A selective and sensitive fluorescent albumin probe for the determination of urinary albumin

Article abstract of DOI:10.1039/c4cc04236k

In this communication, we report a simple albumin probe based on a fluorescent molecular rotor for the detection of trace albumin levels in urine. In the presence of albumin, the probe exhibits remarkable 400-fold fluorescence enhancement with high selectivity and sensitivity. The probe was successfully applied in the quantitative detection of urinary albumin.

Full text of DOI:10.1039/c4cc04236k

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
A selective and sensitive fluorescent albumin
probe for the determination of urinary albumin†
Cite this: Chem. Commun., 2014,
50, 11507
Ying-Yi Wu,a Wan-Ting Yu,a Tai-Cheng Hou,a Tao-Kai Liu,a Chi-Ling Huang,a
I-Chia Chenab and Kui-Thong Tan*ab
Received 3rd June 2014,
Accepted 5th August 2014
DOI: 10.1039/c4cc04236k
www.rsc.org/chemcomm
In this communication, we report a simple albumin probe based on multistep operation procedures have limited their application
a fluorescent molecular rotor for the detection of trace albumin for the routine screening of urine samples.6 Although colori-
levels in urine. In the presence of albumin, the probe exhibits metric dyes such as bromocresol purple have been used exten-
remarkable 400-fold fluorescence enhancement with high selec- sively in clinical laboratories to quantify albumin in blood
tivity and sensitivity. The probe was successfully applied in the plasma, they are not suitable to determine trace urinary
quantitative detection of urinary albumin.
albumin due to the low sensitivity and selectivity of the dyes.7
On the other hand, fluorescent probes have several advantages
Human serum albumin (HSA), with a normal concentration over immunoassays and absorption techniques, particularly in
range between 35 and 50 g LÀ1 in blood plasma, is the most the gain of sensitivity, lower cost of production and technical
abundant protein in the circulatory systems. In contrast, levels of simplicity. Many fluorescent probes, especially those based on
albumin in urine are normally below 30 mg LÀ1, as the kidneys environment-sensitive fluorophores have been reported to
can prevent useful substances such as albumin and other detect HSA by binding to the HSA hydrophobic domain.8
proteins from entering urine. However, when the filtering ability However, many of them either do not show high selectivity
of the kidneys is damaged, elevated urinary albumin levels or toward albumin or require long synthetic steps to prepare the
albuminuria can occur.1 Several studies have identified albumi- fluorescent probes.
nuria as a key indicator for the intensified treatment of diabetes
mellitus and for the early diagnosis of renal disease.2 Further-
more, albuminuria is also one of the most important predictors
of cardiovascular disease in non-diabetic individuals.3 Therefore,
the quantitative detection of albumin in biological fluids espe-
cially in urine, has gained much importance in diagnosis and
preventive medicine during the past few decades.4
Although various methods have been utilised successfully
for the detection of albumin, including immunoassays, capil-
lary electrophoresis, colorimetric and fluorescent probes, these
methods are inadequate for the practical routine analysis of
urine samples to quantify low concentrations of albumin,
particularly in the 30 mg LÀ1 region that covers the usual cutoff
limits between normal and increased albumin excretion.5
While immunoassays are sufficiently responsive and selective
for detecting albuminuria, the high costs of antibodies and
a Department of Chemistry, National Tsing Hua University, 101 Sec. 2, Kuang Fu Rd,
Scheme 1 (a) Schematic illustration of the fluorescent probe AL-1 for
HSA detection. In the absence of HSA, AL-1 shows weak fluorescence due
to unrestricted torsional rotation. Binding of AL-1 to HSA imposes
restricted bond rotation to the probe and lead to a dramatic fluorescence
increase. The images show 2 mM AL-1 in a cuvette before (left) and after
addition of 20 mM HSA (right) under excitation using a UV lamp (365 nm).
(b) Synthesis of AL-1.
Hsinchu 30013, Taiwan, Republic of China. E-mail: kttan@mx.nthu.edu.tw
b Frontier Research Center on Fundamental and Applied Sciences of Matters,
National Tsing Hua University, 101 Sec. 2, Kuang Fu Rd, Hsinchu 30013, Taiwan,
Republic of China
† Electronic supplementary information (ESI) available. See DOI: 10.1039/
c4cc04236k
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 11507--11510 | 11507

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
View Article Online
Communication
ChemComm
In this communication, we report on the use of a simple and
novel fluorescent probe AL-1 for the determination of albumin
levels in urine samples (Scheme 1a). AL-1 belongs to a class of
fluorescent dyes called molecular rotors. Previously, fluorescent
molecular rotors have been used to investigate the viscosity of
blood plasma,9 plasma membranes and cytoplasm of living
cells10 as well as for the rapid detection of biomolecular
interactions.11 These molecules are characterized by a charge-
transfer-excited singlet state which can be rapidly deactivated
through intramolecular rotation about the donor–acceptor
bond. The fluorescence of molecular rotors is mainly sensitive
to the viscosity of the environment and to a smaller extent,
solvent polarity. Thus, in a highly constrained environment
such as glycerol, a large fluorescence increase can be observed
due to the restricted bond rotation (Fig. S1, ESI†). As HSA
contains several hydrophobic binding sites for many hydro-
phobic drugs, we expect that the binding of hydrophobic AL-1
to HSA can impose restricted bond rotation on the probe and
lead to a dramatic fluorescence increase. AL-1 can be prepared
easily in two synthetic steps, requiring only one purification
step with a total yield of 80% (Scheme 1b). AL-1 is water-soluble
and does not form aggregates in aqueous buffer at concentra-
tions under 25 mM (Fig. S2, ESI†). The high solubility of the
hydrophobic AL-1 molecule in pH 7.4 PBS buffer may be due to
the protonation of nitrogen at the bicyclic ring.
In aqueous buffer, AL-1 shows extremely weak fluorescence
(e = 30 531 MÀ1 cmÀ1 and f = 0.0022) as would be expected for
Fig. 1 (a) Fluorescence spectra of 2 mM AL-1 in the presence of 12.5 mM
the probe for completely unrestricted torsional rotation. In the HSA. The fluorescence spectrum of free AL-1 was magnified 10-times.
The inset shows the fluorescent titration curve of 2 mM AL-1 with increasing
presence of HSA, the fluorescence of AL-1 increases signifi-
HSA concentration. lex = 440 nm, lem = 490 nm. (b) Relative fluorescence
cantly (e = 52 342 MÀ1 cmÀ1 and f = 0.5572) and can be
intensity of 2 mM AL-1 with different proteins. The concentrations of HSA are
observed directly under a handheld UV lamp (Scheme 1a).
The binding of AL-1 to HSA induces fluorescence enhancement
at 2 mM and 0.5 mM while all others are at 20 mM except BSA which was
tested at 2 mM. HCA = human carbonic anhydrase, RNase A = Ribonuclease
of about 400-fold which is higher than most of the other HSA A, ConA = Concanavalin.
fluorescent probes reported to date (Fig. 1a). In comparison, a
previously reported fluorescent albumin probe based on a
different type of fluorescent molecular rotor exhibited only increased hydrophobic interaction and better positioning of the
moderate fluorescence enhancement of around 36-fold in the probe in the binding pocket of HSA, hence contributing signifi-
presence of albumin.12
cantly to the high fluorescence increase.
When HSA was added in increasing concentrations, the probe
Besides the superior sensitivity and an extremely high
showed a concentration dependent fluorescence enhancement fluorescence activation ratio, AL-1 also shows excellent selec-
(Fig. 1a, inset). We also observed a slight blue-shift in the tivity toward HSA when tested against a collection of 11
absorption and emission spectra of AL-1 upon binding to HSA. different proteins (Fig. 1b). In the presence of 0.5 mM and
In the absence of HSA, AL-1 displays an absorption maximum at 2 mM HSA, AL-1 exhibits fluorescence enhancement of around
464 nm and an emission maximum at 500 nm, as compared to a 30- and 100-fold, respectively, while other proteins displayed
maximum at 456 nm and 490 nm, respectively, in the presence only a marginal fluorescence increase even with a 10-fold
of HSA (Fig. S3, ESI†). The limit of detection (LOD) for AL-1 to increase in concentration at 20 mM. It is also remarkable to
detect HSA was determined to be as low as 6 nM (0.4 mg LÀ1
)
note that AL-1 is highly selective between HSA and BSA, as
and the fluorescence response was linear in the range of 0–1 mM the three-dimensional structures of albumins from different
(Fig. S4, ESI†). In an attempt to optimize the fluorescence species are assumed to be very similar based on their highly
enhancement for the HSA detection, we synthesized several homologous primary sequences.13
AL-1 derivatives and found that the hydrophobic bicyclic ring
A Job’s plot analysis was performed to determine the stoi-
and the ester moiety of AL-1 are important to induce large chiometry of the complex formed between AL-1 and HSA
fluorescence enhancement, as derivatives without these moieties (Fig. 2a). The fluorescence intensity of AL-1 peaked at 1 : 1 mole
generally gave lower fluorescence turn-on ratios in the presence fraction of AL-1 to HSA which indicates that the probe binds
of albumin (Fig. S5, ESI†). We believe that the ester functional mainly at one site of the protein. A relatively high HSA binding
group might also participate in the bonding with HSA, leading to affinity with the dissociation constant (Kd) of 1.77 Æ 0.65 mM
11508 | Chem. Commun., 2014, 50, 11507--11510
This journal is ©The Royal Society of Chemistry 2014

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
View Article Online
ChemComm
Communication
Fig. 2 (a) Job plot analysis. AL-1 was mixed with HSA at different ratios in
PBS buffer (pH = 7.4) while maintaining the total concentration at 10 mM.
(b) Displacement of AL-1 from the HSA–AL-1 complex by the addition of
site-specific drugs, warfarin, ibuprofen and digitoxin, respectively.
Fig. 3 Fluorescence response of 10 mM AL-1 in a urine sample spiked with
was obtained, when a one-site binding model was fitted to the various concentrations of HSA (0.67 mg LÀ1 to 67 mg LÀ1). y = 29x + 429,
R2 = 0.98. The albumin level was determined to be 14.7 mg LÀ1
.
titration of AL-1 with increasing concentrations of HSA. The
characterization of the drug binding sites on HSA is an impor-
tant step to understand its binding properties for predicting
potential drug interactions, X-ray crystallography studies have The urine samples were also tested by immunoassay for the
identified several drug binding sites on HSA with subdomains validation of the albumin levels determined by AL-1, which
IIA and IIIA being identified as its primary drug binding sites.14 were determined to be 17.3 mg LÀ1, 3.2 mg LÀ1, and 9.7 mg LÀ1
,
To verify the binding site of AL-1, a competitive assay to respectively (Fig. S7, ESI†). As the background fluorescence
displace AL-1 (2 mM) from HSA (5 mM) was conducted with emission spectrum of urine is similar to that of the HSA bound
three site-specific drugs including warfarin (domain IIA), AL-1, this might potentially lead to errors in determining
ibuprofen (domain IIIA) and digitoxin (domain IIIA). Warfarin albumin levels. However, upon comparing the albumin levels
exhibited concentration dependent displacement of AL-1 and obtained by using AL-1 and immunoassay, we found that the
about 30% of AL-1 was displaced by 75 mM warfarin, while both albumin levels detected by our probe are in a close range to the
ibuprofen and digitoxin showed almost no change in fluores- values obtained by using immunoassay. Thus, we believe that
cence (Fig. 2b). These results clearly indicate that the fluores- our probe can provide precise albumin levels in urine.
cence turn-on response of AL-1 is due to its specific binding to
In addition, we also compared our AL-1 probe with another
the domain IIA of HSA. Furthermore, we also determined the albumin binding dye, bromocresol purple (BCP), for the detec-
fluorescence lifetime of AL-1 and found that the probe exhibits tion of urinary albumin. Another three urine samples collected
bi-exponential fluorescence decay, which is longer in the from the same healthy male donors on different days were used
presence of HSA (Fig. S6, ESI†). In the absence of HSA, AL-1 for the comparison. As BCP has lower sensitivity for albumin, the
exhibits fluorescence lifetimes of t1 = 0.028 ns and t2 = 0.89 ns, urine samples were concentrated 10-fold before BCP measure-
while in the presence of HSA, the probe displays fluorescence ment. Using the BCP dye, we determined the albumin concen-
lifetimes of t1 = 0.19 ns and t2 = 1.19 ns, respectively. The tration in the three samples to be 8.6 mg LÀ1, 4.5 mg LÀ1
,
results are consistent with the nature of the fluorescent mole- and 14.9 mg LÀ1, respectively (Fig. S8, ESI†), while the albumin
cular rotors which show longer fluorescence lifetimes in a levels of the same samples using AL-1 were determined to be
restricted environment.15
3.2 mg LÀ1, 20 mg LÀ1, 24.8 mg LÀ1, respectively. The results
The direct quantification of albumin levels in urine is showed that the albumin levels detected by our fluorescent
usually hampered by the interference of other biological sub- probe AL-1 displayed some deviations from the BCP method.
stances and high background fluorescence of urine. Being As compared to the BCP dye, we believe that our probe yielded
highly selective and sensitive with high fluorescence turn-on more accurate results as it is more sensitive and does not require
upon binding to HSA, we believe that AL-1 can be applied the concentration of the urine samples, which might have
directly to determine the albumin concentration of urine. To incurred the loss of some albumin during concentration steps.
validate the practicality of the application, three urine samples Thus, AL-1 is a robust and rapid albumin probe which can be
collected from three healthy male donors without medical applied to routine urine screening for albuminuria. The albumin
history were spiked with various concentrations of HSA (from levels of the three samples show much different values because
10 nM to 1 mM; 0.67 mg LÀ1 to 67 mg LÀ1) and the fluorescence the urinary albumin concentration is dependent on the hydra-
response was measured in microtiter plates without further tion status of the person. When the person is well hydrated, the
dilution. We observed an excellent linear correlation between urinary albumin concentration will be low because of dilution
the fluorescence increase of AL-1 and the amount of HSA in the and vice versa.
urine samples (Fig. 3). With the standard addition method,
In conclusion, we have developed a highly sensitive and
the urinary albumin levels of the male donors were determined selective fluorescent probe for albumin quantification in urine.
to be 12.9 mg LÀ1, 2.6 mg LÀ1, and 14.7 mg LÀ1, respectively. Our fluorescent molecular rotor probe AL-1 exhibits remarkable
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 11507--11510 | 11509

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
View Article Online
Communication
ChemComm
3 (a) Microalbuminuria, a marker for organ damage, ed. C. E. Mogensen,
Science Press, London, 1993; (b) J. S. Yudkin, R. D. Forrest and
C. A. Jackson, Lancet, 1988, 332, 530.
fluorescence enhancement of 400-fold in the presence of HSA
with a detection limit of around 6 nM. AL-1 displayed high
binding affinity with HSA and was found to bind specifically to
the domain IIA of HSA by displacement assay. The probe was
used in the quantitative detection of albumin level in urine
samples from several healthy donors and the results were
validated by immunoassay. As compared to the immunoassay,
quantifying urinary albumin with AL-1 is advantageous due to
the low-cost, rapid detection time and a facile operation
procedure, which can be completed within 30 minutes. Further-
more, AL-1 is stable for long-term storage as no degradation
was observed for AL-1 in the DMSO stock solution for at least
6 months (Fig. S9, ESI†). Finally as AL-1 is simple in structure,
very easy to synthesize and stable for long-term storage, we
believe that this fluorescent albumin probe presents a low-cost
alternative to immunoassay for the practical application in the
analysis of low level urinary albumin.
4 K. M. Ward, Anal. Chem., 1995, 67, 383.
5 D. J. Rowe, A. Dawnay and G. F. Watts, Ann. Clin. Biochem., 1990, 27, 297.
6 (a) P. F. Ruhn, J. D. Taylor and D. S. Hage, Anal. Chem., 1994,
66, 4265; (b) A. Silver, A. Dawnay, J. Landon and W. R. Cattell, Clin.
Chem., 1986, 32, 1303; (c) Q.-P. Qin, O. Peltola and K. Pettersson,
Clin. Chem., 2003, 49, 1105.
7 (a) F. L. Rodkey, Clin. Chem., 1965, 11, 478; (b) B. T. Doumas and
T. Peters Jr., Clin. Chem., 2009, 55, 583.
8 (a) G. Sudlow, D. J. Birkett and D. N. Wade, Mol. Pharmacol., 1975,
11, 824; (b) J. Min, J. W. Lee, Y. H. Ahn and Y.-T. Chang, J. Comb.
Chem., 2007, 9, 1079; (c) J. C. Er, M. K. Tang, C. G. Chia, H. Liew,
M. Vendrell and Y.-T. Chang, Chem. Sci., 2013, 4, 2168; (d) Y. Hong,
C. Feng, Y. Yu, J. Liu, J. W. Lam, K. Q. Luo and B. Z. Tang, Anal.
Chem., 2010, 82, 7035; (e) M. A. Kessler, A. Meinitzer, W. Petek and
O. S. Wolfbeis, Clin. Chem., 1997, 43, 996.
9 (a) M. A. Haidekker, A. G. Tsai, T. Brady, H. Y. Stevens, J. A. Frangos,
E. Theodorakis and M. Intaglietta, Am. J. Physiol.: Heart Circ.
Physiol., 2002, 282, H1609; (b) W. J. Akers, J. M. Cupps and
M. A. Haidekker, Biorheology, 2005, 42, 335.
´
10 (a) M. L. Viriot, M. C. Carre, C. Geoffroy-Chapotot, A. Brembilla,
We are grateful to the Ministry of Science and Technology
(Grant No. 102B2011I5) and the Ministry of Education (Grant
No. 102N2011E1), Taiwan (ROC) for financial support.
S. Muller and J. F. Stoltz, Clin. Hemorheol. Microcirc., 1998, 19, 151;
(b) M. A. Haidekker, T. Ling, M. Anglo, H. Y. Stevens, J. A. Frangos
and E. A. Theodorakis, Chem. Biol., 2001, 8, 123.
11 W. L. Goh, M. Y. Lee, T. L. Joseph, S. T. Quah, C. J. Brown, C. Verma,
S. Brenner, F. J. Ghadessy and Y. N. Teo, J. Am. Chem. Soc., 2014,
136, 6159.
12 Y.-H. Ahn, J.-S. Lee and Y.-T. Chang, J. Comb. Chem., 2008, 10, 376.
13 (a) I. Petitpas, A. A. Bhattacharya, S. Twine, M. East and S. Curry,
J. Biol. Chem., 2001, 276, 22804; (b) S. Curry, H. Mandelkow, P. Brick
and N. Franks, Nat. Struct. Biol., 1998, 5, 827.
Notes and references
1 T. Peters, All About Albumin: Biochemistry, Genetics, and Medical Applica-
tion, Academic Press, San Diego, CA, 1996, pp. 234–240.
2 (a) E. R. Mathiesen, T. Deckert, K. Johansen, B. Oxenboll and 14 (a) X. M. He and D. C. Carter, Nature, 1992, 358, 209; (b) A. J. Ryan,
P. A. Svendsen, Diabetologia, 1984, 26, 406; (b) G. C. Viberti,
R. J. Jarrett, U. Mahmud, R. D. Hill, A. Argyropoulos and H. Keen,
J. Ghuman, P. A. Zunszain, C.-W. Chung and S. Curry, J. Struct. Biol.,
2001, 174, 84.
Lancet, 1982, 319, 1430; (c) C. E. Mogensen and C. K. Christensen, 15 T. Iwaki, C. Torigoe, M. Noji and M. Nakanishi, Biochemistry, 1993,
N. Engl. J. Med., 1984, 311, 89.
32, 7589.
11510 | Chem. Commun., 2014, 50, 11507--11510
This journal is ©The Royal Society of Chemistry 2014