Selective fluorescence labeling of lipids in living cells

Article abstract of DOI:10.1002/anie.200805507

Click chemistry in vivo: Three phosphatidic acid derivatives with alkyne groups in their fatty acid chains weresynthesized and incorporated into mammalian cell membranes. Cooper(l) catalyzed and strain-promoted azide-alkyne cycloaddition reactions were us

Full text of DOI:10.1002/anie.200805507

Communications
DOI: 10.1002/anie.200805507
Live-Cell Imaging
Selective Fluorescence Labeling of Lipids in Living Cells**
Anne B. Neef and Carsten Schultz*
Lipids play important roles in a wide variety of cellular
processes, from membrane fusion to signal transduction, and
are involved in the formation and transport of microdo-
mains.[1] However, their analysis in living cells is complicated
by their extremely dynamic behavior.[2] As intermediates in
metabolism, they are frequently converted into other lipid
species or free fatty acids.[3] They are components of cellular
membranes and quickly diffuse from one membrane compart-
ment to another. Furthermore, the analysis of lipid fluxes is
often problematic as a result of the similarity of lipid
structures. A considerable amount of lipid research relies on
the use of tagged lipid analogues. However, given the size of
an average lipid molecule, even small probes, such as
fluorophores, can have a dramatic effect on the properties
of the lipid, particularly if lipid trafficking and sorting are to
be investigated.[2,4] Hence, the labeled lipid should resemble
its natural counterpart as closely as possible. The modification
should preferably be a small hydrophobic group.
was based on the assumption that a terminal triple bond in a
fatty acid chain should alter neither the overall structure of
the lipid nor its polarity.
The bioorthogonal chemical reporter strategy is becoming
increasingly popular for the labeling of all kinds of biomol-
ecules in their native environment. The method comprises
two steps: First, a unique chemical functionality is incorpo-
rated into the target biomolecule, preferably by the biosyn-
thetic machinery of the cell. In the second step, the functional
group is labeled with a nondisruptively delivered probe in a
specific chemical reaction.[5] The most commonly used
reactions are the reaction of a bisarsenite with a tetracysteine
peptide motif,[5b] the Staudinger ligation, copper(I)-catalyzed
azide–alkyne cycloaddition reactions, and strain-promoted
azide–alkyne cycloaddition reactions.[6] Proteins,[7] polynuc-
leotides,[8] and glycoconjugates[9] have been labeled success-
fully in or on living or fixed cells by using these reactions. A
few studies have been concerned with the detection of
lipidated proteins through the metabolic incorporation of w-
azido fatty acids.[10] However, the detection of the azidoacy-
lated proteins through a Staudinger ligation was only possible
after cell lysis. No studies on the chemical labeling of lipids of
any kind in living systems have been reported to date. Herein,
we report the selective fluorogenic labeling of the alkyne-
containing phospholipid derivatives 1, 2, and 5 (SATE =
S-acetylthioethyl) in various mammalian cells. Our approach
We chose phosphatidic acid (PA) as a model lipid. PA is a
key intermediate of phospholipid metabolism and an impor-
tant lipid second messenger. It is mainly generated by
phospholipase D mediated hydrolysis of phosphatidylcholine
in response to a variety of intracellular stimuli. Phosphatidic
acid regulates a number of target proteins, such as mTOR and
Raf-1, protein phosphatase-1, and cAMP-specific phospho-
diesterases. It affects various cellular functions as diverse as
cell proliferation, metabolism, cytoskeletal rearrangement,
and exocytosis.[11] Methods for studying PA dynamics and
interactions with effector proteins in living cells could reveal
further roles of PA in cell signaling, or confirm findings from
in vitro studies. Furthermore, any labeling approach estab-
lished for PA could be extended to the investigation of other,
more complex lipids by simply varying the lipid headgroup to
monitor signaling events with spatial resolution.
The diacyl PA derivative 1 was synthesized in six steps and
50% yield from commercially available 6-heptynoic acid. The
esterification of 6-heptynoic acid with 2,3-O-isopropylidene-
sn-glycerol was followed by ketal cleavage, dimethoxytrityl
(DMT) protection, ester formation with myristic acid, and
DMT removal. The resulting diacyl glycerol derivative was
transformed into the PA derivative 1 by phosphorylation with
bis(S-acetyl-2-thioethyl)-N,N-diisopropylphosphoramidite
(see Scheme S1 in the Supporting Information). Protection of
the otherwise negatively charged phosphate headgroup with
SATE groups was necessary for membrane penetration and to
prevent lipid aggregation. SATE-protected phosphate groups
are resistant to many conditions used in synthetic organic
chemistry, but are rapidly cleaved inside living cells by cellular
esterases to yield a free phosphate group. The resulting
compounds cannot leak from the cell.[12]
[*] A. B. Neef, Dr. C. Schultz
Cell Biology and Biophysics Unit
European Molecular Biology Laboratory
Meyerhofstrasse 1, 69117 Heidelberg (Germany)
Fax: (+49)6221-387-206
E-mail: schultz@embl.de
[**] We thank Heike Stichnoth for providing cells and Dr. Adrian Neal for
providing TBTA and 3-azido-7-(diethylamino)coumarin.
The PA derivative 2 with a nonhydrolyzable ether-coupled
alkyne-containing alkyl chain was synthesized from 10-
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.200805507.
1498
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 1498 –1500

Angewandte
Chemie
undecyn-1-ol in seven steps and 28% overall yield. Appel
halogenation of 10-undecyn-1-ol and ether formation with
2,3-O-isopropylidene-sn-glycerol were followed by further
steps analogous to those in the synthesis of 1 (see Scheme S2
in the Supporting Information).
To test their general suitability for copper(I)-catalyzed
azide–alkyne cycloaddition, 1 and 2 were treated with
fluorogenic 3-azido-7-(diethylamino)coumarin[13] in vitro.
The corresponding fluorescent triazole derivatives 3 and 4
were obtained in virtually quantitative yield (Scheme 1).
Compounds 3 and 4 showed excitation maxima at around
410 nm and emission maxima at around 500 nm. They were
applied to living cells, and their temporal and spatial dynamics
were observed (see Figure S1 in the Supporting Information).
RAW macrophages turned out to be the most suitable cell
type for the visualization of lipids, since their globular
morphology enables individual membrane compartments to
be distinguished. Shortly after addition, both 3 and 4 stained
the plasma membrane; however, they quickly relocated to
intracellular membranes. Especially in the case of 4, a small,
perinuclear region was exceptionally fluorescent.
The cell-permeant, alkyne-containing PA derivatives 1
and 2 were applied to living cells in a similar way and labeled
with 3-azido-7-(diethylamino)cou-
Scheme 1. Synthesis of the fluorescent phosphatidic acid derivatives 3
and 4: a) 3-azido-7-(diethylamino)coumarin, CuSO4, sodium ascorbate,
EtOH/H2O (1:1).
marin[13] in the presence of cop-
per(II) sulfate, sodium ascorbate,
and tris(benzyltriazolylmethyl)a-
mine (TBTA). Since copper ions
are toxic, cells had to be fixed prior
to labeling. Paraformaldehyde fix-
ation was chosen to restrict lateral
and vesicular movement through
the immobilization of lipid-binding
proteins. After just ten minutes of
labeling, cell membranes fluor-
esced brightly. Negative control
cells, which were not treated with
alkyne-modified lipids, but which
were otherwise treated identically,
remained virtually nonfluorescent
even after longer labeling times
(Figure 1). By using different cell
types, such as HeLa and MDCK
cells, the labeling method was
shown to be widely applicable
(see Figure S2 in the Supporting
Information).
Figure 1. Fluorescence microscopic images of fixed RAW macrophages treated with the PA
derivatives 1 or 2 for 1 h and with 3-azido-7-(diethylamino)coumarin, CuSO4, sodium ascorbate, and
TBTA for 30 min. a) Cells treated with 1 and the nuclear marker TO-PRO-3 iodide. b) Cells treated
with 2 and the nuclear marker TO-PRO-3 iodide. c) Negative control cells treated with 3-azido-7-
(diethylamino)coumarin, CuSO4, sodium ascorbate, and TBTA only. DIC=differential interference
contrast.
In contrast to the results
obtained with the prelabeled
lipids 3 and 4, the labeling of 1
and 2 in living cells led to more or
less equal staining of all cellular
membranes. Whereas the sn-1
ether derivative 4 stained the peri-
nuclear region exceptionally brightly, the sn-1 ether deriva-
tive 2 appeared to be distributed evenly, which indicates that
the localization of fluorophore-tagged lipid analogues is often
assumed wrongly. We also showed by fluorescence recovery
after photobleaching (FRAP) that paraformaldehyde fixation
resulted in lipid immobilization, as no movement of
unbleached, fluorescent molecules into bleached regions of
cells was observed even after 30 min (see Figure S3 in the
Supporting Information). The results indicated that the
intracellular localization of the labeled PA derivatives was
Angew. Chem. Int. Ed. 2009, 48, 1498 –1500
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 1499

Communications
not directed by the fluorophore but reflected the position of
cellular distribution, lipid-interaction partners, and behavior
in response to endogenous or environmental stimuli.
the alkyne lipid at the moment of fixation. It cannot be
completely ruled out that the ester functionality of 1 is
cleaved prior to fixation and labeling. However, in vivo
labeling of hexynoic acid resulted in a slightly different
fluorescence distribution (see Figure S4 in the Supporting
Information).
Received: November 11, 2008
Published online: January 14, 2009
Keywords: alkynes · click chemistry · fluorescent probes ·
.
phospholipids
Owing to the toxicity of copper ions, cells needed to be
fixed prior to labeling. However, fixation prevents the
investigation of dynamic cellular processes. We therefore
designed the cyclooctyne-containing PA derivative 5, to label
lipids in living cells through strain-promoted azide–alkyne
cycloaddition.[14] A cyclooctyne-containing fatty acid was
synthesized from commercially available cycloheptene in
three steps and 39% yield. Further synthetic steps according
to the procedure described for 1 gave 5 in 46% yield after six
steps (see Scheme S3 in the Supporting Information). Com-
pound 5 was then applied to living RAW macrophages. Fixed
(see Figure S5 in the Supporting Information) and living cells
were labeled successfully with 3-azido-7-(diethylamino)cou-
marin[13] within 3 h in the absence of auxiliaries (Figure 2).
The staining pattern was very similar in both cases. No cell
toxicity was observed even after several hours, and mitosis
continued to occur.
In conclusion, we have described herein the first success-
ful labeling of lipids by azide–alkyne cycloaddition in fixed
and living cells. The localization differences associated with a
simple ester-to-ether modification and, more importantly, the
differences in the distribution of prelabeled lipids and those
labeled in vivo demonstrate the relevance of the technique.
We assume that the effect of a terminal triple bond on the
biophysical properties of a membrane lipid is negligible, but
the extent to which the cyclooctynyl group influences lipid
diffusion and location remains to be shown. In the future,
more reactive triple bonds will be beneficial for staining
procedures in the minute range.[14b,c] The labeling technique
developed herein should enable the monitoring of lipid
dynamics in fixed and living cells and give new insight into
[1] R. Woscholski, Signal Transduction 2006, 6, 77 – 79.
[2] O. Maier, V. Oberle, D. Hoekstra, Chem. Phys. Lipids 2002, 116,
3 – 18.
[3] J. van der Waal, R. Habets, P. Varnai, T. Balla, K. Jalink, J. Biol.
Chem. 2001, 276, 15337 – 15344.
[4] R. D. Klausner, D. E. Wolf, Biochemistry 1980, 19, 6199 – 6203.
[5] a) J. A. Prescher, C. R. Bertozzi, Nat. Chem. Biol. 2005, 1, 13 –
21; b) S. R. Adams, R. E. Campbell, L. A. Gross, B. R. Martin,
G. K. Walkup, Y. Yao, J. Llopis, R. Y. Tsien, J. Am. Chem. Soc.
2002, 124, 6063 – 6076.
[6] V. V. Fokin, ACS Chem. Biol. 2007, 2, 775 – 778.
[7] a) K. E. Beatty, J. C. Liu, F. Xie, D. C. Dieterich, E. M. Schuman,
Q. Wang, D. A. Tirrell, Angew. Chem. 2006, 118, 7524 – 7527;
Angew. Chem. Int. Ed. 2006, 45, 7364 – 7367; b) A. E. Speers,
B. F. Cravatt, Chem. Biol. 2004, 11, 535 – 546.
[8] a) A. Salic, T. J. Mitchison, Proc. Natl. Acad. Sci. USA 2008, 105,
2415 – 2420; b) C. Y. Jao, A. Salic, Proc. Natl. Acad. Sci. USA
2008, 105, 15779 – 15784.
[9] a) L. K. Mahal, K. J. Yarema, C. R. Bertozzi, Science 1997, 276,
1125 – 1128; b) D. Rabuka, S. C. Hubbard, S. T. Laughlin, S. P.
Argade, C. R. Bertozzi, J. Am. Chem. Soc. 2006, 128, 12078 –
12079; c) M. Sawa, T.-L. Hsu, T. Itoh, M. Sugiyama, S. R.
Hanson, P. K. Vogt, C.-H. Wong, Proc. Natl. Acad. Sci. USA
2006, 103, 12371 – 12376; d) T.-L. Hsu, S. R. Hanson, K. Kishi-
kawa, S.-K. Wang, M. Sawa, C.-H. Wong, Proc. Natl. Acad. Sci.
USA 2007, 104, 2614 – 2619.
[10] a) Y. Kho, S. C. Kim, C. Jiang, D. Barma, S. W. Kwon, J. Cheng, J.
Jaunbergs, C. Weinbaum, F. Tamanoi, J. Falck, Y. Zhao, Proc.
Natl. Acad. Sci. USA 2004, 101, 12479 – 12484; b) H. C. Hang, E.-
J. Geutjes, G. Grotenbreg, A. M. Pollington, M. J. Bijlmakers,
H. L. Ploegh, J. Am. Chem. Soc. 2007, 129, 2744 – 2745; c) M. A.
Kostiuk, M. M. Corvi, B. O. Keller, G. Plummer, J. A. Prescher,
M. J. Hangauer, C. R. Bertozzi, G. Rajaiah, J. R. Falck, L. G.
Berthiaume, FASEB J. 2008, 22, 721 – 732.
[11] a) M. N. Hodgkin, T. R. Pettitt, A. Martin, R. H. Michell, A. J.
Pemberton, M. J. O. Wakelam, Trends Biochem. Sci. 1998, 23,
200 – 204; b) K. Athenstaedt, G. Daum, Eur. J. Biochem. 1999,
266, 1 – 16; c) X. Wang, S. P. Devaiah, W. Zhang, R. Welti, Prog.
Lipid Res. 2006, 45, 250 – 278; d) C. L. Stace, N. T. Ktistakis,
Biochim. Biophys. Acta Mol. Cell Biol. Lipids 2006, 1761, 913 –
926.
[12] a) I. Lefebvre, C. Pꢀrigaud, A. Pompon, A.-M. Aubertin, J.-L.
Girardet, A. Kirn, G. Gosselin, J.-L. Imbach, J. Med. Chem. 1995,
38, 3941 – 3950; b) O. Wichmann, J. Wittbrodt, C. Schultz,
Angew. Chem. 2006, 118, 522 – 527; Angew. Chem. Int. Ed.
2006, 45, 508 – 512.
[13] K. Sivakumar, F. Xie, B. M. Cash, S. Long, H. N. Barnhill, Q.
Wang, Org. Lett. 2004, 6, 4603 – 4606.
[14] a) N. J. Agard, J. A. Prescher, C. R. Bertozzi, J. Am. Chem. Soc.
2004, 126, 15046 – 15047; b) J. M. Baskin, J. A. Prescher, S. T.
Laughlin, N. J. Agard, P. V. Chang, I. A. Miller, A. Lo, J. A.
Codelli, C. R. Bertozzi, Proc. Natl. Acad. Sci. USA 2007, 104,
16793 – 16797; c) X. Ning, J. Guo, M. A. Wolfert, G.-J. Boons,
Angew. Chem. 2008, 120, 2285 – 2287; Angew. Chem. Int. Ed.
2008, 47, 2253 – 2255.
Figure 2. Fluorescence microscopic images of living RAW macro-
phages. a) Cells treated with 5 for 1 h and 3-azido-7-(diethylamino)-
coumarin for 3 h. b) Negative control cells treated with 3-azido-7-
(diethylamino)coumarin for 3 h only.
1500
www.angewandte.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 1498 –1500