Protein assembly directed by synthetic molecular recognition motifs

Article abstract of DOI:10.1039/c1ob05998j

Tris-functionalized cyanuric acid (TCA) and melamine (TM) selectively recognize each other in aqueous solution with 1:1 stoichiometry. We have coupled biotin to TCA and TM to allow pseudo-tetrahedral display of TCA and TM on streptavidin through biotin-ligand binding. Synthetic cyanuric acid/melamine recognition is found to drive selective protein-protein assembly.

Full text of DOI:10.1039/c1ob05998j

View Article Online / Journal Homepage / Table of Contents for this issue
Organic &
Biomolecular
Chemistry
Dynamic Article Links
Cite this: Org. Biomol. Chem., 2011, 9, 7296
www.rsc.org/obc
COMMUNICATION
Protein assembly directed by synthetic molecular recognition motifs†
Mingming Ma and Dennis Bong*
Received 21st June 2011, Accepted 12th August 2011
DOI: 10.1039/c1ob05998j
Tris-functionalized cyanuric acid (TCA) and melamine (TM)
selectively recognize each other in aqueous solution with
1 : 1 stoichiometry. We have coupled biotin to TCA and
TM to allow pseudo-tetrahedral display of TCA and TM
on streptavidin through biotin–ligand binding. Synthetic
cyanuric acid/melamine recognition is found to drive selective
protein–protein assembly.
direct consequence of the complementary hydrogen bond donor–
acceptor patterning of each heterocycle. We have previously
examined cyanuric acid/melamine recognition anchored at the
lipid–water interface with a phospholipid or peptide anchor.38–40
Assembly at hydrophobic-water interfaces is known to enhance
weak hydrogen bonding interactions by shielding H-bond donor–
acceptor sites from competitive aqueous solvent. This has been
demonstrated with a number of reports on the aqueous-phase
assembly of hydrogen bonding amphiphiles. We found that the
cyanuric acid/melamine system could be induced to assemble in
water without an amphiphilic scaffold; when displayed on the Tris
scaffold (Fig. 1), these molecules could bury sufficient surface area
on recognition to drive nanoparticle assembly.41 We were curious
whether sterically demanding protein assembly could be mediated
by TCA and TM recognition at the protein–protein interface;
macromolecular assembly was clearly possible at the lipid–water
interface but interfacial stoichiometry was not controlled. Defined
protein functionalization may be achieved in a number of ways, but
to test the assembly strategy, it was most expedient to couple TCA
and TM to biotin, which binds tightly (Kd~10-15 M) to streptavidin
protein (Scheme 1). Streptavidin is a homotetramer with 4 biotin
binding sites and pseudo-tetrahedral protein symmetry.42–44 Biotin
site saturation with TCA–Btn and TM–Btn should place either
four TCA or four TM recognition modules at each pseudo-
tetrahedral site of streptavidin, thus transforming those ligand
sites into protein–protein assembly42 interfaces (Scheme 1). We
first set out to confirm that biotin functionalization of TCA
Bio-based materials have potential applications as useful struc-
tural or functional biomaterials.1–9 We describe herein non-
covalent protein assembly triggered by small molecule recognition
at the protein surface. Native protein assemblies exhibit sophisti-
cated function and play crucial roles in biological processes (e.g.
actins, tubulins and ribosome proteins);10–14 protein misassembly
conversely is associated in toxic loss or gain of function, as found
in neurodegenerative diseases such as Alzheimer’s, Parkinson’s
and Huntington’s disease.15–18 Thus, the function and pathology of
native protein assemblies has inspired considerable investigation
of both native19–22 and artificial systems6,23–25 in an effort to recreate
and understand these desirable and undesirable aspects of protein
assemblies. Two general strategies of fabricating artificial protein
assemblies have been explored: protein-mediated connection24,25
and multivalent ligand-mediated connection.23 The ligand-based
strategy permits a more straightforward method for tuning the
properties of the artificial protein assemblies, as compared to
the task of retooling a protein–protein interface;26–28 we have
explored this concept by using the synthetic cyanuric acid and
melamine recognition motif to construct an artificial protein–
protein assembly interface.
Cyanuric acid (CA) and melamine (M) recognition has been
studied extensively in organic solvents and the solid state.29–37
We have recently reported on the utility of these simple hete-
rocycles as recognition motifs in water. The driving force for
assembly in water appears to derive from favorable polar–polar
surface area burial along with hydrophobic surface area burial
when Tris is used as a scaffold; these effects are likely to be
less important in organic solvents where hydrogen-bonding is
a strong contributor to recognition. Regardless of the origins
and solvent, CA/M recognition is highly selective and is a
Department of Chemistry, The Ohio State University, Columbus, USA.
E-mail: bong@chem.osu.edu; Fax: +1 614-292-1685; Tel: +1 614-247-8404
† Electronic supplementary information (ESI) available: Detailed experi-
mental procedures, compound characterization, peptide characterization
(MS, CD), SDS-PAGE. See DOI: 10.1039/c1ob05998j
Fig. 1 Synthetic recognition modules used in this study. Peptide sequences
for cc1 and cc2 are shown using the single letter code. The cysteine
residue used to form the thioether bond is shown in bold in cc1. Aba is
acetamidobenzamide, used as a chromophore to determine concentration.
7296 | Org. Biomol. Chem., 2011, 9, 7296–7299
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
the aggregates formed have a kinetic stability while formation
of an “assembly nucleus” from soluble protein requires cooling
to 15 ◦C. These protein aggregates were clearly derived from
streptavidin conjugates as indicated by SDS-PAGE (Supporting
Information).
We further probed the TCA-Btn-SA/TM-Btn-SA interaction
by surface plasmon resonance (SPR) experiments, in which TCA–
Btn or TM–Btn was used to surface saturate a streptavidin SPR
chip the complementary streptavidin biotin conjugate was flowed
over the surface. While control SA protein resulted in no signal,
complementary SA conjugate produced strong, concentration-
dependent signals consistent with protein–protein association
interaction (Fig. 3A); this behavior was similar regardless of which
SA conjugate was on the surface and which was in solution. Slow
equilibration, indicated by thermal hysteresis in the DLS experi-
ment (Fig. 3A), also prevented direct fitting of kinetic constants;
nevertheless, a submicromolar estimate of Kd, was obtained by
assumption of a simple binding model and observation of analyte
concentration at 50% surface saturation (Fig. 3B). This Kd value
is consistent with the apparent Kd of the TCA–Btn/TM–Btn
complex in PBS buffer, (if a 1 : 1 complex is assumed) which
supports the notion that the aggregation of functionalized SA
proteins is mediated by TCA/TM recognition. Thus, it was clear
that synthetic recognition modules can induce protein assembly
in a reversible way. This was qualitatively underscored by the
ability of soluble TCA and TM modules to competitively inhibit
protein binding in the SPR experiment (Fig. 3D). Notably, there
are two different noncovalent linkages in this assembly; the tighter
biotin–streptavidin binding anchors display of a weaker but more
reversible synthetic CA/M recognition motif, thus resulting in a
supramolecular synthon using a protein building block.
Scheme 1 Illustration of streptavidin (SA, blue) being saturated with
TCA–Btn and TM–Btn and formation of protein assembly mediated by
TCA/TM recognition.
and TM modules did not affect recognition. Indeed, simple
coupling with a diamine resulted in poorly soluble products and
it was necessary to use diaminopropionic acid as a linker, which
preserved the solubilizing negative charge of the carboxylate group
(Fig. 1). The resulting biotin conjugates, TCA–Btn and TM–Btn
were readily purified and characterized. Biotin substitution did not
affect molecular recognition by TCA and TM modules as judged
by isothermal titration calorimetry (ITC), yielding exothermic
◦
interaction enthalpies of -29 to -30 kcal mol-1 at 25 and 10 C,
respectively, in line with what was observed with other TCA/TM
derivatives41 (Fig. 2). Interestingly, this data fits well to the 1 : 1
heterodimeric binding model, yielding a Kd of 0.59 and 0.070 mM
at 25 and 10 ◦C, respectively. Although we have no direct structural
evidence for this complex other than this data fit.
Fig. 2 Isothermal titration calorimetry data for titration of TCA–Btn
into a solution of TM–Btn. In (A) raw data taken at 10 ◦C is shown and
integrated data (B) taken at 10 ( ) and 25 ◦C (ꢀ). Solid lines represent fits
᭹
to a 1 : 1 heterodimer binding model.
With this functioning recognition pair in hand, we treated
streptavidin separately with either TCA–Btn or TM–Btn in excess
to saturate biotin-binding sites with these synthetic recogni-
tion modules. Unbound biotin derivatives were purified away
from protein by protein precipitation and washing. In isolation,
streptavidin-conjugates were soluble protein samples as judged
by minimal scattering. However, mixing TCA–Btn–SA with TM–
Btn–SA in a 1 : 1 ratio in buffer resulted in rapid formation of large
aggregates as detected by DLS. This finding echoes prior results in
which mixtures of TCA/TM formed nanoparticles with rigorously
maintained 1 : 1 stoichiometry. These protein aggregates could
be thermally dissociated in a reversible fashion, though with
considerable hysteresis. The melting temperature of the aggregates
is slightly higher than room temperature (~30 ◦C), which is
signifi◦cantly below streptavidin Tm (~75 ◦C for apostreptavidin,
~110 C for biotin-saturated streptavidin),45 so this aggregation
is clearly associated with the non-native interaction between
the synthetic TCA and TM ligands. On cooling, the system
must be cooled to 15 ◦C before assembly begins. The hysteresis
indicates non-equilibrium assembly conditions and suggests that
Fig. 3 (A) DLS of a 1 : 1 mixture of TCA–Btn–SA/TM–Btn–SA as a
function of temperature, with heating and cooling traces shown in red and
blue, respectively. The first heating-cooling cycle is shown in a solid line
and the second cycle is shown in a dashed line. (B) SPR signal as function
of soluble TM–Btn–SA concentration over a TCA–Btn–streptavidin SPR
chip. Inhibition of TM–Btn–SA protein binding to a TCA–Btn–SA chip
with TCA is shown in (C), while inhibition of TCA–Btn–SA binding to a
TM–Btn–SA chip with TM is shown in (D). A 100-fold excess of soluble
ligand (TCA or TM) to protein in solution was used for inhibition.
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 7296–7299 | 7297
©

View Article Online
It is known that streptavidin itself has a propensity to as-
semble into oligomers beyond the tetramer.46 We suspect that
our recognition modules are working cooperatively with this
native oligomerization interface to form a stabilized aggregate.
It is possible that partial or complete burial of the TCA/TM
complex at this protein–protein interface facilitates assembly. To
probe this notion, we synthesized a system with more “exposed”
TCA/TM modules by coupling these groups via thioether for-
mation with the sulfhydryl sidechain of the N-terminal cysteine
of a dimeric coiled-coil peptide (Fig. 4).47 These peptides were
readily synthesized on solid phase, purified, then coupled to
TCA/TM chlorides in PBS buffer. While these peptides retained
their ability to form dimeric coiled-coils as judged by analytical
ultracentrifuge, no higher oligomerization states were observed by
AUC under mid to high micromolar conditions (Fig. 4). Thus,
it appears that the dimeric coiled coil platform does not medi-
ate assembly, possibly because of insufficient interfacial surface
area.
Acknowledgements
This work was supported in part by the NSF and the Institute of
Materials Research at the Ohio State University. We thank Deniz
Yu¨ksel and Prof. Krishna Kumar for assistance with analytical
ultracentrifuge experiments. The Analytical Ultracentrifugation
facility at Tufts University is supported by the NIH.
Notes and references
1 N. Huebsch and D. J. Mooney, Nature, 2009, 462, 426–432.
2 R. V. Lewis, Chem. Rev., 2006, 106, 3762–3774.
3 L. Little, K. E. Healy and D. Schaffer, Chem. Rev., 2008, 108, 1787–
1796.
4 A. J. Ragauskas, C. K. Williams, B. H. Davison, G. Britovsek, J. Cairney,
C. A. Eckert, W. J. Frederick, J. P. Hallett, D. J. Leak, C. L. Liotta, J. R.
Mielenz, R. Murphy, R. Templer and T. Tschaplinski, Science, 2006,
311, 484–489.
5 D. N. Woolfson and Z. N. Mahmoud, Chem. Soc. Rev., 2010, 39, 3464–
3479.
6 S. I. Stupp, Nano Lett., 2010, 10, 4783–4786.
7 T. J. Deming, Prog. Polym. Sci., 2007, 32, 858–875.
8 K. Rajagopal and J. P. Schneider, Curr. Opin. Struct. Biol., 2004, 14,
480–486.
9 C. Q. Yan and D. J. Pochan, Chem. Soc. Rev., 2010, 39, 3528–3540.
10 D. A. Fletcher and D. Mullins, Nature, 2010, 463, 485–492.
11 T. D. Pollard and J. A. Cooper, Science, 2009, 326, 1208–1212.
12 H. Y. Kueh and T. J. Mitchison, Science, 2009, 325, 960–963.
13 M. Schliwa and G. Woehlke, Nature, 2003, 422, 759–765.
14 M. G. L. van den Heuvel and C. Dekker, Science, 2007, 317, 333–
336.
15 C. M. Dobson, Nature, 2003, 426, 884–890.
16 D. J. Selkoe, Nature, 2003, 426, 900–904.
17 R. Wetzel, Cell, 1996, 86, 699–702.
18 R. Roychaudhuri, M. Yang, M. M. Hoshi and D. B. Teplow, J. Biol.
Chem., 2009, 284, 4749–4753.
19 T. H. Bayburt and S. G. Sligar, FEBS Lett., 2010, 584, 1721–
1727.
20 D. B. Teplow, N. D. Lazo, G. Bitan, S. Bernstein, T. Wyttenbach, M. T.
Bowers, A. Baumketner, J. E. Shea, B. Urbanc, L. Cruz, J. Borreguero
and H. E. Stanley, Acc. Chem. Res., 2006, 39, 635–645.
21 Y. Levy and J. N. Onuchic, Acc. Chem. Res., 2006, 39, 135–142.
22 T. Pawson and P. Nash, Science, 2003, 300, 445–452.
23 A. Fegan, B. White, J. C. T. Carlson and C. R. Wagner, Chem. Rev.,
2010, 110, 3315–3336.
24 E. N. Salgado, R. J. Radford and F. A. Tezcan, Acc. Chem. Res., 2010,
43, 661–672.
25 P. Ringler and G. E. Schulz, Science, 2003, 302, 106–109.
26 M. R. Arkin, M. Randal, W. L. DeLano, J. Hyde, T. N. Luong, J. D.
Oslob, D. R. Raphael, L. Taylor, J. Wang, R. S. McDowell, J. A. Wells
and A. C. Braisted, Proc. Natl. Acad. Sci. U. S. A., 2003, 100, 1603–
1608.
Fig. 4 (Top) Illustration of the formation of peptide heterodimers
functionalized with TCA and TM that do not assemble into higher
aggregates. (Bottom) Analytical ultracentrifuge data of a mixture of the
TCA and TM heterodimers fit to a heterodimer model (solid line).
In summary, we have demonstrated the ability of synthetic
recognition groups based on cyanuric acid and melamine recogni-
tion to trigger the assembly of streptavidin protein in a reversible
manner. We have noncovalently conjugated these recognition
modules to streptavidin in a defined manner using the native
biotin ligand binding sites through biotin coupling to TCA
and TM and shown thermally reversible assembly. Thus, this
synthetic recognition motif, which has been shown to be functional
in aqueous solution41 and at the lipid–water interface, also
functions at a protein–protein interface. However, our experiments
with coiled-coil peptides suggest that an interface that provides
sufficient burial of the recognition module is necessary for recog-
nition. This is entirely consistent with our studies on TCA/TM
recognition in isolation, and bodes well for future efforts using
these synthetic motifs for directing biomolecular interactions in
the synthesis of novel biomaterials and functional molecular
assemblies.
27 S. Atwell, M. Ultsch, A. M. DeVos and J. A. Wells, Science, 1997, 278,
1125–1128.
28 W. L. DeLano, M. H. Ultsch, A. M. de Vos and J. A. Wells, Science,
2000, 287, 1279–1283.
29 M. G. J. ten Cate, J. Huskens, M. Crego-Calama and D. N. Reinhoudt,
Chem.–Eur. J., 2004, 10, 3632–3639.
30 K. Damodaran, G. J. Sanjayan, P. R. Rajamohanan, S. Ganapathy and
K. N. Ganesh, Org. Lett., 2001, 3, 1921–1924.
31 A. Ranganathan, V. R. Pedireddi and C. N. R. Rao, J. Am. Chem. Soc.,
1999, 121, 1752–1753.
32 J. A. Zerkowski, J. C. MacDonald, C. T. Seto, D. A. Wierda and G. M.
Whitesides, J. Am. Chem. Soc., 1994, 116, 2382–2391.
33 J. P. Mathias, E. E. Simanek, C. T. Seto and G. M. Whitesides,
Macromol. Symp., 1994, 77, 157–166.
34 C. T. Seto and G. M. Whitesides, J. Am. Chem. Soc., 1993, 115, 905–
916.
35 J. A. Zerkowski, C. T. Seto and G. M. Whitesides, J. Am. Chem. Soc.,
1992, 114, 5473–5475.
36 Y. Wang, B. Wei and Q. Wang, J. Chem. Cryst., 1990, 20, 79–84.
37 C. T. Seto and G. M. Whitesides, J. Am. Chem. Soc., 1990, 112, 6409–
6411.
7298 | Org. Biomol. Chem., 2011, 9, 7296–7299
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
38 M. Ma and D. Bong, Langmuir, 2011, 27, 1480–1486.
39 M. Ma, Y. Gong and D. Bong, J. Am. Chem. Soc., 2009, 131, 16919–
16926.
44 W. A. Hendrickson, A. Pahler, J. L. Smith, Y. Satow, E. A. Merritt and
R. P. Phizackerley, Proc. Natl. Acad. Sci. U. S. A., 1989, 86, 2190–2194.
45 M. J. Waner, I. Navrotskaya, A. Bain, E. D. Oldham and D. P. Mascotti,
Biophys. J., 2004, 87, 2701–2713.
46 J. Huskens, A. Mulder, T. Auletta, C. A. Nijhuis, M. J. Ludden and
D. N. Reinhoudt, J. Am. Chem. Soc., 2004, 126, 6784–6797.
47 P. B. Harbury, T. Zhang, P. S. Kim and T. Alber, Science, 1993, 262,
1401–1407.
40 M. M. Ma, A. Paredes and D. Bong, J. Am. Chem. Soc., 2008, 130,
14456.
41 M. Ma and D. Bong, Langmuir, 2011, 27, 8841–8853.
42 P. Ringler and G. E. Schulz, Science, 2003, 302, 106–109.
43 C. M. Paleos and D. Tsiourvas, Adv. Mater., 1997, 9, 695–710.
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 7296–7299 | 7299
©