Arrays for the combinatorial exploration of cell adhesion

Article abstract of DOI:10.1021/ja0474291

A new method for the fabrication of arrays of self-assembled monolayers (SAMs) of alkane thiols (ATs) on gold to combinatorially assay surfaces for cell adhesion is reported. A fluorous SAM, which is both cytophobic and solvophobic, was used as the backgr

Full text of DOI:10.1021/ja0474291

Published on Web 08/14/2004
Arrays for the Combinatorial Exploration of Cell Adhesion
Brendan P. Orner,†,‡ Ratmir Derda,‡ Rachel L. Lewis,§ James A. Thomson,§ and
Laura L. Kiessling*,†,‡
Departments of Biochemistry, Chemistry, and Anatomy, UniVersity of Wisconsin, Madison, Wisconsin 53706
Received May 2, 2004; E-mail: kiessling@chem.wisc.edu
The surfaces with which cells interact are important for maintain-
ing cellular viability and localization. The features of these surfaces
can act as signals that influence cellular behavior. Dissecting the
critical features of physiological surfaces is difficult given their
complexity and heterogeneity. Synthetic, homogeneous, surfaces
presenting specific ligands for cell surface receptors can be used
to unravel key features of physiological surfaces responsible for
eliciting a specific cellular response.1,2 Here we present a method
to screen diverse surfaces to identify those with illuminating cell
adhesive or signaling properties.
Combinatorial chemistry and diversity-oriented synthesis can be
used to generate ligands that bind a specific receptor. Parameters
beyond binding specificity must be considered when synthesizing
surfaces that interact with cells. Cell binding to another cell or
matrix is influenced not only by the presence of a specific ligand
but also by its density on the surface. Moreover, cell surface ligands
can act synergistically. Thus, it would be valuable to display
combinations of ligands at defined densities. To explore how surface
presentation of ligands influences cell viability, adhesion, and
response, we developed a method to generate patterned surfaces
that present arrays of ligands (Figure 1).
Useful arrays for testing cell attachment exist;2 however, none
Figure 1. (A) Arrays for the combinatorial evaluation of cell adhesion
and signaling. Each square represents a different surface element. The
magnified section illustrates that elements may afford cell adhesion and/or
differentiation. (B) Illustration of the solvophobicity of a fluorous SAM
can be used to present many different ligands and/or ligand
(the colored shapes represent different solvents). (C) SAM composed of
combinations at defined densities. To design a strategy to accom-
fluoro-AT patterned to reveal bare gold holes. ATs containing ligands are
modate our needs, we chose to employ self-assembled monolayers
then spotted to create the array. The blue and green ovals represent solutions
(SAMs). SAMs are well-characterized surfaces3 that have been used
of different ATs. (D) ATs employed.
for cell attachment and patterning.4 SAMs are attractive because
they are well-packed, are homogeneous across the assembly, and
can be reproducibly assembled and patterned. With SAMs, differ-
ences in cell binding can be attributed to the structure and density
of the ligands presented.
We sought to develop a method to pattern SAMs for the
combinatorial study of cell adhesion (Figure 1A). Compounds that
Figure 2. Two cell lines plated on arrays formed by spotting various
bind cell surface receptors include proteins, peptides, carbohydrates,
and small molecules; the solubilities of these ligands vary. Thus,
we required a method to pattern the SAMs that would resist cross-
contamination of spots containing ligands of different solutes;
therefore, the alkane thiol (AT) that is used between array elements
(background) must create a “solvophobic” surface (i.e., one that
resists spreading of solvents) (Figure 1B). The background SAM
also must be cytophobic; it serves as a barrier to cell attachment
and growth (Figure 1A). With these criteria in mind, we reasoned
that a SAM formed from a perfluoro AT would serve as an excellent
background. Such surfaces are known to be cytophobic4a,4c,5,6 and
solvophobic.7,8 By taking advantage of the features of the fluorous
SAM, we could employ an “assembly after conjugation” strategy
to fabricate the target arrays. Thus, we envisioned assembling SAMs
by spotting ATs with preattached ligands on bare gold “holes” in
fluorous SAMs (Figure 1C).
percentages of PEG acid-AT and polyol-AT on bare gold squares surrounded
by a fluoro-AT SAM (percentage refers to the amount of PEG acid-AT).
(A) Fixed (2% paraformaldehyde) and stained (Coomassie) Swiss 3T3
fibroblast cells (the 100% spot was spatially separated but in the same array).
(B) Fixed and stained SH-SY5Y cells. All squares are 750 µm in length.
To test this strategy, we generated a pattern containing an AT
known to bind cells, PEG acid-AT,4b using the cytophobic fluoro-
AT9 as the background (Figure 1D). To test the importance of the
density of the cytophilic AT, solutions of varying molar ratios of
PEG acid-AT diluted with a hydrophilic AT that does not interact
with cells, polyol-AT,7 were spotted on bare gold areas formed by
photopatterning a SAM. Neuroblastoma (SH-SY5Y) and fibroblast
(Swiss 3T3) cell lines were plated on the patterned surfaces, allowed
to proliferate, fixed, and stained (Figure 2). For both cell lines,
adhesion decreases as the mole fraction of the acid within the spots
approaches 0.2. These data illustrate the feasibility of the spotting
method, and they indicate that SAMs composed of mixtures of ATs
can be screened using this approach.
† Department of Biochemistry.
‡ Department of Chemistry.
§ Department of Anatomy.
9
10808
J. AM. CHEM. SOC. 2004, 126, 10808-10809
10.1021/ja0474291 CCC: $27.50 © 2004 American Chemical Society

C O M M U N I C A T I O N S
Figure 4. Human embryonic stem cells plated on arrays formed by spotting
various percentages of PEG acid-AT (indicated) and polyol-AT on bare
gold squares surrounded by a fluoro-AT SAM. (A) Fixed (2% paraform-
aldehyde) and stained (Coomassie) Human ES cells (the 100% spot was
spatially separated but in the same array). (B) Human ES cells on 100%
PEG Acid-AT SAM stained for alkaline phosphatase activity. All squares
are 750 µm in length.
The data indicate that our method for fabricating ligand-
presenting arrays can be used for cell-based screens. We have shown
that arrays generated by this strategy can present combinations of
ligands; different cell lines can “read” the same surfaces differently.
Given the flexibility of our design, we anticipate that it can be used
to present virtually any single AT or combination of synthetic ATs.
We expect that this method can be used to fabricate arrays that
address fundamental questions critical for understanding and
controlling the adhesion, proliferation, and differentiation of cells.
Figure 3. Two cell lines plated on arrays formed by spotting with various
percentages of peptide-ATs and polyol-AT on bare gold squares surrounded
by a fluoro-AT SAM. (A) Fixed (2% paraformaldehyde) and stained
(Coomassie) Swiss 3T3 fibroblast cells. (B) Fixed and stained SH-SY5Y
neuroblastoma cells. “100% -COOH” indicates PEG acid-AT. The label
“-GYIGSR” denotes the peptide-AT derivatized with the sequence
GSDPGYIGSR. All squares are 750 µm in length.
In physiological settings, cells interpret cues from the extra-
cellular matrix. Different cell types interact with different matrix
proteins. The sequence RGD, which is found in a number of
proteins, including fibronectin and vitronectin, binds a family of
integrins on the surface of some cell lines, including Swiss-3T3
fibroblasts. This binding is specific; a simple Arg to Lys substitution
obliterates adhesion.10 Alternatively, the peptide YIGSR, a sequence
found in the matrix protein laminin, interacts with neuroblastoma
cell lines but not with fibroblasts.11 Therefore, surfaces displaying
YIGSR are expected to bind the neuroblastoma cell line and the
fibroblast cell line is expected to adhere preferentially to RGD
surfaces over those presenting KGD.
Acknowledgment. We thank P. J. Belshaw, E. Endler, and R.
Corn for helpful discussions for use of equipment. This research
was supported by DARPA and the W. M. Keck Foundation. B.P.O.
thanks the NIH (AG19550) for a postdoctoral fellowship.
Supporting Information Available: Synthetic methods and ex-
perimental details for the development of the methodology (PDF). This
material is available free of charge via the Internet at http://pubs.acs.org.
References
To test whether surfaces displaying different peptides interact
selectively with different cell types, a surface displaying fluoro-
AT SAM was photopatterned. Mixtures of polyol-AT and peptide-
AT12 were spotted on the bare gold. An array element known to
bind cells, 100% PEG acid-AT, and one that resists cell binding,
100% polyol-AT, were also spotted. Swiss 3T3 and SH-SY5Y cells
were plated on the resulting chips, allowed to proliferate, fixed,
and stained. The Swiss 3T3 fibroblasts (Figure 3A) preferentially
bound the RGD surface; minimal binding was detected to the KGD
element. Modest fibroblast adhesion to the YIGSR-modified surface
was observed; as the percentage of polyol-AT was increased,
however, cell binding fell off more rapidly than did binding to RGD-
presenting surfaces. The neuroblastoma cell line, SH-SY5Y (Figure
3B), adheres to the YIGSR-substituted surface but not to either
the RGD or KGD elements. Thus, cell-line specific binding profiles
can be established using these arrays.
Human embryonic stem (ES) cells are developmental precursors
to all cell types and thus termed “pluripotent”.13 The ability to
culture human ES cells on synthetic surfaces would eliminate the
complications of using mouse embryonic fibroblasts as a feeder
layer during ES expansion.14 To determine whether our method
can be used to identify surfaces upon which to culture human ES
cells, solutions of different molar ratios of PEG acid-AT and polyol-
AT were employed to generate arrays upon which human ES cells
were plated. Again, a gradient of cell attachment was observed
(Figure 4). The human ES cells, however, require a more acidic
surface than the two previously investigated cell lines. Interestingly,
alkaline phosphatase staining of human ES cells indicates that these
cells remain undifferentiated for at least 2 days after plating (Figure
4B). It is intriguing that such a simple surface can propagate human
embryonic stem cells. This result bodes well for finding optimized
surfaces for long-term ES cell proliferation and controlled dif-
ferentiation.
(1) (a) Singhvi, R.; Kumar, A.; Lopez, G. P.; Stephanopoulos, G. N.; Wang,
D. I. C.; Whitesides, G. M.; Ingber, D. E. Science 1994, 264, 696-698.
(b) Yousaf, M. N.; Houseman, B. T.; Mrksich, M. Proc. Natl. Acad. Sci.
U.S.A. 2001, 98, 5992-5996. (c) Jung, D. R.; Kapur, R.; Adams, T.;
Giuliano, K. A.; Mrksich, M.; Craighead, H. G.; Taylor, D. L. Crit. ReV.
Biotechnol. 2001, 21, 111-154.
(2) (a) Falsey, J. R.; Renil, M.; Park, S.; Li, S.; Lam, K. Bioconjugate Chem.
2001, 12, 346-353. (b) Hoff, A.; Andre, T.; Schaffer, T. E.; Jung, G.;
Wiesmuller, K.-H.; Brock, R. ChemBioChem 2002, 2, 1183-1191. (c)
Aina, O. H.; Sroka, T. C.; Chen, M.-L.; Lam, K. S. Biopolymers 2002,
66, 184-199. (d) Ito, Y.; Nogawa, M. Biomaterials 2003, 24, 3021-
3026.
(3) Bain, C. D.; Troughton, E. B.; Tao, Y.-T, Evall, J.; Whitesides, G. M.;
Nuzzo, R. G. J. Am. Chem. Soc. 1989, 111, 321-335.
(4) (a) Stenger, D. A.; Georger, J. H.; Dulcey, C. S.; Hickman, J. J.; Rudolph,
A. S.; Nielsen, T. B.; McCort, S. M.; Calvert, J. M. J. Am. Chem. Soc.
1992, 114, 8435-8442. (b) Lopez, G. P.; Albers, M. W.; Schreiber, S.
L.; Carroll, R.; Peralta, E.; Whitesides, G. M. J. Am. Chem. Soc. 1993,
115, 5877-5878. (c) Spargo, B. J.; Testoff, M. A.; Nielsen, T. B.; Stenger,
D. A.; Hickman, J. J.; Rudolph, A. S. Proc. Natl. Acad. Sci. U.S.A. 1994,
91, 11070-11074.
(5) (a) Margel, S.; Vogler, E. A.; Firment, L.; Watt, T.; Haynie, S.; Sogah,
D. Y. J. Biomed. Mater. Res. 1993, 27, 1463-1476. (b) Ranieri, J. P.;
Bellamkonda, R.; Jacob, J.; Vargo, T. G.; Gardella, J. A.; Aebischer, P.
J. Biomed. Mater. Res. 1993, 27, 917-925.
(6) See Supporting Information.
(7) Luk, Y. Y.; Kato, M.; Mrksich, M. Langmuir 2000, 16, 9604-9608.
(8) Sorribas, H.; Padeste, C.; Tiefenauer, L. Biomaterials 2002, 23, 893-
900.
(9) (a) Graupe, M.; Koini, T.; Wang, V. Y.; Nassif, G. M.; Colorado, R.;
Villazana, R. J.; Dong, H.; Miura, Y. F.; Shmakova, O. E.; Lee, T. R. J.
Fluorine Chem. 1999, 93, 107-115. (b) Naud, C.; Calas, P.; Blancou H.;
Commeyras A. J. Fluorine Chem. 2000, 104, 173-183.
(10) Pierschbacher, M. D.; Ruoslahti, E. Nature 1984, 309, 30-33.
(11) Graf, J.; Iwamoto, Y.; Sasaki, M.; Martin, G. R.; Kleinman, H. K.; Robey,
F. A.; Yamada, Y. Cell 1987, 48, 989-996.
(12) Houseman, B. T.; Mrksich, M. J. Org. Chem. 1998, 63, 7552-7555.
(13) Thomson, J. A.; Itskovitz-Eldor, J.; Shapiro, S. S.; Waknitz, M. A.;
Swiergiel, J. J.; Marshall, V. S.; Jones, J. M. Science 1998, 282, 1145-7.
(14) Xu, C.; Inokuma, M. S.; Denham, J.; Golds, K.; Kundu, P.; Gold, J. D.;
Carpenter, M. K. Nat. Biotechnol. 2001, 19, 971-974.
JA0474291
9
J. AM. CHEM. SOC. VOL. 126, NO. 35, 2004 10809