Cancer Research Meeting Calendar Genetics and Biology of Brain Cancer
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Nummela, P.
Right arrow Articles by Hölttä, E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Nummela, P.
Right arrow Articles by Hölttä, E.
[Cancer Research 66, 701-712, January 15, 2006]
© 2006 American Association for Cancer Research

Molecular Biology, Pathobiology, and Genetics

Thymosin ß4 Is a Determinant of the Transformed Phenotype and Invasiveness of S-Adenosylmethionine Decarboxylase–Transfected Fibroblasts

Pirjo Nummela1, Miao Yin1, Mari Kielosto1, Virna Leaner2, Michael J. Birrer2 and Erkki Hölttä1

1 Haartman Institute and Helsinki University Central Hospital, Department of Pathology, University of Helsinki, Helsinki, Finland and 2 Cell and Cancer Biology Department, Center for Cancer Research, National Cancer Institute, Bethesda, Maryland

Requests for reprints: Erkki Hölttä, Haartman Institute and Helsinki University Central Hospital, Department of Pathology, University of Helsinki, P.O. Box 21, FIN-00014 Helsinki, Finland. Phone: 358-9-19126516; Fax: 358-9-19126675; E-mail: erkki.holtta{at}helsinki.fi.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
S-adenosylmethionine decarboxylase (AdoMetDC) is a key enzyme in the synthesis of polyamines essential for cell growth and proliferation. Its overexpression induces the transformation of murine fibroblasts in both sense and antisense orientations, yielding highly invasive tumors in nude mice. These cell lines hence provide a good model to study cell invasion. Here, the gene expression profiles of these cells were compared with their normal counterpart by microarray analyses (Incyte Genomics, Palo Alto, CA, and Affymetrix, Santa Clara, CA). Up-regulation of the actin sequestering molecule thymosin ß4 was the most prominent change in both cell lines. Tetracycline-inducible expression of thymosin ß4 antisense RNA caused a partial reversal of the transformed phenotype. Further, reversal of transformation by dominant-negative mutant of c-Jun (TAM67) caused reduction in thymosin ß4 mRNA. Interestingly, a sponge toxin, latrunculin A, which inhibits the binding of thymosin ß4 to actin, was found to profoundly affect the morphology and proliferation of the AdoMetDC transformants and to block their invasion in three-dimensional Matrigel. Thus, thymosin ß4 is a determinant of AdoMetDC-induced transformed phenotype and invasiveness. Up-regulation of thymosin ß4 was also found in ras-transformed fibroblasts and metastatic human melanoma cells. These data encourage testing latrunculin A–like and other agents interfering with thymosin ß4 for treatment of thymosin ß4–overexpressing tumors with high invasive and metastatic potential. (Cancer Res 2006; 66(2): 701-12)


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
S-adenosylmethionine decarboxylase (AdoMetDC; EC 4.1.1.50) is one of the two key regulatory enzymes in the biosynthetic pathway of polyamines (putrescine, spermidine, and spermine), the other being ornithine decarboxylase (ODC; EC 4.1.1.17). ODC catalyzes the synthesis of putrescine, and AdoMetDC catalyzes the formation of decarboxylated S-adenosylmethionine, which is used as an aminopropyl donor in the conversion of putrescine to spermidine and spermine. Polyamines are essential for normal cell growth. If cells are depleted of their polyamine reservoirs, by using inhibitors of the polyamine biosynthetic enzymes or polyamine biosynthesis-deficient mutants, they cease to proliferate. The activities of both AdoMetDC and ODC are rapidly increased after growth factor stimulus, and these inductions together with the exceptionally fast intracellular turnover rates of these enzymes provide the cells with a means of rapidly regulating their polyamine levels (1–4). A tight regulation of polyamine biosynthesis is important for normal cell growth. Indeed, aberrant elevation of the activities and expression levels of ODC and AdoMetDC may be associated with the development of cancers, including prostatic, breast, colorectal, and hepatocellular tumors (2, 5–11). Furthermore, overexpression of both ODC and AdoMetDC, by themselves, has been shown to induce tumorigenic cell transformation of rodent cells (12–17). Murine NIH3T3 fibroblast cells transfected with AdoMetDC cDNA in either sense or antisense orientation were paradoxically both transformed and highly invasive, invading rapidly into the peritoneal cavity when inoculated s.c. into the flanks of nude mice (17). As these cell lines are even more invasive than the HT-1080 fibrosarcoma cells, MDA-MB-231 breast cancer cells, and c-Ha-rasVal12 oncogene-transformed fibroblasts (18),3 they provide a good model for studies of invasive growth.

The aim of the current work was to elucidate the gene expression alterations behind the transformed morphology and invasive behavior of the two AdoMetDC-transfected cell lines compared with their normal counterpart. In particular, we were interested in studying whether the apparently different signaling pathways in these sense and antisense cDNA-expressing cells could ultimately lead to common alterations in gene expression, typifying the transformed phenotype. The gene expression levels were compared by two different microarray systems: cDNA array (Incyte Genomics LifeArrays, Palo Alto, CA, 9,596 expressed elements) and oligonucleotide array (Affymetrix GeneChip, Santa Clara, CA, 12,654 elements). The former represents a competitive hybridization of two differently fluorescently labeled RNA samples to cDNA fragments (average length, 1,000 bp) immobilized onto glass slides, and the latter is based on two independent hybridizations of biotinylated cRNA samples to glass-immobilized oligonucleotides (25-mers) and fluorescent detection using antibody amplification.

Our studies revealed that the most marked gene expression change in both cell lines was a large increase in the mRNA level of the actin sequestering protein thymosin ß4. The functional significance of this increase was further studied by in vitro transfection assays using reverse tetracycline-inducible retroviral expression vector-driven thymosin ß4 antisense RNA expression and small interfering RNAs (siRNA) to thymosin ß4. In addition, the association of thymosin ß4 overexpression with transformation was also detected in AdoMetDC-sense cell lines carrying a tetracycline-inducible expression system of a transactivation domain deletion mutant of the oncogenic transcription factor c-Jun (TAM67), allowing reversible regulation of transformation. Further, the studies on thymosin ß4 expression were extended to c-Ha-ras oncogene-transformed fibroblasts and to normal human melanocytes and melanoma cell lines isolated from primary and metastatic tumors and clinical melanoma metastases. Finally and most interestingly, an actin-binding sponge toxin, latrunculin A, which inhibits the binding of thymosin ß4 to actin (19), was discovered to have a remarkable effect on the morphology, proliferation, and invasion of the AdoMetDC transformants, whereas normal fibroblasts were only slightly affected.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Cell culture. NIH3T3 cells transfected with a pLTRpoly-vector carrying human AdoMetDC cDNA in sense and antisense orientations (17) or a vector carrying c-Ha-rasVal12 oncogene (pGEJ6.6; ref. 20) have been described previously. The resulting transfectants with the empty vector (4N) and AdoMetDC cDNA in sense and antisense orientations (Amdc-s and Amdc-as, respectively) and the ras transformants (E4) were cultured in DMEM supplemented with antibiotics, G418, and 5% newborn calf serum or FCS (both from Invitrogen, Carlsbad, CA). To prevent the counterselection of cells expressing AdoMetDC-antisense mRNA at the levels sufficient to block the synthesis of essential polyamines, a low concentration (1 µmol/L) of spermidine was added to the growth medium of Amdc-as cells (17). Human primary melanocytes (42V; kindly provided by Dr. O. Saksela, University of Helsinki, Helsinki, Finland) were cultured as described previously (21). The melanoma cell lines, WM793 and WM239, established from a vertical growth phase melanoma primary tumor and melanoma metastasis, respectively, were kindly provided by Dr. M. Herlyn (Wistar Institute, Philadelphia, PA). They were both cultured in RPMI 1640 supplemented with 10% FCS and antibiotics.

Cell lines carrying a tetracycline-inducible expression system of the transactivation domain deletion mutant of c-Jun (pLRT-TAM67). The tetracycline-inducible TAM67 expression construct (pLRT-TAM67) was generated and transfected into Amdc-s cells identically to that described for the ODC and ras transformants (22). Several clones (transformed foci) of the stable transfectants were picked up by cylinder cloning, and after initial screening, the ones with the best induction were further subjected to single-cell cloning in 96 wells. For induction of TAM67 expression, 1 µg/mL doxycycline (Sigma, St. Louis, MO) was added the day after plating.

Microarray analysis with Incyte Genomics mouse GEM 2/Unigene 1 LifeArrays. Polyadenylated RNA was extracted from the exponentially growing cells by oligo(dT) cellulose affinity chromatography (Roche Diagnostics GmbH, Mannheim, Germany) as reported previously (23), and the possible genomic DNA carryover was destroyed by 1-hour RNase-free DNase I treatment (Roche Diagnostics GmbH). After that, the samples were further purified with Oligotex mRNA Mini kit (Qiagen, Hilden, Germany) according to the manufacturer's instructions. Polyadenylated RNA was then reverse transcribed with Cy3 and Cy5 end-labeled random 9-mers to generate fluorescent single-stranded cDNA probes, which were competitively hybridized to Incyte Genomics mouse GEM 2/Unigene 1 LifeArrays. The analyses were repeated twice. Mouse GEM 2/Unigene 1 contained 9,596 elements, corresponding to 9,307 total unique genes/clusters.

Microarray analysis with Affymetrix GeneChip murine genome arrays. For the analysis with Affymetrix microarrays, total RNA (>200 bases) was extracted from the exponentially growing cells with the RNeasy Midi kit and the RNase-Free DNase Set (Qiagen) according to the manufacturer's instructions. Biotinylated cRNA targets were then prepared from the samples and hybridized to MGU74A and MOE430 Set arrays according to the Affymetrix GeneChip Expression Analysis Technical Manual. The analyses were repeated twice. The total number of genes was 12,654 in MGU74A arrays and >30,000 in MOE430 Set arrays.

Thymosin ß4 vectors. To construct tetracycline-inducible retroviral thymosin ß4 antisense RNA expression vector, the 135-bp rat thymosin ß4 cDNA (coding region) cloned in pT7-7 vector (between NdeI and EcoRI restriction sites; a generous gift from Dr. D. Safer, University of Pennsylvania, Philadelphia, PA) was released by XbaI and EcoRI, subcloned into pBluescript, and finally inserted in reversed orientation into the inducible pLRT expression vector (24) using XhoI and NotI. The correctness of the cloned thymosin ß4 antisense construct was confirmed by DNA sequencing (ABI PRISM 3100 Genetic Analyzer).

Northern blot analysis. The thymosin ß4 mRNA levels of the normal 4N fibroblasts, Amdc-s and Amdc-as cells, ras-transformed cells, and tetracycline-inducible TAM67-expressing derivatives of Amdc-s as well as those of the 42V melanocytes, WM793 and WM239 melanoma cells, and clinical melanoma metastases were confirmed by Northern blot analysis. The polyadenylated RNA samples (10 µg; fibroblasts) or total RNAs (30 µg; melanocytes and melanoma cells) were resolved in agarose gels and transferred to Hybond-N nylon membrane (Amersham Pharmacia Biotech, Buckinghamshire, United Kingdom) as described previously (25). The hybridizations were done according to Clontech's (Palo Alto, CA) "Hybridization with cDNA Probes" manual using the ExpressHyb hybridization solution (Clontech, Palo Alto, CA). The single-stranded thymosin ß4 antisense RNA probe was generated from the linearized pBlue-thymosin ß4-vector (thymosin ß4 flanked by T3 and T7 RNA polymerase promoters) by using MAXIscript T7/T3 In vitro Transcription kit (Ambion, Inc., Austin, TX) and [{alpha}-32P]UTP (Amersham Pharmacia Biotech). Hybridizations with human ß-actin cDNA control probe (Clontech), labeled with Redivue [32P]dCTP using Rediprime II Random Prime Labeling System (both from Amersham), were used to check the ß-actin expression levels. Hybridization signals were detected and quantified by using Fuji's imaging plates (Fuji, Tokyo, Japan) and MacBAS 2.5 software.

Transfection experiments. The transformed Amdc-s and Amdc-as cells were transfected with the pLRT-thymosin ß4-antisense vector (2 µg/5 cm diameter plate) using LipofectAMINE Plus reagent (Invitrogen) according to the instructions. To obtain stable cell lines, blasticidin S selection (5 µg/mL; Invitrogen) was started 2 days after transfection and continued for 2 weeks, after which the concentration was diminished to 1.25 µg/mL. The expression of thymosin ß4 antisense RNA was induced by the addition of 1 µg/mL doxycycline immediately after transfection, the next day or later, to study the effects of the timing of induction. As a control, doxycycline was also added to normal Amdc-s and Amdc-as cells (transfected or not with empty thymosin ß4 vector) with no effects. To maximize the effect of doxycycline induction, the cells were grown with Tet System–approved fetal bovine serum (BD Biosciences, Palo Alto, CA). The pools of transfected cells were further subjected to cylinder cloning to study individual clones.

Transfection with siRNA oligonucleotides. The siRNA oligonucleotides to thymosin ß4 were synthesized by Dharmacon Research, Inc. (Lafayette, CO). The two siRNA oligonucleotides tested were targeted to the 5'-AACAAGAGAAGCAAGCUGGCC-3' and 5'-AAGAGAAGCAAGCUGGCGAAU-3' sequences of the mouse thymosin ß4 mRNA coding region. The negative control siRNA was targeted to a scrambled sequence 5'-AAAGGAGAGGAUCGCACGCAC-3'. The Amdc-s and Amdc-as cells (1 x 105) were transfected with siRNAs (50-200 nmol/L) in six-well plates using Oligofectamine and LipofectAMINE 2000 (both from Invitrogen) or Fugene 6 (Roche Diagnostics GmbH) reagents following the manufacturer's protocols. The cells were observed daily, changing medium with the siRNAs every 3 days. Total RNAs were extracted from the cells with RNeasy Mini columns after 3, 6, and 9 days of culture.

Reverse transcription-PCR analysis. The expression of thymosin ß4 and thymosin ß10 mRNAs and thymosin ß4 antisense RNA were analyzed by reverse transcription-PCR (RT-PCR). Total RNA was extracted from the exponentially growing cells and reverse transcribed to cDNA using avian myeloblastosis virus reverse transcriptase (20 units; Finnzymes, Espoo, Finland) and oligo(dT) primer (12.5 pmol; Genset, Paris, France). After the reverse transcription reaction, 1:10 (v/v) aliquots of the reverse-transcribed RNAs were subjected to PCR amplification. The primers for PCR were from Proligo (Paris, France) unless otherwise specified. The primers were 5'-CCTCATCCTCCTCGTCCTTA-3' (forward) and 5'-TGATCCAACCTCTTTGCATC-3' (reverse) for thymosin ß4 mRNA and 5'-GTAAGAAAATGGCAGACAAGCC-3' and 5'-AGTCCGATTAGTGGAGGG-3' for thymosin ß10 mRNA. For thymosin ß4 antisense RNA, the forward primer was specific for the pLRT-vector (5'-CCTACAGGTGGGGTCTTTCA-3'; Genset) and the reverse was specific for the thymosin ß4 antisense RNA (5'-TCGCCAGCTTGCTTCTCTTG-3'). As a control, the samples were also amplified using glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA primers 5'-GGGTGTGAACCACGAGAAAT-3' and 5'-GGTCCTCAGTGTAGCCCAAG-3'. A negative control contained all constituents but not the template cDNA (H2O or RNA instead). The PCR reactions, optimized to be in the linear range, consisted of initial denaturation (94°C for 2 minutes) followed by 23 cycles (thymosin ß4-antisense) or 30 cycles (thymosin ß4, thymosin ß10, and GAPDH) of 94°C for 30 seconds, 50°C, 56°C, 59°C, or 62°C (thymosin ß10, thymosin ß4, GAPDH, and thymosin ß4-antisense, respectively) for 45 seconds, and 72°C for 40 seconds and a final elongation at 72°C for 7 minutes. Products were resolved in 2% agarose gel and the gel was stained with SYBR Green I nucleic acid gel stain (Molecular Probes, Leiden, the Netherlands) to allow visualization of the bands under UV light. The gels were documented with Gel Doc 2000 and Quantity One 4.2.3 software (Bio-Rad, Hercules, CA).

Immunofluorescence staining. Cells were grown on glass coverslips, fixed with 3.5% paraformaldehyde, permeabilized with 0.1% Triton X-100, and stained with rabbit anti-thymosin ß4 antibody (amino acids 1-43; Biodesign, Saco, ME) and TRITC-labeled anti-rabbit antibody (DAKO, Glostrup, Denmark).

Effect of latrunculin A on cell morphology and growth. To test the effect of latrunculin A on the cell morphology, latrunculin A (100 and 350 ng/mL; Molecular Probes) was added to 4N, Amdc-s and Amdc-as cells 2 days after plating. The morphology of the cells was followed at 5- to 20-minute intervals during the first hours and then checked daily. In a longer follow-up, medium with latrunculin A was changed twice weekly and the detached cells were harvested by centrifugation. To test the ability of the cells to restore the normal morphology, the medium with latrunculin A was replaced with medium without the toxin at different time points. To see the effect of latrunculin A on the growth of the normal and AdoMetDC transfected cells, 10,000 to 15,000 cells were plated onto 12-well plates 2 days before addition of latrunculin A (50, 100, and 350 ng/mL) and the total number of cells was counted 0, 1, 2, and 3 days after the addition with a Coulter particle counter (Beckman Coulter, Fullerton, CA). All measurements were made in triplicates. Growth assays were repeated four times with similar results.

Three-dimensional collagen gel migration/invasion assay. Three-dimensional collagen lattices were prepared in 24-well plates. Rat tail collagen type 1 solution (4 mg/mL; Upstate Biotechnology, Lake Placid, NY) was mixed in a ratio of 1:1 with 2x DMEM containing 10% FCS. About 400 µL of this collagen solution (supplemented or not with 1 µg/mL doxycycline) were added to the wells and allowed to polymerize at 37°C for 30 minutes. Then, 20,000 Amdc-s or Amdc-s-pLRT-thymosin ß4-as cells (pretreated or not with 1 µg/mL doxycycline for 36 hours) were added in 100 µL DMEM containing 5% FCS. The cells were allowed to attach for 1 hour at 37°C, and the excess medium was removed. Next, an upper collagen gel (300 µL) was cast. After polymerization at 37°C for 30 minutes, DMEM (500 µL) containing 5% FCS, supplemented or not with 1 µg/mL doxycycline, was added on the top. The growth medium was changed after every 3 days.

Three-dimensional Matrigel invasion assay. Invasiveness of Amdc-s and Amdc-as cell lines in growth factor–reduced Matrigel (BD Biosciences) was inspected without and with latrunculin A (25, 50, and 100 ng/mL) essentially as reported earlier (18), but with latrunculin A included in the Matrigel and the upper medium. The toxin was not present in the medium during adhesion of the cells. The growth patterns of the cells were observed under a microscope daily for at least 1 week and the growth medium was changed every third day. To see the effect of latrunculin A on the already formed invasive colonies, a final concentration of 100 ng/mL was added to the culture. After 4 days, the Matrigel was then solubilized by 4.5-hour incubation on ice and dilution with ice cold medium, and the cells were recovered by centrifugation and plated onto six-well plates without the toxin. The Matrigel assays were repeated thrice with similar results. As a control, 4N cells were treated in the same way.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
AdoMetDC-sense– and AdoMetDC-antisense–expressing cells show similar gene expression patterns. Microarray analysis of the AdoMetDC-sense– and AdoMetDC-antisense–expressing transformed fibroblasts (Amdc-s and Amdc-as, respectively) compared with wild type cells (4N) showed strikingly similar changes in their gene expression profiles. We used two microarray systems, the Incyte Genomics Mouse GEM2/Unigene 1 cDNA array and the Affymetrix MGU74A oligonucleotide array. Both systems identified thymosin ß4 and cellular retinoic acid–binding protein I as genes showing the highest increase in expression (Tables 1 and 2). Several other changes common to both systems were detected, including increased expression of cathepsin L, integrins ß7 and {alpha}6, chondroitin sulfate proteoglycan 2, proliferin, decorin, kit ligand, protocadherin 7, and osteopontin and decreased expression of serine protease (hypothetical 43-kDa protein), microfibrillar-associated protein 5, thrombospondin 1, procollagen type I {alpha}1, Ly-6 alloantigen, cysteine-rich protein 61, serum-inducible kinase, and phosphatidic acid phosphatase 2a (marked in bold in Tables 1 and 2). The level of AdoMetDC mRNA expression served as a good control and was ~4 to 10 times up-regulated in the Amdc-s cells compared with the 4N control cell line (in the Affymetrix arrays containing the gene; data not shown). Our microarray analysis thus identified a defined set of genes with differential expression associated with the transformed phenotype of AdoMetDC-transfected cells. It is noteworthy that thymosin ß4 and cellular retinoic acid–binding protein I were also the two most up-regulated genes in the global gene expression analyses by Affymetrix MOE430 Set arrays (data not shown). The functional significance of thymosin ß4 up-regulation was further explored in this study, because cellular retinoic acid–binding protein I was not up-regulated in c-Ha-ras oncogene-transformed cells (data not shown), implicating that it is not a universal change in invasive fibroblasts.


View this table:
[in this window]
[in a new window]
  		Table 1. The most marked gene expression changes by Incyte Genomics Mouse GEM 2/Unigene 1 LifeArrays

 

View this table:
[in this window]
[in a new window]
  		Table 2. The most marked gene expression changes by Affymetrix MGU74A arrays

 
Validation of thymosin ß4 overexpression. The overexpression of thymosin ß4 mRNA in AdoMetDC transfectants was verified by Northern blot (Fig. 1) and RT-PCR (data not shown) analyses. A substantial increase in thymosin ß4 mRNA expression was observed in both Amdc-s and Amdc-as cells compared with the normal 4N control (Fig. 1A). A similar increase in thymosin ß4 mRNA was also detected in ras-transformed fibroblasts (Fig. 1A). In addition, the levels of thymosin ß4 protein were found to be increased in the transformed cells by immunofluorescence staining (Fig. 1D).


Figure 1
View larger version (45K):
[in this window]
[in a new window]
  		Figure 1. Thymosin ß4 (Tß4) overexpression in AdoMetDC-transfected fibroblasts, c-Ha-ras oncogene-transformed fibroblasts, and human melanoma cells. Northern blot analysis of (A) NIH3T3 cells transfected with empty pLTRpoly-vector (4N), vector carrying AdoMetDC cDNA in sense (Amdc-s) and antisense (Amdc-as) orientations, and vector carrying ras oncogene (Ras); (B) Amdc-s cells transfected with inducible construct of a transactivation domain deletion mutant of c-Jun (Amdc-s-pLRT-TAM67) before (–) and after (+) induction with doxycycline (dox) for 6 days; and (C) normal human melanocytes (42V), and melanoma cell cultures from primary (WM793) and metastatic (WM239) tumors, and clinical metastasis (M). ß-Actin was used as a loading control. D, immunofluorescence analysis of thymosin ß4 protein in normal 4N (a), Amdc-s (b), Amdc-as (c), and ras-transformed (d) fibroblasts.

 
We have shown previously that the c-Jun/activator protein-1 (AP-1) dominant-negative mutant TAM67 can reverse the ODC- and ras-induced transformation of fibroblasts (22). Here, the same was found to be true for the AdoMetDC-induced transformation. To determine whether this occurred via the regulation of thymosin ß4, we used tetracycline-inducible Amdc-s-pLRT-TAM67 cells and found that TAM67 inhibited the expression of thymosin ß4 mRNA (Fig. 1B). We further studied the expression of thymosin ß4 in normal human melanocytes and malignant melanoma cells. The expression of thymosin ß4 was low in normal melanocytes (42V) and early vertical growth phase melanoma cells (WM793) but high in metastasized melanoma cells (WM239) and clinical metastases (Fig. 1C). These findings are in line with a recent study (26) and suggest that thymosin ß4 overexpression is associated with a more aggressive transformed cell type.

Thymosin ß4 antisense expression affects the transformed phenotype. To determine if thymosin ß4 expression is required for the transformed phenotype of AdoMetDC-overexpressing fibroblasts, cells with tetracycline-inducible antisense thymosin ß4 were prepared by stable transfection. In RT-PCR analyses, thymosin ß4 antisense RNA-specific primers detected a 0.4-kb product that was amplified from total RNA of thymosin ß4 antisense-transfected Amdc-s and Amdc-as cells (Fig. 2A, shown for Amdc-s cells). The band was detectable in uninduced transfectants albeit at a much lower level than that observed in doxycycline-treated cells. This was probably due to leakiness of the pLRT-thymosin ß4-antisense construct. Control cells transfected with the empty pLRT vector had no corresponding band. These results confirm the specific expression of thymosin ß4 antisense message. Inducible inhibition of thymosin ß4 expression in these cells resulted in a morphology change to more flattened cells (Fig. 2B). The degree of flattened morphology varied, however, among the cells even in isolated clones. The effect was most extensive immediately after transfection and induction. At times long after transfection, the inducibility was weakened. Similar results were observed with siRNA to thymosin ß4 (data not shown). This is likely due to some compensating mechanisms taking place in the thymosin ß4 antisense RNA-expressing cells. The possibility that the thymosin ß10 isoform (expressed at the same levels in normal and AdoMetDC-transfected cells) would show a compensatory increase was excluded by RT-PCR (data not shown). We also studied the cellular behavior of the transfectants in three-dimensional collagen gel assays, mimicking more closely the physiologic growth conditions than the two-dimensional culture on plastic. As shown in Fig. 2C, the expression of thymosin ß4 antisense RNA inhibited dispersion/migration of the Amdc-s cells in the three-dimensional system, resulting in cells growing adhered to each other as multicellular aggregates.


Figure 2
View larger version (103K):
[in this window]
[in a new window]
  		Figure 2. Reversal of the transformed phenotype of the Amdc-s and Amdc-as cells by expression of thymosin ß4 antisense RNA. A, RT-PCR analysis of thymosin ß4 antisense (Tß4-as) RNA expression in Amdc-s cells transfected with empty pLRT vector and pLRT-thymosin ß4-antisense expression construct (pLRT-Tß4-as). The cells were grown in the presence or absence of doxycycline, RNA extracted, and subjected to RT-PCR analysis as described in Materials and Methods. Slight leakage of the construct can be seen in the uninduced cells. Similar results were obtained with Amdc-as cells (data not shown). B, morphologic changes of Amdc-s (a and b) and Amdc-as (c and d) cells stably transfected with inducible thymosin ß4 antisense RNA construct. The cells were cultured in the absence (a and c) and presence (b and d) of doxycycline (1 µg/mL) and photographed by phase-contrast microscopy. Magnification, x200. C, growth of thymosin ß4 antisense RNA-expressing Amdc-s cells in three-dimensional collagen gel. The cells were grown without (a) and with (b) doxycycline for 10 days (see Materials and Methods) and photographed by phase-contrast microscopy. Magnification, x80.

 
Latrunculin A, an inhibitor of thymosin ß4 binding to actin, affects the morphology of Amdc-s and Amdc-as cells. To further study the potential functional significance of thymosin ß4 overexpression, we used latrunculin A, a sponge toxin shown to interfere with thymosin ß4 binding to actin (19). Interestingly, the effect of latrunculin A on cell morphology was very different between the normal fibroblasts (Fig. 3A and B) and the AdoMetDC transformants (Fig. 3C-H, shown for Amdc-s). Latrunculin A treatment at a concentration of 350 ng/mL for 35 minutes had little effect on the morphology of normal 4N cells (Fig. 3A and B). Amdc-s and Amdc-as cells started to round up from the cell body region in <20 minutes after addition of only 100 ng/mL (Fig. 3C and D), progressively becoming ball-and-stick model–looking cells with a round center and thin protrusions. The effect of latrunculin A was found to be cell density dependent, with the more sparse cultures being more sensitive to the toxin. With 350 ng/mL of the drug, the morphology changes occurred sooner and the protrusions further contracted, resulting in rounding of the cells that started to detach in <1 hour and become swollen (Fig. 3E and F). When the cell growth medium was replaced with toxin-free medium at this point, the attached cells recovered as fast as they initially had started to round up, resulting in nearly full recovery of the morphology after an overnight culture (data not shown). One day after the addition of 350 ng/mL latrunculin A, almost all the cells had rounded and detached. Replating of the detached cells in fresh medium without toxin resulted in attachment in <1 hour and the cells started to divide. Interestingly, the swollen cells seemed to rapidly give rise to groups of cells (Fig. 3G), implicating arrested cell division, cytokinesis, as the cause of the swelling. When the detached cells were collected after 2-day culture in 350 ng/mL latrunculin A, they attached somewhat more inefficiently and did not recover fully. Many of the attached cells had multiple nuclei (data not shown). After 5 days of exposure to latrunculin A, the recovery of the cells was poor and nearly all the attached cells had multiple nuclei (Fig. 3H). However, the detached cells were not apoptotic, as no activation/cleavage of caspase-3 was detected by immunoblotting (data not shown).


Figure 3
View larger version (144K):
[in this window]
[in a new window]
  		Figure 3. Effect of latrunculin A on the morphology of the normal fibroblasts and AdoMetDC-transformed cells. Cultures of normal fibroblasts (4N) without (A) and with 350 ng/mL latrunculin A for 35 minutes (B); Amdc-s cells without (C) and with 100 ng/mL for 35 minutes (D) and 350 ng/mL for 35 minutes (E) and 2 hours (F); Amdc-s cells grown 1 day with 350 ng/mL latrunculin A and then plated into toxin-free medium and photographed 1 hour thereafter (G); and Amdc-s cells grown 5 days with 350 ng/mL latrunculin A, replated into toxin-free medium, and photographed 8 hours after plating (H). The morphologic changes of Amdc-as cells were similar to those of Amdc-s (data not shown). Magnification, x200.

 
Latrunculin A inhibits the growth of AdoMetDC transformants. The effect of latrunculin A on cell growth was analyzed at different time points after addition of 0, 50, 100, and 350 ng/mL latrunculin A. Latrunculin A at these concentrations had little effect on the growth of normal 4N cells (Fig. 4A). The growth of both Amdc-s and Amdc-as cells was effectively inhibited with as little as 50 ng/mL latrunculin A (Fig. 4B, shown for Amdc-s cells). In addition, the motility of the cells, as determined by wound-scratch assay, was found to be differentially inhibited by latrunculin A (data not shown). As with the effect on morphology, the sensitivity of the cells to the latrunculin A–induced growth inhibition seemed to be cell density dependent, with less effect seen with more dense cultures (data not shown). The untreated 4N and Amdc-s cells had similar growth rates (Fig. 4A and B), indicating that thymosin ß4 overexpression itself does not affect the growth rate of fibroblasts.


Figure 4
View larger version (16K):
[in this window]
[in a new window]
  		Figure 4. Effect of latrunculin A on the growth of normal 4N cells (A) and Amdc-s transformants (B). The cells were seeded into 12-well plates in triplicates and exposed to latrunculin A after 2 days of culture. The cell numbers were counted at the day of toxin addition and 1 and 3 days thereafter using a Coulter counter. The growth curves of Amdc-as were similar to Amdc-s (data not shown). Points, mean growth curves; bars, SD.

 
Latrunculin A blocks invasion of Amdc-s and Amdc-as cells in three-dimensional Matrigel. When inoculated into nude mice, the thymosin ß4–overexpressing Amdc-s and Amdc-as cell lines gave rise to aggressively growing tumors, which rapidly invaded into the peritoneal cavity (17). Similarly, they effectively invaded the surrounding basement membrane matrix in a three-dimensional Matrigel invasion assay (Fig. 5A, F, and G, shown for Amdc-s), correlating with the in vivo invasiveness in nude mice (18, 22). When latrunculin A was added to the Matrigel matrix and the growth medium, the invasion was inhibited in a dose-dependent manner (Fig. 5A-C). At 100 ng/mL of the toxin, the inhibition was almost complete (>95%) with both Amdc-s and Amdc-as cells (Fig. 5C). The normal 4N control cells, which are not invasive in Matrigel, were not affected by latrunculin A (Fig. 5D and E). When 100 ng/mL concentration of latrunculin A was added to the already formed invasive colonies, not only the invasion was inhibited but also the colonies started to shrink (Fig. 5F-H). When Matrigel was solubilized 4 days after the latrunculin A addition, the cells were harvested and plated, they appeared enlarged and aberrantly shaped with multiple nuclei (Fig. 5I and J). Their growth was also significantly impaired. Taken together, these results suggest that thymosin ß4 functional activity is essential for the transformed phenotype and migration/invasion of AdoMetDC-expressing cells.


Figure 5
View larger version (119K):
[in this window]
[in a new window]
  		Figure 5. Effect of latrunculin A on invasiveness of AdoMetDC transformants in three-dimensional Matrigel. Amdc-s cells cultured without (A) and with 50 ng/mL latrunculin A for 2 days (B) and with 100 ng/mL for 4 days (C); normal fibroblasts (4N) cultured without (D) and with 100 ng/mL latrunculin A for 4 days (E); and Amdc-s cells cultured without latrunculin A for 3 days (F) and 6 days (G) and without latrunculin A for 3 days and then with 100 ng/mL final concentration of latrunculin A for 3 days (H). Magnifications, x40. Similar results were obtained with Amdc-as cells (data not shown). Amdc-as cells recovered from Matrigel after culturing without latrunculin A for 8 days (see Materials and Methods) and photographed after 1 day of growth in tissue culture dish (I); Amdc-as cells recovered from Matrigel after culturing without latrunculin A for 4 days and then with 100 ng/mL latrunculin A for 4 days and photographed as above (J). Magnifications, x200.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Overexpression of AdoMetDC cDNA in both sense and antisense orientations results in morphologically transformed fibroblasts and induces highly invasive tumors in nude mice (17). In this study, we compared the gene expression profiles of these two cell lines with their normal counterpart by microarray analyses and found only a surprisingly small set of common changes potentially relevant for transformation/invasion. They included genes regulating cytoskeletal structure, cell adhesion, and membrane trafficking: thymosin ß4 and cytohesin-binding protein (increased); cell surface receptors/adhesion molecules: integrins {alpha}6 and ß7 and protocadherin 7 (increased); extracellular matrix (ECM) components/modifiers: decorin, chondroitin sulfate proteoglycan 2, and hyaluronan synthase 2 (increased) and procollagen type I {alpha}1, microfibrillar-associated protein 5, and thrombospondin 1 (decreased); and ECM-degrading enzymes: cathepsin L and type IV collagenase (increased). Whether some of these diverse genes could be functionally coupled with some of the molecules acting as central linkers/mediators of the signaling in and out between the ECM, cell surface, and cytoskeleton remains a highly interesting object of further studies. For example, the cytohesin-binding protein might be one such integrator molecule, because in leukocytes cytohesin-1, a guanine nucleotide exchange activator of ADP ribosylation factor GTPases, is known to bind to integrin ß2 and actin cytoskeleton and to regulate cell adhesion and membrane trafficking (27–29). Moreover, a couple of unknown genes [expressed sequence tags (EST), with increased or decreased expression] and as yet functionally uncharacterized proteins, such as serine protease (hypothetical 43-kDa protein) and LR8 (showing the largest decreases), were detected and will require further elucidation of their potential significance for transformation/invasion. Here, the functional significance of thymosin ß4 was studied in more detail, as it showed the most striking increase in expression in the Amdc-s and Amdc-as cells and it was also highly increased in the c-Ha-rasVal12 oncogene-transformed fibroblasts and other cells (see below). The second most increased gene in the AdoMetDC-transformed cells, cellular retinoic acid–binding protein I, was proven not to be increased in the ras transformants and thus does not seem to represent a common transformational target in fibroblasts.

Thymosin ß4, first characterized in 1981 (30), is the most abundant member of the highly conserved family of ß-thymosins, strongly polar 5-kDa polypeptides. Thymosin ß4 was originally isolated from calf thymus and thought to be a secreted hormone but was later shown to be present in several mammalian tissues and cell lines (31–34). Thymosin ß4 seems to be a major actin sequestering molecule, which specifically binds monomeric actin (G-actin) forming a 1:1 complex (35, 36). Due to its high concentration in the cell, thymosin ß4 sequesters a large pool of G-actin, from which the actin monomers can then easily be released for the polymerization of actin filaments (F-actin). This dynamic equilibrium of G-actin and F-actin is important for the rapid reorganization of the cytoskeleton, a process essential for cell proliferation, migration, and differentiation. Thymosin ß4 seems to have numerous roles in the cells, both intracellular and extracellular. It has been reported to be involved in adhesion and spreading of fibroblasts (37, 38), differentiation of endothelial cells (39, 40), directional migration of endothelial cells and keratinocytes (41–44), angiogenesis (39–42, 45, 46), wound healing (42, 43, 47, 48), hair follicle growth (44), and apoptosis (49, 50), and to possess anti-inflammatory properties (47, 51). Interestingly, elevated thymosin ß4 expression levels have been observed in various malignant cell lines and tumors (52–55) and its expression seems to be related to increased tumorigenicity and metastatic potential (26, 38, 46, 50, 56, 57).

In our study, besides the AdoMetDC transfectants, we also identified thymosin ß4 overexpression in ras-transformed NIH3T3 fibroblasts as well as in metastatic human melanoma cell lines and melanomas. In contrast, a cell line transformed by ODC, which is also tumorigenic in nude mice (13), did not show a significant increase in thymosin ß4 expression or high invasiveness (only local invasion). In our study, the increased expression of thymosin ß4 correlated with the invasive capability of the cells as well as with the degree of morphologic transformation and disintegration of actin filaments (17).3 Recent work has also correlated increased thymosin ß4 expression with increased cell growth (50, 57), but this seems not to be universal (this study; refs. 38, 46).

When the increased thymosin ß4 expression of the AdoMetDC transfectants was inhibited, at least partly, by induced expression of thymosin ß4 antisense RNA, the cell morphology became more flattened, resembling that of normal fibroblasts. In addition, the migration/invasion of the cells in three-dimensional collagen was inhibited, with the blocking of thymosin ß4 possibly affecting the cytoskeletal regulation of cell-cell and cell-matrix interactions. The reversal of the transformed phenotype was, however, not complete in all the asynchronously growing cells probably because the inducible antisense message did not fully block the increased expression of thymosin ß4 in all of them or due to some compensating mechanisms. Further, an apparent counterselection resulted in impairment of the inducibility of antisense RNA expression during prolonged culturing. The mechanism(s) of counterselection, noted also by others (58), remains to be elucidated. It also seems that the rapidity of the changes in thymosin ß4 may have an effect on the outcome. Indeed, although microinjection of thymosin ß4 has been found to cause disintegration of the actin filaments (59, 60), a slower (and probably less intense) increase in thymosin ß4 by cDNA transfection may result in adaptation of the cells with formation of even stronger actin filaments through a concomitant increase in actin (37). The same may conversely be reflected in the antisense/siRNA transfection experiments and inhibitor (e.g., latrunculin A) studies (see below).

Our data with a dominant-negative mutant of c-Jun/AP-1 (TAM67) showed the morphologic reversal of Amdc-s and diminished thymosin ß4 expression levels. These results suggest a connection between thymosin ß4 and c-Jun. Our kinetic analyses show that the decrease in thymosin ß4 mRNA following TAM67 induction is a relatively slow process, suggesting that thymosin ß4 may not be a direct target of c-Jun (data not shown).

As thymosin ß4 up-regulation seems to be associated with cell transformation and high invasiveness, it could serve as a target for antitumor strategies. Recently, latrunculin A, an actin sequestering macrolide isolated from a Red Sea sponge, has been reported to inhibit the binding of thymosin ß4 to actin (19). In this study, we show that the addition of latrunculin A to thymosin ß4–overexpressing Amdc-s and Amdc-as cells caused a rapid and dramatic change in the morphology of the cells, whereas minor effects were observed in normal fibroblasts. Previously, Spector et al. (61) have similarly reported that latrunculin A rapidly changes the morphology of mouse fibroblasts and neuroblastomas. A 10-fold difference existed in the sensitivity of the studied cell lines, with neuroblastomas being more sensitive. In the light of our results, the differential sensitivity could be explained by differences in the thymosin ß4 levels. Indeed, Hall et al. (62, 63) found thymosin ß4 protein to be abundant in human and rat neuroblastoma cell lines. However, it is possible that latrunculin A has also other sites of action than thymosin ß4 binding of actin, making it important to develop more specific derivatives of the toxin (19). Following latrunculin A addition, the morphology of thymosin ß4–overexpressing AdoMetDC transfectants changed to a ball-and-stick appearance. The outermost long protrusions of the cells were much more resistant to the drug action. Similarly, in the work of Spector et al. (61), the long and thick processes characteristic of morphologically differentiated neuroblastoma cells were also minimally affected even after hours in high toxin concentration. Further, two populations of actin filaments, sensitive stress fibers at the center and more resistant cortical filaments at the cell periphery, have been seen following thymosin ß4 overexpression-induced actin disassembly (59, 60, 64).

In our study, latrunculin A seemed to prevent cytokinesis. Importantly, latrunculin A was found to fully inhibit the invasive growth of Amdc-s and Amdc-as cells in three-dimensional Matrigel invasion assays. Even when the toxin was added to the already formed invasive colonies, the invasion was blocked and the colonies further regressed. In earlier studies, thymosin ß4 has been suggested to facilitate tumor angiogenesis and metastasis, but the mechanisms involved have remained unclear. Recently, Philp et al. (65) found that the actin-binding site of thymosin ß4 is important for angiogenesis. Our data with latrunculin A, inhibiting the binding of thymosin ß4 to actin, show that the actin-binding site could be similarly important for invasion. As both invasion and angiogenesis are essential for cancer spreading and metastasis, thymosin ß4 might be an excellent target to interfere with these two processes. These data should encourage the development and testing of specific drugs for thymosin ß4, interfering either with its up-regulation or actin sequestering function, for the treatment of thymosin ß4–overexpressing tumors with high metastatic potential.


   Acknowledgments
 
Grant support: University of Helsinki, Finnish Cancer Organizations, Academy of Finland, Sigrid Juselius Foundation, Helsinki University Central Hospital Research Funds, and Ida Montin Foundation (P. Nummela and M. Kielosto). P. Nummela and M. Kielosto are predoctoral fellows of the Helsinki Biomedical Graduate School.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

We thank Drs. Olga Aprevikov and Ari Ora for assistance with the Affymetrix microarray and the siRNA experiments, respectively, and Taina Nieminen and Merja Haukka for technical assistance.


   Footnotes
 
3 Unpublished data. Back

Received 7/11/05. Revised 9/23/05. Accepted 11/ 3/05.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Tabor CW, Tabor H. Polyamines. Annu Rev Biochem 1984;53:749–90.[CrossRef][Medline]
  2. Pegg AE. Polyamine metabolism and its importance in neoplastic growth and a target for chemotherapy. Cancer Res 1988;48:759–74.[Abstract/Free Full Text]
  3. Heby O, Persson L. Molecular genetics of polyamine synthesis in eukaryotic cells. Trends Biochem Sci 1990;15:153–8.[CrossRef][Medline]
  4. Cohen SS. A guide to the polyamines. New York: Oxford University Press; 1998. p. 595.
  5. Romano M, Santacroce MA, Bonelli P, Cecco L, Cerra M. Differences in polyamine metabolism between carcinomatous and uninvolved human breast tissues. Int J Biol Markers 1986;1:77–80.[Medline]
  6. Porter CW, Herrera-Ornelas L, Pera P, Petrelli NF, Mittelman A. Polyamine biosynthetic activity in normal and neoplastic human colorectal tissues. Cancer 1987;60:1275–81.[CrossRef][Medline]
  7. Radford DM, Nakai H, Eddy RL, et al. Two chromosomal locations for human ornithine decarboxylase gene sequences and elevated expression in colorectal neoplasia. Cancer Res 1990;50:6146–53.[Abstract/Free Full Text]
  8. Gan FY, Gesell MS, Alousi M, Luk GD. Analysis of ODC and c-myc gene expression in hepatocellular carcinoma by in situ hybridization and immunohistochemistry. J Histochem Cytochem 1993;41:1185–96.[Abstract]
  9. Tamori A, Nishiguchi S, Kuroki T, et al. Relationship of ornithine decarboxylase activity and histological findings in human hepatocellular carcinoma. Hepatology 1994;20:1179–86.[CrossRef][Medline]
  10. Mohan RR, Challa A, Gupta S, et al. Overexpression of ornithine decarboxylase in prostate cancer and prostatic fluid in humans. Clin Cancer Res 1999;5:143–7.[Abstract/Free Full Text]
  11. Bettuzzi S, Davalli P, Astancolle S, et al. Tumor progression is accompanied by significant changes in the levels of expression of polyamine metabolism regulatory genes and clusterin (sulfated glycoprotein 2) in human prostate cancer specimens. Cancer Res 2000;60:28–34.[Abstract/Free Full Text]
  12. Auvinen M, Paasinen A, Andersson LC, Holtta E. Ornithine decarboxylase activity is critical for cell transformation. Nature 1992;360:355–8.[CrossRef][Medline]
  13. Auvinen M, Laine A, Paasinen-Sohns A, et al. Human ornithine decarboxylase-overproducing NIH3T3 cells induce rapidly growing, highly vascularized tumors in nude mice. Cancer Res 1997;57:3016–25.[Abstract/Free Full Text]
  14. Moshier JA, Dosescu J, Skunca M, Luk GD. Transformation of NIH/3T3 cells by ornithine decarboxylase overexpression. Cancer Res 1993;53:2618–22.[Abstract/Free Full Text]
  15. Shantz LM, Pegg AE. Overproduction of ornithine decarboxylase caused by relief of translational repression is associated with neoplastic transformation. Cancer Res 1994;54:2313–6.[Abstract/Free Full Text]
  16. Clifford A, Morgan D, Yuspa SH, Soler AP, Gilmour S. Role of ornithine decarboxylase in epidermal tumorigenesis. Cancer Res 1995;55:1680–6.[Abstract/Free Full Text]
  17. Paasinen-Sohns A, Kielosto M, Kaariainen E, et al. c-jun activation-dependent tumorigenic transformation induced paradoxically by overexpression or block of S-adenosylmethionine decarboxylase. J Cell Biol 2000;151:801–10.[Abstract/Free Full Text]
  18. Ravanko K, Jarvinen K, Helin J, Kalkkinen N, Holtta E. Cysteine cathepsins are central contributors of invasion by cultured adenosylmethionine decarboxylase-transformed rodent fibroblasts. Cancer Res 2004;64:8831–8.[Abstract/Free Full Text]
  19. Yarmola EG, Somasundaram T, Boring TA, Spector I, Bubb MR. Actin-latrunculin A structure and function. differential modulation of actin-binding protein function by latrunculin A. J Biol Chem 2000;275:28120–7.[Abstract/Free Full Text]
  20. Holtta E, Sistonen L, Alitalo K. The mechanisms of ornithine decarboxylase deregulation in c-Ha-ras oncogene-transformed NIH 3T3 cells. J Biol Chem 1988;263:4500–7.[Abstract/Free Full Text]
  21. Alanko T, Rosenberg M, Saksela O. FGF expression allows nevus cells to survive in three-dimensional collagen gel under conditions that induce apoptosis in normal human melanocytes. J Invest Dermatol 1999;113:111–6.[CrossRef][Medline]
  22. Kielosto M, Nummela P, Katainen R, Leaner V, Birrer MJ, Holtta E. Reversible regulation of the transformed phenotype of ornithine decarboxylase- and ras-overexpressing cells by dominant-negative mutants of c-jun. Cancer Res 2004;64:3772–9.[Abstract/Free Full Text]
  23. Schwab M, Alitalo K, Varmus HE, Bishop JM, George D. A cellular oncogene (c-Ki-ras) is amplified, overexpressed, and located within karyotypic abnormalities in mouse adrenocortical tumour cells. Nature 1983;303:497–501.[CrossRef][Medline]
  24. Watsuji T, Okamoto Y, Emi N, Katsuoka Y, Hagiwara M. Controlled gene expression with a reverse tetracycline-regulated retroviral vector (RTRV) system. Biochem Biophys Res Commun 1997;234:769–73.[CrossRef][Medline]
  25. Paasinen-Sohns A, Holtta E. Cells transformed by ODC, c-Ha-ras and v-src exhibit MAP kinase/Erk-independent constitutive phosphorylation of sos, raf and c-jun activation domain, and reduced PDGF receptor expression. Oncogene 1997;15:1953–66.[CrossRef][Medline]
  26. Clark EA, Golub TR, Lander ES, Hynes RO. Genomic analysis of metastasis reveals an essential role for RhoC. Nature 2000;406:532–5.[CrossRef][Medline]
  27. Klarlund JK, Guilherme A, Holik JJ, Virbasius JV, Chawla A, Czech MP. Signaling by phosphoinositide-3,4,5-trisphosphate through proteins containing pleckstrin and Sec7 homology domains. Science 1997;275:1927–30.[Abstract/Free Full Text]
  28. Dierks H, Kolanus J, Kolanus W. Actin cytoskeletal association of cytohesin-1 is regulated by specific phosphorylation of its carboxyl-terminal polybasic domain. J Biol Chem 2001;276:37472–81.[Abstract/Free Full Text]
  29. Moss J, Vaughan M. Cytohesin-1 in 2001. Arch Biochem Biophys 2002;397:156–61.[CrossRef][Medline]
  30. Low TL, Hu SK, Goldstein AL. Complete amino acid sequence of bovine thymosin ß4: a thymic hormone that induces terminal deoxynucleotidyl transferase activity in thymocyte populations. Proc Natl Acad Sci U S A 1981;78:1162–6.[Abstract/Free Full Text]
  31. Hannappel E, Xu GJ, Morgan J, Hempstead J, Horecker BL. Thymosin ß4: a ubiquitous peptide in rat and mouse tissues. Proc Natl Acad Sci U S A 1982;79:2172–5.[Abstract/Free Full Text]
  32. Hannappel E, Leibold W. Biosynthesis rates and content of thymosin ß4 in cell lines. Arch Biochem Biophys 1985;240:236–41.[CrossRef][Medline]
  33. Goodall GJ, Morgan JI, Horecker BL. Thymosin ß4 in cultured mammalian cell lines. Arch Biochem Biophys 1983;221:598–601.[CrossRef][Medline]
  34. Erickson-Viitanen S, Ruggieri S, Natalini P, Horecker BL. Distribution of thymosin ß4 in vertebrate classes. Arch Biochem Biophys 1983;221:570–6.[CrossRef][Medline]
  35. Safer D, Golla R, Nachmias VT. Isolation of a 5-kilodalton actin-sequestering peptide from human blood platelets. Proc Natl Acad Sci U S A 1990;87:2536–40.[Abstract/Free Full Text]
  36. Safer D, Elzinga M, Nachmias VT. Thymosin ß4 and fx, an actin-sequestering peptide, are indistinguishable. J Biol Chem 1991;266:4029–32.[Abstract/Free Full Text]
  37. Golla R, Philp N, Safer D, et al. Co-ordinate regulation of the cytoskeleton in 3T3 cells overexpressing thymosin-ß4. Cell Motil Cytoskeleton 1997;38:187–200.[CrossRef][Medline]
  38. Kobayashi T, Okada F, Fujii N, et al. Thymosin-ß4 regulates motility and metastasis of malignant mouse fibrosarcoma cells. Am J Pathol 2002;160:869–82.[Abstract/Free Full Text]
  39. Grant DS, Kinsella JL, Kibbey MC, et al. Matrigel induces thymosin ß4 gene in differentiating endothelial cells. J Cell Sci 1995;108:3685–94.[Abstract]
  40. Grant DS, Rose W, Yaen C, Goldstein A, Martinez J, Kleinman H. Thymosin ß4 enhances endothelial cell differentiation and angiogenesis. Angiogenesis 1999;3:125–35.[CrossRef][Medline]
  41. Malinda KM, Goldstein AL, Kleinman HK. Thymosin ß4 stimulates directional migration of human umbilical vein endothelial cells. FASEB J 1997;11:474–81.[Abstract]
  42. Malinda KM, Sidhu GS, Mani H, et al. Thymosin ß4 accelerates wound healing. J Invest Dermatol 1999;113:364–8.[CrossRef][Medline]
  43. Sosne G, Chan CC, Thai K, et al. Thymosin ß4 promotes corneal wound healing and modulates inflammatory mediators in vivo. Exp Eye Res 2001;72:605–8.[CrossRef][Medline]
  44. Philp D, Nguyen M, Scheremeta B, et al. Thymosin ß4 increases hair growth by activation of hair follicle stem cells. FASEB J 2004;18:385–7.[Abstract/Free Full Text]
  45. Koutrafouri V, Leondiadis L, Avgoustakis K, et al. Effect of thymosin peptides on the chick chorioallantoic membrane angiogenesis model. Biochim Biophys Acta 2001;1568:60–6.[Medline]
  46. Cha HJ, Jeong MJ, Kleinman HK. Role of thymosin ß4 in tumor metastasis and angiogenesis. J Natl Cancer Inst 2003;95:1674–80.[Abstract/Free Full Text]
  47. Sosne G, Szliter EA, Barrett R, Kernacki KA, Kleinman H, Hazlett LD. Thymosin ß4 promotes corneal wound healing and decreases inflammation in vivo following alkali injury. Exp Eye Res 2002;74:293–9.[CrossRef][Medline]
  48. Philp D, Goldstein AL, Kleinman HK. Thymosin ß4 promotes angiogenesis, wound healing, and hair follicle development. Mech Ageing Dev 2004;125:113–5.[CrossRef][Medline]
  49. Niu M, Nachmias VT. Increased resistance to apoptosis in cells overexpressing thymosin ß four: a role for focal adhesion kinase pp125FAK. Cell Commun Adhes 2000;7:311–20.
  50. Wang WS, Chen PM, Hsiao HL, Wang HS, Liang WY, Su Y. Overexpression of the thymosin ß-4 gene is associated with increased invasion of SW480 colon carcinoma cells and the distant metastasis of human colorectal carcinoma. Oncogene 2004;23:6666–71.[CrossRef][Medline]
  51. Young JD, Lawrence AJ, MacLean AG, et al. Thymosin ß4 sulfoxide is an anti-inflammatory agent generated by monocytes in the presence of glucocorticoids. Nat Med 1999;5:1424–7.[CrossRef][Medline]
  52. Hall AK. Differential expression of thymosin genes in human tumors and in the developing human kidney. Int J Cancer 1991;48:672–7.[Medline]
  53. Yamamoto T, Gotoh M, Kitajima M, Hirohashi S. Thymosin ß-4 expression is correlated with metastatic capacity of colorectal carcinomas. Biochem Biophys Res Commun 1993;193:706–10.[CrossRef][Medline]
  54. Santelli G, Califano D, Chiappetta G, et al. Thymosin ß-10 gene overexpression is a general event in human carcinogenesis. Am J Pathol 1999;155:799–804.[Abstract/Free Full Text]
  55. Xie D, Jauch A, Miller CW, Bartram CR, Koeffler HP. Discovery of over-expressed genes and genetic alterations in breast cancer cells using a combination of suppression subtractive hybridization, multiplex FISH and comparative genomic hybridization. Int J Oncol 2002;21:499–507.[Medline]
  56. Ji P, Diederichs S, Wang W, et al. MALAT-1, a novel noncoding RNA, and thymosin ß4 predict metastasis and survival in early-stage non-small cell lung cancer. Oncogene 2003;22:8031–41.[CrossRef][Medline]
  57. Wang WS, Chen PM, Hsiao HL, Ju SY, Su Y. Overexpression of the thymosin ß-4 gene is associated with malignant progression of SW480 colon cancer cells. Oncogene 2003;22:3297–306.[CrossRef][Medline]
  58. Gibbons DL, Shashikant C, Hayday AC. A comparative analysis of RNA targeting strategies in the thymosin ß4 gene. J Mol Biol 2004;342:1069–76.[CrossRef][Medline]
  59. Sanders MC, Goldstein AL, Wang YL. Thymosin ß4 (fx peptide) is a potent regulator of actin polymerization in living cells. Proc Natl Acad Sci U S A 1992;89:4678–82.[Abstract/Free Full Text]
  60. Sanger JM, Golla R, Safer D, et al. Increasing intracellular concentrations of thymosin ß4 in PtK2 cells: effects on stress fibers, cytokinesis, and cell spreading. Cell Motil Cytoskeleton 1995;31:307–22.[CrossRef][Medline]
  61. Spector I, Shochet NR, Kashman Y, Groweiss A. Latrunculins: novel marine toxins that disrupt microfilament organization in cultured cells. Science 1983;219:493–5.[Abstract/Free Full Text]
  62. Hall AK, Hempstead J, Morgan JI. Thymosin ß10 levels in developing human brain and its regulation by retinoic acid in the HTB-10 neuroblastoma. Brain Res Mol Brain Res 1990;8:129–35.[Medline]
  63. Hall AK, Chen SC, Hempstead JL, Morgan JI. Retinoic acid regulates thymosin ß10 levels in rat neuroblastoma cells. J Neurochem 1991;56:462–8.[CrossRef][Medline]
  64. Yu FX, Lin SC, Morrison-Bogorad M, Yin HL. Effects of thymosin ß4 and thymosin ß10 on actin structures in living cells. Cell Motil Cytoskeleton 1994;27:13–25.[CrossRef][Medline]
  65. Philp D, Huff T, Gho YS, Hannappel E, Kleinman HK. The actin binding site on thymosin ß4 promotes angiogenesis. FASEB J 2003;17:2103–5.[Abstract/Free Full Text]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Nummela, P.
Right arrow Articles by Hölttä, E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Nummela, P.
Right arrow Articles by Hölttä, E.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online