Cationic poly(2-oxazoline) hydrogels for reversible DNA binding

Article abstract of DOI:10.1039/c3sm00114h

A new 2-oxazoline monomer with a Boc-protected amino group in the side chain (BocOx) was synthesized. Homopolymerization as well as copolymerization with 2-ethyl-2-oxazoline (EtOx) revealed a pseudo first order kinetic. A series of homopolymers was synthesized, deprotected and characterized regarding their structure and thermal properties. The copolymerization with EtOx yielded a series of water soluble polymers with varying amino contents. After deprotection it was shown by the ethidium bromide assay that these polymers were able to form complexes with DNA. Treatment with epichlorohydrin leads to the formation of hydrogels. The swelling properties of the gels were investigated and it could be demonstrated that also the polymeric scaffolds were able to immobilize DNA from aqueous solution. Furthermore, the release of the DNA was accomplished using heparin.

Full text of DOI:10.1039/c3sm00114h

Soft Matter
View Article Online
View Journal | View Issue
PAPER
Cationic poly(2-oxazoline) hydrogels for reversible DNA
binding
Cite this: Soft Matter, 2013, 9, 4693
Matthias Hartlieb,ab David Pretzel,ab Kristian Kempe,†ab Carolin Fritzsche,ab
ab
Renzo M. Paulus,ab Michael Gottschaldtab and Ulrich S. Schubert
*
A new 2-oxazoline monomer with a Boc-protected amino group in the side chain (BocOx) was synthesized.
Homopolymerization as well as copolymerization with 2-ethyl-2-oxazoline (EtOx) revealed a pseudo first
order kinetic. A series of homopolymers was synthesized, deprotected and characterized regarding their
structure and thermal properties. The copolymerization with EtOx yielded a series of water soluble
polymers with varying amino contents. After deprotection it was shown by the ethidium bromide assay
that these polymers were able to form complexes with DNA. Treatment with epichlorohydrin leads to
the formation of hydrogels. The swelling properties of the gels were investigated and it could be
demonstrated that also the polymeric scaffolds were able to immobilize DNA from aqueous solution.
Furthermore, the release of the DNA was accomplished using heparin.
Received 11th January 2013
Accepted 8th March 2013
DOI: 10.1039/c3sm00114h
www.rsc.org/softmatter
immobilization of DNA and other biomolecules within
hydrogel-like structures.15–17
Introduction
DNA chips and microarrays are effective tools for life science
research, for instance in modern clinical diagnostics,1 genetic
analysis,2 drug discovery,3 and pathogen detection.4 These effi-
cient analytical devices are mostly characterized by a low
detection level combined with a high specicity and reproduc-
ibility as well as low costs allowing high-throughput experi-
ments.1 The basic principle of DNA chips relies on the
hybridization of target DNA molecules from the sample with
specic single stranded oligonucleotides attached to a 2-
dimensional solid support like glass,5–7 silicon wafers, gold
surfaces,8 or polymers such as poly(methyl methacrylate).9,10
One major drawback of two-dimensional DNA chips is the
limited loading capacity of surface materials and the restricted
hybridization efficiency inuenced by the probe density
resulting in a diminished/low test signal.11,12 To overcome
these issues, a challenging alternative approach is the immo-
bilization of single stranded DNA molecules within a 3-
dimensional hydrogel matrix. Thereby, a multifold increased
loading capacity compared to 2D supports can be achieved.13,14
The high water content of the hydrogel and the resulting
rapid diffusion of target DNA within the 3D matrix predestine
such material attractive for DNA immobilization and hybrid-
ization. Hence, various reports describe methods for the
Beside the above described detection unit of common DNA
chips and microarrays, another interesting application for DNA
binding hydrogels aims at a step prior to chip analysis: the
isolation of DNA from the sample. It is of particular interest for
fully integrated “lab on a chip” solutions that in a single device a
DNA binding/purication/enrichment phase, which releases
isolated DNA on a certain stimulus, can be combined with the
detection area, where specic oligonucleotides are immobilized
on a 3-dimensional substrate to catch and hybridize the
analyte DNA.
The extraction of DNA from living or conserved tissues, cells,
virus particles, or other samples is the starting point for
downstream processes including diagnostic kits.18 The general
steps of nucleic acid purication include cell lysis, inactivation
of cellular nucleases and separation of desired nucleic acid
from cell debris.19 Currently, there are many specialized
methods for extracting DNA, which are generally divided into
solution-based (e.g. guanidinium thiocyanate–phenol–chloro-
form extraction; alkaline extraction, ethidium bromide–
caesium chloride gradient centrifugation) or solid phase-based
methods (mostly spin columns with matrices containing silica,
nitrocellulose and polyamide or anion exchange resins e.g. with
diethylaminoethyl cellulose groups).20–24 Whereas solid-phase
nucleic acid purication systems allow quick and efficient
purication compared to conventional liquid phase methods25
and display the current state of the art, similar limitations as
described for 2-dimensional DNA chip technologies occur such
as limited accessible reactive surfaces. The use of functional
DNA binding hydrogels for the isolation, enrichment, purica-
tion and stimuli responsive release of these biomolecules
aLaboratory of Organic and Macromolecular Chemistry (IOMC), Friedrich Schiller
University Jena, Humboldtstrasse 10, 07743, Jena, Germany. E-mail: ulrich.
schubert@uni-jena.de
bJena Center for So Matter (JCSM), Friedrich Schiller University Jena, Philosophenweg
7, 07743, Jena, Germany
† Current address: Department of Chemical and Biomolecular Engineering, The
University of Melbourne, Victoria 3010, Australia.
This journal is ª The Royal Society of Chemistry 2013
Soft Matter, 2013, 9, 4693–4704 | 4693

View Article Online
Soft Matter
Paper
displays an ideal possibility to combine advantages of common Furthermore, hydrogels are produced on the basis of these
liquid and solid phase extraction systems, such as a high polymers in order to act as solid substrates for the reversible
diffusion rate, tuneable functionalities and a high reaction binding of DNA, which is investigated using the uorescence of
surface for the binding of target molecules.
ethidium bromide (EB).
For the synthesis of DNA binding hydrogels, inspiration can
be found in the exploited eld of non-viral gene delivery, where
cationic polymers, such as poly(ethylene imine)s (PEI),26,27
poly(L-lysin)s28,29 and poly(methacrylate)s containing amino
Experimental part
Materials and instrumentation
functionalities30–32 have been extensively studied in the last few All chemicals and solvents were purchased from Sigma-Aldrich,
decades. Branched PEI (BPEI) has become the gold standard, Merck, Fluka, and Acros. 2-Ethyl-2-oxazoline (EtOx) and methyl
due to its efficient polyplex formation.33,34 The charge density tosylate (MeTos) were distilled to dryness prior to use.
and the length of the polymer backbone represent crucial
The Initiator Sixty single-mode microwave synthesizer from
factors for efficient DNA binding.35 The synthesis path, which Biotage, equipped with a noninvasive IR sensor (accuracy: 2%),
leads to linear PEI (LPEI), the acidic/basic hydrolysis of poly(2- was used for polymerizations under microwave irradiation.
oxazoline)s (POx)s in particular of poly(2-methyl-2-oxazoline) Microwave vials were heated overnight to 110 ꢀC and allowed to
and poly(2-ethyl-2-oxazoline) (PEtOx),36 limits the structural cool to room temperature under an argon atmosphere before
versatility of these gene delivery systems due to the incompati- usage. All polymerizations were carried out under temperature
bility of most of the functional groups under these conditions. control.
Additionally, PEI shows an unsatisfying water solubility for
biological applications.37
Size-exclusion chromatography (SEC) of protected polymers
was performed on a Shimadzu system equipped with a SCL-10A
Various reports have demonstrated the conjugation of PEI to system controller, a LC-10AD pump, a RID-10A refractive index
other biochemically relevant polymers such as polyethylene detector and a PSS SDV column with chloroform–triethylamine
glycol (PEG),9–11 POxs,38 and poly(caprolactone)39,40 in post- (TEA)–2-propanol (94 : 4 : 2) as eluent. The column oven was set
polymerization reactions to overcome these issues. But also for to 50 ꢀC. SEC of the deprotected statistical copolymers was
this approach, the variety of possible end-groups for selective performed on a Shimadzu system with a LC-10AD pump, a RID-
coupling reactions and the structural control over the PEI block 10A refractive index detector, a system controller SCL-10A, a
are strongly limited under the hydrolysis conditions.41
degasser DGU-14A, and a CTO-10A column oven using N,N-
However, the possibilities provided by the cationic ring dimethylacetamide with 2.1 g Lꢁ1 LiCl as the eluent and the
opening polymerization (CROP) of 2-oxazolines are very versatile. column oven set to 50 ꢀC. Poly(styrene) (PS) samples were used
It is possible to introduce functionalities into the polymer using as calibration standards for both solvent systems. SEC
functional initiators, end capping reagents and variation of the measurements in water were performed on a Jasco system
substituent at the 2-position of the monomer.42–44 The living equipped with a RI detector using a 0.1% aqueous solution of
character of the reaction leads to an excellent control over the triuoro acetic acid (TFA) with 0.05 mol Lꢁ1 NaCl using a pul-
polymer length and the polydispersity index (PDI).45 This variety lulan standard.
is only limited by the requirements of the cationic polymeriza-
tion, where nucleophiles can quench the propagating chain.
Proton NMR spectroscopy (1H NMR) measurements were
performed at room temperature on a Bruker AC 300 and 400
Cesana et al. reported the synthesis of cationic POx in 2006.46 MHz spectrometer, using CDCl3, MeOD or DMSO-d6 as solvents.
They used a 2-oxazoline monomer bearing a protected amino The chemical shis are given in ppm relative to the signal from
group at the 2-position to create polymers, which exhibit the residual non-deuterated solvent. Fourier transform infrared
cationic side chains aer deprotection. To the best of our (FTIR) spectroscopy was performed on an Affinity-1 FT-IR from
knowledge, this represents the only study about cationic POxs to Shimadzu, using the reection technique. Gas chromatography
date. With this type of monomer it is possible to adjust the (GC) was performed on a GC-2010 from Shimadzu. Acetonitrile
sequence, length and charge density rather easily by the CROP was used as an internal standard to determine the monomer
technique. However, the potential of this polymeric system has conversion.
never been used for biochemical applications.
High resolution electrospray ionization (HR-ESI) mass
Moreover, (co)polymers containing amino side chains spectrometry (MS) was performed on a micrOTOF Q-II (Bruker
exhibit enormous potential for the design of smart hydrogels. Daltonics) mass spectrometer equipped with an automatic
While the synthesis of hydrogels from partially47–49 or fully syringe pump from KD Scientic for sample injection at 4.5 kV
cleaved50–52 POxs is known in the literature, the application of at a desolvation temperature of 180 ꢀC. The mass spectrometer
these systems for the reversible binding of DNA, e.g. for the was operating in the positive ion mode. Matrix assisted laser
enrichment of genes in biochip devices, has never been desorption ionization (MALDI) time of ight (ToF)-MS spectra
described to the best of our knowledge. Besides the swelling were recorded on an Ultraex III TOF/TOF (Bruker Daltonics
value, a basic structural requirement for such hydrogels GmbH, Bremen, Germany). The instrument was equipped with
displays the presence of amino groups in the gel matrix to a frequency-tripled Nd:YAG operating at a wavelength of
interact with the DNA in solution.
355 nm. Spectra were recorded in the positive reector and in
This report focuses on the synthesis and detailed charac- linear mode. 2,5-Dihydroxybenzoic acid was used as a matrix for
terization of amino bearing POxs and their ability to bind DNA. the sample preparation.
4694 | Soft Matter, 2013, 9, 4693–4704
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Differential scanning calorimetry (DSC) was performed on a
Soft Matter
FTIR: ꢀn (cmꢁ1) ¼ 3377, 3312, 3275, 2980, 2944, 1747, 1691,
Netzsch DSC 204 F1 Phoenix under a nitrogen atmosphere with 1671, 1646, 1547, 1524, 1481, 1366, 1277, 1249, 1219, 1160.
a heating rate of 20 K minꢁ1 from ꢁ100 to 200 ꢀC. 3 Cycles were
recorded for each sample. The glass transition temperature (Tg)
values are reported for the second heating run. Thermo-gravi-
metric analysis (TGA) was performed under a nitrogen atmo-
2-(4-((tert-Butoxycarbonyl)amino)butyl)-2-oxazoline (BocOx) (4)
3 (1.93 g, 6.9 mmol) was dissolved in DMF (150 mL) and
potassium carbonate (1.91 g, 13.8 mmol) was added. The
suspension was stirred at 70 ꢀC for 5 h. Aer evaporation of the
solvent, the yellow oil was dissolved in dichloromethane and
washed with distilled water (3 ꢂ 100 mL). The solution was
dried using sodium sulfate and the solvent was evaporated
sphere on a Netzsch TG 209 F1 Iris in the range froꢁm1 room
ꢀ
temperature to 800 C with a heating rate of 10 K min
.
5-((tert-Butoxycarbonyl)amino)valeric acid53 (2)
5-amino valeric acid (1, 5.15 g, 50.4 mmol) was dissolved in a under reduced pressure. The crude product was puried by
mixture of aqueous sodium hydroxide solution (2 wt%, 100 mL) distillation (6 ꢂ 10ꢁ2 mbar, 115 ꢀC) to yield BocOx as a colorless
and dioxane (100 mL). Di-tert-butyldicarbonate (DiBoc) (11 g, viscous liquid (1 g, 4.14 mmol, 60%, r ¼ 0.97 g cmꢁ3).
50.4 mmol) was dissolved in dioxane (50 mL) and added drop-
1H NMR (CDCl3, 300 MHz): d ¼ 4.65 (1H, s, NH), 4.21 (2H, t,
wise to the yellow solution. The reaction mixture was stirred CH2–oxazoline, 9.3 Hz), 3.80 (2H, t, CH2–oxazoline, 9.8 Hz), 3.12
for 20 h at room temperature and acidied using aqueous HCl (2H, t, CH2–CH2–NH, 6.3 Hz), 2.28 (2H, t, CH2–CH2–CON, 7.2
(1 mol Lꢁ1) until a pH of 3 was reached. The turbid mixture was Hz), 1.65 (2H, q, CH2–CH2–CH2, 7.40 Hz), 1.53 (2H, q, CH2–
washed with dichloromethane (3 ꢂ 50 mL), the combined CH2–CH2, 6.6 Hz), 1.42 (9H, s, CH3) ppm.
organic phases were washed with water (3 ꢂ 100 mL) and dried
HR-ESI: m/z calc. for C12H22N2O3Na [M + Na]: 265.1523;
over sodium sulfate. Aer evaporation of the solvent and drying found: 265.1534 (error: 4 ppm).
in a high vacuum, the product was obtained as a white crys-
talline solid (8.41 g, 39 mmol, 77%).
FTIR: ꢀn (cmꢁ1) ¼ 3325, 2978, 2932, 1694, 1663, 1524, 1451,
1366, 1246, 1165.
1H NMR (DMSO-d6, 300 MHz): d ¼ 5.77 (0.3H, s, COOH), 4.60
(0.77H, s, NH), 3.10 (2H, s, CH2–CH2–N), 2.38 (2H, t, CH2–CH2–
COO), 1.67 (2H, q, CH2–CH2–CH2, 7.10 Hz), 1.54 (2H, q, CH2–
CH2–CH2, 6.39 Hz), 1.45 (9H, s, CH3) ppm.
Kinetic studies
For kinetic investigations an in situ oligoinitiator was synthe-
sized from a stock solution of EtOx, MeTos (3 : 1) and aceto-
nitrile ([M] ¼ 1 mol Lꢁ1) by microwave-assisted polymerization
HR-ESI: m/z calc. for C10H19NO4Na [M + Na]: 240.1206,
found: 240.1243 (error: 15.4 ppm).
ꢀ
(140 C, 90% conversion). Subsequently, BocOx (and EtOx for
FTIR: ꢀn (cmꢁ1) ¼ 3374, 2983, 2951, 2879, 1720, 1683, 1520,
1486, 1462, 1444, 1431, 1415, 1388, 1361, 1330, 1277, 1242,
1204, 1165.
the copolymerization) was added to the initiator mixture. The
stock solution was aliquoted into microwave vials (1 mL per
ꢀ
vial) and heated in the microwave synthesizer (140 C, varying
reaction times). Aer polymerization, the conversions of the
monomers were determined by GC with acetonitrile as an
tert-Butyl (5-((2-chloroethyl)amino)-5-oxopentyl)carbamate (3)
To a solution of 2 (2 g, 9.21 mmol) in tetrahydrofuran (THF) internal standard. The reaction rate constant kP of the
(80 mL), TEA (1.276 mL, 9.21 mmol) was added and the mixture monomers was determined using eqn (1) and (2) assuming
was cooled to 0 ꢀC. Aer a dropwise addition of ethyl chlor- that the slope of the linear t of ln([M]0/[M]t) ¼ f(t) complies
oformate (0.87 mL, 9.21 mmol) the reaction solution was with keff
.
allowed to reach room temperature and stirred for 1 h. Aer
cooling again to 0 C, a mixture of 2-chloroethylamine hydro-
chloride (1.067 g, 9.21 mmol) and TEA (1.276 mL, 9.21 mmol) in
dimethylformamide (DMF) (10 mL) was added and the reaction
mixture was stirred for 1.5 h.
ꢀ
ln M0 ꢁ ln Mt ¼ keff
keff ¼ kp[I]
t
(1)
(2)
Aer evaporation of the solvent, the residue was re-dissolved
in dichloromethane and washed with aqueous sodium bicar-
Homopolymerization of BocOx (PBocOx) (5–8)
bonate solution (3 ꢂ 50 mL) and brine (3 ꢂ 50 mL). The solu- In a microwave vial, EtOx (121 mL, 1.2 mmol), MeTos (60.6 mL,
tion was dried over sodium sulfate and the solvent was 0.4 mmol) and acetonitrile (2.95 mL) were mixed under inert
evaporated. The intermediate product was obtained as a white conditions. Aer heating the vial in the microwave at 140 ꢀC for
crystalline solid (1.93 g, 6.9 mmol, 75%) and used without 6 min, BocOx (403.7 mL, 4 mmol) was added under an argon
further purication.
stream and the mixture was heated again in the microwave
1H NMR (CDCl3, 300 MHz): d ¼ 6.13 (1H, s, CH2–NH–CO), synthesizer (140 C, 5 min). The polymer solution was diluted
4.64 (1H, s, Boc–NH–CO), 3.61 (4H, m, CH2Cl–CH2–NH), 3.13 with dichloromethane (50 mL) and extracted with saturated
(2H, t, CH2–CH2–NHBoc, 6 Hz), 2.25 (2H, t, CH2–CH2–CON, 7.14 aqueous solution of sodium bicarbonate (3 ꢂ 50 mL) and water
Hz), 1.68 (2H, q, CH2–CH2–CH2, 7.40 Hz), 1.52 (2H, q, CH2– (3 ꢂ 50 mL). Subsequently, the solution was concentrated and
ꢀ
CH2–CH2, 6.6 Hz), 1.44 (9H, s, CH3) ppm.
the polymer was precipitated in 150 mL ice-cold diethyl ether.
HR-ESI: m/z calc. for C12H23ClN2O4Na [M + Na]: 301.1289; The white precipitate was ltered and dried in a high vacuum
found: 301.1318 (error: 9.6 ppm).
(392 mg, 85%).
This journal is ª The Royal Society of Chemistry 2013
Soft Matter, 2013, 9, 4693–4704 | 4695

View Article Online
Soft Matter
Paper
1H NMR (CDCl3, 300 MHz): d ¼ 5.17 (s, 1H, NH), 3.44 (s, 4H,
1H NMR (MeOD, 400 MHz): d ¼ 3.52 (s, 4H, backbone), 2.96
backbone), 3.10 (s, 2H, CH2–CH2–NH), 2.35 (s, 2H, CH2–CH2– (0.23H, s, CH2–CH2–NH2), 2.42 (2H, s, CH2 (EtOx) + CH2–CH2–
CO), 1.62 (s, 2H, CH2–CH2–CH2), 1.52 (s, 2H, CH2–CH2–CH2), CO (AmOx)), 1.69 (0.41H, s, CH2–CH2–CH2–CH2 (AmOx)), 1.11
1.41 (s, 9H, CH3) ppm.
(2.8H, s, CH3 (EtOx)) ppm.
SEC (5) (eluent: CHCl3–i-propanol–TEA, PS-standard): Mn ¼
SEC (18) (eluent: DMAc–LiCl, PS-standard): Mn ¼ 8100 g
3700 g molꢁ1, Mw ¼ 4700 g molꢁ1, PDI ¼ 1.25.
molꢁ1, Mw ¼ 10 200 g molꢁ1, PDI ¼ 1.26.
Hydrogel synthesis (23–27)
Copolymerization of EtOx and BocOx (P(EtOx-stat-BocOx)) (9–13)
P(EtOx-stat-AmOx) (18, 100 mg, 0.1 mmol NH2-groups) was
dissolved in a 5 wt% aqueous solution of sodium hydroxide.
Epichlorohydrin (3.92 mL, 0.05 mmol) was added to the reaction
In a microwave vial, EtOx (181.7 mL, 1.8 mmol), MeTos (90.8 mL,
0.600 mmol) and acetonitrile (11.5 mL) were mixed under inert
conditions. Aer heating in the microwave at 140 ꢀC for 16 min
EtOx (2543 mL, 25.2 mmol) and BocOx (749 mL, 3 mmol) were
added under an argon stream and the mixture was heated again
in the microwave synthesizer (140 ꢀC, 11 min). The solution was
diluted with dichloromethane (100 mL) and extracted with
saturated aqueous solution of sodium bicarbonate (3 ꢂ 100 mL)
and water (3 ꢂ 100 mL). Subsequently, the solution was
concentrated and the polymer was precipitated in 400 mL ice-
cold diethyl ether. The white precipitate was ltered and dried
in a high vacuum (2.77 g, 78%).
ꢀ
solution and the mixture was heated at 50 C for 1 h. Gelation
occurred aer 10 min. The gel was washed with methanol (5 ꢂ
20 mL) and water (2 ꢂ 20 mL) and freeze dried at ꢁ35 ꢀC and 0.2
mbar (55 mg, 52%).
FTIR: ꢀn (cmꢁ1) ¼ 2978, 2936, 1628, 1466, 1420, 1373, 1315,
1238, 1196, 1126, 1061.
Swelling studies
The water uptake of the freeze dried hydrogel samples was
measured gravimetrically using centrifuge lter tubes. The lter
tube was saturated with water and the excess solvent was
removed by centrifugation (3000 rpm, 10 min). The tube was
weighed to yield m0. Aer addition of the hydrogel sample (10 to
20 mg), the tube was weighed again (m0,gel) and the sample
weight (mgel) was determined using eqn (3).
1H NMR (CDCl3, 300 MHz): d ¼ 4.99 (s, 0.2H, NH), 3.45 (s,
4H, backbone), 3.11 (s, 0.3H, CH2–CH2–NH (BocOx)), 2.40 (s,
1.9H, CH2 (EtOx)), 1.92 (s, 0.3H, CH2–CH2–CO (BocOx)), 1.64
(s, 0.3H, CH2–CH2–CH2 (BocOx)), 1.53 (s, 0.3H, CH2–CH2–CH2
(BocOx)), 1.42 (s, CH3 (BocOx)), 1.21 (s, 1.3H, CH3 (EtOx)) ppm.
SEC (9) (eluent: CHCl3–i-propanol–TEA, PS-standard): Mn ¼
5300 g molꢁ1, Mw ¼ 5700 g molꢁ1, PDI ¼ 1.08.
mgel ¼ m0,gel ꢁ m0
(3)
Deprotection of PBocOx (PAmOx) (14–17)
Aer swelling of the sample in water for 24 h the lter tube
was centrifuged (3000 rpm, 10 min) and weighed (mwet) to
determine the mass of the swollen gel (msw) using eqn (4).
PBocOx (5, 195 mg, 0.8 mmol of amino groups) was dissolved in
dichloromethane (2.5 mL) and TFA (617 mL, 8 mmol) was added
subsequently. Aer heating at 60 ꢀC for 1 h and stirring for 24 h
at room temperature, the mixture was diluted with 5 mL of
methanol and precipitated in 150 mL of ice-cold diethyl ether.
The yellow precipitate was re-dissolved in methanol (50 mL) and
stirred with Amberlyst A21 for 24 h. Subsequently, the solution
was concentrated and the polymer was precipitated in ice-cold
diethyl ether (150 mL), ltered, dried in a high vacuum and
obtained as a yellowish powder (103 mg, 90%).
msw ¼ mwet ꢁ m0
(4)
The swelling degree (Qeq) was calculated according to
eqn (5).54
msw ꢁ mgel
Qeq
¼
ꢂ 100%
(5)
msw
The gelated polymer fraction (FP) was determined using
eqn (6).54
1H NMR (D2O, 400 MHz): d ¼ 3.51 (s, 4H, backbone), 2.85 (s,
2H, CH2–CH2–NH2), 2.42 (2H, CH2–CH2–CO), 1.65 (s, 4H, CH2–
CH2–CH2–CH2) ppm.
mth
md
FP ¼
ꢂ 100%
(6)
SEC (14) (eluent: water–TFA–NaCl, pullulan-standard): Mn ¼
15 100 g molꢁ1, Mw ¼ 18 500 g molꢁ1, PDI ¼ 1.23.
mth: theoretical mass of the polymer in the gel (based on the
mass of the polymer precursor and cross-linker), and md: mass
of the dried sample.
Deprotection of P(EtOx-stat-BocOx) (P(EtOx-stat-AmOx)) (18–22)
P(EtOx-stat-BocOx) (9, 2 g) was dissolved in TFA (5 mL) and
Ethidium bromide assay (EBA) of P(EtOx-stat-PAmOx)
ꢀ
heated at 60 C for 1 h. Aer stirring for 24 h at room temper-
ature, the mixture was diluted with 10 mL methanol and The complex formation of plasmid-DNA (pDNA, 4700 base
precipitated in 400 mL of cold diethyl ether. The yellowish pairs) with cationic polymers was detected by quenching of the
precipitate was re-dissolved in methanol (200 mL) and stirred EB uorescence as described in the literature.55
with Amberlyst A21 for 24 h. Subsequently, the solution was
pDNA (7.5 mg mLꢁ1) and EB (0.4 mg mLꢁ1) were dissolved in
concentrated and the polymer was precipitated in cold diethyl HBG-buffer (HEPES buffered glucose, pH 7) and incubated for 10
ether (400 mL), ltered, dried in a high vacuum and obtained as min at room temperature. 100 mL of the pDNA–EB solution were
a yellowish powder (1.58 g, 87%).
transferred to the wells of a black 96-well plate (Nunc,
4696 | Soft Matter, 2013, 9, 4693–4704
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Soft Matter
Langenselbold, Germany) containing different polymer concen-
trations. Fluorescence was measured aer 20 min of incubation
with the polymer solution using a Tecan M200 Pro uorescence
microplate reader (Crailsheim, Germany) at the following wave-
lengths: Ex 525 nm/Em 605 nm. An internal standard containing
only pDNA and EB was used to calibrate the measurements.
EBA of hydrogel samples and release studies
Between 0.8 and 2.25 mg of hydrogels per well (12-well plate)
were used depending on the amino content of the samples to
reach a nitrogen to phosphate (N/P) value of 1000 for every
sample. The gels were swollen overnight in 250 mL of HBG-
buffer (pH 7). Subsequently, 1 mL of pDNA–EB solution (con-
taining 7.5 mg pDNA per mL and 0.4 mg EB per mL) was added
and the sample aliquots of 50 mL for every time interval were
taken. Fluorescence was measured in black 96-well plates using
a Tecan M200 Pro uorescence microplate reader.
For release studies, 1 mL of a mixture of heparin (6 mg mLꢁ1
,
2 eq.) and EB (0.4 mg mLꢁ1) in HPG-buffer (pH 7) was added to
each of the swollen gel samples. For every measuring point,
50 mL were collected and the uorescence was determined as
described above.
Microscopic detection of hydrogel supported DNA binding
and release
Approximately 0.18 mg of hydrogels per well were transferred
into the cavity of a clear bottom black walled 96 well plate and
the gels were swollen overnight in 50 mL of HBG-buffer (pH 7).
Next, 100 mL of pDNA–EB (containing 7.5 mg pDNA per mL and
0.4 mg EB per mL) solution were added and the uorescence
signal was captured directly in the wells using a uorescence
microscope (Cell Observer Z1, Carl Zeiss, Jena, Germany)
equipped with a mercury arc UV lamp and the appropriate lter
combinations for excitation and detection of emission. Images
of a series (11 ꢂ 11 pictures per well) were captured with a 10ꢂ
objective using identical instrument settings (e.g. UV lamp
power, integration time, and camera gain) and spots of the 96
well plate were addressed using an automated XY table. Control
samples contained only 100 mL of pDNA–EB (7.5 mg pDNA per
mL and 0.4 mg EB per mL) and 50 mL of HBG-buffer. Microscopic
detection of the resulting uorescence signal was performed at
different time points (0 min, 5 min, 30 min and 120 min).
For the release of the DNA from the hydrogel, 50 mL of
heparin (60 mg mLꢁ1) were added directly to the wells and the
resulting uorescence signal was immediately analyzed micro-
scopically as described above.
Scheme 1 Schematic overview of the synthesis of the amino bearing 2-oxazo-
line monomer and its polymerization with and without a comonomer. (a) diBoc,
dioxane–water, NaOH; (b) (1) ethylchloroformate, chloroethylamine, TEA, DMF,
(2) DMF, K2CO3, 60 ꢀC; (c) mW, 140 ꢀC, acetonitrile; (d) TFA (10 eq.), dichloro-
methane, 60 ꢀC, Amberlyst A21 and (e) TFA, 60 ꢀC, Amberlyst A21.
reversibly bind DNA. For this purpose, (POx)s suitable for cross-
linkingreactionsandbindingofDNAare necessary. Bothfeatures
can be accomplished by amino groups in the side chain of the
polymers. Because free amino functionalities would quench a
CROP immediately, it is necessary to use a monomer with a
protected amino group, which is separated from the oxazoline
ring by a spacer. The distance between the amino group and the
ring should be sufficiently large to ensure that the polymerization
isnotinuencedbythefunctionalgroup.Inaddition, itshould be
short enough to yield polymers with a certain hydrophilicity.
To this end, 5-amino valeric acid was chosen as a raw material
for the monomer synthesis. The butoxy-carbonyl (Boc)-protection
of the amino functionality was carried out according to a patent
from Fino et al.53 The amidation with chloroethylamine hydro-
chloride was performed in a similar fashion as reported by
Cesana et al.46 in a THF–DMF solvent mixture with ethyl chlor-
Quantication of the relative uorescence intensity signal
was performed using 8 bit, grayscale-converted images and
ImageJ soware. The average uorescence intensity per pixel was
obtained and the samples were normalized to the references.
1
oformate as an activation agent. The H NMR spectrum of the
intermediate shows that 10% of the nal monomer had already
been formed as indicated by the appearance of the respective
oxazoline ring signals. Eventually, BocOx was obtained by treat-
ment of the intermediate mixture with potassium carbonate in
Results and discussion
Monomer synthesis
The aim of this work was the synthesis of precursors for the DMF. The monomer was distilled to yield a product with a purity
production of hydrogels (Scheme 1), which possess the ability to that fullls the requirements of CROP.
This journal is ª The Royal Society of Chemistry 2013
Soft Matter, 2013, 9, 4693–4704 | 4697

View Article Online
Soft Matter
Paper
Fig. 1 Kinetic study of the homopolymerization of BocOx. (A) First-order kinetic plot, (B) SEC traces of the kinetic samples and (C) Mn against conversion plot as well as
PDI values of PBocOx kinetic samples.
Cesana et al. reported that an initiation using small cations
To investigate the properties of this polymer, four homo-
such as the methyl cation of the common initiators, MeTos and polymers with different lengths were synthesized using the data
methyl triate, is not suitable for this type of monomer because obtained from the kinetic study (Table 1). The values for the
the cationic species will also attack the amide nitrogen of the degree of polymerization (DP) of PBocOxs were determined by
oxazoline side chain. Preliminary tests supported these results 1H NMR, comparing the signal of the polymer backbone with
yielding polymers with multimodal SEC traces (data not the peaks of the initiator (tosylate) before purication. Aer
shown). Therefore, a two-step initiation was used for the poly- extraction and precipitation SEC measurements revealed high
merization of BocOx. First, the desired amount of initiator was PDI values for DP > 35 and, therefore, a loss of control for higher
mixed with three equivalents of EtOx in acetonitrile. This DP values. Similar observations were made for PEtOx as well,
solution was heated under microwave conditions until a however, at higher molar masses.57 MALDI-ToF analysis of the
conversion of ꢃ90% was reached (calculated from rate polymers revealed complex spectra, which is attributed to the
constants reported in the literature56). In a second step, BocOx initiation method and the resulting formation of copolymer like
was introduced into the microwave vial under inert conditions systems. Surprisingly, the main distribution of the spectrum of
and polymerized in the microwave. The EtOx-oligomer 5 is an H-initiated PBocOx without an EtOx unit as a starting
produced in step one acts as an initiator for the polymerization group. The thermal behavior of these polymers was investigated
of the BocOx monomer and the sterical hindrance of the oxa- via DSC and TGA (Table 1). The Tg values of PBocOxs range
zolinium species prevents an attack at the side chain. A ratio of between 36 and 47 ꢀC with a dependence on the molar mass. It
EtOx to MeTos of 3 : 1 was necessary to ensure that all methyl is known that the Tg is increasing with increasing molar mass
cations are consumed before the addition of BocOx. Tests with a until a plateau is reached.58 For the present polymers, the rather
ratio of 1 : 1 resulted in signicantly higher PDI values.
small change of 1 ꢀC from DP ¼ 35 to DP ¼ 125 indicated that
the value of 47 ꢀC is close to the maximum Tg for this polymer
class. The thermal degradation of PBocOx was investigated by
TGA under a nitrogen atmosphere. The decomposition was
dened as the temperature where 5 wt% of the substance is
destroyed. The difference between the degradation tempera-
tures of the samples is most likely attributed to the uncertainty
of the measurement. The differing amount of EtOx cannot be
Synthesis and characterization of BocOx homopolymers
The rst step to well-dened polymers by CROP is the knowl-
edge about the reaction kinetic of the polymerization under the
given conditions. Consequently, a kinetic study of the homo-
polymerization of BocOx was performed using the initiation
method described above. The conversion of the monomer was
determined by GC using the polymerization solvent (acetoni-
trile) as an internal standard. The resulting pseudo rst-order
kinetic plot is depicted in Fig. 1A. The linearity of the graph
demonstrates a constant concentration of the propagating
species during the polymerization, which is indicative of a living
polymerization mechanism of BocOx. The polymerization rate
was determined from the slope of the linear t of the log plot
using eqn (1) and (2). The kp value was found to be 153 L molꢁ1
sꢁ1 ꢂ 10ꢁ3, which is in the same order of magnitude as EtOx in
Table 1 Detailed characterization data for PBocOx and PAmOx. The Mn values
were calculated from the M/I values of the polymerization
Thermal
Sample
number Polymer
Mn (calc.) Mn (SEC)
Yield Tg stability
PDI [%]
[g molꢁ1
]
[g molꢁ1
]
[ꢀC] [ꢀC]
5
6
7
8
14
15
16
17
PBocOx10
PBocOx25
PBocOx35
2000
5600
8000
2600
3700
6400
10 000
15 700
21 100
32 200
52 800
1.18 82
1.25 85
1.25 62
1.51 77
1.19 90
1.23 96
1.38 95
2.07 94
36 179
40 206
46 212
47 209
acetonitrile under the same conditions (105 L molꢁ1 sꢁ1
ꢂ
PBocOx125 30 000
10ꢁ3).56 The livingness of this polymerization is supported by
the linear increase of the molar mass with conversion (Fig. 1C),
as well as monomodal SEC curves and low PDI values (1.2 < PDI
< 1.3) (Fig. 1B and 1C).
PAmOx10
PAmOx25
PAmOx35
1800
3900
5300
—
192
16 197
27 201
25 190
PAmOx125 18 100
4698 | Soft Matter, 2013, 9, 4693–4704
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Soft Matter
responsible for the change since PEtOx itself decomposes deliberately, also because the length of the polymer should not
thermally at around 350 ꢀC.59 The degradation takes place in have changed signicantly during deprotection. The MALDI-
two steps. The rst one is associated with a mass loss of 41 wt% ToF spectra of PAmOx show even more distributions than the
at 250 ꢀC and indicates that the protection group is cleaved protected polymers. A reason could be the raised laser intensity,
thermally.
which was necessary to desorb and ionize the polymers,
together with an attributed degradation. However, the most
prominent difference between peaks of the major distributions
is 142 g molꢁ1, which is equal to the mass of AmOx. The glass
transition of the polymers changed signicantly aer depro-
Deprotection of the BocOx homopolymers
The chemical deprotection of the synthesized polymers was
carried out under acidic conditions as commonly applied in the
literature.60 The polymers were dissolved in a 1 : 1 mixture of
TFA and dichloromethane under elevated temperatures (60 ꢀC)
to yield the respective deprotected polymers (PAmOx). The
resulting TFA-anion was removed with Amberlyst A21, because
PAmOx is too hydroscopic for an extraction with aqueous
ꢀ
tection to lower values (16 to 27 C).
The Tg of polymer 14 could not be determined because it
interferes with a dip at around 0 ꢀC, which is present in the DSC
curves of all PAmOx samples. Most likely this drop-off is asso-
ciated with adherent water molecules. PAmOx exhibits a high
hydroscopic tendency. Aer some minutes under ambient
conditions, the consistency of the polymers changes from a
powder to a highly viscous liquid. Also TGA shows approxi-
mately 10% of water content for all polymers. The thermal
stability does not change signicantly in comparison to the
protected samples.
1
sodium bicarbonate solution as usually applied for PEtOx. H
NMR spectroscopy proves the successful deprotection since the
signal of the tert-butyl group has vanished in all cases (Fig. 2).
SEC measurements of PAmOx were performed in water under
acidic conditions. However, the standard for the measurements
was pullulan which is due to its structural difference and the
lack of cationic charges a poor comparison for these polymers.
Thus, the obtained molar masses should be considered
Statistical copolymerization
Because the water solubility of PAmOx was found to be unsat-
isfying, it was necessary to introduce a second monomer to yield
polymers which combine functionality with hydrophilicity. The
kp value of BocOx already indicates the possibility of a statistical
copolymerization with EtOx. Nevertheless, a kinetic study was
performed to conrm this expectation. Using the same initia-
tion method as described before, a stock solution of the initi-
ating species was produced. A dened amount of BocOx and
EtOx (1 : 4) was added and the solution was distributed over
several microwave vials to investigate the conversions of both
monomers depending on the reaction time by GC. The analyt-
ical data of the investigations are depicted in Fig. 3.
The copolymerization under microwave irradiation at 140 ꢀC
fullls the characteristics for a living polymerization: linear
pseudo rst-order kinetics, a linear increase of the molar mass
with conversion as well as narrow molar mass distributions.
The calculated rate constants of the monomers (kp(EtOx) ¼
159 (L molꢁ1 sꢁ1) ꢂ 10ꢁ3, kp (BocOx) ¼ 175 (L molꢁ1 sꢁ1) ꢂ
10ꢁ3) increased in comparison to the homo-polymerization.
Fig. 2 Comparison of the 1H NMR spectra of P(BocOx) (8, in CDCl3, 300 MHz)
and P(AmOx) (14, in D2O, 400 MHz).
Fig. 3 Kinetic study of the statistical copolymerization of BocOx with EtOx. (A) First-order kinetic plot, (B) SEC traces of the kinetic samples and (C) Mn against
conversion plot as well as PDI values of P(EtOx-stat-BocOx) kinetic samples.
This journal is ª The Royal Society of Chemistry 2013
Soft Matter, 2013, 9, 4693–4704 | 4699

View Article Online
Soft Matter
Paper
Table 2 Detailed characterization data for P(EtOx-stat-BocOx) and P(EtOx-stat-
polymers (P(EtOx-stat-BocOx)). However, the measured molar
mass values appear to be higher in comparison to the protected
polymers, which can be attributed to the positive charges of the
nal polymeric material. Also for the deprotected polymers,
complex MALDI-ToF spectra were obtained, in which both
monomer masses could be identied.
AmOx)
Sample
number Polymer
Mn (SEC)
Yield BocOx/
[g molꢁ1
] PDI [%] AmOx [%]
9
P(EtOx43-stat-BocOx5)
P(EtOx39-stat-PBocOx7) 5200
P(EtOx32-stat-PBocOx9) 4400
P(EtOx38-stat-PBocOx15
P(EtOx32-stat-PBocOx16
P(EtOx43-stat-PAmOx5) 8100
P(EtOx39-stat-PAmOx7) 8200
P(EtOx32-stat-PAmOx9) 8700
5400
1.08 78
1.14 61
1.20 75
1.17 79
1.17 81
1.26 95
1.24 90
1.25 93
1.23 89
1.24 96
10
15
22
27
33
10
15
22
27
33
10
11
12
13
18
19
20
21
22
Hydrogel synthesis
)
)
4600
5400
To perform DNA binding and release on solid supports it was
necessary to form scaffolds out of the synthesized copolymers.
For this purpose, a certain amount of the amino functionalities
was consumed during the reaction with a cross-linker. However,
the remaining primary amines and the secondary and ternary
amines which are formed during the process of linking
should still possess the ability to bind negatively charged gene
material.
P(EtOx38-stat-PAmOx15
P(EtOx32-stat-PAmOx16
)
)
9500
9400
With this knowledge, it was possible to synthesize P(EtOx-stat-
BocOx) with varying amounts of protected amino groups in the
side chain. An overview of the produced polymers is provided in
Table 2. It was aimed to synthesize P(EtOx-stat-BocOx) with a
constant DP of 50 but varying amounts of BocOx between 10 and
30%. The DP, as well as the ratio between the monomers, was
Epichlorohydrin was chosen as a cross-linker. The epoxide
functionality of the molecule reacts with amines under elevated
temperature. Aer coupling to a polymer chain, a second
epoxide function is formed under basic conditions, which can
react with a second polymer chain (Scheme 2). The amount of
cross-linker for the synthesis of hydrogels was chosen for every
copolymer to consume only 5 amino functions per chain
(Table 3). This value is sufficient to form a network while NH2-
groups are le for the DNA interaction for most of the samples.
The mixture of polymer and cross-linker in a 5 wt% sodium
hydroxide solution showed no gelation aer 3 h at room
1
determined by H NMR by comparison of the backbone signal
with either the aromatic peaks of the tosylate or the integral of
the protection group. SEC measurements reveal narrow PDI
values and molar mass values in the expected range. The
MALDI-ToF analysis of these polymers is rather complicated
because of the broad variety of possible combinations of both
monomers. However, it was possible to identify both types of
repeating units in the spectra. The deprotection of the statistical
copolymers was carried out in pure TFA since the reaction in a
mixture of acid with dichloromethane yielded only a partial
cleavage of the protection group. The TFA anion was removed
in a similar fashion as for the homopolymers using Amberlyst
A21. Quantitative deprotection was proven by 1H NMR spec-
troscopy (Fig. 4).
ꢀ
temperature. However, heating the sample to 50 C for 10 min
resulted in the formation of a polymer network. Subsequently
the gels were washed with methanol and water to extract
unlinked polymer chains, residual cross-linker and to
neutralize the hydrogels.
The swelling degrees were measured according to the liter-
ature by centrifugation of the swollen sample for 10 min at 3000
rpm.54 The results are depicted in Table 3. No clear trend in the
swelling degree with the ratio between the two monomers is
visible, although a strong dependency on the gelated polymer
fraction was observed. With an increasing amount of cross-
linking, more polymer chains are incorporated into the scaffold,
while the stiffness of the network increases. Therefore, an
increased gelated polymer fraction indicates a more efficient
cross-linking which leads to lower swelling degrees, because the
balance of the Gibbs free energy of mixing and the Gibbs free
energy associated with the elastic nature of the polymer network
is changed.61 However, the most outstanding property for the
binding of DNA is the content of amino groups which is linearly
increasing within the series of hydrogels.
Beside the disappearance of the tert-butyl peak, the signals of
both middle CH2 groups of the side chain of AmOx shi to
higher ppm values and form one singlet. The SEC analysis of the
copolymers was performed in DMAc (calibration against PS
standard) which should enable a comparison with the protected
DNA-binding and release
The ability of the synthesized polymers to complex DNA in
aqueous solution is crucial for the aimed application. The
ethidium bromide assay offers the possibility to investigate this
interaction by the measurement of the uorescence intensity of
the dye. The intensity differs depending on the environment of
the ethidium bromide. The dye, incorporated into a DNA helix,
shows enhanced uorescence intensity in comparison to the
Fig. 4 Comparison of the NMR data of P(EtOx-stat-BocOx) (in CDCl3) and the
deprotected polymers P(EtOx-stat-AmOx) (in MeOD).
4700 | Soft Matter, 2013, 9, 4693–4704
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Soft Matter
Scheme 2 Schematic representation of the hydrogel synthesis using P(EtOx-stat-AmOx) and epichlorohydrin.
Table 3 Overview of P(EtOx-stat-PAmOx) based hydrogels (per gel 100 mg of polymer were used as the starting material)
Sample number
Feed material
Epichlorohydrin [mL]
Ratio (NH2 : ECH)
Swelling degree [%]
Gelated polymer fraction [%]
23
24
25
26
27
P(EtOx43-stat-PAmOx5)
P(EtOx39-stat-PAmOx7)
P(EtOx32-stat-PAmOx9)
P(EtOx38-stat-PAmOx15
P(EtOx32-stat-PAmOx16
3.92
3.92
4.31
3.33
3.73
5 : 5
7 : 5
9 : 5
15 : 5
16 : 5
97.4 ꢄ 0.2
94.6 ꢄ 0.7
91.1 ꢄ 0.3
95.7 ꢄ 0.1
97.7 ꢄ 0.4
34
45
49
42
32
)
)
free, dissolved species. By displacement of EB with cationic to its higher charge density. The point of saturation of BPEI is
species, such as polycations, the uorescence intensity reached at an N/P ratio of 15 in this assay. To investigate the
decreases and the quality of the complexation can be deter- DNA binding potential of the hydrogels, the procedure for the
mined. BPEI (10 kDa) was used as a reference for the performed EBA was changed. With an insoluble substrate it is not possible
tests. The results of the EBA for P(EtOx-stat-AmOx) are depicted to dilute a stock solution and reach relatively low concentra-
in Fig. 5.
tions as performed for the copolymers. However, for compar-
Most of the copolymers show an interaction with DNA. While ison reasons, the chosen N/P value of 1000 was equal for all
a content of 10% amino bearing monomer per chain seems to hydrogel samples in the tests. The amount of gel varied between
be insufficient for a replacement of ethidium bromide, all other 0.8 and 2.25 mg per well depending on the content of amines in
samples decrease the uorescence intensity. Furthermore, a the scaffold. Subsequently, the hydrogel samples were swollen
dependency on the strength of interaction with the amount of in a 24-well plate in 250 mL of HBG buffer solution. The amount
amino functions is clearly visible. With an increase of 5% of of buffer was not adjusted to the swelling degree to ensure that
amino groups per chain it is possible to decrease the uores- the dilution of DNA–EB stock solution is identical for each well.
cence intensity by around 10%. The plateau, where a further
Aer 24 h of swelling, 1 mL of the mentioned stock solution
excess of positive charges is not able to replace more dye, is was added. For every time interval a sample of 50 mL was taken
reached for all samples at an N/P ratio of 5. However, BPEI and investigated regarding its uorescence intensity. The
shows a higher complexation efficiency which can be attributed results of these measurements are depicted in Fig. 6. It is clearly
visible that all gel samples decrease the uorescence intensity
Fig. 5 EBA of P(EtOx-stat-AmOx) (branched polyethylene imine (BPEI) was used
as positive control).
Fig. 6 EBA of P(EtOx-stat-AmOx) hydrogels.
This journal is ª The Royal Society of Chemistry 2013
Soft Matter, 2013, 9, 4693–4704 | 4701

View Article Online
Soft Matter
Paper
over time. A solution of EB–DNA was used as a reference for
each measuring point and treated in the same way as the other
samples. Therefore, the decrease of intensity cannot be attrib-
uted to a degradation of the EB–DNA complex over time. In
contrast to the polymeric precursors, no clear trend concerning
the percental amount of amino groups is visible, though the gel
with the highest content of AmOx (27) reaches the plateau of
constant uorescence in 2 h before the other samples. More-
over, it can be observed that the uorescence intensity reached
by the gel with the lowest amount of comonomer (23) is in
contrast to the other samples 10% higher. A possible explana-
tion is that this gel is saturated with DNA at the point where the
plateau is reached, while the absorbance capacity of the other
gel samples is high enough to bind all DNA from the solution.
The residual uorescence signal of about 10% is attributed to
the remaining uorescence of the free dye in solution. The
relatively large error bars, in particular in the region of strong
changes of the intensity, are attributed to the solid substrate for
DNA uptake. The long adsorption period of all gels indicates a
diffusion control for the process of DNA immobilization.
Differences of the surface/volume ratio lead to a large deviation
between the three measurements which were performed for
each gel.
Fig. 8 Investigation of DNA binding and release using a fluorescence micro-
scope (Gel sample 27). Left side: absorption and release of gene material in single
wells (96-well plate), indicated by fluorescence (red). Right side: transmission
picture of the well with the swollen gel sample.
For biochip applications it is necessary to absorb but also to
release the DNA, using a certain stimulus. For the release
studies, heparin, a polyanion with multiple negative charges per and a potential loss of hydrogel during the assay can be
repeating unit, was used to replace the DNA in the polymeric responsible for this observation.
scaffold. The heparin was dissolved in an aqueous EB solution
As a nal proof of the reversible absorption of DNA into
of the same concentration as applied before. The released DNA P(EtOx-stat-AmOx) hydrogels, the process of immobilization
should bind to the EB in solution and increase its uorescence. and release was investigated using a uorescence microscope
As a standard for these measurements, an EB–DNA solution was (Fig. 8). The experiment clearly shows the decrease of uores-
used and dened as 100%. The concentration of DNA was equal cence intensity of DNA–EB aer the addition to the hydrogel as
to the amount, which should be bound in the gel if all material compared to the reference, containing a DNA–EB solution of the
was adsorbed during the rst assay. The results of the release same concentration. The relative intensity values were calcu-
studies are depicted in Fig. 7. The intensity increases rapidly lated from the gray scale values of the samples normalized to
aer the addition of heparin and reaches a plateau at ca. 30 min the references. In contrast to the previous measurements,
for all samples. The uorescence value of around 60% indicates the uorescence intensity only decreases to a value of 50%,
an incomplete release. But also the dilution of the test solution which is related to the method. In contrast to the photometric
by the buffer, which was still present in the swollen gel sample, measurements of the supernatant of P(EtOx-stat-AmOx) copoly-
mers and the respective hydrogels, the microscopic study was
performed with the complete sample including the hydrogel.
Since the DNA, which is bound to the gel, has still free coordi-
nation sites for EB, a uorescence signal is still detectable in the
area associated with the hydrogel location. This observation was
conrmed by the fact that the red uorescence is more domi-
nant in the boundary area of the well, where the hydrogel is
mainly located, as indicated by the transmission light micro-
graphs (Fig. 8).
The disintegration of the hydrogel-coordinated DNA turned
out to be even faster than in the previous investigations. As
indicated by the obtained gray scale values, a full release of DNA
occurs immediately aer the addition of heparin. The pictures
clearly show an increased intensity in the area of the hydrogel
which proves that the DNA is replaced by heparin but not
diffused out of the gel (Fig. 8). This observation supports the
assumption that the release of gene material from the hydrogels
Fig. 7 Heparin induced release of DNA from P(EtOx-stat-AmOx) hydrogels.
4702 | Soft Matter, 2013, 9, 4693–4704
is diffusion controlled.
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Soft Matter
4 C. W. Wong, T. J. Albert, V. B. Vega, J. E. Norton, D. J. Cutler,
T. A. Richmond, L. W. Stanton, E. T. Liu and L. D. Miller,
Genome Res., 2004, 14, 398–405.
5 L. G. Britcher, D. C. Kehoe, J. G. Matisons, R. S. C. Smart and
A. G. Swincer, Langmuir, 1993, 9, 1609–1613.
6 B. Joos, H. Kuster and R. Cone, Anal. Biochem., 1997, 247, 96–
101.
7 Y.-H. Rogers, P. Jiang-Baucom, Z.-J. Huang, V. Bogdanov,
S. Anderson and M. T. Boyce-Jacino, Anal. Biochem., 1999,
266, 23–30.
8 A. B. Steel, R. L. Levicky, T. M. Herne and M. J. Tarlov,
Biophys. J., 2000, 79, 975–981.
9 F. Fixe, M. Dufva, P. Telleman and C. B. V. Christensen,
Nucleic Acids Res., 2004, 32, e9.
10 E. Waddell, Y. Wang, W. Stryjewski, S. McWhorter,
A. C. Henry, D. Evans, R. L. McCarley and S. A. Soper, Anal.
Chem., 2000, 72, 5907–5917.
11 A. W. Peterson, R. J. Heaton and R. M. Georgiadis, Nucleic
Acids Res., 2001, 29, 5163–5168.
12 J. B. Rampal, Microarrays: Synthesis Methods, Humana Press,
Totowa, 2001.
Conclusion and outlook
In summary it was shown that amino bearing POx can be
synthesized starting from an oxazoline monomer with a pro-
tected amino functionality at the 2-position (BocOx). Kinetic
studies were performed for a homopolymerization and a series
of polymers (PBocOx) were synthesized. The deprotected poly-
mers (PAmOx) were characterized regarding their structure, Tg
values and thermal degradation behavior. Because of a lack of
water solubility of the resulting homopolymers, EtOx was used
as a comonomer. Kinetic investigations revealed that both
monomers possess similar polymerization rates in a copoly-
merization. P(EtOx-stat-BocOx) were synthesized with differing
amounts of BocOx (10 to 33%) and deprotected subsequently. It
was shown that these polymers are able to form complexes with
DNA using the EB-assay. The strength of the interaction corre-
lates directly with the amount of amino groups per polymer.
P(EtOx-stat-AmOx)s were cross-linked using epichlorohydrin to
form hydrogels. These polymeric scaffolds were investigated
regarding their swelling degrees and ability to absorb DNA from
aqueous solution in a similar manner as investigated for the
polymer precursors. It was shown that all hydrogels possess the
ability to immobilize DNA from solution. It was also possible to
release DNA from the scaffolds by the addition of heparin as a
polyanion. Amounts larger than 60% of the initial DNA quantity
could be set free again by the replacement with heparin. The
process was also investigated using a uorescence microscope,
proving the previous ndings and indicating quantitative
release.
13 P. Anzenbacher Jr, Y.-l. Liu and M. E. Kozelkova, Curr. Opin.
Chem. Biol., 2010, 14, 693–704.
14 D. Guschin, G. Yershov, A. Zaslavsky, A. Gemmell, V. Shick,
D. Proudnikov, P. Arenkov and A. Mirzabekov, Anal.
Biochem., 1997, 250, 203–211.
15 O. Okay, J. Polym. Sci., Part B: Polym. Phys., 2011, 49, 551–556.
16 F. Kivlehan, M. Paolucci, D. Brennan, I. Ragoussis and
P. Galvin, Anal. Biochem., 2012, 421, 1–8.
Further studies will focus on the release of gene material
from hydrogels by a certain stimulus. While heparin is suffi-
cient to replace the DNA in the rst tests, a release induced by
temperature or pH change would be more meaningful with
regard to the targeted applications. Furthermore, biochips will
be equipped with the produced hydrogels by inkjet printing to
check their ability for DNA enrichment also on a smaller scale.
17 J. Liu, So Matter, 2011, 7, 6757–6767.
18 M. Wink, An Introduction to Molecular Biotechnology:
Molecular Fundamentals, Methods and Applications in
Modern Biotechnology, Wiley-VCH, Weinheim, Germany,
2006.
19 K. Doyle, The Source of Discovery: Protocols and Applications
Guide, PROMEGA, Madison, Wis, USA, 1996.
20 L. Buckingham and M. L. Flaws, Molecular Diagnostics:
Fundamentals, Methods, & Clinical Applications, F. A. Davis,
Philadelphia, Pa, USA, 2007.
Acknowledgements
¨
The authors would like to thank the Bundesministerium fur
21 J. Sambrook and D. Russel, Molecular Cloning: A Laboratory
Manual, Cold Spring Harbor Laboratory Press, New York,
NY, USA, 2001.
Bildung und Forschung (Germany) for funding (project: BASIS,
03WKCB01C). Kristian Kempe acknowledges the Alexander
von Humboldt foundation. Grateful thanks to Esra Altuntas
(Friedrich Schiller University Jena) for ESI-MS, Sarah Crotty
(Friedrich Schiller University Jena) for MALDI-ToF-MS
measurements and Dr. Stephanie Schubert for correction of the
manuscript.
22 Pierce Strong Ion Exchange Spin Columns, Thermo Scientic,
New Hampshire, NH, USA, 2007.
23 QIAGEN Genomic DNA Handbook, QIAGEN, Valencia, Calif,
USA, 2001.
24 D. T. Gjerse, L. Hoang and D. Hornby, RNA Purication and
Analysis: Sample Preparation, Extraction, Chromatography,
Wiley-VCH, Weinheim, Germany, 2009.
25 K.-H. Esser, W. H. Marx and T. Lisowsky, BioTechniques,
2005, 39, 270–271.
Notes and references
1 J. Peeters and P. Van der Spek, Cell Biochem. Biophys., 2005,
43, 149–166.
2 T. R. Gingeras, D. Y. Kwoh and G. R. Davis, Nucleic Acids Res.,
1987, 15, 5373–5390.
26 H. Pollard, J.-S. Remy, G. Loussouarn, S. Demolombe,
J.-P. Behr and D. Escande, J. Biol. Chem., 1998, 273, 7507–
7511.
3 J. M. Anderton, R. Tokarz, C. D. Thill, C. J. Kuhlow, 27 O. Boussif, F. Lezoualc'h, M. A. Zanta, M. D. Mergny,
C. S. Brooks, D. R. Akins, L. I. Katona and J. L. Benach,
Infect. Immun., 2004, 72, 2035–2044.
D. Scherman, B. Demeneix and J. P. Behr, Proc. Natl. Acad.
Sci. U. S. A., 1995, 92, 7297–7301.
This journal is ª The Royal Society of Chemistry 2013
Soft Matter, 2013, 9, 4693–4704 | 4703

View Article Online
Soft Matter
Paper
28 G. Y. Wu and C. H. Wu, J. Biol. Chem., 1988, 263, 14621–14624. 45 N. Adams and U. S. Schubert, Adv. Drug Delivery Rev., 2007,
29 G. Y. Wu and C. H. Wu, J. Biol. Chem., 1987, 262, 4429–4432. 59, 1504–1520.
30 J. Luten, N. Akeroyd, A. Funhoff, M. C. Lok, H. Talsma and 46 S. Cesana, J. Auernheimer, R. Jordan, H. Kessler and
W. E. Hennink, Bioconjugate Chem., 2006, 17, 1077–1084. O. Nuyken, Macromol. Chem. Phys., 2006, 207, 183–192.
31 A. M. Funhoff, C. F. van Nostrum, A. P. C. A. Janssen, 47 Y. Chujo, Y. Yoshifuji, K. Sada and T. Saegusa,
M. H. A. M. Fens, D. J. A. Crommelin and W. E. Hennink,
Pharm. Res., 2004, 21, 170–176.
32 P. van de Wetering, J.-Y. Cherng, H. Talsma and
W. E. Hennink, J. Controlled Release, 1997, 49, 59–69.
33 M. A. Mintzer and E. E. Simanek, Chem. Rev., 2008, 109, 259–
302.
Macromolecules, 1989, 22, 1074–1077.
48 Y. Chujo, K. Sada and T. Saegusa, Macromolecules, 1990, 23,
2636–2641.
49 Y. Chujo, K. Sada and T. Saegusa, Polym. J., 1993, 25, 599–
608.
50 J.-J. Yuan and R.-H. Jin, Langmuir, 2005, 21, 3136–3145.
34 D. D. Dunlap, A. Maggi, M. R. Soria and L. Monaco, Nucleic 51 D. Kim and K. Park, Polymer, 2004, 45, 189–196.
Acids Res., 1997, 25, 3095–3101.
35 H. P. C. Van Kuringen, J. Lenoir, E. Adriaens, J. Bender,
52 F. Khan, R. S. Tare, R. O. C. Oreffo and M. Bradley, Angew.
Chem., Int. Ed., 2009, 48, 978–982.
B. G. De Geest and R. Hoogenboom, Macromol. Biosci., 53 Fino, et al., US Pat., US 4476229, 1984.
2012, 12, 1114–1123. 54 A. Koschella, M. Hartlieb and T. Heinze, Carbohydr. Polym.,
36 L. Tauhardt, K. Kempe, K. Knop, E. Altuntaxs, M. Jager, 2011, 86, 154–161.
S. Schubert, D. Fischer and U. S. Schubert, Macromol. 55 A. N. Zelikin, E. S. Trukhanova, D. Putnam, V. A. Izumrudov
¨
Chem. Phys., 2011, 212, 1918–1924.
and A. A. Litmanovich, J. Am. Chem. Soc., 2003, 125, 13693–
37 A. Kichler, J. Gene Med., 2004, 6, 3–10.
13699.
38 G.-H. Hsiue, H.-Z. Chiang, C.-H. Wang and T.-M. Juang, 56 R. Hoogenboom, M. W. M. Fijten, H. M. L. Thijs, B. M. van
Bioconjugate Chem., 2006, 17, 781–786.
39 R. Arote, T.-H. Kim, Y.-K. Kim, S.-K. Hwang, H.-L. Jiang,
Lankvelt and U. S. Schubert, Des. Monomers Polym., 2005,
8, 659–671.
H.-H. Song, J.-W. Nah, M.-H. Cho and C.-S. Cho, 57 F. Wiesbrock, R. Hoogenboom, M. A. M. Leenen,
Biomaterials, 2007, 28, 735–744.
40 X. Shuai, T. Merdan, F. Unger, M. Wittmar and T. Kissel,
Macromolecules, 2003, 36, 5751–5759.
M. A. R. Meier and U. S. Schubert, Macromolecules, 2005,
38, 5025–5034.
´
´
58 P. Claudy, J. M. Letoffe, Y. Camberlain and J. P. Pascault,
¨
41 M. Jager, S. Schubert, S. Ochrimenko, D. Fischer and
Polym. Bull., 1983, 9, 208–215.
U. S. Schubert, Chem. Soc. Rev., 2012, 41, 4755–4767.
42 B. Guillerm, S. Monge, V. Lapinte and J.-J. Robin, Macromol.
Rapid Commun., 2012, 33, 6000–6016.
59 C. Weber, A. Krieg, R. M. Paulus, H. M. L. Lambermont-
Thijs, C. R. Becer, R. Hoogenboom and U. S. Schubert,
Macromol. Symp., 2011, 308, 17–24.
43 K. Kempe, R. Hoogenboom, M. Jaeger and U. S. Schubert, 60 P. G. M. Wuts and T. W. Greene, Greene's Protective Groups in
Macromolecules, 2011, 44, 6424–6432. Organic Synthesis, Wiley-VCH, Weinheim, 4th edn, 2006.
44 K. Kempe, M. Lobert, R. Hoogenboom and U. S. Schubert, 61 M. E. Byrne, K. Park and N. A. Peppas, Adv. Drug Delivery Rev.,
J. Polym. Sci., Part A: Polym. Chem., 2009, 47, 3829–3838. 2002, 54, 149–161.
4704 | Soft Matter, 2013, 9, 4693–4704
This journal is ª The Royal Society of Chemistry 2013