Chen,[a]and Gerd-Uwe Flechsig*[a, b] Abstract: This communication reports on electrochemical detection of thrombin based on labeling with osmium tetroxide bipyridine [OsO4bipy].. Using t
Trang 12
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
DOI: 10.1002/elan.201700734
Voltammetric Detection of Thrombin by Labeling with
Osmium Tetroxide Bipyridine and Binding with Aptamers
on a Gold Electrode
Sarasi K K Galagedera,[a]Loan Huynh,[a]Falko Wachholz,[b]Martin Jacobsen,[b]
Phensinee Haruehanroengra,[a]Jia Sheng,[a]Alan A Chen,[a]and Gerd-Uwe Flechsig*[a, b]
Abstract: This communication reports on electrochemical
detection of thrombin based on labeling with osmium
tetroxide bipyridine [OsO4(bipy)] Tryptophan amino
acids can be labeled at the C C-double bond, and at least
some tryptophan moieties are accessible for labeling in
thrombin Using the catalytic hydrogen signal from
adsorptive stripping voltammetry performed on hanging
mercury drop electrode, we could detect as little as
1.47 nM [OsO4(bipy)]-modified thrombin We also tested
the binding of [OsO4(bipy)]-modified thrombin with the
classic thrombin binding aptamer (TBA) on gold
electro-des This preliminary study revealed that even after
modification, a major part of the affinity was conserved,
and that the aptamer self-assembled monolayer (SAM) could be regenerated several times Molecular simulations confirm that [OsO4(bipy)]-modified thrombin largely preserves the high binding affinity also of the alternative HD22 aptamer to thrombin, albeit at slightly reduced affinities due to steric hindrance when tryptophans 96 and
237 are labelled Based on these simulations, compensa-tory modifications in the aptamer should result in significantly improved binding with labelled thrombin This combined experimental-computational approach lays the groundwork for the rational design of improved aptamer sensors for analytical applications
Keywords: Hanging mercury drop electrode · Osmium tetroxide-2,2-bipyridyl · Thrombin binding aptamer · Gold electrodes · Adsorptive stripping voltammetry
Thrombin as a central protease of hemostasis represents a
major analytical challenge No methods for direct
detec-tion are available in routine molecular-diagnostic
labora-tories Recent methods for international normalized ratio
(INR) blood testing include indirect techniques such as
enzyme-immuno-assays for the detection of
prothrombin-fragments 1 + 2 (i e side products formed during
biosyn-thesis of thrombin) The CoaguChek INR test system
from Roche is based on such an indirect electrochemical
thrombin assay [1]
The thrombin binding aptamer (TBA) system is a
widely used model to test new approaches for
bioanalyt-ical assays [2, 3] Recent reviews on aptamers for thrombin
or other clinical target molecules revealed that routine
diagnostic tests are still not available [4, 5] Chemical
modification of amino acids is being used extensively to
characterize proteins and their reaction mechanisms in
biological systems Tryptophan can be specifically
modi-fied with an electrochemically active label [OsO4(bipy)] at
its C C-double bond [6–8] This was used by Palecek et al
for electrochemical detection of peptides [9] The osmium
complex, however, despite the C C-double bond, did not
attack histidine At least, no signal could be attributed to
this amino acid The reaction was also studied by CE and
MALDI-TOF MS [10] Fojta et al have applied this
method to investigate accessibility of tryptophan in
streptavidin and avidin It turned out that in the
com-plexes of biotin with streptavidin or avidin, the tryptophan moieties were not accessible for labeling with [OsO4 (bipy)], and that the avidin and streptavidin tetrameric structures would be affected by the labeling [11] Multi-labelling capability is a clear advantage of [OsO4(bipy)] as
a label for both nucleic acids and proteins, and we have used this label to detect DNA hybridization on gold electrodes [12, 13]
Here, we modified thrombin with [OsO4(bipy)] for detection at both mercury drop and gold electrodes The aptamer immobilized on the gold electrode serves as a capture probe for the modified thrombin molecule It is expected that the binding constant is affected by multiple labeling of the analyte, making it weakly bind to the aptamer It is therefore intended to predict an optimized aptamer sequence taking the known TBA-thrombin co-crystal as a starting point
[a] S K K Galagedera, L Huynh, P Haruehanroengra, J Sheng,
A A Chen, G.-U Flechsig Department of Chemistry, University at Albany-SUNY, 1400 Washington Ave, Albany, NY 12222, United States
E-mail: gflechsig@albany.edu [b] F Wachholz, M Jacobsen, G.-U Flechsig Department of Chemistry, University of Rostock, A.-Einstein-Str 3a, D-18055 Rostock, Germany
E-mail: gflechsig@albany.edu
Trang 22
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
Preliminary results with both adsorptive stripping
voltammetry (AdSV) on the classic hanging mercury drop
electrode (HMDE), as well as gold electrode modified
with self-assembled monolayers (SAM) of the TBA are
shown in Figure 1 and 2, respectively The electrocatalytic
signal at 1.25 V using the HMDE allows for trace
thrombin detection, with 1.5 nM limit of detection
(LOD) This result demonstrated that at least some
tryptophan moieties in thrombin are unprotected and
accessible for labeling with [OsO4(bipy)] However,
adsorption on the Hg surface is not specific We also performed preliminary experiments with the TBA as capture probe layer on gold electrodes as illustrated in Scheme 1 As it may be difficult to perform SELEX with osmium tetroxide-labeled aptamers, a very promising approach seems to be computer-aided re-design of the existing aptamers
Figure 2 depicts the detection of [OsO4(bipy)]-labeled thrombin at a TBA-SAM modified gold electrode Three repetitive hybridization-dehybridization cycles are
exhib-Fig 1 A) Electrocatalytic response in adsorptive stripping voltammetric determination of [OsO 4 (bipy)]-modified thrombin on HMDE
in 0.5 M Acetate buffer, pH 4.5; deposition potential 0.8 V, pulse amplitude 50 mV, pulse time 50 ms & sweep rate 20 mV/s B) Calibration plot of [OsO 4 (bpy)]-modified thrombin; The LOD of 1.47 nM has been obtained using the prediction band (95 %) of the calibration plot.
Fig 2 A) Voltammetric response after repetitive 1 hour hybridization, 120 nM Os-labelled thrombin (a, b & c) and de-hybridization (d,
e & f) of [OsO4(bipy)]-modified thrombin at gold electrode with the immobilized TBA in 0.1 M Britton-Robinson buffer at pH 3.8 B) Response obtained upon repetitive hybridization and dehybridization with three different gold disk electrodes, with error bars indicating the standard deviation.
Short Communication
These are not the final page numbers! ÞÞ
Trang 32
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
ited The voltammetric scan has been performed after
hybridization and dehybridization, respectively On gold
electrodes, the catalytic Os-signal cannot be used due to
the large hydrogen background Instead, we use for the
first time the reversible Os(VI/IV) signal to detect a
protein after hybridization on gold The square
wave-voltammetric (SWV) signal appears at 0.42 V
(Fig-ure 2A) Good reproducibility with significant differences
of signals was obtained after hybridization and
dehybrid-ization of the surface-confined TBA with Os-labeled
thrombin using three different gold electrodes, and three
repetitive cycles as illustrated in Figure 2B These
prelimi-nary results indicate, that osmium tetroxide-labeled
thrombin can be voltammetrically detected on gold
electrodes following a hybridization step with immobilized
thiol-linked aptamers The question is now, whether the
common thrombin aptamers are still the optimal fit for
osmium-modified thrombin Fojta et al had observed that
[OsO4(bipy)]-labeled avidin and streptavidin both lost
their ability to bind biotin, and the tetrameric structure of
these proteins was affected On the other hand, the
presence of biotin in the binding site protected the
involved tryptophan moieties from reacting with [OsO4
(bipy)] [11] In that case, the binding site was a bit hidden
and somewhat narrow Thrombin obviously, does not
completely lose its affinity to the aptamer after the
labeling with [OsO4(bipy)] However, we expected at least
a slight decline in binding affinity
Since the HD22 sequence has been identified as a
better aptamer for thrombin than the classical TBA (KD
smaller by a factor of 20 or more [14]), we performed
simulations on the new HD22 sequence to investigate the effect of osmium-labeling upon the binding free energy Using molecular simulations, we were able to identify which specific [OsO4(bipy)] label sites most negatively affect the binding of the thrombin-HD22 complex In prior work [15], we have demonstrated that this method is capable of folding non-canonical RNAs to sub-angstrom resolution [16], as well as achieve quantitative agreement with mass-spectrometry binding experiments [17] Using all-atom molecular dynamics simulation, the free energy for dissociating the thrombin-HD22 complex was calcu-lated both with and without the [OsO4(bipy)] labels at different sites The simulations found that only labels at tryptophans 96 and 237 interfered with aptamer binding, confirming the partial loss of binding affinity (Table 1) It
is significant that even if [OsO4(bipy)] labels at the thrombin binding interface occur at both sites simulta-neously, a majority of the HD22 binding affinity is retained, thus explaining why even a fully modified thrombin is electrochemically detectable at nanomolar concentrations This agreement hints that it may be possible to recover the lost binding affinity via compensa-tory aptamer mutations identified via a combined in-silico and experimental approach, which is the subject of ongoing studies Such computer-optimized aptamer design may be a more general approach to obtain strongly and specifically binding aptamers, even in cases of labelled targets and aptamers, where SELEX cannot be used
Experimental
Thrombin from bovine plasma, osmium tetroxide, 2,2’-bipyridyl and Tris(hydroxymethyl) amino-methane were purchased from Sigma All other chemicals were of analytical grade Solutions were prepared using mega-pure water The used thiol-linked aptamer was based on the TBA (the exosite I-binding aptamer), was delivered
by FRIZ Biochem GmbH (Munich, Germany), and had the following sequence including a spacer of 10 adenines and a thiol-linker with three dithian groups: 5’-GGTTGGTGTGGTTGG aaaaaaaaaa[Dithio]3-3’ First, labelling of 6 mM thrombin with 4 mM [OsO4 (bipy)] solution was done in an unstirred 100 mM Tris buffer (pH adjusted with HCl to 7.0) at room temperature for 2 hours Then the solutions containing labelled thrombin were dialyzed using “Slide-A-Lyzer Mini Dialy-sis units, with a first buffer exchange performed after
3 hours and a second for overnight (19 h) dialysis The
Scheme 1 [OsO 4 (bipy)]-modified thrombin bound to
surface-confined aptamer containing spacer nucleotides (green color)
coupled to a thiol-linker “S” Structure reproduced from [18] with
permission of the International Union of Crystallography
(CC-BY License), DOI: 10.1107/S0907444913022269.
Table 1 Thrombin/DNA aptamer binding free energies.
System [a]
DG bind (kcal/mol)
[a] (T) Thrombin and (A) aptamer (OT) [OsO 4 (bipy)]-con-jugated thrombin Subscripts relax indicates without crystal-lographic restraints.
Trang 42
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
dialysis step was followed by buffer exchange (3 times)
with 10 K filter tubes to further remove any free [OsO4
(bipy)] The electrochemical response of the labelled
thrombin was measured by means of adsorptive stripping
differential pulse (DPV) voltammetry at a hanging
mercury drop electrode using a 663 VA stand controlled
by an Autolab PGSTAT204 potentiostat and software
NOVA 1.11 (Metrohm-Autolab, Switzerland) A
three-electrode system was used for the measurements with a
glassy carbon electrode as the counter electrode and Ag/
AgCl in 3 M KCl solution as the reference electrode
Detection of labelled thrombin on gold electrodes has
been performed by SWV after repetitive 1 h hybridization
and de-hybridization (1 min at 50 8C) of osmium
tetrox-ide-labelled thrombin at a gold electrode with
immobi-lized aptamer in 0.1 M Britton-Robinson buffer at pH 3.8
The simulations are based on the HD22 sequence (the
exosite II-binding aptamer):
5’-GTCCGTGGTAGGG-CAGGTTGGGGTGAC-3’ as illustrated in Figure 3
NPT simulations at 300 K and 1 atm were conducted
using GROMACS 4.5.5, and included 11,800 TIP3P water
molecules as well as 0.38 M NaCl The AMBER-99
parameters were used, including the Chen-Garcia [15]
modifications for nucleic acids Dissociation free energies
were calculated using virtual replica-exchange umbrella
sampling [16], using 29 umbrellas spanning intermolecular
distances of 2.4–3.8 nm with a spring constant ku, of
2000 kJ/mol/nm2
Acknowledgement
The authors are grateful to SUNY Albany for start-up support and to Deutsche Forschungsgemeinschaft (DFG Heisenberg Fellowship FL 384/7-1, FL 384/8-1, and FL 384/11-1) for financial support This work used the Extreme Science and Engineering Discovery Environ-ment (XSEDE), which is supported by National Science Foundation grant number ACI-1548562
References
[1] L R Bereznicki, S L Jackson, G M Peterson, E C Jeffrey,
K A Marsden, D M Jupe, J Clin Pathol 2007, 60, 311– 315.
[2] L C Bock, L C Griffin, J A Latham, E H Vermaas, J J Toole, Nature 1992, 355, 564–566.
[3] B Deng, Y Lin, C Wang, F Li, Z Wang, H Zhang, X.-F Li,
X C Le, Anal Chim Acta 2014, 837, 1–15.
[4] G S Baird, Am J Clin Pathol 2010, 134, 529–531.
[5] A V Lakhin, V Z Tarantul, L V Gening, Acta Naturae
2013, 5, 34–43.
[6] J S Deetz, E J Behrman, J Org Chem 1980, 45, 135–140 [7] J S Deetz, E J Behrman, Int J Pept Protein Res 1981, 17, 495.
[8] M Emerman, E J Behrman, J Histochem Cytochem 1982,
30, 395–397.
[9] S Billova, R Kizek, E Palecek, Bioelectrochemistry 2002,
56, 63–66.
[10] O Sedo, S Billova, E M Pena-Mendez, E Palecek, J Havel, Anal Chim Acta 2004, 515, 261–269.
[11] M Fojta, S Billova, L Havran, H Pivonkova, H Cernocka,
P Horakova, E Palecek, Anal Chem 2008, 80, 4598–4605 [12] G.-U Flechsig, T Reske, Anal Chem 2007, 79, 2125–2130 [13] K Biala, A Sedova, G.-U Flechsig, ACS Appl Mater Interfaces 2015, 7, 19948–19959.
[14] D M Tasset, M F Kubik, W Steiner, J Mol Biol 1997, 272, 688–698.
[15] W Stephenson, P N Asare-Okai, A A Chen, S Keller, R Santiago, S A Tenenbaum, A E Garcia, D Fabris, P T X.
Li, J Am Chem Soc 2013, 135, 5602–5611.
[16] A A Chen, A E Garcı´a, Proc Natl Acad Sci USA 2013,
110, 16820–16825.
[17] S Rauscher, C Neale, R Pomes, J Chem Theory Comput.
2009, 5, 2640–2662.
[18] I Russo Krauss, A Pica, A Merlino, L Mazzarellaa, F Sica, Acta Crystallogr 2013, D69, 2403–2411.
Received: November 10, 2017 Accepted: December 20, 2017 Published online on &&&, &&&&
Fig 3 Snapshot of (orange ribbons and multicolored nucleotide
bases) HD22-27mer DNA aptamer interacting with (grey)
thrombin, with modeled (yellow) osmium conjugated to
trypto-phan residues 96 and 237.
Short Communication
These are not the final page numbers! ÞÞ
Trang 54
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
S K K Galagedera, L Huynh, F Wachholz, M Jacobsen, P Harue-hanroengra, J Sheng, A A Chen, G.-U Flechsig*
1 – 5 Voltammetric Detection of Thrombin by Labeling with Osmium Tetroxide Bipyridine and Binding with Aptamers on a Gold Electrode