Site-specific, reversible and fluorescent immobilization of proteins on CrAsH-modified surfaces for microarray analytics

Article abstract of DOI:10.1039/c4cc04120h

A novel technique for protein immobilization onto CrAsH-modified surfaces is presented. This approach enables an efficient, reversible and fluorogenic immobilization of proteins. Moreover, expressed proteins can also be directly immobilized from cellular lysates without prior purification. The immobilized proteins are suitable for protein-protein interaction studies and the fluorescence enhancement upon immobilization can be employed for the direct detection of the immobilized protein without the need for secondary detection methods. This journal is

Full text of DOI:10.1039/c4cc04120h

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Site-specific, reversible and fluorescent
immobilization of proteins on CrAsH-modified
surfaces for microarray analytics†
Cite this: Chem. Commun., 2014,
50, 12761
Received 29th May 2014,
Accepted 5th September 2014
Janine Schulte-Zweckel,‡a Federica Rosi,‡a Domalapally Sreenu,a
Hendrik Schroder,b Christof M. Niemeyerbc and Gemma Triola*ad
¨
DOI: 10.1039/c4cc04120h
www.rsc.org/chemcomm
A novel technique for protein immobilization onto CrAsH-modified requires previous protein biotinylation. Hence, the development
surfaces is presented. This approach enables an efficient, reversible of novel immobilization methods is still in high demand.
and fluorogenic immobilization of proteins. Moreover, expressed
In general, a protein immobilization technique should ideally
proteins can also be directly immobilized from cellular lysates combine the best features of all the currently existing methods: a
without prior purification. The immobilized proteins are suitable high-affinity but reversible binding, a site-specific immobilization
for protein–protein interaction studies and the fluorescence under simple and mild conditions and a genetic incorporation of
enhancement upon immobilization can be employed for the the required tag, thus avoiding additional chemical or enzymatic
direct detection of the immobilized protein without the need for transformations and enabling a straightforward immobilization
secondary detection methods.
of proteins-of interest directly from cell lysates. In addition to
these features, an ideal protein immobilization technique would
In the last decade protein immobilization techniques have become also allow one to directly monitor successful surface binding
an essential tool for the understanding of protein function and without the need for secondary detection steps (i.e. labelled
interactions by enabling a rapid profile of proteins with a minimal antibodies or proteins) or without requiring label-free methods
amount of sample. The major challenge in the generation of such as Surface Plasmon Resonance (SPR) or Mass Spectrometry
protein biochips is the development of suitable conditions for (MS), not available in every laboratory. Here, we have applied
the mild immobilization of proteins onto surfaces. Site-specific the interaction between biarsenical derivatives of fluorescein
attachment is usually preferred to random immobilization and tetracysteine (TC)-motifs as a novel strategy for protein
because it ensures a homogeneous protein layer with a defined microarray generation which meets all these demands.
orientation that does not affect the active site of the proteins.
Tsien and co-workers reported a method for site-specific
A variety of chemical transformations have provided powerful protein labeling based on the genetic insertion of a so-called TC
opportunities for the generation of site-specifically oriented motif (CCPGCC) that can selectively form a stable complex with
chips.1 Other immobilization strategies exploit the affinity existing biarsenical-functionalized fluorescent dyes such as FlAsH or
between known interacting pairs such as the polyhistidine-Nickel CrAsH resulting in a extraordinary enhancement of fluorescence
nitriloacetate acid complex or the biotin–avidin interaction.2 How- upon binding.4 This methodology has found numerous applications
ever, the low affinity of the His–NTA complex (KD 10 mM) may result in fluorescence microscopy localization of proteins,5 protein imaging,
in undesired protein detachment3 and the biotin–avidin interaction purification,5,6 labelling and interactions in living cells7 (recently
reviewed in ref. 8). This high affinity interaction (KD 1011 MÀ1
)
proceeds fast under mild conditions, it is highly selective for the
presence of the TC motif and it can conveniently be reversed by
incubation with dithiols. Moreover, since TC-FlAsH binding leads to a
strong enhancement of fluorescence, we reasoned that this method
might be advantageous for use in protein immobilization because it
would make secondary detection steps unnecessary.
To investigate whether a FlAsH-based strategy might be suitable
for protein microarray fabrication, a biarsenical dye derived from
carboxyfluorescein (CrAsH) was chosen for slide-modification.
CrAsH presents a less pH-dependent fluorescence, as compared
to FlAsH and contains an additional carboxy group which enables
a
b
¨
Max-Planck-Institut fur Molekulare Physiologie, Abteilung Chemische Biologie,
Otto-Hahn-Strasse 11, 44227 Dortmund, Germany
¨
Fakultat Chemie, Biologisch-Chemische Mikrostrukturtechnik, Technische
¨
Universitat Dortmund, Otto-Hahn-Strasse 6, 44227 Dortmund, Germany
c Karlsruhe Institute of Technology (KIT), Institute of Biological Interfaces (IBG1),
Hermann-von-Helmholtz Platz, D-76344 Eggenstein-Leopoldshafen, Germany
d Institute of Advanced Chemistry of Catalonia, Department of Biomedicinal
Chemistry, Jordi Girona 18-26, 08034 Barcelona, Spain.
E-mail: gemma.triola@iqac.csic.es; Fax: +34 932-045-904
† Electronic supplementary information (ESI) available. See DOI: 10.1039/
c4cc04120h
‡ These authors contributed equally to this work.
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 12761--12764 | 12761

View Article Online
Communication
ChemComm
with Cy5-labeled streptavidin. No significant signal could be
detected on carboxyfluorescein(4)-modified slides treated with
peptide 3 whereas this same peptide could be detected up to a
concentration of 10 mM on CrAsH-modified slides, thereby indicat-
ing that the immobilization of the peptide indeed occurred due to
specific TC-CrAsH interaction (Fig. 1B and ESI† Fig. S2).
We next investigated the immobilization efficiency at variable
incubation times after spotting. Although a major fluorescence
signal was already detected after 30 min, a significant increase of
the average fluorescence intensities was already remarkable after
one-hour incubation (ESI† Fig. S3). Longer incubation times did
not result in a substantial increase of the fluorescence.
Scheme 1 PEG-CrAsH (1), PEG-Fluorescein (4) and TC-containing
peptides 2–3.
Since recycling of biarsenical-(EDT)2 modified surfaces would
be attractive from an economical point of view, we also investi-
gate the reversibility of peptide attachment. To this end, peptide
3 was spotted and the immobilization efficiency was quantified
by fluorescence detection. Subsequently, the slide was incubated
with a 2 M solution of EDT in DMSO for different incubation
times. Whereas after one-hour the fluorescence intensity could
still be detected, the immobilization could be fully reverted after
4 h, as indicated by the absence of a signal, and the regenerated
slide could be successfully reused for a second step of spotting
achieving similar fluorescence intensities to that in the first
spotting round (Fig. 2A and B).
To test whether our method is suitable for mild immobiliza-
tion of proteins, we next investigated the immobilization of
TC-tagged proteins onto biarsenical-coated slides. For a proof-of
principle study we chose Ras proteins, which play an essential
role in regulating cell signaling and proliferation. To this end, a
TC-containing peptide was synthesized and installed C-terminally
into an H-Ras protein via MIC ligation.11 The resulting semi-
synthetic protein was then spotted onto CrAsH-modified slides
together with a H-Ras protein lacking the TC-tag as a negative
control. After incubation for one hour the immobilized protein
was directly quantified by fluorescence detection of the formed
complex. As expected, the H-Ras protein was efficiently immobilized
the incorporation of the required linker for surface modification.
Hence, a CrAsH derivative conveniently quenched as ethandithiol
(EDT)-complex and bearing a polyethyleneglycol (1) equipped with
an amino group (Scheme 1) was initially prepared in an overall
yield of 17% starting from carboxyfluorescein. Interaction with the
tetracysteine-containing peptides 2 and 3 was studied in solution
confirming that the presence of the linker had no effect on the
fluorescence enhancement upon binding that was in alignment
with the previously reported data (ESI† Fig. S1).9
The immobilization process is depicted in Fig. 1A. To generate
the bisarsenical modified slides, glass slides activated with
N-hydroxysuccinimide (NHS) esters were coated with 1 and sub-
sequently blocked with ethanolamine. As proof-of-principle
peptides 2 and 3 were initially spotted onto the modified slides
at two different concentrations and incubated for one-hour
at room temperature. Direct detection of both immobilized
peptides was successfully performed by determining the levels
of CrAsH fluorescence intensity upon formation of the
complex, thus proving the applicability of the method. The
scanning of the slides was performed with white light to reduce
the reported photobleaching of FlAsH and other derivatives.6,10
Repetitive scannings resulted in signals of similar intensity, thus
indicating the stability of the complex during the measurements.
Additional specific detection of immobilized peptides was achieved
for the biotinylated peptide 3 which was subsequently detected
Fig. 2 (A) CrAsH-coated surfaces can be regenerated and reused after 4 h
incubation of slides with a 2 M solution of EDT in DMSO. (B) Fluorescence
intensities at 100 mM were of the same range in the reused slide compared
to a new one; approximately 30–50% of signal loss could be observed at
lower concentrations. (C) Immobilization of a TC-containing H-Ras protein.
Primary detection through complex formation (yellow) and secondary
detection by an anti-Ras antibody (green). The Protein immobilization can
be performed at different pH rates (pH 5.5 to 10.5) with similar efficiencies.
Fig. 1 (A) Immobilization of peptides 2 and 3 onto CrAsH-modified
surfaces results in fluorescence enhancement that can be monitored.
(B) Peptide 3 in MOPS or Tris buffer was spotted at different concentrations
(0.01–500 mM) onto slides functionalized with PEG-CrAsH (1, 0.5 mM)
or PEG-Fluorescein (4, 0.5 mM) and incubated for 1 h. The detection of
the immobilized biotinylated peptide (3) was performed by incubation with
Cy5-streptavidin (200 nM, 30 min).
12762 | Chem. Commun., 2014, 50, 12761--12764
This journal is ©The Royal Society of Chemistry 2014

View Article Online
ChemComm
Communication
in a concentration-dependent manner while only trace amounts of
the control protein were detected (ESI† Fig. S5A), thus clearly
demonstrating not only the applicability of the approach for the
immobilization of proteins but also the suitability of the complex
formation as a direct detection method. To prove the correct
folding of the immobilized protein we next incubated the slides
with a Cy3-labeled antibody directed against Ras isoforms, which
recognizes the distinct a-helix of the Ras proteins close to the active
site. Therefore binding of this antibody is indicative of correctly
folded H-Ras on the slide.12 Detection by fluorescence scanning
showed efficient immobilization of the TC-tagged recombinant Ras
protein up to a concentration of 25 mM and a minimal signal in the
negative control (signal-to-noise ratio 10 : 1) (ESI† Fig. S5B).
Methods that can be employed in a long range of pH are of
most interest. Therefore, we also investigated the pH tolerance
of the method by immobilizing the semisynthetic H-Ras protein at
different pH rates ranging from 10.5 to 3.5. Although no immobi-
lization was observed at lower pH values, the results indicated that
the immobilization strategy works efficiently at a wide range of pH
values (10.5 to 5.5) (Fig. 2C).
Fig. 3 (A)
A
purified TC-containing PDEd can be immobilized and
detected up to a 10 mM concentration. (B and C) A TC-containing PDEd
was efficiently immobilized from cell lysates without prior purification
(total protein concentration ranges from 1.03 mg mlÀ1 to 0.26 mg mlÀ1).
The immobilized protein could be directly recognized (yellow) as well as
indirectly by an anti-PDEd antibody (purple) and employed to detect
protein–protein interactions with a bodipy-labeled farnesylated N-Ras
protein (blue).
To further investigate this novel immobilization method, a
TC-tag was fused to PDEd a prenyl-binding protein known to
selectively recognize farnesylated proteins, and the recombinant
protein was overexpressed in E. coli. After protein purification,
PDEd could also be successfully immobilized at different concen-
trations with a detection limit of 10 mM as observed by scanning
for the fluorescent complex formed upon interaction (Fig. 3A).
Direct spotting of the TC-tagged protein from the cell lysates
without prior purification would simplify the whole process of
protein microchip fabrication and enable a rapid and general
approach for the generation of multiprotein microarrays. Hence,
we next investigate if a TC-tagged PDEd could be directly immo-
bilized from the cell lysates. With this aim, cell lysates from PDEd
expressing cells were centrifuged and after determining the total
protein concentration were spotted onto CrAsH-modified slides.
As shown in Fig. 3C, cell lysates containing the TC-PDEd could be
efficiently immobilized and directly detected by the formation of
an interacting complex. No detectable signal was observed when
cell lysates lacking the PDEd-TC protein were employed as a
negative control. We also confirmed that the generated protein
microarrays can be employed for measuring protein–protein
interactions. To this end, PDEd-containing slides were incubated
with a PDEd antibody and treated with a fluorescently-labeled
secondary antibody. Subsequent signal quantification indicated
the successful immobilization and the intact folding of all three
proteins bound to the surface.
between peptides or proteins bearing TC-tags and CrAsH-modified
surfaces. This immobilization technique is mild, rapid, it requires
only one-hour incubation, and it is compatible with the sensitive
nature of proteins. Moreover, our approach overcomes important
limitations. Briefly, slide reuse and more importantly, direct
detection of immobilized proteins have always been challenging
issues in the fabrication of protein microarrays. The His–NTA
complex enables slide regeneration but the low affinity of
this interaction may result in undesired protein detachment.
Alternatively, direct detection usually requires label-free methods
that are mostly found only in specialized laboratories. Hence, the
approach described here presents additional important features
compared to the previously described methods, i.e. a high affinity
but reversible binding that can be employed for slide reuse and a
fluorescence enhancement upon immobilization that enables the
direct detection of the immobilized proteins. Finally, this strategy
can be employed for protein immobilization directly from cell
lysates with high efficiency, thus enhancing the practicability
of protein microarray fabrication. Our approach expands the
repertoire of immobilization methods by providing additional
features that can strongly contribute to the application of protein
microarrays in life sciences.
The Dortmund Protein Facility is acknowledged for assis-
tance in cloning, protein expression and purification. We also
thank Christine Nowak for excellent technical help. G.T. would
like to thank Prof. Herbert Waldmann for his generous and
unconditional support.
PDEd is known to specifically bind C-terminal farnesylated
proteins.13 Therefore, as an additional example of protein–protein
interaction studies, a semisynthetic bodipy-labeled farnesylated
N-Ras protein was prepared and incubated with a slide containing
immobilized PDEd. Detection of the interacting Ras was achieved
by fluorescence scanning thus proving that the immobilized PDEd
was functional and that the protein–protein interaction was also
taking place on the chip (Fig. 3C).
Notes and references
1 C. Gauchet, G. R. Labadie and C. D. Poulter, J. Am. Chem. Soc., 2006,
128, 9274; P. Jonkheijm, D. Weinrich, H. Schroder, C. M. Niemeyer
and H. Waldmann, Angew. Chem., Int. Ed., 2008, 47, 9618; P. C. Lin,
S. H. Ueng, M. C. Tseng, J. L. Ko, K. T. Huang, S. C. Yu, A. K. Adak,
Y. J. Chen and C. C. Lin, Angew. Chem., Int. Ed., 2006, 45, 4286;
T. Pauloehrl, G. Delaittre, V. Winkler, A. Welle, M. Bruns,
In summary, we describe here a novel one-step method for
protein immobilization based on the stable complex formation
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 12761--12764 | 12763

View Article Online
Communication
ChemComm
H. G. Borner, A. M. Greiner, M. Bastmeyer and C. Barner-Kowollik, Angew.
Chem., Int. Ed., 2012, 51, 1071; M. B. Soellner, K. A. Dickson, B. L. Nilsson
and R. T. Raines, J. Am. Chem. Soc., 2003, 125, 11790; L. Yi, Y. X. Chen,
P. C. Lin, H. Schroder, C. M. Niemeyer, Y. W. Wu, R. S. Goody, G. Triola
and H. Waldmann, Chem. Commun., 2012, 48, 10829.
2 A. Vaish, V. Silin, M. L. Walker, K. L. Steffens, S. Krueger,
A. A. Yeliseev, K. Gawrisch and D. J. Vanderah, Chem. Commun.,
2013, 49, 2685.
and C. Zimmer, Proc. Natl. Acad. Sci. U. S. A., 2012, 109, 8564;
N. W. Luedtke, R. J. Dexter, D. B. Fried and A. Schepartz, Nat. Chem.
Biol., 2007, 3, 779; A. Rutkowska, C. H. Haering and C. Schultz,
Angew. Chem., Int. Ed., 2011, 50, 12655.
8 A. Pomorski and A. Krezel, ChemBioChem, 2011, 12, 1152;
S. Uchinomiya, A. Ojida and I. Hamachi, Inorg. Chem., 2014,
53, 1816.
9 S. R. Adams and R. Y. Tsien, Nat. Protoc., 2008, 3, 1527.
3 S. Lata, A. Reichel, R. Brock, R. Tampe and J. Piehler, J. Am. Chem. 10 A. K. Bhunia and S. C. Miller, ChemBioChem, 2007, 8, 1642.
Soc., 2005, 127, 10205.
4 B. A. Griffin, S. R. Adams and R. Y. Tsien, Science, 1998, 281, 269.
11 B. Bader, K. Kuhn, D. J. Owen, H. Waldmann, A. Wittinghofer and
J. Kuhlmann, Nature, 2000, 403, 223.
5 S. R. Adams, R. E. Campbell, L. A. Gross, B. R. Martin, G. K. Walkup, 12 I. S. Sigal, G. M. Smith, J. B. Gibbs, F. Jurnak and E. M. Scolnick,
Y. Yao, J. Llopis and R. Y. Tsien, J. Am. Chem. Soc., 2002, 124, 6063.
6 L. Q. Ying and B. P. Branchaud, Bioconjugate Chem., 2011, 22, 987.
Ann. Biomed. Eng., 1986, 14, 88; M. E. Furth, L. J. Davis, B. Fleurdelys
and E. M. Scolnick, J. Virol., 1982, 43, 294.
7 C. Hoffmann, G. Gaietta, A. Zurn, S. R. Adams, S. Terrillon, 13 S. A. Ismail, Y. X. Chen, A. Rusinova, A. Chandra, M. Bierbaum,
M. H. Ellisman, R. Y. Tsien and M. J. Lohse, Nat. Protoc., 2010,
5, 1666; M. Lelek, F. Di Nunzio, R. Henriques, P. Charneau, N. Arhel
L. Gremer, G. Triola, H. Waldmann, P. I. H. Bastiaens and
A. Wittinghofer, Nat. Chem. Biol., 2011, 7, 942.
12764 | Chem. Commun., 2014, 50, 12761--12764
This journal is ©The Royal Society of Chemistry 2014