Synthesis of autophagosomal marker protein LC3-II under detergent-free conditions

Article abstract of DOI:10.1002/anie.201209523

Just add oil: A new detergent-free method was developed to synthesize lipidated proteins using a light-activatable solubilizing side chain (in dashed circle, see scheme) to assist the ligation of the lipopeptides. This method allows the efficient preparation of a phosphatidylethanolamine-conjugated autophagosomal marker protein (LC3-II) as well as labeled derivatives of LC3-II, which can be used to study autophagy regulation. Copyright

Full text of DOI:10.1002/anie.201209523

.
Angewandte
Communications
DOI: 10.1002/anie.201209523
Protein Lipidation
Synthesis of Autophagosomal Marker Protein LC3-II under Detergent-
Free Conditions**
Yi-Chao Huang, Yi-Ming Li, Yang Chen, Man Pan, Yi-Tong Li, Li Yu, Qing-Xiang Guo, and
Lei Liu*
Autophagy is a catabolic process for the bulk degradation of
intracellular materials, such as abnormal protein aggregates
and damaged organelles.[1] It normally occurs under nutrient
deprivation, causing cytoplasmic components to be engulfed
by the double-membrane-bound autophagosomes (Figure 1).
Autophagosomes then fuse with lysosomes to form autolyso-
somes where the sequestered materials are degraded by the
lysosomal hydrolases. As a pivotal cellular housekeeping
system, autophagy plays important roles in many physiolog-
ical processes including bacterial and viral infection, cell
death, and antigen presentation. Dysfunction of autophagy is
involved in a number of pathologies such as hepatic inflam-
mation, neurodegenerative diseases, and cancer.[2] Owing to
these reasons, there has been growing interest to understand
the mechanisms of autophagy at the molecular level.
The formation of autophagosomal membranes is a key
event in autophagy. During this process, a cytosolic micro-
tubule-associated protein 1A/1B-light chain 3 (LC3-I) is
conjugated to 1,2-distearoyl-sn-glycero-3-phosphoethanol-
amine (DSPE) through an amide bond with the C-terminal
residue, Gly120.[3] The resulting lipid-anchored protein is
called LC3-II. It promotes the elongation of the phagophore
(the autophagosome precursor) and engulfment of cytoplas-
mic materials targeted for degradation. Later on LC3-II is
either enzymatically converted back to LC3-I by a cysteine
protease, Atg-4, or degraded by the lysosomal proteases.[4]
Previous studies have shown that the cellular LC3-II level can
be connected to the autophagic activity.[5] Accordingly, LC3-
II is a good marker for the study of the machinery of
autophagy. Although traditional methods such as Western
blotting can be used to monitor LC3-II,[6] more advanced
studies on autophagy require preparative amounts of LC3-II,
ideally with reporter groups and tags incorporated.
Figure 1. Inter-conversion of LC3-I and LC3-II during autophagy.
Previous attempts to produce LC3-II have relied on
in vitro conjugation of LC3-I with PE using E1- and E2-like
enzymes, Atg7 and Atg3.[7] Unfortunately, this enzymatic
approach can only afford LC3-II in a protein–lipid mixture
and in the low microgram scale. Herein, we report the first
synthesis of homogeneous LC3-II in practical quantities,
using the strategy of expressed protein ligation.[8] This study is
part of our on-going efforts to elucidate the mechanism of
autophagy regulation.[9] Previously chemical synthesis has
been used for the preparation of lipidated proteins with
attached isoprenoid lipids[10] and glycosylphosphatidylino-
sitols.[11] However, we recognized that LC3-II presents an
even more challenging synthetic target, because the PE tail is
too hydrophobic for the previous detergent-assisted ligation
methods. To solve this problem, we developed a new method
for the synthesis of lipidated proteins using a removable
solublizing side chain. With this approach, lipidated proteins
can be prepared under detergent-free conditions, without the
laborious screening of detergents and solvents. This strategy
not only enables the preparation of LC3-II with good
[*] Y.-C. Huang,[+] M. Pan, Y.-T. Li, Prof. Dr. L. Liu
Tsinghua-Peking Center for Life Sciences, Key Laboratory of
Bioorganic Phosphorus Chemistry and Chemical Biology (Ministry
of Education), Department of Chemistry, Tsinghua University
Beijing 100084 (China)
E-mail: lliu@mail.tsinghua.edu.cn
Prof. Dr. Y.-M. Li[+]
School of Medical Engineering, Hefei University of Technology
Hefei, Anhui 230009 (China)
Dr. Y. Chen, Prof. Dr. L. Yu
School of Life Sciences, Tsinghua University
Beijing 100084 (China)
Prof. Q.-X. Guo
Department of Chemistry, University of Science and Technology of
China
Hefei 230026 (China)
[+] These authors contributed equally to this work.
[**] This work was supported by the National Basic Research Program of
China (973 program No. 2013CB932800) and NSFC (No. 21225207,
20932006, 21102083).
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201209523.
4858
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 4858 –4862

Angewandte
Chemie
efficiency, but also may provide a more general and user-
friendly method for making diverse lipidated proteins.[12]
We began by searching for a ligation site suitable for
expressed protein ligation in the C-terminal flexible region of
LC3-I. The DNA sequences corresponding to the first 113,
114, 115, 116, or 118 amino acid residues of LC3-I were cloned
into the pTXB1 bacterial expression vector. The truncated
proteins fused to a C-terminal gyrA intein–chitin binding
domain were expressed in E. coli ER2566 cells.[13] They were
collected and loaded on a chitin affinity column, incubated
overnight, and eluted with a MESNa (sodium 2-mercapto-
ethanesulfonic acid) cleavage buffer (100 mm, pH 7.8) to
produce thioesters of the truncated protein. According to
SDS-PAGE analysis (Supporting Information, Figures S2,S3)
LC3-I[Met1–Ala114]-MESNa thioester was produced with
a higher efficiency (5–6 mgLÀ1) than the other thioesters.
Thus we chose the Ala114-Ser115 bond as the ligation site,
which required the mutation of Ser115 to Cys115 in the target
protein.
We next synthesized the [Cys115–Gly120]-PE hexapep-
tide with a PE tail (Figure S4). For this purpose, we first used
the 2-Cl-trityl resin to prepare the fully protected hexapep-
tide. This hexapeptide was then condensed with DSPE in
chloroform using N,N’-diisopropylcarbodiimide (DIC)
assisted by HOAt (1-hydroxy-7-azabenzotriazole). Subse-
quently, the protecting groups were removed using the
standard trifluoroacetic acid (TFA) cocktail and the desired
product (i.e. [Cys115–Gly120]-PE) was obtained through
precipitation in cold diethyl ether. Unfortunately,
the resulting lipopeptide was barely soluble in up to
80% aqueous acetonitrile containing 0.1% TFA,
rendering its handling and purification very difficult.
Moreover, this lipopeptide could not be dissolved in
the solutions used for expressed protein ligation
(aqueous buffers containing 6m guanidinium chloride
or 8m urea mixed with 0–50% dimethyl sulfoxide or
trifluoroethanol (TFE)). To solve this problem,
a detergent-assisted ligation approach has been
established for the synthesis of lipidated proteins.
However, this strategy failed in the present case,
because many commonly used detergents (e.g. sodium
dodecyl sulphate,[14] b-octyl glucoside,[15] Triton X-
114,[16] dodecylphosphocholine,[17] cetyl trimethylam-
monium bromide[18]) still gave almost no ligation. A
possible explanation for the difficulty encountered
with this ligation is that DSPE (with two C17H35 tails)
is much more hydrophobic than previously used lipid
chains (e.g. the farnesyl group with only one C15H25
tail). We therefore concluded that the detergent-
assisted ligation strategy may not guarantee success
for the synthesis of every lipidated protein.
Figure 2. Development of a removable solubilizing side chain.
solubility, we initially attached three to five arginine residues
(i.e. Arg3, Arg4, Arg5) at the side chain of Thr118 through
a simple ester bond (Figure 2; Figure S5). Our experiments
revealed that the PE-modified lipopeptide was still poorly
soluble with an Arg3 tag (Figure S6). On the other hand,
when an Arg4 or Arg5 tag was attached, the lipopeptide
became soluble in 6m guanidinium chloride buffer and could
be ligated with LC3-I[Met1–Ala114]-MESNa thioester with-
out the need for a detergent (Figures S6,S7).
Unfortunately, the diacyl moiety of PE was destroyed
under the saponification conditions used to cleave the ester
linker. Therefore, we had to design an alternative linker to
attach the Arg4 tag onto the side chain of the PE-modified
lipopeptide. It is important to note that the linker must be
To overcome this solubility problem, we decided
to use a removable solubilizing side chain on the PE-
modified lipopeptide. Because oligoarginine has often
been used to improve the solubility of hydrophobic
peptides[19] (mainly transmembrane protein frag-
ments), we examined the attachment of oligoargi-
Scheme 1. Synthesis of light-activatable lipopeptide 6 with an observed ESI-MS
of 2330.2 Da (calcd: 2330.4 Da). Trt=trityl protecting group; Alloc=allyloxycar-
nines of different lengths to the lipopeptide. To ease
the synthesis of the compounds used for testing bonyl protecting group; Pbf=pentamethyldihydrobenzofurane protecting group.
Angew. Chem. Int. Ed. 2013, 52, 4858 –4862
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 4859

.
Angewandte
Communications
stable for both Fmoc-protected solid
phase peptide synthesis (SPPS) and
native chemical ligation, yet its removal
should not affect the amide and ester
bonds in the final lipidated protein.
With this in mind, we designed a photo-
sensitive nitrobenzyl linker that can be
cleaved using ultraviolet light.[20]
The nitrobenzyl linker involved the
use of an ortho-nitrobenzyl modified
glutamine residue (compound 4, see
Supporting Information). It was syn-
thesized from Fmoc-Glu-OMe (1) in
three steps with an overall yield of 51%
(Figure S1). We used the 2-Cl-trityl
resin to assemble the fully protected
hexapeptide through standard Fmoc
SPPS. Without cleavage of the peptide
from the solid support, the Alloc group
on the side chain of 4 was removed
using Pd(PPh3)4/PhSiH3 in a dimethyl-
formamide (DMF)/CH2Cl2 mixture.
The liberated amino group was then
used to assemble the Arg4 tag capped
with a Boc-Gly-OH residue on the N-
terminus, once again using Fmoc SPPS.
Subsequently, the branched hexapep-
tides with their protecting groups were
cleaved off the resin with a 1:4 mixture
of TFE/CH2Cl2. This intermediate (5)
was condensed with DSPE and treated
with reagent B (TFA/phenol/H2O/tri-
isopropylsilane = 88:5:5:2) to give the
final lipopeptide (6) suitable for the
expressed protein ligation (Scheme 1).
6 could be easily handled, with good
solubility in both 50% aqueous aceto-
nitrile and 6m guanidinium chloride
buffer.
Figure 3. a) Analytical HPLC traces for the ligation of the LC3-I[Met1–Ala114]-MESNa thioester
with lipopeptide 6; b) Characterization of the purified LC3-II precursor 7 by analytical HPLC and
ESI-MS (inset; obs: 15674.4 Da, calcd: 15674.6 Da); c) Characterization of LC3-II after photolysis
under 365 nm UV light. Indet: the observed MALDI-TOF mass for synthetic LC3-II is 14760.1 Da
(calcd [M+1]: 14759.6 Da).
The ligation of LC3-I[Met1–Ala114]-MESNa thioester
(2.4 mg) with 6 (3.1 mg) was carried out in 200 mL ligation
buffer (6m guanidinium chloride, 100 mm NaH2PO4, 50 mm 4-
mercaptophenylacetic acid (MPAA), pH 7.4), with final
reaction concentrations of approximately 0.9 mm and
6.7 mm, respectively. Again, no detergents were needed for
this ligation. The desired ligation product 7 (which is the LC3-
II precursor still carrying an Arg4 tag) was detected within
five minutes by HPLC (Figure 3a) and we stopped the
reaction after four hours. 7 appeared as a double peak in the
HPLC profile because the nitrobenzyl linker was prepared as
an isomeric mixture. 7 could be easily purified with a semi-
preparative reverse-phase HPLC column to afford 2.1 mg of
white powder with a yield of the isolated product of 75%. The
purity and identity of 7 was characterized by analytical HPLC
and ESI-MS (Figure 3b). The use of less than four equivalents
of 6 significantly slowed down the ligation reaction and led to
much lower ligation yields. This may indicate that 6 still
aggregates in the solution despite the attachment of an Arg4
tag. Fortunately, the excess 6 could be easily recovered
following the ligation. This is an advantage of the detergent-
free method for the synthesis of lipidated proteins, because
the use of detergents in the previous methods often made it
difficult to purify the product by reverse-phase HPLC or to
recover the reactants.[21]
To remove the Arg4 tag in the synthetic protein, we
dissolved 7 (to a final concentration of 0.5 mgmLÀ1) in the
photolysis buffer (pH 5.5) containing 6m guanidinium chlo-
ride, 100 mm NaH2PO4, 300 mm MeONH2·HCl, and 10 mm
sodium ascorbate. The solution was irradiated with 365 nm
UV light and monitored by HPLC (Figure 3c). Almost all of
compound 7 was consumed after 45 minutes irradiation, but
the desired product LC3-II was too hydrophobic to be eluted
off of the C4 column. This observation indicated that it is
difficult, if not impossible, to purify LC3-II using HPLC. To
solve this problem, we chose to successively dialyze the
photolysis buffer against renaturation buffers containing 6m,
4m, 2m, 0m urea, according to the standard method.[22] Using
this approach, we expected to separate all the small molecules
from the desired product. We then characterized LC3-II using
4860 www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 4858 –4862

Angewandte
Chemie
MALDI-TOF MS after ESI-MS was unsuccessful (Figure 3c).
In the MALDI-TOF spectrum, the major peak corresponded
to LC3-II but a peak of 7 still appeared. One possible
explanation for this is that the ionization of 7 was better than
that of LC3-II, owing to the Arg4 tag.
To further characterize the synthetic LC3-II, we per-
formed urea/SDS-PAGE experiments (Figure 4a). We found
that LC3-II migrated faster (at approximately 15 kD) on urea/
SDS-PAGE as compared to LC3-I (at approximately 16 kD).
Scheme 2. Generation of fluorescently labeled LC3-II.
To test the biological activity of synthetic LC3-II, we
labeled it with a fluorescent dye by conjugating Alexa
Fluor 488 maleimide to the Cys115 residue (Scheme 2). This
conjugation reaction was completed in two hours at 48C in
aqueous buffer (10 mm NaH2PO4, pH 7.4). Meanwhile, we
prepared liposomes as in vitro mimics of the autophagosome
using the previous method of mixing 1-palmitoyl-2-oleoyl-sn-
glycero-3-phosphocholine,
1,2-dioleoyl-sn-glycero-3-phos-
phoethanolamine, and phosphatidylinositol in a molar ratio
of 30:55:15.[23] Without the addition of LC3-II, the prepared
liposomes were approximately 400 nm in diameter, as mea-
sured by transmission electron microscopy. Following incu-
bation of the fluorescently labeled LC3-II with the above
liposomes, we found that the synthetic LC3-II co-localized
with the liposomes, as shown by confocal fluorescence
microscopy (Figure 4c). The LC3-II anchored liposomes
formed irregularly shaped cluster-like structures with
increased diameters to several micrometers. This observation
was consistent with Ohsumi and co-workers previous finding
that Atg8-PE (the yeast homologue of LC3-II), prepared by
way of enzymatic conjugation, was also caused clustering of
liposomes in vitro.[23] Thus, our synthetic LC3-II was success-
fully anchored to the liposome membranes leading to the
fusion of membranes to create larger liposome clusters. This
experiment confirmed the correct biological activity of the
synthetic LC3-II. Moreover, synthetic LC3-II labeled with
a small fluorophore could provide a useful replacement for
green fluorescent protein (GFP)-fused LC3.[6b]
To summarize, phosphatidylethanolamine conjugated
LC3-II is an important lipidated protein that is needed for
mechanistic studies on autophagy. Our investigation of the
synthesis of LC3-II revealed that this particular protein posed
a unique challenge for the existing technology of expressed
protein ligation owing to the very high hydrophobicity of the
PE moiety. To solve this problem, we developed a new
method to use a light-activatable solubilizing side chain to
assist the ligation of the lipopeptides. This strategy allowed for
the synthesis of lipidated proteins under detergent-free
conditions without laborious screening of the solvents and
additives. It also made the handling and separation of the
synthetic intermediates by the standard reverse-phase HPLC
methods easier. We showed that LC3-II could be readily
prepared using this newly developed method. Physicochem-
Figure 4. a) Urea/SDS-PAGE (top) and Western blot (bottom) of
recombinant LC3-I and synthetic LC3-II. b) CD spectra of the synthetic
LC3-II as compared with that of recombinant LC3-I and the LC3-
I[Met1–Ala114]-MESNa thioester. The protein concentration was
approximately 0.2 mgmLÀ1. c) Liposomes anchored with Alexa
Fluor 488 labeled LC3-II. Differential interference contrast (DIC)
images showed the presence of liposomes (first column), whereas
images in the second column were measured by confocal microscopy
using a 488 nm laser. Images from the first and second lines show two
independent windows under confocal microscopy. For colored pictures,
see the Supporting Information. Scale bars=10 mm.
This extraordinary behavior was in agreement with the
previous observations for LC3-II generated in vivo,[3] which
may be explained by the sticking of negatively charged SDS
molecules to the PE tail of LC3-II. Moreover, a Western-blot
assay showed that our synthetic LC3-II could be detected by
the antibody against LC3 (Figure 4a). Note that the Western-
blot assay was performed with SDS-PAGE instead of urea/
SDS-PAGE, in which the migrations of LC3-I and LC3-II
were similar to each other. To confirm the proper folding of
synthetic LC3-II, we conducted circular dichroism (CD)
measurements. Synthetic LC3-II exhibited the absorptions of
both a-helices and b-sheet structures, similar to the recombi-
nant LC3-I (Figure 4b). On the other hand, the recombinant
LC3-I[Met1–Ala114]-MESNa thioester showed a slightly
different CD spectrum from LC3-I and LC3-II, possibly
owing to the absence of the C-terminal residues.
Angew. Chem. Int. Ed. 2013, 52, 4858 –4862
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
4861

.
Angewandte
Communications
[13] R. K. McGinty, C. Chatterjee, T. W. Muir, Methods Enzymol.
2009, 462, 225.
[14] F. I. Valiyaveetil, R. MacKinnon, T. W. Muir, J. Am. Chem. Soc.
2002, 124, 9113.
[15] a) C. L. Hunter, G. G. Kochendoerfer, Bioconjugate Chem. 2004,
15, 437; b) M. J. Grogan, Y. Kaizuka, R. M. Conrad, J. T. Groves,
C. R. Bertozzi, J. Am. Chem. Soc. 2005, 127, 14383; c) M. G.
Paulick, A. R. Wise, M. B. Forstner, J. T. Groves, C. R. Bertozzi,
J. Am. Chem. Soc. 2007, 129, 11543.
ical and biological tests verified that synthetic LC3-II had the
desired structure and activity. Synthetic LC3-II equipped with
reporter groups or tags is expected to be useful for studying
the regulation of autophagy. We also expect that the light-
activatable solubilizing side chain strategy may provide
a general approach for the synthesis of lipidated proteins.
Received: November 28, 2012
Revised: February 6, 2013
Published online: April 2, 2013
[16] B. Bader, K. Kuhn, D. J. Owen, H. Waldmann, A. Wittinghofer,
J. Kuhlmann, Nature 2000, 403, 223.
[17] a) D. Clayton, G. Shapovalov, J. A. Maurer, D. A. Dougherty,
H. A. Lester, G. G. Kochendoerfer, Proc. Natl. Acad. Sci. USA
2004, 101, 4764; b) D. Olschewski, R. Seidel, M. Miesbauer, A. S.
Rambold, D. Oesterhelt, K. F. Winklhofer, J. Tatzeit, M.
Engelhard, C. F. W. Becker, Chem. Biol. 2007, 14, 994; c) S.
Lahiri, M. Brehs, D. Olschewski, C. F. W. Becker, Angew. Chem.
2011, 123, 4074; Angew. Chem. Int. Ed. 2011, 50, 3988.
[18] a) K. Alexandrov, I. Heinemann, T. Durek, V. Sidorovitch, R. S.
Goody, H. Waldmann, J. Am. Chem. Soc. 2002, 124, 5648; b) T.
Durek, K. Alexandrov, R. S. Goody, A. Hildebrand, I. Heine-
mann, H. Waldmann, J. Am. Chem. Soc. 2004, 126, 16368;
c) Y. X. Chen, S. Koch, K. Uhlenbrock, K. Weise, D. Das, L.
Gremer, L. Brunsveld, A. Wittinghofer, R. Winter, G. Triola, H.
Waldmann, Angew. Chem. 2010, 122, 6226; Angew. Chem. Int.
Ed. 2010, 49, 6090; d) Y. W. Wu, L. K. Oesterlin, K. T. Tan, H.
Waldmann, K. Alexandrov, R. S. Goody, Nat. Chem. Biol. 2010,
6, 534.
[19] a) T. Sato, Y. Saito, S. Aimoto, J. Pept. Sci. 2005, 11, 410;
b) E. C. B. Johnson, S. B. H. Kent, Tetrahedron Lett. 2007, 48,
1795; c) E. C. B. Johnson, E. Malito, Y. Shen, D. Rich, W. J. Tang,
S. B. H. Kent, J. Am. Chem. Soc. 2007, 129, 11480; d) Z. P. Tan,
S. Y. Shang, S. J. Danishefsky, Proc. Natl. Acad. Sci. USA 2011,
108, 4297.
[20] a) H. S. Li, J. Hah, D. S. Lawrence, J. Am. Chem. Soc. 2008, 130,
10474; b) N. Bindman, R. Merkx, R. Koehler, N. Herrman, W. A.
van der Donk, Chem. Commun. 2010, 46, 8935.
Keywords: autophagy · LC3-II · lipoproteins ·
protein expression · protein ligation
.
[1] E. L. Eskelinen, Autophagy 2005, 1, 1.
[2] a) T. Wileman, Science 2006, 312, 875; b) D. C. Rubinsztein, M.
Difiglia, N. Heintz, Autophagy 2005, 1, 11; c) C. Munz, Cell.
Microbiol. 2006, 8, 891.
[3] Y. Kabeya, N. Mizushima, T. Ueno, A. Yamamoto, T. Kirisako,
T. Noda, E. Kominami, Y. Ohsumi, T. Yoshimori, EMBO J. 2000,
19, 5720.
[4] Y. Kabeya, N. Mizushima, A. Yamamoto, S. Oshitani-Okamoto,
Y. Ohsumi, T. Yoshimori, J. Cell Sci. 2004, 117, 2805.
[5] I. Tanida, T. Ueno, E. Kominami, Methods Mol. Biol. 2008, 445,
77.
[6] a) N. Mizushima, Methods Enzymol. 2009, 452, 13; b) N.
Mizushima, T. Yoshimori, B. Levine, Cell 2010, 140, 313.
[7] Y. Sou, I. Tanida, M. Komatsu, T. Ueno, E. Kominami, J. Biol.
Chem. 2006, 281, 3017.
[8] a) R. R. Flavell, T. W. Muir, Acc. Chem. Res. 2009, 42, 107; b) M.
Vila-Perellꢀ, T. W. Muir, Cell 2010, 143, 191.
[9] a) Y. G. Rong, M. Liu, L. Ma, W. Q. Du, H. S. Zhang, Y. Tian, Z.
Cao, Y. Li, H. Ren, C. M. Zhang, L. Li, S. Chen, J. Z. Xi, L. Yu,
Nat. Cell Biol. 2012, 14, 924; b) W. J. Huang, W. Choi, W. Q. Hu,
N. Mi, Q. Guo, M. S. Ma, M. Liu, Y. Tian, P. L. Lu, F. L. Wang,
H. T. Deng, L. Liu, N. Gao, L. Yu, Y. G. Shi, Cell Res. 2012, 22,
473.
[10] L. Brunsveld, J. Kuhlmann, K. Alexandrov, A. Wittinghofer,
R. S. Goody, H. Waldmann, Angew. Chem. 2006, 118, 6774;
Angew. Chem. Int. Ed. 2006, 45, 6622.
[11] Y. Tsai, X. Liu, P. H. Seeberger, Angew. Chem. 2012, 124, 11604;
Angew. Chem. Int. Ed. 2012, 51, 11438.
[12] a) H. C. Hang, J. P. Wilson, G. Charron, Acc. Chem. Res. 2011,
44, 699; b) H. C. Hang, M. E. Linder, Chem. Rev. 2011, 111, 6341.
[21] The use of detergents in the reaction mixture may significantly
decrease the resolution of the HPLC separation. In addition,
many detergents can be strongly absorbed on the surface of the
reversed-phase columns and thereby, irreversibly damage the
columns.
[22] Protein Expression Analysis Impact Kit Instruction Manual,
New England Biolabs Inc., 2009, 22.
[23] H. Nakatogawa, Y. Ichimura, Y. Ohsumi, Cell 2007, 130, 165.
4862 www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 4858 –4862