Pretubulysin derived probes as novel tools for monitoring the microtubule network via activity-based protein profiling and fluorescence microscopy

Article abstract of DOI:10.1039/c2mb25144b

Microtubules (mt) are highly dynamic polymers composed of alpha- and beta-tubulin monomers that are present in all dividing and non-dividing cells. A broad variety of natural products exists that are known to interfere with the microtubule network, by eit

Full text of DOI:10.1039/c2mb25144b

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Molecular
BioSystems
Cite this: Mol. BioSyst., 2012, 8, 2067–2075
www.rsc.org/molecularbiosystems
PAPER
Pretubulysin derived probes as novel tools for monitoring the microtubule
network via activity-based protein profiling and fluorescence microscopyw
Ju
¨
rgen Eirich,a Jens L. Burkhart,b Angelika Ullrich,b Georg C. Rudolf,a
Angelika Vollmar,c Stefan Zahler,c Uli Kazmaierb and Stephan A. Sieber*a
Received 13th April 2012, Accepted 18th May 2012
DOI: 10.1039/c2mb25144b
Microtubules (mt) are highly dynamic polymers composed of alpha- and beta-tubulin monomers
that are present in all dividing and non-dividing cells. A broad variety of natural products exists
that are known to interfere with the microtubule network, by either stabilizing or de-stabilizing
these rope-like polymers. Among those tubulysins represent a new and potent class of cytostatic
tetrapeptides originating from myxobacteria. Early studies suggested that tubulysins interact with
the eukaryotic cytoskeleton by inhibition of tubulin polymerization with EC50 values in the
picomolar range. Recently, pretubulysins have been described to retain the high tubulin-
degradation activity of their more complex tubulysin relatives and represent an easier synthetic
target with an efficient synthesis already in place. Although tubulin has been suggested as the
dedicated target of tubulysin a comprehensive molecular target analysis of pretubulysin in the
context of the whole proteome has not been carried out so far. Here we utilize synthetic chemistry
to develop two pretubulysin photoaffinity probes which were applied in cellular activity-based
protein profiling and imaging studies in order to unravel and visualize dedicated targets.
Our results clearly show a remarkable selectivity of pretubulysin for beta-tubulin which we
independently confirmed by a mass-spectrometry based proteomic profiling platform as well as
by tubulin antibody based co-staining on intact cells.
academic research. In 2000 Sasse et al. discovered a new class
Introduction
of cytostatic tetrapeptides originating from myxobacteria
termed tubulysins 1 (Fig. 1).10 These molecules are assembled
by hybrid polyketide synthases/non-ribosomal peptide syn-
thetases and consist of one proteinogenic, isoleucine (Ile),
and three non-proteinogenic amino acids including N-methyl-
pipecolic acid (Mep), tubuvaline (Tuv) and a chain extended
analogue of either phenylalanine or tyrosine called tubu-
phenylalanine (Tup) or tubutyrosine (Tut), respectively. Early
studies suggested that tubulysins interact with the eukaryotic
cytoskeleton by inhibition of tubulin polymerization with
EC50 values in the picomolar range.10–12 This potency even
exceeds that of other tubulin modifiers such as taxol, vinblastine
and epothilone by 20- to 100-fold13 and consequently rank
tubulysins as the most promising lead structures for pharma-
ceutical application. However, several prerequisites have to be
considered in order to initiate a drug development program.
First, easy synthetic access for natural product derivatization
has to be established in order to investigate structure–activity
relationships (SAR) and provide sufficient compound quanti-
ties for pharmacological testing. Moreover, although tubulin
has been established as the dedicated target, no information
about undesired off-targets in eukaryotic cells is available.
Previous SAR studies already revealed several positions with
increased tolerance for derivatization.13–17 One important
Microtubules (mt) are highly dynamic polymers composed of
a- and b-tubulin monomers that are present in all dividing and
non-dividing cells.1,2 These components of the cytoskeleton
play a major role in forming the spindle apparatus during
mitosis, which was recognized early on as an interesting drug
target for selective inhibition of the growth of fast dividing
cells, e.g. in cancer.3–7
A broad variety of natural products originating from
different organisms exists which are known to interfere with
the microtubule network, by either stabilizing or de-stabilizing
these rope-like polymers already since the 1970s.5,6,8,9 Some
of them are marketed drugs, in clinical trials or subject to
a Center for Integrated Protein Science Munich CIPSM
Department of Chemistry, Institute of Advanced Studies IAS,
,
Technische Universitat Munchen, Lichtenbergstrasse 4,
¨
85747 Garching, Germany
¨
b Institut fur Organische Chemie, Universitat des Saarlandes,
¨
¨
Im Stadtwald, Geb. C4.2, 66123 Saarbrucken, Germany
¨
c Department of Pharmacy, Pharmaceutical Biology,
Ludwig-Maximilians-Universitat Munchen, Butenandtstr. 5-13,
81377 Munchen, Germany
¨
¨
¨
w Electronic supplementary information (ESI) available: Additional
fluorescent gel scans, MS data, compound structures and synthesis of
trifunctional linkers, synthetic procedures for pretubulysin probes. See
DOI: 10.1039/c2mb25144b
c
This journal is The Royal Society of Chemistry 2012
Mol. BioSyst., 2012, 8, 2067–2075 2067

View Article Online
Fig. 1 Probe design – structures of the naturally occurring tubulysins 1 and pretubulysin (Pt) 2 (left) as well as derived probes (Pt propargyl amide
3, benzophenone probes 4 and 5) (right) with common core motive indicated in blue.
finding included the replacement of a chemically labile
N,O-acetal of tubulysin 1 by alkyl groups and the removal
of the acetoxy group at the central amino acid tubuvaline
(Tuv). The resulting chemically less complex compounds
are called pretubulysins (Pt) 2 and have been also found in
nature as biosynthetic precursors of the tubulysins (Fig. 1).18
Interestingly, pretubulysins retain the high tubulin-degradation
activity of their more complex tubulysin relatives and repre-
sent an easier synthetic target with an efficient synthesis
already in place.13,19 Here we utilize this well-established route
to develop two pretubulysin photoaffinity probes 4 and 5
which are applied in cellular activity based protein profiling
and imaging studies in order to unravel and visualize dedicated
targets (Fig. 2).20–24 Our results clearly show a remarkable
selectivity of pretubulysin for tubulin which we independently
confirmed by a mass-spectrometry based proteomic profiling
platform as well as by tubulin antibody based co-staining on
intact cells.
20 mL (5 mg mLÀ1) filtered MTT stock solution in PBS were
added and the medium was mixed by gentle pipetting. The
cells were incubated at 37 1C and 5% CO2 for 1.5 h to allow
the MTT to be metabolized and the reaction was controlled
under the microscope. Subsequently, the medium was
removed and the produced formazan resuspended in 200 mL
DMSO by placing the well-plate on a shaking table for 2 min
and 650 rpm. The optical density was read out at l = 570 nm
and l = 630 nm with a Tecan Infinite 200 PRO NanoQuant
microplate reader and the background was subtracted at
l = 630 nm. Cells incubated with 1% DMSO served as
control. Three independent replicates were conducted.
The EC50 values were determined using the non-linear fit for
dose response of OriginPRO 8.5.1.
In vitro analytical labeling experiments
Cells were detached by scraping from the culture dishes.
The pellet was homogenized in PBS by sonication with
a Bandelin Sonopuls instrument with 5 Â 15 s pulsed at
70% max. power under ice cooling. Proteome samples
were adjusted to a final concentration of 2 mg protein mLÀ1
by dilution in PBS prior to probe labeling. The experi-
ments were carried out in 43 mL total volume, such that once
click chemistry (CC) reagents were added, a total reaction
volume of 50 mL was reached. In the case of competitive
displacement experiments Pt was added and the proteomes
were preincubated with the compound for 15 min at 22 1C.
Reactions were initiated by addition of varying concentrations
of probes and allowed to incubate (additionally) for 15 min at
22 1C. Then they were irradiated in open polystyrene micro-
well plates for 60 min on ice with 366 nm UV light (Benda
UV hand lamp NU-15 W). For heat controls the proteome
was denatured with 2% SDS (4 mL of 21.5% SDS) at 95 1C
for 6 min and cooled to 22 1C before the probe was
applied. Following incubation, 13 mM rhodamine-azide
(Rh-N3) (1 mL) followed by 1 mM TCEP (1 mL) and 100 mM
TBTA ligand (3 mL) were added. Samples were gently vortexed
Methods and materials
Synthesis
For synthetic details please refer to the ESI.w
Cell culture
The cell lines Hela, H460 and Jurkat were grown in RPMI-
1640, A549 in DMEM (high glucose) with 10% FBS in 5%
CO2 at 37 1C. The adherently growing cell lines were detached
with trypsin-EDTA.
MTT assay
Cells were grown in 96-well-plates at a concentration of
5000 cells per cavity. After removing the growth medium by
suction compounds 2, 3, 4 and 5 in 100 mL medium (without FBS)
with a final DMSO concentration of 1% were added and the cells
incubated for 24 h at 37 1C and 5% CO2. After 24 h of exposure,
c
2068 Mol. BioSyst., 2012, 8, 2067–2075
This journal is The Royal Society of Chemistry 2012

View Article Online
Fig. 2 Experimental design – general labeling procedure with UV-induced cross-link between target protein and probe in living cells (left) enables
a broad variety of down-stream applications such as MS based proteomics as well as fluorescence imaging depending on the reporter tag
introduced via bio-orthogonal click-chemistry (right).
and the cycloaddition was initiated by the addition of 1 mM
CuSO4 (1 mL).
protein binding on avidin-agarose beads. After CC proteins
were precipitated using a 5-fold volume of prechilled acetone,
samples were stored at À20 1C for 60 min and centrifuged at
16 000g for 10 min. The supernatant was discarded and the
pellet washed two times with 200 mL of prechilled methanol
and resuspended by sonication. Subsequently, the pellet was
dissolved in 1 mL of PBS with 0.2% SDS by sonication and
incubated under gentle mixing with 50 mL of avidin-agarose
beads (Sigma-Aldrich) for 1 h at room temperature. The beads
were washed three times with 1 mL of PBS/0.2% SDS, twice
with 1 mL of 6 M urea, and three times with 1 mL of PBS.
50 mL of 2 Â SDS loading buffer were added and the proteins
were released for preparative SDS-PAGE by 6 min incubation
at 95 1C. Gel bands were isolated, washed, and tryptically
digested as described previously.33
The reaction mixtures were incubated at 22 1C for 1 h.
For gel electrophoresis, 50 mL of 2 Â SDS loading buffer was
added, and 50 mL applied on the gel. Fluorescence was
recorded in a Fujifilm LAS4000 luminescent image analyzer
with a Fujinon VRF43LMD3 lens and a 575DF20 filter.
In situ analytical and preparative experiments
For preparative in situ studies cells were grown to 80–90%
confluence in cell culture petri dishes. After suction and
washing with PBS, compound 5 was added in 15 mL medium
without FCS and a final DMSO concentration of 0.1%. The
cells were incubated with the probe for different periods of
time at 37 1C and 5% CO2. The medium was removed and the
cells were covered with PBS prior to UV irradiation (0 1C, 1 h,
366 nm). The cells were detached with a cell scraper, washed
with PBS and the pellet homogenized by sonication. The
proteomes were separated into cytosolic and membrane frac-
tions, followed by CC either in 43 mL for analytical ABPP
experiments or in a final volume of 1 mL of the proteome
sample with 20 mM trifunctional linker (2 mL), 1 mM TCEP
(10 mL) and 100 mM TBTA ligand (30 mL) for preparative
ABPP. Samples were gently vortexed, and the cycloaddition
was initiated by the addition of 1 mM CuSO4 (10 mL).
Mass spectrometry and bioinformatics
Tryptic peptides were loaded onto a Dionex C18 Nano Trap
Column (100 mm) and subsequently eluted and separated by a
Dionex C18 PepMap 100 mm (3 mm) column for analysis
by tandem MS followed by high-resolution MS using a
coupled Dionex Ultimate 3000 LC-Thermo LTQ Orbitrap
XL system. The mass spectrometry data were searched using
the SEQUEST algorithm against the corresponding databases
via the software ‘‘Proteome Discoverer 1.3’’. The search
was limited to only tryptic peptides, two missed cleavage
sites, monoisotopic precursor ions, and a peptide tolerance
of o5 ppm.33,41
Reactions for enrichment were carried out together with a
control lacking the probe to compare the results of the biotin-
avidin-enriched samples with the background of unspecific
c
This journal is The Royal Society of Chemistry 2012
Mol. BioSyst., 2012, 8, 2067–2075 2069

View Article Online
Fluorescence imaging
Hela cells were seeded in carriers compatible with high resolu-
tion microscopy (m-slides, Ibidi). After two days, cells were
incubated with the respective pretubulysin derivatives for 1 to
30 min. Subsequently the cells were irradiated (366 nm) at
37 1C (in order to keep the microtubules intact) for 10 min,
and then rapidly fixed in methanol (À20 1C). After three
washing steps in PBS, cells were incubated for 20 min with a
freshly prepared solution containing 775 ml PBS, 20 ml CuSO4
(50 mM), 5 ml dye-azid (5 mM), and 200 ml ascorbic acid
(500 mM). After three times washing with PBS, the cells were
incubated with primary tubulin antibody (1 : 400, abcam
rabbit anti a tubulin) for 60 min, washed again three times,
and then incubated with a secondary antibody (goat anti
rabbit AlexaFluor 488, Invitrogen, 1 : 400) and 1 mg mLÀ1
HOECHST for nuclear staining for 45 min. After three final
washing steps, the samples were mounted with Permafluor
(Thermo Scientific) and glass cover slides. The samples were
observed on a confocal laser scanning microscope (LSM510
meta, Zeiss) with a 63Â lens, using a pinhole diameter, which
generates optical slices of 1 mm thickness.
Scheme 2 Synthtic route leading to pB-Tup precursor 10.
reaction to yield the p-iodo-tubuphenylalanine derivative (9).
This compound was further modified with the above-
mentioned carbonylative Suzuki coupling to yield derivative
10 which closely resembles the desired building block with
the exception of the double bond that is saturated in the
natural product (Scheme 2). All attempts to selectively hydro-
genate this bond failed due to the loss of the carbonyl function
within the benzophenone moiety. We therefore used the
unsaturated precursor building block (10 – Scheme 2) together
with the shorter pB-Phe in order to assemble two structurally
different pretubulysin probes 4 and 5. The synthesis of both
probes is outlined in Schemes 3 and 4. In brief, the two
benzophenone precursors 7 and 10 were deprotected with
HCl in dioxane and subsequently coupled with N-Boc-
tubu-valine which was activated as a mixed anhydride. The
N-terminal tripeptide was prepared according to literature
procedures.18 Subsequent deprotection and coupling with
Cbz-(S,S)-isoleucine as well as (R)-N-methyl-pipecolic acid-
pentafluorophenyl ester ((R)-Mep-OPFP), after Cbz-deprotection,
gave the two N-methylated tetrapeptides (13a/b) in good yields
(Scheme 3). In the next step, the methyl, respectively, ethyl
ester was saponified with NaOH and the free tetrapeptides
were converted to the TFA-salts to get the benzophenone
derivatives 14 in excellent yield. Finally, propargylamine was
coupled to the C-terminus yielding the desired probes 4 and 5
(Scheme 4).
Results and discussion
Chemical probe design and synthesis
One major challenge in the design of photoaffinity probes for
target discovery is the introduction of the photolinker as well
as the tag which is required as a reporter for protein identifi-
cation. While the tag is usually a small alkyne moiety that can
be modified by post-labeling via the Huisgen–Sharpless–Meldal
cycloaddition25–28 the photolinker is bulky and may cause a
steric clash in the active site pocket/binding site of the target
protein (Fig. 2).29,30 Therefore, a careful chemical design is
required in order to incorporate the photolinker in a way that
it is at least partially embedded into the compound scaffold.
As pretubulysin 2 exhibits a C-terminal tubuphenylalanine
(Tup) moiety which can be modified with only little loss of
activity according to SAR14–17 studies we decided to expand its
aromatic system by carbonylative cross coupling to a photoactive
benzophenone moiety. In order to assemble the pretubulysin
photo-probe we first established a reliable synthetic procedure
for the p-benzoyl-phenylalanine (pB-Phe) building block
(Scheme 1) before the more complex p-benzoyl-tubuphenyl-
alanine (pB-Tup) moiety was approached. Carbonylative
Suzuki coupling of Boc-p-iodophenylalanine methylester (6),
prepared in three steps according to the literature procedure,31
with phenylboronic acid gave the desired product 7 in very
good yield (83%). For the synthesis of pB-Tup, 6 was first
reduced to the corresponding aldehyde with DibalH and
subsequently coupled with Wittig reagent 8 in a one-pot
In order to gain insights into the structure–activity relationship
(SAR) of the probe scaffolds we also prepared pretubulysin D (2)
as well as its C-terminal alkynylated derivative (3) as reference
compounds according to previous procedures.18
Bioactivity
Pretubulysin 2 exhibits very potent cytotoxicity against various
cell-lines with EC50 values in the low nM range (Table 1).18,19,32
In order to investigate the effect of structural remodeling on
probe potency we tested all three derivatives against various
cell lines and obtained a significant drop in potency for probes
4 and 5. The bioactivity of alkynylated derivative 3 was
significantly better compared to the photo-probes emphasizing
a higher tolerance for small modifications at the C-terminus.
Although the introduction of the additional benzoyl moiety at
the phenyl ring of tubuphenylalanine represented the least
structural perturbation in order to gain access to benzo-
phenone possible, the drop in potency indicated that this position
is quite sensitive for alterations. Nevertheless, the obtained
Scheme 1 Carbonylative cross-coupling leading to pB-Phe 7.
2070 Mol. BioSyst., 2012, 8, 2067–2075
c
This journal is The Royal Society of Chemistry 2012

View Article Online
Scheme 3 Reactions leading to cross-linkable Pt derivatives 14 with deprotected C-terminus.
Scheme 4 Final amide coupling reaction to introduce alkyne tag yielding desired probes 4 and 5.
EC50s of 8–13 mM for both probes still reflect elevated
Pretubulysin target analysis
cytotoxicities with an unresolved mechanism of action. Since
all results of cellular profiling will be validated with the
unmodified natural product, e.g. in competitive profiling
experiments (see below), the probes represent suitable tools
and a promising starting point for target discovery.
With the tool compounds in hand we next analyzed the
cellular target(s) that are responsible for the observed cyto-
toxicity. Both probes were incubated for 1 h with intact Hela
cells under in situ conditions (Experimental section) and
c
This journal is The Royal Society of Chemistry 2012
Mol. BioSyst., 2012, 8, 2067–2075 2071

View Article Online
Table 1 EC50 values for compounds 2, 3, 4 and 5 determined via
MTT (24 h incubation time) in four different cell lines
protein and the probe (Fig. S2, ESIw). A comparison of both
concentration dependent labeling experiments suggests that
probe 4 with the shorter C-terminal end exhibits a more
pronounced selectivity for the above-mentioned protein
target compared to probe 5 (Fig. S3, ESIw). Based on the
better labeling profile as well as the slightly higher cytotoxicity
we conducted all subsequent labeling experiments with
probe 4. In order to evaluate the optimal pre-incubation time
probe 4 was incubated with intact Hela cells for several
durations ranging from 5 to 120 min. Subsequent irradia-
tion for 1 h and in gel fluorescent scanning revealed that
60 min were already sufficient for effective labeling (Fig. S4,
ESIw).
In order to evaluate which of the labeled protein bands in
Hela cells correspond to pretubulysin binding, we added the
unmodified natural product in varying excess to intact cells
followed by the probe and applied the standard irradiation
labeling procedure. Interestingly, a 4-fold excess of pretu-
bulysin was already sufficient to totally block the labeling of
the prominent 60 kDa band which emphasizes a high affinity
of the unmodified natural product for the protein binding site.
Profiling of several other cell lines including H460, Jurkat and
A549 with probe 4 at a concentration close to the EC50 (10 mM)
revealed an intense protein band at the same molecular weight
as the one observed in Hela emphasizing that the target is highly
conserved throughout the different cell lines.
Compound/mM
Cell line
2 (Pt)
3
4
5
Hela
0.02
0.02
1.2
1.0
0.5
40
7.9
8.1
8.5
27
13.4
9.6
9.2
43
H460
Jurkat
A549
o0.01
3.9
irradiated for 1 h at 360 nm in order to establish a covalent
link between the probe molecule and all bound proteins.33
Cells were lyzed by sonication and cytosolic as well as
membrane fractions separated by centrifugation. Subsequent
CC with a fluorescent rhodamine-azide tag followed by SDS
polyacrylamide gel separation and fluorescent scanning
revealed distinct bands that varied in intensity depending on
the applied probe concentration (Fig. 3 and Fig. S1, ESIw).34–36
While 500 mM, the highest concentration applied, led to the
labeling of several proteins, a gradual reduction of probe
concentration down to 1–10 mM significantly reduced the
number of protein bands. One intense band with a molecular
weight of about 60 kDa was labeled under all probe concen-
trations applied emphasizing a significant specificity and affi-
nity of the probes for this protein target. Contrarily, labeling
of this band disappeared when the Hela cell lysate was first
denatured by heat treatment and subsequently labeled by the
probe emphasizing a specific interaction between the folded
To reveal the identity of this protein band we utilized a
proteomic enrichment strategy and incubated Hela cells with
Fig. 3 Fluorescence scans of the in situ labeling in Hela cells with probe 4 (1–500 mM) in competition with Pt 2 (2 mM) (A) and in comparison
with other cell lines (10 mM 4) (B), selective enrichment of tubulin in H460 (5 mM 4) (C).
c
2072 Mol. BioSyst., 2012, 8, 2067–2075
This journal is The Royal Society of Chemistry 2012

View Article Online
Fig. 4 Fluorescent microscopy: A–D: top left – blue Hoechst, top right – red Rh-N3, bottom left – green Antibody-Alexa488-conjugate, bottom
right – overlay. Probe treatment of cells indicated individually.
probe 4 followed by irradiation, lysis and CC in order to
attach a trifunctional rhodamine-biotin-azide tag (ESIw).33
The labeled protein was selectively enriched by binding to
avidin beads and separated on a SDS gel after heat cleavage
(Fig. 3c). The enriched band was isolated, washed, tryptically
digested and subject to mass spectrometric analysis. Peptide
fragments were analyzed via the proteome discoverer software
with the SEQUEST search algorithm. All experiments were
carried out with a control in which no probe was added in
order to subtract proteins that bind unspecifically to avidin.
All experiments conclusively revealed tubulin as the primary
target of the probes (Table S1, ESIw). While inhibition of
tubulin polymerization by tubulysin has been shown in in vitro
assays previously,10–12,37 it is an important result that the
structurally less complex precursor pretubulysin also interacts
specifically with tubulin explaining its potent bioactivity.
Although our probes are significantly less potent compared
to the unmodified natural product, competitive labeling
demonstrates that pretubulysin specifically binds to tubulin
with high affinity. However, based on our ABPP experi-
ments we cannot exclude that other pretubulysin targets may
exist which are not detectable by our probes. Due to the
restrictions of ABPP, e.g. photocrosslinker, this problem
cannot be easily resolved. Since tubulin represents at least
a major target of pretubulysin which was detected for the
first time via full proteome analysis we will focus on this
protein and study the cellular effect of tubulin binding in more
detail.
Fluorescence imaging
Besides target identification another advantage of the probe
molecules is their utility in imaging studies which is suitable
to independently confirm the target and reveal insights into
c
This journal is The Royal Society of Chemistry 2012
Mol. BioSyst., 2012, 8, 2067–2075 2073

View Article Online
the function and possible effects of binding. We therefore
incubated living Hela cells with various concentrations of
probe 4, irradiated the cells to covalently attach the molecule
to the target protein and subsequently fixed the cells with
methanol after different time points ranging from 1 to 30 min.
Fixed cells were reacted with CC reagents (rhodamine-azide,
ascorbic acid, ligand and CuSO4), Hoechst DNA stain as well
as tubulin antibodies for the visualization of the probe, the
nucleus as well as microtubuli, respectively. A control experi-
ment without the probe but in the presence of CC reagents
revealed low background binding of the rhodamine dye which
is mandatory for the visualization of specific binding partners
(Fig. 4a). While these pretubulysin untreated cells reveal a
tight mesh of microtubuli, a 1 min incubation of 3 mM probe 3
already initiates significant damage of the filament assembly
(Fig. 4b). Interestingly, the same phenotype has been reported
for pretubulysin previously38–40 but so far the direct binding to
tubulin could not be directly confirmed. Here we demonstrate
by co-staining with the tubulin directed antibody as well as the
fluorescent probe that pretubulysin derivative 4 binds to mono-
meric tubulin which appear as little spots in the image (Fig. 4c
and d). The sequestration of tubulin monomers prevents the
filament assembly and explains its time dependent dissolution.
The results of all imaging experiments are in line with our MS
based target analysis and demonstrate that monomeric tubulin
is the point of pretubulysin attachment leading to inhibition of
filament assembly and subsequent cell death.
J.E. was supported by the Deutsche Forschungsgemeinschaft
(DFG) FOR1406. G.C.R. was supported by an ERC starting
grant. J.E. and G.C.R. were supported by the TUM Graduate
School (Faculty Center Chemistry). J.L.B., A.U., A.M.V.,
S.Z. and U.K. were supported by the DFG FOR1406.
S.A.S. was supported by the DFG Emmy Noether, SFB749
and FOR1406, an ERC starting grant and the Center for
Integrated Protein Science Munich CIPSM.
References
1 J. B. Olmsted and G. G. Borisy, Biochemistry, 1973, 12,
4282–4289.
2 J. B. Olmsted and G. G. Borisy, Annu. Rev. Biochem., 1973, 42,
507–540.
3 C. Dumontet and M. A. Jordan, Nat. Rev. Drug Discovery, 2010, 9,
790–803.
4 L. Wilson and M. A. Jordan, J. Chemother. (Firenze, Italy), 2004,
16(Suppl. 4), 83–85.
5 R. A. Stanton, K. M. Gernert, J. H. Nettles and R. Aneja, Med.
Res. Rev., 2011, 31, 443–481.
6 A. Jordan, J. A. Hadfield, N. J. Lawrence and A. T. McGown,
Med. Res. Rev., 1998, 18, 259–296.
7 P. Giannakakou, D. Sackett and T. Fojo, J. Natl. Cancer Inst.,
2000, 92, 182–183.
8 P. B. Schiff, J. Fant and S. B. Horwitz, Nature, 1979, 277,
665–667.
9 M. C. Wani, H. L. Taylor, M. E. Wall, P. Coggon and
A. T. McPhail, J. Am. Chem. Soc., 1971, 93, 2325–2327.
10 F. Sasse, H. Steinmetz, J. Heil, G. Hofle and H. Reichenbach,
J. Antibiot., 2000, 53, 879–885.
11 M. W. Khalil, F. Sasse, H. Lunsdorf, Y. A. Elnakady and
H. Reichenbach, ChemBioChem, 2006, 7, 678–683.
12 H. Steinmetz, N. Glaser, E. Herdtweck, F. Sasse, H. Reichenbach
and G. Hofle, Angew. Chem., Int. Ed., 2004, 43, 4888–4892.
13 A. Ullrich, Y. Chai, D. Pistorius, Y. A. Elnakady, J. E. Herrmann,
K. J. Weissman, U. Kazmaier and R. Muller, Angew. Chem.,
Int. Ed., 2009, 48, 4422–4425.
14 R. Balasubramanian, B. Raghavan, J. C. Steele, D. L. Sackett and
R. A. Fecik, Bioorg. Med. Chem. Lett., 2008, 18, 2996–2999.
15 B. Raghavan, R. Balasubramanian, J. C. Steele, D. L. Sackett and
R. A. Fecik, J. Med. Chem., 2008, 51, 1530–1533.
16 P. Wipf and Z. Y. Wang, Org. Lett., 2007, 9, 1605–1607.
17 A. W. Patterson, H. M. Peltier and J. A. Ellman, J. Org. Chem.,
2008, 73, 4362–4369.
From a chemical biology perspective photocrosslinking
probe 4 revealed superior imaging properties over compound
3 that lacks the photoreactive moiety. Although compound 3
exhibited a stronger effect on microtubule assembly, the
reversible binding was not stable enough to survive the washing
procedures. Since photo-linkage is mandatory for the visuali-
zation of pretubulysin binding, a next generation of photo-
crosslinking probes should focus on the introduction of
smaller linker moieties such as diazirines in order to maintain
the original potency.
In conclusion we designed and synthesized customized
pretubulysin probes that revealed tubulin as the main target
in living human cells. Although the introduction of the
photocrosslinking moiety reduced the biological potency,
competitive labeling studies revealed that beta tubulin is the
dedicated binding partner of the unmodified natural product.
Moreover, the application of both probes at concentrations
close to their EC50 value (10 mM) revealed only beta tubulin as
the predominant target emphasizing that this interaction is
solely responsible for cell toxicity of these molecules. However,
we cannot exclude that additional targets of pretubulysin may
exist that could not be identified here due to the steric bulk of
the photolinker. The results obtained in this study demon-
strate the utility of our approach to identify tubulin as the
main target within living cells of different origins as well as to
monitor tubulin binding in cells and its associated inhibition of
microtubuli assembly.
18 (a) A. Ullrich, J. Herrmann, R. Muller and U. Kazmaier, Eur. J.
Org. Chem., 2009, 6367–6378; (b) J. Herrmann, Y. A. Elnakady,
R. M. Wiedmann, A. Ullrich, M. Rohde, U. Kazmaier, A. M.
Vollmar and R. Muller, PLoS ONE, 2012, DOI: 10.1371/journal.
pone.0037416, accepted manuscript.
19 J. L. Burkhart, R. Muller and U. Kazmaier, Eur. J. Org. Chem.,
2011, 3050–3059.
20 B. F. Cravatt and E. J. Sorensen, Curr. Opin. Chem. Biol., 2000, 4,
663–668.
21 M. J. Evans, A. Saghatelian, E. J. Sorensen and B. F. Cravatt, Nat.
Biotechnol., 2005, 23, 1303–1307.
22 M. J. Evans and B. F. Cravatt, Chem. Rev., 2006, 106, 3279–3301.
23 T. Bottcher, M. Pitscheider and S. A. Sieber, Angew. Chem., Int. Ed.,
2010, 49, 2680–2698.
24 M. Fonovic and M. Bogyo, Expert Rev. Proteomics, 2008, 5,
721–730.
25 R. Huisgen, 1,3 Dipolar Cylcoaddition Chemistry, Wiley, New York,
1984.
26 V. V. Rostovtsev, J. G. Green, V. V. Fokin and K. B. Sharpless,
Angew. Chem., Int. Ed., 2002, 41, 2596–2599.
27 Q. Wang, T. R. Chan, R. Hilgraf, V. V. Fokin, K. B. Sharpless and
M. G. Finn, J. Am. Chem. Soc., 2003, 125, 3192–3193.
28 C. W. Tornøe, C. Christensen and M. Meldal, J. Org. Chem., 2002,
67, 3057–3064.
Acknowledgements
29 Y. Tanaka, M. R. Bond and J. J. Kohler, Mol. BioSyst., 2008, 4,
473–480.
30 M. B. Nodwell and S. A. Sieber, Top. Curr. Chem., 2011, 1–42.
J.E., G.C.R. and S.A.S. thank Mona Wolff for excellent
support and Matthew Nodwell for comments on the manuscript.
c
2074 Mol. BioSyst., 2012, 8, 2067–2075
This journal is The Royal Society of Chemistry 2012

View Article Online
31 H. Y. Lei, M. S. Stoakes, K. P. B. Herath, J. H. Lee and
A. W. Schwabacher, J. Org. Chem., 1994, 59, 4206–4210.
32 A. Ullrich, Y. Chai, D. Pistorius, Y. A. Elnakady, J. E. Herrmann,
K. J. Weissman, U. Kazmaier and R. Muller, Angew. Chem., Int. Ed.,
2009, 48, 4422–4425.
37 Y. A. Elnakady, F. Sasse, H. Lunsdorf and H. Reichenbach,
Biochem. Pharmacol., 2004, 67, 927–935.
38 S. Rath, R. Furst, A. Ullrich, J. Burkhart, A. M. Vollmar,
U. Kazmaier and S. Zahler, Naunyn—Schmiedeberg’s Arch.
Pharmacol., 2011, 383, 28–28.
33 J. Eirich, R. Orth and S. A. Sieber, J. Am. Chem. Soc., 2011, 133,
12144–12153.
34 A. E. Speers, G. C. Adam and B. F. Cravatt, J. Am. Chem. Soc.,
2003, 125, 4686–4687.
35 A. E. Speers and B. F. Cravatt, ChemBioChem, 2004, 5,
41–47.
36 A. E. Speers and B. F. Cravatt, Chem. Biol., 2004, 11, 535–546.
39 V. Kretzschmann, A. Ullrich, S. Zahler, A. M. Vollmar,
U. Kazmaier and R. Furst, J. Vasc. Res., 2011, 48, 113–113.
40 S. Rath, A. Ullrich, J. L. Burkhardt, J. Herrmann, M. Gunther,
U. Kazmaier, A. M. Vollmar, J. Liebl, R. Furst and S. Zahler,
J. Vasc. Res., 2011, 48, 231–231.
41 T. Bottcher and S. A. Sieber, Angew. Chem., Int. Ed., 2008, 47,
4600–4603.
c
This journal is The Royal Society of Chemistry 2012
Mol. BioSyst., 2012, 8, 2067–2075 2075