Communications
HeLa cells for 3 h with 20 mm P9 induced only caspase 9 activi-
ty, this compound initiates apoptosis mainly by a mitochondrial
pathway (Figure 1B). To gain an insight in the subcellular local-
isation of potential targets, an in situ fluorescent-dye-genera-
tion strategy was applied (Figure S11); this avoids direct influ-
ence of the fluorophore on the peptoid localisation. To this
end, a benzothiazole-based “click-on” dye was chosen, due to
the low affinity of the pro-dye towards proteins or cell organ-
transportin-1, Exp1 and Exp2 to P9-A but not to inactive P30-
A was confirmed by immunoblotting after the pulldown ex-
periment with HeLa lysates. In addition, competition of immo-
bilised and non-immobilised ligand for binding to Imp5 and
transportin-1 was observed (Figure 2B). Pulldown experiments
employing the purified proteins Imp5, Exp1 and Exp2 instead
of cell lysate confirmed binding of the active probe P9-A but
not of the control probe P30-A to all three proteins (Fig-
ure S16). In order to monitor binding of P9 to Imp5 by fluores-
cence polarisation (FP), fluorescein-labelled derivatives of
active P9 (compound P9-D) and inactive P30 (compound P30-
D) were employed (Figure 2A) and binding of P9-D to Imp5,
with an apparent KD of 1.9Æ0.1 mm, was confirmed (Fig-
ure 2C).
[
33]
elles, and azido derivatives of P20 and P30 (i.e., P20-B and
P30-B, respectively) were prepared. Fixation of P20-B-treated
cells and a subsequent click reaction resulted in an intense
staining of the perinuclear region (Figure 1C), whereas P30-B-
treated cells displayed a weak fluorescent signal mainly located
in the nucleus. P20-B was primarily localised in the endoplas-
mic reticulum (ER) and mitochondria in further colocalisation
assays with suitable trackers (Figures S12 and S13).
Nuclear export inhibitors such as the Exp1 inhibitor leptomy-
[
43]
cin B exhibit anticancer activity by disruption of the trans-
port of growth regulatory proteins and of tumour suppressor
proteins, and examples of import inhibitors have also been de-
These results raised the question of whether the P9-induced
apoptosis was initiated through impairment of the ER func-
tions. Disruption of the ER homeostasis can result in an accu-
mulation of misfolded and unfolded proteins: that is, ER stress.
In response to ER stress, cells try to sustain survival through
the activation of an integrated intracellular signal transduction
pathway termed unfolded protein response (UPR). However,
[
44–45]
scribed.
Proper nucleocytoplasmic protein transport is es-
sential to maintain cell homeostasis. Most of the cargoes iden-
tified for each karyopherin use multiple transport pathways to
ensure reliable flux through the nuclear pore. Therefore, com-
bined inhibition of several karyopherins might be needed to
impair nuclear transport, and this would result in a “fast” apop-
totic catastrophe as observed for P9.
[
34]
under chronic ER stress cells can undergo apoptosis. Thus,
[35]
mitochondrial apoptosis can be triggered and consequently
[36–37]
accompanied by caspase 9 activation.
Downstream effec-
tors of the UPR are usually measured as indicators of ER
Target identification for antimitotic derivatives
[
38]
stress. Immunoblotting of a selection of downstream pro-
teins from the IRE1 (inositol-requiring enzyme 1) and PERK
The antimitotic activity of P26 was confirmed by quantification
[
29]
(
protein kinase RNA-like endoplasmic reticulum kinase) signal-
as described above
(Figure 3B). At 20 mm concentration,
[34]
ling pathways of the UPR was assessed, and only a slight
increase in phosphorylated p-eIf2a was observed (Figures S14
and S15).
apoptotic cell death led to a reduction in the number of mitot-
ic cells. Antimitotic compounds frequently exert their bioactiv-
ity by induction of mitotic arrest through interaction with tu-
bulin, and immunostaining of tubulin in the presence of 5 mm
P26 indeed revealed impaired microtubule cytoskeletons in
HeLa cells (Figure 4A or Figure S19). Subsequent analysis of tu-
bulin polymerisation through turbidity measurements showed
that P26 dose-dependently inhibited microtubule formation
Salubrinal is a known stress suppressor that exhibits an anti-
apoptotic effect through inhibition of p-eIf2a dephosphoryla-
[
39]
tion.
Interestingly, salubrinal pretreatment protected cells
from P9-induced apoptosis (Figure 1D).
(
Figure 4A) and almost completely prevented tubulin polymer-
Identification of cellular targets of P9
isation at a concentration as low as 2.5 mm. In addition, related
peptoids P27, P29 and P33 also inhibited tubulin polymeri-
sation in vitro (data not shown). Peptoid P9 was again not
active in these assays, thus suggesting a different target for
this compound. Tubulin polymerisation in cells in the presence
of P26 was monitored through depolymerisation of microtu-
bules by cold treatment and subsequent rewarming at 378C to
initiate microtubule assembly. In DMSO-treated cells microtu-
bule-organising centres become visible 2 min after rewarming,
then microtubules started to regrow 5 min after rewarming,
and the microtubule network was recovered after 10 min. In
contrast, P26 delayed the formation of microtubule-organising
centres, which appeared 5 min after rewarming.
To identify the cellular targets of P9 by means of a proteomics-
based approach, probes P9-A and P30-A were immobilised on
NHS-activated Sepharose beads, and both label-free qualitative
identification and SILAC-based quantification of proteins en-
riched during affinity purification (“pulldown”) were ex-
[
40]
plored. In the label-free approach, proteins that bound only
to the active compound P9-A were considered target candi-
dates. SILAC-based identification allows relative quantification
[
41–42]
of proteins;
therefore, proteins that were statistically sig-
nificantly enriched during the affinity purification with the
active probe relative to the inactive control were considered
hits. Comparison of the results of both strategies revealed sev-
eral nucleocytoplasmic transport proteins, in particular karyo-
pherins [e.g., importin-5 (Imp5), exportin-1 (Exp1), exportin-2
Of the known binding sites for microtubule-destabilising
agents, the colchicine site and the vinblastine site are charac-
[
46]
(
Exp2), transportin-1] as potential targets. Because receptor-
terised best. We explored whether P26 binds to any of these
mediated nucleocytoplasmic transport is essential for cell via-
sites through fluorescence-based assays that monitor the dis-
[43]
[47]
[48]
bility, its disruption can cause cell death. Binding of Imp5,
placement of colchicine or BODIPY-FL-vinblastine, respec-
ChemBioChem 2015, 16, 1580 – 1587
1585
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim