Involvement of RSK1 activation in malformin-enhanced cellular fibrinolytic activity

Pharmacological interventions to enhance fibrinolysis are effective for treating thrombotic disorders. Utilizing the in vitro U937 cell line-based fibrin degradation assay, we had previously found a cyclic pentapeptide malformin A1 (MA1) as a novel activating compound for cellular fibrinolytic activity. The mechanism by which MA1 enhances cellular fibrinolytic activity remains unknown. In the present study, we show that RSK1 is a crucial mediator of MA1-induced cellular fibrinolysis. Treatment with rhodamine-conjugated MA1 showed that MA1 localizes mainly in the cytoplasm of U937 cells. Screening with an antibody macroarray revealed that MA1 induces the phosphorylation of RSK1 at Ser380 in U937 cells. SL0101, an inhibitor of RSK, inhibited MA1-induced fibrinolytic activity, and CRISPR/Cas9-mediated knockout of RSK1 but not RSK2 suppressed MA1-enhanced fibrinolysis in U937 cells. Synthetic active MA1 derivatives also induced the phosphorylation of RSK1. Furthermore, MA1 treatment stimulated phosphorylation of ERK1/2 and MEK1/2. PD98059, an inhibitor of MEK1/2, inhibited MA1-induced phosphorylation of RSK1 and ERK1/2, indicating that MA1 induces the activation of the MEK-ERK-RSK pathway. Moreover, MA1 upregulated the expression of urokinase-type plasminogen activator (uPA) and increased uPA secretion. These inductions were abrogated in RSK1 knockout cells. These results indicate that RSK1 is a key regulator of MA1-induced extracellular fibrinolytic activity.

. Rhodamine-conjugated MA 1 was observed mainly in the cytoplasm of U937 cells ( Fig. 1C and Fig. S1), and enrichment of MA 1 -rhodamine was also observed in cytoplasmic organelles in some of the cells (Fig. 1C). These results suggest that MA 1 may interact with intracellular molecule(s).

Antibody macroarray screening reveals that MA 1 induces phosphorylation of RSK1.
To address what intracellular mechanisms by which MA 1 induces fibrinolytic activity, we carried out a screening to identify differentially phosphorylated kinases (26 targets) between MA 1 -treated and non-treated cell lysates using membrane-based antibody macroarrays ( Fig. 2A). The result was represented as a Volcano plot (Fig. 2B). The negative log 10 function of the p-value was plotted against the log 2 function of the fold change, representing the up-and downregulated phosphorylated kinases in response to MA 1 treatment. All kinases with increasing or decreasing fold changes by at least 2-fold at a p-value cut-off of p < 0.05 were considered differentially phosphorylated as a result of the treatment. MA 1 significantly elevated the signal intensity of phosphorylated RSK1 (Ser380) (Fig. 2B,C, Table S1).

MA 1 -induced RSK1 phosphorylation is independent of fibrin and plasma components.
We further examined the phosphorylation status of RSK1 by western blotting. Phosphorylation of Ser380 residue of RSK1 was increased by MA 1 in U937 cells cultured in the presence of plasma components on the fibrin-coated plate (Fig. 3A). As we had previously shown that the uPA/plasminogen system and vitronectin were involved in MA 1 -induced fibrinolysis 19 , we next examined whether extracellular components affected the phosphorylation status of RSK1. When we cultured U937 cells in the absence of plasma components on non-coated plates, MA 1 also stimulated phosphorylation of RSK1 (Fig. 3B). These data show that fibrin and/or plasma components are not necessary for the ability of MA 1 to stimulate RSK1 phosphorylation in U937 cells. RSK1 is essential for MA 1 -enhanced fibrinolytic activity. The phosphorylation of RSK1 at several sites including Ser380 is important for its kinase activation and downstream signaling [22][23][24][25][26][27] . To address whether RSK1 activation was involved in MA 1 -enhanced cellular fibrinolytic activity, we co-treated U937 cells with an Macroarray membranes were blocked with blocking buffer and then simultaneously reacted with cell lysates and a biotinylated antibody cocktail. After washing, the membranes were incubated with streptavidin-conjugated HRP. Subsequently, multiple proteins were detected using the ECL system. (B) Volcano plot of antibody macroarray data created by quantifying the mean spot pixel densities for MA 1 -treated versus non-treated control U937 cell lysate. The x-axis is the fold change between the two samples represented as log 2 ([MA 1 ]/ [Control]) and the y-axis is the significance between the two samples represented as −log 10 (p-value). Statistical analyses were performed using the Student's t test. Green lines show cut-off values (fold change >2, p-value <0.05). The macroarray was conducted with n = 3. (C) Macroarray images showing the spots of phospho-RSK1 (Ser380). Upper; non-treated control, lower; MA 1 -treated.
Biologically active MA 1 derivative also induces phosphorylation of RSK1. Our previous structure-activity relationship study showed that the disulfide bond and the bulky hydrophobic side chains in MA 1 play a crucial role in exerting pro-fibrinolytic activity 29 . We further examined the effects of synthetic MA 1 derivatives on RSK1 phosphorylation. Three MA 1 derivatives (reduced MA 1 , dimethyl MA 1 , and 3-LysMA 1 ), which have been found not to retain fibrinolysis-enhancing activity, did not induce phosphorylation of RSK1 (Fig. 5A,B). On the other hand, a biologically active derivative, 3-BocLysMA 1 , induced RSK1 phosphorylation (Fig. 5A,B). Thus, there was a positive correlation between the biological activity and the induction of RSK1 phosphorylation in MA 1 derivatives. MA 1 induces the activation of the MEK-ERK-RSK pathway. RSK1 is activated by MAP kinases of the ERK family in response to growth factors, hormones, and other stimuli 21 . We investigated whether ERK1/2, an upstream kinase of RSK1, was activated in MA 1 -treated cells. MA 1 treatment caused the phosphorylation of ERK1/2 (Fig. 6A). Furthermore, MA 1 also upregulated phosphorylation levels of MEK1/2, an upstream kinase . Statistical analyses were performed using one-way ANOVA with Tukey's post-hoc test ( ** p < 0.01, versus control). (C) Effect of SL0101 on MA 1 -enhanced fibrinolytic activity. For the in vitro fibrin degradation assay, U937 cells with human platelet-poor plasma were added to each well of the 125 I-fibrin-coated 96-well microplate. After incubation for 3 h with MA 1 and SL0101 at the indicated concentrations, 125 I-fibrin degradation products released into the supernatant were quantified using a γ-counter. Data were presented as means ± SD (n = 3). Statistical analyses were performed using two-way ANOVA with Tukey's post-hoc test ( ** p < 0.01, versus control).

MA 1 increases the expression of the uPA gene and secretion of uPA in an RSK1-dependent manner.
Although MA 1 -induced RSK activation increased fibrinolytic activity, it remains unknown what fibrinolytic factors were elevated. Our previous study using neutralizing antibodies has shown that uPA was involved in MA 1 -enhanced activity 1 . Therefore, we examined uPA gene expression levels. The expression of the uPA gene was upregulated in MA 1 -treated cells (Fig. 7A). Furthermore, when the uPA protein level was examined, MA 1 induced an increase in uPA secretion into the conditioned medium (Fig. 7B). Next, we investigated the effects of RSK1 KO. In contrast to the RSK2 KO clone (RSK2#24), MA 1 did not significantly upregulate the expression of the uPA gene in RSK1 KO clones (RSK1#9 and RSK1#14) (Fig. 7C). Similarly, MA 1 could not increase the secretion of uPA in RSK1 KO clones (Fig. 7D). These results indicate that MA 1 induced the activation of RSK1 followed by an increase in uPA expression, and enhanced extracellular fibrinolytic activity (Fig. 8).

Discussion
In our previous report, we had shown that MA 1 -treated U937 cells exhibited changes in actin organization and focal accumulation of plasminogen on the cell surface 19 . In the present study, the fluorescence-labeled MA 1 , when added to the culture of U937 cells, was primarily localized to the intracellular compartment (Fig. 1C). Based on these observations, we hypothesized that MA 1 may affect the function of intracellular molecules. Indeed, we showed here that RSK1 was essentially activated by MA 1 and thereby enhanced fibrinolytic activity. Activation of intracellular signaling by MA 1 had also been previously reported to have other effects including cytotoxic effects of MA 1 on prostate cancer cells involving the accumulation of reactive oxygen species, a decrease in Data were presented as means ± SD (n = 3). Statistical analyses were performed using one-way ANOVA with Tukey's post-hoc test ( * p < 0.05, ** p < 0.01, versus control).
ScIeNtIfIc RepoRts | (2018) 8:5472 | DOI:10.1038/s41598-018-23745-0 mitochondrial transmembrane potential, and induction of autophagy 10 . Importantly, MA 1 induced the activation of ERK, an upstream kinase of RSK, and its activation was considered to occur in a MEK-dependent pathway (Fig. 6). Localization at the intracellular compartment of MA 1 may be necessary for efficient activation of the MEK-ERK-RSK pathway, which increases cellular fibrinolytic activity. The identification of direct target molecules of MA 1 that activates MEK is currently under investigation.
The RSK family is comprised of a group of highly related serine/threonine kinases that regulate diverse cellular processes, including cell growth, proliferation, survival, and motility. This family includes four vertebrate isoforms (RSK1-4). Mutations in the RSK2 isoform are known to be associated with Coffin-Lowry syndrome 31 , and a gene knockout study of RSK2 in mice shows that RSK2 is a modulator of craniofacial development 32 . Results from these previous reports plus the results of this study that RSK1 KO specifically suppresses MA 1 -induced fibrinolytic activity (Fig. 4B, and Fig. 7C,D) lead us to speculate that RSK family proteins are exhibit isoform-specific functions. As the phenotypes of RSK1 KO mice have not yet been reported, the significance of RSK1 in fibrinolysis in vivo remains elusive.
Mammalian RSK was reported to function as an important effector of ERK in global transcriptional regulation 33 . RSK was shown to regulate ~20% of mRNAs controlled by ERK in Madin-Darby canine kidney cells through direct and indirect mechanisms, such as the induction of transcription factors. On the other hand, there have been several reports describing the regulation of uPA activity via the activation of RSK 33,34 . Activated RSK regulates epithelial cell motility and invasiveness by stimulating the gene expression of the uPA/uPA receptor 33 , and activated RSK in breast carcinoma enhances cellular invasion through phosphorylation of IκBα, allowing NF-κB to translocate into the nucleus and induce uPA expression 34 . We had previously shown that uPA rather than tPA was likely to be involved in MA 1 -enhanced fibrinolytic activity 1 , and in this study, MA 1 consistently upregulated uPA mRNA levels and increased uPA protein secretion (Fig. 7A,B). Target gene clusters for MA 1 -induced RSK1-mediated transcription activation will be further elucidated in the future.
In summary, we demonstrated that the phosphorylation of RSK1 at Ser380 was crucial for MA 1 -induced cellular fibrinolysis in U937 monocytoid cells. Pharmacological and genetic intervention of RSK1 suppressed MA 1 -enhanced fibrinolytic activity. In addition, MA 1 resulted in the activation of the MEK-ERK pathway followed by the activation of RSK. Further, MA 1 led to the induction of uPA expression in an RSK-dependent manner. Although further studies are needed, the potential applications of MA 1 derivatives or RSK1-activating compounds as therapeutics may contribute to the development of better antithrombotic treatments for severe ischemic diseases.
Malformin A 1 and its synthetic derivatives. MA 1 was purified from the culture broth of Aspergillus niger F7586 as previously described 19 . Reduced MA 1 , Dimethyl MA 1 , 3-LysMA 1 , and 3-BocLysMA 1 were synthesized using solid-phase peptide synthesis according to previous work 29 . MA 1 -rhodamine was synthesized by conjugation of 3-LysMA 1 with NHS-rhodamine in 20 mM phosphate buffer (pH 7.4) at 37 °C for 2 h, and then purified by preparative HPLC.
Cell culture. U937 cells were obtained from Japanese Collection of Research Bioresources, and were cultured in RPMI-1640 including 10% FBS, 100 units/mL penicillin, and 100 µg/mL streptomycin. 293FT cells were obtained from Thermo Fisher Scientific, and were cultured in complete medium including DMEM, 10% FBS, 6 mM L-glutamine, 0.1 mM non-essential amino acids, 1 mM sodium pyruvate, 100 units/mL penicillin, and 100 µg/mL streptomycin. Cells were cultured in a humidified 5% CO 2 atmosphere at 37 °C.
Fluorescence imaging. For MA 1 -rhodamine staining, U937 cells (5 × 10 6 cells/ml) were incubated with 5 µM MA 1 -rhodamine in 10% human platelet-poor plasma on fibrin-coated 8-chamber slide at 37 °C for 1 h. After the reaction, cells were fixed with 3.7% paraformaldehyde at room temperature for 20 min. The cells were U937 cells were treated with MA 1 at the indicated concentration at 37 °C for 3 h. Conditioned media were concentrated ~20-fold and subjected to western blot analysis for the detection of uPA. Data were presented as means ± SD (n = 3). Statistical analyses were performed using one-way ANOVA with Tukey's post-hoc test ( * p < 0.05, ** p < 0.01, versus control). (C) Quantitative real-time PCR for the determination of uPA gene expression levels in CRISPR/Cas9-mediated U937 knockout clones. Knockout clones (RSK1#9, RSK1#14, or RSK2#24) were treated with 5 μM MA 1 at 37 °C for 1 h. RNA was then extracted and the expressions of uPA gene (PLAU) were quantified by qPCR. (D) Western blot analysis of uPA in the conditioned medium of CRISPR/Cas9-mediated U937 knockout clones. Knockout clones (RSK1#9, RSK1#14, or RSK2#24) were treated with 5 μM MA 1 at 37 °C for 3 h. Conditioned media were concentrated ~20-fold and subjected to western blot analysis for the detection of uPA. Data were presented as means ± SD (n = 3). Statistical analyses were performed using two-way ANOVA with Tukey's post-hoc test ( * p < 0.05, ** p < 0.01, versus control).

Membrane-based antibody macroarray.
Membrane-based antibody macroarrays were performed using Proteome Profiler Human phosphorylated kinase Arrays (ARY002B; R&D Systems). The procedure was carried out according to the manufacturer's instruction. Briefly, U937 cells (5 × 10 6 cells/ml) were treated with 5 µM MA1 in 10% human platelet-poor plasma on a fibrin-coated 24-well microplate at 37 °C for 1 h. After treatment, cell lysates were prepared in lysis buffer (contained in the kit) and clarified by centrifugation. After determination of protein concentration, equal amounts of protein were subjected to macroarrays. Macroarray membranes were blocked with blocking buffer (contained in the kit) and then simultaneously reacted with cell lysates and a biotinylated antibody cocktail. After washing, the membranes were incubated with streptavidin-conjugated horseradish peroxidase. Subsequently, multiple phosphorylated kinases were detected using the ECL system (GE Healthcare). The macroarray was conducted with n = 3.
Western blot analysis. U937 cells (5 × 10 6 cells/ml) were treated with test compounds at the indicated concentrations in the presence or absence of 10% human platelet-poor plasma on a fibrin-coated or non-coated dish at 37 °C for 1 h. After treatment, cell lysates were prepared in lysis buffer (50 mM Tris-HCl [pH 7.5], 150 mM NaCl, 1 mM EDTA, 1% NP-40, 20 mM NaF, 2 mM Na 3 VO 4 , and protease inhibitor cocktail [Roche Diagnostics]) and clarified by centrifugation. After determining the protein concentration, equal amounts of protein were subjected to SDS-PAGE under reducing conditions. For uPA detection, U937 cells were treated with MA 1 for 3 h in RPMI-1640 medium containing 0.5% FBS. After treatment, conditioned media were concentrated ~20-fold using Amicon ultra filter devices (Merck Millipore), and then subjected to SDS-PAGE under non-reducing conditions. Next, proteins were transferred to a nitrocellulose membrane, and then blocked with 5% skim milk and then incubated with specific antibodies. After washing, the membranes were incubated with peroxidase-conjugated secondary antibody. Subsequently, targeted proteins were detected using the ECL system.
In vitro fibrin degradation assay. The In vitro fibrin degradation assay was performed using an 125 I-labeled fibrin-coated plate as previously described 1,19 . Briefly, each well of a 96-well microplate was coated with 125 I-fibrinogen (ca. 50,000 cpm/20 µg/well) and then treated with 0.68 units/ml thrombin to convert 125 I-fibrinogen to 125 I-fibrin. Fifty microliters of U937 cells at a concentration of 5 × 10 6 cells/ml in 10% human platelet-poor plasma were added to each well of the 125 I-fibrin-coated 96-well microplate. After incubating at 37 °C for 3 h with samples at the indicated concentrations, 125 I-fibrin degradation products released into the supernatant were quantified using the γ-counter AccuFLEX γ7010 (Hitachi).
Lentivirus preparation, transduction, and single cell cloning. Single guide RNAs (sgRNAs) for RSK1 and RSK2 were designed according to the sgRNA sequences in the genome-scale CRISPR knockout library GeCKOv2 (http://genome-engineering.org/gecko/?page_id=15) (Table 1), and oligonucleotides for sgRNA were annealed and cloned into the lentivirus transfer vector lentiCRISPRv2 at the BsmBI restriction site 35 . For the production of recombinant lentiviral particles, the transfer plasmids were co-transfected with the packaging plasmids pMD2.G and psPAX2. Briefly, for each virus, 80% confluent 293FT cells in 6-well plate were transfected in Opti-MEM using 1 µg of the transfer plasmid, 0.5 µg pMD2.G, 0.5 µg psPAX2, and 6 µl of FuGENE HD. After 12 h, the media were replaced with fresh media. After 48 h, the supernatants containing the viral particles were harvested and filtrated through a 0.45 µm-pore membrane. For each viral construct, 1 × 10 6 U937 cells were transduced in 2 ml of media with 1 ml of viral supernatant in wells of a 6-well plate. At 48 h post-transduction, the media was changed to media containing 2 µg/ml puromycin, and the cells were then further cultured in media with puromycin. For single cell cloning of RSK1-or RSK2-knockout cells, puromycin-selected cells were plated onto a 96-well microplate at a density of 0.5 cells/well and cultured in puromycin-free medium. At 20 days post-plating, the knockout efficiency in 40 single cell clones was examined by western blotting.
Quantitative real-time reverse transcription PCR. U937 cells (2 × 10 6 cells/ml) were treated with MA 1 at the indicated concentrations at 37 °C for 1 h. After treatment, total RNA was isolated using TRIzol reagent according to the manufacturer's instructions. The cDNA was synthesized using the PrimeScript RT reagent kit. Quantitative real-time PCR was performed with SYBR Premix Ex Taq II in a Thermal Cycler Dice Real Time System II (Takara). The relative fold change in mRNA expression was calculated using the ΔΔC t method. GAPDH was used as reference gene. Sequences of the primers used were as followed: PLAU_for, CAC GCA AGG GGA GAT GAA; PLAU_rev, ACA GCA TTT TGG TGG TGA CTT; GAPDH_for, CTT CAC CAC CAT GGA GAA GGC; GAPDH_rev, GGC ATG GAC TGT GGT CAT GAG.
Data availability. All data generated or analyzed during this study are included in this published article and its Supplementary Information files.