Adenosine–oligoarginine conjugate, a novel bisubstrate inhibitor, effectively dissociates the actin cytoskeleton Helin Ra¨ a¨ gel1,2, Marje Lust3, Asko Uri3 and Margus Pooga1,2
1 Institute of Molecular and Cell Biology, University of Tartu, Estonia 2 Estonian Biocentre, Tartu, Estonia 3 Institute of Chemistry, University of Tartu, Estonia
Keywords actin; cell-penetrating peptide; protein kinase inhibitor; Rho-kinase; Y-27632
Correspondence M. Pooga, Estonian Biocentre, Riia 23B-226, University of Tartu, 51010 Tartu, Estonia Fax: +3727420286 Tel: +3727375024 E-mail: mpooga@ebc.ee
(Received 11 April 2008, revised 8 May 2008, accepted 14 May 2008)
doi:10.1111/j.1742-4658.2008.06506.x
Aberrant regulation of protein kinases impairs normal cellular functioning and may lead to disease. The protein kinase involved in the regulation of the dynamics of the actin cytoskeleton, Rho-kinase (ROCK), phosphory- lates various substrates (e.g. myosin light chain, myosin phosphatase), caus- ing the formation of actin fibers and tension inside cells. Hyperactivation of ROCK, for example, causes hypertension and cardiovascular disorders. Thus, the design of highly specific protein kinase inhibitors is of the utmost importance. To date, the majority of inhibitors investigated have been found to mimic and compete with ATP. However, in the present study we characterized the cellular effects of a novel bisubstrate inhibitor – adeno- sine–oligoarginine conjugate (ARC) – designed to interfere simultaneously with the ATP site and the substrate-binding pocket of basophilic kinases. ARC effectively pulled down ROCK from cell lysates, showed no cytotox- icity and suppressed the assembly of the actin cytoskeleton (especially cen- tral actin bundles) as the result of interference with the activity of the kinase. Combination of ARC with chloroquine yielded a stronger inhibi- tory effect and gave results similar to treatment with Y-27632. However, treatment with ARC produced more actin fragments and yielded a longer- lasting effect than treatment with Y-27632. Additionally, quantification of phosphorylated myosin light chain levels in ARC-treated or Y-27632-trea- ted cells implies that ARC is more effective than Y-27632 in suppressing the phosphorylation of at least one of the substrates of ROCK. We believe that the described bisubstrate strategy could be a useful lead for designing novel, highly specific inhibitors for different protein kinases.
Protein kinases act as molecular switches in most cellu- lar processes, and their concentration and activity are tightly controlled [1–3]. Deviation from their normal levels of activation may lead to dysfunction of cellular processes and the manifestation of different diseases [4,5]. The normal functioning of cells is based on the complex regulation of the cytoskeleton, comprising the microtubule network and the actin filaments. The pro-
teins responsible for the majority of processes involved in the regulation and the dynamics of the actin cyto- skeleton are the members of the Rho GTPase family: RhoA, Rac1 and Cdc42 [6–10]. These small GTPases switch between the inactive GDP-bound form and the active GTP-bound form [11]. Upon activation by external stimuli [e.g. lysophosphatidic acid (LPA)] they associate with their cellular substrates and activate
Abbreviations ARC, adenosine-oligoarginine conjugate; ARC-343, TAMRA-labeled ARC; CPP, cell-penetrating peptide; EGF, epidermal growth factor; GFP, green fluorescent protein; LDH, lactate dehydrogenase; LPA, lysophosphatidic acid; MLC, myosin light chain; PKA, protein kinase A; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide; p-MLC, phosphorylated myosin light chain; ROCK, Rho-associated serine ⁄ threonine kinase; TAMRA, tetramethylrhodamine.
FEBS Journal 275 (2008) 3608–3624 ª 2008 The Authors Journal compilation ª 2008 FEBS
3608
ARC effectively dissociates the actin cytoskeleton
H. Ra¨a¨ gel et al.
strate that extracellularly applied ARC efficiently enters the cells, where it interferes with the formation of the actin cytoskeleton. The inhibitory action of ARC is potentiated by chloroquine, a reagent known the acidification of endosomes and to to inhibit enhance the activity of molecules transduced into cells with CPPs [31,32]. Additionally, we demonstrate that ARC is not cytotoxic, even at high concentrations, and that it is able to associate with and pull down ROCK from cell lysates.
their functional pathways [12]. An important target for RhoA-GTPase is a Rho-associated serine ⁄ threonine kinase (ROCK), which phosphorylates myosin light chain (MLC) and myosin phosphatase. The subsequent activation of MLC and the inactivation of myosin phosphatase result in the formation and strengthening of the actin fibers and cytoskeleton [13]. Therefore, ROCK plays a crucial role in several cellular processes, which require the assembly, regulation and dynamics of the actin cytoskeleton (e.g. cellular adhesion and motility, processes during cytokinesis, etc.) [13–15].
Results
ARC hinders the formation of the actin cytoskeleton
Inhibitors of protein kinases have attracted much attention as potential drugs, but they are also valuable tools for investigating the functions of different kin- ases. A selective ROCK inhibitor, Y-27632, competes with ATP for the ATP-binding pocket of the enzyme molecule, thereby inhibiting the activity of the kinase to phosphorylate its substrates [16,17]. Y-27632 has been shown to abolish the formation of the actin cyto- skeleton and consequently the processes requiring the re-organization and existence of the cytoskeleton, for example, smooth muscle [18,19] and nonmuscle cell contraction, adhesion and motility [20,21]. However, a comparable potency of Y-27632 has been reported towards some other kinases, for example the protein kinase C-related protein kinase PRK2 [22], emphasiz- ing the need for novel specific inhibitors for ROCK and other protein kinases.
(ARC)
A new class of protein kinase bisubstrate-analog inhibitors was designed on the basis of adenosine-5¢- carboxylic acid derivates, where a short peptide was attached to the 5¢ carbon atom of the adenosine sugar moiety via a linker chain (Fig. 1A). This bisubstrate inhibitor was designed to compete simultaneously with the binding of ATP and the substrate protein, being capable of associating with the protein kinase at both locations [23,24]. The bisubstrate inhibitor with oligo- arginine as the peptide sequence was named an aden- osine–oligoarginine [23]. The conjugate oligoarginine sequence of ARC belongs to the class of cell-penetrating peptides (CPPs). The overall strong positive charge of the peptide, and especially the high number of arginine residues, provides the inhibitor with the ability to cross the plasma membrane barrier and effectively enter the cells [25,26]. Additionally, the planar structure of the adenosine, in analogy with the tryptophan residue in CPPs, might promote the entry into cells [27,28]. ARC translocates into several types of cells [24,29] and efficiently inhibits ROCK, and to a lesser extent also inhibits other basophilic protein kinases in vitro (Table 1) [30].
It has been shown previously that ARC efficiently inhibits ROCK in vitro (yielding 96% inhibition of ROCK-II at a 1 lm concentration of ARC-341 (new compound structure in Fig. 1A), as estimated in the protein kinase selectivity panel in the presence of 100 lm ATP; KinaseProfilerTM; Upstate Biotech Inc., Lake Placid, NY, USA; Table 1) [30]; however, the actions of ARC in living cells have not previously been documented. Thus, the inhibitory effect of ARC-341 on its target kinase – ROCK – was determined using HeLa and NIH 3T3 cells transiently expressing b-actin–eGFP. The loss of actin fibers as a result of the inhibition of ROCK activity was assessed in live cells by fluorescence microscopy. In these experiments, Y-27632, a commonly used selective inhibitor of ROCK, was utilized for comparison. The effect of ARC-341 on the actin fibers in HeLa cells was first incubation (Fig. 2), when evident after 15 min of slightly thinner fibers running across the cells were seen (in 100% of cells). However, no loss of fibers was detected. Compared with ARC-341, Y-27632 had a stronger inhibitory effect on the re-organization of actin fibers after only 15 min. The majority of fibers were lost upon treatment with Y-27632, and a diffuse b-actin–eGFP signal was evident in the cytoplasm, which tended to converge in the perinuclear region (characteristic for 80% of cells). Moreover, the accu- mulation of b-actin–eGFP at the plasma membrane, and especially in lamellipodia, was recorded. A much stronger effect of ARC-341 was visible after 1 h of incubation, and distinct thin, but frayed, actin fibers were detected across the cell body. From 1 h of incu- bation onwards the actin fibers became progressively more frayed, and by 7 h of incubation the majority of fibers had disappeared (in 80–90% of cells). A typical effect of ARC-341 was that the integrity of the actin fibers in the center of cells was affected, whereas the
This study characterizes the effects, in living cells, of a six-arginine-residue-containing ARC. We demon-
FEBS Journal 275 (2008) 3608–3624 ª 2008 The Authors Journal compilation ª 2008 FEBS
3609
ARC effectively dissociates the actin cytoskeleton
H. Ra¨ a¨gel et al.
C
A
D
(L-Arg)6-X-NH2
NH
Cytosolic fluorescence intensity
Vesicluar fluorescence intensity
O
*
***
*
***
**
N
***
O
N
NH2
O
U F R
U F R
N
N
OH
OH
3000 2500 2000 1500 1000 500 0
90 80 70 60 50 40 30 20 10 0
-
-
i
i
i
i
i
i
i
i
+
+
+
+
n m 0 3
i
i
3 4 3
n m 0 3
3 4 3 - C R A
l
+ 3 4 3
l
l
l
l
l
+ 3 4 3 - C R A
e n u q o r o h c
e n u q o r o h c
e n u q o r o h c
e n u q o r o h c
e n u q o r o h c
e n u q o r o h c
3 4 3 - C R A h 2
3 4 3 - C R A h 2
3 4 3 - C R A h 2
3 4 3 - C R A h 1
3 4 3 - C R A h 1
C R A n m 0 3
C R A n m 0 3
3 4 3 - C R A h 1
3 4 3 - C R A h 1
3 4 3 - C R A h 2
ARC-341, X = - ARC-342, X = Lys ARC-343, X = Lys(5-TAMRA) Affi-Gel-(ARC-342), X = Lys(Affi-Gel)
B
3 µM ARC-343
3 µM ARC-343 + 100 µM chloroquine
h 1
h 2
E
30 min
1 h
2 h
ARC-343 LAMP-2-AlexaFluor488
FEBS Journal 275 (2008) 3608–3624 ª 2008 The Authors Journal compilation ª 2008 FEBS
3610
ARC effectively dissociates the actin cytoskeleton
H. Ra¨a¨ gel et al.
Table 1. Residual activities of protein kinases in the presence of the inhibitor ARC-341 [assayed in the protein kinase selectivity panel (KinaseProfilerTM; Upstate Biotech Inc.) in the presence of 100 lM ATP].
Protein kinase
Residual activitya (%)
ROCK II PKA PKBa PKCbII CK2
4 ± 2 11 ± 2 30 ± 3 76 ± 4 86 ± 1
a Percentage of the residual activity of kinases (at 1 lM ARC-341) relative to that in control incubations where the inhibitor was omit- ted (results are expressed as the means of duplicate determina- tions) (from Enkvist et al. [30], modified).
NIH 3T3 cells showed similar results, giving rise to numerous filopodia during treatment with ARC, and producing lamellipodia and protrusions during treat- ment with Y-27632 (supplementary Fig. S1A). In accordance with the results obtained using HeLa cells, treatment with ARC-341 resulted in dissociation of mainly the central actin cables and no dissociation at all, even during longer incubation times, of the cortical actin fibers. By contrast with the results from HeLa cells, the loss of central actin bundles was observed after only 15 min (in 70% of cells) and the effect lasted for at least 7 h in NIH 3T3 cells (100% of cells), although frayed pieces of actin were still evident in the cytoplasm. Additionally, the induction of lamellipodia in the interfilopodial region (in 60% of cells) was observed for higher concentrations of ARC-341 (20 lm), but this was not as widespread as seen with Y-27632 (in 100% of cells) (supplementary Fig. S1B). Treatment with Y-27632 led to the formation of lamel- lipodia, which, over time, grew to longer protrusions tipped with spreading lamellipodia (in 90–100% of cells) (supplementary Fig. S1A,B), as seen in HeLa cells. Although some similarities were present as a result of treatment with both ARC-341 and Y-27632 (e.g. dissociation of actin fibers, formation of lamelli- podia and microspikes), ARC-341 tended to produce a villous phenotype, whereas Y-27632 gave rise to lamel- lipodia and protrusions.
The effects of ARC and Y-27632 were also con- firmed with phalloidin staining, where similar different phenotypes were observed for the inhibitors used (sup- plementary Fig. S1C). Treatment with ARC abolished the formation and existence of central actin cables (in 75% of cells), but the intense cortical fibers were not dissociated, even at a higher concentration of the inhibitor. Additionally, long actin-rich outgrowths, and also filopodial bridges between the cells, were detected in about 80–90% of the cells. Y-27632, on the
cortical actin fibers were not affected. The peripheral actin fibers were unaffected, even at an ARC-341 concentration of 20 lm (supplementary Fig. S1B). Upon 7 h of treatment with ARC-341 the cells lost their normal shape and shifted to a villous morphology (more than 20 filopodia per cell detected in 90% of the cells). Y-27632, on the other hand, had a marked effect on the integrity of the actin fibers starting from 15 min of incubation. The morphology of cells was similar at all time points – most of the fibers, including the cortical fibers, were missing, and an induction of lamellipodia was observed in approxi- mately 80% of cells. However, after 7 h of incuba- their normal tion with Y-27632 the cells had lost morphology, and numerous protrusions of the plasma membrane had formed (at least three protrusions per cell in 90% of the cells analysed). The protrusions were lined with some microspikes, which were scat- tered more widely than, and were not as long as seen with, ARC-341. Additionally, shrinkage of cells was observed after longer treatment with Y-27632 (in 40– 50% of cells).
Fig. 1. (A) Structure of ARC. (B) Uptake of tetramethylrhodamine (TAMRA)-labeled ARC into HeLa cells. HeLa cells were incubated with 3 lM ARC-TAMRA (ARC-343) in the absence (left panels) or presence of 100 lM chloroquine (right panels) at 37 (cid:2)C for 1 or 2 h. Vesicular and diffuse localization in cells was characterized in both cases at the 1 h time point, but vesicles appeared slightly larger and had a higher intensity of fluorescence in chloroquine-treated cells. However, at 2 h the ARC-343-treated cells resembled the chloroquine cotreated cells. Bar: 50 lm. (C, D) Quantification of ARC fluorescence intensity in cytosol and vesicles. The fluorescence signal of ARC-343, with or without chloroquine (images of Fig. 1B) was measured in the cytosolic fraction (C) and inside the vesicular structures (D) of live cells from 30 min to 2 h of incubation. The cytosolic signal of ARC-343 was evident after only 30 min of incubation, resulting possibly from nonvesicular uptake of the compound. At 1 h, the fluorescence of the cytosolic fraction of ARC had increased by 15–20% in chloroquine cotreated cells, and no detectable change was observed in cells incubated with ARC-343 only. At 2 h of incubation the ARC-343 signal in the cytosol of ARC-343- treated cells reached a level comparable to that of chloroquine cotreated cells. The same trend was followed measuring the intensity of the fluorescence signals inside the vesicles. The figures represent 50 analyzed cells (125 regions) of three separate experiments. * – padj value <0.05, *** – padj value <0.001. (E) Colocalization of ARC with the lysosomal marker, LAMP-2. HeLa cells were incubated with 3 lM ARC- 343 at 37 (cid:2)C for 30 min to 2 h, fixed and stained for lysosomal structures with LAMP-2 antibody. Maximum 20% colocalization between ARC and LAMP-2 was established at 2 h. Because overlapping of these structures was rare, it is possible that ARC does not follow the classical endolysosomal pathway inside the cells. Bar: 50 lm.
FEBS Journal 275 (2008) 3608–3624 ª 2008 The Authors Journal compilation ª 2008 FEBS
3611
ARC effectively dissociates the actin cytoskeleton
H. Ra¨ a¨gel et al.
IMDM
7 h sfIMDM
15 min
1 h
7 h
1 4 3 - C R A
15 min
1 h
7 h
2 3 6 7 2 - Y
Fig. 2. Dissociation of the actin cytoskeleton with ARC or Y-27632. HeLa cells transfected with b-actin–eGFP were incubated with 10 lM ARC-341 or Y-27632 in serum-free medium at 37 (cid:2)C, for 15 min, 1 h or 7 h. Dissociation of actin fibers was detected for both compounds, but Y-27632 was more potent at all time points. Treatment with ARC led to frayed actin fibers and a villous phenotype and treatment with Y-27632 induced lamellipodia and, upon longer incubation times, visible protrusions. Additionally, shrinkage was observed for Y-27632- treated cells. Untreated cells (IMDM) and serum-starved cells (sfIMDM) are included as controls. Bar: 50 lm.
other hand, dissociated the actin fibers completely (in 90% of cells), including the cortical actin structures, giving rise to protrusions lined with microspikes and
lamellipodia. As seen with b-actin–eGFP in live cells, phalloidin staining revealed the accumulation of actin to the plasma membrane of Y-27632-treated cells.
FEBS Journal 275 (2008) 3608–3624 ª 2008 The Authors Journal compilation ª 2008 FEBS
3612
ARC effectively dissociates the actin cytoskeleton
H. Ra¨a¨ gel et al.
with ARC-343 alone (Fig. 1C). However, 2 h after treatment, the cytosolic fluorescence intensity of ARC- 343-treated cells became drastically elevated, reaching the level of cells cotreated with ARC and chloroquine. Furthermore, the size and fluorescence intensity of the vesicles increased by about 1.5-fold, giving a markedly stronger signal than that of cells cotreated with ARC- 343 and chloroquine (Fig. 1D).
ARC-341 was not only able to induce the dissocia- tion of the existing actin cytoskeleton, but it was also capable of hindering the de novo formation of actin fibers upon induction with LPA in cells where the fibers had been previously lost as a result of 24 h of incubation in serum-free medium. Treatment with ARC-341 led to a marked decrease in the number and thickness of the actin fibers compared with the control cells. The effect was not as strong as detected with Y-27632, and a higher concentration of ARC-341 was needed to achieve the loss of most of the actin fibers. ARC-341 blocked the formation of the actin cytoskele- ton when used at a concentration of 20 lm, yielding results with 10 lm Y-27632 (data not analogous shown).
Internalization of ARC is enhanced by chloroquine
The ability of ARC to enter the cells has been demon- strated previously [24], where the internalization of ARC was shown to be mostly endocytic, along with a clearly detectable diffuse staining of the cytoplasm.
laser
The internalization efficiency of ARC was similar to that of other highly cationic CPPs (e.g. Arg9 or Tat), and seemingly the same endocytic pathways were used (data not shown). Even though a type of endocytic route was used for internalization, we were unable to detect a markable colocalization of ARC-343 with the lysosomal marker, LAMP-2, even after 2 h of incuba- tion (Fig. 1E). The ARC-343-containing vesicles were often found near the LAMP-2-positive structures. Nev- ertheless, overlapping of these structures was rarely found (maximum 20% colocalization). This means that the typical endocytic pathway, which in 2 h leads to degradative organelles (i.e. lysosomes), was not used in the uptake of ARC. To elaborate more on the earlier internalization steps of the inhibitor, we co-incubated live cells with ARC-343 and epidermal growth factor (EGF) and visualized them by confocal laser scanning microscopy (supplementary Fig. S2). At 30 min, both signals were mostly found at the plasma membrane, where about 50% of the colocalization between ARC and EGF was evident. Moreover, some of the cells containing strong ARC fluorescence on their mem- brane gave only a minor EGF-positive signal, and vice versa, suggesting that a similar uptake mechanism was utilized by both compounds. Over time (45 min to 1 h), a greater number of the marked vesicles became located more deeply within the cytoplasm, and a remarkable 70–80% colocalization between ARC and EGF was found inside the cells. However, at the 2 h time point, the fraction of overlapping vesicles had decreased dramatically and only 30% colocalization was detectable.
Chloroquine potentiates the effect of ARC
The results of the fluorescence microscopy experiments suggested that chloroquine increases the cytosolic con- centration of labeled ARC; therefore, we also analyzed the effect of chloroquine on the ability of ARC-341 to suppress the formation of actin fibers in b-actin–eGFP- expressing HeLa cells (Fig. 3). After 15 min of incuba- tion at a 10 lm concentration of the inhibitor together with 100 lm chloroquine, the cells had lost most of their actin fibers and the still-existing cytoskeleton was very frayed. Additionally, some of the cells had lost
To assess the effect of chloroquine on the internali- zation of ARC, HeLa cells were incubated with 3 lm fluorophore-labeled ARC in the presence of 100 lm chloroquine for 30 min to 2 h, and live cells were examined by confocal scanning microscopy (Fig. 1B). At the 30 min time point, most of the TAM- RA-labeled ARC (ARC-343) was still bound to the plasma membrane, but some of the signal was already detectable inside cells (data not shown). Effective inter- nalization of ARC-343 was evident after 1 h of incuba- tion, when small punctuate structures were observed in the cortical cytoplasm, although some of the signal could have also been found deeper in the central region of the cell. Over time, the number and the size of the vesicles grew, and more of the ARC-positive structures became located in the perinuclear region. Cotreatment with ARC-343 and chloroquine led to slightly larger vesicles; however, after 2 h of incubation the vesicles were of similar size and the cells were mostly indistinguishable from the cells treated with ARC-343 alone. During 1 h of incubation, only a weak diffuse fluorescence signal was measured in the cytosol (Fig. 1C). Cotreatment with ARC-343 and chloroquine resulted in slightly larger vesicles, which, judging by the intensity of the fluorescence, contained more labeled compound than the vesicles treated with ARC-343 alone (Fig. 1D). Additionally, at 1 h the the cytosolic fraction of fluorescence intensity of ARC-343 was about 1,2-fold higher in cells cotreated with ARC-343 and chloroquine than in cells treated
FEBS Journal 275 (2008) 3608–3624 ª 2008 The Authors Journal compilation ª 2008 FEBS
3613
ARC effectively dissociates the actin cytoskeleton
H. Ra¨ a¨gel et al.
15 min
30 min
60 min
e n i u q o r o l h c
1 4 3 - C R A
M µ
0 1
M µ 0 0 1 +
Fig. 3. Effect of chloroquine on the ability of ARC to inhibit the formation of actin fibers. HeLa cells transfected with actin-GFP were co-incu- bated with 10 lM ARC-341 and 100 lM chloroquine in serum-free medium for 15, 30 or 60 min at 37 (cid:2)C. Cells had already lost most of their actin fibers after 15 min of incubation, and a number of villi were present. After 30 min the cells had lost cortical fibers and started to resemble Y-27632-treated cells. No protrusions were detected. Bar: 50 lm.
bated with Triton X-100 to create holes in the plasma membrane. This allows the efflux of shorter actin frag- ments (e.g. monomeric and dimeric actin) but prevents longer actin filaments from escaping the cell interior. The amount of actin in the supernatants was quanti- fied by western blot analysis.
Interestingly, 20 lm ARC-341 produced more short actin oligomers than 20 lm Y-27632. The effect was evident after only 15 min of incubation (where the effi- ciency of ARC was found to be approximately three- fold greater) following through to 7 h of treatment (a difference in efficiency of about twofold; Fig. 4B). The actin-dissociating effect of both ARC-341 and Y-27632 increased over time, peaking at 30 min and persisting up to 7 h. Additionally, while the effect of Y-27632 decreased with length of incubation time (7 h), ARC- 341 effectively sustained the high level of actin-dissoci- ating activity, up to at least 7 h of incubation.
ARC is not toxic to cells
their cortical actin fibers. The outgrowth of numerous long filopodia was detected in cells with the strongest inhibitory effect. Although the majority of the cells were strongly affected by the synergistic effect of ARC- the cells were not, 341 and chloroquine, some of because the actin fibers were still visible, though thinner than in control cells. After 30 min of treatment, most cells resembled the cells incubated with Y-27632, with the actin cytoskeleton completely absent and, in addi- tion, possessed some of the typical lamellipodia. The b-actin–eGFP was located diffusely in the cells, and the long, numerous filopodia had been replaced with much shorter spikes. After 1 h of incubation, the cells resem- bled the cells treated with Y-27632, except that they had more spikes and fewer lamellipodia on their plasma membrane. Even though the cells co-incubated with ARC-341 and chloroquine looked mostly like the cells treated with Y-27632, no cells with numerous pro- trusions were seen at any incubation time point. When chloroquine was co-administered with Y-27632, no additional effect was observed by comparison with the cells treated with Y-27632 only.
ARC produces more monomeric actin than Y-27632
(MTT)
Even though visually ARC-341 had a milder and more spatially defined dissociating effect on the actin fibers, as seen in Fig. 2, we felt compelled to assess the quan- tity of dissociated actin produced by both inhibitors. For that we treated HeLa cells with 20 lm ARC-341 or Y-27632 for 15 min, 30 min, 1 h or 7 h, and incu-
Arginine oligomers at high concentrations are known to be cytotoxic to cells [33,34]. Because the ARC mole- cule contains six arginine residues, it was important to evaluate the putative toxicity of the compound on the cells used in this study. The 3-(4,5-dimethylthiazol-2- yl)-2,5-diphenyltetrazolium bromide assay revealed that ARC-341 did not suppress the metabolic activity of HeLa (Fig. 4A) or NIH3T3 (data not shown) cells after 1 h of incubation with up to 30 lm ARC-341. Comparable results with control cells were obtained for all concentrations of ARC-341 used.
FEBS Journal 275 (2008) 3608–3624 ª 2008 The Authors Journal compilation ª 2008 FEBS
3614
ARC effectively dissociates the actin cytoskeleton
H. Ra¨a¨ gel et al.
A
B
15 min 30 min 1 h
7 h 15 min 30 min 1 h
7 h
140
Actin
120
)
100
20 µM ARC-341
20 µM Y-27632
%
(
y t i v
80
Release of actin into extracellular environment
i t c a
c
-
i l
60
1400
o b a t e M
1200
40
1000
15 min
20
800
30 min
60 min
600
0
Control 1 µM
s l l e c d e v r a t s
420 min
400
Y27632
3 µM Y27632
10 µM Y27632
30 µM Y27632
1 µM ARC
3 µM ARC
10 µM ARC
30 µM ARC
200
m u r e s h 7 m o r f e s a e l e r f o
0
%
20 µM ARC
20 µM Y-27632
D
1
2
3
4
C ROCK
Qualification of p-MLC levels in cells after 2 h treatment
Qualification of kinases bound to the affinity resin
120
n i s e r y t i
100
n i f f a
o t
80
d e i l
1.04 1.02 1 0.98 0.96 0.94 0.92 0.9 0.88 0.86 0.84
ROCK
p p a
60
PKA
n o i t c a r f
40
10 µM Y-27632
20 µM Y-27632
10 µM ARC- 341
20 µM ARC- 341
sf IMDM IMDM + 30 min 5µM LPA
20
1
2
3
4
5
6
p-MLC
0
m o r f e s a n i k f o %
Wash 1
Wash 3
Bound to affinity resin
Unbound to affinity resin
Actin
Fig. 4. (A) Effect of ARC on the metabolic activity of HeLa cells. Cells were incubated with different concentrations of ARC-341 or Y-27632 at 37 (cid:2)C for 1 h. The MTT assay was used to determine the metabolic activity of cells. No significant deviation from the metabolic activity of control cells was observed for ARC-treated cells. (B) Quantification of the actin dissociation efficiency of ARC and Y-27632 by western blot- ting. Cells were incubated with 20 lM ARC-341 or 20 lM Y-27632 for 15 min, 30 min, 1 h, or 7 h in serum-free medium, treated with Triton X-100 to create pores in the plasma membrane to allow the efflux of shorter actin fragments. The amount of actin that escaped the cell inte- rior was analyzed by western blot. ARC effectively produced short actin oligomers, starting from 15 min, peaking at 30 min and persisting from 1 h until 7 h. Treatment with Y-27632 showed only traces of dissociated actin after 15 min of incubation. The effect of Y-27632 was maximal at the 30 min and 1 h time points, but the dissociation was not as strong as seen with ARC, and it started to weaken after a longer treatment period (7 h). (C) Affinity purification of ROCK from cell lysates using ARC–agarose. The affinity resin Affi-Gel-ARC-342 was used to pull down ROCK from cell lysates, and quantification of western blot analysis is shown in the graph. The majority of ROCK ((cid:2)80% of ROCK from the cell lysate) was bound by ARC–agarose (lane 1) and only a small fraction (approximately 20%) did not associate with the matrix (lane 2). The complex formed was stable, and extensive washing under stringent conditions did not dissociate ROCK from ARC–agarose. The first wash and the last wash are shown in lanes 3 and 4, respectively. The majority of PKA did not bind to ARC, and relatively high amounts of PKA were detected in the wash solutions. (D) Quantification of the phosphorylated (p)-MLC level in cells treated with ARC or Y-27632. The p-MLC level in HeLa cells incubated with 10 lM ARC-341 (lane 3), 10 lM Y-27632 (lane 4), 20 lM ARC-341 (lane 5), or 20 lM Y-27632 (lane 6) at 37 (cid:2)C for 2 h was analyzed by western blotting. Serum-starved (lane 1) and LPA-induced (30 min with 5 lM LPA) cells (lane 2) were used as controls and the level of p-MLC in serum-starved cells was taken as 1 (100%). Quantification of MLC phosphorylation states revealed a slight decrease in p-MLC levels for both inhibitors and that the efficiency of ARC was greater than that of Y-27632.
bolic activity was identical to that of the untreated cells (supplementary Fig. S3A).
Additionally, the long-term impact (7 h of incubation) of ARC-341 (up to 30 lm) was tested by using the MTT assay. No change was detected in the cells trea- ted with any concentration of ARC-341, and the meta-
Because some of the penetrating peptides destabilize the plasma membrane at high concentrations [34], it
FEBS Journal 275 (2008) 3608–3624 ª 2008 The Authors Journal compilation ª 2008 FEBS
3615
ARC effectively dissociates the actin cytoskeleton
H. Ra¨ a¨gel et al.
ARC is more effective than Y-27632 in decreasing the phosphorylation of MLC in cells
To evaluate the efficiency of both ARC and Y-27632 in inhibiting ROCK, we quantified the effects of the compounds used on the phosphorylation levels of a known ROCK substrate – MLC. As a control, we incubated the cells in a serum-free medium or induced the activity of the RhoA ⁄ ROCK pathway with 5 lm LPA for 30 min, and the level of p-MLC in serum-free cells was taken as 1 (100%).
was also necessary to assess the integrity of the plasma membrane of cells treated with ARC-341 using the lac- tate dehydrogenase (LDH) assay. When following the same experimental set-up as in the MTT assay, the results obtained showed that ARC-341 did not desta- bilize the plasma membrane at concentrations up to 30 lm because no efflux of LDH was induced (supple- mentary Fig. S3B). Surprisingly, treatment with ARC- 341 seemed to rather enhance the barrier function of the membrane because the leakage of LDH into the surrounding medium was slightly reduced in comp- arison to the control cells.
ARC precipitates ROCK from cell lysates
The ability of ARC to bind to and form a stable complex with ROCK was corroborated using agarose- immobilized ARC (the affinity resin Affi-Gel-ARC- 342), in an affinity-chromatography analysis. In order to characterize the stability of the ARC–ROCK com- plex, the affinity matrix was extensively washed under stringent conditions.
Analysis of the fractions obtained from the lysis of cells revealed that the majority of ROCK was located in the cytosol, and only a small amount of ROCK was associated with the cell membrane and the actin cyto- skeleton (data not shown). Therefore, the cytosolic fraction of lysed cells was used for affinity-chromatog- raphy assay.
As expected, treatment with LPA elevated the phos- phorylated state of MLC, giving a slightly higher p-MLC level in cells. However, incubation of cells with either inhibitor led to a significant decrease in MLC phosphorylation, and the observed decline was, in both cases, concentration dependent. A 2-h incubation of cells with Y-27632 reduced the phosphorylation of MLC, resulting in a decrease of p-MLC by 2% (at 10 lm Y-27632) and 5.5% (at 20 lm Y-27632) com- pared with cells incubated in serum-free medium. ARC-341, on the other hand, exhibited a stronger inhibitory effect on the level of p-MLC, and a decrease of 4 and 8.5% in phosphorylation of MLC was calcu- lated for 10 and 20 lm ARC, respectively. This clearly shows that even minor changes in the phosphorylation state of MLC can generate differences in the actin cytoskeleton, hinder its modeling and cause the above described phenotypic diversity.
Discussion
cellular
Protein kinases are the key regulators of most cellular processes. The overproduction or unregulated activity of several kinases contributes to the development of severe disorders and diseases [35–37]. Therefore, the specific inhibitors for these targets are of utmost importance. Most inhibitors, designed to date, mimic ATP and compete with it in occupying the ATP-bind- ing site of the enzyme [16,38,39]. The present study characterized the effects of an inhibitor designed to compete with both ATP and the protein substrate for the respective binding sites. This novel bisubstrate inhibitor has been found to be highly effec- tive in interfering with the activity of several basophilic protein kinases (e.g. ROCK, PKA, akt ⁄ PKBa and PKCg) in vitro [30]. Although the ability of the ARCs to enter different cell types has already been docu- mented [24,29], their potency to modulate the activity of targeted kinases in living cells has not previously been shown.
In this study we used ARC (also called ARC-341) that contains six arginine residues, which evokes its
Agarose-immobilized ARC effectively bound and concentrated ROCK, as confirmed by western blot analysis, where more than 75% of ROCK from the fraction applied was detected to be bound to the affinity resin. A minor fraction of ROCK (<25%) did not bind to Affi-Gel-ARC-342, and trace amounts of ROCK were detected in the solution that was removed from the matrix (Fig. 4C). The ARC– ROCK complex on agarose resin was highly stable because the concentrated samples of the wash-off solutions showed only traces of ROCK (Fig. 4C). The binding of another basophilic kinase, protein kinase A (PKA) [which also showed relatively high affinity towards ARC in an in vitro assay (Table 1)] to the affinity resin was also assessed. Only about 5% of PKA from the fraction applied to the affinity resin was able to bind to ARC, and most of the kinase was washed off from the resin (about 40% during the first wash and about 25% during the third wash), determining the fragility of the bond between the inhibitor and PKA. The fact that ARC binds sta- leads us to believe that bly ROCK, but not PKA, ROCK is its primary target inside the cells [as well as in vitro (Table 1)].
FEBS Journal 275 (2008) 3608–3624 ª 2008 The Authors Journal compilation ª 2008 FEBS
3616
ARC effectively dissociates the actin cytoskeleton
H. Ra¨a¨ gel et al.
quine changes the morphology of ARC-containing ves- icles, probably by making the membrane of the vesicles more permeable, thus (a) inducing an increased escape of ARC from all the vesicles or (b) causing some of the vesicles to break and release their contents into the cytosol. However, during 2 h of incubation, most of the vesicles observed in ARC-treated cells were of similar size and contained an even stronger fluores- cence signal than those in chloroquine cotreated cells. This leads us to believe that ARC, alone, is also capa- ble of inducing swelling of the vesicles, rupturing the membrane and leaking out from the entrapping struc- tures to the cytosol, and chloroquine treatment only accelerates this process. The fact that in 2 h only a minor fraction of ARC was found in LAMP-2-positive structures and that only about 30% of ARC colocal- ized with EGF at that time point demonstrates that ARC does not follow the classical endolysosomal path- way inside the cell, giving the compound the advantage of leaking into the cytosol, where it is able to bind its target(s) and perform its inhibitory effects.
uptake by cells. The internalization of the TAMRA- labeled ARC was mainly endocytic, as has been dem- onstrated for other arginine-rich compounds and highly charged cationic CPPs [40–43]. Co-incubation of ARC with EGF (a marker of typical endocytotic pathway leading to lysosomes in 2 h [44]) illustrated the endocytic internalization of ARC because a strong colocalization between the two signals was first evident on the plasma membrane following the uptake and formation of endocytic vesicles. The apparent colocal- ization reached a maximum at 1 h, when up to 80% of the EGF and ARC signals were detected inside the same vesicles. However, thereafter the level of overlap- ping decreased to about 30%, indicating that different types of mechanisms are used to sort them from multi- vesicular bodies, and that they have different intra- cellular destinations within the cell. Nevertheless, we cannot rule out the fact that the prevalence of endo- cytosis could also be increased by the presence of the TAMRA label because it has been shown that some CPPs (e.g. Tat) tend to remain trapped in endosomes to a higher extent when labeled with TAMRA than with fluorescein isothiocyanate (FITC) [45].
from these
vesicular
signal
Faint, diffuse, overall staining of the cells was also always detected, suggesting the presence of ARC in the cytoplasm, which was first seen after only 30 min, but was more evident at 2 h. A fraction of ARC probably escaped from the endocytic vesicles into the cytosol, a situation most likely to occur during longer incubation times (between 1 and 2 h). Nonendocytic uptake could not be excluded, which would explain the presence of diffuse staining of the cytoplasm observed at early time points (up to 1 h). This may have been (a) facilitated by the planar structure of the adenosine, which might induce translocation without the formation of endo- somes [27,28] or (b) induced by oligo-arginine itself [46–48]. During the first hour, co-incubation of cells with ARC and chloroquine enlarged the ARC-contain- ing vesicles and increased the intensity of the fluores- cence structures. Chloroquine interferes with the acidification of endoso- mal structures (data not shown), leading to the swell- ing of vesicles or possibly inducing the fusion of early endosomes. As the concentration of ARC seemed to be higher in these vesicles (at the 1 h time point), the stability of the membrane might be compromised, resulting in greater leakage, as described in previous studies [49]. Indeed, a slightly stronger diffuse staining of the cytoplasm was recorded in cells treated with TAMRA-labeled ARC along with chloroquine at 1 h, indicating that the trapped ARC was able to escape from the vesicles more efficiently in the presence of the lysosomotrophic agent. This clearly shows that chloro-
A number of previous studies have demonstrated the effect of Y-27632 on the actin cytoskeleton [16,50– 52]. Our results show that Y-27632, a potent ROCK inhibitor, dissociates the actin fibers markedly after only 15 min of incubation. The inhibition of ROCK enzymatic activity with Y-27632 resulted in the loss of the actin cytoskeleton and the concentration of b-actin–eGFP in numerous lamellipodia, indicating the prevalence of Rac1 activity. The antagonistic regula- tions of RhoA and Rac1 have been previously described also by other groups [53–55], thus explaining the observations that the inhibition of ROCK by Y-27632 leads to the domination of the Rac1 pathway and the consequent induction of membrane ruffling. The long-term treatment of cells with Y-27632 gave rise to protrusions, a phenotype prone to become motile [21,56] (H. Ra¨ a¨ gel, unpublished results). Addi- tionally, shrinkage of cells was observed for Y-27632- treated cells, which is probably caused by the focal adhesions (H. Ra¨ a¨ gel, weakening and loss of unpublished results) along with dissociation of the the RhoA ⁄ ROCK actin cytoskeleton. Inhibition of pathway interferes with the maturation of focal con- tacts to focal adhesions, thus hindering cell spreading and the achievement of normal cell size and morphol- ogy. ARC, on the other hand, started to show visible inhibitory effects from the 60-min time point onwards, when thinner and frayed actin fibers were seen across the cell body. After 7 h, most cells had completely lost their central actin fibers, yet distinct, strong cortical fibers were still evident. Another notable difference was that instead of inducing lamellipodia and protru-
FEBS Journal 275 (2008) 3608–3624 ª 2008 The Authors Journal compilation ª 2008 FEBS
3617
ARC effectively dissociates the actin cytoskeleton
H. Ra¨ a¨gel et al.
rule out the possibility that Y-27632 dissociates the actin cytoskeleton into longer fragments, which are not able to pass through the pores of the plasma membrane generated by Triton X-100. Thus, we can speculate that ARC might be more effective in producing shorter fragments of actin. Furthermore, the treatment of cells with ‘classical’ CPPs (Tat or Arg9) had no effect on the integrity of the actin structures, indicating that dissoci- ation of the actin cytoskeleton is specific for ARC and not for its peptidic moiety (supplementary Fig. S4).
The lower inhibitory effect of the compound in living cells, compared with a cell-free system, could be explained by the fact that the majority of ARC enters cells mainly via endocytosis. First, the escape of ARC from the vesicles might be the limiting step and, second, the endocytic cellular uptake targets the inhibitor to proteolytic degradation by proteases capable of digest- ing the peptide moiety of the molecule. Pegylation of ARC increases its half-life in vivo and could improve its proteolytic stability [29]. On the other hand, the added group makes ARC bulkier and interferes with cellular uptake and perhaps also with endosomal escape.
existed in distinct
strong cables
In sum, we describe here the actions of ARC, a novel bisubstrate analog-type inhibitor of ROCK and other basophilic protein kinases, in living cells. ARC enters cells effectively and is able to escape from the vesicles into the cytosol, which leads to the loss of central actin fibers in b-actin–eGFP-transfected HeLa cells. Potentia- tion of the inhibitory effect of ARC by co-administra- tion with chloroquine at early time points (up to 1 h) indicated that the endosomal release could be the major factor limiting the activity of the compound in living cells, although a significantly higher extent of actin dis- sociation was quantified with ARC, rather than with Y-27632, and a slightly lower level of p-MLC was detected in ARC-treated cells compared with Y-27632- treated cells. Additionally, we show here that ARC does not exert toxicity on cells in culture, even at high con- centrations, and is able to pull down ROCK from cell lysates efficiently. To our knowledge, the results of the experiments described in the present communication constitute the first demonstration of the applicability of the novel bisubstrate-analog approach in a cellular envi- ronment for the development of inhibitors of protein kinases for the regulation of cellular processes.
Materials and methods
Cell lines
sions, ARC-treated cells obtained a villous phenotype. The long and numerous villi point to the canalization of the signal transduction to Cdc42 pathway known to promote the formation of filopodia [57–59]. Chloro- quine amplified the inhibitory effect of ARC, and visi- ble changes in the actin cytoskeleton were recognizable after only 15 min of treatment, yet some cells remained unaffected. The results obtained in the internalization experiment corroborate that the cytoplasmic concen- tration of ARC is higher when it is applied to cells along with chloroquine at an earlier time point (up to 1 h), and thus ARC might affect its targets inside the cytosolic compartment faster and more efficiently. The combined effect of ARC and chloroquine increased over time until the cells started to resemble Y-27632- treated cells. This implies a slight drift towards Rac1, although Cdc42 activity was not completely lost, as shorter microspikes were still visible and more evident than in Y-27632-treated cells. Also, the shift towards the Rac1 pathway was not as strong as seen with Y-27632-treated cells, inasmuch as no protrusions and excess motility (data not shown) were recorded, even after longer incubations. Although visually, treatment with Y-27632 led to a more complete dissociation of the actin cytoskeleton, quantification of the phosphor- ylation state of MLC (the protein causing the tension and existence of the actin fibers) revealed that ARC is more potent in inhibiting the RhoA ⁄ ROCK pathway at the molecular level. This could be the result of a higher affinity of ARC towards ROCK, whereas Y-27632 interacts with other cellular kinases (e.g. PKN ⁄ PRK-1 [22,60]), being responsible for the loss of the cortical actin cytoskeleton [61,62] and the develop- it ment of abnormal cell morphology. Additionally, has previously been demonstrated that ROCK-medi- ated actin remodeling is spatially restricted inside the cells, being active more in the central regions [63]; the effect was seen in most of the ARC-treated cells, where the central actin cytoskeleton was affected but cortical actin still left unchanged. Preservation of the organization of the cortical actin, and therefore also the normal cellular shape and size, could be an advantage for ARC in an in vivo environment. Moreover, even though Y-27632 showed a greater extent of actin dissociation through- out the GFP-actin-expressing cells, quantification of dissociated actin revealed that ARC produces notably higher levels of short actin fragments and is able to sustain the effect also after longer incubation times. This could result from the extensive incorporation of monomeric actin into lamellipodia in the case of treatment with Y-27632, which was detected by fluo- rescence microscopy. On the other hand, we cannot
FEBS Journal 275 (2008) 3608–3624 ª 2008 The Authors Journal compilation ª 2008 FEBS
3618
The human cervical cancer cell line, HeLa, and the mouse fibroblast cell line, NIH 3T3, were cultured in a humidified atmosphere, containing 5% CO2, at 37 (cid:2)C in IMDM
ARC effectively dissociates the actin cytoskeleton
H. Ra¨a¨ gel et al.
Reagents
(Gibco, Invitrogen, Paisley, UK). Culture medium was supplemented with 10% fetal bovine serum (PAA Labora- tories, Pasching, Austria), 100 IUÆmL)1 of penicillin and 100 lgÆmL)1 of streptomycin (Gibco, Invitrogen).
Fluorescence microscopy
1.5 mL eppendorf tube (Axygen, Union City, CA, USA). ARC-342 (8.5 mg) and triethylamine (10 lL) were dissolved in dimethylsulfoxide (150 lL), and the solution was added to the gel suspension. After a 1-h reaction, 30 lL of etha- nolamine was added to the reaction mixture to block excess reactive groups of gel, and the reaction was continued for 2 h. The affinity gel was washed three times with iso- propanol, water and acetate buffer (0.1 m, pH 4.5).
Synthesis of ARCs
Effect of chloroquine on the cellular uptake of ARC HeLa cells were seeded onto eight-well chambered cover- (Lab-Tek, #155411; Nalge Nunc International, glasses Rochester, NY, USA) and grown to 60–70% confluence. Cells were incubated with 3 lm TAMRA-labeled ARC (ARC-343), with or without 100 lm chloroquine (Sigma), in serum-free IMDM at 37 (cid:2)C. Images of live cells were recorded, after incubation for 30 min, 1 h and 2 h, using an Olympus Fluoview FV1000 confocal laser-scanning micro- scope, the fluorescence of cytoplasm and vesicles were mea- sured using Olympus FV1000 software FV10-ASW and the representative results of 25 cells from three different expe- riments were analyzed using Microsoft excel and the statistical program r 2.6.0.
Colocalization experiments with EGF
Y-27632 was purchased from Tocris (Ellisville, MO, USA). Protease inhibitor cocktail used in affinity chromatography was obtained from Roche (Basel, Switzerland). Phen- ylmethanesulfonyl fluoride was obtained from Sigma (St Louis, MO, USA). Peptide synthesis chemicals were purchased from Neosystem (Strasbourg, France), Novabio- chem (La Jolla, CA, USA), Advanced ChemTech (Louisville, KY, USA) and AnaSpec (San Jose, CA, USA). All other reagents were from AppliChem (Darmstadt, Germany) (if not stated otherwise).
freeze-dried. Authenticity and
HeLa cells were seeded onto eight-well chambered cover- glasses, as described above. At 60–70% confluence, the cells co-incubated with ARC-343 and EGF-Oregon- were Green514 (Molecular Probes, Eugene, OR, USA) at 37 (cid:2)C and the images of live cells were recorded as described above.
Colocalization experiments with LAMP-2
Synthesis of the fluorescent probe AdcAhx(Arg)6Lys (5-TAMRA)NH2, TAMRA-labeled ARC (ARC-343)
C-terminally amidated adenosine–peptide conjugates [Ad- (ARC-341) and AdcAhx(Arg)6LysNH2 cAhx(Arg)6NH2 (ARC-342)] were prepared by conventional Fmoc solid- phase peptide synthesis procedures on Rink amide MBHA resin, as described previously [30]. Compounds were purified with the Gilson HPLC system using the C18 reversed-phase column Inertsil ODS-3 (5 lm, 25 · 0.46 cm; GL Sciences, Tokyo, Japan), eluted with water-acetonitrile gradient (5–40% MeCN in 30 min, 0.1% trifluoroacetic acid, flow 1 mLÆmin)1) of rate verified using MALDI TOF mass compounds was spectrometry, yielding the following: AdcAhx(Arg)6NH2, calculated Mr = 1330, experimental Mr(M+H) = 1331; AdcAhx(Arg)6LysNH2, calculated Mr = 1459, experimental Mr(M+H) = 1460.
Synthesis of the affinity resin Affi-Gel-ARC-342
A solution of 1 lmol of AdcAhx(Arg)6LysNH2 (ARC-342) in 100 lL of dimethylsulfoxide (containing 10 lL of tri- ethylamine) was added to the solution of 5-carboxytetra- methylrhodamine, succinimidyl ester (0.53 mg, 1.0 lmol; Anaspec, CA, USA) in dimethyl formamide (50 lL). After a 3-h reaction time at room temperature, the product was puri- fied by HPLC and characterized as described above. MALDI TOF mass spectrometry analysis yielded the following results: calculated Mr = 1871, experimental Mr(M+H) = 1872, and UV–VIS spectroscopy: kmax = 558 nm.
FEBS Journal 275 (2008) 3608–3624 ª 2008 The Authors Journal compilation ª 2008 FEBS
3619
HeLa cells were seeded onto coverslips (Ø 12 mm; Menzel- Glazer, Braunschweig, Germany) in 24-well plates (Greiner bio-one, Frickenhausen, Germany). At 80% confluence, the cells were incubated with ARC-343 for 30 min, 1 h or 2 h, washed with NaCl ⁄ Pi, fixed with 4% paraformaldehyde in NaCl ⁄ Pi on ice for 30 min, permeabilized with 0.1% Triton X-100 on ice for 5 min and blocked with 10% nonfat dry milk in NaCl ⁄ Pi for 30 min. Fixed cells were incubated with a 1 : 100 dilution of LAMP-2 antibody (DSHB) in 1% nonfat dry milk in NaCl ⁄ Pi for 1 h, washed five times, for 5 min each wash with NaCl ⁄ Pi, incubated with a 1 : 400 dilution of anti-mouse-AlexaFluor488 in 1% nonfat dry milk in NaCl ⁄ Pi for 30 min and washed again (five times, for 5 min each wash). Cells were mounted with 30% glyc- erol in NaCl ⁄ Pi to microscope slides (Menzel-Glazer); the images were recorded with an Olympus FV1000 confocal laser scanning microscope and then processed using Adobe photoshop 6.0. A 0.5 mL volume of settled Affi-Gel 10 (BioRad, Hercules, CA, USA) was washed three times with isopropanol in a
ARC effectively dissociates the actin cytoskeleton
H. Ra¨ a¨gel et al.
LDH release assay
Dissociation of actin fibers in actin-eGFP transfected cells
Affinity chromatography of ROCK
HeLa cells were seeded onto 24-well tissue culture plates, as described above. Cells were incubated with 1, 3, 10 or 30 lm ARC-341 or Y-27632 in serum-free DMEM at 37 (cid:2)C for 1 h, medium was collected and the amount of LDH that leaked out from the cells was analyzed using the CytoTox-ONE Homogenius Membrane Integrity Assay (Promega, South- ampton, UK), following the manufacturer’s protocol. As a positive control (100% LDH in medium), cells were lysed with 0.1% Triton X-100 on ice for 5 min. The fluorescence signal was measured with Genios Plus (Tecan, Gro¨ dig, Austria) at 540 nm and the results were processed using Microsoft excel.
Dissociation of actin fibers in cells (phalloidin staining)
HeLa or NIH 3T3 cells were seeded onto coverslips (Ø 12 mm; Menzel-Glazer) in 24-well plates (Greiner bio- one). When 60–70% confluent, the cells were transfected with pAcGFP1-actin (Clontech Laboratories, Mountain View, CA, USA) using ExGen 500 (Fermentas, Vilnius, Lithuania) according to the protocol provided by the sup- plier. Transfected cells were incubated with 10 or 20 lm ARC or Y-27632, with or without 100 lm chloroquine, in serum-free medium at 37 (cid:2)C for 15 min to 7 h. Cells were washed three times with NaCl ⁄ Pi, for 3 min each wash, covered with Fluoromount G (Electron Microscopy Sci- ences, Hatfield, PA, USA), mounted to microscope slides (Menzel-Glazer) and the images were recorded using an Olympus BX61 microscope equipped with a CCD camera DP70. The images thus obtained were processed using Adobe photoshop 6.0.
MTT assay
HeLa cells were seeded onto coverslips, as described above. At 80% confluence, the cells were treated with 20 lm ARC-341 or Y-27632 in serum-free medium for 1 h, starved in serum-free medium for 18 h (negative con- trol), or incubated with 5 lm LPA for 30 min (positive control) at 37 (cid:2)C. The cells were then washed, fixed and blocked as described above. Actin cytoskeleton-containing structures were visualized with a 1 : 150 dilution of phal- loidin-AlexaFluor594 in NaCl ⁄ Pi containing 1% nonfat dry milk for 30 min. The cells were washed extensively and mounted with 30% glycerol in NaCl ⁄ Pi to microscope slides. Images were recorded with the Olympus Fluoview FV1000 confocal microscope and processed using Adobe photoshop 6.0.
FEBS Journal 275 (2008) 3608–3624 ª 2008 The Authors Journal compilation ª 2008 FEBS
3620
Cells were grown on tissue culture plates (Ø 10 cm; Greiner bio-one) to 80–90% confluence, detached by scraping and then collected by centrifugation (200 g, 20 (cid:2)C, 5 min). The cells were frozen and lysed with ice-cold lysis buffer con- taining 50 mm Tris–HCl, pH 7.2, 100 mm NaCl, 5 mm MgCl2, 0.5 mm EDTA, 1 mm dithiothreitol, 2.5· protease inhibitor cocktail, 2 mm phenylmethanesulfonyl fluoride and 1% Triton X-100 on ice for 10 min. Nuclei and cell debris were pelleted by centrifugation (3000 g, 4 (cid:2)C, 5 min). The supernatant was collected and recentrifuged (16 000 g, 4 (cid:2)C, 10 min). ATP was removed from the resulting super- natant by gel filtration in a Sephadex G-25 10 ⁄ 10 column (Pharmacia Fine Chemicals AB, Uppsala, Sweden). Prote- (1·), 1 mm phenylmethanesulfonyl ase inhibitor cocktail fluoride and 1 mm dithiothreitol were added to the frac- tions and analyzed bywestern blot. The fraction containing the majority of ROCK (1.5 mL) was applied to the ARC– agarose matrix (Affi-Gel-ARC-342) (150 lL) and incubated for 1 h at 4 (cid:2)C by continuous end-over-end mixing. The separated by centrifugation (900 g, 4 (cid:2)C, matrix was 5 min), washed three times with 200 lL of Pull-Down buf- fer containing 25 mm Tris–HCl, pH 7.2, 100 mm KCl, 10 mm MgCl2, 0.5 mm EDTA, 1 mm dithiothreitol, 0.1% Tween-20, 1· protease inhibitor cocktail and 1 mm phen- ylmethanesulfonyl fluoride, and separated by centrifugation (900 g, 4 (cid:2)C, 5 min). The supernatants were collected for further analysis, and matrix-bound proteins were eluted with 0.5 gel-volumes of colorless 4· SDS sample buffer. Samples were analyzed by western blotting using rabbit polyclonal antibodies against ROCK2 [Rock-2 (H-85), Santa Cruz Biotechnology, Santa Cruz, CA, USA] or the PKA catalytic domain [PKAa cat (C-20) (Santa Cruz Bio- technology)]. For quantification, the amount of ROCK or PKA in the fraction applied to affinity resin was taken as 100%, and the percentage of kinase bound to the matrix was determined by calculating the difference between the amount of applied and of unbound kinase measured from western blots using image j software. A total of 5 · 104 cells per well were seeded onto 24-well tissue culture plates (Greiner bio-one) and grown in serum- containing medium for 1 day. The cells were treated with 1–30 lm ARC or Y-27632 at 37 (cid:2)C for 1 or 7 h, washed and the medium changed. To measure the metabolic activ- ity of cells, MTT (final concentration 0.5 mgÆmL)1 in H2O; Sigma) was added and incubated for 4 h at 37 (cid:2)C. Precipi- tated MTT formazan was dissolved overnight by the addi- tion of an equal volume of 20% SDS in 0.02 m HCl. The absorbance at 540 nm was measured by spectrophotometry (Genios, Tecan, Austria). To determine the long-lasting effect of the reagents used, cells were treated as described above. After 1 h of incubation, the medium was changed and the cells were grown in medium containing 10% serum for an additional 24 h, after which the MTT assay was performed.
ARC effectively dissociates the actin cytoskeleton
H. Ra¨a¨ gel et al.
Western blot
Quantification of the p-MLC level in cells
Western blot analysis of the association of ROCK with the ARC–agarose matrix Proteins were separated by electrophoresis through an 8% SDS polyacrylamide gel and transferred to poly(vinylidene difluoride) (PVDF) membrane. Sites of nonspecific binding on the membrane were saturated with 6% BSA in NaCl ⁄ Tris for 3 h, and the membrane was rinsed with washing solution (0.05% Tween-20 and 1 mgÆmL)1 of PEG 20000 in NaCl ⁄ Tris). The membrane was incubated with 1– 2 lgÆmL)1 of rabbit polyclonal antibody ROCK2 [Rock-2 (H-85), Santa Cruz Biotechnology)] in NaCl ⁄ Tris at 4 (cid:2)C overnight, washed three times (10 min each wash) and incu- bated with alkaline phosphatase anti-rabbit IgG (1 : 2000 dilution in 1% BSA in NaCl ⁄ Tris) (Vector Laboratories, Peterborough, UK) at room temperature for 1 h. The mem- brane was washed and rinsed with alkaline phoshatase buf- (100 mm NaCl, 5 mm MgCl2, 100 mm Tris–HCl, fer pH 9.5). The signal was developed with Nitro Blue tetrazo- lium and 5-bromo-4-chloroindol-2-yl phosphate as sub- strates in the same buffer. Quantification of western blot images was carried out using image j and the results were analyzed with Microsoft excel, where the amount of ROCK or PKA in the fraction applied to the affinity col- umn was taken as 100% and the amount of kinase bound to affinity resin was calculated as the difference between the amount of applied and the amount of unbound kinase.
Dissociation of actin fibers into monomeric b-actin by ARC and Y-27632
A total of 1 500 000 cells were seeded per culture dish (Ø 6 cm; Grenier, Bio-One). The following day, cells were incubated with 10 or 20 lm ARC-341 or Y-27632 in serum- free medium at 37 (cid:2)C for 2 h, washed with ice-cold NaCl ⁄ Tris and collected by scraping. A 1 mL volume of 2· SDS sample buffer [containing 1· protease inhibitor cocktail, 1 mm phenylmethanesulfonyl fluoride, 2 mm sodium ortho- vanadate (S6508; Sigma) and 20 mm sodium pyrophosphate (S6422; Sigma)] was added to 50 mg of cells and homoge- nized with an ultrasound device three times at maximum intensity (Ultrasonic Homogenizer 4710 Series; Cole- IL, USA). Proteins were separated on 15% Palmer, acrylamide gel and transferred to PVDF membrane, as described above. Membranes were then blocked for 3 h with 6% BSA in NaCl ⁄ Tris, washed and incubated with p-MLC (Ser-19) antibody (1 : 250 dilution in 1% BSA in NaCl ⁄ Tris; Santa Cruz Biotechnology) at 4 (cid:2)C overnight with gentle rocking. Membranes were washed three times, for 10 min each wash, with 0.01% Tween-20 in NaCl ⁄ Tris and incubated with horseradish peroxidase-conjugated anti- rabbit polyclonal antibody (Vector Laboratories) (1 : 2500 dilution in 1% BSA in NaCl ⁄ Tris) for 1 h at room temper- ature with gentle rocking and washed again three times, for 10 min each wash. Horseradish peroxidase-conjugated substrate was activated with the ECL Plus Western Blot Detection System (RPN3132; Amersham, Little Chalfont, UK) following the manufacturer’s protocol and the signal was recorded onto medical X-ray film (Super RX; Fujifilm Corporation, Tokyo, Japan). Quantification of the western blot was carried out using image j and the results were analyzed using Microsoft excel.
Acknowledgements
The study was supported by grants from the European Commission (QLK3-CT-2002-01989), the Estonian Science Foundation (ESF 5588, 6710 and 7058) and the Estonian Ministry of Education and Research (0182611s03 and 0182691505). H. R. was supported by the Artur Lind Scholarship Fund.
References
1 Smits VA & Medema RH (2001) Checking out the
G(2) ⁄ M transition. Biochim Biophys Acta 1519, 1–12. 2 Galderisi U, Jori FP & Giordano A (2003) Cell cycle regulation and neural differentiation. Oncogene 22, 5208–5219.
FEBS Journal 275 (2008) 3608–3624 ª 2008 The Authors Journal compilation ª 2008 FEBS
3621
3 Gerthoffer WT (2005) Actin cytoskeletal dynamics in smooth muscle contraction. Can J Physiol Pharmacol 83, 851–856. A total of 100 000 cells per well were seeded onto 12-well plates (Greiner bio-one). The following day, cells were incu- bated with 20 lm ARC-341 or Y-27632 in serum-free med- ium for 15 min, 30 min, 1 h, or 7 h. The cells were washed with NaCl ⁄ Pi, treated with 0.1% Triton X-100 on ice for 5 min and then the supernatants were collected and concen- trated [64]. The amount of b-actin released from the cells was quantified by western blotting, where the membrane was first blocked with 5% nonfat dry milk in NaCl ⁄ Pi at 4 (cid:2)C overnight and then incubated with 0.05 lgÆmL)1 of murine polyclonal antibodies against b-actin {beta actin antibody [ACTN05 (C4)], Abcam, UK} in NaCl ⁄ Pi con- taining 5% nonfat dry milk, at room temperature for 1 h. The membrane was washed three times, for 7 min each wash, with 0.05% Tween-20 in NaCl ⁄ Pi, incubated with alkaline phosphatase anti-mouse IgG (Vector Laboratories) (1 : 2500 dilution in 5% nonfat dry milk in NaCl ⁄ Pi) at room temperature for 1 h, washed again with 0.05% Tween 20 in NaCl ⁄ Pi, and developed with Nitro Blue tetrazolium and 5-bromo-4-chloroindol-2-yl phosphate. Western blot quantification was carried out using image j and Microsoft excel.
ARC effectively dissociates the actin cytoskeleton
H. Ra¨ a¨gel et al.
4 Testa JR & Tsichlis PN (2005) AKT signaling in 20 Shiokawa S, Iwashita M, Akimoto Y, Nagamatsu S,
normal and malignant cells. Oncogene 24, 7391–7393. 5 Budzyn K, Marley PD & Sobey CG (2006) Targeting
Rho and Rho-kinase in the treatment of cardiovascular disease. Trends Pharmacol Sci 27, 97–104. 6 Clark EA, King WG, Brugge JS, Symons M & Hynes Sakai K, Hanashi H, Kabir-Salmani M, Nakamura Y, Uehata M & Yoshimura Y (2002) Small guanosine tri- phospatase RhoA and Rho-associated kinase as regula- tors of trophoblast migration. J Clin Endocrinol Metab 87, 5808–5816.
RO (1998) Integrin-mediated signals regulated by mem- bers of the rho family of GTPases. J Cell Biol 142, 573– 586. 7 Evers EE, Zondag GC, Malliri A, Price LS, ten Kloo- 21 Koga T, Awai M, Tsutsui J, Yue BY & Tanihara H (2006) Rho-associated protein kinase inhibitor, Y-27632, induces alterations in adhesion, contraction and motility in cultured human trabecular meshwork cells. Exp Eye Res 82, 362–370.
ster JP, van der Kammen RA & Collard JG (2000) Rho family proteins in cell adhesion and cell migration. Eur J Cancer 36, 1269–1274.
22 Davies SP, Reddy H, Caivano M & Cohen P (2000) Specificity and mechanism of action of some com- monly used protein kinase inhibitors. Biochem J 351, 95–105. 8 Nikolic M (2002) The role of Rho GTPases and associ- ated kinases in regulating neurite outgrowth. Int J Bio- chem Cell Biol 34, 731–745.
23 Loog M, Uri A, Raidaru G, Ja¨ rv J & Ek P (1999) Adenosine-5¢-carboxylic acid peptidyl derivatives as inhibitors of protein kinases. Bioorg Med Chem Lett 9, 1447–1452. 24 Uri A, Raidaru G, Subbi J, Padari K & Pooga M 9 Moreau V, Tatin F, Varon C & Genot E (2003) Actin can reorganize into podosomes in aortic endothelial cells, a process controlled by Cdc42 and RhoA. Mol Cell Biol 23, 6809–6822.
10 Aspenstrom P, Fransson A & Saras J (2004) Rho GTP- ases have diverse effects on the organization of the actin filament system. Biochemistry 377, 327–337. 11 Boguski MS & McCormick F (1993) Proteins regulating (2002) Identification of the ability of highly charged nanomolar inhibitors of protein kinases to cross plasma membranes and carry a protein into cells. Bioorg Med Chem Lett 12, 2117–2120. 25 Mitchell DJ, Kim DT, Steinman L, Fathman CG & Ras and its relatives. Nature 366, 643–654. 12 Kjoller L & Hall A (1999) Signaling to Rho GTPases. Exp Cell Res 253, 166–179. Rothbard JB (2000) Polyarginine enters cells more effi- ciently than other polycationic homopolymers. J Pept Res 56, 318–325.
13 Amano M, Fukata Y & Kaibuchi K (2000) Regulation and functions of Rho-associated kinase. Exp Cell Res 261, 44–51. 14 Riento K & Ridley AJ (2003) Rocks: multifunctional 26 Futaki S (2005) Membrane-permeable arginine-rich pep- tides and the translocation mechanisms. Adv Drug Deliv Rev 57, 547–558. 27 Derossi D, Joliot AH, Chassaing G & Prochiantz A kinases in cell behaviour. Nat Rev Mol Cell Biol 4, 446– 456. 15 Nelson CM, Pirone DM, Tan JL & Chen CS (2004) (1994) The third helix of the Antennapedia homeodo- main translocates through biological membranes. J Biol Chem 269, 10444–10450. 28 Christiaens B, Grooten J, Reusens M, Joliot A, Goe- Vascular endothelial-cadherin regulates cytoskeletal ten- sion, cell spreading, and focal adhesions by stimulating RhoA. Mol Biol Cell 15, 2943–2953.
thals M, Vandekerckhove J, Prochiantz A & Rosseneu M (2004) Membrane interaction and cellular internali- zation of penetratin peptides. Eur J Biochem 271, 1187– 1197. 29 Viht K, Padari K, Raidaru G, Subbi J, Tammiste I, 16 Ishizaki T, Uehata M, Tamechika I, Keel J, Nonomura K, Maekawa M & Narumiya S (2000) Pharmacological properties of Y-27632, a specific inhibitor of rho-associ- ated kinases. Mol Pharmacol 57, 976–983. 17 Narumiya S, Ishizaki T & Uehata M (2000) Use
and properties of ROCK-specific inhibitor Y-27632. Methods Enzymol 325, 273–284. 18 Uehata M, Ishizaki T, Satoh H, Ono T, Kawahara Pooga M & Uri A (2003) Liquid-phase synthesis of a pegylated adenosine-oligoarginine conjugate, cell-perme- able inhibitor of cAMP-dependent protein kinase. Bioorg Med Chem Lett 13, 3035–3039. 30 Enkvist E, Lavo˜ gina D, Raidaru G, Vaasa A, Viil I,
T, Morishita T, Tamakawa H, Yamagami K, Inui J, Maekawa M et al. (1997) Calcium sensitization of smooth muscle mediated by a Rho-associated protein kinase in hypertension. Nature 389, 990– 994. Lust M, Viht K & Uri A (2006) Conjugation of adeno- sine and hexa-(d-arginine) leads to a nanomolar bisub- strate-analog inhibitor of basophilic protein kinases. J Med Chem 49, 7150–7159. 31 Turner JJ, Ivanova GD, Verbeure B, Williams D, 19 Tomasek JJ, Vaughan MB, Kropp BP, Gabbiani G,
FEBS Journal 275 (2008) 3608–3624 ª 2008 The Authors Journal compilation ª 2008 FEBS
3622
Arzumanov AA, Abes S, Lebleu B & Gait MJ (2005) Cell-penetrating peptide conjugates of peptide nucleic acids (PNA) as inhibitors of HIV-1 Tat-dependent trans- activation in cells. Nucleic Acids Res 33, 6837–6849. Martin MD, Haaksma CJ & Hinz B (2006) Contraction of myofibroblasts in granulation tissue is dependent on Rho ⁄ Rho kinase ⁄ myosin light chain phosphatase activ- ity. Wound Repair Regen 14, 313–320.
ARC effectively dissociates the actin cytoskeleton
H. Ra¨a¨ gel et al.
32 El-Andaloussi S, Johansson HJ, Lundberg P & Langel (2006) Induction of splice correction by cell-penetrat-
U¨ ing peptide nucleic acids. J Gene Med 8, 1262–1273. 33 Jones SW, Christison R, Bundell K, Voyce CJ, Brock- 47 Rothbard JB, Jessop TC, Lewis RS, Murray BA & Wender PA (2004) Role of membrane potential and hydrogen bonding in the mechanism of translocation of guanidinium-rich peptides into cells. J Am Chem Soc 126, 9506–9507.
bank SM, Newham P & Lindsay MA (2005) Charac- terisation of cell-penetrating peptide-mediated peptide delivery. Br J Pharmacol 145, 1093–1102. 34 Pooga M, Elmquist A & Langel U¨ (2002) Toxicity and
48 Kosuge M, Takeuchi T, Nakase I, Jones AT & Futaki S (2008) Cellular internalization and distribution of arginine-rich peptides as a function of extracellular pep- tide concentration, serum, and plasma membrane asso- ciated proteoglycans. Bioconjug Chem 19, 656–664. 49 Fuchs SM & Raines RT (2004) Pathway for polyargi-
nine entry into mammalian cells. Biochemistry 43, 2438– 2444.
side effects of cell-penetrating peptides. In Cell-Penetrat- ing Peptides: Processes and Applications (Langel U¨ , ed.), pp. 245–262. CRC Press LLC, Boca Raton, FL. 35 Senderowicz AM (2003) Small-molecule cyclin-depen- dent kinase modulators. Oncogene 22, 6609–6620. 36 Michie AM & Nakagawa R (2005) The link between PKCalpha regulation and cellular transformation. Immunol Lett 96, 155–162. 37 Anderson ME, Higgins LS & Schulman H (2006) Dis- 50 Honjo M, Tanihara H, Inatani M, Kido N, Sawamura T, Yue BY, Narumiya S & Honda Y (2001) Effects of rho-associated protein kinase inhibitor Y-27632 on intraocular pressure and outflow facility. Invest Ophthal- mol Vis Sci 42, 137–144.
ease mechanisms and emerging therapies: protein kinas- es and their inhibitors in myocardial disease. Nat Clin Pract Cardiovasc Med 3, 437–445. 38 Senderowicz AM (2002) The cell cycle as a target for 51 Katoh K, Kano Y, Amano M, Kaibuchi K & Fujiwara K (2001) Stress fiber organization regulated by MLCK and Rho-kinase in cultured human fibroblasts. Am J Physiol Cell Physiol 280, C1669–C1679.
cancer therapy: basic and clinical findings with the small molecule inhibitors flavopiridol and UCN-01. Oncolo- gist 7(Suppl. 3), 12–19. 39 Takami A, Iwakubo M, Okada Y, Kawata T, Odai H, 52 Katoh K, Kano Y, Amano M, Onishi H, Kaibuchi K & Fujiwara K (2001) Rho-kinase-mediated contraction of isolated stress fibers. J Cell Biol 153, 569–584.
Takahashi N, Shindo K, Kimura K, Tagami Y, Miyake M et al. (2004) Design and synthesis of Rho kinase inhibitors (I). Bioorg Med Chem 12, 2115–2137.
53 Moorman JP, Luu D, Wickham J, Bobak DA & Hahn CS (1999) A balance of signaling by Rho family small GTPases RhoA, Rac1 and Cdc42 coordinates cytoskele- tal morphology but not cell survival. Oncogene 18, 47– 57. 54 Rottner K, Hall A & Small JV (1999) Interplay between 40 Vive´ s E, Richard J-P, Rispal C & Lebleu B (2003) TAT peptide internalization: seeking the mechanism of entry. Curr Protein Pept Sci 4, 125–132.
Rac and Rho in the control of substrate contact dynamics. Curr Biol 9, 640–648. 55 Salhia B, Rutten F, Nakada M, Beaudry C, Berens M,
41 Futaki S, Niwa M, Nakase I, Tadokoro A, Zhang Y, Nagaoka M, Wakako N & Sugiura Y (2004) Arginine carrier peptide bearing Ni(II) chelator to promote cellu- lar uptake of histidine-tagged proteins. Bioconjug Chem 15, 475–481. 42 Nakase I, Niwa M, Takeuchi T, Sonomura K, Kawaba- Kwan A & Rutka JT (2005) Inhibition of Rho-kinase affects astrocytoma morphology, motility, and inva- sion through activation of Rac1. Cancer Res 65, 8792– 8800.
ta N, Koike Y, Takehashi M, Tanaka S, Ueda K, Simpson JC et al. (2004) Cellular uptake of arginine- rich peptides: roles for macropinocytosis and actin rearrangement. Mol Ther 10, 1011–1022.
56 Holtje M, Hoffmann A, Hofmann F, Mucke C, Grosse G, Van Rooijen N, Kettenmann H, Just I & Ahnert- Hilger G (2005) Role of Rho GTPase in astrocyte mor- phology and migratory response during in vitro wound healing. J Neurochem 95, 1237–1248. 57 Kozma R, Ahmed S, Best A & Lim L (1995) The 43 Fotin-Mleczek M, Fischer R & Brock R (2005) Endocy- tosis and cationic cell-penetrating peptides – a merger of concepts and methods. Curr Phar Des 11, 3613–3628. 44 Chen X & Wang Z (2001) Regulation of intracellular
Ras-related protein Cdc42Hs and bradykinin promote formation of peripheral actin microspikes and filopo- dia in Swiss 3T3 fibroblasts. Mol Cell Biol 15, 1942– 1952. trafficking of the EGF receptor by Rab5 in the absence of phosphatidylinositol 3-kinase activity. EMBO Rep 2, 68–74. 45 Fischer R, Waizenegger T, Kohler K & Brock R (2002)
58 Nobes CD & Hall A (1995) Rho, rac and cdc42 GTPas- es: regulators of actin structures, cell adhesion and motility. Cell 81, 53–62. 59 Takai Y, Sasaki T & Matozaki T (2001) Small
FEBS Journal 275 (2008) 3608–3624 ª 2008 The Authors Journal compilation ª 2008 FEBS
3623
GTP-binding proteins. Physiol Rev 81, 153–208. 60 Amano M, Chihara K, Nakamura N, Kaneko T, Matsuura Y & Kaibuchi K (1999) The COOH terminus A quantitative validation of fluorophore-labelled cell-per- meable peptide conjugates: fluorophore and cargo depen- dence of import. Biochim Biophys Acta 1564, 365–374. 46 Zaro JL & Shen WC (2003) Quantitative comparison of membrane transduction and endocytosis of oligopep- tides. Biochem Biophys Res Commun 307, 241–247.
ARC effectively dissociates the actin cytoskeleton
H. Ra¨ a¨gel et al.
Supplementary material
is available
of Rho-kinase negatively regulates rho-kinase activity. J Biol Chem 274, 32418–32424.
61 Lim MA, Yang L, Zheng Y, Wu H, Dong LQ & Liu F (2004) Roles of PDK-1 and PKN in regulating cell migration and cortical actin formation of PTEN-knock- out cells. Oncogene 23, 9348–9358.
The following supplementary material online: Fig. S1. (A) Dissociation of the actin cytoskeleton with 10 lm ARC in NIH 3T3 cells. (B) Dissociation of the actin cytoskeleton with 20 lm ARC in NIH 3T3 cells. (C) Effect of ARC on actin cytoskeleton in HeLa cells. Fig. S2. Co-localization of ARC with EGF. Fig. S3. Toxicity of ARC and Y-27632. Fig. S4. Effect of ‘classical’ CPPs on the actin cyto- skeleton of HeLa cells.
This material is available as part of the online article
from http://www.blackwell-synergy.com
62 Dong LQ, Landa LR, Wick MJ, Zhu L, Mukai H, Ono Y & Liu F (2000) Phosphorylation of protein kinase N by phosphoinositide-dependent protein kinase-1 medi- ates insulin signals to the actin cytoskeleton. Proc Natl Acad Sci USA 97, 5089–5094.
Please note: Blackwell Publishing are not responsible for the content or functionality of any supplementary materials supplied by the authors. Any queries (other than missing material) should be directed to the corre- sponding author for the article.
63 Totsukawa G, Yamakita Y, Yamashiro S, Hartshorne DJ, Sasaki Y & Matsumura F (2000) Distinct roles of ROCK (Rho-kinase) and MLCK in spatial regulation of MLC phosphorylation for assembly of stress fibers and focal adhesions in 3T3 fibroblasts. J Cell Biol 150, 797–806. 64 Darbre A (1986) Analytical methods. In Practical
FEBS Journal 275 (2008) 3608–3624 ª 2008 The Authors Journal compilation ª 2008 FEBS
3624
Protein Chemistry – A Handbook (Darbre A, ed.), pp. 230–335. John Wiley & Sons, Hoboken, NJ.
