Practical fluorescence detection of acrolein in human plasma via a two-step tethering approach

Article abstract of DOI:10.1039/c4cc02578d

Acrolein, a cytotoxic α,β-unsaturated aldehyde and disease biomarker, was determined in plasma by means of a novel tethering strategy using Michael addition of the compound to a fluorescent dye, followed by immobilization of the product on microbeads via

Full text of DOI:10.1039/c4cc02578d

Showcasing research from the laboratory of
Prof. Tetsuo Nagano and Prof. Yasuteru Urano, Graduate School
of Pharmaceutical Sciences, The University of Tokyo, Japan
As featured in:
Practical fluorescence detection of acrolein in human plasma via
a two-step tethering approach
A simple and high-throughput method to measure the
concentration of acrolein, a cytotoxic α,β-unsaturated aldehyde
and disease biomarker, in human plasma was developed. The
method relies on a novel tethering approach using hydrazine
microbeads and a thiol-containing fluorophore.
Cover illustration was drawn by Ryo Fujikake.
See Yasuteru Urano,
Tetsuo Nagano et al.,
Chem. Commun., 2014, 50, 14946.
www.rsc.org/chemcomm
Registered charity number: 207890

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Practical fluorescence detection of acrolein in
human plasma via a two-step tethering approach†
Cite this: Chem. Commun., 2014,
50, 14946
Masataka Togashi,a Takuya Terai,a Hirotatsu Kojima,b Kenjiro Hanaoka,a
Kazuei Igarashi,c Yasunobu Hirata,d Yasuteru Urano*ae and Tetsuo Nagano*b
Received 8th April 2014,
Accepted 9th June 2014
DOI: 10.1039/c4cc02578d
www.rsc.org/chemcomm
Acrolein, a cytotoxic a,b-unsaturated aldehyde and disease biomarker, reaction with 3-aminophenol.6 In this method, the plasma of
was determined in plasma by means of a novel tethering strategy using patients is reacted with 3-aminophenol under acidic conditions, and
Michael addition of the compound to a fluorescent dye, followed by the fluorescent product, 7-hydroxyquinoline, is quantified. In most
immobilization of the product on microbeads via the aldehyde moiety. cases,6a,b detection after HPLC separation is necessary to obtain a
Elevation of blood acrolein was detected in mice treated with an sufficient S/N ratio, because (i) the background fluorescence of
anticancer agent cyclophosphamide, which releases acrolein upon plasma is high upon excitation at 350–400 nm (the excitation
activation. This method should be suitable for high-throughput wavelength of 7-hydroxyquinoline), and (ii) the fluorescence
diagnostic and clinical application.
quantum yield of 7-hydroxyquinoline is low. However, HPLC is
unsuitable for high-throughput assay of multiple samples. In
Acrolein (CH2CHCHO) is a strong electrophile that disrupts addition, the reaction conditions of the conventional assay are
important cellular functions by covalently modifying biomolecules, quite severe (1 N HCl aq., reflux), which is also inconvenient for
including proteins and DNA.1 Formerly, it was thought that practical measurement on site. Acrolein–protein or –DNA adducts
acrolein is generated from unsaturated lipids and reactive oxygen can be detected using monoclonal antibodies,2a,7 but the method
species (ROS),2 but recent reports indicate that acrolein is mainly is costly and requires complicated procedures. Moreover, there is
produced from polyamines, such as spermine, by polyamine a delay of several hours between the production of acrolein and
oxidases (PAOs).3 Acrolein and its protein conjugate are of interest the formation of acrolein adducts. Therefore, in order to overcome
as disease markers. For example, they were reported to be increased in the limitations of current methods, we have developed a novel,
plasma of patients with chronic renal failure4 or brain infarction.3b,5 inexpensive, and practical fluorescence-based method, using a
Therefore, quantification of acrolein in biological samples such as two-step tethering strategy, for detection of acrolein under mild
plasma or urine is potentially of diagnostic value. However, con- conditions without need for HPLC.
ventional methods for detection of acrolein are not suitable for
For fluorescence detection of acrolein at the low concentrations
high-throughput assay, and hence are not applicable for diagnosis. that exist in biological samples, including plasma and urine, we
A practical detection method for pathological concentrations of free considered that it would be essential to remove background signals
acrolein (sub to low micromolar range)3a,4 is expected to find many by isolation of the target compound. As the use of HPLC is
applications in the medical and biological fields.
undesirable for the reasons mentioned above, we considered that
The most widely used method for detecting free acrolein in a two-step tethering procedure to a solid phase, taking advantage of
biological samples is fluorescence analysis based on the Skraup the characteristic a,b-unsaturated carbonyl reactivity, would be an
effective alternative. Our design strategy is summarized in Fig. 1.
First, a fluorescent compound with a thiol functional group (I) is
added to acrolein to yield the Michael addition product (II). Second,
a Graduate School of Pharmaceutical Sciences, The University of Tokyo,
7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan
the mixture is incubated with plastic microbeads bearing hydrazine
functional groups on their surface, resulting in concentration-
dependent covalent attachment of the fluorophore-bearing addition
product to the microbeads (III). Unbound dye and biological compo-
nents (proteins, sugars, nucleic acids, etc.) can then be readily removed
simply by washing the microbeads, as already well established.8
We selected tetramethylcarboxyrhodamine (TAMRA) as the
fluorophore because it emits strong red fluorescence and has
b Open Innovation Center for Drug Discovery, The University of Tokyo, 7-3-1 Hongo,
Bunkyo-ku, Tokyo 113-0033, Japan. E-mail: tlong@mol.f.u-tokyo.ac.jp
c Amine Pharma Research Institute, 1-8-15 Inohana, Chuo-ku, Chiba 260-0856, Japan
d Department of Advanced Clinical Science and Therapeutics, Graduate School of
Medicine, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8655, Japan
e Graduate School of Medicine, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku,
Tokyo 113-0033, Japan. E-mail: uranokun@m.u-tokyo.ac.jp
† Electronic supplementary information (ESI) available: Table S1, Fig. S1, and
experimental section. See DOI: 10.1039/c4cc02578d
14946 | Chem. Commun., 2014, 50, 14946--14948
This journal is ©The Royal Society of Chemistry 2014

View Article Online
Communication
ChemComm
Fig. 1 Schematic representation of the strategy for detection of acrolein
in biological samples and the structure of the thiol-containing fluorescent
compound used in this work, TAMRA-C2-SH.
high solubility in water, and synthesized TAMRA-C2-SH (Fig. 1). The
synthetic procedures and photophysical properties of TAMRA-C2-SH
are described in the ESI.† As microbeads, we selected a PEG-
attached resin (TentaGel) with hydrazine functional groups, because
it has excellent swelling properties in aqueous solution. Then we
optimized the reaction conditions of acrolein, TAMRA-C2-SH,
and microbeads. The optimum conditions were as follows:
TAMRA-C2-SH and acrolein in sodium phosphate buffer (pH 7.4)
were first incubated at 40 1C for 1 h, then the pH was adjusted to 3.5
with sodium phosphate buffer (pH 1.4). Second, TentaGel-NH-NH2
microbeads were added to the mixture, followed by incubation
at 40 1C for 3 h and brief washing with DMF and sodium
phosphate buffer (pH 7.4). Fig. 2 shows confocal fluorescence
microscopic images of microbeads after the two-step tethering
process. The beads incubated in the acrolein-containing solution
emitted strong fluorescence, while those incubated in the acrolein-
free solution emitted almost no fluorescence. These results showed
that our tethering strategy, in which acrolein serves as a linker
between the microbeads and TAMRA-C2-SH, is effective, and that
non-specific adsorption of the fluorophore on the beads is negligible.
Therefore, we next investigated the selectivity of our method. As
shown in Fig. 3, the method was very selective for a,b-unsaturated
Fig. 3 Fluorescence assays of various a,b-unsaturated carbonyl com-
pounds, propionaldehyde, N-acetylcysteine and glutathione (5 mM each).
(A) Structures of the analytes used. (B) Fluorescence intensity (ex./em. =
554/570–700 nm) of the microbeads, which was recorded using confocal
fluorescence microscopy; n.d.: not detected (F.I. o 0.1).
carbonyl compounds, and the highest reactivity was observed for
acrolein, presumably due to its small size. We also carried out HPLC
analysis to confirm that the reaction proceeded as we expected (see
the ESI†). Since the biological concentration of acrolein is higher
than those of other a,b-unsaturated carbonyl compounds such as
4-hydroxynonenal (HNE),5 we consider that our method has sufficient
specificity for acrolein for practical purposes in a biological context.
We then applied this detection method to human plasma.
Plasma itself has absorbance in the UV-visible region, and emits
fluorescence in the range of 300–500 nm. Therefore, the conven-
tional 7-hydroxyquinoline method gives poor results in plasma.
However, with our method, the background signal is easily elimi-
nated by washing the microbeads. As can be seen in Fig. 4A, as
little as 1 mM acrolein could be reliably detected with our method,
and the fluorescence increased linearly as a function of added
acrolein concentration in plasma.9 In contrast, 1 mM acrolein
could not be detected after derivatization with 3-aminophenol
Fig. 4 (A) Fluorescence assays of acrolein (0, 1, 2, 3 mM) in plasma with
TAMRA-C2-SH and TentaGel-NH-NH2 microbeads. Fluorescence intensity
(ex./em. = 554/575 nm) was recorded after washing the microbeads.
(B) Fluorescence assays in plasma using the conventional 7-hydroxyquinoline
method. Data are shown as mean Æ SD (n = 3).
Fig. 2 Confocal fluorescence microscopic images of microbeads incubated
with acrolein-containing solution (acrolein: 10 mM, left) and acrolein-free solu-
tion (right). Upper panel; fluorescence images, lower panel; bright-field images.
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 14946--14948 | 14947

View Article Online
ChemComm
Communication
in the absence of HPLC separation (Fig. 4B). With the conven- Nevertheless, high selectivity and sensitivity are obtained because
tional method, quantitative assays in this concentration range unsaturated aldehyde works as a linker to tether the fluoro-
are not possible because of the intrinsic background fluores- phore on microbeads. Although this is not the first application of
cence of plasma. The calculated detection limit of our method a tethering strategy, only very limited examples have so far been
was 0.54 mM, which is one order of magnitude smaller than that reported for a few other biological molecules.8 Here, we used plastic
of the conventional method (5.4 mM) without HPLC separation. microbeads as a solid phase because of their ready commercial
Finally, we applied the method for detection of acrolein in availability, but it should be straightforward to employ other
plasma of mice treated with cyclophosphamide (CPA). CPA and its materials, such as multi-well plates or glass slides functionalized
analog ifosfamide are well-known anticancer agents.10 However, with hydrazine for convenient high-throughput application. A poten-
these drugs are metabolically activated by the liver P450 system, and tial drawback of this strategy may be the lack of strict discrimination
then decomposed into phosphoramide mustards and acrolein.10a,e of acrolein from other a,b-unsaturated carbonyl compounds, but we
Although a good method to quantify activated CPA has been believe this is not a bar to practical application for diagnostic and
reported with the aid of HPLC,11 it would be preferable to monitor clinical purposes, as discussed above. Some of the present authors
the total acrolein concentration in the blood of CPA-treated are currently validating this method for biological and clinical use,13
patients because acrolein is likely to be responsible for the adverse with the expectation that it will improve patient safety. It should also
effect of hemorrhagic cystitis observed in some patients.10b
Therefore, to demonstrate that our method has practical value
be useful in studies of the pathological mechanisms of acrolein.
This work was in part supported by KAKENHI (Grant no. 23651231
and is sufficiently sensitive for clinical use, we used it to determine and 25104506 to T.T., 22000006 to T.N., 24689003 and 24659042 to
the acrolein concentration in plasma of CPA-treated and non-treated K.H., and 20117003 and 23249004 to Y.U.). T.T. was also supported by
mice. CPA (0.4 mg per body) in saline was injected intravenously, Astellas Foundation for Research on Metabolic Disorders.
and then 30 min later, the mice were sacrificed and the plasma was
collected. As shown in Fig. 5, the acrolein concentration in plasma
was markedly increased in CPA-treated mice. The results clearly
demonstrate that our method can detect acrolein generated in vivo.
Notes and references
1 (a) A. Furuhata, T. Ishii, S. Kumazawa, T. Yamada, T. Nakayama and
K. Uchida, J. Biol. Chem., 2003, 278, 48658; (b) Y. Kawai, A. Furuhata,
S. Toyokuni, S. Aratani and K. Uchida, J. Biol. Chem., 2003, 278, 50346.
In conclusion, we have developed a novel, inexpensive, and
practical fluorescence-based method for detecting acrolein in
human plasma. Our method relies on the characteristic reactivity of
a,b-unsaturated aldehydes to selectively tether a fluorophore
(TAMRA-C2-SH) to a solid phase (TentaGel microbeads), using
the analyte as a linker. It enables sensitive (submicromolar) and
practical detection of acrolein in plasma without the need for HPLC
separation. We confirmed its utility by detecting the elevation of
acrolein in plasma of CPA-treated mice, as a model of a serious side
effect in human cancer patients.
Compared with most fluorescent probes for specific bio-
molecules,12 which are elaborately designed conjugates of
fluorophore and receptor/substrate moieties, the detection strategy
used in this work is quite different, in that reaction with the target
does not alter the fluorescence properties of the probe itself.
2 (a) K. Uchida, M. Kanematsu, Y. Morimitsu, T. Osawa and N. Noguchi,
J. Biol. Chem., 1998, 273, 16058; (b) K. Uchida, Trends Cardiovasc. Med.,
1999, 9, 109.
3 (a) K. Igarashi and K. Kashiwagi, Mol. Nutr. Food Res., 2011, 55, 1332;
(b) R. Saiki, K. Nishimura, I. Ishii, T. Omura, S. Okuyama, K. Kashiwagi
and K. Igarashi, Stroke, 2009, 40, 3356.
4 K. Sakata, K. Kashiwagi, S. Sharmin, S. Ueda, Y. Irie, N. Murotani
and K. Igarashi, Biochem. Biophys. Res. Commun., 2003, 305, 143.
5 R. Saiki, H. Park, I. Ishii, M. Yoshida, K. Nishimura, T. Toida,
H. Tatsukara, S. Kojima, Y. Ikeguchi, A. E. Pegg, K. Kashiwagi and
K. Igarashi, Biochem. Biophys. Res. Commun., 2011, 404, 1044.
6 (a) R. A. Alarcon, Anal. Chem., 1968, 40, 1704; (b) F. Bohnenstengel,
M. Eichelbaum, E. Golbs and H. K. Kroemer, J. Chromatogr. B:
Biomed. Sci. Appl., 1997, 692, 163; (c) D. Masurel, P. J. Houghton,
C. L. Young and I. W. Wainer, Cancer Res., 1990, 50, 252.
7 J. Pan, B. Awoyemi, Z. Xuan, P. Vohra, H.-T. Wang, M. Dyba, E. Greenspan,
Y. Fu, K. Creswell, L. Zhang, D. Berry, M.-S. Tang and F.-L. Chung,
Chem. Res. Toxicol., 2012, 25, 2788.
8 (a) M. D. Sørensen, R. Martins and O. Hindsgaul, Angew. Chem.,
Int. Ed., 2007, 46, 2403; (b) S. Akita, N. Umezawa and T. Higuchi,
Org. Lett., 2005, 7, 5565; (c) E. E. Carlson and B. F. Cravatt,
Nat. Methods, 2007, 4, 429.
9 Native plasma contains large amounts of thiol-containing proteins
that react with acrolein. To exclude consumption of acrolein by
these proteins, we treated plasma with a spin column for protein
removal before subjecting it to the assay (see ESI† for details).
10 (a) P. J. Cox, Biochem. Pharmacol., 1979, 28, 2045; (b) D. E. Bjorling,
J. E. Elkahwaji, W. Bushman, L. M. Janda, K. Boldon, W. J. Hopkins
and Z. Y. Wang, BJU Int., 2007, 99, 1523; (c) S. Takamoto, N. Sakura,
A. Namera and M. Yashiki, J. Chromatogr. B: Anal. Technol. Biomed.
Life Sci., 2004, 806, 59; (d) M. Gu¨nther, E. Wagner and M. Ogris,
J. Cell. Mol. Med., 2008, 12, 2704; (e) T. Wagner, D. Heydrich, T. Jork,
G. Voelcker and H. J. Hohorst, J. Cancer Res. Clin. Oncol., 1981, 100, 95.
11 L. Griskevicius, L. Meurling and M. Hassan, Ther. Drug Monit., 2002, 24, 405.
12 (a) T. Terai and T. Nagano, Curr. Opin. Chem. Biol., 2008, 12, 515;
(b) T. Terai and T. Nagano, Pflugers Arch.: Eur. J. Physiol., 2013,
465, 347; (c) The Molecular Probes Handbook, ed. I. Johnson and
M. T. Z. Spence, Life Technologies, Carlsbad, 11th edn, 2010.
13 T. Uemura, Y. Tanaka, K. Higashi, D. Miyamori, T. Takasaka,
T. Nagano, T. Toida, K. Yoshimoto, K. Igarashi and H. Ikegaya,
Toxicology, 2013, 310, 1.
Fig. 5 Determination of acrolein in plasma of CPA-treated and non-
treated mice. Data are shown as mean
Æ
SD (n = 3). * indicates
p o 0.05 (Student’s t-test).
14948 | Chem. Commun., 2014, 50, 14946--14948
This journal is ©The Royal Society of Chemistry 2014