Supported bilayer lipid membrane arrays on photopatterned self-assembled monolayers

Article abstract of DOI:10.1002/chem.200700521

This work demonstrates the use of photocleavable cholesterol derivatives to create supported bilayer lipid membrane arrays on silica. The photocleavable cholesteryl tether is attached to the surface by using the reaction of an amine-functionalized self-assembled monolayer (SAM) and the N- hydroxysuccinimide-based reagent 9. The resultant SAM contains an orthonitrobenzyl residue that can be cleaved by photolysis by using soft (365 nm) UV light regenerating the original amine surface, and which can be patterned using a mask. The photoreaction yield was ≈75% which was significantly higher than previously found for related ortho-nitrobenzyl photochemistry on gold substrates. The SAMs were characterized by means of contact angle measurements, ellipsometry and X-ray photoelectron spectroscopy. Patterned surfaces were characterized with SEM and AFM. After immersing the patterned surface into a solution containing small unilamellar vesicles of egg phosphatidylcholine (PC), supported lipid membranes were formed comprised of lipid bilayer over the amine functionalized "hydrophilic" regions and lipid monolayer over the cholesteryl "hydrophobic" regions. This was confirmed by fluorescence microscopy and AFM. FRAP studies yielded a lateral diffusion coefficient for the probe molecule of 0.14±0.05 μm ²s^-1 in the bilayer regions and ≈0.01 μm ²s^-1 in the monolayer regions. This order of magnitude difference in diffusion coefficients effectively serves to isolate the bilayer regions from one another, thus creating a bilayer array.

Full text of DOI:10.1002/chem.200700521

FULL PAPER
DOI: 10.1002/chem.200700521
Supported Bilayer Lipid Membrane Arrays on Photopatterned
Self-Assembled Monolayers
Xiaojun Han,[a] Singh N. D. Pradeep,[b] Kevin Critchley,[a] Khizar Sheikh,[a]
Richard J. Bushby,[b] and Stephen D. Evans*[a]
Abstract: This work demonstrates the
use of photocleavable cholesterol de-
rivatives to create supported bilayer
lipid membrane arrays on silica. The
photocleavable cholesteryl tether is at-
tached to the surface by using the reac-
tion of an amine-functionalized self-as-
sembled monolayer (SAM) and the N-
hydroxysuccinimide-based reagent 9.
The resultant SAM contains an ortho-
nitrobenzyl residue that can be cleaved
by photolysis by using soft (365 nm)
UV light regenerating the original
amine surface, and which can be pat-
terned using a mask. The photoreac-
tion yield was ꢀ75% which was signifi-
cantly higher than previously found for
related ortho-nitrobenzyl photochemis-
try on gold substrates. The SAMs were
characterized by means of contact
angle measurements, ellipsometry and
X-ray photoelectron spectroscopy. Pat-
terned surfaces were characterized with
SEM and AFM. After immersing the
patterned surface into a solution con-
taining small unilamellar vesicles of
egg phosphatidylcholine (PC), support-
ed lipid membranes were formed com-
prised of lipid bilayer over the amine
functionalized “hydrophilic” regions
and lipid monolayer over the cholester-
yl “hydrophobic” regions. This was
confirmed by fluorescence microscopy
and AFM. FRAP studies yielded a lat-
eral diffusion coefficient for the probe
molecule of 0.14Æ0.05 mm2 sÀ1 in the
bilayer regions and ꢀ0.01 mm2 sÀ1 in
the monolayer regions. This order of
magnitude difference in diffusion coef-
ficients effectively serves to isolate the
bilayer regions from one another, thus
creating a bilayer array.
Keywords: lipid bilayers · photo-
patterning · self-assembly · vesicles
Introduction
solid supported bilayer lipid membranes (sBLMs).[3–13] It is
hoped that such sBLMs will be prove suitable membrane
mimics to find application in membrane-based biosens-
ing[14–17] and drug discovery.[18] With such applications in
mind there has been significant interest in the development
of patterned sBLMs. A variety of approaches have been
used, to date, including: 1) the creation of macroscopic, 3D,
barriers to confine lipid bilayers to discrete areas;[19–24] 2)
the polymerization of diacetylene containing lipids to create
stable corals that could be filled with phospholipids;[25,26] 3)
the direct UV irradiation (184–257 nm) of preformed sBLM
through a photomask[27,28] where the exposed lipids were
photodecomposed while the unexposed lipids remained
fluid; 4) the patterning of self-assembled monolayer (SAM)
anchor layers on gold to create regions of “fluid” phospho-
lipid bilayer separated by regions of essentially “immobile”
adsorbed lipid monolayer. In our earlier work we used mi-
crocontact printing (mCP) of cholesterol derivatives followed
by “backfilling” with short-chain hydroxy terminated thiols
to create the patterned SAMs, and it was found that as long
as the hydroxylated region was not too large vesicles would
adsorb, rupture and spread over the surface to yield bilayers
Phospholipid bilayers are a critical component of all cell-
based biological systems, forming both the barrier between
the cytosol and the cellꢀs exterior as well as forming the sub-
compartments within cells. Furthermore, the membrane pro-
teins within the lipid bilayer serve to control cell signalling,
the transfer of metabolites and waste products, and the cel-
lular response to a range of external factors. The importance
of these bilayer systems has made the formation of “artifi-
cial” membranes an important target for many years, both
as freely suspended membranes[1,2] and more recently as
[a] Dr. X. Han, Dr. K. Critchley, Dr. K. Sheikh, Prof. S. D. Evans
School of Physics and Astronomy, University of Leeds
Leeds, LS2 9JT (UK)
Fax : (+44)113-343-3900
E-mail: s.d.evans@leeds.ac.uk
[b] Dr. S. N. D. Pradeep, Prof. R. J. Bushby
Self-Organising Molecular Systems (SOMS) Centre
University of Leeds, Leeds, LS2 9JT (UK)
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author.
Chem. Eur. J. 2007, 13, 7957 – 7964
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
7957

above the hydroxylated regions of the surface and monolay-
ers above the cholesterol regions of the surface.[29,30] More
recently we have introduced a variation on this approach by
using photocleavable fluorocarbon molecules to form hydro-
phobic SAMs on gold substrates. Upon irradiation, through
a mask, it was possible to cleave the fluorocarbon deriva-
tives to leave hydrophilic patches separated by hydrophobic
regions.[31] These systems behaved in a similar fashion to the
mCP systems mentioned above.
In this paper, we introduce some important variations on
this approach. Firstly, we introduce a simple tethering
system that is designed to work on both gold and silicon
oxide based systems. Silicon oxide surfaces are important
since they allow the use of complementary techniques, such
as fluorescence microscopy, which is not possible on gold
due to quenching. Secondly, rather than using the fluorocar-
bon derivative, as in our previous study on gold, we now in-
troduce a cholesterol derivative since these are known to
insert into the lipid layer and have proved successful in pre-
vious studies on the tethering of lipid bilayer mem-
branes.[13,29,30,32–34]
Results and Discussion
In comparison with our previous studies[31] the introduction
of N-hydroxysuccinimide chemistry, shown in Figure 1, in-
troduces an additional SAM-functionalisation step. This,
however, brings the advantage that the same reagents can
be used to form photocleavable SAMs on metallic or oxide
surfaces. The synthesis of the reagent 9 is shown in outline
in Scheme 1 (see also Experimental Section) and is straight-
forward. Because of the susceptibility of the cholesteryl nu-
cleus to attack by NO2+, it was necessary to introduce the
cholesteryl group after the introduction of the 2-nitro
group; hence the need for the amine protection and depro-
tection steps.
Figure 1. Schematic of the fabrication steps required to form a patterned
lipid bilayer on silica surface.
APMES (3-aminopropyldimethylethoxysilane) was used
to form amine-terminated SAMs (Figure 1a). Ellipsometry
of the APMES layer yielded an average thickness value of
7Æ1 Š (Table 1) over five samples which is in agreement
with that expected from molecular models 7 Š and to the
value reported by Moon et al., 7 Š.[35] The advancing and
receding water contact angles were 78Æ18 and 32Æ18 (aver-
aged over five samples), respectively. XPS of the N 1s
region showed two peaks (Figure 2) at 400.4 and 402.3 eV
(Table 2), which are associated with the amine nitrogen
(NH2) and ammonium nitrogen (NH3+), respectively.[36–38]
From the integrated intensities we estimate that ꢀ40% of
the amine groups are protonated, which is comparable to
43% protonation reported by Kowalcyk.[39]
receding contact angles. The significant hysteresis suggests
that the water penetrated into the film or that the film is
heterogeneous. From molecular models we would expect the
increase in thickness to be 20 Š for a tightly-packed layer.
However, the change in ellipsometric thickness of 12 Š was
60% of this value. Further from the integrated intensity of
the NO2 peak we estimate the reaction yield to be
ꢀ50%.[42] The ellipsometry and XPS analysis suggests that
the reaction is sterically hindered due to the larger cross-
section area of the cholesterol molecules and the nitroben-
zyl aromatic ring in comparison to the alkyl chains associat-
ed with the APMES.
Reaction of the APMES SAM with the reagent 9 gave
rise to a 12Æ2 Š increase in thickness and the appearance
of a new peak at 406.6 eV (Figure 2b) in the XPS spectrum
associated with the NO2 group of the tether.[40,41] After the
reaction the water contact angles increased from 78 to 1018
for the advancing contact angle and from 32 to 588 for the
Soft UV photopatterning: The ortho-nitrobenzyl moiety can
be cleaved using 365 nm UV light.[43–47] Figure 3 shows the
variation of the water contact angle as a function of irradia-
tion time. It is evident that the limiting values obtained after
prolonged irradiation (Table 1) were higher than those of
the “freshly” prepared APMES SAM indicating that the re-
7958
www.chemeurj.org
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 7957 – 7964

Supported Bilayer Lipid Membrane Arrays
FULL PAPER
molecules 10 have undergone a
competitive side reaction and
remain attached to the sur-
face.[32,48] As shown in Scheme 2
the probable cause is a compet-
itive photoreduction of 10 to
12. The final film thickness fol-
lowing prolonged exposure was
10Æ1 Š, that is, 3 Š greater
than that of the initial APMES
layer, and also consistent with
ꢀ25% of 12 remaining bound
to the surface. In the case of
photocleavable ortho-nitroben-
zyl SAMs on gold it was shown
that a competing photoreduc-
tion of the NO2 groups to NH2
reduced the efficacy of photo-
cleavage to ꢀ50% thus on
silica we have a significant in-
crease in yield.[44] As expected,
Scheme 1. Synthesis of the photocleavable tether 9. i) 70% HNO3, 90%; ii) Raney Ni/NH4Cl/H2O, 80%;
iii)acetic anhydride/acetic acid, 95%; iv) 70% HNO3, 75%; v) 2n HCl, 70%; vi) cholesteryl chloroformate/di-
isopropylethylamine/DMAP/toluene, 55%; vii) NaBH4/THF, 90%; viii) N,N’-disuccinimidyl carbonate/triethyl-
the
N 1s (NO2) peak (Fig-
1
amine/DMF, 85%. H and 13C NMR Spectra of compounds 3–9 are available in the Supporting Information.
ure 2c) has almost completely
Table 2. Binding energy shift observed in the N 1s region.
Table 1. Ellipsometry and contact angle data of the APMES SAM, after
reaction with reagent 9 and finally after photolysis.
Band
Chemical species
Binding energy [eV]
Surfaces
Ellipsometric thickness [Š]
Contact angle [8]
advancing receding
1A
1B
2
NH2
NH2/-NHCO2-
NH3
NO2
400.4
400.4–400.8
402.3
+
APMES
7Æ1
19Æ2
10Æ1
78Æ1
101Æ2
81Æ1
32Æ1
58Æ1
41Æ1
3
406.6
APMES + 9
APMES + 9
after 2 h photolysis
Figure 3. Contact angle variation during photolysis using soft UV light
(365 nm, 7 mWcmÀ2). a) Advancing water contact angle, b) receding
water contact angle, and c) the average of the advancing and receding
water contact angle. In cases (a) and (b) the lines are guides to the eye
whilst in case (c) the line represents a fit to the data using the Cassie
equation (see text).
Figure 2. X-ray photoelectron spectra of the N 1s region a) APMES
SAM, b) after reaction with reagent 9, c) after irradiation at 365 nm for
3600 s.
disappeared at the end of the reaction. This is also consis-
tent with the suggestion that the side reaction is the reduc-
tion of NO2 to NH2.
From the fit to the contact angle data we can estimate the
rate constants for the photocleavage reaction, k1, and the
action did not go 100%. The photocleavage reaction scheme
is shown in Schem 2. Using the Cassie equation and the lim-
iting values obtained from a pure APMES film and the uni-
rradiated film we estimate that ꢀ25% of original tether
Chem. Eur. J. 2007, 13, 7957 – 7964
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7959

S. D. Evans et al.
The patterned samples were imaged by friction mode
AFM in air, which clearly shows 30 mm diameter circles of
the amine surfaces (Figure 5a). Using an unmodified tip, the
friction force observed on hydrophilic area of amine group
was significantly higher than the friction force on the hydro-
phobic tether regions. Following incubation with SUVs (Fig-
ure 5b), the contrast due to the patterned SAM is not ob-
served because the outer leaflet of the bilayer is the same as
the lipid monolayer over the tether region. The white specks
in Figure 5b are physically adsorbed, and in some cases par-
tially ruptured, vesicles. Force volume imaging was then car-
ried out, which did not show a significant difference in the
adhesion force. However, analysis of the penetration dis-
tance (taken at the point of positive gradient in force-dis-
tance data over 90 mm2 region of surface) shows two distri-
butions (Figure 5c). The first, centered at 4.4 nm is consis-
tent with the penetration of the lipid bilayer. A typical
waveform in this region is shown as the dot plot in Figure 5d
which shows a 4 nm transition followed by a small 1 nm
compression. The second distribution was centered at
3.4 nm. A typical waveform in this region shown as the solid
line in Figure 5d shows a 2 nm transition which is consistent
with the penetration of the lipid monolayers, followed by a
2 nm compression of the underlying SAM. The area fraction
corresponding to bilayer formation was calculated (from the
areas of the Gaussian fits to the penetration distance distri-
bution data) to be 30.5% which is in the range of 8.7 to
34.8% calculated as the fractional surface area of the amine
pattern on the SAM.
Following incubation of the patterned SAMs with egg
phosphatidylcholine (PC) vesicles (containing 1 mol%
Texas Red-labeled 1,2-dihexadecanoyl-sn-glycero-3-phos-
phoethanolamine) fluorescence microscopy shows bright re-
gions corresponding to lipid bilayer separated by darker re-
gions in-between corresponding to areas of adsorbed lipid
monolayer. The very bright specks are unruptured vesicles
attached to the surface. In Figure 6a one of the bilayer re-
gions has been partially photobleached. Monitoring the in-
tensity recovery in this region (Figure 6b) we calculate the
diffusion coefficient, D, to be ꢀ0.14Æ0.05 mm2 sÀ1 and the
mobile fraction of lipid to be only around 40%. This value
of D is more than an order of magnitude lower than that ob-
served for lipid bilayers adsorbed directly onto glass sup-
ports, 4.6Æ1.2 mm2 sÀ1[20] and approximately a factor of 2
lower than that observed for bilayer formed directly onto
non-patterned APMES SAMs, (D ꢀ0.36Æ0.15 mm2 sÀ1; Sup-
porting Information, Figure S1). There are two points to
note about this slow recovery. Firstly, in a typical FRAP ex-
periment lipid diffusion is isotropic, permitting the diffusion
of bleached and unbleached species equally rapidly in all di-
rections. Here, however, the bleached region, B, consists of
a bilayer portion B1 and a lipid monolayer portion B2. Diffu-
sion to and from portion B1 will be relatively rapid from
region A, but much slower from regions B2 and C. Lipid dif-
fusion in the adsorbed lipid monolayer region is
ꢀ0.01 mm2 sÀ1 (see Supporting Information Figure S2), that
is, over an order of magnitude lower than that found in the
Scheme 2. The photocleavage process. By analogy with previous work the
competing side reaction is identified as a reduction of the ortho-NO2 to
NH2. The fact that this leads to a product which can no longer be photo-
cleaved from the surface certainly indicates that the NO2 is no longer
present.
competing side reaction, k2, to be (1.5Æ0.1)”10À3 sÀ1 and
(0.5Æ0.1)”10À3 sÀ1, respectively.[44] The rate constant for
photocleavage is significantly larger than that found for the
thiol/gold system, k1 for gold was (0.33Æ0.05)”10À3 sÀ1,
while that of the competing side reaction is similar to that
found on the gold-based system.[44]
Patterns can readily be formed by irradiation through a
mask. Figure 4 shows an SEM image of a SAM photopat-
terned by using soft UV light (365 nm). The dark circles
areas are where photolysis has occurred leaving regions of
amine-functionalized surface; the bright regions show the
tethered cholesterol moieties. The contrast is reversed from
patterned fluorocarbon SAM.[43] This is due to reversal of
the net dipole in the relevant SAMs.
Figure 4. A scanning electron microscope (SEM) image of 30 mm diame-
ter circles patterned surface. The dark regions correspond to areas where
photodeprotection has occurred.
7960
www.chemeurj.org
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 7957 – 7964

Supported Bilayer Lipid Membrane Arrays
FULL PAPER
surfaces can then be used to
supported lipid bilayer arrays.
The significant difference in dif-
fusion coefficients between
lipid bilayer and monolayer re-
gions leads each of these re-
gions to be effectively isolated
from one another. This method
for fabricating supported pat-
terned lipid bilayers is very
flexible and versatile, allowing
to any required shape or size of
lipid bilayer to be formed
simply by changing the mask.
Further it has the advantage
over the mCP method,[29,30] that
the patterned bilayers can be
formed in fewer steps. Because
it is applicable to silica sub-
strates, the method also enables
us to study the lipid bilayer by
fluorescence microscopy which
is difficult on gold substrates.
The diffusion coefficient of pat-
terned lipid bilayer was found
to be significantly slower than
that of lipid bilayer on glass.
Most of this difference is attrib-
utable to the fact that it is sup-
Figure 5. a) Friction mode AFM image of 30 mm-circle photopatterned SAM; b) tapping mode AFM image
after lipid membrane formation; c) the histogram of penetration distance over 90 mm2 region of surface (c).
Dash and dot lines are Gaussian fits and sum of the two Gaussian fits, respectively; d) typical force curve for
lipid bilayer regions (g) and for lipid monolayer–tether regions (c).
bilayer regions. The upper curve (Figure 6b) shows the
result of a finite element analysis simulation of the expected
recovery behavior based on the pattern geometry used ex-
perimentally and the diffusion coefficients determined from
lipid bilayers on “pure” amine SAMs and adsorbed lipid
monolayers on pure cholesterol tether surface. The faster re-
covery and higher mobile fraction obtained in the simula-
tion curve than those from experimental curve implies a
second factor for slow recovery, that is, the ꢀ25% remain-
ing cholesterol tethers. In a typical bilayer on glass, the lipid
in both leaflets is “bleached”, and is assumed to have rough-
ly the same diffusion constants. Here, however, only lipid in
the out leaflet (upper surface) could be regarded as free to
diffuse unrestricted. Whilst the lipid in the lower leaflet will
encounter ꢀ25% coverage of immobile cholesterol tethers,
which presents a significant restriction to lipid diffusion in
the lower leaflet.
ported over an amine-functionalized SAM.
Experimental Section
Materials: Dichloromethane 99.9% and toluene were supplied by Fisher
scientific. 3-Aminopropyldimethylethoxysilane (APMES) was purchased
from Fluorochem Limited. Diisoproylethylamine 99.5% was obtained
from Sigma-Aldrich. Silicon substrates, supplied by Rockwood Electronic
Materials, were cut into approximately 1.2 cm2 from n-tpye doped, (100)
wafers. Egg yolk phosphatidylcholine (egg PC) was purchased from
Avanti Polar Lipids. Texas Red-labeled 1,2-dihexadecanoyl-sn-glycero-3-
phosphoethanolamine triethylammonium salt (Texas Red DHPE) was
purchased from Molecular Probes. Millipore Milli-Q water with a resis-
tivity of 18.0MWcm was used throughout.
Synthesis
4-Methoxy-3-nitrobenzaldehyde (2): 4-Methoxybenzaldehyde (1) (2.6 g,
2 mmol) was slowly added to nitric acid (70%, 20 mL) at room tempera-
ture, and the mixture stirred for a further 12 h. The reaction mixture was
poured into ice water (50 mL). The resultant yellow solid was collected
by filtration and washed by cold water, dried and purified by recrystalli-
sation using ethyl acetate to afford the product (2.7 g, 90%) as pale
yellow prisms. Further the product was authenticated by comparing the
Conclusion
1
IR and H NMR values with the authentic sample.[51]
We have demonstrated that a surface can be formed by at-
taching cholesterol through photocleavable tether to an
APMES SAM on silica. Subsequent photolysis regenerates
the APMES surface. This system has a significantly better
photolysis yield than equivalent thiol-on-gold SAMs. The
functionalized surface can be patterned by exposing to
365 nm UV light through a mask. The resulting patterned
3-Amino-4-methoxybenzaldehyde (3): Raney nickel (3 mL, 1.1 mmol,
50% slurry in water) was added to a stirred solution of 3-nitro-4-methox-
ybenzaldehyde (2; 1.8 g, 1 mmol) and ammonium chloride (1 g, 2 mmol)
in distilled water (120 mL). The reaction mixture was heated under
reflux for 2 h and cooled to room temperature. To this reaction mixture,
ethyl acetate (20 mL) was added and the solution was stirred for 30 min.
The reaction mixture was filtered, the organic layer was separated and
the aqueous layer extracted with ethyl acetate (3”10 mL). The combined
Chem. Eur. J. 2007, 13, 7957 – 7964
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7961

S. D. Evans et al.
5-Acetamido-4-methoxy-2-nitrobenzaldehyde (5): 3-Acetamido-4-me-
thoxybenzaldehyde (4; 1.5 g, 0.77 mmol) was slowly added to nitric acid
(70%, 15 mL) with stirring at room temperature, and the reaction mix-
ture was stirred at room temperature for a further 12 h. The reaction
mixture was poured into ice/water (30 mL). The resultant yellow solid
was collected by filtration and washed with cold water, dried and purified
by recrystallisation with ethyl acetate to afford 5 (1.3 g, 75%) as pale
1
yellow prisms. M.p. 214–2158C; H NMR (300 MHz, CDCl3, 258C, TMS):
d=10.43 (s, 1H; CHO), 8.98 (s, 1H; ArH), 7.97 (s, 1H; NHCO), 7.63 (s,
1H; ArH), 4.23 (s, 3H; OMe), 2.32 ppm (s, 3H; COCH3); 13C NMR
(75 MHz, [D6]DMSO, 258C): d=188.9, 170.0, 152.0, 145.5, 132.8, 124.3,
119.8, 107.4, 57.3, 24.4 ppm; ESI-HRMS: m/z: calcd for C10H10N2O5+Na:
261.0482, found: 261.0472 [M+Na]+; elemental analysis calcd (%) for
C10H10N2O5: C 50.40, H 4.20, N 11.80; found: C 49.4, H 3.95, N 12.0.
2-Nitro-5-amino-4-methoxybenzaldehyde (6): 5-Acetamido-4-methoxy-2-
nitrobenzaldehyde (5; 1.1 g, 0.5 mmol) was added to hydrochloric acid
(2n, 10 mL) with stirring at room temperature, and the reaction mixture
was heated under reflux for 3 h. The reaction mixture was extracted with
ethyl acetate (3”10 mL) and the combined organic layers were dried
over anhydrous magnesium sulfate. After filtration, the solvent was re-
moved under reduced pressure to give a solid which was purified by flash
chromatography on silica gel eluting with ethyl acetate/hexane 1:1 to
afford 6 (0.74 g, 70%). M.p. 193–1948C; 1H NMR (300 MHz, CDCl3,
258C, TMS): d=10.46 (s, 1H; CHO), 7.59 (s, 1H; ArH), 7.10 (s, 1H;
ArH), 4.64 (brs, 2H; NH2) 4.01 (s, 3H; OMe), 2.32 ppm (s, 3H;
COCH3); 13C NMR (75 MHz, [D6]acetone, 258C): d=190.0, 149.0, 145.4,
140.0, 129.7, 111.8, 107.7, 57.2 ppm; ESI-HRMS (negative mode) calc. for
(M-H)À:195.0411, found: 195.0412; elemental analysis calc (%) for
C8H8N2O4: C 48.97, H 4.08, N 14.28; found: C 49.0, H 4.25, N 14.10.
Carbamate 7: Diisopropylethylamine (0.52 mL, 0.3 mmol) and a few mg
of DMAP were added to a stirred solution of 2-nitro-5-amino-4-methoxy-
benzaldehyde (6; 0.39 g, 0.2 mmol), cholesteryl chloroformate (0.99 g,
0.22 mmmol) in anhydrous toluene (40 mL). The reaction mixture was
heated under reflux overnight followed by removal of solvent under re-
duced pressure resulting in a reddish-brown residue. The residue was pu-
rified by flash chromatography on silica gel eluting with 5% ethyl ace-
tate/hexane to afford 7 (0.66 g, 55%) as a yellow solid. M.p. 231–2328C;
1H NMR (300 MHz, CDCl3, 258C, TMS): d=10.41 (s, 1H; CHO), 8.71 (s,
1H; NHCO), 7.60 (s, 1H; ArH), 7.40 (s, 1H; ArH), 5.40 (d, J=5.4 Hz,
1H; vinyl cholesteryl), 4.61 (m, 1H; O-CH- of cholesteryl), 4.05 (s, 3H;
OMe), 0.85–2.47 (m, 22H; cholesteryl), 1.03 (s, 6H; cholesteryl), 0.92 (d,
J=6.6 Hz, 3H; cholesteryl), 0.87 (d, J=7.8 Hz, 6H; cholesteryl),
0.68 ppm (s, 6H; cholesteryl); 13C NMR(75 MHz, CDCl3, 258C, TMS):
d=187.9, 152.4, 150.1, 144.4, 139.6, 133.8, 126.7, 123.5, 117.5, 106.4, 76.5,
57.1, 57.0, 56.5, 50.4, 42.7, 40.1, 39.9, 38.6, 37.3, 36.9, 36.5, 36.2, 32.3, 28.6,
28.4, 28.3, 24.6, 24.2, 23.2, 22.9, 21.4, 19.7, 19.1, 12.2 ppm; ESI-HRMS
(negative mode): m/z: calcd for C36H51N2O6: 607.3748 [MÀ], found:
607.3753; elemental analysis calcd (%) for C36H52N2O6: C 71.05, H 8.63,
N 4.60; found: C 69.9, H 8.65, N 4.45.
Figure 6. Lipid bilayer array on photopatterned surface: a) 30 mm diame-
ter regions of bilayer separation by region of lipid monolayer. The dotted
circle shows the area bleached. b) Fluorescence recovery after photo-
bleaching (FRAP). Solid squares represent actual data, whilst the solid
circles represent prediction based on modeling. From actual data, we cal-
culate the diffusion coefficient, D, to be ꢀ0.14Æ0.05 mm2 sÀ1 and the im-
mobile fraction of lipid to be only around 60%.
organic layers were dried over anhydrous magnesium sulfate. After filtra-
tion, the solvent was removed under reduced pressure to give a solid
which was purified by flash chromatography on silica gel eluting with
ethyl acetate/hexane 1:3 to afford 3 (1.2 g, 80%). M.p. 89.88C; 1H NMR
(300 MHz, CDCl3, 258C, TMS): d=9.78 (s, 1H; CHO), 7.27 (d, J=
9.6 Hz, 1H; ArH), 7.23 (s, 1H; ArH), 6.88 (d, J=9.6 Hz, 1H; ArH), 3.99
(br, 2H; NH2), 3.92 ppm (s, 3H; OMe); 13C NMR (75 MHz, CDCl3,
258C, TMS): d=191.8, 152.7, 137.3, 130.7, 123.9, 113.4, 110.0, 56.1 ppm;
ESI-HRMS: m/z: calcd for C8H9NO2: 151.0633 [M]+, found: 151.0632;
elemental analysis calcd (%) for C8H9NO2: C 63.60, H 6.00, N 9.3;
found: C 62.45, H 5.70, N 10.2.
Carbamate 8: Sodium borohydride (0.11 g, 0.3 mmol) was added to a so-
lution of the carbamate (7) (0.6 g, 0.1 mmol) in anhydrous THF (10 mL)
and the mixture stirred for 10 h. The reaction was quenched by addition
of water (10 mL). The organic layer was separated and the aqueous layer
was extracted with ethyl acetate (3”10 mL).The combined organic layers
were dried over anhydrous magnesium sulfate. After filtration, the sol-
vent removed under reduced pressure to give a solid which was purified
3-Acetamido-4-methoxybenzaldehyde (4): Acetic acid (0.8, 1.2 mmol)
was added to a stirred solution of 3-amino-4-methoxybenzaldehyde (3;
1.5 g, 1 mmol) in acetic anhydride (1.2 mL, 1.2 mmol). After 1h of stir-
ring, water (10 mL) was added and the aqueous layer extracted with
ethyl acetate (3”20 mL).The combined organic layers were washed with
brine (2”10 mL) followed by water (10 mL) and dried over anhydrous
magnesium sulfate. After filtration, the solvent was removed under re-
duced pressure to give a solid which was purified by flash chromatogra-
phy on silica gel eluting with ethyl acetate/hexane 1:4 to afford the prod-
uct 4 (1.8 g, 95%). M.p. 106–1078C; 1H NMR (300 MHz, CDCl3, 258C,
TMS): d=9.90 (s, 1H; CHO), 8.88 (s, 1H; ArH), 7.77 (s, 1H; NHCO),
7.65 (d, J=8.4 Hz, 1H; ArH), 7.00 (d, J=8.4 Hz, 1H; ArH), 3.98 (s, 3H;
OMe), 2.23 ppm (s, 3H; COMe); 13C NMR (CDCl3, 258C, TMS): d=
191.5, 168.8, 152.7, 130.4, 128.6, 125.8, 122.1, 110.4, 56.5, 25.8 ppm; ESI-
HRMS: m/z: calcd for C10H11NO3+Na: 216.0631; found: 216.0635
[M+Na]+; elemental analysis calcd (%) for C10H11NO3: C 62.20, H 5.70,
N 7.2; found: C 62.0, H 5.55, N 7.0.
by flash chromatography eluting with dichloromethane to afford
8
(0.54 g, 90%) as yellow solid. M.p. 211–2128C; 1H NMR (300 MHz,
CDCl3, 258C, TMS): d=8.48 (s, 1H; NHCO), 7.68 (s, 1H; ArH), 7.42 (s,
1H; ArH), 5.41 (d, J=5.4 Hz, 1H; vinyl cholesteryl), 4.91 (d, J=6.9 Hz,
2H; benzylic -CH2), 4.61 (m, 1H; O-CH- of cholesteryl), 3.97 (s, 3H;
OMe), 2.70 (t, J=14.0, 7.0 Hz, 1H; benzylic -OH), 0.85–2.47 (m, 22H;
cholesteryl), 1.03 (s, 6H; cholesteryl), 0.92 (d, J=6.6 Hz, 3H; cholester-
yl), 0.87 (d, J=7.8 Hz, 6H; cholesteryl), 0.68 ppm (s, 6H; cholesteryl);
13C NMR (75 MHz, CDCl3, 258C, TMS): d=152.7, 146.5, 141.4, 139.6,
134.1, 132.5, 123.4, 118.3, 107.3, 76.2, 63.4, 57.1, 57.0, 56.5, 50.4, 42.7, 40.1,
39.9, 38.6, 37.3, 36.9, 36.5, 36.2, 32.3, 28.6, 28.4, 28.3, 24.6, 24.2, 23.2, 22.9,
21.4, 19.7, 19.1, 12.2 ppm; ESI-HRMS (negative mode): m/z: calcd for
7962
www.chemeurj.org
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 7957 – 7964

Supported Bilayer Lipid Membrane Arrays
FULL PAPER
C36H53N2O6: 609.3909 [MÀ], found: 609.3915; elemental analysis calcd
(%) for C36H54N2O6: C 70.81, H 8.85, N 4.59; Found: C 70.75, H 8.95, N
4.45.
between 300 and 800 nm in steps of 15 nm. DeltaPsi2 software was used
to model and fit the acquired data assuming a simple three-layer system.
Values for the base layer (silicon dioxide) were obtained from a freshly
cleaned silicon substrate. The SAM was modelled as a transparent thin
Reagent 9: N,N’-Disuccinimidyl carbonate (0.15 g, 0.058 mmol), triethyl-
amine (0.23 mL,0.17 mmol) were added to a solution of the carbamate 8
(0.3 g, 0.05 mmol) in anhydrous DMF (5 mL), and the reaction mixture
was stirred for 5 h. Water (10 mL) was added. The organic layer was sep-
arated and the aqueous layer was extracted with ethyl acetate (3”10 mL)
and the combined organic layers were dried over anhydrous magnesium
sulfate. After filtration, the solvent removed under reduced pressure to
give a solid which was purified by flash chromatography eluting with di-
4
CÁ109
film using the Cauchy approximation, n(l)=A+BÁ10
+
and k(l)=0.
l2
l4
Where l is the wavelength in units of nanometres and A, B and C are
the Cauchy parameters dependant on optical properties of the material.
Parameter, A, was restricted to a value of 1.46 and parameters B and C
were allowed to vary between 0–1 nm2 and 0–0.2 nm4, respectively. The
ambient, air, was assumed to have n=1 and k=0. At least six ellipsomet-
ric measurements were made per sample (and on more than one sample
of each type).
chloromethane to afford
9 (318 mg, 85%) as yellow powder. M.p.
1
Scanning electron microscopy (SEM): SEM images were obtained using
a Zeiss Gemini 1500 UHV SEM attached to an Omicron Nanoprobe
(Germany) with EHT of 10 kV and a probe current of 100 pA (working
distance approx. 13 mm).
236.88C; H NMR (CDCl3, 258C, TMS): d=8.49 (s, 1H; NHCO), 7.73 (s,
1H; ArH), 7.42 (s, 1H; ArH), 5.72 (s, 2H; benzylic CH2), 5.41 (d, J=
5.4 Hz, 1H; vinyl cholesteryl), 4.65 (m, 1H; O-CH- of cholesteryl), 3.99
(s, 3H; OMe), 2.84 (s, 4H; -N-CH2-CH2-N-), 0.85–2.47 (m, 22H; choles-
teryl), 1.03 (s, 6H; cholesteryl), 0.92 (d, J=6.6 Hz, 3H; cholesteryl), 0.87
(d, J=7.8 Hz, 6H; cholesteryl), 0.68 ppm (s, 6H; cholesteryl); 13C NMR
(75 MHz, CDCl3, 258C, TMS): d=168.7, 152.6, 151.5, 147.2, 141.2, 139.7,
134.2, 124.6, 123.4, 117.6, 107.5, 76.5, 63.4, 57.1, 57.0, 56.5, 50.4, 42.7, 40.1,
39.9, 38.6, 37.3, 36.9, 36.5, 36.2, 32.3, 25.8, 28.6, 28.4, 28.3, 24.6, 24.2, 23.2,
22.9, 21.4, 19.7, 19.1, 12.2 ppm; ESI-HRMS (negative mode): m/z: calcd
for C41H56N3O10: 750.3971 [MÀ]; found: 750.3998.
Vesicle preparation: Small unilamellar vesicles (SUVs) were prepared by
tip sonication. 5 mg of egg PC containing 1 mol% TR-DHPE chloroform
solution was dried under nitrogen in a small glass vial over night. The
dried egg PC film was then hydrated in 1 mL 0.1m KCl by vortexing. The
cloudy solution was tip-sonicated at less than 58C until the solution
became clear. In order to remove any metal particulates from the tip, the
clear solution was centrifuged at a speed of 13000 rev per s for 30 min.
Finally, the vesicle solution was diluted to egg PC concentration of
Substrate cleaning: Silicon substrates were cleaned by ultrasonication in
dichloromethane for 15 min, dried under a stream of nitrogen, and rinsed
in Milli-Q water before being immersed in Piranha solution (70:30, v/v,
H2SO4/H2O2; CAUTION: Piranha solution reacts violently with organic
materials and should be treated with great care) for 5 min. The substrates
were rinsed in Milli-Q water and dried under nitrogen. The water contact
angles were typically much less than 58 for freshly cleaned substrates.
2 mgmLÀ1
.
Fluorescence microscopy and Fluorescence recovery after photobleaching
(FRAP): A Nikon E600 fluorescence microscope equipped with a Hama-
matsu (ORCA-ER) digital camera was used to image the patterned bi-
layers and to carry out FRAP measurements. Image-Pro Express data ac-
quisition software (Media Cybernetics) was used to obtain and analyze
images. The lateral diffusion coefficient (D) was calculated from D=
0.224 w2/t1/2, where w is the radius of the bleached spot and t1/2 is the half-
life of the fluorescence recovery.[49,50]
Silanization: The clean dried substrates were immersed into anhydrous
toluene solution with 1% APMES under a nitrogen atmosphere for 18 h
at room temperature. After silanization they were washed with toluene
and CH2Cl2, and dried under nitrogen. The silanized substrates were
used immediately.
Atomic force microscopy (AFM) measurement: AFM images were taken
using an Asylum Research Inc. MFP-3D, using Veeco NPS cantilevers
having nominal spring constants of 0.06 NmÀ1. Friction mode images
were taken in contact mode in air and tapping mode images were taken
in Pipes buffer, at pH 7, with a cantilever resonant frequency of 9 kHz.
Force volume images were taken in Pipes buffer, at pH 7, and with an ap-
proach rate of 400 nmsÀ1 (this value was chosen to minimizing the effect
of approach velocity of the tip to the penetration distance[52]) using a 64
by 64 unit grid and analysed using an in-house program.
APMES Derivatization: The APMES SAMs were immediately placed in
anhydrous CH2Cl2 containing a 20 mm solution of the reagent 9 (under a
nitrogen atmosphere) and 1% (by volume) diisoproylethylamine added.
The solution was maintained at room temperature, 228C, and stored in
the dark. After 6 h, the samples were removed, rinsed with CH2Cl2, ultra-
sonicated in CH2Cl2, and dried under nitrogen.
UV irradiation of the functionalized surface: A 365 nm UV lamp (Blak-
Ray model
B 100 AP) with a nominal power, at the sample, of
7 mWcmÀ2 was used to irradiate the samples. After the UV exposure,
samples were rinsed with CH2Cl2, followed by Milli-Q water, and finally
dried under nitrogen. The emission spectrum of the lamp displayed three
prominent lines, 309, 331, and 365 nm, with integrated intensities of
0.005:0.03:1, respectively. Photopatterning was achieved by irradiation of
samples through a chromium/quartz mask, for one hour. Patterned sam-
ples were rinsed with CH2Cl2, followed by Milli-Q water, and finally
dried under a stream of nitrogen.
Acknowledgement
We thank EPSRC for financial support: GR/S87195/01 and
EP/C006755/1.
Wetting measurements: Contact angles were measured using a First Ten
Angstroms (FTA4000) instrument. Advancing and receding contact
angles were obtained using the “Dynamic Contact Angle Analyzer”.
Measurements were made on at least three different points on each
sample.
[1] M. Montal, P. Mueller, Proc. Natl. Acad. Sci. USA 1972, 69, 3561–
3566.
[2] P. Mueller, D. O. Rudin, H. T. Tien, W. C. Wescott, Nature 1962, 194,
979–980.
[3] Z. Leonenko, E. Finot, D. Cramb, Biochim. Biophys. Acta 2006,
1758, 487–492.
X-ray photoelectron spectroscopy: Spectra were obtained using
Thermo VG ESCALAB 250 with a base pressure maintained below 5”
10À9 mbar during acquisition.
monochromated AlKa X-ray source
a
A
(15 kV 150 W) irradiated the samples, with a spot diameter of approxi-
mately 0.5 mm. The spectrometer was operated in Large Area XL mag-
netic lens mode, using pass energies of 150 and 20 eV for survey and de-
tailed scans, respectively. Spectra were obtained with an electron takeoff
angle of 908. High-resolution spectra were fitted using Avantage
(Thermo VG software) peak fitting algorithms.
[4] S. Lingler, I. Rubinstein, W. Knoll, A. Offenhausser, Langmuir
1997, 13, 7085–7091.
[5] R. Naumann, S. M. Schiller, F. Giess, B. Grohe, K. B. Hartman, I.
Karcher, I. Koper, J. Lubben, K. Vasilev, W. Knoll, Langmuir 2003,
19, 5435–5443.
[6] E. Sackmann, Science 1996, 271, 43–48.
[7] E. Sackmann, Mol. Biotechnol. 2000, 74, 135–136.
[8] E. Sackmann, M. Tanaka, Trends Biotechnol. 2000, 18, 58–64.
Ellipsometry: A Jobin-Yvon UVISEL spectroscopic ellipsometer was
used to measure the thickness of the SAMs. The wavelength was varied
Chem. Eur. J. 2007, 13, 7957 – 7964
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7963

S. D. Evans et al.
[9] S. M. Schiller, R. Naumann, K. Lovejoy, H. Kunz, W. Knoll, Angew.
Chem. 2003, 115, 219–222; Angew. Chem. Int. Ed. 2003, 42, 208–
211.
[34] L. M. Williams, S. D. Evans, T. M. Flynn, A. Marsh, P. F. Knowles,
R. J. Bushby, N. Boden, Supramol. Sci. 1997, 4, 513–517.
[35] J. H. Moon, J. W. Shin, S. Y. Kim, J. W. Park, Langmuir 1996, 12,
4621–4624.
[36] I. George, P. Viel, C. Bureau, J. Suski, G. Lecayon, Surf. Interface
Anal. 1996, 24, 774–780.
[37] H. L. Cabibil, V. Pham, J. Lozano, H. Celio, R. M. Winter, J. M.
White, Langmuir 2000, 16, 10471–10481.
[38] R. R. Sahoo, A. Patnaik, J. Colloid Interface Sci. 2003, 268, 43–49.
[39] D. Kowalczyk, S. Slomkowski, M. M. Chehimi, M. Delamar, Int. J.
Adhes. Adhes. 1996, 16, 227–232.
[10] M. Tanaka, E. Sackmann, Nature 2005, 437, 656–663.
[11] S. Terrettaz, M. Mayer, H. Vogel, Langmuir 2003, 19, 5567–5569.
[12] S. Terrettaz, H. Vogel, MRS Bull. 2005, 30, 207–210.
[13] L. M. Williams, S. D. Evans, T. M. Flynn, A. Marsh, P. F. Knowles,
R. J. Bushby, N. Boden, Langmuir 1997, 13, 751–757.
[14] B. A. Cornell, V. L. B. BraachMaksvytis, L. G. King, P. D. J. Osman,
B. Raguse, L. Wieczorek, R. J. Pace, Nature 1997, 387, 580–583.
[15] H. Bayley, P. S. Cremer, Nature 2001, 413, 226–230.
[16] O. Worsfold, N. H. Voelcker, T. Nishiya, Langmuir 2006, 22, 7078–
7083.
[40] J. H. Moon, J. H. Kim, K. Kim, T. H. Kang, B. Kim, C. H. Kim, J. H.
Hahn, J. W. Park, Langmuir 1997, 13, 4305–4310.
[17] M. A. Cooper, J. Mol. Recognit. 2004, 17, 286–315.
[18] J. T. Groves, Curr. Opin. Drug Discovery Dev. 2002, 5, 606–612.
[19] P. S. Cremer, T. L. Yang, J. Am. Chem. Soc. 1999, 121, 8130–8131.
[20] J. T. Groves, S. G. Boxer, Biophys. J. 1995, 69, 1972–1975.
[21] J. T. Groves, N. Ulman, S. G. Boxer, Science 1997, 275, 651–653.
[22] J. T. Groves, N. Ulman, P. S. Cremer, S. G. Boxer, Langmuir 1998,
14, 3347–3350.
[23] B. L. Jackson, J. T. Groves, Langmuir 2007, 23, 2052–2057.
[24] L. A. Kung, L. Kam, J. S. Hovis, S. G. Boxer, Langmuir 2000, 16,
6773–6776.
[25] K. Morigaki, T. Baumgart, A. Offenhausser, W. Knoll, Angew.
Chem. 2001, 113, 184–186; Angew. Chem. Int. Ed. 2001, 40, 172–
174.
[26] K. Morigaki, T. Baumgart, U. Jonas, A. Offenhausser, W. Knoll,
Langmuir 2002, 18, 4082–4089.
[41] Y. C. Liu, R. L. Mccreery, J. Am. Chem. Soc. 1995, 117, 11254–
11259.
[42] If we assume that there are x0 amine groups on the surface prior to
step 1 (Figure 1) and that following the reaction the NO2 signal, y, is
proportional to the number of reacted amines. Then the reaction
yield is given by y/x0, where x0 can be determined from the total
amine signal (after attachment), z, which is given by x0 +y because
one reacted reagent 9 introduced one NH group. Thus the yield is
given by y/
A
[43] K. Critchley, J. P. Jeyadevan, H. Fukushima, M. Ishida, T. Shimoda,
R. J. Bushby, S. D. Evans, Langmuir 2005, 21, 4554–4561.
[44] K. Critchley, L. Zhang, H. Fukushima, M. Ishida, T. Shimoda, R. J.
Bushby, S. D. Evans, J. Phys. Chem. B 2006, 110, 17167–17174.
[45] J. Nakanishi, Y. Kikuchi, T. Takarada, H. Nakayama, K. Yamaguchi,
M. Maeda, J. Am. Chem. Soc. 2004, 126, 16314–16315.
[46] M. C. Pirrung, Angew. Chem. 2002, 114, 1326–1341; Angew. Chem.
Int. Ed. 2002, 41, 1276–1289.
[27] C. K. Yee, M. L. Amweg, A. N. Parikh, J. Am. Chem. Soc. 2004, 126,
13962–13972.
[28] C. K. Yee, M. L. Amweg, A. N. Parikh, Adv. Mater. 2004, 16, 1184–
1189.
[47] E. Besson, A. M. Gue, J. Sudor, H. Korri-Youssoufi, N. Jaffrezic, J.
Tardy, Langmuir 2006, 22, 8346–8352.
[29] A. T. A. Jenkins, R. J. Bushby, N. Boden, S. D. Evans, P. F. Knowles,
Q. Y. Liu, R. E. Miles, S. D. Ogier, Langmuir 1998, 14, 4675–4678.
[30] A. T. A. Jenkins, N. Boden, R. J. Bushby, S. D. Evans, P. F. Knowles,
R. E. Miles, S. D. Ogier, H. Schonherr, G. J. Vancso, J. Am. Chem.
Soc. 1999, 121, 5274–5280.
[31] X. H. Han, K. Critchley, L. Zhang, S. N. D. Pradeep, R. J. Bushby,
S. D. Evans, Langmuir 2007, 23, 1354–1358.
[32] Y. L. Cheng, N. Boden, R. J. Bushby, S. Clarkson, S. D. Evans, P. F.
Knowles, A. Marsh, R. E. Miles, Langmuir 1998, 14, 839–844.
[33] L. J. C. Jeuken, S. D. Connell, P. J. F. Henderson, R. B. Gennis, S. D.
Evans, R. J. Bushby, J. Am. Chem. Soc. 2006, 128, 1711–1716.
[48] A. B. D. Cassie, S. Baxter, Trans. Faraday Soc. 1944, 40, 0546–0550.
[49] D. M. Soumpasis, Biophys. J. 1983, 41, 95–97.
[50] L. Q. Zhang, M. L. Longo, P. Stroeve, Langmuir 2000, 16, 5093–
5099.
[51] J. A. R. Rodrigues, R. A. Abramovitch, J.D.F de Sousa, G. C. Leiva,
J. Org. Chem., 2004, 69, 2920–2928.
[52] V. Franz, S. Loi, H. Muller, E. Bamberg, H. H. Butt, Colloids Surf.
B 2002, 23, 191–200.
Received: April 3, 2007
Published online: July 4, 2007
7964
www.chemeurj.org
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 7957 – 7964