Novel epothilone analogs were obtained from Biostar Technologies (Beijing, China). All the cell culture reagents were purchased from Invitrogen (San Diego, USA). MTT, Triton X-100, β-tubulin antibody, Hoechst 33342, and Matrigel were purchased from Sigma-Aldrich (St. Louis, USA). Rac1, Cdc42 and p-PAK1 primary antibodies were obtained from Santa Cruz Biotechnology (Santa Cruz, USA). p-ERK1/2, p-JNK, p-p38, p-Akt, PAK1, ERK1/2, JNK, Akt and p38 antibodies were purchased from Abcam (Cambridge, UK). Horseradish peroxidase (HRP)-conjugated IgG and FITC-labeled mouse IgG were acquired from Jackson ImmunoResearch Laboratories (West Grove, USA). bFGF and fibronectin were obtained from R&D systems (Minneapolis, USA). The ECL regent was purchased from GE Healthcare Life Science (Pittsburgh, USA). The polyvinylidene fluoride (PVDF) membrane was acquired from Millipore (Massachusetts, USA). Superscript II reverse transcriptase was the product of Takara (Dalian, China). UTD2 was initially dissolved in dimethyl sulfoxide (DMSO) as a stock solution and subsequently diluted to the desired experimental concentrations with culture media.

The human breast cancer cell lines MCF-7, MDA-MB-231 were purchased from the American Type Culture Collection (ATCC) (Manassas, USA). The stable cell lines expressing constitutively active Rac1 or dominant negative Rac1 were established by transfecting MCF-7 cells with the myc-tagged Rac1 mutant expression plasmids, that is, myc-tagged Rac1V12 (mRac1V₁₂)and myc-tagged Rac1N17 (mRac1N₁₇), and selected for stable expressing clones. The expression of the myc-tagged Rac1 mutants was screened by Western blot using anti-myc tag antibody. The cells were cultured in Dulbecco's modified Eagle's medium (DMEM), supplemented with 10% fetal calf serum (FCS), 100 U/mL penicillin and 100 µg/mL streptomycin at 37℃ in 5% CO₂.

The antiproliferative effects of novel epothilone analog UTD2 were assessed by 3-(4,5-dimethylthiazol-2-yl)-2,5 diphenyl tetrazolium bromide (MTT) assay as described before [6]. The UTD2 stock solution was further diluted to the appropriate concentrations in culture media immediately before use. The cells were plated in culture medium in 96-well plates (3000 cells per well, in quadruplicate) and were treated with UTD2 at concentrations ranged from 0 to 1000 nM. The cells were incubated with the drug for up to 72 hours and the number of viable cells was then determined. The results were expressed in terms of IC₅₀ and IC₁₀ (the highest non-toxic concentration of a drug, HNTC) values, which represent the drug concentrations responsible for 50% and 10% inhibition in the growth of cells, respectively, compared to the control cell growth. Both the IC₅₀ and the IC₁₀ concentrations were used in further experiments.

Breast cancer cells were seeded on glass coverslips in a 24-well plate. At 80% confluence, the cells were treated with or without UTD2 (IC₁₀, IC₅₀) for 24 hours and then were fixed with 4% Polyoxymethylene. Before fixation, cells were not washed with phosphate-buffered saline (PBS) to preserve their cytoskeletal integrity. The fixed cells were then incubated with 5% bovine serum albumin (BSA) in PBS for 30 min to block non-specific reactions. After blocking with BSA, the cells were permeated with 0.1% Triton X-100 in PBS and incubated with β-tubulin antibody (1:50) for 2 hours. After three PBS washes for 15 min, the cells were incubated with FITC-labeled anti-mouse IgG (1:120) for 30 min to visualize β-tubulin. After PBS washes, the cells were incubated with Hoechst 33342 (10 µg/mL) for 10 min to stain the nuclei. The images were captured by confocal laserscan microscope through a 100×objective lens and subsequently analyzed with SlidebookTM software (Version 4.0; Intelligent Imaging Innovations Inc., USA).

MCF-7 cells were seeded in a 6-well plate at a density of 1.5×10⁶ cells per well. Before plating, two parallel lines were drawn at the underside of each well with a marker. These lines will serve as fiducial marks for the wound areas to be analyzed. The cells were serum starved overnight before performing the assay. On the day of analysis, the monolayer was completely confluent. Two parallel scratches that were approximately 200 µm in width were made perpendicular to the marker lines with a sterile pipette tip. This procedure allows the entire width of the wound to be imaged using a 10× objective lens. The wounds were observed using phase contrast microscopy on an inverted microscope.

The cells migration was measured in a 24-well plate transwell system containing inserts with a fluorescence blocking filter and a pore size of 8 µm. MCF-7 cells were seeded in the insert at 10⁵ cells per well. The cells were treated with UTD2 (IC₁₀, IC₅₀) for 16 hours or culture media as control. The upper media was serum free, culture media containing 20% FCS and 30 ng/mL Basic fibroblast growth factor (bFGF) were added to the lower compartment of the system to elicit tumor cell migration. The cells were allowed to migrate for 16 hours at 37℃ in 5% CO₂. Cotton swabs were used to clean off the non-migrated cells on the top of the transwell. To fix the migrated cells, 4% Polyoxymethylene dissolved in PBS was applied for 15 min at room temperature, and 0.1% crystal violet was used to stain the cells for 10 min. The migrated cells are observed using phase contrast microscopy on an inverted microscope.

The breast cancer cells (MCF-7) invasion was measured in a 24-well plate transwell system containing inserts with a fluorescence blocking filter and a pore size of 8 µm. The inserts were coated on the bottom with 2 mg/mL fibronectin, washed with PBS and coated on the upper side with 5 mg Matrigel in 100 mL medium. A number of 10⁵ tumor cells were seeded on top of the Matrigel layer in each well and were allowed to settle for 16 hours. Meanwhile, cells were exposed to UTD2 (IC₁₀, IC₅₀), or were left untreated (control) followed by replacement by culture media without FCS in the upper compartment. Culture media containing 20% FCS and 30 ng/mL bFGF was added to the lower compartment of the system to elicit tumor cell invasion through the Matrigel layer. 16 hours later, the non-invasive cells on the top of the transwell were clean off by cotton swabs. 4% Polyoxymethylene dissolved in PBS fixed the invasive cells for 15 min at room temperature, and 0.1% crystal violet stained the cells for 10 min. The invasive cells are observed using phase contrast microscopy on an inverted microscope.

The assays of cell adhesion to immobilized Matrigel were based on that described in Faull et al. [7]. 200 µL dilutions of 4 mg/mL Matrigel in media without serum were incubated inside the wells of a 96-well plate 30 min at 37℃. After washing with PBS, the wells were blocked with 2% BSA in PBS for 2 hours at room temperature. Cells were washed three times in PBS and resuspended at 10⁶ cells/mL, and 50 µL aliquots were added to each coated well. To test the effect of UTD2 on the cell adhesion, the cells were preincubated with UTD2 (IC₁₀, IC₅₀) for 30 min at room temperature and washed twice, before addition to the wells. After 2 hours incubation at 37℃ and 5% CO₂, the non-adherent cells were washed off with two rounds of gentle pipetting, the residual adherent cells were fixed with 4% Polyoxymethylene dissolved in PBS for 15 min at room temperature, and the cells were stained with 0.1% crystal violet for 10 min. Stained cells were washed three times with PBS. After washing, the staining reaction was terminated by the addition of 50 µL of 1% SDS and read in an ELISA plate reader with a 570 nm filter. Background values were determined in wells coated with 2% BSA.

The breast cancer cells were grown to 80% confluence in 10 cm petri-dishes and rendered quiescent in culture media without FCS overnight. Rac1/Cdc42 activation was initiated by the addition of complete culture media (10% FCS), while 5 nM, 10 nM or 20 nM UTD2 were added concurrently. After four hours, Rac1/Cdc42 activity was analyzed using a Rac1/Cdc42 activation assay. Briefly, after a washing step, the cells were lysed in the assay buffer, which was supplemented with 0.5 mM trypsin inhibitor, 0.5 mg/mL leupeptin, 1 mM PMSF, and subsequently centrifuged to remove cell debris. The cell lysate was incubated with 10 mg of agarose that was conjugated with p21-binding domain of PAK1 which binds both activated (GTP-bound) Rac1 and Cdc42, overnight at 4℃. Agarose beads bound with activated Rac1/Cdc42 were washed four times in the assay buffer, resuspended in 30 µL SDS-sample buffer (10% glycerol, 3% SDS, 20 mg/mL bromophenol blue, 25 mg/mL β-mercaptoethanol, 0.6 M Tris, pH 6.7). The samples were subjected to 12% polyacrylamide gel electrophoresis. The activated Rac1/Cdc42 protein levels in each sample were analyzed by Western blot using a rabbit polyclonal antibody against Rac1 (1:1000), or a mouse monoclonal antibody against Cdc42 (1:1000).

The activity of matrix metalloproteins (MMPs) MMP-2 was determined using slightly modified substrate-impregnated gels. The breast cancer cells MDA-MB-231 were plated in a 24-well plate with 3×10⁵ cells per well and then treated with UTD2 (0 nM, 5 nM, 10 nM or 20 nM) for 24 hours. After each treatment, the culture media was replaced with serum-free media and incubated overnight. The culture supernatants were collected and proteins were separated on a gelatin-impregnated (1 mg/mL) 10% SDS-PAGE. The gels were then incubated for 18 hours at 37℃ in 50 mmol/L Tris, 0.2 mol/L NaC1, 5 mmol/L CaC1₂, and 1% Triton X-100 (pH 7.6) and stained with 0.5% Coomassie blue G 250 in methanol/acetic acid/H₂O. The gelatinolysis was visualized as transparent bands.

The MDA-MB-231cells were seeded in 6-well plate, and then treated with UTD2 (0 nM, 5 nM, 10 nM or 20 nM) for 24 hours. The total RNA was isolated using Trizol reagent on the treated cells. Aliquots of 1 µg of total RNA were used for the first-strand cDNA synthesis. The reaction volume was 50 µL and used 100 units of Superscript II reverse transcriptase. The primers [8] that were used for human MMP-2 amplification were the following: 5'-GTGCTGAAGGACACACTAAAGAAGA-3' (sense) and 5'-TTGCCATCCTTCCAAAGTTGTAGG-3' (antisense). The primers used for β-actin were the following: 5'-GTGGGGCGCCCCAGGCACCA-3' (sense) and 5'-CCTTAATGTCACGCACG-3' (antisense). The PCR amplifications were run for 30 cycles at 94℃ for 30 seconds, 55℃ for 30 seconds, 72℃ for 1 min, and 72℃ for 7 min; the samples were then held at 4℃. The PCR products were analyzed via electrophoresis in 1% agarose gels and staining with ethidium bromide.

The focus formation assay was carried out according to that described by Qiu et al. [9]. NIH 3T3 cells were transfected with plasmid constructs expressing activated Rac1 and Raf, and the transfected cells were incubated in culture media containing UTD2 (IC₅₀) over night. The cells were then maintained in drug-free media. The number of transformed foci induced by Rac1, Raf alone or by Rac1 plus Raf were quantified about 17 days post transfection. Transfections were performed in triplicate, and data represented the average of three independent experiments.

Cells were plated in 10 cm petri-dishes with 8×10⁶ cells per dish. The cells were grown to about 80% confluency. The culture media was replaced by media containing the novel epothilone analog UTD2 (0 nM, 5 nM, 10 nM or 20 nM), and the cultures were further incubated for 24 hours. The treated cells were lysated as described in Ahn et al. [10]. Briefly, the treated cells were gently washed three times with cold PBS, scraped into ice-cold lysis buffer (150 mM NaCl, 20 mM MgCl₂, 2 mM EGTA, 50 mmol/L Tris pH 8.0, 1%Triton X-100) and lysed for 20 min on ice. The lysates were centrifuged at 12,000 ×g and 4℃ for 10 min. Samples were first separated by 12% SDS-PAGE and then transferred to a PVDF. The membrane was blocked with 5% nonfat milk in PBS for 1 hour, incubated overnight with primary antibodies (either anti-phosphorylated PAK1 (pS199/204), anti-phosphorylated JNK (pT183/Y185), antiphosphorylated ERK1/2 (pT202/Y204), anti-phosphorylated p38 (pT180/Y182), anti-phosphorylated Akt (pS473), anti-GAPDH), and then probed for 1 hour with secondary HRP-conjugated anti-mouse or anti-rabbit IgG. After extensive washing with PBST, the target proteins were detected on the membranes by enhanced ECL reagent.

The results are presented as the mean±SE for each treatment group in each experiment. The data were subjected to statistical analysis by one-way ANOVA, which was followed by the Scheffe post-hoc test, which used SPSS software (SPSS Inc., USA). A probability value of p<0.05 was considered statistically significant.

The cells were exposed to UTD2 for 72 hours. The antiproliferative effects of the drug on MCF-7 and MDA-MB-231 cells were determined by the MTT assay. The HNTC (IC₁₀) and the IC₅₀ values are shown in Table 1. UTD2 inhibited 50% of the growth of breast cancer cells at 16 nM. To ensure the validity of the IC₁₀ value, the inhibition of cell growth had to be less than 10% as compared to the control cells. The last column of Table 1 gives the mean percentage of cell growth when breast cancer cells were treated with the IC₁₀ concentration of UTD2.

The microtubule cytoskeletal changes in the MCF-7 were examined after 24 hours of drug exposure by confocal laserscan microscopy using an anti-tubulin antibody. The untreated MCF-7 cells showed good distribution and integrity of the microtubules throughout the cell, especially in the direction of movement (Fig. 1A arrowhead). The treated cells were more compact. The microtubules were still visible after exposure to UTD2 at IC₁₀, but the filiforms were shorter, and tubulin was primarily present closer to the nucleus (Fig. 1B arrowhead). The MCF-7 cells treated with UTD2 at IC₅₀ showed no clear movement in any direction, only a thick layer of tubulin around the nucleus (Fig. 1C arrowhead). These results indicate that UTD2 treatment led to significantly higher signal intensity of tubulin around the nucleus as compared with that of control cells. Furthermore, exposure to very low concentrations of UTD2 appears to be sufficient to impair microtubule integrity and localization of the tubulin.

Cell movement is coordinated by the actin cytoskeleton [11]. UTD2 treatment seemed to eliminate the movement of MCF-7 cells, suggesting that the drug may act on the actin cytoskeleton. We then investigated the effect of UTD2 on actin cytoskeleton-related cell processes, such as motility, migration, adhesion and invasion, as well as on the activity of the critical modulators of the actin cytoskeletal organization.

The wound closure was imaged under microscope at different times (up to 24 hours) and was illustrated by a representative experiment (Fig. 2A). The control cells filled the wound by 50% after 24 hours. Treatment with UTD2 significantly inhibited the wound closure abilities of the MCF-7 cells at the IC₅₀. Fig. 2B depicts the wound closure at 24 hours after treatment as a percentage of control wound width. Only 70% of the wound was filled by the cells treated with the IC₁₀ of UTD2 after 24 hours compared to the control wound width. After exposure to the IC₅₀, the wound closure was only 40% of the control width (Fig. 2B, p<0.01).

The transwell system was used to investigate the effects of UTD2 on migration and invasion. Our results demonstrated that UTD2 inhibited the migration of MCF-7 (Fig. 2C). Compared to the control cells, UTD2 treated cells exhibited only approximately 70% cell migration at the IC₁₀ and 50% at IC₅₀ concentration (Fig. 2D, p<0.01).

UTD2 also showed strong inhibition of MCF-7 cell invasion (Fig. 2E), which agreed with the results from migration assays. The reduction of invasion caused by the novel epothilone analog was dramatic in MCF-7 cells (Fig. 2F, p<0.01).

To study the effect of UTD2 on the ability of the breast cancer cells to attach to the extracellular matrix (ECM), an adhesion assay was performed. The results are displayed in Fig. 2G. Without the addition of UTD2, cells were able to attach well to ECM. However, when cells were grown in media containing UTD2, the cells showed a lower affinity for ECM and were not able to attach; the relative adhesion was 70% at IC₅₀ compared to the controls.

Comprehensive studies have elaborated the interaction between Rho GTPases and the cytoskeleton [5]. Rac1 and Cdc42 are two well-known critical regulators of the actin cytoskeleton; they play important roles in control of cell motility, migration and adhesion. The activity of both Rac1 and Cdc42 can be tightly regulated through the dynamics of actin and tubulin. To elucidate the mechanism for the UTD2-mediated inhibition of the actin related cellular processes, a pull-down assay was carried out to measure the levels of activated Rac1 and Cdc42 after drug treatment. After 4 hours of drug exposure, the GTP bound active Rac1 and Cdc42 from the total cell lysates were pulled down using the Rac1/Cdc42-binding domain of PAK1; the data was then analyzed by Western blot (Fig. 3). The UTD2 treatment resulted in the inhibition of activated GTP-bound Rac1, while the GTP-bound Cdc42 remained unchanged. Addition of UTD2 yielded no change in the levels of both total Rac1 and Cdc42 proteins. These results reflected differential effects of the novel epothilone analog on Rac1 and Cdc42 activation.

Many previous studies using dominant negative mutants of Rac1 have demonstrated that inhibition of Rac1 would lead to inhibition of actin cytoskeleton and cell motility [12]. Therefore, if the effect of UTD2 on cell motility is due to inhibition of Rac1, then it should be expected that the dominant negative mutant of Rac1 will potentiate the UTD2 effect, and constitutively active mutant of Rac1 should antagonize the effect. To confirm these possibilities, MCF-7 cell lines stably expressing myc-tagged Rac1 mutants or empty vector were established. Expression of correct myc-tagged Rac1 mutants in these clones was confirmed by Western blot analysis (Fig. 4A). The effect of UTD2 on migration of these cell lines was tested. As shown in Fig.4B, UTD2-induced inhibition of migration was neutralized in mRac1V₁₂-expressing cells, but was augmented in mRac1N₁₇-expressing cells when compared to parental MCF-7 or vector control cells. Relative migration of mRac1V₁₂-expressing cells was 1.5-fold higher than of MCF-7 cells following UTD2 treatment (Fig. 4B). In contrast, the migration of mRac1N₁₇-expressing cells was about only 2/3 of the control cells (Fig. 4B). These findings demonstrated that UTD2 decreased tumor cell migration via suppressing Rac1 activity.

To further elucidate the downstream signaling pathway of Rac1 that attributes to the effects of UTD2 on actin cytoskeleton and cell motility, we tested the effect of UTD2 on the phosphorylation of PAK1 through Western blotting. PAK1 is an effector of Rac1 which has been shown to mediate Rac1 controlled pathways that lead to actin cytoskeletal-related cellular processes, such as cell spreading, cell motility and cell migration. The phosphorylation of PAK1 is essential for its activation and signaling [13]. As shown in Fig. 5A, the novel epothilone analog significantly reduced the phosphorylation of PAK1 in human breast cancer cells (p<0.01). The total expression levels of PAK1 remained the same in MCF-7 cells with or without treatment of UTD2. These results demonstrate that UTD2 efficiently blocked the phosphorylation of PAK1 in human breast cancer cells in a dose-dependent manner.

Akt is a serine/threonine protein kinase that has been implicated in cell migration and is an important regulator of cell survival. It has been previously reported that inhibition of PAK results in attenuation of Akt phosphorylation and decreased tumor cell viability [14]. To determine if the Akt-mediated signaling pathway was affected by UTD2, we tested the effect of UTD2 on the phosphorylation of Akt. As shown in Fig. 5B, 20 nM and 10 nM of UTD2 significantly reduced Akt phosphorylation in human breast cancer cells (p<0.01), and 5 nM of the epothilone only slightly decreased Akt phosphorylation (p<0.05), UTD2 efficiently inhibited activation of cell survival kinase Akt in human breast cancer cells in a dose-dependent manner. The total expression levels of Akt remained the same in MCF-7 cells with or without treatment by UTD2. It has been demonstrated that PI3K/Akt pathway can be downstream of PAK1, and not only is involved in the regulation of cell survival, but also in cell motility [15]. Therefore, our findings indicated that UTD2 may inhibit both cancer cell growth and motility via the Rac1/PAK/PI3K/Akt pathway.

MAPK signaling cascades control mitogenic and apoptotic responses. It has been reported that the three types of MAPKs, ERK1/2, p38 and JNK, could be regulated by Rac1 [14]. To investigate if UTD2 affects MAPK signaling, we measured the phosphorylation levels of ERK1/2, p38 and JNK. Western blot analysis showed that the phosphorylation of ERK1/2 and p38 was reduced to 20% after 24 hours of treatment with 20 nM UTD2 in MCF-7 cells compared to the control cells (Fig. 6B and Fig. 6C). However, JNK phosphorylation remained unaffected (Fig. 6D). These results indicate differential effects of UTD2 on ERK1/2, p38 MAPK and JNK.

It was previously demonstrated that Rac1 could interact synergistically with Raf kinase to induce transformed foci in a focus formation assay [9]. Raf is an effector of Ras and mediate a signaling pathway that is crucial for both cell proliferation and malignant transformation. To examine the effect of UTD2 on the synergized transformation capacity of Rac1 and Raf, we performed a focus formation assay with NIH3T3 cells, which are sensitive to this assay. Table 2 showed that the number of tansformed foci induced by Rac1 plus Raf were dramatically decreased. Approximately 90% foci were inhibited when cells were exposed to UTD2 compared to untreated cells. UTD2 demonstrated a smaller inhibitory effect on focus formation caused by Raf alone (40% inhibition). UTD2 also abolished focus formation induced by activated Rac1. These findings revealed that UTD2 efficiently affects the synergistic signaling between Rac1 and Raf kinase, and also has a moderate effect on the signaling of Raf. These results further support the notion that UTD2 down-regulates Rac1 activity. These results also suggest that UTD2 can also affect the signaling of Raf kinase, consistent with a inhibitory effect of UTD2 on ERK1/2.

Matrix metalloproteins (MMPs) play a key role in tumor cell invasion, metastasis, and angiogenesis because they promote the degradation of the extra cellular matrix. Rac1 activity has also been connected to the regulation of MMP-2. To determine whether UTD2's inhibition of Rac1 activation results in the down-regulation of MMP-2, we performed a gelatin zymography assay to analyze the activity of MMP-2 in the UTD2-treated MDA-MB-231 cells. As shown in Fig. 7A, the activity of MMP-2 was clearly reduced after incubation with UTD2 compared to the cells grown in normal culture media. To investigate whether the reduced MMP-2 activity was caused by down-regulation at the gene transcriptional level, we performed a reverse transcription PCR experiment to analyze the mRNA expression of MMP-2 in the MDA-MB-231 cells both with and without the treatment of UTD2. For this experiment, β-actin was used as the control gene. The results indicated that UTD2 significantly down-regulated MMP-2 gene transcription (Fig. 7B). The inhibition of MMP-2 activity and expression was dose dependent.