Microarray glass slides coated with block copolymer brushes obtained by reversible addition chain-transfer polymerization

Article abstract of DOI:10.1021/ac0521091

The reversible addition-fragmentation chain-transfer polymerization was used to prepare microarray slides grafted with polymer brushes for DNA-based applications. Block copolymer brushes of N,N-dimethylacrylamide (DMA) and glycidyl methacrylate (GMA), poly(DMA-b-GMA) were prepared by extending living poly(dimethylacrylamide) chains. The functional surface was used as a substrate for oligonucleotide hybridization experiments. The results were compared to those provided by glass slides coated by a self-assembled monolayer made of (3-glycidyloxypropyl)trimethoxysilane. Surfaces coated with block polymer brushes bearing oxirane groups are more efficient as substrates for oligonucleotide hybridization than surfaces coated with nonpolymeric self-assembled monolayers containing the same functional group. The high probe grafting density and hybridization efficiency achieved with this polymeric coating reveal the importance of the block architecture to ensure good accessibility of the immobilized probe. The new surface was characterized by static angle measurements and diffuse reflectance FT-IR spectroscopy on a silica model system.

Full text of DOI:10.1021/ac0521091

Anal. Chem. 2006, 78, 3118-3124
Microarray Glass Slides Coated with Block
Copolymer Brushes Obtained by Reversible
Addition Chain-Transfer Polymerization
Giovanna Pirri,* Marcella Chiari,* Francesco Damin, and Alessandra Meo
Istituto di Chimica del Riconoscimento Molecolare, CNR, Milano, Italy
The reversible addition-fragmentation chain-transfer po-
lymerization was used to prepare microarray slides grafted
with polymer brushes for DNA-based applications. Block
copolymer brushes of N,N-dimethylacrylamide (DMA)
and glycidyl methacrylate (GMA), poly(DMA-b-GMA) were
prepared by extending living poly(dimethylacrylamide)
chains. The functional surface was used as a substrate
for oligonucleotide hybridization experiments. The results
were compared to those provided by glass slides coated
by a self-assembled monolayer made of (3-glycidyloxypro-
pyl)trimethoxysilane. Surfaces coated with block polymer
brushes bearing oxirane groups are more efficient as
substrates for oligonucleotide hybridization than surfaces
coated with nonpolymeric self-assembled monolayers
containing the same functional group. The high probe
grafting density and hybridization efficiency achieved with
this polymeric coating reveal the importance of the block
architecture to ensure good accessibility of the immobi-
lized probe. The new surface was characterized by static
angle measurements and diffuse reflectance FT-IR spec-
troscopy on a silica model system.
Organosilane-based self-assembled monolayers (SAM) bearing
different reactive groups4 have been widely used in the field of
microarrays due to the availability of a large variety of reagents
and well-established coating procedures. However, it is experi-
mentally found that, with this type of coating, the hybridization
efficiency for an oligonucleotide target to its complementary probe
is significantly lower than that expected due to the low accessibility
-7
8
,9
of the immobilized probe, which lies flat on the surface.
With respect to the problem of probe accessibility, three-
dimensional coatings are advantageous in that they leave the
immobilized biomolecules free to interact with their counterpart
in solution. Different strategies have been proposed to realize
three-dimensional surfaces. Macroporous silicon has been recently
introduced as a substrate for highly sensitive protein chip
applications. The formation of 3D porous silicon structures was
performed by electrochemical dissolution of monocrystalline
10
silicon. In another recent article, a three-dimensional, uniformly
porous substrate featuring a regular array of discrete, ordered,
11
high aspect ratio microchannels has been described. Systems
of the type reported above, although very effective in creating 3D
surfaces, require the use of technologies that are not suitable for
large-scale application of the microarray support at low cost.
A simple way to create an aqueous three-dimensional environ-
ment on a surface is by depositing a polymeric coating of thick-
nesses ranging from a few nanometers to a few micrometers.
The development of reactive surfaces and their use as
substrates for oligonucleotides (ODNs), DNA, and protein mi-
croarrays has had a great acceleration in the past decade.1
Microarrays consist of regularly repeating arrays of DNA or
protein spots. The availability of functional surfaces, able to
efficiently bind biomolecules, is mandatory to achieve specific
discrimination in molecular recognition events that take place at
12-15
16,17
Thin films of cross-linked polyacrylamide gels,
agarose,
(
4) Maskos, U.; Southern, E. M. Nucleic Acids Res. 1992, 20 (7), 1679-1684.
(5) Guo, Z.; Guilfoyle, R. A.; Thiel, A. J.; Wang, R.; Smith, L. M. Nucleic Acids
Res. 1994, 22 (24), 5456-5465.
6) Chrisey, L. A.; Lee, G. U.; Ferrall, C. E. Nucleic Acids Res. 1996, 24 (15),
2
the surface/solution interface. The amount of biological material
(
present within each spot determines the levels of sensitivity of
microarray technology as well as its dynamic range and analysis
speed. However, the parameter that ultimately affects mixed-phase
hybridization efficiencies is not the total amount of probe im-
mobilized within each spot but its grafting density. It is important
to avoid steric crowding that can hinder hybrid formation.3
Different types of surfaces have already been explored for arraying
biomolecules, but the search for new supports giving superior
characteristics is still a challenge.
3
031-3039.
(7) Rogers, Y.-H.; Jiang-Baucom, P.; Huang, Z.-J.; Bogdanov, V.; Anderson, S.;
Boyce-Jacino, M. Anal. Biochem. 1999, 266, 23-30.
(
(
8) Steel, A. B.; Herne, T. M.; Tarlov, M. J. Anal. Chem. 1998, 70, 4670-4677.
9) Hong, B. J.; Oh, S. J.; Youn, T. o.; Kwon, S. H.; Park, J. W. Langmuir 2005,
21, 4257-4261.
(
(
(
(
10) Ressine, A.; Ekstrom, S.; Marko-Varga, G.; Laurell, T. Anal. Chem. 2003,
5, 6968-6974.
7
11) Benoit, V.; Steel, A.; Torres, M.; Yu, Y.-Y.; Yang, H.; Cooper, J. Anal. Chem.
2001, 73, 2412-2420.
12) Timofeev, E.; Kochetkova, S. V.; Mirzabekov, A.; Florentiev, V. L. Nucleic
Acids Res. 1996, 24 (16), 3142-3148.
13) Rehman, F. N.; Audeh, M.; Abrams, E. S.; Hammond, P. W.; Kenney, M.;
Boles, T. C. Nucleic Acids Res. 1999, 27 (2), 649-655.
*
To whom correspondence should be addressed. E-mail: giovanna.pirri@
icrm.cnr.it; marcella.chiari@icrm.cnr.it. Fax: 0039 02 28901239.
(14) Kumar, A.; Larsson, O.; Parodi, D.; Liang, Z. Nucleic Acids Res. 2000, 28
(14), e71.
(15) Kumar, A.; Liang, Z. Nucleic Acids Res. 2002, 29, e2.
(16) Afanassiev, V.; Hanemann, V.; W o¨ lfl, S. Nucleic Acids Res. 2000, 28, e66.
(1) Pirrung, M. C. Angew. Chem., Int. Ed. 2002, 41, 1276-1289.
(2) Wong, E. L. S.; Chow, E.; Gooding, J. J. Langmuir 2005, 21, 6957-6965.
(3) Southern, E.; Mir, K.; Shchepinov, M. Nat. Genet. 1999, 21, 5-9.
3118 Analytical Chemistry, Vol. 78, No. 9, May 1, 2006
10.1021/ac0521091 CCC: $33.50 © 2006 American Chemical Society
Published on Web 04/01/2006

Scheme 1. Chain-Transfer Mechanism Responsible for RAFT Polymerization
electropolymerized polypyrrole,18 and dendritic polymers19,20 have
been proposed to immobilize biological molecules with a density
in both organic and aqueous media,30-35 as a robust and industry
friendly route to produce living homopolymers block and star
12
15
2
36-38
per surface unit ranging from 10 to 10 molecules/cm depend-
polymers.
The process involves a conventional free radical
1
ing on the characteristics of the polymer.
polymerization in the presence of thiocarbonylthio compounds of
general structure SdC(Z)SR, which convert chain-propagating
radicals into a “dormant” form in equilibrium with the “active”
form (Scheme 1). This category of molecules has the role of chain-
transfer agent (CTA), being able to control the polymer growing
Macromolecular chains tethered in solution, linked to the solid
surface by one end, commonly named “brushes”, provide a system
in which the reactive groups are segregated at the end of the chain
2
1
and spaced from the substrate by a long spacer. To produce
tethered chains on the surface, two different approaches are
commonly employed: physisorption or covalent attachment of
process. In the literature, there are surprisingly few reports on
the application of RAFT techniques3
9-43
to the synthesis of polymer
22
linear chains. PEG-based polymer brushes, introduced by form-
ing a self-assembled monolayer, are widely used in many biological
brushes, probably due to the difficulty of preparing RAFT agent
anchored substrates.44 The aim of this work was to develop a user-
friendly method to initiate, from the glass surface, living polymer
segments that terminate with a dithiobenzoic group (CTA). The
CTA at the end of the chain allows the initiation of a second
polymer segment comprising a reactive monomer. The confine-
ment of reactive groups on the external portion of the polymer is
a considerable advantage as it allows the binding of low or high
molecular mass molecules with a conformation that resembles
the one they have in free solution, thus facilitating the hybridiza-
tion reaction with their biological counterpart.
2
3-26
applications.
to a substrate involves living polymerization of PEG-containing
An alternative approach to attach PEG polymers
27
acrylate monomers by a variety of mechanisms such as ATRP
or nitroxide-mediated CRP.28 However, PEG brushes usually bear
only one reactive group per chain. To increase the density of
reactive groups, the combination of surface-initiated poly-
merizations and controlled polymerization techniques has been
explored.
In this paper, we provide the first report on the successful
immobilization of an oligonucleotide on a microarray glass slide
coated with block copolymer brushes obtained by a grafting from
approach, based on a radical addition-fragmentation transfer
The proposed brush block coating was compared with a SAM
containing a 3-glycidyloxypropyl residue, the group that constitutes
the functional block of the polymer. Coated glass slides were
characterized by tensiometry and by diffuse reflection FT-IR
(RAFT) polymerization. RAFT is a controlled/living polymerization
(
DRIFT) spectroscopy.45 The analysis by DRIFTS, carried out on
process used to synthesize well-defined polymers with low
29
d
polydispersity index (I ). In this approach, polymer chains grow
(
(
(
(
30) Ladavi e` re, C.; Dorr, N.; Claverie, J. P. Macromolecules 2001, 34, 5370-
directly from an initiator immobilized on the surface leading to
formation of chains covalently attached, by one end, to the
substrate. RAFT has been widely used with a variety of monomers
including acrylamide, and N- and N,N-disubstituted acrylamides,
5372.
31) Sumerline, B. S.; Donovan, M. S.; Mitsukami, Y.; Lowe, A. B.; McCormick,
C. Macromolecules 2001, 34, 6561-6564.
32) D’Agosto, F.; Hughes, R.; Charreyre, M.-T.; Picot, C.; Gilbert, R. G.
Macromolecules 2003, 36, 621-629.
33) Thomas, D. B.; Sumerline, B. S.; Lowe, A. B.; McCormick, C. L. Macromol-
ecules 2003, 36, 1436-1439.
(
17) Wang, H.; Li, J.; Liu, H.; Liu, Q.; Mei, Q.; Wang, Y.; Zhu, J.; He, N.; Lu, Z.
Nucleic Acids Res. 2002, 30 (12), e61.
(34) Vosloo, J. J.; Tonge, M. P.; Fellows, C. M.; D’agosto, F.; Sanderson, R. D.;
Gilbert, R. G. Macromolecules 2004, 37, 2371-2382.
(35) Convertine, A. J.; Ayres, N.; Scales, C. W.; Lowe, A. B.; McCormick, C.
Biomacromolecules 2004, 5, 1177-1180.
(
18) Peng, H.; Soeller, C.; Vigar, N.; Kilmartin, P. A.; Cannell, M. B.; Bowmaker,
G. A.; Cooney, R. P.; Travas-Sejdic, J. Biosens. Bioelectronics 2005, 20 (9),
1
821-1828.
(
19) Le Berre, V.; Tre’visiol, E.; Dagkessamanskaia, A.; Sokol, S.; Caminade, A.-
(36) Le, T. P.; Moad, G.; Rizzardo, E.; Thang, S. H. PCT Int. Appl. WO 9801478,
M.; Mafujoral, J.-P.; Meunier, B.; Francois, J. Nucleic Acids Res. 2003, 31
1998.
(
16), e88.
(37) Kanagasabapathy, S.; Sudalai, A.; Benicewicz, B. C. Macromol. Rapid
Commun. 2001, 22, 1076-1080.
(
20) Park, J. W. Langmuir 2005, 21, 4257-4261.
(
21) Senaratne, W.; Andruzzi, L.; Ober, C. K. Biomacromolecules 2005, 6 (5),
(38) Li, C.; Benicewicz, B. C. J. Polym. Sci.. Part A: Polym. Chem. 2005, 43,
1058-1063.
2
427-2448.
(
(
22) Zhao, B.; Brittain, W. J. Prog. Polym. Sci. 2000, 25, 677-710.
23) Harris, J. M. Poly(ethylene glycol) chemistry: Biotechnical and Biomedical
Applications; Plenum Press: New York, 1992.
(39) Lowe, A. B.; Sumerline, B. S.; Donovan, M. S.; McCormick, C. L. J. Am.
Chem. Soc. 2002, 125, 11562-11563.
(40) Shan, J.; Nuopponen, M.; Jiang, H.; Kauppinen, E.; Tenhu, T. Macromolecules
(
(
24) Prime, K. L.; Whitesides, G. M. J. Am. Chem. Soc. 1993, 115, 10714-10721.
25) Mrksich, M.; Sigal, G. B.; Whitesides, G. M. Langmuir 1995, 11, 4383-
2003, 36, 4526-4533.
(41) Matsumoto, K.; Tuji, R.; Yonemushi, Y.; Yoshida, T. J. Nanopart. Res. 2004,
6, 649-659.
(42) Baum, M.; Brittain, W. J. Macromolecules 2002, 35, 610-615.
(43) Tsujii, Y.; Ejaz, M.; Sato, K.; Goto A.; Fukuda, T. Macromolecules 2001,
34, 8872-8878.
(44) Li, C.; Benicewicz, B. C. Macromolecules 2005, 38 (14), 5929-5936.
(45) Heise, H. M. Spectroscopic methods of analysis-diffuse reflectance spec-
troscopy. In Encyclopedia of Pharmaceutical Technology, 2nd ed.; Swarbrick,
J., Boylan, J. C., Eds.; Marcel Dekker: New York, 2002; pp 2499-2510.
4
385.
26) Sofia, S. J.; Premnath, V.; Merrill, E. W. Macromolecules 1998, 31, 5059-
070.
(
5
(
(
27) Ma, H.; Hyun, J.; Stiller, P.; Chilkoti, A. Adv. Mater. 2004, 16, 338-341.
28) Andruzzi, L.; Senaratne, W.; Hexemer, A.; Sheets, E. D.; Ilic, B.; Kramer, E.
J.; Baird, B.; Ober, C. K. Langmuir 2005, 21, 2495-2504.
(29) Mayadunne, R. T. A.; Rizzardo, E.; Chiefari, J.; Pristina, J.; Moad, G.; Postma,
A.; Tang, S. H. Macromolecules 2000, 33, 243-245.
Analytical Chemistry, Vol. 78, No. 9, May 1, 2006 3119

a silica model system, provided information on the film composi-
tion whereas tensiometry allowed us to monitor contact angle
variations resulting from the growth of different polymer chains.
In addition, a functional test based on an ODN hybridization
experiment confirmed that the proposed method leads to an
dipped in this solution, under nitrogen atmosphere. The reactor
containing the slides was sealed and the polymerization carried
out at 60 °C in an oxygen-free environment. After 48 h, the slides
were cooled and washed with DMF for 1 h at room temperature,
with THF for 30 min, and with acetone for 30 min. They were
then dried in an oven at 60 °C for 10 min.
13
effective functionalization of the surface with at least 0.3 × 10
accessible ODN molecules immobilized per square centimeter.
Poly(DMA-b-GMA) Graft Polymerization. The grafted poly-
(DMA)-SC(S)Ph slides were placed in a reactor filled with a
EXPERIMENTAL SECTION
solution of 0.7 M GMA and 0.0065 M AIBN in dry and degassed
DMF. The reactor was heated at 60 °C, and the polymerization
was carried out for 16 h. Cooled slides were washed and dried as
reported above.
Materials. Glycidyl methacrylate (GMA), N,N-dimethylacry-
lamide (DMA), (3-mercaptopropyl)trimethoxysilane (γ-MPS), (3-
glycidyloxypropyl)trimethoxysilane (γ-GPS), benzoic acid, phos-
phorus pentasulfide, cyclohexane and petroleum benzine were
supplied from Aldrich (Stanheim, Germany). GMA and DMA were
purified on a neutral alumina column in order to remove the
stabilizer. Sodium dodecyl sulfate (SDS), ammonium sulfate, tris-
Silica Bead Functionalization. Organosilanization with (3-
Mercaptopropyl)trimethoxysilane. Silica beads, without any
pretreatment, were incubated in a solution of 10% v/v (3-
mercaptopropyl)trimethoxysilane in dry toluene for 12 h at room
temperature under nitrogen and gentle stirring. Silanized silica
particles were recovered, extensively washed with toluene, cy-
clohexane, and petroleum benzine, and then dried as described
above. Polymer growth on silica was carried out as reported above.
Characterization of Functionalized Glass Slides. Tensi-
ometry Measurements. Static contact angle measurements were
carried out on microscope slides by an OCA20 instrument, by a
sessile method according to the Young-Laplace equation.
Characterization of Functionalized Silica Beads. DRIFT
Measurements. Diffuse reflectance measurements were made
by dispersing coated silica beads in ground and dry KBr at a 10%
w/w concentration. The samples were inserted in a 10-mm sample
cup without applying any pressure. The surface was leveled off
by a sharp-edged spatula. A total of 32 scans was measured in
(hydroxymethyl)aminomethane (TRIS), tetrahydrofuran (THF),
ethanolamine, bovine serum bovine (BSA), and 20× SSC (3.0 M
sodium chloride, 0.3 M sodium citrate) were supplied from Sigma
(St. Louis, MO). Sodium hydroxide and chloridric acid were
supplied from Serva (Heidelberg, Germany). N,N-Dimethylfor-
mamide (DMF), toluene, ethanol, potassium bromide (spectro-
scopic grade), and R,R′-azoisobutyronitrile (AIBN) were from
Merck (Darmstadt, Germany). Silica (Davisil 663XWP, 500-Å pore
2
size, 78-85 m /g surface area) was obtained by Supelco (Belle-
fonte, PA).
4
6
Synthesis of Cyanoisopropyl Dithiobenzoate. Benzoic acid
(600 mg, 5.13 mmol) was dissolved in dry and degassed toluene.
4
P S10 (550 mg, 1.2 mmol) and AIBN (2.5 g, 15 mmol) were added
to this solution, under stirring and nitrogen. The reaction flask
was warmed at 110 °C for 1 h. After cooling, the solution was
poured into cyclohexane and then filtered. A crude sample,
containing the cyanoisopropyl dithiobenzoate (R
was recovered and purified by flash chromatography first on silica
-1
the range from 4000 to 400 cm for each sample, at a resolution
of 2 cm-1. The spectra, in reflectance format, were converted in
f 3
0.8 in CHCl ),
Kubelka-Munch units.
Functional Test of Microscope Glass Slides. Oligonucle-
otide Immobilization. A 23-mer, 5′-amine-modified oligonucle-
2
(CH Cl
2
) and then on neutral alumina (petroleum benzine/CH
2
-
):
1
Cl
2
9/1) with an overall yield of 40%. H NMR (δ ppm, CDCl
3
2 2 6
otide (100 mM stock solution, 5′-NH -(CH ) -GCC CAC CTA TAA
7
1
.93 (d, 2H, o-ArH), 7.56 (dd, 1H, p-ArH), 7.4 (dd, 2H, m-ArH),
.95 (s, 6H, 2 CH ).
Microscope Slide Functionalization. Slide Pretreatment.
GGT AAA AGT GA) was dissolved in 150 mM sodium phosphate
buffer at pH 8.5. A 10 µM solution of this oligonucleotide was
printed on coated slides to form 10 × 10 spot subarrays using a
3
Glass slides were immersed in EtOH for 1 h, washed with distilled
water, dried, and dipped in 1 M NaOH solution for 30 min. After
these treatments, the slides were immersed in 1 M HCl for 30
min, washed with water to neutrality, and dried in an oven at 80
2
QArray spotter from Genetix (Hampshire, UK). Printed slides,
in an uncovered storage box, were placed in a sealed chamber,
saturated with NaCl, and incubated overnight at room tempera-
ture.
°
C.
Hybridization Density. After spotting the slides with a 10 µM
solution of 23-mer 5′-amino-modified oligonucleotide as reported
above, the residual reactive groups of the coating were blocked
by dipping the printed slides in 50 mM ethanolamine/0.1 M Tris
pH 9.0 at 50 °C for 15 min. After discarding the blocking solution,
the slides were rinsed two times with water and shaken for 15
min in 4× SSC/0.1% SDS buffer, prewarmed at 50 °C, and briefly
rinsed with water.
Organosilanization with (3-Mercaptopropyl)trimethoxysilane.
The slides were covered with a 10% v/v (3-mercaptopropyl)-
trimethoxysilane solution in dry THF and left in this solution for
1
2 h at room temperature under nitrogen. After the reaction was
completed, the slides were washed with THF and with acetone,
and then dried under nitrogen in oven at 60 °C for 20 min.
Organosilanization with (3-Glycidyloxypropyl)trimethoxysilane.
The procedure is analogous to the one reported above.
Poly(DMA) Graft Polymerization. A polymerization solution was
prepared by dissolving N,N-dimethylacrylamide (1.0 M), cyanoiso-
propyl dithiobenzoate (5 mM), and AIBN (1 mM) in degassed
and dry DMF. The slides derivatized with mercapto groups were
An oligonucleotide, complementary to the one spotted on the
2
surface, at a 1.0 µM concentration (2.5 µL/cm of coverslip) was
dissolved in the hybridization buffer (5× SSC, 0.1% SDS and 0.2%
BSA) and immediately applied to microarrays. The complementary
oligonucleotide has the sequence 5′-TCA CTT TTA CCT TAT AGG
TGG GC-3′ and is labeled with Cy3 in 5′ position. The hybridized
duplex has a melting temperature of 64.5 °C. This test allows us
(46) Dur e´ ault, A.; Gnanou, Y.; Taton, D.; Destarac, M.; Leising, F. Angew. Chem.,
Int. Ed. 2003, 42, 2869-2872.
3120 Analytical Chemistry, Vol. 78, No. 9, May 1, 2006

Scheme 2. Schematic Representation of the Hypothetical Mechanism Responsible for Block Brush
a
Formation
a
•
Surface thiols, in the presence of a radical initiator (AIBN), form S radicals that (pathway a) transesterify with cyanoisopropyl
dithiobenzoate in solution or (pathway b) originate radical chains.
to evaluate the stability of the surface at high temperature, under
conditions that are challenging for the coating. The slides, placed
in a hybridization chamber, were transferred to a humidified
incubator at a temperature of 65 °C for 4 h.
After hybridization, the slides were first washed with 4× SSC
at room temperature to remove the coverslip and then with 2×
SSC/0.1% SDS at hybridization temperature for 5 min. This
operation was repeated two times and was followed by two
washing steps with 0.2× SSC and 0.1× SSC for 1 min at room
temperature. The slides were scanned with a Scan Array Express
fluorescence scanner from Packard Bioscience at 22% laser power
and 64% PMT.
reaction discourages their use in large-scale applications. In this
work, we demonstrate that polymer brushes can be obtained by
a RAFT process also with thiol molecules immobilized on the
surface. A schematic representation of the hypothetical mecha-
nism responsible for block brushes formation is reported in
Scheme 2. Surface thiols, in the presence of a radical initiator
•
(AIBN) form S radicals that either transesterify with cyanoiso-
propyl dithiobenzoate in solution (pathway a) or originate radical
chains (pathway b). Both pathways lead to formation of polymer
chains initiated from the surface and terminated by a dithioben-
zoate group according to the equilibrium reaction reported in
Scheme 1. A significant portion of polyDMA chains are “living”
and thus able to reinitiate polymerization to grow a second block
of polyGMA. The growth of the second block has been confirmed
by tensiometry, DRIFT spectroscopy, and functional tests.
Polymer Characteristics. Polydispersity and molecular weights
of the polymers recovered from the solution are listed in Table 1.
The row data of spot fluorescence intensity were converted to
2
molecules/cm by using a standard curve of fluorescence made
from a serial dilution of known amounts of fluorescent oligonucle-
otide (in the 0.025-15 µM concentration range). The data reported
are replicates of 400 spots.
It’s important to note that the polydispersity index (I
in solution is very close to unity, as reported for a living
polymerization process. The polydispersity index, I , is the ratio
of the weight-average molecular weight (M ) to the number-
average molecular weight (M ). It indicates the distribution of
d
) of polyDMA
RESULTS AND DISCUSSION
An innovative procedure was devised to generate living chains
d
on the surface of a glass slide. In this innovative process, the
surface was, initially, derivatized with a thiol-bearing organosilane
w
n
(γ-MPS) and then dipped in a solution of monomer (dimethyl-
individual molecular weights in a batch of polymers. According
acrylamide) and initiator (AIBN) in the presence of a chain-transfer
agent (cyanoisopropyl dithiobenzoate). Grafting of RAFT living
chains commonly requires silanizing the surface with asymmetric
organosilane RAFT agents or radical initiators. These types of
molecules are difficult to synthesize, and the low yield of their
47
to Fukuda et al., we assume that molecular weights and
polydispersity index of the free polymer are similar to those
(47) Tsujii, Y.; Ejaz, M.; Sato, K.; Goto, A.; Fukuda, T. Macromolecules 2001,
34, 8872-8878.
Analytical Chemistry, Vol. 78, No. 9, May 1, 2006 3121

Table 1. Molecular Weights and Polydispersitya of
Solution Polymers
Mn(expt) “free”
Mw(expt) “free”
Id of “free”
polymer
polymer (g/mol) polymer (g/mol)
polymer
pDMA (brush)
pGMA (brush)
3620
15620
4270
40900
1.18
2.61
a
Polymer molecular weights were determined by gel permeation
chromatography (RI detector), by using HR 1, 2, 3 water and 4 columns
in THF at 40 °C, relatively at polystyrene standards ranging from 1 to
1
860 kDa (Id ∼1).
Table 2. Values of Static Contact Anglesa for Different
Surfaces Determined in Water
Figure 1. DRIFT spectra: overlapping of grafted polyDMA spectra,
grown in the presence (s) and in absence (- - -) of CTA. The content
of DMA is higher in the sample where the grafted polymer growth
was mediated by the CTA molecules. The silica content, expressed
by the intensity of the couple of peaks at 1877-1994 cm , is
constant in all spectra.
solvent/
[CTA]/
[AIBN]
static CA
b
entry
group
T (°C)
(deg)/σ (N)
1
2
3
4
5
6
glass
13.0/3.4 (17)
56.7/1.8 (15)
66.4/1.4 (17)
56.0/1.8 (17)
72.0/2.8 (10)
53.0/1.0 (12)
-
1.
γ-GPS
THF/RT
THF/RT
DMF/60
DMF/60
DMF/60
thiol (MPS)
g-pDMA
g-pDMASC(S)Ph
g-p(DMA-b-GMA)
5/1
a
SCA measurements were done by an OCA20 instrument, by a
b
sessile method according to the Young-Laplace equation. N indicates
number of measurements and σ standard deviation on these.
of the grafted polymer. In the second step, the polyGMA growth
proceeds without CTA molecules in solution; therefore, the only
dithiobenzoate molecules present are those end grafted to poly-
DMA chains. Their concentration is too low to control the
polydispersity, and this explains why a value higher than 2, which
is typical for free radical polymerization, was obtained for
polyGMA chains.
Figure 2. DRIFT spectra of polyDMA grown in the presence of CTA
(- - -) and next grown with polyGMA (s).
Surface Characterization by Physical Methods. Tensio-
metry. After each step of the derivatization process, a significant
variation of the static contact angle value was observed, as
reported in Table 2. The static contact angle in water of a surface
coated with polyDMA grafted chains and not terminated by the
CTA was 56° whereas that of a surface coated with polyDMA and
terminated by a dithiobenzoate moiety was higher, up to 72° due
to the hydrophobicity of this group. This is consistent with what
The surface of silica was functionalized with a 10% v/v solution of
mercaptosilane in THF. The growth of polyDMA in the presence
and in the absence of cyanoisopropyl dithiobenzoate was moni-
tored by DRIFT (Figure 1). In absence of diothiobenzoic ester,
the mercapto groups immobilized on the surface produced a
significant termination effect on the growth of solution chains
(- - -line). Instead, when the polymerization was carried out in the
presence of CTA (s line), the termination effect was significantly
reduced, providing a more intense signal for the amide carbonyl
stretching and for the C-H backbone stretching.
4
8
reported in the literature and is an indication of the presence of
CTA at the end of grafted chains. The growth of a second block
of polyGMA on polyDMA living chains was carried out by adding
AIBN to a GMA solution in the absence of CTA. Also, in this case,
the growth of the second block was monitored by contact angle
variations. The contact angle found on slides in which the
polyDMA block was reinitiated with a GMA monomer was 53°, a
value slightly lower than that of the surface coated with a CTA-
terminated polyDMA homopolymer. A reduction of the contact
angle (from 72° to 53°) was expected due to the hydrophilic
character of GMA and confirms the growth of a block of
The growth of the second polyGMA block was also monitored
by DRIFTS. In Figure 2, the spectra of the silica coated with
polyDMA and with poly(DMA-b-GMA) (- - - line) are shown. The
signal corresponding to the CdO stretching of methacrylate group
-
1
(
1731.7 cm ) is clearly visible in the spectrum of the poly(DMA-
b-GMA)-coated silica. This is evidence of the growth of the second
layer. To assess the molar ratio of the two monomers on the
surface, two calibration curves were built for the two monomers,
by depositing known amounts of polyGMA and polyDMA on
known amounts of silica. As shown in Figure 3, the ratio between
the peak area of CdO (ester and amide) and the characteristic
4
9
polyGMA.
DRIFT Spectroscopy. Silica beads subjected to the coating
procedure reported above were analyzed by DRIFT spectroscopy.
2
signal of SiO (couple of peaks) increased linearly with the amount
of polymer deposited. The coated silica was stored in a LiBr
humidity chamber for 24 h, to control the water content. The area
of water signal, at 1630 cm was added to that of carbonyl,
(
(
48) Baum, M.; Brittain, W. J. Macromolecules 2002, 35, 610-615.
49) Draper, J.; Luzinov, I.; Minko, S.; Tokarev, I.; Stamm, M. Langmuir 2004,
-
1
2
0, 4064-4075.
3122 Analytical Chemistry, Vol. 78, No. 9, May 1, 2006

Figure 4. Cy3-oligonucleotide-NH2 deposited at increasing con-
centration: (A) fluorescence measured after the spotting; (B) fluo-
rescence remained on slide after washing (covalent linking).
Figure 3. Calibration curves for (A) polyGMA and (B) polyDMA.
Known amounts of the two polymers dissolved in chloroform were
deposited on known amounts of silica. The ratio between CdO
-
1
stretching area peak (at 1631.48 cm for amide on polyDMA and
-
1
1
731.7 cm for ester on polyGMA) and SiO2 network signal (area
was spotted at this concentration in the subsequent hybridization
experiments.
-
1
peak) at 1840-1994 cm is plotted versus the amount of polymer
deposited on silica beads.
The second functional test consists of a hybridization experi-
ment and requires two steps: an amino-modified oligonucleotide
was coupled overnight to the polymer brush, in a phosphate buffer
at pH 8.5 at room temperature in a humid chamber (70%
humidity). In the second step, a specific hybridization between
the oligonucleotide arrayed on the surface and its complementary
strand was carried out, by depositing on the glass slide a solution
of the Cy3-oligonucleotide. After removing the excess of oligo-
nucleotide solution, the slide was washed and scanned by a
fluorescence detector. This test provides information on ODN
density and accessibility. The amount of oligonucleotide bound
to the surface upon hybridization was extrapolated from a
fluorescence calibration curve built by depositing 0.2 nL of a Cy3
amino-modified oligonucleotide ranging in concentration from
keeping the humidity content constant. This explains why at zero
concentration of polymer deposited, the ratio of the peak area is
different from zero. Each measurement of the calibration curve
was carried out on two replicates, and each sample was analyzed
by DRIFTS at three different measurement angles (0°, 60°, 120°).
2
The R values for polyDMA and polyGMA calibration curves were
0
.998 and 0.930, respectively.
By using these two calibration curves, the amount of polymer
2
grafted to silica was calculated. Values of 0.923 mg/m for
2
polyDMA and 0.037 mg/m for polyGMA, were determined.
n
Assuming that the M values of the polymers in solution are
representative of those grafted to the surface (47), a value of 0.24
0
.025 to 15 µM.
2
pmol/cm immobilized polyGMA was calculated. This value that
Both functional tests were carried out either on a poly(DMA-
1
1
2
corresponds to 1.4 × 10 molecules/cm and provides an
indication on the number of polyDMA chains terminated by a
dithiobenzoate group.
b-GMA) coating or on a γ-GPS silanized surface. The different
fluorescence obtained in identical experiments carried on the two
surfaces clearly demonstrates that the density of accessible probes
strictly depends on the characteristics of the coating. In a self-
assembled monolayer, there is only one reactive group per silane
molecule whereas in the brush block polymer coating, each
segment of the GMA homopolymer contains several reactive
groups (∼100).
Surface Characterization by Functional Test. In a first test,
increasing concentrations of a fluorescent oligonucleotide were
arrayed on the surface and reacted with substrate. The covalent
attachment of amino-modified oligonucleotide is due to the
nucleophile addition of amine to the oxirane functionality50 by the
primary amino group of the oligonucleotide. This type of reaction
is widely employed in the production of protein and DNA arrays
and a number of commercial slides bearing oxirane groups are
available. The fluorescence of the spots, 16 replicates for each
concentration, after washing the unreacted excess of oligonucle-
otide was used to evaluate the binding sites saturation concentra-
tion. In Figure 4, the curves representing the fluorescence of
the Cy3-oligonucleotide amino-modified recorded immediately
after the spotting (curve A) and after washing the excess of
unreacted molecules (curve B) are shown. For each concentration,
above 1 µM, the fluorescence after washing is lower than that
detected after spotting the oligonucleotide. In curve B, the
fluorescence tends to a plateau when the oligonucleotide concen-
tration is higher than 10 µM. Above this value, the binding sites
are saturated; therefore, the amino-modified oligonucleotide probe
In Figure 5, a typical image of an ODN hybridization experi-
ment carried out on the two surfaces is shown. The polymeric
spacer between the long segment of polyGMA and the glass
dramatically changes the binding capacity of the functional coating
and the accessibility of the immobilized probes leading to a higher
spot fluorescence intensity.
The number of oligonucleotide molecules linked to the surface
and accessible for the hybridization, extrapolated from the
calibration curve in the linear range between 0.025 and 5.0 µM
2
13
2
(
R 0.989), corresponds to 0.3 × 10 molecules/cm . This value
is based on the fluorescence of 400 spots whose average intensity
was 15 400 arbitrary units of fluorescence. By combining the
information provided by DRIFT on the number of polyGMA
chains, 1.4 × 10 /cm , with the oligonucleotide density given by
the functional test, one can easily assume that, on average, each
chain bears ∼20 probes, which corresponds to about one oligo-
nucleotide every five monomer residues.
11
2
(
50) Tanaka, Y. In Synthesis and Characteristics of Epoxides, Epoxy Resins:
Chemistry and Technology, May, C. A., Ed.; Marcel Dekker: New York, 1988.
Analytical Chemistry, Vol. 78, No. 9, May 1, 2006 3123

Figure 5. Oligonucleotide hybridization on GPS silanized slides (1) and poly(DMA-b-GMA) slides (2). The higher fluorescence obtained in (2)
demonstrates that the density of accessible probes depends on the characteristics of the coating.
CONCLUSIONS
w/v γ-GPS and a block polymer obtained by silanization shows
that the performance provided by the polymeric coating is superior
than that of the organosilanized surface.
In this work, we provide an efficient sequential synthetic
procedure to introduce block polymers on a glass surface by using
RAFT process. A simple method, that involves silanization of glass
by the commercial product γ-MPS, followed by the sequential
introduction of two different polymeric layer was proposed. The
functional surface obtained was characterized by tensiometry and
DRIFTS. Both techniques demonstrated the presence of the block
copolymer of the desired composition whereas a functional test
based on the hybridization of complementary strands of DNA
allowed us to quantify the density of immobilized and hybridized
molecules. A comparison between a surface silanized with 10%
ACKNOWLEDGMENT
Published with the support of the European Commission, Sixth
Framework program, Information Society Technologies. NANO-
SPAD (016610).
Received for review December 1, 2005. Accepted February
27, 2006.
AC0521091
3124 Analytical Chemistry, Vol. 78, No. 9, May 1, 2006