaller amounts of 5-FU were
released, but in each irradiation more than 1.0 ␮g were released
and proving the multidose capability of these polymers.
nMol
˚
SPA
=
nPhot
The number of cleaved molecules was calculated from the ini-
tial reaction rate v0 (t → 0) according to Lambert–Beer’s law.
The cleavage of 1,1-dimethylnaphtalenone causes an absorption
band with a maximum at ꢀ = 331 nm. The extinction coefficient
is ε331 = 10,250 L × mol−1 cm−1[11]. Neither the H5FU nor the het-
erodimer show any absorption in this region. The initial rea−ct1ion
rate was determined to be v0 = 1.01547 × 1013 molecules s as
shown in Fig. 3b. The quantum yield ˚SPA was calculated to be
4. Summary and conclusions
The synthesis and dimerisation of monomers and drug loading
of HEMA/MMA photopolymer with 5-FU was successfully carried
out. The 7-hydroxy-1,1-dimethylnaphtalenone revealed to be a
suitable linker for reversible covalent attachment of the cytotoxin
5-FU to the polymer backbone. The 5-FU is reversible attached to
the linker via [2+2] cyclo-addition and -reversion. Single-photon-
absorption as well as two-photon-absorption cycloreversions of
the dimer were successfully carried out, the released doses are
more than adequate for cataract treatment and repeating the
drug release with sufficient drug doses is possible. The SPA of
the novel linker system was found to be ˚SPA = 0.010. The TPA
˚SPA = 0.010 = 1% (Fig. 4).
Irradiation of the dimer (7) in acetonitrile (c = 1.964 mmol/L)
with intense 532 nm pulses (20 Hz, 3 ns) causes TPA-induced cleav-
age of the heterodimer (Fig. 5). The formation of (5) is monitored
at ꢀ = 331 nm. An increase in the absorption indicates the cleavage
of the cyclobutane-linker.
The change in absorption should be linearly proportional to the
applied irradiation energy. In order to verify this, the TPA cross
section was measured at four different energies (24.7 mJ, 34.7 mJ,
45.7 mJ and 55.8 mJ) (Fig. 5b). The change in concentrations of (5)

234
J. Liese, N.A. Hampp / Journal of Photochemistry and Photobiology A: Chemistry 219 (2011) 228–234
cross section in solution was determined to be 1.81 GM, which
is lower than the TPA for the naphtalenone homodimer but
higher than for the coumarin dimers. This continuative process-
ing of photocontrolled building blocks into polymer matrices takes
another step towards its final application for TPA triggered drug
delivery.
[7] H.-C. Kim, S. Härtner, M. Behe, T. Behr, N. Hampp, Two-photon absorption-
controlled multidose drug release: a novel approach for secondary cataract
treatment, J. Biomed. Optics 11 (2006) 1–9.
[8] C. Sinkel, A. Greiner, S. Agarwal, Synthesis, characterization and properties eval-
uation of methylcoumarin end-functionalized poly(methylmethacrylate) for
photoinduced drug release, Macromolecules 41 (2008) 3460–3467.
[9] N. Yonezawa, M. Hasegawa, Synthesis of lactone-opened derivatives of anti and
syn head-to-head coumarin dimers, Bull. Chem. Soc. Jpn. 56 (1983) 367–368.
[10] N. Yonezawa, T. Yoshida, M. Hasegawa, Symmetric and asymmetric photo-
cleavage of the cyclobutane rings in head-to-head coumarin dimers and their
lactone-opened derivatives, J. Chem. Soc. Perkin Trans. I (1983) 1083–1086.
[11] J. Liese, N. Hampp, 1,1-dimethylnaphtalenon-dimers as photocleavable link-
ers with improved two-photon-absorption efficiency and hydrolitic stability, J.
Photochem. Photobiol. A 209 (2010) 128–134.
[12] J.M. Ruiz, M. Medrane, J.L. Alio, Inhibition of posterior capsule opacification by
5-fluoruracile in rabbits, Opthalmic Res. 22 (1990) 201–209.
[13] R.P. Wayne, Principles and Applications of Photochemistry, Oxford University
Press, Oxford, 1988.
[14] S. Härtner, H.-C. Kim, N. Hampp, Photodimerized 7-hydroxycoumarin with
improved solubility in PMMA: single-photon and two-photon-induced pho-
tocleavage in solution and PMMA films, J. Photochem. Photobiol. A 187 (2007)
242–246.
[15] L. Parma, N. Omenetto, Two-photon absorption of 7-hydroxycoumarine, Chem.
Phys. Lett. 54 (1978) 541–543.
References
[1] L. Lu, K. Huang, Preparation of poly(propylene-co-y-butyrolactone carbonate)
and release profiles of drug loaded microcapsules, J. Poly. Sci. Part A: Polym.
Chem. 43 (2005) 2468–2475.
[2] C. Bogdan, Current status and future of implantable insulin pumps
for the treatment of diabetes, Expert Rev. Med. Devices
181–185.
1 (2004)
[3] N. Ibaraki, A brighter future for cataract surgery, Nat. Med. 3 (1997) 958–960.
[4] D.A. Schaumberg, M.R. Dana, W.G. Christen, R.J. Glynn, A systematic overview
of the incidence of posterior capsule opacification, Opthalmology 105 (1998)
1213–1221.
[5] D.J. Apple, K.D. Solomon, M.R. Tetz, E.I. assia, E.Y. Holland, U.F. Legler, J.C. Tsai,
V.E. Casteneda, J.P. Hoggatt, A.M. Kosick, Posterior capsule opacification, Surv.
Opthalmol. 37 (1992) 73–116.
[6] S. Härtner, H.-C. Kim, N. Hampp, Phototriggered release of photolabile drugs
via two-photon absorption-induced cleavage of polymer-bound dicoumarin, J.
Polym. Sci. A 12 (2007) 2443–2452.
[16] M. Goeppert-Mayer, Elementary processes with two-quantum transitions, Ann.
Phys. 9 (1931) 273–294.
[17] J.D. Bhawalkar, G.S. He, P.N. Prasad, Nonlinear multiphoton processes in organic
and polymeric materials, Rep. Prog. Phys. 59 (1996) 1041–1070.