Hydrophobic surfaces for enhanced differentiation
of embryonic stem cell-derived embryoid bodies
Bahram Valamehr*, Steven J. Jonas†, Julien Polleux†, Rong Qiao*‡, Shuling Guo§¶, Eric H. Gschweng?, Bangyan Stiles*,
Korey Kam†, Tzy-Jiun M. Luo†, Owen N. Witte*§¶?**, Xin Liu*‡, Bruce Dunn†¶**, and Hong Wu*¶**
*Department of Molecular and Medical Pharmacology, David Geffen School of Medicine,†Department of Materials Science and Engineering, Henry Samueli
School of Engineering and Applied Sciences,‡Department of Pathology and Laboratory Medicine and Embryonic Stem Cell/Transgenic Mice Shared
Resource,§Department of Microbiology, Immunology, and Molecular Genetics,¶Eli and Edythe Broad Center of Regenerative Medicine and Stem
Cell Research, and?Howard Hughes Medical Institute, University of California, Los Angeles, CA 90095
Contributed by Owen N. Witte, July 25, 2008 (sent for review October 17, 2007)
With their unique ability to differentiate into all cell types, em-
bryonic stem (ES) cells hold great therapeutic promise. To improve
the efficiency of embryoid body (EB)-mediated ES cell differenti-
ation, we studied murine EBs on the basis of their size and found
that EBs with an intermediate size (diameter 100–300 ?m) are the
most proliferative, hold the greatest differentiation potential, and
have the lowest rate of cell death. In an attempt to promote the
formation of this subpopulation, we surveyed several biocompat-
ible substrates with different surface chemical parameters and
identified a strong correlation between hydrophobicity and EB
development. Using self-assembled monolayers of various lengths
of alkanethiolates on gold substrates, we directly tested this
correlation and found that surfaces that exhibit increasing hydro-
phobicity enrich for the intermediate-size EBs. When this approach
was applied to the human ES cell system, similar phenomena were
observed. Our data demonstrate that hydrophobic surfaces serve
as a platform to deliver uniform EB populations and may signifi-
cantly improve the efficiency of ES cell differentiation.
hydrophobicity ? self-assembled monolayers ? serum-free differentiation
models for studying the mechanisms of lineage commitment and
has opened pathways for regenerative medicine (1). Various in
vitro strategies have been developed for differentiation of ES
cells into populations of specific cell types (2). Of these strate-
gies, the formation of three-dimensional cell aggregates known
as embryoid bodies (EBs) is a common and critical intermediate
to the induction of lineage-specific differentiation (3, 4). In
addition, lineage differentiation programs within the EB closely
resemble lineage commitment in vivo in the developing embryo,
further highlighting the importance of the ES cell–EB culture
Although EBs can be generated through several methodolo-
gies, the suspension culture technique allows for easy access to
the cultured EBs and can be scaled for expansion (9). In this
method, EBs are formed when ES cells are removed from feeder
contact and dispersed on low-attachment tissue culture plates,
supplemented by culture medium absent of key factors necessary
for the maintenance of undifferentiated ES cell growth. Low-
hydrogels to prevent protein adsorption and subsequent cell
attachment, facilitating the initial aggregation of ES cells that is
critical to EB formation (10). The cellular aggregates formed by
this procedure will develop simple EBs that consist of an outer
layer of endoderm cells within 2–4 days (3). At this point, two
differentiation strategies can be applied. If suspension culture is
continued, simple EBs will differentiate further to form cystic
EBs that typically contain an inner layer of columnar ectoderm-
(3). However, the most commonly used multistage differentia-
tion protocols use the second strategy in which simple EBs are
he potential of embryonic stem (ES) cells to differentiate
into all specialized cell types has made them attractive
transferred onto adherent tissue culture surfaces after day 4 of
EB development and are subsequently supplemented with key
factors necessary for lineage-specific differentiation (5, 8, 11).
One shortcoming of this suspension culture system is the pro-
duction of heterogeneous EBs, varying in size and morphology,
which may limit homogeneous differentiation and impede pro-
duction yields (12–15).
In this study, we investigated alternative surface conditions to
promote uniform EB formation and enhance the differentiation
yields of ES cells. Indeed, altering surface properties is known to
significantly affect cell growth, attachment, and differentiation
in various culture systems (16–19). For example, by using a
polyacrylamide gel in which Young’s Modulus could be tuned
based on the degree of cross-linking, Engler et al. (17) provided
compelling evidence that matrix elasticity can specify lineage
commitment toward neurons, myoblasts, and osteoblasts. Our
findings demonstrate the benefit of using hydrophobic surfaces
for monodispersed EB formation. The materials tested in our
study, such as polydimethylsiloxane (PDMS) or self-assembled-
monolayer (SAM) surfaces presenting terminal hydrophobic
moieties, are easily adapted by, and widely accessible to, the
general research community, which should enable more labo-
ratories to better pursue ES cell research.
EB Size Determines Cellular Viability, Proliferation, and Differentia-
tion Potential. EBs produced in suspension culture are known to
and morphology influence subsequent differentiation and pro-
duction yield is not clear. To understand this critical issue, we
conducted a systematic study by first manually separating day 4
murine EBs into three subpopulations based on their diameter:
small (?100 ?m), intermediate (100–300 ?m), and large (?300
?m) (Fig. 1A). We then compared their respective potentials for
cellular survival, proliferation, and differentiation of these three
Cellular survival can be determined by monitoring the per-
centage of cells undergoing apoptosis. Indicated by annexin V
(AV) and propidium iodide (PI) positive staining, apoptosis can
be categorized into three temporal stages: early (AV?/PI?),
intermediate (AV?/PI?), and late (AV?/PI?) (20). Very little
late-stage apoptosis was detected during day 4 of EB formation
for all three subgroups [supporting information (SI) Fig. S1A].
Author contributions: B.V., S.J.J., J.P., S.G., E.H.G., B.S., K.K., T.-J.M.L., O.N.W., B.D., and
S.J.J., J.P., S.G., E.H.G., B.S., O.N.W., X.L., B.D., and H.W. analyzed data; and B.V., S.J.J., J.P.,
S.G., E.H.G., O.N.W., X.L., B.D., and H.W. wrote the paper.
The authors declare no conflict of interest.
firstname.lastname@example.org, or email@example.com.
whom correspondence maybe addressed.E-mail: firstname.lastname@example.org,
This article contains supporting information online at www.pnas.org/cgi/content/full/
© 2008 by The National Academy of Sciences of the USA
September 23, 2008 ?
vol. 105 ?
no. 38 ?
This observation was confirmed independently with 7-amino-
actinomycin D staining (data not shown), indicating that during
the first stage of EB formation, the majority of the compromised
cells have only entered the initial stages of apoptosis. When all
stages of apoptosis were considered, the intermediate-size EB
than the small (25.0%) and large (35.5%) EB populations,
respectively (Fig. 1B and Fig. S1A).
To address whether EB size can also influence cellular prolifer-
ation, we evaluated proliferation rates by using flow cytometry-
mediated cell cycle analysis because percentage population in
G0/G1is inversely, and in G2/M is directly, correlated to the cellular
proliferation rate. The large-EB subgroup contains a reduced
and intermediate-size EBs, suggesting that large EBs are the least
proliferative subpopulation (Fig. S1B). Differences in cell cycle
distribution between small- and intermediate-size EB subgroups
are marginal but remain statistically significant (P ? 0.05).
In addition to cellular proliferation and viability, EB size also
influences differentiation potential. Upon differentiation, EBs
ectoderm, endoderm, and mesoderm (21). We therefore exam-
ined the expression levels of genes associated with lineage
differentiation in each subpopulation. Quantitative reverse-
transcription–polymerase chain reaction (qRT-PCR) analysis
revealed that EBs of intermediate size express the highest levels
of those genes associated with lineage differentiation (Fig. 1C
and Table S1). Small EBs display a significant attenuation of
gene expression for all three somatic lineages, whereas gene
expression levels of the large-EB subgroup fall in between (Fig.
1C). These findings thus demonstrate that intermediate-size EBs
are the most proliferative, display the lowest rate of cell death,
and may have the greatest differentiation potential.
enrich for EBs of the preferred size, we surveyed several
biocompatible coatings, such as agarose, polyethylene glycol
silica sol gel (PEG sol), polyhydroxyethylmethacrylate
(pHEMA), and polydimethylsiloxane (PDMS) (Table S2). Two
commonly used tissue culture plates—low-binding culture plates
(Nunc) and ultra-low-attachment culture (LAC; Corning)
plates—were also incorporated into the study (Fig. 2A). These
surfaces were selected based on their neutral charge and their
ability to resist cellular adhesion and spreading, which is a
principal requirement for suspension culture. The major differ-
ences among these surfaces are the terminal functional group
and hydrophobicity, which can be assessed by water contact
angle; contact angles ?90° are considered hydrophobic. By
varying these two parameters, we tested whether surface chem-
ical properties could influence EB formation.
were plated onto the abovementioned surfaces, a variety of EB
sizes and morphologies were observed (Fig. 2A). Interestingly,
PDMS—the sole hydrophobic surface in our study—promoted
the development of EBs that were more uniform in size, with a
reduction in cellular clumps and debris (Fig. 2A Lower Right).
Based on the results seen in this initial screen, we decided to
perform an in-depth study of the PDMS-coated surface.
PDMS Surface Promotes Formation of Intermediate-Size EBs. When
we compared EBs formed on PDMS-coated surfaces with EBs
derived from LAC plates—which are routinely used for EB
formation—we found that PDMS surfaces promoted the for-
mation of intermediate-size EBs by ?2-fold (Fig. 2B). To
confirm that this finding was not cell line-dependent, other
independent ES cell lines were subjected to the same test and
analyzed during the first 4 days of differentiation (Fig. S2).
During day 1 of differentiation, cell aggregates formed on LAC
plates varied in size and shape, with visible large clumps (Fig.
S2A), whereas aggregates developed on PDMS appeared to be
similar in shape and size and evenly distributed throughout the
culture well (Fig. S2B). This even distribution, combined with a
reduction in debris and larger aggregates, translated to morpho-
logically uniform EBs after 4 days of culture (Fig. S2 C–H),
suggesting that PDMS surfaces may better support EB devel-
Further study revealed an increase in total viable cells on
PDMS as compared with EBs derived from LAC plates during
25 days of continuous EB culture, with a 2-fold increase at day
Cell Death (%)
potential. (A) Day 4 EBs were manually separated into three size categories:
small, intermediate, and large. (Scale bars, 250 ?m.) (B) EBs of the intermedi-
ate size contain the lowest number of apoptotic cells. n ? 3. (C) qRT-PCR
analysis of lineage marker expression for ectoderm (Nestin and Tau), meso-
derm [Brachyury (Bry) and Flk1], and endoderm (Pdx1 and Sox17) in each
subpopulation. The level of 18S serves as an internal control. n ? 3. Data
presented as mean ? SD.**, P ? 0.01.
EB size influences cellular viability, proliferation, and differentiation
4 EBs were measured. n ? 4. (C) Enrichment for the intermediate-size EBs on
PDMS-coated surface results in increased number of viable cells. EBs were mon-
viable cells were counted after trypan blue staining and plotted over a 25-day
culture period. n ? 5. Data presented as mean ? SD;*, P ? 0.05;***, P ? 0.001.
www.pnas.org?cgi?doi?10.1073?pnas.0807235105 Valamehr et al.
8 (Fig. 2C). The increase in total viable cells also coincides with
the period when the majority of commitment toward all three
germ cell layers and the appearance of specialized cell types are
detected during EB-mediated cell differentiation (1, 2, 7, 25–27).
In addition, the percentage of cells in G0/G1cell cycle phase was
lower in EBs formed on PDMS (36.2%; P ? 0.01) than in EBs
formed on LAC (42.8%) (Fig. S3A). The decrease in G0/G1and
increase in S/G2/M phases displayed by EBs formed on PDMS
held true at days 4, 8, and 16 of EB development (Fig. S3A).
Furthermore, EBs formed on PDMS exhibited increased total
metabolic activity, consistent with increased cell viability and
proliferation (Fig. S3B).
We then used three primer sets per germ layer to monitor EB
differentiation over time. In general, these lineage-specific dif-
ferentiation markers appeared earlier and sustained longer in
EBs formed on PDMS than in their counterparts formed on
LAC plates (Fig. S3C). We next studied whether EBs formed on
PDMS readily undergo hematopoietic differentiation by assess-
ing the presence of hematopoietic progenitors by means of
quantitative FACS analysis and methylcellulose assay. Similar to
a previous report (28), we observed a sequential appearance of
Flk1- and c-Kit-positive populations in day 4 EBs, then CD34-
and CD45-positive populations in day 7 EBs. To this extent,
FACS analysis reveals that Flk1?, CD45?, and CD34?popula-
tions are significantly increased in EBs generated on PDMS, as
compared with those on LAC plates (Fig. S3D). When day 7 EBs
were dissociated and plated in methylcellulose (Stem Cell Tech-
nologies), we also detected a 4.3-fold increase in the number of
hematopoietic colonies (Fig. S3E), suggesting that EBs formed
on PDMS indeed have greater differentiation potential.
Role of Hydrophobicity in EB Formation. By analyzing the surface
properties of the materials used, we found that one potential
determining factor for uniform EB formation may be surface
hydrophobicity (Table S2). To test whether hydrophobicity plays
a critical role in controlling EB size distribution, we used SAMs
of alkanethiolates on gold-coated substrates as model surfaces
(29). We chose to vary the alkyl chain length of the SAM films,
where longer chained alkanethiol structures (R in Fig. 3A) would
lead to increases in hydrophobicity at the cell culture interface
as measured by the contact angle (? in Fig. 3A). SAMs composed
of butanethiol (C4), heptanethiol (C7), dodecanethiol (C12),
and octadecanethiol (C18) were selected to form a hydrophobic
gradient (Fig. 3A). It is important to note that the contact angles
of the C12 and C18 SAMs were similar to that of PDMS
substrates (Table S2): between 100° and 110°. Additionally, an
unmodified gold surface (Au) was incorporated as a control for
the influence of the gold foundation on EB formation and a
carboxylic acid-derivatized SAM, 16-mercaptohexadecanoic
acid (MHA), was incorporated to test the influence of a hydro-
philic terminal functional group (Fig. 3A).
Interestingly, as the surface hydrophobicity was augmented,
enrichment for the intermediate size was observed (Fig. 3B).
Moreover, the 18-carbon alkanethiolate SAM, which has a
hydrophobic character similar to that of PDMS, offers the best
surface for promoting efficient intermediate-size EB formation
(Fig. 3B). Conversely, the majority of the aggregates counted on
the more hydrophilic surfaces, including Au, MHA, and C4,
were ?100 ?m in size.
The increasing surface hydrophobicity and contact angle also
translated to a significant decrease in cell attachment (Fig. 3C,
lower images). After 4 days of culture, the percentages of cell
attachment under each culture condition were quantified and
graphed against the respective surface contact angle. As dem-
onstrated in Fig. 3C Right, hydrophobicity, or contact angle, is
negatively correlated with cell attachment (R2? 0.9514). For
example, the MHA surface, which is charged and hydrophilic
under our culture conditions, promotes cell adherence and
produces the largest quantity of small EBs. In contrast, the most
hydrophobic surface in our study, C18, impeded cell attachment
and delivered the largest quantity of intermediate-size EBs.
Overall, it appears that increasing surface hydrophobicity can
cell-surface interaction and attachment.
Formation of Uniform-Size EBs from Human ES Cell Cultures. We next
investigated whether the relationship between hydrophobicity
cells. When HSF1 [University of California, San Francisco
(UCSF)] human ES cells from the same stock were seeded on
either LAC or C18 surfaces, we found that EBs cultured on C18
appear to be more uniform in size and morphology, as compared
with LAC controls (Fig. 4A). Also compared with LAC cultures,
human EBs formed on C18 surfaces display reduced cellular
death (4.7% vs. 8.2%; P ? 0.01; Fig. 4B) and enhanced cellular
proliferation (S/G2/M, 43.2% vs. 41.0%; P ? 0.01; Fig. 4C).
Similar to what was observed in the murine EB cultures, cell
surface attachment was inhibited as hydrophobicity increased
(Fig. S4). Collectively, it appears that human EB formation and
development are also facilitated by hydrophobic surfaces, in-
cluding C18 and PDMS (Fig. 4 and data not shown).
4060 80 100
Au MHA C4C7C12
45 ± 3°
82 ± 4°
96 ± 2°
100 ± 2°
103 ± 2°
θ < 90°
θ > 90°
69 ± 5°
size EBs. (A) Surface characterization of SAM films of alkanethiolates. R,
terminal functional group; ?, water contact angle. (B) Day 4 EB size distribu-
results in an enrichment of intermediate-size EBs. n ? 4. (C) Negative corre-
lation between surface attachment and contact angle. (Left) Cells in suspen-
were fixed and stained with crystal violet. (Right) Linear regression analysis
was performed for the relationship between contact angle and cell surface
attachment. n ? 3.
Hydrophobic surfaces play a key role in the development of desired-
Valamehr et al.
September 23, 2008 ?
vol. 105 ?
no. 38 ?
PDMS Promotes Intermediate-Size EB Formation in the Absence of
Serum. The aforementioned studies were carried out in the
presence of fetal bovine serum (FBS), which is known to impact
the efficiencies of EB formation and EB-mediated lineage
differentiation (2). Additionally, most studies in the field are
moving toward serum-free defined culture conditions with ap-
propriate inducers (30). To test whether PDMS can promote
intermediate-size EB formation in the absence of FBS, we first
adapted LW1 ES cells to a serum-free culture system (ESGRO;
Chemicon). After 15 passages in ESGRO medium, LW1 ES cells
had fully adapted to the serum-free culture system and main-
tained their germ-line potential, the gold-standard of pluripo-
tency (data not shown). These adapted LW1 ES cells were then
passaged onto either LAC plates or PDMS-coated wells and
cultured in ESGRO basal medium (Chemicon) without the key
growth. Less cellular debris and cell death (32.5% vs. 16.4%,
12.3%, and 14.8%; Fig. S5 A and B) were observed on PDMS-
coated wells, regardless of the soaking regimen. qRT-PCR
analysis further revealed that EBs formed on PDMS in the
absence of serum express the highest levels of genes associated
with lineage differentiation (P ? 0.05; Fig. S5C).
We next studied lineage-specific differentiation under defined
conditions with appropriate inducers for each of the three
somatic lineages and tested whether EBs formed on PDMS
showed enhanced differentiation yields. When mouse and hu-
man ES cells were directed toward neuronal differentiation by
means of EB intermediates (31), 2.4- and 1.6-fold increases in
total number of neurons for mouse and human, respectively,
were detected (Fig. 5A and Fig. S6A). We also detected signif-
icant increases when mouse and human EBs formed on PDMS
were directed toward hematopoietic lineages (Fig. 5B) (31, 32).
Similarly, when endoderm differentiation was attempted (31,
32), significant increases in gene expression associated with
early- and late-stage endoderm differentiation were detected for
both mouse and human when EBs were initially formed on a
PDMS surface as compared with LAC plates (Fig. 5C and Fig.
S6B). Collectively, by providing a more uniform population of
intermediate-size EBs, PDMS serves as a suitable surface for
promoting ES cell-mediated cellular proliferation and lineage
differentiation, independent of FBS.
To address whether EB size is a bona fide factor in determining
differentiation efficiencies, we studied the first stage of EB
development and demonstrated that EBs can be categorized into
three subgroups according to their diameters: (i) aggregates
(?100 ?m) that are too small to develop into EBs; (ii) an
agglomeration of several EBs that develop into large clumps
(?300 ?m), resulting in enhanced cell death; and (iii) aggregates
of intermediate size (100–300 ?m) for proper EB formation and
development. Therefore, culture systems that can enrich for the
intermediate-size EB population may significantly improve the
yield of ES cell differentiation.
Our studies suggest that hydrophobic surfaces, such as those
obtained from PDMS, C12, and C18, provide a local microen-
vironment that promotes the development of intermediate-size
EB populations and inhibits cellular surface attachment. In
Day 3Day 6Day 8
ison of EBs derived from HSF1 human ES cells cultured either on LAC plates
(Upper) or on C18 surfaces (Lower) and maintained in continuous culture for
lower apoptosis profile as analyzed by TUNEL staining. n ? 5. (C) A higher
percentage of cells derived from EBs cultured on C18 resides in S/G2/M after 7
days of culture. n ? 5. Data presented as mean ? SD;**, P ? 0.01.
Uniform human EB formation on hydrophobic surfaces. (A) Compar-
Relative Gene Expression
Relative Number of Colonies
Relative Number of Neurons
ES cells were directed toward hematopoietic lineage for a total of 12 and 20 days, respectively, and scored for the appearance of various colonies in
The existence of macrophage colonies was confirmed by cytospin for mouse and human. (Right Bottom) Erythroid-containing colonies were confirmed by DAB
staining for mouse and human. (Scale bars, 100 ?m.) (C) Mouse and human ES cells were directed toward endoderm lineage for a total of 12 and 16 days,
respectively, and assayed for associated gene expression. n ? 6. Data presented as mean ? SD; all data, P ? 0.05.
Serum-free differentiation is enhanced by PDMS into the three somatic lineages. (A) Mouse and human ES cells were directed toward neuronal lineage
www.pnas.org?cgi?doi?10.1073?pnas.0807235105Valamehr et al.
contrast, hydrophilic surfaces such as C4 or MHA could not
inhibit surface attachment and produced mainly small aggre-
gates. These observations suggest that SAMs designed for
intermediate-size EB enrichment can be optimized by two
parameters—(i) alkane chain length and (ii) chemical function-
alities or terminal molecules presented at the culture interface—
which, combined, ultimately govern surface wettability. For
example, although MHA has a 16-carbon chain spacer, its
carboxyl terminal group determines its wettability and produces
EBs that are mainly of the small size.
Hydrophobic surfaces have previously been used as a nonad-
hesive background for cell patterning studies (33). Either PDMS
or hydrophobic SAMs can serve as an efficient barrier for cell
attachment over 1 week of cell culture (34). Compared with
techniques that involve multistep procedures to immobilize
surfaces have the advantage of exhibiting naturally low adhesive
character. Moreover, the fact that PDMS is widely used in
microfluidic systems offers promising opportunities for large-
scale culturing technologies. In addition to PDMS, hydrophobic
SAMs can also support the formation of a monodispersed
population of EBs derived from human ES cells. To better
ments in EB formation, we used attenuated total reflectance–
Fourier transform infrared spectroscopy (ATR-FTIR) to ana-
lyze the nature of the culture surface. The resulting spectra
indicate that proteins from the cell culture medium adsorb onto
the surface of PDMS with no evidence of substrate degradation
(Fig. S7). A more detailed analysis of the composition, texture,
and mechanical response of this adsorbed film will be required
in order to fully discern the role that substrate–media interac-
tions play in contributing to the observed hydrophobicity-
mediated enhancement in EB formation.
It is clear that the methyl-terminated hydrophobic surfaces
studied here can provide significant benefits over currently used
EB culture systems, with the production of uniform EB forma-
tion and enhanced cellular survival, proliferation, and differen-
tiation. Importantly, our observations hold true regardless of
FBS usage in the medium recipe, which will play a vital role as
serum-free systems become the standard. In addition, PDMS
can be readily applied as a coating for standard tissue culture
plates. It does not require significant effort, added equipment, or
experience in chemical synthesis for stem cell biologists who wish
to carry out EB-mediated ES cell-lineage-specific differentiation
studies. Collectively, these observations provide insight into the
design of devices and culture systems intended for large-scale
manufacturing of potential ES cell therapies.
Materials and Methods
PDMS. PDMS Sylgard 184 (Dow Corning) was prepared by mixing the prepoly-
mer component with its cross-linker in a 10:1 weight ratio, in accordance with
manufacturer recommendations. Surfaces for suspension culture of ES cells
were then cast directly into six-well plates (Falcon) without degassing to form
?2-mm substrates. Samples were then cured overnight at room temperature.
For some cell lines, 1 mg/ml fresh BSA (MP Biomedical) solution was used to
block residual and long-term culture adhesion to PDMS surfaces. Briefly, BSA
was dissolved in Milli-Q water overnight at 4°C. The mixture was warmed in a
37°C water bath for 1 h, then heated in a 56°C water bath for 1 h. Next, it was
filtered through a 0.22-?m PVDF filter (Steriflip; Millipore) and cooled at 4°C
for 1–2 h. To coat, 2 ml of the BSA solution was added to each well and kept
at 37°C overnight.
Preparation of SAM Substrates. Transparent gold-coated substrates were
initially prepared in a clean-room facility (36). Glass coverslides (34-mm diam-
eter; Fisher) were cleaned in a Piranha solution (70% H2SO4, 30% H2O2) at
100°C for 20 min, followed by thorough rinsing in DI water, and blown dry
under a stream of nitrogen. Immediately after this cleaning procedure, a
10-nm layer of gold with a 5-nm titanium adhesion layer was deposited onto
the cleaned cover glasses with a CHA E-beam evaporator. The result is an
optically transparent, gold-coated substrate similar to materials reported
On completion of the gold evaporation, the slides were immediately
transferred to our laboratory, where the samples were once again cleaned in
Piranha at room temperature for 20 min and subsequently rinsed in DI water
and blown dry with N2. SAMs were then prepared by immersing the cleaned
substrates in a 2 mM ethanolic solution of the appropriate alkanethiol over-
1-propanol for 2 days. On completion of the SAM formation, samples were
samples were then moved to the tissue culture facility, where they were
inserted into six-well tissue culture plates. Before use, each well containing
samples was washed three times with PBS (Gibco).
Additional Methods. The preparation of the various biocompatible surfaces,
and the methods used, are described in SI Materials and Methods.
ACKNOWLEDGMENTS. We thank Drs. Kathrin Plath, Rudolf Jaenisch, Hanna
Mikkola, and Laurraine Gereige, and members of our laboratories, for valu-
able suggestions. S.J.J. is supported by National Science Foundation Integra-
tive Graduate Education and Research Traineeship: Materials Creation Train-
ing Program Grant DGE-0114443, University of California (Los Angeles)
O.N.W. is a Howard Hughes Medical Institute investigator. This work was
Medical Research Grant PN2 EY018228 (to H.W.).
1. Smith AG (2001) Embryo-derived stem cells: Of mice and men. Annu Rev Cell Dev Biol
and medicine. Genes Dev 19:1129–1155.
4. Doetschman TC, Eistetter H, Katz M, Schmidt W, Kemler R (1985) The in vitro devel-
opment of blastocyst-derived embryonic stem cell lines: Formation of visceral yolk sac,
blood islands and myocardium. J Embryol Exp Morphol 87:27–45.
5. Bibel M, et al. (2004) Differentiation of mouse embryonic stem cells into a defined
neuronal lineage. Nat Neurosci 7:1003–1009.
6. Keller G, Kennedy M, Papayannopoulou T, Wiles MV (1993) Hematopoietic commit-
ment during embryonic stem cell differentiation in culture. Mol Cell Biol 13:473–486.
7. Keller GM (1995) In vitro differentiation of embryonic stem cells. Curr Opin Cell Biol
8. Lumelsky N, et al. (2001) Differentiation of embryonic stem cells to insulin-secreting
structures similar to pancreatic islets. Science 292:1389–1394.
9. Dang SM, Kyba M, Perlingeiro R, Daley GQ, Zandstra PW (2002) Efficiency of embryoid
body formation and hematopoietic development from embryonic stem cells in differ-
ent culture systems. Biotechnol Bioeng 78:442–453.
J Am Chem Soc 122:8303–8304.
11. Gouon-Evans V, et al. (2006) BMP-4 is required for hepatic specification of mouse
embryonic stem cell-derived definitive endoderm. Nat Biotechnol 24:1402–1411.
12. Dang SM, Gerecht-Nir S, Chen J, Itskovitz-Eldor J, Zandstra PW (2004) Controlled,
scalable embryonic stem cell differentiation culture. Stem Cells (Dayton) 22:275–282.
numbers of human embryonic stem cells into embryoid bodies fosters robust, repro-
ducible hematopoietic differentiation. Blood 106:1601–1603.
micropattern surfaces for cell-based assays. Biomaterials 27:3044–3063.
cells. Biomaterials 27:6032–6042.
16. Elisseeff J, Ferran A, Hwang S, Varghese S, Zhang Z (2006) The role of biomaterials in
stem cell differentiation: Applications in the musculoskeletal system. Stem Cells Dev
17. Engler AJ, Sen S, Sweeney HL, Discher DE (2006) Matrix elasticity directs stem cell
lineage specification. Cell 126:677–689.
18. Hinz B, Celetta G, Tomasek JJ, Gabbiani G, Chaponnier C (2001) Alpha-smooth muscle
actin expression up-regulates fibroblast contractile activity. Mol Biol Cell 12:2730–
19. Tidwell CD, et al. (1997) Endothelial cell growth and protein adsorption on terminally
functionalized, self-assembled monolayers of alkanethiolates on gold. Langmuir
accidental cell death (necrosis). Cytometry 27:1–20.
21. Biswas A, Hutchins R (2007) Embryonic stem cells. Stem Cells Dev 16:213–222.
efficiency, yield and homogeneity of embryoid body differentiation. Stem Cells (Day-
23. Thomson H (2007) Bioprocessing of embryonic stem cells for drug discovery. Trends
Valamehr et al.
September 23, 2008 ?
vol. 105 ?
no. 38 ?
24. Burridge PW, et al. (2007) Improved human embryonic stem cell embryoid body Download full-text
homogeneity and cardiomyocyte differentiation from a novel V-96 plate aggregation
system highlights interline variability. Stem Cells (Dayton) 25:929–938.
embryonic stem cell lines. Stem Cells (Dayton) 19:193–204.
27. Schuldiner M, Yanuka O, Itskovitz-Eldor J, Melton DA, Benvenisty N (2000) Effects of
eight growth factors on the differentiation of cells derived from human embryonic
stem cells. Proc Natl Acad Sci USA 97:11307–11312.
the initiation of definitive hematopoiesis in the mouse embryo. Blood 101:508–516.
ers of thiolates on metals as a form of nanotechnology. Chem Rev 105:1103–1170.
30. Murry CE, Keller G (2008) Differentiation of embryonic stem cells to clinically relevant
populations: Lessons from embryonic development. Cell 132:661–680.
31. Irion S, et al. (2007) Identification and targeting of the ROSA26 locus in human
embryonic stem cells. Nat Biotechnol 25:1477–1482.
for the induction of an in vitro model of primitive streak formation using embryonic
stem cells. Proc Natl Acad Sci USA 103:16806–16811.
33. Orner BP, Derda R, Lewis RL, Thomson JA, Kiessling LL (2004) Arrays for the combina-
torial exploration of cell adhesion. J Am Chem Soc 126:10808–10809.
34. De Silva MN, Desai R, Odde DJ (2004) Micro-patterning of animal cells on PDMS
substrates in the presence of serum without use of adhesion inhibitors. Biomed
35. Kim P, et al. (2005) Fabrication of nanostructures of polyethylene glycol for
applications to protein adsorption and cell adhesion. Nanotechnology 16:2420–
36. DimillaPA, etal.(1994)Wettingandproteinadsorptionofself-assembledmonolayers
of alkanethiolates supported on transparent films of gold. J Am Chem Soc 116:2225–
www.pnas.org?cgi?doi?10.1073?pnas.0807235105 Valamehr et al.