Protumorigenic overexpression of stathmin/Op18 by gain-of-function mutation in p53 in human hepatocarcinogenesis


  • Potential conflict of interest: Nothing to report.


The microtubule (MT)-destabilizing protein stathmin/Op18 has previously been described to be negatively regulated by p53 and to be highly expressed in several tumor entities. However, little is known about its expression profile, functional or therapeutic relevance, and regulation in human hepatocarcinogenesis. Here we demonstrate cytoplasmic overexpression of stathmin in premalignant lesions (dysplastic nodules; DNs) and hepatocellular carcinomas (HCCs), which significantly correlated with tumor progression, proliferation, and activation of other protumorigenic factors (e.g., nuclear p53). Inhibition of stathmin expression by gene-specific short interfering RNA (siRNA) was associated with a significant reduction of MT-dependent cellular functions such as tumor cell viability, proliferation, migration, and increased apoptosis in HCC cells. Loss of stathmin expression increased responsiveness of tumor cells to the treatment with cytostatic drugs targeting MT-stability (paclitaxel, vinblastine) and to DNA cross-linking agents (cisplatin). Surprisingly, inducible expression of p53wt in p53-negative HCC cells as well as a reduction of p53wt by siRNA in p53wt-positive cells did not alter stathmin expression. However, stathmin was down-regulated after siRNA-based reduction of p53mut/Y220C and p53mut/R213Q expression in different tumor cell types. Conclusion: Our results demonstrate that overexpression of stathmin is an early protumorigenic event in human hepatocarcinogenesis, and its up-regulation can be mediated by gain-of-function mutations in p53. Thus, stathmin represents a potential therapeutic target, for example, by increasing responsiveness of tumor cells to treatment with chemotherapeutic agents after reduction of stathmin bioactivity. (HEPATOLOGY 2007.)

Hepatocellular carcinoma (HCC) is one of the most frequent malignant human cancers worldwide with increasing incidence in Western countries such as Europe and the United States. In most cases, a well-defined cause (mainly chronic viral hepatitis and chronic alcohol abuse) is associated with its development1; however, therapeutic options for HCC are sobering, and the number of specific tumor-relevant targets is limited. In an attempt to define novel therapeutic targets for HCC, we have previously performed expression profiling of human HCCs, revealing altered expression of various tumor-relevant genes involved in hepatocarcinogenesis.2 Among these, STMN1, which encodes for the cytosolic phosphoprotein stathmin (synonym: oncoprotein-18/Op18, metablastin) was significantly overexpressed in most cases. Stathmin belongs to a family of proteins that also contains stathmin-like 2 (STMN2, SCG10), stathmin-like 3 (STMN3, SCLIP), and stathmin-like 4 (STMN4, RB3), which are involved in the regulation of microtubule (MT) dynamics.3 Although all family members share a highly conserved stathmin/Op18 domain, differences in the number of phosphorylation sites and sequence variations may facilitate diverse biofunctionality in different cell types.4

Stathmin modulates MT dynamics through 2 different mechanisms. First, it sequesters α/β-tubulin-heterodimers, preventing MT polymerization. Second, it directly binds to MTs, subsequently promoting the so called MT-‘catastrophe’, which describes the transition from MT growth to MT shrinkage. In response to a number of cellular stimuli (e.g., leading to proliferation), stathmin is phosphorylated on up to 4 serine residues (Ser16, Ser25, Ser38, and Ser63) by various kinases (e.g., cyclin-dependent kinases), which inactivates stathmin and facilitates mitotic spindle assembly.4 Accordingly, the dephosphorylation of stathmin (e.g., by protein phosphatases type 1, 2A, and 2B; PP1, PP2A, PP2B) mediates proper disassembly of the spindle apparatus and the subsequent exit from mitosis.5 Besides its role in cell cycle progression, emerging evidence indicates that stathmin affects cell migration, because cell motility depends on MT–cytoskeleton reorganization in distinct subcellular regions.6

Overexpression of stathmin has been reported for several tumor entities such as leukemia and breast cancer.7, 8 Because the tumor suppressor gene TP53 negatively regulates stathmin expression in various cell types,9 and expression of p53mut is associated with elevated stathmin levels in breast cancer cells,10, 11 functional p53-deficiency (loss-of-function) was discussed to be the most probable reason for increased stathmin expression in cancer cells. Recently, immunohistochemical analyses of HCCs showed that elevated stathmin levels correlated with clinical parameters such as tumor size and decreased 5-year survival. Furthermore, stathmin expression correlated with the detection of p53mut, also suggesting loss-of-function–dependent mechanisms for its accumulation in HCC cells.12

In the current work, we show that elevated stathmin expression is an early event in hepatocarcinogenesis, its overexpression is essential for MT-dependent cellular processes, and that loss of stathmin expression increases the responsiveness of HCC cells for chemotherapeutic treatment. In contrast to commonly accepted concepts, stathmin is not regulated by p53wt in HCC cells; however, p53mutled to the overexpression of stathmin, which implicates that increased stathmin levels are at least in part mediated by p53 gain-of-function mutations.


DN, dysplastic nodule; HCC, hepatocellular carcinoma; MT, microtubule; PCR, polymerase chain reaction; siRNA, small interfering RNA; TMA, tissue microarray.

Materials and Methods

Cell Culture, Transfection, and Reagents.

Hep3B, Huh-7, HepG2, and U138-MG cells were maintained in DMEM, MEM, RPMI 1640 medium (PAA, Cölbe, Germany) supplemented with 10% fetal bovine serum and 1% penicillin/streptomycin at 37°C in a 5% CO2atmosphere. Hep3B cells expressing the temperature-sensitive mutant of p53 (4Bv) and control cells (BT-2E) were provided by M. Oren.13 All transfection assays were performed using Oligofectamine (Invitrogen, Karlsruhe, Germany) according to the manufacturer's protocol.

Aqueous stock solutions of chemotherapeutic drugs were obtained from the local pharmacy (University of Heidelberg, Germany) and applied in final concentrations for the indicated time. HepG2 cells were treated with the MDM-2 antagonist nutlin-3 (4 μM; Calbiochem/Merck Biosciences, Schwalbach/Ts., Germany) for 24 hours.

Sample Preparation, Short Interfering RNA Sequences, Primer Sequences, and Real-Time Polymerase Chain Reaction.

Total RNA isolation from 34 HCCs, reverse transcription, and total protein lysate extraction from HCCs were performed as previously described.2

The sequences for the chemical synthesis of short interfering RNAs (siRNA) were: stathmin siRNA #1 (sense: 5′-AGG-CAA-UAG-AAG-AGA-ACA-A-dtdt-3'), stathmin siRNA #2 (sense: 5′-AAG-AGA-AAC-UGA-CCC-ACA-A-dtdt-3′), p53 siRNA #1 (sense: 5′-UGU-UCC-GAG-AGC-UGA-AUG-A-dtdt-3′), p53 siRNA #2 (sense: 5′-AGA-CCU-AUG-GAA-ACU-ACU-U-dtdt-3′), and nonsense siRNA (sense: 5′-UUC-UCC-GAA-CGU-GUC-ACG-U-dtdt-3′).

The following primers and probes were used for real-time polymerase chain reaction (PCR) analyses: stathmin probe: 5′-Fam-CCC-AGT-GTG-GTT-TGC-ATT-GTC-TCC-Tamra-3′; stathmin primer forward: 5′-GTT-CTC-TGC-CCC-GTT-TCT-TG-3′; stathmin primer reverse: 5′-TAA-CAG-CTG-ACC-TGG-GCT-GA-3′. 18S-rRNA probe: 5′-Fam-AGC-AGG-CGC-GCA-AAT-TAC-CC-Tamra-3′; 18S-rRNA primer forward: 5′-AAA-CGG-CTA-CCA-CAT-CCA-AG-3′; 18S-rRNA primer reverse: 5′-CCT-CCA-ATG-GAT-CCT-CGT-TA-3′. The following cycling program was applied using the TaqMan Universal Master-Mix: 95°C for 10 minutes, followed by 40 cycles of 95°C for 15 seconds and 60°C for 60 seconds (Prism 7300; Applied Biosystems, Darmstadt, Germany).

Cell Viability, Proliferation-, Tubulin-, and Migration Assay.

Cell viability was analyzed using the MTT(3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl-tetra- zolium-bromide)-assay according to the manufacturer's instructions (EZ4U, Biomedica, Vienna, Austria) at indicated time points. Cell proliferation was examined using a bromodeoxyuridine enzyme-linked immunosorbent assay (Cell proliferation ELISA Biotrak, GE Healthcare/Amersham, Freiburg, Germany) according to the manufacturer's instructions. In addition, the percentage of surviving cells was assessed by staining with crystal violet.14

Semiquantitative analysis of polymerized/depolymerized tubulin was performed according to a modified protocol.15 Protein fractions were resuspended in Laemmli-buffer, quantified (40 μg; 280 nm), and used for immunoblotting.

Migratory activity was analyzed 3 days after transfection of siRNAs. Cells were treated with mitomycin-c (5 μg/ml) for 3 hours to repress proliferation. Subsequently, the cell monolayer was damaged with a ”scratch”, using a pipette tip. Cells were treated with HGF (hepatocyte growth factor) (10 ng/ml) and incubated for a further 18 hours. Scratches were digitally documented, and relative migratory activity was ascertained by calculating the cell-free areas.

Western Immunoblotting.

Total protein extracts were separated on a 10% to12% sodium dodecyl sulfate polyacrylamide gel electrophoresis (20 to 50 μg/lane), and electro-transferred to a PVDF (polyvinylidene fluoride) membrane. Anti-stathmin (1:1000; Cell Signaling Technology), anti-p53 (1:1000; Santa Cruz Biotechnology, Heidelberg, Germany), anti-p21 (1:1000; Pharmingen, Heidelberg, Germany), anti-survivin (1:200; R&D Systems, Wiesbaden, Germany), anti-MDM-2 (1:200; Novocastra, Berlin, Germany), and anti-actin antibody (1:5000; MP-Biomedicals, Aurora, OH) were diluted in TBST (Tris-buffered saline/Tween) (5% milk powder) and incubated at 4°C overnight. The appropriate secondary antibody was applied [1:2000; horseradish peroxidase anti-mouse and horseradish peroxidase anti-rabbit] at room temperature for 1 hour. Visualization was performed by enhanced chemiluminescence (Amersham Bioscience, Freiburg, Germany).

Flow Cytometry.

Analyses for DNA-index, cell cycle distribution, and apoptotic cell fractions were performed using a FACS-Calibur flow cytometer (Becton-Dickinson, Heidelberg, Germany) equipped with a 488 nm air-cooled argon laser with filter combinations for propidium iodide. After preparation according to Nicoletti with modifications,16 measurements were acquired in Fl-2 in logarithmic mode and calculated by setting gates over the first 3 decades to detect apoptotic cells. DNA-index and cell cycle analyses were acquired in Fl-3 in linear mode.

Tissue Micro-Array Analysis and Immunohistochemistry.

The tissue micro-array (TMA) contained 2 representative areas (diameter: 0.6 mm) of 15 ”normal” livers, 37 dysplastic nodules (DNs), 161 HCCs (Grading: 34 × G1, 96 × G2, 24 × G3, and 7 × G4), 17 cirrhosis, and 27 peritumorous tissue samples without significant pathological alterations. Cases were surgically resected at the University of Heidelberg and histologically classified according to established criteria by 2 experienced pathologists. Immunohistochemistry was performed as previously described.2 The primary antibodies were as follows: rabbit anti-human stathmin (1:50, Santa Cruz Biotechnology); rabbit anti-human Ki-67 (1:75; A0047, DAKO, Hamburg, Germany); and rabbit anti-human p53 (1:50, Santa Cruz Biotechnology).

For assessing the staining for stathmin, the product of cytoplasmic intensity and quantity was calculated based on the following scoring system: the intensity ranged from 0 = negative; 1 = low; 2 = medium to 3 = high, the quantity comprised 0 = no expression; 1 = positivity in less than 1%; 2 = positivity in 1% to 9%; 3 = positivity in 10% to 50%; 4 = positivity in more than 50% of tumor cells. The evaluation was performed independently by 2 experienced investigators.

Statistical Analysis and Software.

Data are presented as the mean ± standard deviation. The Spearman rank coefficient was used as a statistical measure of association. To account for multiple testing (TMA-data), the α-level was adjusted according to Bonferroni. The statistical comparison between 2 groups was accomplished with the nonparametric Mann-Whitney U-test, and the significance levels were defined as *P < 0.05 and **P < 0.01.

For densitometric quantification of signals, the AlphaEase FC software was used (V4.0; Alpha Innotech/Biozym, Hess. Oldendorf, Germany). For DNA and cell cycle histogram analysis, the Multicycle program (Phoenix Flow Systems, San Diego, CA), the ModFit LT-software (Flow cytometry modelling software, Varity Software House, Topsham, ME), and for calculation of apoptotic cell fractions Cell Quest software (Becton & Dickinson) was used. Area determination was performed using an analysis imaging system (Olympus, Hamburg, Germany).


Stathmin Overexpression in Human Hepatocarcinogenesis.

Initial expression profiling indicated a more than 2-fold induction of stathmin expression in most human HCCs (86%; 37/43) compared with representative nontumorous liver tissue.2 Using more sensitive semiquantitative real-time PCR analyses, we detected increased stathmin transcript levels in 38% of all analyzed HCCs as compared with normal liver samples (Fig. 1A). These elevated amounts of stathmin mRNA significantly correlated with tumor de-differentiation (r = 0.44; P = 0.009). Next we tested whether stathmin was detectable in total protein extracts of primary human HCCs and corresponding nontumorous liver samples. In 44% (4/9), a clear expression of stathmin was observed, whereas no signals were detected in appropriate liver samples or normal liver tissues of healthy donors (Fig. 1B).

Figure 1.

Elevated expression of stathmin/Op18 in human HCCs. (A) Using real-time PCR the relative transcript levels of stathmin were analyzed in primary human HCCs (gray bars). The horizontal line represents the threshold for the 2-fold induction of stathmin. The white bar indicates the expression of stathmin in a mixture of 3 normal livers and was used for mathematical calibration. For normalization, 18S-rRNA transcript levels were analyzed. The tumor grading of each sample is depicted under the appropriate column. (B) Exemplary stathmin protein levels in 3 indicated tumor (T) and peritumor (PT) sample pairs. Protein fractions of Hep3B cells served as positive and normal liver (n.l.) as negative controls.

To evaluate the expression of stathmin in the course of human hepatocarcinogenesis, we analyzed TMAs containing normal liver, premalignant lesions (DNs), carcinomas (grading: G1 to G4), cirrhotic liver tissues, and peritumorous liver samples. Whereas nearly no nuclear nor cytoplasmic staining was detected in normal hepatocytes (Fig. 2A), a clear cytoplasmic signal was observed in DNs (51%, 19/37; Fig. 2A) and HCCs (53%, 85/161; Fig. 2A). The remaining tumor samples (DNs: 49%; HCCs: 47%) did not exhibit elevated expression levels for stathmin. In the group of normal livers, DNs, and HCCs, stathmin expression significantly increased during tumor progression and de-differentiation (Spearman coefficient: r = 0.408; P < 0.001; Fig. 2B). No correlation with tumor etiology, especially chronic viral hepatitis (B and C) and clinical parameters such as sex, age, and tumor staging, were observed. In cirrhotic liver tissues and noncirrhotic peritumorous tissue samples, nearly no stathmin staining was detected. In the group of HCCs, the accumulation of stathmin correlated with increased expression of the proliferation marker Ki-67 (r = 0.558; P < 0.001), the nuclear staining of p53 (r = 0.518; P < 0.001), and the elevated expression of p27 (r = 0.301; P < 0.001).

Figure 2.

Overexpression of stathmin correlates with de-differentiation in human hepatocarcinogenesis. (A) TMA analyses revealed an increased cytoplasmic expression of stathmin in premalignant lesions (DNs) and carcinomas (HCCs; grading: G1-G4). No staining was observed in normal liver. Original magnification: ×200. (B) Boxplot illustration and statistical analyses demonstrate significantly elevated expression in premalignant lesions (DNs) and in the group of carcinomas as compared with normal liver. **P < 0.01

In summary, these results demonstrate that stathmin is increasingly expressed in approximately 50% of premalignant hepatic lesions and HCCs. Therefore, stathmin accumulation is likely to be an early pro-tumorigenic event in human hepatocarcinogenesis.

Inhibition of Stathmin Expression Leads to Reduced Viability and Migration of HCC Cells.

To analyze whether elevated stathmin expression is of functional relevance in hepatocarcinogenesis, we reduced stathmin expression using RNA interference in different HCC cell lines (Hep3 and HuH-7). No nonspecific interferon response (illustrated by STAT1-phosphorylation) was observed after application of different siRNAs (data not shown).

After transfection of 2 independent gene-specific siRNA sequences (si#1, si#2), a marked decrease of stathmin transcript levels was observed as compared with untreated cells, cationic carrier-treated cells, and nonsense-siRNA transfected cells (up to 80% reduction; Hep3B: Fig. 3A). Equally, a reduction of protein production below the level of detection was observed 3 days after stathmin-specific siRNA application (Hep3B: Fig. 3B; HuH-7: Supplementary Fig. 1A). The functional relevance of stathmin inhibition was analyzed using a tubulin polymerization assay.17, 18 A signal shift from the depolymerized (soluble) to the polymerized tubulin protein fraction was observed after siRNA-mediated reduction of stathmin as compared with nonsense siRNA-treated cells (Hep3B: Fig. 3C). These results were confirmed by densitometric signal quantification showing that the ratio of polymerized/depolymerized tubulin increased in cells with stathmin inhibition (#si1: 1.231 and #si2: 1.027) as compared with nonsense-siRNA transfected cells (ratio: 0.852). Paclitaxel- (ratio: 1.982; MT-stabilizing) and vinblastine-treated cells (ratio: 0.001; MT-destabilizing) served as control reactions.

Figure 3.

Inhibition of stathmin expression by gene-specific siRNAs in HCC cells. (A) Real-time PCR revealed the reduction of stathmin transcript levels by independent siRNAs (si#1, si#2) as compared with untreated cells (control), cationic carrier treated cells (oligofect.), and “nonsense”-siRNA (nons.) transfected Hep3B cells. (B) A complete reduction of stathmin expression on protein levels was observed compared with the appropriate controls. (C) Functional consequences of stathmin inhibition on MT integrity. The tubulin assay revealed a moderately increased accumulation of polymerized tubulin for both stathmin-specific siRNAs as compared with nonsense-siRNA transfected cells. Treatment with vinblastine (1 μM) and paclitaxel (1 μM) served as controls. S, soluble tubulin fraction; P, polymerized tubulin fraction.

Transient inhibition of stathmin expression for 3 days was associated with diminished tumor cell viability (up to 42% in Hep3B: Fig. 4A; HuH-7: Supplementary Fig. 1B), which was based on a significantly reduced number of proliferating cells (up to 44% in Hep3B: Fig. 4B; HuH-7: Supplementary Fig. 1C) and an increase in the number of apoptotic cells (up to 50% in Hep3B: Fig. 4C; HuH-7: Supplementary Fig. 1D). Cell cycle analyses showed a moderate increase of cells in the G2/M-phase after inhibition of stathmin expression (relative differences up to 18% in Hep3B: Fig. 4D; HuH-7: Supplementary Fig. 1E). Moreover, the number of S-phase cells was reduced after specific siRNA inhibition whereas the percentage of G0/G1-phase cells was unchanged (data not shown). Because altered MT-dynamics is of central relevance for cell motility,6 we further analyzed the effects of stathmin inhibition on cell migration using scratch assays. Indeed, HGF-dependent induction of tumor cell motility was significantly attenuated after transfection with stathmin-specific siRNAs as compared with nonsense siRNA-transfected cells (up to 46% in Hep3B: Fig. 4E; HuH-7: Supplementary Fig. 1F).

Figure 4.

Functional effects of reduced stathmin expression on HCC cells. SiRNA-mediated inhibition of stathmin (si#1, si#2) affects Hep3B (A) cell viability, (B) proliferation, (C) apoptosis, and (D) G2/M-phase cell cycle distribution as compared with nonsense-siRNA–treated controls. (E) Equally, HGF-induced tumor cell migration was significantly reduced after stathmin inhibition. Representative scratches are shown. Arrows highlight the ”wound”-margin of migrating cells. Bar: 200 μm. Mann-Whitney U-test: *P < 0.05; **P < 0.01.

In summary, these data show that increased stathmin expression in HCC cells affects a variety of MT-dependent cellular functions. These results do not exclude further stathmin effects that may have been compensated by additional factors with redundant biofunctionality, such as other stathmin family members.

Inhibition of Stathmin Expression Sensitizes HCC Cells to Chemotherapeutic Substances.

The multiple functional effects of reduced stathmin expression in HCC cells suggest that it may synergize with chemotherapeutic substances. We therefore tested whether stathmin-specific inhibition using siRNA increases the responsiveness of HCC cells for drugs affecting MT-stability such as paclitaxel (MT-stabilizing) and vinblastine (MT-destabilizing) or to substances with MT-independent molecular mechanisms such as cisplatin (cross-linking of DNA).

HCC cells treated with low doses of MT-interacting drugs (100 nM) showed decreased viability as compared with controls (paclitaxel: 30%, vinblastine: 23%), whereas the relative viability of cells treated with cisplatin was not reduced (Hep3B: Fig. 5; HuH-7: Supplementary Fig. 2). As shown above, the sole inhibition of stathmin (si#1 and si#2) also diminished tumor cell viability. Combined treatment with cytostatic substances and stathmin-specific siRNAs further reduced tumor cell viability in all cases; however, MT-affecting drugs led to the strongest biological consequences. Similar results were observed in experiments using alternative viability assays (crystal violet assay) as well as other drug concentrations (10 nM and 1 μM) in combination with siRNA-mediated stathmin inhibition (Supplementary Fig. 2).

Figure 5.

Reduction of stathmin bioactivity increased responsiveness to cytotoxic substances in HCC cells after 24 hours. Combination treatment of Hep3B cells with stathmin-specific siRNA and different chemotherapeutic drugs (paclitaxel, vinblastine, and cisplatin; each 100 nM) is associated with reduced tumor cell viability. Analyses were reproduced and performed using other drug concentrations (10 nM, 1 μM), showing similar results. Mann-Whitney U-test: **P < 0.01

In conclusion, the specific reduction of stathmin bioactivity increased the responsiveness of HCC cells for the treatment with different cytotoxic drugs. However, the combination of stathmin inhibition and the application of substances affecting MT stability in very low concentrations may be the most promising therapeutic approach.

Stathmin Expression Is Regulated by p53mut But Not by p53wt in HCC Cells.

p53wt has been described to be a negative transcriptional regulator of stathmin expression in different cell types.9 Furthermore, stathmin overexpression correlates with the mutational status of p53,12 and nuclear accumulation of p53 significantly correlates with stathmin staining in HCC TMAs (our study). Therefore, we also expected p53 loss-of-function–dependent regulation of stathmin expression in HCC cells.

Indeed, high expression levels of stathmin were observed in liver tumor cells harboring p53mut(HuH-7 cells)19 and p53del (Hep3B)20 but not in cells with p53wt (HepG2; Fig. 6A). This constellation was also present in other non-HCC cell lines (T47D, MCF7, U138-MG, HT29; Fig. 6A). Treatment of HepG2 cells with apoptosis-inducing doxorubicin caused intense p53wt induction and subsequent regulation of p53 target genes, but, surprisingly, the expression levels of stathmin remained unchanged (Fig. 6B). Furthermore, stathmin expression was not altered in Hep3B cells carrying a temperature-sensitive, transcriptionally active mutant of p53 after shifting to the permissive temperature; however, several known p53 target genes (p21CIP1, survivin, and MDM-2) were differentially regulated (Fig. 6C).13 To further substantiate these findings, we incubated HepG2 cells with a small molecule antagonist of MDM-2 (nutlin-3), which stabilized p53wt. Although a clear induction of p21CIP1 was observed after nutlin-3 treatment, no regulation of stathmin was detected (Fig. 6D). Moreover, efficient siRNA-mediated inhibition of p53wt in HepG2 cells (and the lung tumor cell line A549; data not shown) did not increase stathmin expression (Fig. 6E). Because changes in the short half-lives of stathmin transcripts accounts for differences in amount of stathmin protein, the steady stathmin expression cannot be explained by prolonged half-life of the protein.21 In summary, multiple experimental approaches clearly show that stathmin was not negatively regulated by p53wtin different HCC cells.

Figure 6.

p53wt-independent and p53mut-dependent regulation of stathmin. (A) Western blot analysis of p53 and stathmin expression levels in liver tumor (HuH-7, HepG2, and Hep3B), ductal breast cancer (T47D), breast adenocarcinoma (MCF7), glioblastoma (U138-MG), and colon adenocarcinoma (HT29) cell lines. Mutational status of p53 is indicated. (B) Stathmin and p53wt expression after doxorubicin (1 μM) treatment of HepG2 cells. p21CIP1 expression served as control for proper p53wt activation. (C) Expression of stathmin in 4Bv cells expressing the temperature sensitive p53 mutant (p53val135) in Hep3B cells. Transcriptionally active p53 is expressed after incubation at 32°C. Control cells (BT-2E) were analyzed under identical conditions. Equally, 4Bv and BT-2E cells were incubated at 37 °C (negative controls). Expression of survivin (negative target of p53wt) and MDM-2/p21CIP1 (positive targets of p53wt) served as controls for proper p53 activation. (D) Stathmin, p53wt, and p21CIP1expression in HepG2 cells treated with the MDM-2 inhibitor nutlin-3 (4 μM). (E) Expression of stathmin after siRNA-mediated inhibition of p53wt/mut in HepG2 (p53wt), HuH-7 (p53mut: codon 220), and U138-MG (p53mut: codon 213) cells. Protein extracts were subjected to Western blot analysis 72 hours after transfection with 2 different p53-specific siRNAs and 2 ”nonsense” siRNAs.

Alternatively, not p53wt-deficency (loss-of-function) but additional biofunctionality attributable to distinct mutations in the core domain of p53 may facilitate elevated stathmin levels in a subset of HCCs. We therefore tested whether expression of p53mut affects stathmin expression in tumor cells. SiRNA-based reduction of p53mut in HuH-7 cells (p53mut: codon 220, position 2, A:T to G:C)19 results in an clear decline in stathmin expression as compared with nonsense transfected cells (Fig. 6E). In addition, inhibition of another mutant p53 isoform in glioblastoma cells (p53mut: codon 213, position 2, G:C to A:T) equally caused decreased stathmin levels (Fig. 6E).

These results demonstrate that stathmin expression is not negatively regulated by p53wt in HCC cells but is upregulated by distinct mutations in p53. Thus, these data indicate that stathmin represents a novel effector for p53 gain-of-function mutations, not only in HCC but also in other tumor entities.


MT-cytoskeleton dynamics are essential for the adjustment of several cellular processes under physiological and pathological conditions. In the normal liver, stathmin/OP18 is only expressed by proliferating hepatocytes during liver regeneration after partial hepatectomy or hepatic ischemia–reperfusion injury, but not by resting hepatocytes.22–24 Because high stathmin levels are required for liver development, its expression is strictly regulated during organogenesis.23 In HCC, its neo-expression has been linked to several clinical parameters such as tumor size, vascular invasion, intrahepatic metastasis, differentiation, and decreased 5-year survival.12 Stathmin has also been identified to be highly expressed in hepatitis B virus–associated HCC samples using proteomic screening approaches.25 However, we were not able to confirm a link between stathmin expression and cause in our analyses (microarray and TMA analyses),2 which might be attributable to different methodical approaches and heterogenous patient populations. We found that stathmin is highly expressed in the group of carcinomas [approximately 50%: analyzed by quantitative PCR (38%), Western blotting (44%), and immunohistochemistry (53%)] as compared with normal, cirrhotic, and peritumorous liver tissues, which is supported by previously performed RT-PCR studies.12 Likewise, we observed elevated stathmin levels in premalignant lesions (approximately 50%, by immunohistochemistry), which suggests that its aberrant expression has to be classified as an early protumorigenic event.

As shown in this study for HCC cells, biofunctionality of stathmin has been associated with cellular effects directly depending on MT integrity and MT dynamics.6, 26 Equally, the modulation of apoptosis is indirectly linked to MT stability because tubulin polymerization affects phosphorylation of BCL (B-cell lymphoma) family members and subsequent subcellular translocation.27 However, inhibition of stathmin below the level of detection incompletely modulated all analyzed tubulin-dependent cellular effects. These data suggest that additional MT integrity modifiers such as other stathmin family members (STMN2-4) may functionally compensate the lack of stathmin expression after siRNA-mediated inhibition. Although the expression of these stathmin-related proteins has initially been published solely in neuronal cells,28 increasing data indicate further functional relevance in other tissue types. Indeed, STMN2/SCG10 has been described to be dysregulated in liver fibrosis,29 cirrhotic liver, and HCCs in a β-catenin/TCF (transcription factor)-dependent manner.30

Stathmin inhibition increases responsiveness to different cytotoxic substances in HCC cells, which may affect the treatment with chemotherapeutic drugs in HCC patients. Monotherapy and combination therapy using, for example, cisplatin, doxorubicin, and fluorouracil with or without interferon, have been studied; however, the prognosis of patients with inoperable tumors remains poor. Similarly, monotherapy with substances that stabilize MT, such as paclitaxel, did not show anti-cancer effects in HCC patients.31 Our in vitro data suggest that inhibition of stathmin may render HCC cells accessible to MT-interacting and DNA-damaging drugs. Here, the most obvious effects on tumor cell viability were observed after combining stathmin inhibition and MT-affecting substances. Therefore, in accordance to breast cancer cells,10 targeting of identical cellular mechanisms (MT integrity) seems to be the most promising approaches for the treatment of HCC. In this context, the specific reduction of stathmin biofunctionality, for example, by small molecules that inhibit stathmin bioactivity or siRNA-mediated approaches, may offer new therapeutic options for HCC patients in combination with low doses of cytostatics.32

An increasing number of data suggest that stathmin is a direct negative target of the tumor suppressor p53, which is frequently mutated in most cancer types.9, 11 A mechanistic link between the mutational status of p53 and the elevated expression of stathmin has been proposed, for example, in breast cancer.10, 11 Surprisingly, we were not able to modulate stathmin levels after reduction or induction of p53wt in different tumor cell lines (HCC: HepG2; lung tumor: A549). This is supported by independent experimental setups (nutlin-3 treatment, application of doxorubicin, induction of trancriptionally active p53) that suggest that ”simple” loss-of-function mutations disturbing the transcriptional activity of p53wt are not the relevant aberrations for stathmin expression in all tumor cells. In contrary, the reduction of mutated p53 (HuH-7: Y220C)19 in HCC cells is followed by a significant decrease of stathmin levels, which suggests that gain-of-function mutations directly or indirectly regulate stathmin expression. Indeed, mutations in codon 220 of TP53 are close to aflatoxin-induced codon 249 aberrations frequently found in HCCs. Because p53 gain-of-function mutations (R213Q) in glioblastoma cells also result in elevated stathmin levels, this protumorigenic mechanism is not restricted to hepatocarcinogenesis.

These findings are supported by different models explaining the possible molecular mechanisms for p53 gain-of-function mutations. First, p53 mutants may downregulate pro-apoptotic activities of the p53 family members p63 and p73 in vivo.33 Especially for p53mut/Y220C, a functionally relevant physical interaction with several isoforms of p63 and p73 has been shown,34 whereas p73, in turn, was described to control stathmin expression.35 Second, mutated p53 might directly influence target gene expression, possibly by recruiting transcription factors. Depending on the mutational status of p53, the transcriptional response to Sp1 and Ets1 is inhibited (p53wt) or elevated (p53mut) with subsequent opposite effects on target gene expression.36 Interestingly, some functional consequences are valid for DNA contact-defective p53 mutants (class I mutations) and conformational stability mutants (class II mutations), which suggests that irrespective of the type of missense mutation, gain-of-function activity may affect different p53mut-carrying cells.37 Future studies will have to investigate to which extent stathmin may confer the p53 gain-of-function properties in tumor cells.

However, several lines of evidence indicate that additional p53-independent mechanisms may also account for stathmin overexpression in HCCs: (1) Hep3B cells (p53del)20 showed high stathmin expression levels, and (2) some HCCs without nuclear accumulation of mutated p53 exhibited stathmin accumulation. Possible mechanistic explanations exist for a p53mut-uncoupled stathmin expression. For instance, stathmin is a direct target of E2F-transcription factor family members, which themselves are frequently dysregulated in HCCs.38 Similarly, genomic amplification of STMN1 (1p36) may account for a number of HCCs with stathmin overexpression.39

In summary, the data presented here demonstrate that protumorigenic stathmin is dysregulated early during hepatocarcinogenesis, and it represents a favorable target for combination therapeutic treatment using stathmin-specific (MT stability) and cytotoxic approaches. Moreover, stathmin expression was identified as a new target for p53 gain-of-function mutations in tumor cells.


The authors thank Prof. Dr. Moshe Oren (Weizmann Institute of Science, Israel) and Prof. Dr. Finzer (German Cancer Research Center, DKFZ, Germany) for providing the 4Bv/BT-2E cell lines and Michaela Bissinger for excellent technical assistance.