Full Paper
for example, by an acetyltransferase such as SOAT1, which was
also identified as a target of callyspongynic acid.[24]
pling of a synthesis (or isolation) of an alkyne-containing natu-
ral product with a ‘native’ chemical proteomics approach, in
combination with a GO term analysis, is a straightforward
methodology to gain insights into the potential molecular
mechanism of a compound and thus may enable the direct
search for ‘chemical phenotypes’ of natural products. We
therefore anticipate that such an approach may find wide-
spread application in the future.
The other half of the putative targets is predicted to lack en-
zymatic activity. Some targets such as VDAC2 (voltage-depen-
dent anion-selective channel protein 2)[25] or SLC6A8 (sodium-
and chloride-dependent creatine transporter 1)[26] are sup-
posed to be involved in cellular trafficking, and at first sight
might not be obvious targets for callyspongynic acid. However,
the GO annotation enrichment by cellular compartment re-
veals that almost all putative targets are mainly membrane-as-
sociated.
Experimental Section
Altogether, the observed enrichment of the enzymes and
membrane-associated proteins seems to result from direct la-
beling with callyspongynic acid, for example, by a ‘metabolic
labeling’ in which callyspongynic acid is used as a long-chain
fatty acid derivative or by chemical reaction with target pro-
teins, for example, after chemical modification (e.g., acylation)
of the hydroxy moiety, which converts this residue into a leav-
ing group for SN2’-type reactions. Alternatively, the enrichment
of the membrane-associated proteins could be explained by
an ‘unspecific’ detergent-like perturbation of membrane struc-
ture from callyspongynic acid application, resulting in an in-
creased ‘release’ of membrane-associated proteins.
Supplementary figures and synthetic procedures for the synthesis
of callyspongynic acid (1) can be found in the Supporting Informa-
tion.
Live cell tagging and preparation of samples for in-gel fluo-
rescence detection
HeLa cells in DMEM (6 cm round culture dishes, 80% confluent,
10% CO2, 378C, 3 mL media from Gibco supplemented with 10%
FBS from Sigma) were incubated for 6 h with 1 at 0, 1, 2, 5, 10, and
20 mm. Cells were washed with PBS (32 mL), and then harvested
on ice in 100 mL lysis buffer (1PBS, 0.1% SDS, 1% Triton X-100,
1EDTA-free complete protease inhibitor cocktail from Roche Di-
agnostics). Lysates were kept on ice for 20 min and centrifuged at
17000g for 20 min to remove insoluble material. Protein concen-
tration was determined by using the Bio-Rad DC Protein Assay.
Proteins (100 mg) were diluted with lysis buffer to a final concentra-
tion of 1.0 mgmLÀ1 (94 mL final volume). The click reaction was
then started by adding TBTA (final concentration 0.1 mm), capture
reagent AzTB (final concentration 0.1 mm), CuSO4 (final concentra-
tion 1 mm), and TCEP (final concentration 1 mm). The sample was
vortex-mixed for 1 h before the click reaction was stopped by
adding 2 mL 0.5m aqueous EDTA (final concentration 10 mm). To
precipitate all proteins and remove unreacted capture reagent,
methanol–chloroform precipitation was performed. Briefly, 200 mL
MeOH, 50 mL CHCl3, and 100 mL H2O were added to the click reac-
tion. The samples were then centrifuged at 17000g for 5 min
and the pellets were washed with 1 mL MeOH and dried on air.
SDS (30 mL, 2% in PBS) was added to dissolve the proteins by vigo-
rous vortex-mixing and the samples were diluted with 10 mL 4
times sample loading buffer to 2.5 mgmLÀ1. Proteins (25 mg) were
loaded onto 12% polyacrylamide bis-tris gels and run at 150 V for
90 min. Proteins were fixed (40% MeOH, 10% acetic acid, 50%
water) for 5 min and the gels were washed with water (3). In gel
fluorescence was detected with an Ettan DIGE Imager (GE Health-
care) and the protein loading was checked by Coomassie staining.
Conclusion
We have presented a combined experimental approach to
characterize potential targets of a polyacetylene natural prod-
uct. To this end, we described the first enantioselective chemi-
cal synthesis of the polyacetylene natural product callyspon-
gynic acid, thereby proving the correct structural assignment
from the original isolation. By using an inherent feature of the
compound (that is, the presence of multiple alkyne groups
amenable to the click chemistry approach), we subsequently
investigated the putative protein targets of this compound in
human cell cultures. To this end, we affinity-enriched putative
targets after in-cell tagging and used quantitative mass spec-
trometry to identify 37 proteins that bind callyspongynic acid.
Notably, 90% of these putative targets turned out to be local-
ized within the membrane. Based on GO annotations, about
60% of the targets were classified as enzymes, and most of
these are involved in lipid and fatty acid metabolism.
Our findings thus indicate that callyspongynic acid might
represent a valuable chemical tool to perturb and to profile ac-
tivity of multiple different membrane-associated proteins, al-
though further biological studies are required to determine
the overall biological impact of these perturbations. In addi-
tion, our studies indicate that callyspongynic acid may target
organelles featuring large-membrane structures such as the
endoplasmic reticulum (ER), which comprises the largest mem-
brane-bound structure in the cell and harbors many lipid-mod-
ifying enzymes. These findings indicate that the so far un-
known chemical phenotype and thus bioactivity of callyspon-
gynic acid could, for example, be perturbation of ER biology.
On a broader view, our study provides a general roadmap to
characterize putative targets of polyacetylene natural products
(as well as other alkyne-containing natural products). The cou-
Live cell tagging and preparation of samples for proteomics
HeLa cells in R10K8 DMEM (three 10 cm round culture dishes, cells
80% confluent pre-cultured in R10K8 for 7 cell doublings, 10%
CO2, 378C, 10 mL media from Dundee Cell Products supplemented
with 10% dialysed FBS from Sigma) were incubated for 6 h with
1 at 5 mm, whereas control HeLa cells in R0K0 DMEM supplement-
ed with 10% dialysed (Mw =10000 Da) FBS were incubated for 6 h
with DMSO (10 mL, same amount the probe was dissolved in). Cells
were then washed with PBS (35 mL), and then harvested on ice
in the lysis buffer (1PBS, 0.1% SDS, 1% Triton X-100, 1EDTA-
free complete protease inhibitor cocktail from Roche Diagnostics).
Lysates were kept on ice for 20 min and centrifuged at 17000g
for 20 min to remove insoluble matter. Protein concentration was
Chem. Eur. J. 2015, 21, 10721 – 10728
10726
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim