Secondary-ion mass spectrometry of genetically encoded targets

Article abstract of DOI:10.1002/anie.201411692

Secondary ion mass spectrometry (SIMS) is generally used in imaging the isotopic composition of various materials. It is becoming increasingly popular in biology, especially for investigations of cellular metabolism. However, individual proteins are diffi

Full text of DOI:10.1002/anie.201411692

.
Angewandte
Communications
International Edition: DOI: 10.1002/anie.201411692
German Edition: DOI: 10.1002/ange.201411692
Protein Analysis
Secondary-Ion Mass Spectrometry of Genetically Encoded Targets**
Ingrid C. Vreja, Selda Kabatas, Sinem K. Saka, Katharina Krçhnert, Carmen Hçschen,
Felipe Opazo, Ulf Diederichsen,* and Silvio O. Rizzoli*
Abstract: Secondary ion mass spectrometry (SIMS) is gen-
erally used in imaging the isotopic composition of various
materials. It is becoming increasingly popular in biology,
especially for investigations of cellular metabolism. However,
individual proteins are difficult to identify in SIMS, which
limits the ability of this technology to study individual
compartments or protein complexes. We present a method for
specific protein isotopic and fluorescence labeling (SPILL),
based on a novel click reaction with isotopic probes. Using this
method, we added 19F-enriched labels to different proteins, and
visualized them by NanoSIMS and fluorescence microscopy.
The 19F signal allowed the precise visualization of the protein
of interest, with minimal background, and enabled correlative
studies of protein distribution and cellular metabolism or
composition. SPILL can be applied to biological systems
suitable for click chemistry, which include most cell-culture
systems, as well as small model organisms.
sample, where it causes the release of secondary ions. The
isotopic nature of the released ions is analyzed by mass
spectrometry, thereby providing an image of the sample
composition.
NanoSIMS experiments are often descriptive, measuring
the overall distribution of native isotopes in the samples.[6,7]
Other applications are based on introducing a non-native,
isotopically labeled molecule into the biological sample, and
measuring its distribution.[8,9] A related procedure measures
cellular turnover, by pulsing cells with an isotopically labeled
metabolite, such as an amino acid or a nucleotide, and
imaging its incorporation.[10–12]
However, all of these applications suffer from one major
limitation: it is difficult to identify specific proteins or specific
organelles, except for a few that can be identified by their
morphology.[4,10] One proposed solution has been the corre-
lation of NanoSIMS with fluorescence imaging.[4] This
approach is difficult, since it requires the use of two instru-
ments, and a very precise overlap of the two types of images.
Another solution has been to immunostain the samples using
antibodies coupled to isotopically pure metals, such as
lanthanides.[13] These antibodies can be imaged in NanoSIMS,
without the need for correlative microscopy. This technique,
however, undermines to some extent the high resolution of
NanoSIMS, since antibodies, and especially the ones coupled
to large metal tags, incorporate relatively poorly into speci-
mens, and find only a small fraction of the epitopes, thus
resulting in a less-precise image.[14,15]
Anumber of technologies have been recently developed to
investigate the chemical composition of biological samples,
including matrix-assisted laser desorption/ionization
(MALDI),[1,2] laser-ablation inductively coupled plasma
mass spectrometry (LA-ICP-MS),[3] or secondary-ion mass
spectrometry (SIMS).[4,5] SIMS has the highest spatial reso-
lution, approximately 30–50 nm in the lateral direction (x–
y plane), and down to 5–10 nm in the axial direction
(z direction), when operating in the NanoSIMS configura-
tion.[5] NanoSIMS is based on the use of a primary ion beam,
such as Cs+ ions, which is scanned across the surface of the
[*] I. C. Vreja,[+] Dr. S. K. Saka, K. Krçhnert, Dr. F. Opazo,
Prof. Dr. S. O. Rizzoli
[+] These authors contributed equally to this work.
[**] We thank Dr. Vladimir Belov (Department of NanoBiophotonics,
Max Planck Institute for Biophysical Chemistry, Gçttingen (Ger-
many)) for providing Star635P-azide. We also acknowledge the
technical support of Johann Lugmeier (Department Ecology and
Ecosystem Management, Center of Life and Food Sciences
Weihenstephan, TU München (Germany)) with the NanoSIMS
measurements. We also thank William I. Zhang for help with the
manuscript. I.C.V. was supported by a Dorothea Schlçzer Ph.D.
Fellowship from the University of Gçttingen. This work was
supported by grants to S.O.R. and U.D. from the Deutsche
Forschungsgemeinschaft Cluster of Excellence Nanoscale Micros-
copy and Molecular Physiology of the Brain (CNMPB), by grants to
S.O.R. from the DFG (RI1967/3-1) and ERC (CoG NeuroMolAnat-
omy).
Department of Neuro- and Sensory Physiology
University Medical Center Gçttingen
Humboldtallee 23, 37073 Gçttingen (Germany)
E-mail: srizzol@gwdg.de
S. Kabatas,[+] Prof. Dr. U. Diederichsen
Institute for Organic and Biomolecular Chemistry
University of Gçttingen
Tammannstrasse 2, 37077 Gçttingen (Germany)
E-mail: udieder@gwdg.de
Dr. C. Hçschen
Department of Ecology and Ecosystem Management, Center of Life
and Food Sciences Weihenstephan
Technische Universität München
Freising-Weihenstephan (Germany)
I. C. Vreja,[+] S. Kabatas,[+] Dr. S. K. Saka, K. Krçhnert, Dr. F. Opazo,
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201411692.
Prof. Dr. U. Diederichsen, Prof. Dr. S. O. Rizzoli
Center for Nanoscale Microscopy and Molecular Physiology of the
Brain (CNMPB), Gçttingen (Germany)
I. C. Vreja[+]
International Max Planck Research School Molecular Biology
Gçttingen (Germany)
ꢀ 2015 The Authors. Published by Wiley-VCH Verlag GmbH & Co.
KGaA. This is an open access article under the terms of the Creative
Commons Attribution Non-Commercial NoDerivs License, which
permits use and distribution in any medium, provided the original
work is properly cited, the use is non-commercial and no
modifications or adaptations are made.
5784
ꢀ 2015 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 5784 –5788

Angewandte
Chemie
This type of problem has been solved in fluorescence
imaging by the introduction of green fluorescent proteins
(GFPs), which can be coupled to native proteins by genetic
encoding. We sought to perform a comparable strategy for
NanoSIMS imaging by introducing a specific isotope into the
protein of interest. Three major conditions are required. First,
the genetic encoding procedure should be sufficiently flexible
to enable the introduction of the isotope in any protein, in as
many types of biological samples as possible. Second, the
isotope should be linked to the protein of interest through
a highly specific reaction. Third, the isotope should not be
present at high levels in untreated preparations, to enable
a high signal-to-noise ratio.
We tested the specificity of the click-based labeling in
mammalian cells, by expressing tagged versions of several
proteins involved in membrane fusion (SNAREs): the trans-
membrane syntaxin 1 and syntaxin 13, and the membrane-
attached SNAP-25 (Figure S1 in the Supporting Informa-
tion),[23] which we chose based on our experience with their
biology and tagging.[24,25] The UAA propargyl-l-lysine
(PRK), which contains an alkyne group, was introduced
relying on the wild-type tRNA/RS pair from Methanosarcina
mazei.[19,20] All three proteins were detected with high
specificity by this procedure (Figure S2).
To identify the proteins in NanoSIMS, we sought to
couple an isotopically enriched probe to the alkyne group of
the PRK (Figure 1). We found 19F to be easily identified in
NanoSIMS, and to be present at low levels in normal cells, and
should thus provide a high signal-to-noise ratio. We therefore
produced a clickable dipentafluorobenzyl peptide
The first and the second condition are best fulfilled by
what is known as genetic-code expansion, and subsequent
click-chemistry labeling (see Figure 1). This technique incor-
containing 13 19F atoms, which we termed SK155
(Figure 2). To enable convenient testing of this
probe by fluorescence microscopy, we also intro-
duced the fluorophore Abberior Star635, which
contributes three of the 19F atoms.[26] A nona-
peptide sequence was designed to provide proper
solubility under physiological conditions, by
incorporating four negatively charged amino
acids (glutamate and aspartate). Lysine amino
acids with orthogonal protection were used for
Figure 1. Propargyl-l-lysine (PRK) incorporation into proteins of interest and click
reaction with SK155, a dual probe that can be used in both fluorescence and SIMS
imaging.
later side-chain attachment of the Star635-NHS
fluorophore (Abberior GmbH, Germany), and an
N-terminal lysine served for the linkage of two
porates unnatural amino acids (UAAs) into the
proteins of interest.[16,17] The UAAs are then
revealed by coupling to different probes, such as
fluorophores, for example through the copper-
catalyzed azide–alkyne cycloaddition (click
chemistry),[18] or its copper-free variants. To incor-
porate the UAA into a specific protein, a modified
version of the protein containing an Amber stop
codon (TAG) is expressed together with a pair of
suitable tRNA and aminoacyl–tRNA synthetase
(tRNA/RS). The tRNA/RS pair introduces the
UAA selectively into the protein of interest, at the
site determined by the Amber codon.[19,20] A major
advantage of click chemistry over other genetic
encoding procedures is that the tag, the alkyne or
azide moiety present on the UAA, is extremely
Figure 2. SK155 contains 19F atoms as isotopic labels for SIMS, and an Abberior
Star635 fluorescent moiety.
small, of only a few atoms in size. It is therefore
less bulky than tags such as GFP, and offers several
other technical advantages.[21] The coupling of
specific fluorophores or isotopes to the UAAs
takes place only after cellular fixation, when functional
perturbations are no longer a concern. In principle, the
experiment could even be performed without the subsequent
click-chemistry reaction, by incorporating amino acids con-
taining atoms or isotopes that are not naturally abundant in
cells, such as selenocysteine,[22] albeit this only allows the
incorporation of one Se isotope per protein, which would
result in a fairly low signal-to-noise ratio.
2,3,4,5,6-pentafluorobenzoic acid units. Finally, azidolysine
(Fmoc-Lys(N3)-OH) was included, providing the azide for
later attachment of the label SK155 to proteins by click
chemistry. Synthesis of SK155 (Figure S3) was provided by
microwave-mediated solid-phase peptide synthesis (SPPS) of
the nona-peptide on a Sieber amide resin, applying the Fmoc
synthesis method with O-(benzotriazol-1-yl)-N,N,N’,N’-tetra-
methyluronium hexafluorophosphate (HBTU), 1-hydroxy-
Angew. Chem. Int. Ed. 2015, 54, 5784 –5788
ꢀ 2015 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5785

.
Angewandte
Communications
benzotriazole (HOBt) activation. The 2,3,4,5,6-pentafluoro-
benzoic acid units were coupled on solid support.[27] Cleavage
from the resin provided the nona-peptide (C6F5CO)2-Lys-
Asp-Glu-Lys(N3)-Gly-Asp-Glu-Lys-Gly-NH2 that was cou-
pled to the Star635-NHS in solution. After purification by
HPLC, the constitutional integrity of the label SK155 was
indicated by high-resolution mass spectrometry.
optical technique, and is typically 50-fold lower than the axial
resolution of NanoSIMS.[4] This problem is avoided using
SPILL.
In a proof-of-principle experiment, we combined SPILL
imaging with an analysis of cellular turnover, relying on the
incorporation of isotopically labeled metabolites. We treated
the cells with 15N-labeled leucine, which is incorporated into
all newly translated proteins, and thus indicates the turnover
of the protein-based cellular structures (see Figure S5 for
a timeline of the experiments). The ratio between the 15N and
14N isotopes provides an accurate measurement of turnover:
the higher this ratio, the higher the incorporation of new
proteins within the particular structure (Figure 4A). Relying
on the high axial resolution of NanoSIMS, we probed the 15N,
14N, and 19F composition of the sample at different depths (see
syntaxin 1 images in Figure 4A), which enabled us to obtain
much more information than a simple correlation with the
fluorescence image (Figure 4A, inset). We measured the
protein metabolism of the cells at sites marked by the proteins
of interest (identified by the high 19F signal). Interestingly, our
measurements indicated that regions of high protein expres-
sion correlated negatively with the 15N/14N ratio, for both
SNAP-25 and syntaxin 13 (Figure 4B,D), but not for syn-
The 19F distribution overlapped well with the fluorescence
signal (Figure 3A and Figure S4), as expected, and also
enabled the analysis of various other parameters, both in the
fluorescence and in the isotopic domains (Figure 3A). The
labeling procedure was efficient (Figure 3B), and the 19F
signal was found, as expected, within the areas containing the
fluorescence signal associated with the protein of interest
(Figure 3C).
These experiments indicate that specific protein isotopic
and fluorescence labeling (SPILL) is feasible. SPILL offers
two major advantages for NanoSIMS imaging. First, it
alleviates the need for correlation with fluorescence micros-
copy to investigate specific proteins. Second, it enables the
investigator to use the high axial resolution of NanoSIMS.
When employing a correlation with fluorescence imaging, the
axial resolution of the correlated images is determined by the
Figure 3. Genetic incorporation of unnatural amino acids (UAAs) as a tool to label proteins for NanoSIMS. A) Fluorescence and NanoSIMS
images of a representative cell that had incorporated PRK in syntaxin 1, and was subsequently labeled with SK155. A confocal image of the
Star635 fluorescence (left), a 19F NanoSIMS image, and an overlay (Star635 signal in red; 19F in green). Additional fluorescence and NanoSIMS
images are shown: a 14N image, and a confocal image of immunostained calnexin, an endoplasmic reticulum marker. Scale bar: 2 mm. B) The
ratio between 19F and 14N in non-transfected cells was measured and compared with cells expressing different proteins. It was significantly higher
for all expressed proteins: SNAP-25 (p<0.01), syntaxin 1 (p<0.001), and syntaxin 13 (p<0.001; t-tests). The number of analyzed cellular regions
is: 32 for non-transfected, 371 for SNAP-25, 281 for syntaxin 1, and 448 for syntaxin 13 expressing cells. C) Linear relationship between 19F/14N
ratio in a cell and SK155 (Star635) intensity in the corresponding cell regions.
5786
www.angewandte.org
ꢀ 2015 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 5784 –5788

Angewandte
Chemie
Figure 4. Genetic labeling of proteins and SIMS imaging offer detailed information on how protein metabolism correlates with the presence of
a specific protein. A) A series of NanoSIMS images taken at different depth planes in a SK155-labeled cell expressing syntaxin 1. 19F is shown on
the upper row; the 15N/14N ratio on the lower row. Inset: for comparison, the fluorescent signal of SK155 in the same cell. Scale bar: 2 mm. B) An
analysis of 15N/14N ratios as a function of 19F levels. The number of analyzed cellular regions is 371 for SNAP-25, 281 for syntaxin 1, and 448 for
syntaxin 13 samples. A downward trend is observed for all three types of staining, which is statistically significant for SNAP-25 and syntaxin 13
(p<0.01, t-tests), but not for syntaxin 1.
taxin 1 (Figure 4C). Owing to problems in correlating images
with different axial resolutions, no significant trends could be
observed when comparing the fluorescence signal of the
proteins to an average 15N/14N ratio image obtained by
overlapping the images taken at different depths (data not
shown). This also constitutes an advantage of NanoSIMS
imaging over techniques such as stimulated Raman scatter-
ing,[28] which measures turnover with a far lower axial
resolution. Albeit, unlike NanoSIMS, these methods are
non-invasive, and allow the imaging of living cells.
implies that a large field of experimentation is open to this
technology.
Keywords: click chemistry · isotopic labeling ·
protein engineering · secondary-ion mass spectrometry (SIMS) ·
unnatural amino acid
How to cite: Angew. Chem. Int. Ed. 2015, 54, 5784–5788
Angew. Chem. 2015, 127, 5876–5880
The results suggest that the plasma membrane and the
early endosomes, where SNAP-25 and syntaxin 13 are
located, have a slower metabolism than other parts of the
cell. The turnover of the endoplasmic reticulum, where
syntaxin 1 is trapped in BHK cells,[29] appears to be similar
to the average cell turnover. While the biological significance
of these findings awaits testing in other cellular systems, we
conclude that the SPILL procedure is an efficient tool for the
SIMS-based investigation of specific proteins. The require-
ments for SPILL are simple: the biological system should
allow UAA encoding and click chemistry. This has been
already achieved in bacteria, yeast, mammalian cells, Caeno-
rhabditis elegans,[30] and Drosophila melanogaster,[31] which
[1] R. M. Caprioli, T. B. Farmer, J. Gile, Anal. Chem. 1997, 69,
4751 – 4760.
[2] S. Trimpin, M. L. Deinzer, J. Am. Soc. Mass Spectrom. 2005, 16,
542 – 547.
[3] J. S. Becker, J. Mass Spectrom. 2013, 48, 255 – 268.
[4] S. K. Saka, A. Vogts, K. Krçhnert, F. Hillion, S. O. Rizzoli, J. T.
Wessels, Nat. Commun. 2014, 5, 3664.
[5] C. Lechene, F. Hillion, G. McMahon, D. Benson, A. M.
Kleinfeld, J. P. Kampf, D. Distel, Y. Luyten, J. Bonventre, D.
Hentschel, K. M. Park, S. Ito, M. Schwartz, G. Benichou, G.
Slodzian, J. Biol. 2006, 5, 20.
[6] N. Musat, H. Halm, B. Winterholler, P. Hoppe, S. Peduzzi, F.
Hillion, F. Horreard, R. Amann, B. B. Jørgensen, M. M. M.
Kuypers, Proc. Natl. Acad. Sci. USA 2008, 105, 17861 – 17866.
Angew. Chem. Int. Ed. 2015, 54, 5784 –5788
ꢀ 2015 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5787

.
Angewandte
Communications
[7] A. W. Thompson, R. A. Foster, A. Krupke, B. J. Carter, N.
Musat, D. Vaulot, M. M. M. Kuypers, J. P. Zehr, Science 2012,
337, 1546 – 1550.
[19] S. Milles, S. Tyagi, N. Banterle, C. Koehler, V. VanDelinder, T.
Plass, A. P. Neal, E. A. Lemke, J. Am. Chem. Soc. 2012, 134,
5187 – 5195.
[8] J. F. Frisz, K. Lou, H. A. Klitzing, W. P. Hanafin, V. Lizunov,
R. L. Wilson, K. J. Carpenter, R. Kim, I. D. Hutcheon, J.
Zimmerberg, M. L. Kraft, Proc. Natl. Acad. Sci. USA 2013,
110, E613 – E622.
[9] J. F. Frisz, H. A. Klitzing, K. Lou, I. D. Hutcheon, P. K. Weber, J.
Zimmerberg, M. L. Kraft, J. Biol. Chem. 2013, 288, 16855 –
16861.
[20] D. P. Nguyen, H. Lusic, H. Neumann, P. B. Kapadnis, A. Deiters,
J. W. Chin, J. Am. Chem. Soc. 2009, 131, 8720 – 8721.
[21] K. Horisawa, Front. Physiol. 2014, 5, 1 – 6.
[22] J. L. Leonard, D. M. Leonard, Q. Shen, A. P. Farwell, P. E.
Newburger, J. Cell. Biochem. 1996, 61, 410 – 419.
[23] R. Jahn, R. H. Scheller, Nat. Rev. Mol. Cell Biol. 2006, 7, 631 –
643.
[10] M. L. Steinhauser, A. P. Bailey, S. E. Senyo, C. Guillermier, T. S.
Perlstein, A. P. Gould, R. T. Lee, C. P. Lechene, Nature 2012,
481, 516 – 519.
[11] S. E. Senyo, M. L. Steinhauser, C. L. Pizzimenti, V. K. Yang, L.
Cai, M. Wang, T.-D. Wu, J.-L. Guerquin-Kern, C. P. Lechene,
R. T. Lee, Nature 2013, 493, 433 – 436.
[12] D.-S. Zhang, V. Piazza, B. J. Perrin, A. K. Rzadzinska, J. C.
Poczatek, M. Wang, H. M. Prosser, J. M. Ervasti, D. P. Corey,
C. P. Lechene, Nature 2012, 481, 520 – 524.
[24] N. D. Halemani, I. Bethani, S. O. Rizzoli, T. Lang, Traffic 2010,
11, 394 – 404.
[25] P. Hoopmann, A. Punge, S. V. Barysch, V. Westphal, J. Bückers,
F. Opazo, I. Bethani, M. Lauterbach, S. W. Hell, S. O. Rizzoli,
Proc. Natl. Acad. Sci. USA 2010, 107, 19055 – 19060.
[26] K. Kolmakov, V. N. Belov, J. Bierwagen, C. Ringemann, V.
Müller, C. Eggeling, S. W. Hell, Chem. Eur. J. 2010, 16, 158 – 166.
[27] K. Tomita, S. Oishi, H. Ohno, N. Fujii, Biopolymers 2008, 90,
503 – 511.
[13] M. Angelo, S. C. Bendall, R. Finck, M. B. Hale, C. Hitzman,
A. D. Borowsky, R. M. Levenson, J. B. Lowe, S. D. Liu, S. Zhao,
Y. Natkunam, G. P. Nolan, Nat. Med. 2014, 20, 436 – 442.
[14] F. Opazo, M. Levy, M. Byrom, C. Schäfer, C. Geisler, T. W.
Groemer, A. D. Ellington, S. O. Rizzoli, Nat. Methods 2012, 9,
938 – 939.
[28] L. Wei, Y. Shen, F. Xu, F. Hu, J. K. Harrington, K. L. Targoff, W.
Min, ACS Chem. Biol. 2015, DOI: 10.1021/cb500787b.
[29] S. Martinez-Arca, V. Proux-Gillardeaux, P. Alberts, D. Louvard,
T. Galli, J. Cell Sci. 2003, 116, 2805 – 2816.
[30] L. Davis, J. W. Chin, Nat. Rev. Mol. Cell Biol. 2012, 13, 168 – 182.
[31] T. S. Elliott, F. M. Townsley, A. Bianco, R. J. Ernst, A. Sachdeva,
S. J. Elsässer, L. Davis, K. Lang, R. Pisa, S. Greiss, K. S. Lilley,
J. W. Chin, Nat. Biotechnol. 2014, 32, 465 – 472.
[15] J. Ries, C. Kaplan, E. Platonova, H. Eghlidi, H. Ewers, Nat.
Methods 2012, 9, 582 – 584.
[16] C. C. Liu, P. G. Schultz, Annu. Rev. Biochem. 2010, 79, 413 – 444.
[17] S. Milles, E. Lemke, Bioessays 2013, 35, 65 – 74.
[18] H. C. Kolb, M. G. Finn, K. B. Sharpless, Angew. Chem. Int. Ed.
2001, 40, 2004 – 2021; Angew. Chem. 2001, 113, 2056 – 2075.
Received: December 4, 2014
Published online: March 17, 2015
5788
www.angewandte.org
ꢀ 2015 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 5784 –5788