close

Вход

Забыли?

вход по аккаунту

?

Chameleon Labels for Staining and Quantifying Proteins.

код для вставкиСкачать
Communications
Fluorescent Probes
Chameleon Labels for Staining and Quantifying
Proteins
Bianca K. Wetzl, Sergiy M. Yarmoluk, Douglas B. Craig,
and Otto S. Wolfbeis*
The analysis of a protein pattern, its temporal changes, and
the interpretation of its function is one of the fascinating
technologies at present and is often referred to as proteomics.[1] Protein patterns can be analyzed by a variety of
methods including gel electrophoresis, blotting, or by socalled biochips.[2] While biochips (and protein arrays) are
preferably applied to systems of known protein composition
to identify specific proteins, electrophoresis in one- or twodimensional form is readily applied to unknown samples.[2]
Since gel electrophoresis is a separation technique, it also
requires appropriate methods for determination (or visualization) of proteins.[3] Standard methods for visualization
include silver staining or staining with dyes, such as Coomassie Brilliant Blue (CBB) or Amido Black B.[4] Fluorescent
methods for staining and visualization of proteins are of
particular interest because of the high sensitivity of (laserinduced) fluorescence, which has reached the nano- and
picomole (if not zeptomole or single molecule) level, at least
for solutions.
Two types of fluorescent protein stains need to be
distinguished: The first involves covalent linkage of the
stain to a functional group of a protein (such as amino or
thiol), the second involves noncovalent protein–stain interaction. Both have their merits. A covalent linkage to the
protein is stable (i.e., the tag cannot be washed out), while
noncovalent labeling enables, for example, mass spectroscopy
to be performed because no change in the total mass of the
protein occurs on staining.[5] Typical noncovalent protein
stains include the SYPRO dyes (certain organic or organometallic fluorochromes) that give red or pink emissions.[6] The
stains bind to proteins with high affinity which then can be
determined in gels in quantities of 2–10 ng/band. Covalent
fluorescent labeling, in contrast, is widely used in polyacrylamide gel electrophoresis (PAGE). Conjugation is achieved
by either pre-staining (that is, before electrophoresis) or after
electrophoresis. A variety of covalently binding labels are
known.[6–8] However, all of them have spectral properties that
are identical (within a few nm) in the free and the protein-
[*] Dipl.-Chem. B. K. Wetzl, Prof. Dr. D. B. Craig,[+] Prof. Dr. O. S. Wolfbeis
Institute of Analytical Chemistry, Chemo- and Biosensors, University
of Regensburg, 93040 Regensburg (Germany)
Fax: (+ 49) 941-943-4064
E-mail: otto.wolfbeis @chemie.uni-regensburg.de
Dr. S. M. Yarmoluk
Department of Combinatorial Chemistry
Institute of Molecular Biology and Genetics
Academy of Sciences, UI-03187 Kyiv (Ukraine)
[+] Humboldt Fellow; on leave from the Department of Chemistry,
University of Winnipeg, Winnipeg R3B2E9, Canada.
5400
2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
bound form. This situation requires that—in case of staining
after electrophoresis—excess label has to be washed out very
carefully from the gel or blot to minimize fluorescent
background.
Herein we report an entirely new class of labels for gel
electrophoresis and also for general protein quantitation.
Pyrylium groups react with amines to form the respective
pyridinium analogues.[9] We have synthesized the new labels
by attaching a pyrylium group to a small aromatic group
(thereby forming one of the smallest blue and fluorescent
chromophores known) and this resulted in the label Py-1. Py-
1 is obtained in a single reaction step and in good yields from
the respective benzaldehyde derivative and 1,3,6-trimethylpyrylium tetrafluoroborate. Its spectral data are given in
Figure 1 and in Table 1. Py-1 is blue, virtually nonfluorescent,
Figure 1. Normalized absorption and fluorescence emission spectra of
the label Py-1 (1A, 1F) and of its conjugate to human serum albumin
(2A, 2F) at pH 7.2 in a 22 mm phosphate buffer. The molar extinction
coefficient of the free label is 63 000 L cm 1 mol 1 and the conjugated
label is 24 000 L cm 1 mol 1. The dye-to-protein ratio of the conjugate
is 1.7.
Table 1: Photophysical properties of stains Py-1, Py-3, and Py-5 in free
form and after conjugation to human serum albumin.
em
Label labs
max/lmax of free stain
[nm]
em
labs
max/lmax of conjugate
[nm]
f
[%][a,b]
t
[ns][c]
Py-1
Py-3
Py-5
503/602
464/582
465/629
50
10
15
2.8
2.3
2.3
621/665
572/641
644/732
[a] f= quantum yields of the conjugate; the quantum yields of the
unconjugated labels are < 1 %. [b] Dependent on dye-to-protein ratio;
usually best at 0.5 to 1.0. [c] t = Decay times of conjugates. The decay
times of stains Py-1, Py-3, Py-5 are < 0.5 ns.
DOI: 10.1002/anie.200460508
Angew. Chem. Int. Ed. 2004, 43, 5400 –5402
Angewandte
Chemie
and reacts with primary amino groups of proteins (but also of
amino-modified DNA and other primary amines, such as
dopamine) in aqueous solution of pH 8–9 at room temperature to give a covalently stained red conjugate. Simultaneously, the fluorescence quantum yield increases to typically
50 % depending on the protein to be labeled and on the dyeto-protein ratio (DPR). Since these labels undergo such a
significant (and visually detectable) change in both absorption and fluorescence, we refer to them as chameleon labels.
Their spectral properties can be modified by variation of end
groups and chain length, and such changes resulted, for
example, in labels Py-3 and Py-5, both of which also react with
proteins.
The fluorescence of the red conjugate formed between
Py-1 and proteins can be excited between 470 and 530 nm and
therefore matches several standard laser lines, whilst the free
(blue) label is not excited at all at this wavelength. Thus, the
fluorescence of the stained protein is measured against an
almost dark background even if residual free (blue) label is
still present.
The chameleon labels have two additional attractive
features: The first is that labeling causes a relatively small
increase in the mass of the protein (Dm = 288 g mol 1 in case
of mono-labeling with Py-1). Second (and possibly even more
importantly), the electrical charge of a protein does not
change on conjugation (since a positively charged amino
group is replaced by a positively charged pyridinium group).
The second feature is particularly significant, since it has been
reported that multiple labeling of proteins with dyes that have
no (or even a negative) charge results in several differently
charged labeled proteins. As a result, a certain protein can
display different migration rates in (capillary) electrophoresis
and consequently give rise to more than one peak or
substantial band broadening depending on detection
method.[10]
Label Py-1 was tested for its suitability in electrophoresis
on sodium dodecylsulfate (SDS)-PAGE gels. Figure 2 shows
the result of an electrophoretic separation of a variety of
proteins on a standard gel with 10 different proteins.[11] The
gel was analyzed using a standard laser-based scanner[12] and
the results (Table 2) show that even in these initial experiments the limits of detection for most proteins are comparable with the (highly sensitive but tedious) silver staining
method, and often better those obtained with CBB and the
Figure 2. Left: standard SDS polyacrylamide gel with ten lines of a
serial dilution of the mass standard.[11] Right: Right side of the gel
(lines 5–10). Both gels displayed in inverted intensity.
Angew. Chem. Int. Ed. 2004, 43, 5400 –5402
Table 2: Detection limits for proteins (ng/band) in SDS-PAGE as
determined by staining with silver, CBB, and Py-1.
Protein
Silver staining[a]
CBB[a]
Py-1
myosin
b-galactosidase
glycogen-phosphorylase
bovin serum albumin
glutamate dehydrogenase
lactate dehydrogenase
carbonic anhydrase
trypsin inhibitor
lysozyme
aprotinin
1
3
1
3
3
n.d.
6
3
4
6
8–16
8–16
16–30
16–30
16
n.d.
16–30
16–30
16–30
16–30
1
0.8
3
< 2.5
< 0.8
4
4
12
14
4
[a] Data from ref. [6a] n.d. = not determined.
SYPRO stains.[6] However, most covalently binding stains
interfere with mass spectrometry, which is used as a detection
method for proteins after gel electrophoresis. While the stain
Py-1 is rather “small”, it cannot be excluded that more than
one Py-1 is linked to a protein, thus complicating MS analysis.
Stain Py-1 may also be used for the quantitation of
proteins in solution (rather than on gels). To demonstrate this,
various proteins were dissolved in sodium carbonate buffer
solution, and a solution of the Py-1 in aqueous dimethylformamide or aqueous methanol was added.[13] A color change
from blue to red is observed, and either absorption or
fluorescence intensity can be measured and plotted against
protein concentration as shown in Figure 3 for the fluores-
Figure 3. Calibration plot for BSA using label Py-1. The insert reveals
the sensitivity of the assay in giving a limit of detection of approximately 60 ng of BSA per mL, and also demonstrates the good reproducibility (low standard deviations).
cence assay of bovine serum albumin (BSA). As a result of
the sensitivity and brightness of the fluorescence of the
labeled protein, concentrations as low as 60 ng mL 1 can be
quantified. Thus, it is comparable (if not better) in terms of
sensitivity to the most sensitive other assays including the
widely used Lowry, Biuret,[14] or ATTO-TAG[15] assays, with
the notable difference that the protein is not altered by poorly
reproducible redox reactions (as in Lowry and Biuret assays)
and that the highly toxic cyanide need not be added (as in case
of ATTO-TAG).
In conclusion, we believe that Py-1 and its congeners Py-3
and Py-5 are a new class of protein labels that can be applied
in proteomics, for example, in page SDS electrophoresis, but
www.angewandte.org
2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
5401
Communications
also for general protein assay. The features of the labels
include a) ease of preparation, b) a clear color change on
labeling, c) a transition from a nonfluorescent state to a
strongly fluorescent state upon conjugation, d) an increase in
the decay time from < 1 ns to > 2 ns, e) very low limits of
detection for both gel assays and solution assays, and f) the
charge of a protein is not changed on labeling.
[12]
Received: April 29, 2004
[13]
.
Keywords: electrophoresis · fluorescent probes · protein assay ·
proteomics
[1] a) M. P. Washburn, D. Wolters, J. R. Yates, Nat. Biotechnol. 2000,
19, 242 – 247; b) J. Reidl, J. Hacker, Mol. Infect. Biol. 2002, 187 –
193.
[2] a) C. D. OHConnor, K. Rickard, Microarrays Microplates 2003,
61 – 68; b) P. Cutler, Proteomics 2003, 3, 3 – 18.
[3] a) J. H. Issaq, Adv. Protein Chem. 2003, 65, 249 – 269; b) M.
Zhou, L. Yu, Adv. Protein Chem. 2003, 65, 57 – 84.
[4] a) P. J. Wirth, A. Romano, J. Chromatogr. A 1995, 698, 123 – 143;
b) J. P. Goldring, L. Ravaioli, Anal. Biochem. 1996, 242, 197 –
201; c) W. F. Patton, J. Chromatogr. B 2002, 77, 3 – 31;
[5] J. C. Nishihara, K. M. Champion, Electrophoresis 2002, 23,
2203 – 2215.
[6] a) T. H. Steinberg, L. J. Jones, R. P. Haugland, V. L. Singer, Anal.
Biochem. 1996, 239, 223 – 237; b) M. F. Lopez, K. Berggren, E.
Chernokalskaya, A. Lazarev, M. Robinson, W. F. Patton,
Electrophoresis 2000, 21, 3673 – 3683; c) W. F. Patton, Electrophoresis 2000, 21, 1123 – 1144.
[7] R. M. Leimgruber, J. P. Malone, M. F. Radabaugh, M. L.
LaPorte, B. N. Violand, J. B. Monahan, Proteomics 2002, 2,
135 – 144.
[8] a) J. Bergquist, S. D. Gilman, A. G. Ewig, R. Ekman, Anal.
Chem. 1994, 66, 3512 – 3518; b) K. E. Asermely, C. A. Broomfield, J. Nowakowski, B. C. Courtney, M. Adler, J. Chromatogr. B
1997, 695, 67 – 75.
[9] O. M. Kostenko, S. Y. Dmitrieva, O. I. Tolmachev, S. M. Yarmoluk, J. Fluoresc. 2002, 12, 173 – 175.
[10] D. B. Craig, N. J. Dovichi, Anal. Chem. 1998, 70, 2493 – 2494.
[11] a) Protein Mix: myosin (Mr = 220 000), 80 ng/band; b-galactosidase (Mr = 116 000), 50 ng/band; glycogen phosphorylase (Mr =
97 000), 150 ng/band; albumin (Mr = 66 000), 250 ng/band; glutamate dehydrogenase (Mr = 55 600), 80 ng/band; lactate dehydrogenase (Mr = 36 500), 80 ng/band; carbonic anhydrase (Mr =
29 000), 80 ng/band; trypsin inhibitor (Mr = 20 000), 250 ng/band;
lysozyme (Mr = 14 000), 700 ng/band; aprotinin (Mr = 6100),
180 ng/band. The serial dilution factor of this mixture from
line 1 to ten (Figure 2) is 0.0, 1.3, 1.7, 2.0, 2.5, 5.0, 10.0, 12.5, 16.7,
20.0. b) Reagents and Conditions for Electrophoresis: Tris/HCl
(450 mm ; Tris = 2-amino-2-(hydroxymethyl)propane-1,3-diol) as
the probe buffer, glycerol (12 %), SDS (4 %), Coomassie
Brilliant Blue G (0.0025 % in water), phenol red (0.0025 %),
pH 8.45; running buffer (pH 8.3) consisting of Tris/HCl (25 mm),
glycine (192 mm), and SDS (0.1 %); 125 V, current from 80 mA
(at the beginning) to 40 mA (at the end of the run; 90 min);
vertical cell. c) Staining: Following electrophoresis, the gel was
incubated in a solution containing 50 % methanol, 10 % acetic
acid, and 40 % distilled water (3 min), washed with 50 %
aqueous methanol, and then fixed twice with a mixture of
50 % triethylamine/acetate (TEAA) buffer of pH 10 and 50 %
methanol. Staining was performed with a freshly prepared
0.004 % solution of Py-1 (now available from Chromeon) in a 1:1
mixture of methanol and TEAA buffer (Py-1 should be
predissolved in minute quantities of DMF). The staining time
5402
2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
[14]
[15]
depends on the thickness and percentage of the gel. Once the
optimal signal is achieved, additional staining (over night) does
not enhance or degrade the signal. The gel is rinsed a) briefly in a
washing solution of 50 % methanol and 50 % TEAA buffer,
b) several times with a destaining solution containing 5 %
methanol, 7 % acetic acid, and 88 % distilled water, and
subsequently scanned or dried. Gels are stable in the washing
solution for at least two days.
We use a Tecan FL200 fluorescence scanner (excitation at
542 nm (argon laser); emission filter set to 630 nm).
A BSA standard solution (20 mg mL 1) was prepared in sodium
carbonate buffer of pH 9, and diluted to several different
concentrations in the microtiter plate. The stain Py-1 was diluted
from a methanol stock solution (1 L 10 4 mol L 1) with distilled
water to a concentration of 2.5 mm. The insert in Figure 3
illustrates the curve obtained at very low protein concentrations.
Each data point is the average of five determinations (limit of
detection 0.06 mg mL 1).
a) O. H. Lowry, N. J. Rosebrough, A. L. Farr, R. L. Randall, J.
Biol. Chem. 1951, 193, 265 – 275; b) K. Wickelman, R. Braun, J.
Fitzpatrick, Anal. Biochem. 1988, 175, 231 – 237; c) H. Zheng,
Y. X. Mao, D. H. Li, C. Q. Zhu, Anal. Biochem. 2003, 318, 86 –
90; d) C. V. Sapan, R. L. Lundblad, N. C. Price, Biotechnol.
Appl. Biochem. 1999, 29, 99 – 108.
W. W. You, R. P. Haugland, D. K. Ryan, R. P. Haugland, Anal.
Biochem. 1997, 244, 277 – 282.
www.angewandte.org
Angew. Chem. Int. Ed. 2004, 43, 5400 –5402
Документ
Категория
Без категории
Просмотров
2
Размер файла
136 Кб
Теги
chameleon, protein, labels, staining, quantifying
1/--страниц
Пожаловаться на содержимое документа