Stapling monomeric GCN4 peptides allows for DNA binding and enhanced cellular uptake

Article abstract of DOI:10.1039/c4ob02659d

The basic DNA recognition region of the GCN4 protein comprising 23 amino acids has been modified to contain two optimally positioned cysteines which have been linked and stapled using cross-linkers of suitable lengths. This results in stapled peptides wit

Full text of DOI:10.1039/c4ob02659d

Organic &
Biomolecular Chemistry
View Article Online
View Journal | View Issue
COMMUNICATION
Stapling monomeric GCN4 peptides allows for
DNA binding and enhanced cellular uptake†
Cite this: Org. Biomol. Chem., 2015,
3, 3856
1
a
a
a
b
Abhishek Iyer, Dorien Van Lysebetten, Yara Ruiz García, Benoit Louage,
Received 24th December 2014,
Accepted 17th February 2015
b
a
Bruno G. De Geest and Annemieke Madder*
DOI: 10.1039/c4ob02659d
www.rsc.org/obc
8
The basic DNA recognition region of the GCN4 protein comprising adopt a helical fold in solution by itself nor bind to DNA due
9
2
3 amino acids has been modified to contain two optimally posi- to entropic reasons, an external factor forcing the peptide
tioned cysteines which have been linked and stapled using cross- into a helical conformation is needed. For this reason unlike
linkers of suitable lengths. This results in stapled peptides with a the existing synthetic bZip models in which DNA binding is
5
,10
stabilized α-helical conformation which allows for DNA binding induced via dimerisation,
we here aim to stabilize a single
and concurrent enhancement of cellular uptake.
α helix via peptide stapling, reasoning that enhancing the heli-
city within the monomer should sufficiently stabilize the con-
7
formation to allow DNA binding. Stapled peptides have been
1
1
Introduction
used extensively for improving helicity, increasing cell-pene-
1
2,13
tration,
proteolytic stability and enhancing peptide–
Transcription factors (TFs) by definition are responsible for
the decoding of genetic information from dsDNA to mRNA.
14
protein interactions (PPIs). In all the above mentioned cases
the benefits of stapled peptides have been demonstrated.
1
15
In the quest for DNA recognition, TF proteins can be taken as
models to design synthetic DNA binders. They are usually
classified according to the fold of their DNA-binding domains
and grouped into a small number of families, like the bZIP,
bHLH, homeodomain, HTH, and zinc fingers which have
However, the use of stapled peptides in the miniaturisation of
zipper proteins has remained largely unexplored.
2
Results and discussion
already been studied in detail. Most transcription factors are
large proteins possessing complex secondary structures ren-
In this work, we have examined the DNA binding induced by
stapling using the monomeric GCN4 transcription factor basic
region as a model peptide, by comparing the i, i + 4 and i, i + 7
stapling methods and varying the positions of the staple along
the helix. The cellular uptake of the constructs was also investi-
gated using fluorescently labelled versions of the peptide.
dering it difficult to design smaller synthetically accessible ver-
3
sions thereof retaining the DNA binding capacity. The GCN4
leucine zipper is however a relatively easy protein to mimic due
to its well-defined dimerization domain and basic recognition
region.
It has already been shown that dimerization of basic region
peptides through non-peptide scaffolds can allow DNA
binding. However, construction of such smaller dimeric
Selection of the stapling methodology
4
Initial studies were dedicated to selecting the most suitable
method for peptide stapling among the ones reported in
mimics of this TF, although feasible, has proven to be syntheti-
5
,6
cally challenging. Previous attempts at more thorough struc-
tural minimisation and reduction of complexity using the
monomeric GCN4 peptide have shown that DNA binding is
1
2,16–19
literature.
The idea was to adopt a method as general as
possible to be applicable to any potential DNA binding
peptide by increasing its helicity. To ensure easy and cost-
effective modification the use of unnatural amino acids was
avoided. Next to the labour-intensive preparation of the
required modified amino acid building blocks, in case of
the very well-known hydrocarbon stapling e.g., coupling of the
non-natural amino acids as well as peptide folding on resin
7
greatly reduced due to loss of secondary structure. Indeed,
since the basic region of the GCN4 transcription factor cannot
a
Organic and Biomimetic Chemistry Research Group, Krijgslaan 281, S4, B-9000
Gent, Belgium. E-mail: annemieke.madder@ugent.be; Fax: +32-9-264 49 98
b
Department of Pharmaceutics. Ghent University, Ottergemsesteenweg 460, 9000
20
has shown to give problems. Therefore, in the current study
Ghent, Belgium. E-mail: br.degeest@ugent.bel; Tel: +329 264 80 55
1
5
we opted for cysteine cross-linking, in view of the commer-
cial availability and/or easy synthesis of cross-linking moieties,
†
Electronic supplementary information (ESI) available: Experimental pro-
cedures, HPLC and NMR data. See DOI: 10.1039/c4ob02659d
3856 | Org. Biomol. Chem., 2015, 13, 3856–3862
This journal is © The Royal Society of Chemistry 2015

View Article Online
Organic & Biomolecular Chemistry
Communication
mild reactions conditions and easy scalability due to synthesis not bind DNA under the given EMSA conditions. For all the
in solution.
synthetic constructs 1a–c, 2d & 3d we see enhanced DNA
binding as compared to peptide 4. Earlier, DNA binding with
monomeric peptides was observed through the use of a graft-
Design and synthesis of stapled peptides
A detailed analysis of the essential contacts for binding of the ing strategy whereby the crucial contact residues of the GCN4
GNC4 protein to its cognate DNA sequence as derived from the binding region are specifically positioned on an avian pancrea-
2
4
reported crystal structure was described earlier by Ellenberger tic polypeptide. Apparently, our simple and straightforward
2
1
(
pdb file: 1YSA). Various dimeric peptides based on the stapling strategy also allows constraining the peptide into a
D226-Q248 basic region of the GCN4 TF have been shown to suitable conformation for DNA binding. In the case of pep-
1
0,22
retain their DNA binding properties.
Based on these tides 1a–c binding is only observed at higher concentrations of
studies, amino acids within this sequence, indicated as not peptide as compared to 2d and 3d. The more hydrophobic
2
3
involved in DNA contacts, were identified and systematically nature of the biphenyl cross-linker, makes peptide 1a more
replaced by Cys according to an i, i + 4 or i, i + 7 format. Mole- susceptible to aggregation (followed by precipitation) as can
cular modelling aided visualisation based on the pdb file be seen from the complete disappearance of the bands in the
2
1
1
YSA was used to ensure that the introduced linkers point
away from the DNA and not towards it thereby avoiding any
steric repulsion which may arise due to peptide stapling. In
this way three different peptides, comprising the D226–Q248
sequence from the DNA binding basic region of GCN4, con-
taining a double Cys substitution (at positions 237/244 for 1,
2
29/233 for 2 and 233/237 for 3) were synthesized on solid
support, cleaved and subsequently treated with various linkers
YSA yielding a series of five stapled peptides as shown in
Fig. 1.
The stapled peptides and the unmodified WT basic region
1
peptide were successfully synthesized. The cross-linkers a, c &
d are commercially available. The stapling moiety b has not
been used thus far for peptide stapling and was designed and
synthesized as a more polar alternative to the biphenyl and
bipyridine cross-linkers.
DNA binding studies
Fig. 2 EMSA titrations for peptides 1a–c, 2d & 3d. Loading mixture
comprises 5 µL from a mixture of 10 µL mQ, 4 µL sucrose, 2 µL loading
buffer, 2 µL DNA, 2 µL peptide resulting in a total DNA concentration of
Next, the DNA binding capacity of all peptides was evaluated
through electrophoretic mobility shift assay (EMSA) titration of
various peptide concentrations to the DNA sequence 5′ – CGG
1
67 nM. The loading buffer consists of 20 µL Tris 1 M, pH = 7.6, 20 µL
0.1 M, 40 µL EDTA 0.025 M. Peptide concen-
trations from left to right (in µM) are indicated below each gel.
ATG ACG TCA TTT TTT TTC – 3′ (Fig. 2) containing the cognate KCl 0.2 M, 20 µL MgCl
monomeric GCN4 binding site GTCAT. The WT peptide 4 does
2
Fig. 1 Synthetic stapled peptides 1a–c, 2d, 3d for DNA binding and unmodified basic region peptide IV. The peptides are N-terminally capped with
a p-acetylamino benzoic acid (ABA) moiety to ensure UV-based detection and analysis in the case of DNA binding. For cell uptake studies the ABA
moiety was replaced by fluorescein.
This journal is © The Royal Society of Chemistry 2015
Org. Biomol. Chem., 2015, 13, 3856–3862 | 3857

View Article Online
Communication
Organic & Biomolecular Chemistry
last lane. On the other hand, peptides 1b and 1c are less prone
to aggregation but the onset of binding occurs at higher con-
centrations than in the case of 1a. Compared to peptides 1a–c,
the DNA binding pattern in gel 2d is different. We propose
that the appearance of two bands related to DNA–peptide
complex formation is due to the presence of two binding sites
in the CRE sequence (…GTCAT…).
It can be noticed that after full occupation of one binding
site by the peptide, remaining non-bound peptide can interact
with the second binding site. The binding pattern of peptide
2
d is unique and only visible in the case of the two sequences Fig. 3 Confocal microscopy images of the uptake of peptides (A) Ib, (C)
IId, and (C) IV at 37 °C. The upper panel shows accumulated images of
that bind at a lower concentration range. It can further be
noticed that peptide 3d shows a similar binding pattern as
peptide 2d but suffers from non-specific interactions, which
cause aggregation, bands getting blurred or disappearance of
all bands.
DNA in the nucleus (blue), cell membrane (red) and fluorescein (green).
The lower panel shows the fluorescein image.
native GCN4 sequence IV, as can be seen from fluorescence of
From the few examples about DNA binding stapled fluorescein in Fig. 3A–C.
7
,25
described in literature
and from our own data we believe
Quantification of the uptake by flow cytometry shows a con-
the major challenge in case of DNA binding peptides is that centration dependence as well as a temperature dependence
increasing the helicity has to be complemented with a degree on the mean fluorescence values measured (Fig. 4). At 37 °C a
of flexibility in order to account for the conformational change considerably larger uptake is observed at a 1 µM concentration
which occurs when the peptide binds to DNA. Since this con- of peptides. Furthermore, the mean fluorescence of the cells
formational change is more significant in the case of DNA increased considerably for an i, i + 7 staple as can be seen
binding peptides as compared to PPIs, a too tight locking of from the results for Ia–c. Incubation at 4 °C provides insight
the peptide into a helical conformation may result in the into the mode of uptake as endocytic pathways are shut down
inability to achieve DNA binding. Indeed, our data show that at this temperature. Comparison of the data obtained at 4 °C
peptide stapling by providing an N or C terminal helix stabiliz- and 37 °C with flow cytometry (Fig. 4) showed that there is sig-
ation, rather than centrally in the sequence, gives better nificant uptake at 37 °C but almost no uptake at 4 °C. This was
results in terms of DNA binding as observed from the binding further confirmed by confocal microscopy in a separate sys-
pattern of peptides 2d & 3d versus 1a–c. Though in the absence tematic study where we compared cell uptake for the non-
of DNA, a low helical content is observed (see CD measure- stapled reference peptide IV and stapled peptide Ib both at
ment results in ESI†) apparently the conformation can be 4 °C and 37 °C (Fig. 5). Except for slight accumulation in and
adjusted into a structure whereby the contacts between the around the cell membrane there was no uptake at 4 °C. Thus,
positively charged side chains, mainly involving the Lys and active uptake is the main internalization pathway for these
Arg residues, and the negatively charged backbone of the DNA peptides based on the results obtained from flow cytometry
can be maximized without a high entropic penalty. We believe and confocal microscopy.
peptide 2d fits these criteria and hence is the best DNA binder
The cell uptake properties even at low concentrations for
from the constructs synthesized. Moreover, peptide 2d is able the current peptides are combined with low cytotoxicity values
to bind in a dimeric fashion without having been artificially
dimerized.
Cell uptake & toxicity studies
The previously related yeast derived GCN4 peptide (231–252)
2
6,27
has been classified as a membrane permeable peptide.
It
varies slightly from the basic region peptide used in this
article which was specifically chosen for its DNA binding abil-
2
2
ities in its dimeric form. It has further been postulated that
stabilization of the secondary structure via peptide stapling
2
8
can enhance cell uptake. Peptides with hydrocarbon staples
in particular have shown considerable increase in uptake as
1
3,29
compared to their non-stapled counterparts.
Here, the
fluorescently labelled versions of peptides Ia–c & IId were
tested in a cellular environment using RAW 264.7 mouse
macrophages. Confocal microscopy confirmed that cell uptake
is achieved even at a low concentration of 0.25 µM for all pep-
Fig. 4 Mean fluorescence of fluorescently labelled peptides by incu-
bation at (A) 0.25 µM, 4 °C, (B) 1 µM, 4 °C, (C) 0.25 µM, 37 °C, (D) 1 µM,
tides at 37 °C using an incubation time of 3 h, including 37 °C.
3858 | Org. Biomol. Chem., 2015, 13, 3856–3862
This journal is © The Royal Society of Chemistry 2015

View Article Online
Organic & Biomolecular Chemistry
Communication
Fig. 7 Trypsin digest of peptides showing different cleavage sites indi-
cated by the arrows, observed with RP-HPLC analysis.
Fig. 5 Confocal microscopy images of the uptake of (A) peptide IV at
3
4
(
7 °C, (B) peptide Ib at 37 °C, (C) peptide IV at 4 °C and (D) peptide Ib at
°C at 0.25 µM. The upper panel shows accumulated images of DNA
blue), cell membrane (red) and fluorescein (green). The lower panel
shows the fluorescein image.
degraded after 24 h, between 20% (Ic) and 80% (Ib) was left
for the different stapled peptides (see Fig. 50 in ESI†).
Conclusions
The elegant nature of the synthesis in combination with the
observed DNA binding and cellular uptake properties render
these constructs of considerable and specific interest among
the mimics of the GCN4 transcription factor reported to date.
Through this work and due to the nature of the stapled pep-
tides, we believe that a general method is now at our disposal
to allow DNA binding and enhance cellular uptake of a given
DNA binding peptide while avoiding tedious synthetic routes.
It has been further shown that N-terminal helix stabilization
as in the case of 2d is more effective for enhancing DNA
Fig. 6 MTT assay for fluorescently labelled peptides IV, Ia, Ib, Ic & IId binding than stapling in the middle of the sequence (1a–c).
measured at concentrations of 0.25 µM and 1 µM.
For cell uptake, however, the more helical i, i + 7 stapled pep-
tides Ia–c have shown better uptake than IV. Although the two
are not mutually exclusive, for future applications such as DNA
as observed in an MTT-assay (Fig. 6). From the MTT assay we
binding in cellulo, a balance will have to be found between a
can conclude that for peptides IV, Ia, Ib, Ic and IId in general
peptide’s DNA binding and cell penetration abilities.
more than 70% of the cells are viable at a concentration of
0
.25 µM. Cell viability was evaluated according to the
ISO10993-5 norms which state that the compounds are cyto-
toxic if the assay points to a cell viability lower than 70%. Experimental
Clearly, we see that the stapled peptides are not toxic at a con-
centration of 0.25 or 1 µM except for IId which seems to be
Materials
All amino acids, trifluoroacetic acid (TFA) and coupling
slightly toxic at a 1 µM concentration.
reagent HBTU were purchased from Iris GmbH. L-Amino acids
were used throughout the syntheses. The protecting groups for
Peptide stability
t
the amino acids are Bu for Asp, Glu, Ser, Thr, Boc for Lys, Pbf
Many hydrocarbon stapled peptides exhibit increased proteo- for Arg and Trt for Cys, Asn, Gln. 4-Acetamidobenzoic acid
lytic resistance, proportional to the degree of α-helicity and (ABA), 4-4′-bis(bromomethyl)biphenyl, α,α′-dibromo-m-xylene,
number of staples introduced, resulting in enhanced thera- p-phenylenediamine, ethyl acetate, acetonitrile, methanol,
3
0
peutic properties. In case of cysteine cross-linking, no sys- diethyl ether, DIPEA, supplied as extra dry, redistilled, 99.5%
tematic comparison of peptide stability of stapled versus pure and triisopropylsilane were purchased from Sigma
unstapled peptides has been reported. As a high number of Aldrich. Bromoacetylbromide was purchased from Acros
Arg and Lys are present, the peptides will be very susceptible Organics. Dimethylformamide (DMF) and N-methylpyrrolidine
to cleavage of trypsin. Therefore a trypsin digest experiment (NMP) peptide synthesis grade were purchased from Biosolve.
was carried out using a trypsin to peptide ratio of 1/1000 wt% HPLC grade quality hexane and chloroform were purchased
and samples were analysed after 0 min, 30 min, 1 h, 2 h and from Fisher Scientific. Deuterated solvents D O (99.9% atom
2
2
4 h using RP-HPLC and MALDI-TOF. As expected, the control D) and DMSO-d6 (99.8% atom D) were obtained from Euriso-
peptide IV was more susceptible to degradation than the top. Water with the Milli-Q grade standard was obtained in-
stapled peptides which can be seen in the higher number of house either from a Millipore ROs 5 purification system or a
different degradation products (Fig. 7). The degradation pro- Sartorius Arium 611 DI. Rink-Amide ChemMatrix
−
1
ducts clearly reflect the influence of the staple on the cleavage (100–200 μm, manufacturer’s loading: 0.52–0.54 mmol g
)
process. Remarkably, while peptide IV was completely was obtained from Biotage. NHS-fluorescein was purchased
This journal is © The Royal Society of Chemistry 2015
Org. Biomol. Chem., 2015, 13, 3856–3862 | 3859

View Article Online
Communication
Organic & Biomolecular Chemistry
from Thermo Scientific as a 5–6 isomer. All reagents were
Loading mixture: The loading mixture comprised of: 10 µL
acquired from commercial sources and used without prior mQ, 4 µL sucrose, 2 µL loading buffer, 2 µL DNA, 2 µL peptide.
purification. All chemicals were used without further purifi- The loading mixture was prepared only 1–2 h prior to running
cation. All oligonucleotides used were commercially purchased of gels and kept on ice as soon as ready.
from Eurogentec (HPLC purified using RP-cartridge-Gold,
00 nm scale) and were used as such.
Preparation of Gels (for 2 Gels). In a clean falcon tube the
following were added (in given order): 15.595 mL mQ, 0.4 mL
TBE, 4.005 mL of 40% acrylamide solution, 200 µL APS (10%
w/w in mQ). The solution was mixed by sonication to remove
any air bubbles and cooled to 0 °C (1 h under ice). 20 µL of
TEMED was then added to the mixture and was again mixed
properly before pouring it gently along parallel glass plates.
The glass plates were tapped gently to ensure removal of all air
bubbles and the markers were squeezed between the plates to
ensure uniform width of each well. Sufficient time was given
for polymerization (∼1 h).
Gel electrophoresis. A pre-run of the gels was performed
prior to loading them. Care was taken to see that the gels were
properly immersed in 0.2× TBE buffer (non-denaturing gel,
without urea) and the loading wells were free from any air
bubbles. Instrument settings: 150 V, 100 mA, 19 W for 30 min
at 4 °C. The wells were washed after the pre-run. 5 µL of the
loading mixture was then loaded onto the wells. Instrument
settings: 150 V, 100 mA, 19 W for 45 min at 4 °C.
2
Peptide syntheses
All automated peptide synthesis were performed using Rink
Amide Chemmatrix resin, loading = 0.54 mmol g− on a
1
1
2
00 mg scale using 10 eq. of amino acid, 10 eq. of HBTU and
0 eq. of DIPEA in DMF using single couplings of 1 h. Fmoc
deprotection was performed using 20% piperidine in DMF and
shaking for 2, 5 and 15 min. For coupling fluorescein to the
N-terminus 3 eq. of NHS fluorescein along with 3 eq. HOBt
and 6 eq. of DIPEA was used and the reaction mixture was
shaken overnight. The peptides were cleaved off using TFA–
2
TIS–H O = 95/2.5/2.5 and precipitated using cold MTBE.
For peptide stapling, 10 mg of crude peptide was dissolved
in 2–3 mL of NH HCO solution (depending on the solubility)
4
3
with 1.5 eq. of tris(2-carboxyethyl) phosphine (TCEP) and was
stirred for 15 min at room temperature. Then 2.5 eq. of cross-
linker dissolved in 1 mL of dry DMF was added drop-wise to
the reaction mixture and was stirred for another 2–2.5 h after
which the crude reaction mixture was analyzed by MALDI-TOF
to observe the progress. If starting material was detected, 1.5
eq. of cross-linker in 1 mL dry DMF was added and the reac-
tion was continued for another 1.5–2 h. MALDI-TOF samples
were taken every 45 min. The solvents were removed and the
peptides were obtained by precipitation (twice) by centri-
fugation at 4000 rpm at 0 °C in cold MTBE (−20 °C). The pellet
Staining of gels. After the run, the gels were removed from
the glass and were stained using 100 mL of 0.2× TBE buffer +
1
1
0 µL Sybr Gold (commercially purchased stock solution
0 000× in DMSO). The gels were then washed twice with mQ
and gently placed under a UV lamp (dark room) to observe the
gel pattern.
Confocal microscopy
5
RAW264.7 cells cultivated in DMEM (10 per well, 300 µL) were
was then redissolved in mQ H O and purified using RP-HPLC
2
plated in confocal plates and incubated overnight at 37 °C and
to give pure stapled peptides.
2
5% CO . Peptides were added in an overall concentration of
0
.25 µM and incubated for 2 h at the same conditions or on
Electrophoretic mobility shift assay
ice. Cells were washed and fixated with 2% of Paraformalde-
The following stock solutions were prepared (fresh each time, hyde for 30 min at 37 °C. Cells were stained by 30 min incu-
except for DNA and peptide): bation with a 0.2% solution of CTB-AF647 and 0.2% of
DNA: 1.67 µM prepared from CRE (5′ – CGG ATG ACG TCA Hoechst in PBS with 1% BSA. This was done to visualize the
TTT TTT TTC – 3′) & CRE complement (5′ – GAA AAA AAA TGA cell nucleus (blue) and cell membrane (red) respectively. Cells
CGT CAT CCG – 3′) and Random (5′ – GCG CGA GAA GGA AAG were resuspended in PBS and measured with confocal micro-
AAA GCC GG – 3′) & complement (5′ – CCG GCT TTC TTT CCT scope (Leica SP5 equipped with a 63× (1.4 NA) oil immersion
TCT CGC GC – 3′) DNA solutions (commercially purchased) by objective) at 3 different wavelengths (405 nm, 488 nm and
diluting with 20 µL 0.5 M Tris, pH = 8, 40 µL 2.5 M NaCl, 643 nm). The peptides used were labeled with fluorescein
4
0 µL 0.025 M EDTA and then adding milliQ water such that which emits green light. To understand the uptake mechanism
the total volume is 1 mL. The DNA was annealed by heating in better, three samples are made in duplicates and incubated at
a Thermomixer from room temperature to 95 °C and maintain- 4 °C to inhibit active transport. Due to the difference in emis-
ing the temperature at 95 °C in total time of 24 min. The sion wavelength, the three dyes can be detected separately. Pro-
machine was then turned off and the sample was allowed to cessing the data with Image J gave overlay images whereby the
cool down slowly.
Loading buffer: 20 µL Tris 1 M, pH = 7.6, 20 µL KCl 0.2 M, red and green respectively.
0 µL MgCl 0.1 M, 40 µL EDTA 0.025 M.
Sucrose: 30% sucrose in mQ (300 mg mL ).
Peptides: 10 µL stock solutions (10×) were prepared in RAW264.7 cells cultivated in DMEM (10 per well, 1 ml) were
MilliQ water (0, 1.67, 5.01, 6.68, 7.51, 8.35, 10.02, 11.69, 13.36, plated in a 24-well plates and incubated overnight at 37 °C and
nucleus, cell membrane and fluorescein are shown in blue,
2
2
−
1
Flow cytometry
5
1
6.7 µM).
5% CO . Peptides were added in an overall concentration of
2
3860 | Org. Biomol. Chem., 2015, 13, 3856–3862
This journal is © The Royal Society of Chemistry 2015

View Article Online
Organic & Biomolecular Chemistry
Communication
0
.25 µM and 1 µM and incubated for 2 h at the same con-
ditions or on ice. Cells were washed with PBS and detached
with Na EDTA. Cells were re-suspended in PBS and added to
2 E. Pazos, J. Mosquera, M. E. Vázquez and J. L. Mascareñas,
ChemBioChem, 2011, 12, 1958–1973.
3 N. M. Luscombe, S. E. Austin, H. M. Berman and
J. M. Thornton, Genome Biol., 2000, 1, 1–10.
4 L. L. G. Carrette, T. Morii and A. Madder, Eur. J. Org.
Chem., 2014, 14, 2883–2891.
4
the BD Accuri flow cytometer. Experiments were carried out in
duplicate.
MTT assay
5
T. Morii, M. Simomura and S. Morimoto, J. Am. Chem. Soc.,
1993, 32, 1150–1151.
4
RAW264.7 cells cultivated in DMEM (10 per well, 200 µL) were
plated in a 96-well plates and incubated overnight at 37 °C and
6 A. M. Vázquez, M. Eugenio and J. L. Mascareñas, Angew.
Chem., Int. Ed., 2000, 112, 3234–3237.
5
% CO . The peptides and control compounds were added in
2
an overall concentration of 0.25 µM and 1 µM and incubated
for 24 h under the same conditions. The MTT solution was
then added to the aspirated wells and incubated for 3 h. After
removal of the cell medium, purple formazan crystals were dis-
solved in DMSO. Then, UV-measurement at 570 nm was per-
formed with a plate reader to check cell viability quantitatively.
7 M. I. N. Zhang, B. Wu, H. Zhao and J. W. Taylor, J. Pept.
Sci., 2002, 136, 125–136.
8 C. Park, J. L. Campbell and W. A. Goddard, J. Am. Chem.
Soc., 1996, 118, 4235–4239.
9 L. K. Henchey, A. L. Jochim and P. S. Arora, Curr. Opin.
Chem. Biol., 2008, 12, 22–26.
An observation with the naked eye already gave a good idea 10 J. Mosquera, A. Jiménez-Balsa, V. I. Dodero,
about the toxicity of the compounds due to the disappearance
of color in the well. Comparison of the absorbance of the for-
M. E. Vázquez and J. L. Mascareñas, Nat. Commun.,
2013, 4, 1874.
mazan solution of the sample to the absorbance of a positive 11 K. Estieu-gionnet and G. Guichard, Drug Discovery, 2011, 6,
control (incubation with DMSO) and a negative control (incu-
937–963.
bation with ultrapure water) gave quantitative results of cell 12 M. M. Madden, A. Muppidi, Z. Li, X. Li, J. Chen
viability as in the equation:
and Q. Lin, Bioorg. Med. Chem. Lett., 2011, 21,
472–1475.
1
A ꢀ Apos
Cell viability ð%Þ ¼
ꢁ 100
1
1
1
3 H. Inhibitor, H. Zhang, Q. Zhao, S. Bhattacharya,
A. A. Waheed, X. Tong, A. Hong, S. Heck, F. Curreli,
M. Goger, D. Cowburn, E. O. Freed and A. K. Debnath,
J. Mol. Biol., 2008, 378, 565–580.
4 C. J. Brown, S. T. Quah, J. Jong, A. M. Goh, P. C. Chiam,
K. H. Khoo, M. L. Choong, M. a. Lee, L. Yurlova,
K. Zolghadr, T. L. Joseph, C. S. Verma and D. P. Lane, ACS
Chem. Biol., 2013, 8, 506–512.
Aneg ꢀ Apos
Peptide stability
Peptides dissolved in a 50 mM NH
4
HCO
3
buffer at a concen-
−
1
tration of 0.5 mg mL are incubated at 37 °C with a trypsin
solution (trypsin/peptide: 1/1000 wt%) in AcOH 50 mM buffer.
The pH was optimized to 7–9. Samples were taken after 0 min,
0 min, 1 h, 2 h and 24 h and injected on RP-HPLC (Jupiter C4
00A, 0–100% CH CN in 15 min). Peaks were collected and
3
3
5 H. Jo, N. Meinhardt, Y. Wu, S. Kulkarni, X. Hu,
K. E. Low, P. L. Davies, W. F. DeGrado and
D. C. Greenbaum, J. Am. Chem. Soc., 2012, 134, 17704–
3
analysed using MALDI-TOF. Peaks were integrated. The peak
area at 0 min was used as control.
1
7713.
6 Y.-W. Kim, T. N. Grossmann and G. L. Verdine, Nat. Protoc.,
011, 6, 761–771.
1
1
2
Acknowledgements
7 Ø. Jacobsen, H. Maekawa, N. Ge, C. H. Görbitz and
P. Rongved, J. Org. Chem., 2011, 76, 1228–1238.
8 W. S. Horne, Chem. Commun., 2011, 47, 1–4.
9 Y. Demizu, N. Yamagata, S. Nagoya, Y. Sato and M. Doi,
Tetrahedron, 2011, 67, 6155–6165.
0 L. D. Walensky and G. H. Bird, J. Med. Chem., 2014, 57,
Tim Courtin is acknowledged for the determination of the con-
centration of the final compounds by ERECTIC-NMR. Ellen
Gyssels is thanked for her assistance in determining the con-
centration of the DNA strands and fluorescently labeled pep-
tides using the Trinean DropSense 96. Jan Goeman is
acknowledged for the LC-MS analysis. Abhishek Iyer and Yara
Ruiz García are indebted to the Marie Curie Early Stage
Research Training Fellowship of the European Community’s
Seventh Framework Programme under contract number
PITN-GA-2010-238679. The FWO and BOF are also acknowl-
edged for financial support.
1
1
2
2
2
6275–6288.
1 T. E. Ellenberger, C. J. Brandl, K. Struhl and S. C. Harrison,
Cell, 1992, 71, 1223–1237.
2 A. Jimønez-balsa, E. Pazos, B. Martínez-albardonedo,
J. L. Mascareñas and M. E. Vázquez, Angew. Chem., Int. Ed.,
2
012, 51, 8825–8829.
3 W. Keller, P. Ko and T. J. Richmond, J. Mol. Biol., 1995, 254,
57–667.
4 N. J. Zondlo and A. Schepartz, J. Am. Chem. Soc., 1999, 121,
938–6939.
2
2
6
Notes and references
6
1
D. S. Latchman, Eukaryotic Transcription Factors, Elsevier, 25 L. Guerrero, O. S. Smart, G. A. Woolley and R. K. Allemann,
London, 5th edn, 2011, ch. 1, p. 1. J. Am. Chem. Soc., 2005, 127, 15624–15629.
This journal is © The Royal Society of Chemistry 2015
Org. Biomol. Chem., 2015, 13, 3856–3862 | 3861

View Article Online
Communication
Organic & Biomolecular Chemistry
2
6 S. Futaki, T. Suzuki, W. Ohashi, T. Yagami, S. Tanaka, 29 W. Nomura, H. Aikawa, N. Ohashi, E. Urano, M. Métifiot,
K. Ueda and Y. Sugiura, J. Biol. Chem., 2001, 276, 5836–
840.
7 Ü. Langel, Handbook of Cell-Penetrating Peptides, CRC Press,
nd edn, 2006, ch. 25, p. 440.
8 A. Muppidi, Z. Wang, X. Li, J. Chen and Q. Lin, Chem. 30 L. D. Walensky and G. H. Bird, J. Med. Chem., 2013, 57,
Commun., 2011, 47, 9396–9398. 6275–6288.
M. Fujino, K. Maddali, T. Ozaki, A. Nozue, T. Narumi,
C. Hashimoto, T. Tanaka, Y. Pommier, N. Yamamoto,
J. a. Komano, T. Murakami and H. Tamamura, ACS Chem.
Biol., 2013, 8, 2235–2244.
5
2
2
2
3862 | Org. Biomol. Chem., 2015, 13, 3856–3862
This journal is © The Royal Society of Chemistry 2015