Skip Navigation


Carcinogenesis Advance Access originally published online on April 21, 2007
Carcinogenesis 2007 28(10):2172-2183; doi:10.1093/carcin/bgm096
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
28/10/2172    most recent
bgm096v1
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 Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Suryo Rahmanto, Y.
Right arrow Articles by Richardson, D.R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Suryo Rahmanto, Y.
Right arrow Articles by Richardson, D.R.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2007. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

Identification of distinct changes in gene expression after modulation of melanoma tumor antigen p97 (melanotransferrin) in multiple models in vitro and in vivo

Y. Suryo Rahmanto, L.L. Dunn and D.R. Richardson*

Iron Metabolism and Chelation Program, Department of Pathology, Blackburn Building D06, University of Sydney, Sydney, New South Wales, 2006 Australia

* To whom correspondence should be addressed. Tel: +61 2 9036 6548; Fax: +61 2 9036 6549; Email: d.richardson{at}pathology.usyd.edu.au


   Abstract
Top
 Abstract
Introduction
 Materials and methods
 Results
 Discussion
 References
 
Melanoma tumor antigen p97 or melanotransferrin (MTf) is an iron (Fe)-binding protein with high homology to serum transferrin. MTf is expressed at very low levels in normal tissues and in high amounts in melanoma cells although its function remains elusive. To understand the function of MTf, we utilized whole-genome microarray analysis to examine the gene expression profile of five models after modulating MTf expression. These models included two new stably transfected MTf hyper-expression models (SK-N-MC neuroepithelioma and LMTK– fibroblasts) and one cell type (SK-Mel-28 melanoma) where MTf was down-regulated by post-transcriptional gene silencing. These findings were compared with alterations in gene expression identified using the MTf–/– mice. In addition, the changes identified from the microarray data were also assessed in a new model of MTf down-regulation in SK-Mel-2 melanoma cells. In the cell line models, MTf hyper-expression led to increased proliferation, whereas MTf down-regulation resulted in decreased proliferation. Across all five models of MTf down- and up-regulation, we identified three genes modulated by MTf. These included ATP-binding cassette subfamily B member 5, whose change in expression mirrored MTf down- or up-regulation. In addition, thiamine triphosphatase and transcription factor 4 were inversely expressed relative to MTf levels across all five models. The products of these three genes are involved in membrane transport, thiamine phosphorylation and proliferation/survival, respectively. This study identifies novel molecular targets directly or indirectly regulated by MTf and the potential pathways involved in its function, including modulation of proliferation.

Abbreviations: Abcb5, ATP-binding cassette subfamily B member 5; Apod, apolipoprotein D; Mef2a, myocyte enhancer factor 2a; MTf, melanotransferrin; PTGS, post-transcriptional gene silencing; RT-PCR, reverse transcription–polymerase chain reaction; Tcf4, transcription factor 4; Tf, transferrin; TfR1, transferrin receptor 1; Thtpa, thiamine triphosphatase


   Introduction
Top
 Abstract
 Introduction
Materials and methods
 Results
 Discussion
 References
 
Melanoma tumor antigen p97 or melanotransferrin (MTf) is highly expressed on melanoma cells, other tumors and fetal tissues (1–4). Although initially thought to be absent or detected at low levels in normal tissues (2), MTf has since been demonstrated to be differentially expressed across a broad range of normal human and mouse tissues, with relatively high levels being detected in epithelial surfaces (5,6). The MTf protein is typically membrane bound by a glycosylphosphatidylinositol anchor, although a soluble form of MTf has also been reported in urine, cerebrospinal fluid, saliva and serum at very low levels (7). However, the highest expression of MTf is found in melanoma cells (2,3), suggesting that it may have an important role in this cell type.

As a member of the transferrin (Tf) family of proteins, human MTf shares many common properties with these proteins, including (i) a 37–39% sequence homology with human serum Tf, human lactoferrin and chicken ovoTf; (ii) conserved disulfide bonds; (iii) chromosomal co-localization with the transferrin receptor 1 (TfR1) gene and (iv) a high-affinity N-terminal Fe-binding site which is identical to that identified in serum Tf that can bind one atom of Fe (2,3,8,9). These characteristics and the high expression of MTf by melanoma cells led to the hypothesis that MTf may assist rapidly proliferating tumor cells with their increased Fe requirements (3).

However, the recent phenotypic characterization of the MTf knockout (MTf–/–) mouse indicated that MTf does not play an essential role in Fe metabolism (10,11). Other studies using melanoma cells in vitro suggested that MTf does not have a significant role in cellular Fe uptake (5,10,12). Furthermore, studies in cell culture demonstrated that MTf expression was not regulated by intracellular Fe levels and that the pattern of MTf expression was different to that of other molecules involved in Fe uptake, such as serum Tf and the TfR1 (5,6). Taken together, these findings demonstrated that MTf does not have an essential role in Fe metabolism.

It has been suggested that MTf may function in diverse physiological and pathological processes such as Fe transport across the blood brain barrier, eosinophil differentiation, chondrogenesis, Alzheimer's disease and arthritis (7,13–16). However, examination of the MTf–/– mice provided no definitive evidence for the role of MTf in these processes (10,11). More recently, it has been suggested that membrane-bound and soluble MTf may modulate angiogenesis (17), cell migration and plasminogen activation (18–20). However, these preliminary findings require further investigation and definitive evidence for the functional role of MTf remains elusive.

To assist in determining the role of MTf, we previously developed two models of MTf down-regulation, namely the MTf–/– mice and SK-Mel-28 melanoma cells with post-transcriptional gene silencing (PTGS) of MTf expression (10,11). In these latter studies, MTf down-regulation in both models was demonstrated to result in up-regulation of myocyte enhancer factor 2a (Mef2a) and transcription factor 4 (Tcf4) (10). Interestingly, the Mef2a and Tcf4 transcription factors are associated with cell proliferation and survival (21–24). These data, together with the fact that SK-Mel-28 melanoma cells with depressed MTf expression have decreased proliferation, migration and tumorigenesis in vivo (10), suggest the existence of unrecognized pathways that could lead to a better understanding of MTf function.

To further elucidate the molecular pathways involved in MTf function, in this investigation we used genome-wide microarrays to examine patterns of differential gene expression in four models of MTf down- and up-regulation. These included the MTf–/– mice and MTf down-regulated SK-Mel-28 human melanoma cells described previously (10,11), and two new models of stable MTf hyper-expression in LMTK– murine fibroblasts and SK-N-MC human neuroepithelioma cells. Changes in gene expression identified in these four systems were also assessed in an additional new model where MTf was stably down-regulated by PTGS in SK-Mel-2 melanoma cells.

In the current study, across all five models of MTf down- and up-regulation, we identified changes in expression of ATP-binding cassette subfamily B member 5 (Abcb5), thiamine triphosphatase (Thtpa) and Tcf4. Hence, this investigation has identified novel molecules regulated directly or indirectly by MTf expression that have roles in membrane transport, energy metabolism, cell proliferation and survival, suggesting potential pathways through which MTf may play a functional role.


   Materials and methods
Top
 Abstract
 Introduction
 Materials and methods
Results
 Discussion
 References
 
Construction of vectors
The vectors transfected to stably down-regulate MTf expression in the SK-Mel-28 melanoma cell line by PTGS were described previously (10). The same procedure was used to down-regulate MTf expression in the SK-Mel-2 melanoma cell line that naturally expresses high MTf levels (Figure 1A) (6). Stable hyper-expression constructs of the mouse and human MTf cDNAs were generated using the pCI-neo (Promega, Madison, WI) and pCMV-Script® (Stratagene, La Jolla, CA) expression vectors, respectively. The inserts were sequenced and confirmed against mouse and human MTf (Genbank accession no.: NM_013900 and NM_005929, respectively) using the public The National Center for Biotechnology Information database.


Figure 1
View larger version (58K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1. The effect of MTf down-regulation or hyper-expression on proliferation across a variety of cell lines. (A) RT-PCR and western blot analysis of MTf in the human SK-Mel-2 melanoma cell line stably transfected with anti-MTf siRNA and the human SK-N-MC neuroepithelioma and murine LMTK– fibroblast cell lines stably transfected with MTf hyper-expression vectors. SK-Mel-2 melanoma cells were stably transfected with a vector that contained a cloned transgene that encoded a short 19-mer double-stranded hairpin siRNA to specifically target human MTf mRNA. As a control, a second vector containing a non-specific (scrambled) hairpin siRNA was similarly transfected into cells. Clones were selected in G418 and are referred to as ‘A3’ and ‘C1’ for the MTf down-regulated cells and ‘scrambled’ for the control cells. RT-PCR and western analysis showed marked down-regulation of human MTf mRNA and MTf protein in the A3 and C1 clones compared with the scrambled and SK-Mel-2 control cells. Conversely, the SK-N-MC neuroepithelioma and LMTK– fibroblast hyper-expression cell lines showed marked up-regulation of MTf mRNA and MTf protein compared with vector controls. The RT-PCR and western blot analyses are representative of at least three separate experiments and these changes in gene expression were quantitatively assessed via densitometric analysis. (B) The effect of MTf expression on cellular proliferation. Cellular proliferation of SK-Mel-2 melanoma clones was assessed by viable cell counts. The A3 and C1 MTf down-regulated clones had significantly depressed growth when compared with scrambled control cells. In both the SK-N-MC and LMTK– models, the MTf hyper-expressing clones proliferated at a significantly increased rate compared with cells transfected with the vector alone. Results are expressed as mean ± standard error of mean (n = 6).

 
Cell culture, transfection and proliferation assay
Human SK-Mel-2 melanoma, SK-Mel-28 melanoma, SK-N-MC neuroepithelioma, mouse LMTK– fibroblast and L235 hybridoma cell lines were obtained from the American Type Culture Collection (Manassas, VA) and cultured as described previously (5). Transfections of pCI-neo transgenes/vectors into murine LMTK– fibroblasts and pCMV-Script® transgenes/vectors into human SK-N-MC neuroepithelioma cells were performed with Lipofectin® reagent (Invitrogen, Melbourne, Australia). Stable transfection of SK-Mel-2, SK-Mel-28, LMTK– and SK-N-MC cells was achieved through selecting clones in G418 (1000 µg/ml; Alexis Biochemicals, Lausen, Switzerland) (10). Proliferation was assessed by viable cell counts or the 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay (10,25).

Animals
The MTf–/– mice were generated by homologous gene targeting in embryonic stem cells as described previously (11).

RNA isolation, reverse transcription–polymerase chain reaction and western analysis
Total RNA was isolated using TRIzol Reagent® (Invitrogen) (10). Semi-quantitative reverse transcription–polymerase chain reaction (RT-PCR) was performed using the SuperScript III RT/Platinum® Taq Mix (10). The sequences of the primers implemented are listed in Table I. The housekeeping gene, ß-actin, was co-amplified as an internal standard. Protein isolation and western analysis were performed using the established techniques (26). The L235 hybridoma that produces a monoclonal antibody against human MTf was prepared as described (5). A polyclonal antibody was used to detect mouse MTf (11).


View this table:
[in this window]
[in a new window]

  		Table I. Primers for amplification of mouse and human mRNA

 
Microarray processing
Brain tissues from six female littermate mice were used for RNA isolation and RT-PCR confirmation of the microarray results (10). This consisted of four MTf–/– and two MTf+/+ animals. Total RNA from the cell models was processed as reported (10). RNA was hybridized onto the appropriate GeneChip® (Affymetrix, Santa Clara, CA), namely the Human Genome U133 Plus 2.0 Array or Mouse Genome 430 2.0 Array.

Microarray data analysis
Data processing.
Low-level analysis (image to raw expression values) was performed with Affymetrix GeneChip® Operating Software 1.3.0. Absolute expression signals as well as the mean probe level fold changes (expressed as SigLogRatio, i.e. log transformation of fold change using base 2, presence/absence and change calls) were calculated. All chips were scaled to a target intensity of 1000.

Gene filtering.
The criteria for a transcript to be differentially expressed between samples in the analysis were: (i) call = ‘P’ (present) in at least one sample, (ii) DiffCall! = ‘NC’ (no change) and (iii) SigLogRatio ≥ 1.3 (i.e. fold change ≥ 2.5) in comparison between treated and control cell lines and tissues.

Gene expression analysis.
A two-phase strategy was used to identify differentially expressed genes. First, genome-wide screening was performed using Affymetrix GeneChips®. Then, data processing and gene filtering methods were applied to all samples to obtain a list of genes with the greatest fold change. Comparative data analysis combining the SK-Mel-28, SK-N-MC, LMTK– and MTf–/– mouse models were performed to generate a list of 50 genes with the greatest fold change (Table II). A number of genes listed in Table II show inverse differential gene expression between the MTf down-regulation/ablation models (SK-Mel-28 cells and MTf–/– mice) and MTf hyper-expression models (SK-N-MC and LMTK– cells). These analyses were meant only to provide initial screening of genes that showed differential expression. Definitive evidence of differential expression was obtained from RT-PCR assessment of samples used for the microarray analysis and also from at least three other independent samples.


View this table:
[in this window]
[in a new window]

  		Table II. Summary of differential gene expression across all models assessed using Affymetrix GeneChips®

 
Annotation.
Functional annotation of genes was assigned through Gene Ontology (http://www.geneontology.org) and classifications obtained through public databases such as NetAffx (http://www.affymetrix.com/analysis/index.affx) and DAVID (http://david.abcc.ncifcrf.gov).

Data availability.
The complete data set can be accessed on the Gene Expression Omnibus (http://www.ncbi.nlm.nih.gov/geo/), using the accession numbers GSM101412-GSM101417 and GSM157163-157171.

Statistical analysis
Results are expressed as mean ± standard error of the mean or mean ± standard deviation. All experiments were performed at least three times (n = 3). Excluding the statistical analyses of the microarray results, all data were compared using the Student's t-test. Data were considered statistically significant when P < 0.05.


   Results
Top
 Abstract
 Introduction
 Materials and methods
 Results
Discussion
 References
 
Our previous studies established two models of MTf down-regulation, namely an MTf knockout mouse and down-regulation of MTf expression in SK-Mel-28 human melanoma cells mediated via PTGS (10,11). In the current investigation, we created another model of MTf down-regulation mediated by PTGS using human SK-Mel-2 melanoma cells and two models of MTf hyper-expression generated by stable transfection of an MTf expression vector in mouse LMTK– fibroblasts and human SK-N-MC neuroepithelioma cells. Alterations in gene expression were then examined using whole-genome microarray. The assessment of the changes in gene expression between the three models of MTf down-regulation and the two models of MTf up-regulation overcomes potential model-specific changes in gene expression that could result in incorrect conclusions. Hence, this approach allows identification of a common role for MTf across various cell types.

Generation of a melanoma cell model of MTf down-regulation
SK-Mel-2 cells were stably transfected with a pS-MTf vector, which encoded one of four transgenes (A, B, C and D) that transcribe a 19-mer-specific anti-MTf hairpin siRNA. As a control, SK-Mel-2 cells were transfected with the same vector containing a scrambled, non-specific siRNA sequence (pS-scrambled) with no known homology to human, rat or mouse sequences. Of the four siRNAs specifically targeted to the human MTf gene, we found the vector containing either the A or C transgene (pS-MTf/A3 or pS-MTf/C1) consistently provided the most significant and pronounced down-regulation of MTf expression compared with controls with >58% down-regulation of human MTf mRNA expression and >65% down-regulation of protein expression (Figure 1A). These two transgenes were targeted to positions 2046–2064 bp (A) and 2048–2066 bp (C) in the human MTf gene (Genbank accession no.: NM_005929), and share significant overlap. Previously, this region of the human MTf gene was shown to be particularly sensitive to the effects of the transgene B siRNA (position 2031–2049 bp) in SK-Mel-28 cells (10), which targets the mRNA transcript just upstream of the positions targeted by pS-MTf/A3 or pS-MTf/C1.

In good agreement with our previous study using SK-Mel-28 cells (10), SK-Mel-2 cells transfected with pS-MTf/C1 proliferated at a slower rate than control cells (SK-Mel-2 parent and SK-Mel-2 scrambled control cells) over a growth period of 4 days and this was significant on days 2 (P < 0.01), 3 (P < 0.01) and 4 (P < 0.001) (Figure 1B). A significant (P < 0.01) difference in growth rate between the controls (i.e. SK-Mel-2 parent and scrambled control cells) and cells transfected with pS-MTf/A3 was only observed on day 4 (Figure 1B). The greater decrease in growth rate observed in the C1 clone may be attributed to the more effective decrease in MTf protein expression compared with the A3 clone (Figure 1A).

As an additional internal control for potential off-target effects of PTGS, we assessed expression of the interferon-target gene, 2',5'-oligoadenylate synthetase 1 (27) to exclude the possibility that the siRNA was inducing a non-specific phenotypic change. There were no differences in 2',5'-oligoadenylate synthetase 1 expression between cells stably transfected with pS-MTf/A3 or pS-MTf/C1 or the scrambled control and SK-Mel-2 parent cell lines (data not shown). Hence, these SK-Mel-2 cells provide a second model of MTf down-regulation in melanoma cells and were utilized to further test differential gene expression that was identified from the four models on which microarray analyses were performed.

Characterization of cellular models hyper-expressing MTf
Hyper-expression of MTf mRNA and protein was achieved by stably transfecting human and mouse MTf cDNA into human SK-N-MC neuroepithelioma cells and mouse LMTK– fibroblasts, respectively. The G418-resistant clones were tested for mouse and human MTf over-expression by RT-PCR and western blotting (Figure 1A). Two clones from the SK-N-MC (hlMTf-E and hlMTf-4) and LMTK– (mMTf-4 and mMTf-6) cell lines that showed marked hyper-expression were selected for further study. As shown in Figure 1A, over-expression of mouse and human MTf at the mRNA and protein levels in the transfected clones was significantly (P < 0.0001) greater than the vector or parent control cell lines. For example, the LMTK–/mMTf-6 hyper-expressing clone had an approximate 18-fold increase in MTf mRNA expression and 48-fold increase in MTf protein expression compared with the vector control (Figure 1A). The difference in MTf expression between hyper-expressing clones and controls was normalized relative to the expression of the housekeeping gene, ß-actin.

As the studies above and previous experiments in melanoma cells identified changes in cell proliferation relative to decreased levels of MTf expression (10,28), we examined the effect of MTf hyper-expression in our clones on cellular growth. Over 4 days, the MTf hyper-expressing clones of SK-N-MC (hlMTf-E and hlMTf-4) and LMTK– (mMTf-4 and mMTf-6) proliferated at a significantly faster rate from day 2 onwards (P < 0.001) in comparison with their vector transfected controls, namely pCMV-A and pCl-1, respectively (Figure 1B). Since both the cells transfected with plasmid and MTf insert, or plasmid alone were grown in the presence of G418, the comparison is more appropriate than with parental cells.

Identification of new target genes associated with MTf and their potential biological significance
Individual statistical analysis of the four models assessed by whole-genome microarray, namely (i) the MTf–/– mice, (ii) MTf hyper-expressing LMTK– murine fibroblasts, (iii) MTf down-regulated SK-Mel-28 human melanoma cells and (iv) MTf hyper-expressing SK-N-MC human neuroepithelioma cells, resulted in a list of 385, 130, 1152 and 229 potential genes of interest, respectively, that showed significant differential expression within each model (data not shown).

The list of genes from each model were then assessed using gene ontology software as described in Materials and Methods to determine if MTf modulation affected any specific biological processes in particular (Figures 2 and 3). The classification of biological processes, e.g. cellular process, development, etc. (accessed on December 2006; http://www.ebi.ac.uk/), shown in these latter figures were defined by the functional annotation of genes assigned by the Gene Ontology Consortium (29).


Figure 2
View larger version (41K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 2. Biological processes associated with differentially expressed genes from Human Genome U133 Plus 2.0 and Mouse Genome 430 2.0 array GeneChips®. An overall distribution of all differentially expressed genes in four different MTf models, namely MTf–/– knockout mice, LMTK– hyper-expression, SK-Mel-28 down-regulation and SK-N-MC hyper-expression. Individual statistical analysis of the four models assessed by whole-genome microarray resulted in a list of 385, 130, 1152 and 229 genes of interest, respectively, that showed significant differential expression within each model (data not shown). The lists of genes from each model were then assessed using the comprehensive functional annotation software (DAVID, http://david.abcc.ncifcrf.gov).

 


Figure 3
View larger version (39K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 3. Specific biological processes associated with down- and up-regulated genes from Human Genome U133 Plus 2.0 and Mouse Genome 430 2.0 array GeneChips®. A comparison between down- and up-regulated genes in four different MTf models, namely MTf–/– knockout mice, LMTK– MTf hyper-expressing cells, SK-Mel-28 with MTf down-regulation and SK-N-MC MTf hyper-expressing cells. Individual statistical analysis of the four models assessed by whole-genome microarray resulted in a list of 385, 130, 1152 and 229 genes of interest, respectively, that showed significant differential expression within each model (data not shown). The genes were separated into those up-regulated and down-regulated. These lists of genes from each model were then assessed using the comprehensive functional annotation software (DAVID, http://david.abcc.ncifcrf.gov).

 
Examining the biological processes associated with the differentially expressed genes across the four MTf-expression models demonstrated that genes encoding physiological processes (e.g. cell death, locomotion and metabolism) and cellular processes (e.g. adhesion, communication and differentiation) represented the two largest classes of differentially expressed genes (16.9–28.2% and 17.3–29.6%, respectively) (Figure 2). The next largest proportion of genes were associated with regulation of biological processes (4.6–10.5%), development (4.0–5.9%) and response to stimuli (2.6–5.6%) (Figure 2).

The individual analysis of down- and up-regulated genes from the four different models revealed some trends in the number of genes that showed differential expression in a particular biological process (Figure 3). In the human MTf models, comparing the MTf down-regulated SK-Mel-28 model and MTf hyper-expressing SK-N-MC model, we found inverse changes in expression of a number of genes. For example, in the SK-Mel-28 model, for those biological functions associated with cellular processes and development and physiological processes, there were a greater number of up-regulated genes compared with down-regulated genes (Figure 3). On the other hand, the MTf hyper-expressing SK-N-MC model showed completely inverse results, with a higher number of down-regulated genes being observed (Figure 3). Hence, this inverse trend suggests that in human neoplastic cell types, MTf may have roles in the biological processes mentioned above. However, when examining the mouse models, namely the MTf–/– mice and LMTK– hyper-expressing fibroblasts, we observed little correlation in reciprocal expression of the number of genes involved in the functions illustrated in Figure 3, such as development.

It is of interest that despite the fact that MTf is involved in cellular growth (Figure 1B) (10,18), the proportion of genes affected by modulation of MTf expression did not demonstrate a marked alteration in those classically involved in growth as defined by the Gene Ontology Consortium (Figure 3). This could suggest that the MTf-mediated alterations in proliferation are not through traditional pathways.

Combined analysis of all four models was then performed by applying a gene filtering strategy, as described in Materials and Methods, to identify differential gene expression between the MTf down- and up-regulated models. These analyses generated a table of the top 50 genes with the largest positive- and largest negative-fold changes common across the majority of models (Table II). A positive value of the log2-fold change in expression indicates up-regulation in the four models tested relative to the appropriate controls, whereas a negative value demonstrates down-regulation. For example, Fyb-binding protein (Fyb-120/130) was up-regulated compared with controls in models of MTf down-regulation, whereas conversely it was down-regulated when compared with controls in the MTf hyper-expression models (Table II).

Validation of gene expression identified in the microarrays using RT-PCR
To validate expression of the 50 genes showing the largest expression changes in the array data from the four models of MTf down- or up-regulation (Table II), RT-PCR was used. This latter technique confirmed in 81% of cases that the microarray had correctly predicted the change in gene expression. In contrast to the other four models, whole-genome microarray was not performed on SK-Mel-2 cells. However, the genes of interest generated from the comparison of the microarray studies from the four models were also assessed by RT-PCR in SK-Mel-2 cells, providing a further model to examine the effect of MTf levels on gene expression in a melanoma cell type. Although at least two clones and two controls (or animals) were assessed for gene expression in each individual model, for clarity we have shown one clone and one control that are representative of at least three separate experiments performed (Figure 4).


Figure 4
View larger version (38K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 4. RT-PCR analysis of genes showing differential expression across four different MTf models assessed using Affymetrix GeneChips®. RT-PCR of the genes assessed by their log2 value to show differential expression from the microarray data across five models of MTf down-regulation or ablation and hyper-expression. Analyses were performed on the RNA samples submitted for microarray in addition to at least three other RNA samples that were freshly isolated from the tissue or cell lines.

 
As determined previously, MTf expression was not detected in the MTf–/– mice (10,11), and was reduced in MTf down-regulated SK-Mel-28 and SK-Mel-2 human melanoma cells (Figure 4). Conversely, MTf expression was significantly up-regulated in the hyper-expression models of mouse LMTK– and human SK-N-MC cells (Figure 4). The expression of the Fe-regulated gene TfR1 was unchanged within each model (Figure 4), confirming our previous studies demonstrating that MTf expression does not affect Fe metabolism (5,10). However, expression of Tcf4 and Mef2a were up-regulated in response to MTf down-regulation and were decreased when MTf was hyper-expressed (Figure 4). Apolipoprotein D (Apod) was down-regulated in response to MTf down-regulation as found previously in MTf–/– mice (10), but was up-regulated in LMTK– cells hyper-expressing mouse MTf (Figure 4). However, unexpectedly, no significant change in Apod expression was observed in melanoma cells with down-regulated MTf expression and Apod was down-regulated in SK-N-MC cells hyper-expressing human MTf (Figure 4). This could suggest that Apod has roles in context of the entire organism that are not essential for melanoma cells in vitro.

Of the 50 genes that were identified as part of the microarray study, 15 genes were selected based upon the highest fold change in expression relative to the control. This was done to ensure that their change in expression could be sensitively analyzed using RT-PCR. These genes were found to be differentially expressed across most of the four cell lines and the mouse model. These genes included MTf, Abcb5, Btg family member 2 (Btg2), Cd44 antigen (Cd44), contactin 4 (Cntn4), discoidin domain receptor family member 1 (Ddr1), Dickkopf homolog 1 (Dkk1), Fyb-120/130, interleukin 2 receptor gamma (Il2rg), Ptpdc1, Thtpa, Mef2a, glutaminase (Gls), Tcf4 and Apod (Figure 4). Densitometric analysis of the RT-PCR data was used to evaluate the changes in gene expression. In addition to the samples submitted for microarray, we confirmed the data independently using at least three other RNA samples that were isolated from each model. In all further description of results, the fold changes in gene expression are referred to as log2 ratios. The expression of Thtpa was up-regulated between 0.26- and 0.89-fold in all three models of MTf down-regulation (Figure 5). In contrast, Thtpa was down-regulated by –0.16- to –0.33-fold in all MTf hyper-expression models. Similarly, Tcf4 expression was found to be inversely expressed relative to MTf levels across all five models, whereas Mef2a expression was inversely correlated with MTf expression in all models except SK-N-MC cells, where Mef2a was not detected even after 40 PCR cycles (Figure 5). Conversely, Abcb5 expression was found to positively correlate with MTf expression in all models.


Figure 5
View larger version (14K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 5. Comparative data analysis of genes showing differential expression between MTf down-regulation/ablation and hyper-expression compared with controls. Densitometric analysis of RT-PCR results from the genes of interest. The bar graphs demonstrate the extent of the changes in gene expression in each model. The change in gene expression is quantified relative to the housekeeping gene ß-actin that was used as an internal loading control. The results are representative of at least three separate experiments performed.

 
Interestingly, in the human cell lines SK-Mel-2 and SK-Mel-28, Ptpdc1 was up-regulated when MTf was down-regulated, while it was down-regulated when MTf was up-regulated in human SK-N-MC cells (Figure 5). However, a literature search revealed no Ptpdc1 transcripts in the mice and its expression in murine models could not be confirmed. Likewise, although Gls was found to be inversely expressed relative to MTf expression in the mouse models (Figures 4 and 5), its expression was not detected in any of the human models after 40 PCR cycles. There were some discrepancies between models when the expressions of other genes were assessed. For example, Il2rg expression in the MTf–/– mice was up-regulated by 0.26-fold and was down-regulated by –1.12-fold in the LMTK– MTf hyper-expression cell line (Figure 4), but in contrast and unexpectedly, its expression was decreased in the SK-Mel-28 MTf down-regulation cell line (Figure 4). The reason for this discrepancy may relate to cell-type specific differences. Hence, only the most consistent and significant changes in gene expression across the majority of the five models were considered further in our analysis.


   Discussion
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
References
 
Despite over 25 years of research, there is little conclusive evidence regarding the function of MTf. However, this molecule has been suggested to be involved in a diverse range of processes (13,15,17,19,20). In the present study, we examined the role of MTf through whole-genome microarray analysis of various models of MTf down-regulation and hyper-expression generated within our laboratory, namely (i) the MTf–/– mice, (ii) MTf hyper-expressing LMTK– murine fibroblast cells, (iii) MTf down-regulated SK-Mel-28 human melanoma cells and (iv) MTf hyper-expressing SK-N-MC human neuroepithelioma cells.

Our earlier study reported that down-regulation of MTf expression in SK-Mel-28 cells resulted in decreased cell proliferation and migration compared with cells transfected with the scrambled control vectors or the untransfected parent cell line (10). A subsequent study by others has confirmed our latter observations and also suggested that MTf played a role in melanoma cell invasion in vivo (18). Furthermore, we demonstrated that when injected into nude mice, melanoma cells with reduced MTf expression were not as tumorigenic as parent cells that expressed high MTf levels (10). The current study has extended these findings to demonstrate that down-regulation of MTf expression in another melanoma cell type, SK-Mel-2, also decreased proliferation rate compared with controls (Figure 1B). Together, these data strongly suggest that the function of MTf is significant, especially within melanoma cells.

To contrast these models of MTf down-regulation, we engineered MTf hyper-expressing vectors that were used to create stable, MTf hyper-expressing SK-N-MC and LMTK– cell lines (Figure 1A). Both MTf hyper-expressing cell lines demonstrated increased proliferation in vitro compared with cells transfected with the vector without the MTf insert (Figure 1B). Clearly, these phenotypic changes were not clonal effects generated by antibiotic selection, as they were observed over multiple cell types and also in different clones from the same cell type. These data demonstrate that MTf plays a role in proliferation in vitro across a range of cell types, particularly in melanoma cells, where the effects of changes in MTf expression were marked. Evidence from earlier studies by others implementing hyper-expression models of MTf using melanoma cells also demonstrated increased MTf expression results in accelerated growth (28).

To identify the potential molecular pathways involved in MTf function, we used genome-wide microarrays to examine patterns of differential gene expression in four models of MTf down- and up-regulation. The changes in gene expression identified within these systems were then assessed via semi-quantitative RT-PCR. In addition, a further model of MTf down-regulation in SK-Mel-2 cells was generated and the changes in gene expression were assessed within this cell type. It should be noted, that our investigation examined different cell types in vitro (melanoma, neuroepithelioma and fibroblasts) in addition to the whole organism, namely the MTf–/– mice. Clearly, this comparative analysis had the potential to identify a function for MTf that is common to multiple cell types and organisms. However, each model could also be assessed individually to examine the role of MTf in specific cell types such as melanoma, where the protein is known to be endogenously expressed at high levels.

Across all five models of MTf down- and up-regulation, we identified and confirmed via RT-PCR the inverse expression of two genes relative to MTf levels, namely Tcf4 and Thtpa (Figure 5). In contrast, the expression of Abcb5 was directly correlated with MTf expression, across all five models. For instance, when MTf was down-regulated, the expression of Abcb5 was similarly decreased and vice versa. Other notable genes included Gls, Ptpdc1 and Mef2a that were negatively regulated by MTf expression in two, three and four of the MTf models, respectively (Figure 5). The functions of the products of these genes and their relevance to MTf function are discussed below.

Tcf4 and Thtpa
Considering the functions of the genes inversely regulated relative to MTf levels across all five models, namely Tcf4 and Thtpa, it is notable that Tcf4 is a transcription factor that plays roles in proliferation and the regulation of melanocyte differentiation (30–33). The ability of Tcf4 to promote or repress target gene transcription is itself controlled by the Wingless (Wnt)/ß-catenin signaling pathway that plays significant roles in development, growth, proliferation and tumorigenesis (34–36). A potential role for MTf in this pathway is of interest, although further studies are required to confirm this in the whole organism and melanoma cells.

The other negatively regulated gene across the five MTf models, namely Thtpa, is involved in cell metabolism and/or signaling (37). Thtpa is known to de-phosphorylate thiamine triphosphate and is involved in thiamine metabolism (37). This reaction is important for carbohydrate catabolism and is essential for muscular and neuronal function (37). As shown in the current study, down-regulation of MTf leads to increased Thtpa expression, which could potentially lead to decreased levels of the energy currency, thiamine triphosphate. It can be speculated that this effect may result in the down-regulation of proliferation observed. Conversely, up-regulation of MTf resulted in decreased Thtpa levels that may lead to greater thiamine triphosphate and increased proliferation. Clearly, the latter is directly relevant to melanoma cells that hyper-express MTf and could potentially explain the increased proliferation found.

Abcb5
The expression of the Abcb5 gene was directly correlated with MTf expression across all five models (Figure 5). The product of this gene is a membrane transporter that is associated with melanoma and has roles in rhodamine transport, progenitor cell fusion, cellular growth and differentiation (38). Intriguingly, whereas Abcb5 expression appears to be restricted to pigmented cell types, two isoforms of Abcb5 have been suggested to be diagnostic markers for melanoma or therapeutic targets (39,40). Clearly, the direct relationship between Abcb5 and MTf deserves further investigation, as it may play an important role in cellular growth and proliferation.

Gls, Mef2a and Ptpdc1
The other genes showing inverse differential expression relative to MTf levels in some of the models, namely Gls, Mef2a and Ptpdc1, have also been demonstrated to have roles in growth and proliferation (21–24,41–43). The Gls expression in tumor has been shown to reach optimal expression and activity immediately before the maximum proliferation rate (21,22,44). Furthermore, Mef2a expression was regulated inversely to MTf levels across four models except the hyper-expression model of MTf in SK-N-MC cells, where it could not be detected using RT-PCR (Figures 4 and 5). It is known that the Mef2a transcription factor plays a role in muscles and other tissues (45,46) and has been reported to exert important functions in cell growth and survival (23,24). The negative regulation of Gls and Mef2a in most of the MTf models (two and four models, respectively; Figures 4 and 5) indicates a direct or indirect response to the modulation of MTf expression and suggests a role for MTf in proliferation.

Although little is known concerning the role of the Ptpdc1 gene product, its expression was inversely regulated relative to MTf levels across three human models, but it is not reported in the mouse model (Figures 4 and 5). Ptpdc1 is a member of the protein tyrosine phosphatase family that regulates activities of phosphoproteins through dephosphorylation (41). Interestingly, members of the protein tyrosine phosphatase family are predicted to have functions as tumor suppressor genes and to play roles in signaling and melanoma (41,42). In fact, McArdle et al. (43) have demonstrated that loss of protein tyrosine phosphatase activity may stimulate autonomous growth in advanced melanoma and have roles in melanoma progression. Therefore, the inverse expression of Ptpdc1 relative to MTf in the three human models may explain, at least in part, the alterations observed in proliferation (Figure 1B).

In summary, we developed five models of MTf down-regulation or ablation and hyper-expression across a range of cell lines and in the mice. In the cell line models, we observed that MTf hyper-expression leads to increased cellular proliferation, whereas MTf down-regulation resulted in decreased proliferation. Using genome-wide microarray analysis, we identified a number of changes in gene expression across all five models. The most important of these were Abcb5, Thtpa and Tcf4 that were modulated by MTf expression. The products of these three genes are involved in membrane transport, thiamine phosphorylation and cell proliferation/survival, respectively. Hence, this study has identified molecular targets directly or indirectly regulated by MTf and potential pathways involved in its function. These molecular targets could be involved, at least in part, with the role of MTf in modulating proliferation.


   Acknowledgments
 
The authors acknowledge the critical comments on the manuscript prior to submission by Dr Daniel Vyoral, Ms Danuta Kalinowski, Dr David Lovejoy, Dr Dong Fu, Dr Erika Becker, Ms Megan Whitnall, Dr Robert Sutak, Mr Xiangcong Xu and Ms Zaklina Kovacevic of the Iron Metabolism and Chelation Program. For assistance with the interpretation of statistical data from Affymetrix microarrays, we kindly acknowledge Emphron Informatics (Brisbane, Australia). This work was supported by a Fellowship and Project Grants from the National Health and Medical Research Council (NHMRC) and Australian Research Council (ARC) to D.R.R. Y.S.R. acknowledges a PhD Scholarship from the Australian Post-Graduate Award Scheme, University of Sydney, and L.L.D. acknowledges a Dora Lush PhD Scholarship from the NHMRC.

Conflict of Interest Statement: None declared.


   References
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 

  1. Woodbury RG, et al. Identification of a cell surface protein, p97, in human melanomas and certain other neoplasms. Proc. Natl Acad. Sci. USA (1980) 77:2183–2187.[Abstract/Free Full Text]
  2. Brown JP, et al. Quantitative analysis of melanoma-associated antigen p97 in normal and neoplastic tissues. Proc. Natl Acad. Sci. USA (1981) 78:539–543.[Abstract/Free Full Text]
  3. Brown JP, et al. Human melanoma-associated antigen p97 is structurally and functionally related to transferrin. Nature (1982) 296:171–173.[CrossRef][Medline]
  4. Suryo Rahmanto Y, et al. The melanoma tumor antigen, melanotransferrin (p97): a 25-year hallmark from iron metabolism to tumorigenesis. Oncogene (2007) doi:10.1038/sj.onc.1210442.
  5. Richardson DR. The role of the membrane-bound tumour antigen, melanotransferrin (p97), in iron uptake by the human malignant melanoma cell. Eur. J. Biochem. (2000) 267:1290–1298.[Web of Science][Medline]
  6. Sekyere EO, et al. Examination of the distribution of the transferrin homologue, melanotransferrin (tumour antigen p97), in mouse and human. Biochim. Biophys. Acta (2005) 1722:131–142.[Medline]
  7. Desrosiers RR, et al. Expression of melanotransferrin isoforms in human serum: relevance to Alzheimer's disease. Biochem. J. (2003) 374:463–471.[CrossRef][Web of Science][Medline]
  8. Rose TM, et al. Primary structure of the human melanoma-associated antigen p97 (melanotransferrin) deduced from the mRNA sequence. Proc. Natl Acad. Sci. USA (1986) 83:1261–1265.[Abstract/Free Full Text]
  9. Baker EN, et al. Human melanotransferrin (p97) has only one functional iron-binding site. FEBS Lett. (1992) 298:215–218.[CrossRef][Web of Science][Medline]
  10. Dunn LL, et al. The function of melanotransferrin: a role in melanoma cell proliferation and tumorigenesis. Carcinogenesis (2006) 27:2157–2169.[Abstract/Free Full Text]
  11. Sekyere EO, et al. Role of melanotransferrin in iron metabolism: studies using targeted gene disruption in-vivo. Blood (2006) 107:2599–2601.[Abstract/Free Full Text]
  12. Richardson D, et al. The uptake of inorganic iron complexes by human melanoma cells. Biochim. Biophys. Acta (1991) 1093:20–28.[Medline]
  13. Demeule M, et al. High transcytosis of melanotransferrin (P97) across the blood-brain barrier. J. Neurochem. (2002) 83:924–933.[CrossRef][Web of Science][Medline]
  14. Jefferies WA, et al. Reactive microglia specifically associated with amyloid plaques in Alzheimer's disease brain tissue express melanotransferrin. Brain Res. (1996) 712:122–126.[CrossRef][Web of Science][Medline]
  15. McNagny KM, et al. The eosinophil-specific cell surface antigen, EOS47, is a chicken homologue of the oncofetal antigen melanotransferrin. Blood (1996) 87:1343–1352.[Abstract/Free Full Text]
  16. Kawamoto T, et al. Expression of membrane-bound transferrin-like protein p97 on the cell surface of chondrocytes. Eur. J. Biochem. (1998) 256:503–509.[Web of Science][Medline]
  17. Sala R, et al. The human melanoma associated protein melanotransferrin promotes endothelial cell migration and angiogenesis in vivo. Eur. J. Cell Biol. (2002) 81:599–607.[CrossRef][Web of Science][Medline]
  18. Bertrand Y, et al. Melanotransferrin induces human melanoma SK-Mel-28 cell invasion in vivo. Biochem. Biophys. Res. Commun. (2007) 353:418–423.[CrossRef][Web of Science][Medline]
  19. Rolland Y, et al. Melanotransferrin stimulates t-PA-dependent activation of plasminogen in endothelial cells leading to cell detachment. Biochim. Biophys. Acta (2006) 1763:393–401.[Medline]
  20. Demeule M, et al. Regulation of plasminogen activation: a role for melanotransferrin (p97) in cell migration. Blood (2003) 102:1723–1731.[Abstract/Free Full Text]
  21. Medina MA, et al. Relevance of glutamine metabolism to tumor cell growth. Mol. Cell. Biochem. (1992) 113:1–15.[Web of Science][Medline]
  22. Aledo JC, et al. Phosphate-activated glutaminase expression during tumor development. FEBS Lett. (1994) 341:39–42.[CrossRef][Web of Science][Medline]
  23. McKinsey TA, et al. MEF2: a calcium-dependent regulator of cell division, differentiation and death. Trends Biochem. Sci. (2002) 27:40–47.[CrossRef][Web of Science][Medline]
  24. Zhao M, et al. Regulation of the MEF2 family of transcription factors by p38. Mol. Cell. Biol. (1999) 19:21–30.[Abstract/Free Full Text]
  25. Whitnall M, et al. A class of iron chelators with a wide spectrum of potent antitumor activity that overcomes resistance to chemotherapeutics. Proc. Natl Acad. Sci. USA (2006) 103:14901–14906.[Abstract/Free Full Text]
  26. Gao J, et al. The potential of iron chelators of the pyridoxal isonicotinoyl hydrazone class as effective antiproliferative agents, IV: the mechanisms involved in inhibiting cell-cycle progression. Blood (2001) 98:842–850.[Abstract/Free Full Text]
  27. Bridge AJ, et al. Induction of an interferon response by RNAi vectors in mammalian cells. Nat. Genet. (2003) 34:263–264.[CrossRef][Web of Science][Medline]
  28. Estin CD, et al. Transfected mouse melanoma lines that express various levels of human melanoma-associated antigen p97. J. Natl Cancer Inst. (1989) 81:445–448.[Abstract/Free Full Text]
  29. Ashburner M, et al. Gene ontology: tool for the unification of biology. The Gene Ontology Consortium. Nat. Genet. (2000) 25:25–29.[CrossRef][Web of Science][Medline]
  30. Graham TA, et al. Tcf4 can specifically recognize beta-catenin using alternative conformations. Nat. Struct. Biol. (2001) 8:1048–1052.[CrossRef][Web of Science][Medline]
  31. Zhao DH, et al. Aberrant expression and function of TCF4 in the proliferation of hepatocellular carcinoma cell line BEL-7402. Cell Res. (2004) 14:74–80.[CrossRef][Web of Science][Medline]
  32. Furumura M, et al. Involvement of ITF2 in the transcriptional regulation of melanogenic genes. J. Biol. Chem. (2001) 276:28147–28154.[Abstract/Free Full Text]
  33. Mann B, et al. Target genes of beta-catenin-T cell-factor/lymphoid-enhancer-factor signaling in human colorectal carcinomas. Proc. Natl Acad. Sci. USA (1999) 96:1603–1608.[Abstract/Free Full Text]
  34. Behrens J. The role of the Wnt signalling pathway in colorectal tumorigenesis. Biochem. Soc. Trans. (2005) 33:672–675.[CrossRef][Web of Science][Medline]
  35. Gordon MD, et al. Wnt signaling: multiple pathways, multiple receptors, and multiple transcription factors. J. Biol. Chem. (2006) 281:22429–22433.[Free Full Text]
  36. Sato A, et al. Profilin is an effector for Daam1 in non-canonical Wnt signaling and is required for vertebrate gastrulation. Development (2006) 133:4219–4231.[Abstract/Free Full Text]
  37. Makarchikov AF, et al. Thiamine triphosphate and thiamine triphosphatase activities: from bacteria to mammals. Cell. Mol. Life Sci. (2003) 60:1477–1488.[CrossRef][Web of Science][Medline]
  38. Frank NY, et al. Regulation of progenitor cell fusion by ABCB5 P-glycoprotein, a novel human ATP-binding cassette transporter. J. Biol. Chem. (2003) 278:47156–47165.[Abstract/Free Full Text]
  39. Chen KG, et al. Principal expression of two mRNA isoforms (ABCB 5alpha and ABCB 5beta) of the ATP-binding cassette transporter gene ABCB 5 in melanoma cells and melanocytes. Pigment Cell Res. (2005) 18:102–112.[CrossRef][Web of Science][Medline]
  40. Frank NY, et al. ABCB5-mediated doxorubicin transport and chemoresistance in human malignant melanoma. Cancer Res. (2005) 65:4320–4333.[Abstract/Free Full Text]
  41. Tonks NK. Protein tyrosine phosphatases: from genes, to function, to disease. Nat. Rev. Mol. Cell Biol. (2006) 7:833–846.[CrossRef][Web of Science][Medline]
  42. Easty D, et al. Protein tyrosine phosphatases, new targets for cancer therapy. Curr. Cancer Drug Targets (2006) 6:519–532.[CrossRef][Web of Science][Medline]
  43. McArdle L, et al. Microarray analysis of phosphatase gene expression in human melanoma. Br. J. Dermatol. (2005) 152:925–930.[CrossRef][Web of Science][Medline]
  44. Souba WW. Glutamine and cancer. Ann. Surg. (1993) 218:715–728.[Web of Science][Medline]
  45. Wang Q. Advances in the genetic basis of coronary artery disease. Curr. Atheroscler. Rep. (2005) 7:235–241.[Medline]
  46. Moreno H, et al. Differential regulation of the muscle-specific GLUT4 enhancer in regenerating and adult skeletal muscle. J. Biol. Chem. (2003) 278:40557–40564.[Abstract/Free Full Text]
Received January 23, 2007; revised March 17, 2007; accepted April 11, 2007.


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
28/10/2172    most recent
bgm096v1
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 Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Suryo Rahmanto, Y.
Right arrow Articles by Richardson, D.R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Suryo Rahmanto, Y.
Right arrow Articles by Richardson, D.R.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?