July 2012
Volume 53, Issue 8
Free
Biochemistry and Molecular Biology  |   July 2012
Differential Expression of Fourteen Proteins between Uveal Melanoma from Patients Who Subsequently Developed Distant Metastases versus Those Who Did Not
Author Affiliations & Notes
  • Annett Linge
    National Institute for Cellular Biotechnology, Dublin City University, Glasnevin, Dublin, Ireland; the
  • Susan Kennedy
    National Ophthalmic Pathology Laboratory and Research Foundation and the
  • Deirdre O'Flynn
    National Institute for Cellular Biotechnology, Dublin City University, Glasnevin, Dublin, Ireland; the
  • Stephen Beatty
    Macular Pigment Research Group, Department of Chemical and Life Sciences, Waterford Institute of Technology, Waterford, Ireland.
  • Paul Moriarty
    Department of Ophthalmology and Research Foundation, Royal Victoria Eye and Ear Hospital, Adelaide Road, Dublin, Ireland; and the
  • Michael Henry
    National Institute for Cellular Biotechnology, Dublin City University, Glasnevin, Dublin, Ireland; the
  • Martin Clynes
    National Institute for Cellular Biotechnology, Dublin City University, Glasnevin, Dublin, Ireland; the
  • Annemarie Larkin
    National Institute for Cellular Biotechnology, Dublin City University, Glasnevin, Dublin, Ireland; the
  • Paula Meleady
    National Institute for Cellular Biotechnology, Dublin City University, Glasnevin, Dublin, Ireland; the
  • Corresponding author: Paula Meleady, National Institute for Cellular Biotechnology, Dublin City University, Glasnevin, Dublin, Ireland; Paula.Meleady@dcu.ie
Investigative Ophthalmology & Visual Science July 2012, Vol.53, 4634-4643. doi:10.1167/iovs.11-9019
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Annett Linge, Susan Kennedy, Deirdre O'Flynn, Stephen Beatty, Paul Moriarty, Michael Henry, Martin Clynes, Annemarie Larkin, Paula Meleady; Differential Expression of Fourteen Proteins between Uveal Melanoma from Patients Who Subsequently Developed Distant Metastases versus Those Who Did Not. Invest. Ophthalmol. Vis. Sci. 2012;53(8):4634-4643. doi: 10.1167/iovs.11-9019.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Purpose.: To compare the proteomic profiles of two categories of primary uveal melanoma tissue samples; those from patients who have subsequently developed metastatic disease and those who have not.

Methods.: Two-dimensional difference gel electrophoresis (2D DIGE) was performed on 25 uveal melanoma tissue specimens (minimum follow-up of 7 years) comparing nine uveal melanoma tumors from patients who developed metastatic disease and 16 from those who did not. Most of the tumors which metastasized also exhibited chromosome 3 monosomy. Selected differentially expressed proteins were further followed up by immunohistochemistry and functional validation in vitro using siRNA.

Results.: Proteomic analysis revealed 14 statistically significant differentially expressed proteins, with nine showing increased expression (PDIA3, VIM/HEXA, SELENBP1, ENO1, CAPZA1, ERP29, TPI1, PARK7, and FABP3) and five showing decreased expression (EIF2S, PSMA3, RPSA, TUBB, and TUBA1B) in uveal melanomas that subsequently metastasized compared with those that did not. Immunohistochemical analysis was performed for six of the differentially expressed proteins and gave similar results to the 2D DIGE study for two of these proteins, fatty acid–binding protein, heart-type (FABP3) and triosephosphate isomerase (TPI1). siRNA knockdown in the 92.1 uveal melanoma cell line confirmed a functional role for FABP3 and TPI1 in invasion in vitro.

Conclusions.: Proteomic analysis identified proteins differentially expressed in uveal melanoma that will subsequently metastasize, some of which appear to have a functional role in invasion. These results may contribute to better predictive tests (along with genetic analysis) and to the identification of new therapeutic targets.

Introduction
Uveal melanoma is the most common primary intraocular malignancy in adults. The overall incidence of uveal melanoma is approximately five cases per million per year 1,2 and climbs to more than 20 cases per million per year by the age of 70. 3 The 5-year survival rate is approximately 75%. Approximately 50% of patients diagnosed with uveal melanoma die within 15 years because of metastasis. 1,4 In contrast to cutaneous melanoma, uveal melanoma initially spreads hematogenously, preferentially to the liver, but metastases to other distant sites such as lungs, bones, and skin are not uncommon. 5 The occurrence of metastases is primarily detected after disease-free intervals following local treatment. 6,7  
The majority of studies to date attempting to understand the pathogenesis of uveal melanoma have been performed at the genetic level. Gene expression profiling studies have revealed that primary uveal melanoma clusters into two different classes; class 1 tumors are associated with a good prognosis and class 2 tumors with a high metastatic risk. 810 These studies also provide a number of potential genetic markers, such as PTP4A3, 11 which may have the capability to predict the progression and metastasis of uveal melanoma. It has also been shown that deletions and loss of heterozygosity of chromosome 3 correlate with an increased risk of metastasis. 1214 A recent study by Harbour et al. 15 shows that mutational inactivation of BAP1 is a key event in the acquisition of metastatic competence of uveal melanoma. 15 They also demonstrate that histone deacetylase (HDAC) inhibitors can reverse the effects of BAP1 loss. 16 Another recent study, Asnaghi et al., 17 shows that Notch signaling promotes growth and invasion in uveal melanoma. 
To date, only a very limited number of proteomic studies have been carried out investigating the biology of the metastatic phenotype of uveal melanoma. The majority of these studies use cell line models. 18,19 Zuidervaart et al. 18 isolate and compare uveal melanoma cell lines derived from a primary tumor and from two of its liver metastases from the same patient. They identify 24 differentially expressed proteins; most of these have previously been reported to be involved in tumor metastasis. Pardo et al. 19 compare primary uveal melanoma cell cultures with different invasive potentials; this study reveals a number of differentially expressed proteins, such as the melanoma-associated antigen MUC18 and the high mobility group protein 1 HMG-1, which may serve as possible clinical biomarkers of uveal melanoma metastasis. A recent proteomic analysis of seven primary uveal melanoma tissues, four with monosomy 3 and three with disomy 3 status, 20 reveals differential expression of the HSP27 protein. Subsequent immunohistochemical analysis in a larger patient cohort finds that low HSP27 expression is strongly correlated with monosomy 3 and predicts increased mortality. 21  
The purpose of this pilot study was to identify differentially expressed proteins in primary uveal melanoma tissue specimens from patients with a minimum of 7 years clinical follow-up, comparing those who developed metastases to those who did not. Protein biomarkers predicting metastatic uveal melanoma could (along with genetic analysis) improve interim monitoring of patients whose tumors are likely to metastasize. Furthermore, such studies could lead to a better understanding of the biology of the disease and the identification of new therapeutic targets, as few promising targets currently exist for treatment of metastatic disease. 22  
Methods
Uveal Melanoma Sample Collection
Twenty-seven clinical uveal melanoma tissue specimens from patients with a minimum of 7 years clinical follow-up were accessioned from the files of the National Ophthalmic Pathology Laboratory (Royal Victoria Eye and Ear Hospital, Dublin). Of these, 10 patients with uveal melanomas subsequently developed metastatic disease and 17 did not. At the time of diagnosis, the fresh uveal melanoma samples were obtained from patients who had enucleation and stored at −80°C. Samples were also formalin-fixed and paraffin-embedded and cut in 4-μm sections for morphological assessment by immunohistochemistry. Fourteen out of the 27 uveal melanoma tissue specimens were used for both two-dimensional difference gel electrophoresis (2D DIGE) and immunohistochemical studies. Because the tumor samples were small in the case of 11 uveal melanomas, only 2D DIGE, not immunohistochemistry, could be done. In order to increase the number of tumor samples for immunohistochemistry, two additional uveal melanomas were included in the immunohistochemical study (tissue samples outlined in Table 1). Cytogenetic analysis of chromosome 3 status was performed using FISH by the Merseyside and Cheshire Genetics Laboratory, Crown St., Liverpool, UK. The study was approved by the Research and Ethics Committee of the Royal Victoria Eye and Ear Hospital, Dublin. The research adhered to tenets of the Declaration of Helsinki. 
Table 1. 
 
Clinical and Histopathological Features of the Patients Included in the 2D DIGE and Immunohistochemical Studies
Table 1. 
 
Clinical and Histopathological Features of the Patients Included in the 2D DIGE and Immunohistochemical Studies
Sample Sex Age at Diagnosis (y) Follow-up Time (y)* Clinical Characteristics Histopathological Characteristics Study
Metastatic Sites Ciliary Body Involvement Extrascleral Extension Cell Type LTD (mm) No. of Mitotic Cells (per 40 HPF) Chromosome 3 Status
1 F 49 NA Kidney N N S 12 4 Monosomy DIGE, IHC
2 F 51 1 Liver N N E 15 4 Monosomy DIGE, IHC
3 F 73 1 Liver Y N E 10 1 Monosomy DIGE, IHC
4 F 58 5 Liver Y N M 15 0 Monosomy DIGE, IHC
5 F 32 6 Liver Y Y S 8 12 Disomy IHC
6 F 49 5 Liver, skin N Y S 9 1 Monosomy DIGE, IHC
7 M 71 2 Lung N N M 11 16 Monosomy DIGE, IHC
8 M 64 10 Lung Y N M 10 12 Disomy DIGE, IHC
9 F 69 11 Liver N N S 12 16 Monosomy DIGE
10 M 71 7 Liver Y Y E 20 6 Monosomy DIGE
11 F 52 12 N N Y M 10 2 Disomy DIGE, IHC
12 F 55 12 N N N E 20 0 Disomy DIGE, IHC
13 F 76 12 N N N M 8 4 Disomy DIGE, IHC
14 M 53 12 N N N M 10 8 Trisomy DIGE, IHC
15 F 42 11 N N N S 17 1 Disomy DIGE, IHC
16 M 68 9 N Y N S 20 18 Disomy DIGE, IHC
17 M 68 11 N N N E 12 20 Trisomy DIGE, IHC
18 F 64 10 N N N E 8 8 Disomy IHC
19 F 46 11 N N N E 23 4 Disomy DIGE
20 M 34 17 N N N M 15 4 Disomy DIGE
21 F 75 11 N N N S 22 4 Disomy DIGE
22 M 52 10 N N N S 19 0 Disomy DIGE
23 M 50 11 N N N M 17 4 Disomy DIGE
24 M 86 9 N N N S 12 0 NA DIGE
25 M 62 9 N N N S 8 4 NA DIGE
26 F 70 9 N N N M 9 4 Disomy DIGE
27 F 74 7 N N N M 14 1 NA DIGE
Sample Preparation for 2D DIGE Analysis
Uveal melanoma tissue specimens were homogenized using the Sample Grinding Kit (GE Healthcare, Little Chalfont, Buckinghamshire, UK) according to the manufacturer's instructions. The samples were solubilized using 2D lysis buffer containing 7 M urea, 2 M thiourea, 4% (wt/vol) CHAPS, 40 mM dithiothreitol (DTT), and 0.5% immobilized pH gradient (IPG) buffer pH 3 to 11 (GE Healthcare). Insoluble material was removed by centrifugation at 14,000 rpm for 5 minutes at room temperature, and supernatants were stored at −80°C until required. Protein concentration was determined using the thiourea-compatible Quick Start Bradford Protein Assay Kit (Bio-Rad Laboratories Inc., Hercules, CA). 
Protein Labeling and Separation by 2D DIGE
For DIGE experiments, protein labeling was performed as recommended by the manufacturer and as previously described. 23 Briefly, protein lysates of 16 nonmetastasized and nine metastasized uveal melanoma tissues (50 μg of each) were labeled with 300 pm of either Cy3 or Cy5 fluorescent dyes (GE Healthcare) for comparison on the same gel. A pool of all samples (both nonmetastasized and metastasized) was labeled with Cy2 fluorescent dye, and this was used as an internal standard on all gels (50 μg each). 
Nonlinear 24-cm IPG strips (GE Healthcare), pH 3 to 11, were rehydrated in buffer (7 M urea, 2 M thiourea, 4% CHAPS, 0.5% IPG buffer, 50 mM DTT) overnight. 2D DIGE was then performed according to the manufacturer's instructions and as previously described. 23 All of the gels were scanned using a Typhoon 9400 Variable Mode Imager (GE Healthcare) to generate gel images at the appropriate excitation and emission wavelengths from the Cy2-, Cy3-, and Cy5-labeled samples. The resultant gel images were imported into Progenesis SameSpots software (Nonlinear Dynamics Ltd., Newcastle upon Tyne, UK) and cropped; quantification of protein expression and statistical analysis were then carried out comparing uveal melanomas that subsequently metastasized and those that did not. Proteins were defined as differentially expressed if the observed average ratio was ±1.2 with a t-test score <0.05. 
Protein Identification by Mass Spectrometry
Preparative gels containing 300 μg of protein per gel were fixed and poststained with colloidal Coomassie blue stain (Sigma-Aldrich, Dublin, Ireland); 2D gel spots were then excised and peptides extracted as previously described. 23 Extracted peptides were analyzed by liquid chromatography–tandem mass spectrometry (LC-MS/MS) using an Ultimate nanoLC system interfaced with an Orbitrap XL mass spectrometer (Thermo Fisher Scientific, Loughborough, Leicestershire, UK), essentially as previously described. 24 Database searches were performed using the UniProtKB/SwissProt database (taxa, Homo sapiens) downloaded from ftp://ftp.ncbi.nlm.nih.gov/blast/db/FASTA against the MASCOT search algorithm (v2.3, Matrix Science, London, UK) for protein identification. 25 The search parameters that were used allowed one missed cleavage, ±20 ppm for mass-to-charge ratio (m/z) error for MS and 0.5 Da m/z error for MS/MS, fixed modification of cysteine (carbamidomethyl-cysteine) and variable modification of methionine (oxidized) using a 95% confidence interval (CI) threshold (P < 0.05), with a minimum MASCOT score of ≥61. Protein identifications were accepted if they had at least two matched identified peptides and passed relevant statistical criteria. 
Immunohistochemistry
Following dewaxing and antigen retrieval in the Target retrieval solution (Dako, Glostrup, Denmark) pH 9 (for TPI1 and CD68) or pH 6 for 20 minutes at 97°C, immunohistochemical staining was performed using the AutostainerPlus (Dako) according to the manufacturer's instructions. Endogenous peroxidase activity was blocked (Peroxidase Block, Dako), and sections were then incubated with the following rabbit polyclonal antibodies in Dako REAL Antibody Diluent for 30 minutes: anti-cardiac fatty acid–binding protein (FABP3; dilution 1:100 vol/vol; Abcam, Cambridge, UK), anti-CAPZA1 (dilution 1:45 vol/vol; ProteinTech Group, Inc., Chicago, IL), anti-selenium binding protein (dilution 1:50 vol/vol; Abcam), anti-ERp57 (PDIA3; dilution 1:1000 vol/vol; Abcam), anti-PARK7 (Protein DJ-1; dilution 1:50 vol/vol; Atlas Antibodies AB, Stockholm, Sweden), anti-TPI1 (dilution 1:750 vol/vol; GeneTex, Inc., Irvine, CA [GTX104618]), and anti-CD68 (dilution 1:30 vol/vol; Dako). Negative control slides were incubated with Dako REAL Antibody Diluent only; the primary antibody was omitted. Signals were then visualized using the 3-amino-9-ethylcarbazole (AEC) substrate chromogen (Dako) according to the manufacturer's instructions. The sections were counterstained with Mayer's hematoxylin and then coverslipped with Faramount aqueous mounting medium (Dako). 
The immunohistochemical staining for the selected proteins was evaluated semiquantitatively by two independent observers. Scoring of the percentage of cells showing specific immunoreactivity and of the intensity of this immunoreactivity was carried out as previously described. 26 Semiquantitative scores were assigned as follows: Assessment of overall positivity: score 0, no tumor cells showing positive staining; score 1, 1% to 24% of tumor cells showing positive staining; score 2, 25% to 49% of tumor cells showing positive staining; score 3, 50% to 74% of tumor cells showing positive staining; score 4, 75% to 100% of tumor cells showing positive staining. Intensity of immunohistochemistry staining was scored as 0 = none; 1 = weak; 2 = moderate; 3 = strong. A combined score was obtained by multiplying the scores for the overall positivity and the intensity of stained tumor cells, giving a maximum score of 12. 20,27 Immunohistochemistry scores were subjected to statistical analysis using Student's t-tests (two-tailed, two-sample unequal variance). 
Cell Culture
The 92.1 primary uveal melanoma cell line 28 was obtained from the European Searchable Tumour Line Database (ESTDAB, University of Tübingen, Center for Medical Research, Germany) and cultured in RPMI-1640 (Sigma-Aldrich) supplemented with 10% fetal bovine serum (PAA Laboratories GmbH, Pasching, Austria). Cells were cultured without antibiotics and were incubated at 37°C and 5% CO2. The cell line was free of Mycoplasma contamination, as tested with the indirect Hoechst staining method. 
Western Blotting
Protein samples were prepared in Laemmli sample buffer (Sigma-Aldrich), heated at 95°C for 5 minutes, and cooled on ice prior to loading onto 4% to 12% NuPAGE Bis-Tris gels (Invitrogen). Electrophoretic transfer, blocking, and development of Western blots were carried out as described previously. 23 The following primary antibodies were used: rabbit polyclonal anti-TPI1 (dilution 1:4000, vol/vol; GeneTex, Inc.); rabbit polyclonal anti-FABP3 (dilution 1:500, vol/vol; Abcam); mouse monoclonal anti-glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (clone 6C5; dilution 1:10,000, vol/vol; Abcam). Bound antibodies were detected using horseradish peroxidase conjugated secondary antibodies (anti-mouse or anti-rabbit; dilution 1:2000, vol/vol; Dako). 
Transient siRNA Transfection
92.1 uveal melanoma cells (7.5 × 104 cells/well) were transfected using independent predesigned Silencer Select siRNA molecules for knockdown of FABP3 (#s4973, #s4974) and TPI1 (#s14339, #s224747) (Applied Biosystems, Paisley, UK) at a final concentration of 20 nM. Scrambled siRNA (#AM4390843) (Applied Biosystems) treated cells were considered as control. Solutions of siRNAs were prepared in OptiMEM (Invitrogen), and transfections were carried out using Lipofectamine RNAiMAX (Invitrogen Ltd., Carlsbad, CA) following the manufacturer's instructions. After 6 hours, the culture medium was replaced with fresh medium. After a total of 72 hours of transfection, the cells were seeded into invasion and migration chambers for 48 hours to assess the effect of siRNA transfection on invasion and motility of the cell line. At the same time cells were also collected for protein extraction to show protein knockdown by Western blot analysis as a result of siRNA transfection. 
Invasion and Motility Assays
Invasion assays were carried out according to the method previously described. 29 Briefly, matrigel (BD Biosciences, Bedford, MA) was diluted to 1 mg/mL in serum-free RPMI-1640. Boyden chambers (8-μm pore size, BD Biosciences) were coated with 100 μL of matrigel (BD Biosciences) and incubated overnight at 4°C. Cell suspensions were prepared in serum-free basal media at a concentration of 2 × 105 cells/mL. Seven hundred fifty microliters of complete media was added to the lower chamber of the insert in the 24-well plate. A volume of 500 μL of the cell suspension was then added into the insert. The invasion assays were then incubated for 48 hours at 37°C and 5% CO2. After incubation, the noninvading cells were removed by wiping the inner side of the insert with a PBS-soaked cotton swab. The outer side of the insert was then stained with 0.25% crystal violet for 10 minutes. Excess stain was rinsed off the inserts with sterile water and allowed to dry. Motility assays were carried out in a similar manner, except that uncoated 8-μm inserts were used. To determine the total number of invading or migrating cells, the number of cells/field in 15 fields was counted at 200× magnification. The average number of cells per field was then multiplied by a conversion factor of 140 (growth area of membrane divided by field of viewed area at 200× magnification [calibrated using a microscope graticule). All assays were subjected to statistical analysis using Student's t-tests (two-tailed, two-sample unequal variance). 
Results
Patients
Uveal melanoma tissue from a total of 27 Irish patients, 16 females and 11 males, with a median age of 62 years at the time of diagnosis (ranging from 32 to 86 years) were obtained by enucleation at the time of diagnosis and included in our 2D DIGE and pilot immunohistochemical study. Clinical and histopathological characteristics of the patients included in the study are outlined in Table 1. Ten patients subsequently developed metastatic disease; the metastatic sites were the liver (six cases), lungs (two cases), kidney (one case), and both liver and skin (one case). According to the modified Callender classification, 30 of the 10 tumors that subsequently metastasized, 4 were of the spindle cell type, 3 of the epitheloid, and 3 were of the mixed cell type. The histological classification of the 17 tumors that did not metastasize revealed 6 cases as being of the spindle type, 4 as epitheloid, and 7 as mixed cell type. The median of the largest tumor diameter was 11.5 mm (ranging from 8 to 20 mm) for the uveal melanomas that did metastasize and 14 mm (ranging from 8 to 23 mm) for the tumors that did not. Five of the tumors that metastasized and one of the nonmetastasized tumors arose from the ciliary body. Extrascleral extension was found in uveal melanomas of three patients who developed metastatic disease and in one who did not. The median mitotic cell count was 5 per 40 high-power fields (HPF) in the uveal melanomas that did metastasize (ranging from none to 16 mitotic figures) and 4 per 40 HPF (ranging from none to 20 mitotic figures) in those that did not metastasize as determined on hematoxylin and eosin sections. Chromosome 3 status was available for 14 of the 17 patients who did not develop metastatic disease and showed disomy in 12 and trisomy in 2 cases. For the 10 patients who did develop metastatic disease, loss of heterozygosity for chromosome 3 was observed in 8 patients. These results are in agreement with previous studies showing that loss of chromosome 3 heterozygosity results in an increased risk of metastasis. 1214 Two of the patients who developed metastatic disease were found to have disomy 3. This could be possibly due to FISH having a relatively low sensitivity in detecting partial deletions of chromosome 3. Cytogenetic analysis using Multiplex Ligation-Dependent Probe Amplification has been found to be more accurate and sensitive than FISH in detecting monosomy 3. 12  
Identification of Differentially Expressed Proteins in Metastasized Compared to Nonmetastasized Uveal Melanoma
Protein lysates of uveal melanoma tissues from the two patient groups (patients with primary uveal melanomas that did not metastasize and patients with primary uveal melanomas that were subsequently found to have metastasized) were differentially labeled with either Cy3 or Cy5 dyes and separated by 2D DIGE. A pooled mixture of all samples used in the experiment was labeled with Cy2 dye and served as an internal control. Image analysis using Progenesis SameSpots software was performed to compare the differential expression of proteins between the nonmetastasized and subsequently metastasized uveal melanoma groups. Fourteen protein spots were found to be differentially expressed between the two groups that passed a t-test score <0.05 and a fold change of ±1.2. These proteins were identified using mass spectrometry. The list of identified differentially expressed proteins is outlined in Table 2. Their location on a representative 2D DIGE gel is shown in Figure 1
Table 2. 
 
List of Differentially Regulated Proteins from the Comparison of Primary Uveal Melanomas That Were Found To Have Subsequently Metastasized versus Primary Uveal Melanomas That Did Not Metastasize
Table 2. 
 
List of Differentially Regulated Proteins from the Comparison of Primary Uveal Melanomas That Were Found To Have Subsequently Metastasized versus Primary Uveal Melanomas That Did Not Metastasize
Spot No.* Protein Accession No.† Gene Name Protein Name MWt (Da) pI % Coverage MASCOT Score Average Ratio‡ t-test Previously Published§
1 P30101 PDIA3 Protein disulfide-isomerase A3 precursor 57,146 5.98 37 3396 1.5 0.011
2 P08670 VIM Vimentin 53,676 5.06 50 1445 1.8 0.007 32
P06865 HEXA Beta-hexosaminidase subunit alpha 61,106 5.04 24 525 1.8 0.007 18
3 Q13228 SELENBP1 Selenium-binding protein 1 52,907 6.13 39 2426 1.3 0.044
4 P06733 ENO1 Alpha-enolase 47,481 7.01 53 4944 1.4 0.007 18
5 P05198 EIF2S1 Eukaryotic translation initiation factor 2 subunit 1 36,374 5.02 57 1648 −1.6 0.035
6 P52907 CAPZA1 F-actin capping protein subunit alpha-1 33,073 5.45 50 2522 1.3 0.028
7 P25788 PSMA3 Proteasome subunit alpha type 3 28,643 5.19 26 1267 −1.2 0.026
8 P08865 RPSA 40S ribosomal protein SA 32,947 4.79 30 861 −1.4 0.023 20
9 P30040 ERP29 Endoplasmic reticulum protein ERp29 29,032 6.77 34 2592 1.4 0.040
10 P60174 TPI1 Triosephosphate isomerase 26,938 6.45 61 2987 1.6 0.00057
11 P60174 TPI1 Triosephosphate isomerase 26,938 6.45 30 317 1.7 0.00009
12 Q99497 PARK7 Protein DJ-1 20,050 6.33 34 2037 1.2 0.018
13 P07437 TUBB Tubulin beta chain 50,095 4.78 10 587 −1.7 0.017
14 P68363 TUBA1B Tubulin alpha-1B chain 50,804 4.94 13 553 −1.9 0.006
15 P05413 FABP3 Fatty acid-binding protein, heart 14,906 6.29 51 1790 2.2 0.00035
Figure 1. 
 
Representative 2D DIGE gel image of Cy2-labeled pool of protein lysates from primary uveal melanomas that were subsequently found to have metastasized and the primary uveal melanomas that did not metastasize. Differentially expressed proteins that have been successfully identified by LC/MS-MS are represented on the gel. Proteins are numbered for clarity and are outlined in Table 2.
Figure 1. 
 
Representative 2D DIGE gel image of Cy2-labeled pool of protein lysates from primary uveal melanomas that were subsequently found to have metastasized and the primary uveal melanomas that did not metastasize. Differentially expressed proteins that have been successfully identified by LC/MS-MS are represented on the gel. Proteins are numbered for clarity and are outlined in Table 2.
Figure 2. 
 
Progenesis SameSpots software output showing two representative differentially expressed protein spots. Enlarged regions of 2D DIGE representative gels of protein lysates from primary uveal melanomas that did not metastasize (left) and primary uveal melanomas that were found to have subsequently metastasized (right). (A) Spot no. 6 was identified as F-actin capping protein subunit alpha-1 (CAPZA1). Average ratio = 1.3; t-test = 0.028. (B) Spot no. 15 was identified as fatty acid–binding protein, heart type (FABP3). Average ratio = 2.2; t-test = 0.00035.
Figure 2. 
 
Progenesis SameSpots software output showing two representative differentially expressed protein spots. Enlarged regions of 2D DIGE representative gels of protein lysates from primary uveal melanomas that did not metastasize (left) and primary uveal melanomas that were found to have subsequently metastasized (right). (A) Spot no. 6 was identified as F-actin capping protein subunit alpha-1 (CAPZA1). Average ratio = 1.3; t-test = 0.028. (B) Spot no. 15 was identified as fatty acid–binding protein, heart type (FABP3). Average ratio = 2.2; t-test = 0.00035.
In our study, the following proteins were found to show increased expression in primary uveal melanomas that metastasized compared to nonmetastasized uveal melanomas: protein disulphide-isomerase A3 precursor (PDIA3), selenium-binding protein 1 (SELENBP1), alpha-enolase, F-actin capping protein subunit alpha-1 (CAPZA1), endoplasmic reticulum protein ERp29 precursor, TPI1, protein DJ-1 (PARK7), and FABP3. Protein identification of spot number 2 yielded two protein identifications, vimentin and beta-hexosaminidase subunit alpha; vimentin has previously been found to be associated with the metastatic phenotype of uveal melanoma. 31 Eukaryotic translation initiation factor 2 subunit 1, proteasome subunit alpha type 3, 40S ribosomal protein SA, tubulin beta chain, and tubulin alpha-1B chain were shown to have decreased expression in uveal melanoma tissues of patients who subsequently developed metastatic disease. 
Follow-Up of Selected Targets by Immunohistochemistry
For follow-up of the 2D DIGE study, immunohistochemical analysis was performed on eight nonmetastasized uveal melanoma tissue specimens and eight uveal melanoma specimens of patients who did subsequently develop metastatic disease (see Table 1 for clinical and histopathological features of the patients included in the study; note that two samples differ from the 2D DIGE study due to sample availability). Uveal melanoma tissue sections from the two patient groups were analyzed for the expression of FABP3, TPI1, CAPZA1, PDIA3, SELENBP1, and PARK7. For semiquantitative immunohistochemical analysis, a combined score was obtained by multiplying the scores for the overall positivity and the intensity of stained tumor cells, giving a maximum score of 12. 20,21  
FABP3 Immunoreactivity
FABP3 showed positive cytoplasmic and membranous staining in 15 of the 16 of the uveal melanoma cases that were studied. Increased reactivity for FABP3 was observed in the majority of uveal melanomas that were found to have subsequently metastasized compared to those that did not (mean combined score, 8.5 ± 5.1 vs. 5 ± 3.7; P = 0.15) (Fig. 3A, Table 3). 
Figure 3. 
 
Representative images of the immunohistochemical analysis of (A) FABP3 and (B) TPI1 in primary uveal melanomas that did not metastasize (left) and in primary uveal melanomas that have subsequently been found to have metastasized (right). Magnification ×400. Scale bar = 100 μm.
Figure 3. 
 
Representative images of the immunohistochemical analysis of (A) FABP3 and (B) TPI1 in primary uveal melanomas that did not metastasize (left) and in primary uveal melanomas that have subsequently been found to have metastasized (right). Magnification ×400. Scale bar = 100 μm.
Table 3.  
 
Expression Pattern of Proteins in the Immunohistochemical Study
Table 3.  
 
Expression Pattern of Proteins in the Immunohistochemical Study
Sample CAPZA1 FABP3 PARK7 PDIA3 SELENBP1 TPI1
Pos Int Score Pos Int Score Pos Int Score Pos Int Score Pos Int Score Pos Int Score
1 90 1 4 30 1 2 70 1 3 55 2 6 40 1 2 40 2 4
2 75 2 8 70 2 6 95 2 8 20 2 2 100 1 4 100 3 12
3 25 1 2 90 3 12 95 1 4 60 2 6 30 3 6 80 2 8
4 40 1 2 90 3 12 95 1 4 35 2 4 0 0 0 95 1 4
5 75 1 4 0 0 0 80 3 12 95 3 12 95 2 8 80 3 12
6 10 2 2 80 3 12 25 1 2 60 2 6 0 0 0 80 3 12
7 40 2 4 100 3 12 90 2 8 30 2 4 0 0 0 45 3 6
8 90 2 8 90 3 12 95 3 12 35 1 2 20 3 3 75 3 12
11 75 3 12 50 1 3 55 1 3 95 3 12 10 3 3 90 3 12
12 0 0 0 5 2 2 40 1 2 95 3 12 0 0 0 20 2 2
13 75 3 12 95 2 8 95 2 8 80 1 4 20 1 1 0 0 0
14 90 3 12 40 1 2 95 2 8 95 2 8 90 2 8 90 3 12
15 90 2 8 90 3 12 80 2 8 80 2 8 0 0 0 80 2 8
16 90 3 12 80 2 8 60 1 3 70 2 6 25 3 6 100 2 8
17 10 1 1 5 3 3 0 0 0 75 2 8 0 0 0 0 0 0
18 75 2 8 50 1 3 80 2 8 35 1 2 0 0 0 100 2 8
TPI1 Immunoreactivity
TPI1 protein showed expression in the cytoplasm and the nucleus in the majority of the cases analyzed (Fig. 3B; Table 3). Increased positivity for TPI1 was observed in uveal melanomas that were found to have subsequently metastasized compared to those that did not (mean combined score, 8.8 ± 3.7 vs. 6.3 ± 4.9; P = 0.27). Two out of eight of the nonmetastasized uveal melanomas were found to be completely negative for TPI1 protein expression. 
CAPZA1 Immunoreactivity
CAPZA1 showed an overall reduced cytoplasmic expression pattern in the majority of the uveal melanomas that were found to have subsequently metastasized compared to those that did not (mean combined score, 4 ± 2.5 vs. 10 ± 5; P = 0.08) (Figs. 4A, 4B; Table 3). Parallel staining of sections for CAPZA1 and the histiocyte/monocyte/macrophage marker CD68 confirmed CAPZA1 positivity in both tumor cells and tumor-infiltrating macrophages, but not all of those macrophages stained positive for CAPZA1 (Figs. 4C, 4D; parallel staining of CAPZA1 and CD68 staining was carried out using serial sections). 
Figure 4. 
 
Representative images of the immunohistochemical analysis of (A) CAPZA1 expression in primary uveal melanomas that did not metastasize and of (B) CAPZA1 expression in primary uveal melanomas that have subsequently been found to have metastasized (magnification ×400; scale bar = 100 μm). Serial uveal melanoma sections stained with (C) CAPZA1 and (D) CD68 indicate that CAPZA1 is expressed in tumor cells and in a number of tumor-infiltrating macrophages. Magnification ×200. Scale bar = 200 μm.
Figure 4. 
 
Representative images of the immunohistochemical analysis of (A) CAPZA1 expression in primary uveal melanomas that did not metastasize and of (B) CAPZA1 expression in primary uveal melanomas that have subsequently been found to have metastasized (magnification ×400; scale bar = 100 μm). Serial uveal melanoma sections stained with (C) CAPZA1 and (D) CD68 indicate that CAPZA1 is expressed in tumor cells and in a number of tumor-infiltrating macrophages. Magnification ×200. Scale bar = 200 μm.
PDIA3 Immunoreactivity
PDIA3 (Fig. 5A) showed decreased expression in uveal melanomas that were found to have subsequently metastasized compared to those that did not (mean combined score, 5.3 ± 3.2 vs. 7.5 ± 3.5; P = 0.20) (Table 3). 
Figure 5. 
 
Representative images of the immunohistochemical analysis of (A) PDIA3, (B) SELENBP1, and (C) PARK7 in primary uveal melanomas that did not metastasize. Magnification ×400. Scale bar = 100 μm.
Figure 5. 
 
Representative images of the immunohistochemical analysis of (A) PDIA3, (B) SELENBP1, and (C) PARK7 in primary uveal melanomas that did not metastasize. Magnification ×400. Scale bar = 100 μm.
SELENBP1 Immunoreactivity
SELENBP1 (Fig. 5B) showed only weak or no cytoplasmic expression in the majority of both nonmetastasized and subsequently metastasized uveal melanoma groups (mean combined score, 2.9 ± 3.2 vs. 2.3 ± 3; P = 0.69) (Table 3). 
PARK7 Immunoreactivity
For PARK7 (Fig. 5C), positive, heterogeneous staining was observed in both nonmetastasized and subsequently metastasized uveal melanoma groups (mean combined score, 5 ± 3.3 vs. 6.6 ± 4; P = 0.39) (Table 3). 
Down-Regulation of FABP3 or TPI1 Decreases Invasion and Motility of 92.1 Primary Uveal Melanoma Cells
To study the potential effect of FABP3 or TPI1 on invasion and motility of uveal melanoma cells, preliminary down-regulation experiments using siRNA molecules directed against respective target genes were performed in the 92.1 primary uveal melanoma cell line. 
Two independent siRNA molecules were used to silence the FABP3 or TPI1 genes, and the subsequent reduction in protein levels were confirmed by Western blot analysis (Figs. 6A, 6B, i). Following siRNA transfection, invasion and migration assays were carried out. The total number of 92.1 cells invading through the membrane was significantly decreased following transfection with siRNA molecules specific for FABP3 (Fig. 6A, ii). A similar reduction in invasion was obtained following transfection with siRNA molecules specific for TPI1 (Fig. 6B, ii). The reduction of either FABP3 or TPI1 also led to a significant decrease in motility of 92.1 cells (Figs. 6A, 6B, iii). 
Figure 6. 
 
siRNA knockdown of (A) FABP3 and (B) TPI1 decreases invasion and motility. (i) Western blot analysis showing the reduction of (A) FABP3 and (B) TPI1 following siRNA knockdown. Control cells (untreated), transfection reagent treated cells (Lipo), scrambled siRNA treated cells, siRNA #1 and siRNA #2 specific for (A) FABP3 or (B) TPI1 treated cells were lysed 72 hours after transfection and subjected to Western blot analysis. GAPDH served as an internal loading control. (A, B, ii) Invasion assays of 92.1 cells following siRNA transfection. The total number of 92.1 cells invading through membrane of invasion chambers after siRNA transfection is shown. (A, B, iii) Motility assays of 92.1 cells following siRNA transfection. The total number of 92.1 cells migrating through uncoated membrane after siRNA transfection is shown. One representative result out of three independent experiments is shown, and error bars represent standard deviation from data obtained from two technical repeats. *P < 0.05; **P < 0.01 compared with scrambled siRNA treated controls.
Figure 6. 
 
siRNA knockdown of (A) FABP3 and (B) TPI1 decreases invasion and motility. (i) Western blot analysis showing the reduction of (A) FABP3 and (B) TPI1 following siRNA knockdown. Control cells (untreated), transfection reagent treated cells (Lipo), scrambled siRNA treated cells, siRNA #1 and siRNA #2 specific for (A) FABP3 or (B) TPI1 treated cells were lysed 72 hours after transfection and subjected to Western blot analysis. GAPDH served as an internal loading control. (A, B, ii) Invasion assays of 92.1 cells following siRNA transfection. The total number of 92.1 cells invading through membrane of invasion chambers after siRNA transfection is shown. (A, B, iii) Motility assays of 92.1 cells following siRNA transfection. The total number of 92.1 cells migrating through uncoated membrane after siRNA transfection is shown. One representative result out of three independent experiments is shown, and error bars represent standard deviation from data obtained from two technical repeats. *P < 0.05; **P < 0.01 compared with scrambled siRNA treated controls.
Proliferation assays were also carried out following siRNA transfection, and revealed no significant effect on proliferation in cells treated with specific siRNA versus scrambled siRNA treated cells (results not shown). 
Discussion
This is one of the first proteomic studies using primary uveal melanoma tumors in order to understand how the biology differs between uveal melanomas that metastasize and those that do not and to identify possible protein biomarkers that may predict metastasis. Our study successfully identified 14 statistically significant differentially expressed proteins, as outlined in Table 2, from which four proteins (vimentin, beta-hexosaminidase subunit alpha, alpha-enolase, and 40S ribosomal protein SA) have been identified in the context of the metastatic phenotype of uveal melanoma in previous studies. 18,19,31  
Our 2D DIGE analysis revealed a 2.2-fold up-regulation of FABP3 in uveal melanomas that did subsequently metastasize; this trend was confirmed by our immunohistochemical studies. FABP3 is a member of a multigene family including nine FABPs and the cellular retinoid binding proteins. FABPs show a broad tissue distribution, 32 and tissues can contain more than one FABP. 33 Though FABPs are known to be involved in intracellular transport of long-chain fatty acids, 34 their in vivo functions are poorly understood. FABP3 was shown to inhibit cell proliferation in mammary epithelial cells, and the ectopic expression of FABP3 in breast cancer cells led to reduced tumorigenicity in nude mice. 35 In contrast, positive correlations of FABP3 expression with tumor cell invasion, lymph node metastasis, and poor patient survival have been found in gastric carcinomas. 36 Our preliminary functional studies indicate that FABP3 plays a potential role in invasion of uveal melanoma in vitro, as down-regulation of FABP3 using specific siRNAs targeting FABP3 resulted in a significant reduction of invasion and migration of uveal melanoma cells. 
TPI1 was identified twice with a 1.6- and 1.7-fold up-regulation in uveal melanoma samples of patients who subsequently developed metastatic disease; immunohistochemical analysis also showed a trend for up-regulation of TPI1 in uveal melanomas that did metastasize. The double identification of TPI1 as an up-regulated protein is very likely due to its isoforms. 37,38 The glycolytic enzyme TPI1 catalyzes the conversion of dihydroxyacetone phosphate to glyceraldehyde 3-phosphate, 39 and a high rate of glycolysis is required to support tumor growth. 40 TPI1 has previously been shown to be expressed in uveal melanoma primary cell cultures 41 ; our proteomics study provides evidence that TPI1 may be involved in the development of the metastatic phenotype in uveal melanoma. This observation is supported further by our functional studies using siRNA molecules specific for TPI1, where knockdown of TPI1 led to a significant decrease in invasion and motility of uveal melanoma cells in vitro. TPI1 has previously been shown to be involved in the aggressiveness of other malignancies such as breast cancer. 42  
To date, very little is known about the role of the cytoskeletal protein CAPZA1 and any association with cancer. Our immunohistochemical study revealed decreased expression of CAPZA1 in uveal melanomas that did subsequently metastasize compared to those that did not. Previously, CAPZA1 was found to be differentially expressed in renal cell carcinoma compared to normal renal cells, suggesting its possible involvement in tumorigenesis. 43 Functional analysis would be required to confirm a possible involvement of CAPZA1 in the metastatic phenotype of uveal melanoma. 
PDIA3, SELENBP1, and PARK7 also showed statistically significant differential expression in our 2D DIGE study. PDIA3 possibly plays a role in the malignant transformation of esophageal carcinoma. 44 Its expression was also shown in uveal melanoma cell lines. 41 SELENBP1 is suggested to be a tumor suppressor, as its expression is lost in several epithelial cancers. 45 Nagakubo et al. 46 identify PARK7 as a mitogen-dependent oncogene. Its expression is correlated with the transformation and progression of esophageal squamous cell carcinoma. 47 A recent study, Bande et al., 48 finds that elevated serum levels of PARK7 are associated with choroidal nevi transformation risk factors. However, an up-regulation of CAPZA1, PDIA3, SELENBP1, and PARK7 during metastasis, as obtained within our 2D DIGE study, is not reflected in our immunohistochemical analysis. 
In summary, the trend of the expression observed by our immunohistochemical analysis is in agreement with 2D DIGE analysis carried out in the case of FABP3 and TPI1, but not for CAPZA1, PDIA3, SELENBP1, and PARK7. There are several possible reasons for such a lack of correlation. First of all, whole tumor homogenates were used for our 2D DIGE study, so that total than rather localized protein expression was examined. On the other hand, immunohistochemistry examines formalin-fixed, paraffin-embedded tissue, whereas the proteins for 2D DIGE were extracted from snap-frozen tissue, stored at −80°C prior to use. In cases in which the data from the two methods did not agree, it is not possible to determine which result better reflects the protein expression in the tumor. However, the two very different approaches yielded concurrent results for FABP3 and TPI1, which strongly supports the differential expression of both proteins. 
In conclusion, our proteomics study on primary uveal melanoma tissue samples led to successful identification of proteins including FABP3 and TPI1 that are possibly involved in the metastatic phenotype of uveal melanoma. These two proteins showed not only a strong differential expression and high statistical significance from the 2D DIGE analysis, but also their knockdown using specific siRNA significantly reduced invasion, suggesting that they are functionally involved the metastatic phenotype of uveal melanoma. 
For validation of those promising results, we are currently completing our functional studies on the other proteins identified and are expanding our immunohistochemical analysis in a larger cohort of patients. Analysis of a larger number of cases would also allow us to perform comprehensive statistical testing and carry out multivariate analysis on combinations of differentially expressed proteins, which would possibly have potential in stratifying patients in terms of priority for intense surveillance and possible resection of early metastasis. Identification of proteins associated with metastasis may lead to the development of new therapeutic targets and could result in improved treatment for metastatic uveal melanoma. 
References
Singh AD Topham A . Survival rates with uveal melanoma in the United States: 1973–1997. Ophthalmology . 2003;110:962–965. [CrossRef] [PubMed]
Virgili G Gatta G Ciccolallo L Incidence of uveal melanoma in Europe. Ophthalmology . 2007;114:2309–2315. [CrossRef] [PubMed]
Singh AD Rennie IG Seregard S Giblin M McKenzie J . Sunlight exposure and pathogenesis of uveal melanoma. Surv Ophthalmol . 2004;49:419–428. [CrossRef] [PubMed]
Gamel JW McLean IW McCurdy JB . Biologic distinctions between cure and time to death in 2892 patients with intraocular melanoma. Cancer . 1993;71:2299–2305. [CrossRef] [PubMed]
Lorigan JG Wallace S Mavligit GM . The prevalence and location of metastases from ocular melanoma: imaging study in 110 patients. AJR Am J Roentgenol . 1991;157:1279–1281. [CrossRef] [PubMed]
Eskelin S Pyrhonen S Summanen P Prause JU Kivela T . Screening for metastatic malignant melanoma of the uvea revisited. Cancer . 1999;85:1151–1159. [CrossRef] [PubMed]
Koutsandrea C Moschos MM Dimissianos M Georgopoulos G Ladas I Apostolopoulos M . Metastasis rates and sites after treatment for choroidal melanoma by proton beam irradiation or by enucleation. Clin Ophthalmol . 2008;2:989–995. [PubMed]
Onken MD Worley LA Ehlers JP Harbour JW . Gene expression profiling in uveal melanoma reveals two molecular classes and predicts metastatic death. Cancer Res . 2004;64:7205–7209. [CrossRef] [PubMed]
Tschentscher F Hüsing J Hölter T Tumor classification based on gene expression profiling shows that uveal melanomas with and without monosomy 3 represent two distinct entities. Cancer Res . 2003;63:2578–2584. [PubMed]
Zuidervaart W van der Velden PA Hurks MH Gene expression profiling identifies tumour markers potentially playing a role in uveal melanoma development. Br J Cancer . 2003;89:1914–1919. [CrossRef] [PubMed]
Laurent C Valet F Planque N High PTP4A3 phosphatase expression correlates with metastatic risk in uveal melanoma patients. Cancer Res . 2011;71:666–674. [CrossRef] [PubMed]
Damato B Dopierala J Klaasen A van Dijk M Sibbring J Coupland SE . Multiplex ligation-dependent probe amplification of uveal melanoma: correlation with metastatic death. Invest Ophthalmol Vis Sci . 2009;50:3048–3055. [CrossRef] [PubMed]
Hausler T Stang A Anastassiou G Loss of heterozygosity of 1p in uveal melanomas with monosomy 3. Int J Cancer . 2005;116:909–913. [CrossRef] [PubMed]
Lake SL Coupland SE Taktak AF Damato BE . Whole-genome microarray detects deletions and loss of heterozygosity of chromosome 3 occurring exclusively in metastasizing uveal melanoma. Invest Ophthalmol Vis Sci . 2010;51:4884–4891. [CrossRef] [PubMed]
Harbour JW Onken MD Roberson ED Frequent mutation of BAP1 in metastasizing uveal melanomas. Science . 2010;330:1410–1413. [CrossRef] [PubMed]
Landreville S Agapova OA Matatall KA Histone deacetylase inhibitors induce growth arrest and differentiation in uveal melanoma. Clin Cancer Res . 2012;18:408–416. [CrossRef] [PubMed]
Asnaghi L Ebrahimi KB Schreck KC Notch signaling promotes growth and invasion in uveal melanoma. Clin Cancer Res . 2012;18:654–665. [CrossRef] [PubMed]
Zuidervaart W Hensbergen PJ Wong MC Proteomic analysis of uveal melanoma reveals novel potential markers involved in tumor progression. Invest Ophthalmol Vis Sci . 2006;47:786–793. [CrossRef] [PubMed]
Pardo M Garcia A Thomas B The characterization of the invasion phenotype of uveal melanoma tumour cells shows the presence of MUC18 and HMG-1 metastasis markers and leads to the identification of DJ-1 as a potential serum biomarker. Int J Cancer . 2006;119:1014–1022. [CrossRef] [PubMed]
Coupland SE Vorum H Mandal N Proteomics of uveal melanomas suggests HSP-27 as a possible surrogate marker of chromosome 3 loss. Invest Ophthalmol Vis Sci . 2010;51:12–20. [CrossRef] [PubMed]
Jmor F Kalirai H Taktak A Damato B Coupland SE . HSP-27 protein expression in uveal melanoma: correlation with predicted survival Acta Ophthalmologica . 2010;doi:10.1111/j.1755-3768.2010.02038.x
Patel M Smyth E Chapman PB Therapeutic implications of the emerging molecular biology of uveal melanoma. Clin Cancer Res . 2011;17:2087–2100. [CrossRef] [PubMed]
Meleady P Henry M Gammell P Proteomic profiling of CHO cells with enhanced rhBMP-2 productivity following co-expression of PACEsol. Proteomics . 2008;8:2611–2624. [CrossRef] [PubMed]
Meleady P Doolan P Henry M Sustained productivity in recombinant Chinese hamster ovary (CHO) cell lines: proteome analysis of the molecular basis for a process-related phenotype. BMC Biotechnol . 2011;11:78. [CrossRef] [PubMed]
Perkins DN Pappin DJ Creasy DM Cottrell JS . Probability-based protein identification by searching sequence databases using mass spectrometry data. Electrophoresis . 1999;20:3551–3567. [CrossRef] [PubMed]
Larkin A O'Driscoll L Kennedy S Investigation of MRP-1 protein and MDR-1 P-glycoprotein expression in invasive breast cancer: a prognostic study. Int J Cancer . 2004;112:286–294. [CrossRef] [PubMed]
Ghadjar P Coupland SE Na IK Chemokine receptor CCR6 expression level and liver metastases in colorectal cancer. J Clin Oncol . 2006;24:1910–1916. [CrossRef] [PubMed]
De Waard-Siebinga I Blom DJ Griffioen M Establishment and characterization of an uveal-melanoma cell line. Int J Cancer . 1995;62:155–161. [CrossRef] [PubMed]
Albini A Iwamoto Y Kleinman HK A rapid in vitro assay for quantitating the invasive potential of tumor cells. Cancer Res . 1987;47:3239–3245. [PubMed]
McLean IW Foster WD Zimmerman LE Gamel JW . Modifications of Callender's classification of uveal melanoma at the Armed Forces Institute of Pathology. Am J Ophthalmol . 1983;96:502–509. [CrossRef] [PubMed]
Hendrix MJ Seftor EA Seftor RE Biologic determinants of uveal melanoma metastatic phenotype: role of intermediate filaments as predictive markers. Lab Invest . 1998;78:153–163. [PubMed]
Storch J Corsico B . The emerging functions and mechanisms of mammalian fatty acid-binding proteins. Annu Rev Nutr . 2008;28:73–95. [CrossRef] [PubMed]
Storch J McDermott L . Structural and functional analysis of fatty acid-binding proteins. J Lipid Res . 2009;50 (suppl) S126–S131. [PubMed]
Ockner RK Manning JA . Fatty acid-binding protein in small intestine. Identification, isolation, and evidence for its role in cellular fatty acid transport. J Clin Invest . 1974;54:326–338. [CrossRef] [PubMed]
Yang Y Spitzer E Kenney N Members of the fatty acid binding protein family are differentiation factors for the mammary gland. J Cell Biol . 1994;127:1097–1109. [CrossRef] [PubMed]
Hashimoto T Kusakabe T Sugino T Expression of heart-type fatty acid-binding protein in human gastric carcinoma and its association with tumor aggressiveness, metastasis and poor prognosis. Pathobiology . 2004;71:267–273. [CrossRef] [PubMed]
Snapka RM Sawyer TH Barton RA Gracy RW . Comparison of the electrophoretic properties of triosephosphate isomerases of various tissues and species. Comp Biochem Physiol B . 1974;49:733–741. [PubMed]
Kester MV Gracy RW . Alteration of human lymphocyte triosephosphate isomerase isozymes during blastogenesis. Biochem Biophys Res Commun . 1975;65:1270–1277. [CrossRef] [PubMed]
Albery WJ Knowles JR . Free-energy profile of the reaction catalyzed by triosephosphate isomerase. Biochemistry . 1976;15:5627–5631. [CrossRef] [PubMed]
Bui T Thompson CB . Cancer's sweet tooth. Cancer Cell . 2006;9:419–420. [CrossRef] [PubMed]
Pardo M Garcia A Thomas B Proteome analysis of a human uveal melanoma primary cell culture by 2-DE and MS. Proteomics . 2005;5:4980–4993. [CrossRef] [PubMed]
Selicharova I Sanda M Mladkova J 2-DE analysis of breast cancer cell lines 1833 and 4175 with distinct metastatic organ-specific potentials: comparison with parental cell line MDA-MB-231. Oncol Rep . 2008;19:1237–1244. [PubMed]
Kellner R Lichtenfels R Atkins D Targeting of tumor associated antigens in renal cell carcinoma using proteome-based analysis and their clinical significance. Proteomics . 2002;2:1743–1751. [CrossRef] [PubMed]
Qi YJ He QY Ma YF Proteomic identification of malignant transformation-related proteins in esophageal squamous cell carcinoma. J Cell Biochem . 2008;104:1625–1635. [CrossRef] [PubMed]
Scortegagna M Martin RJ Kladney RD Neumann RG Arbeit JM . Hypoxia-inducible factor-1alpha suppresses squamous carcinogenic progression and epithelial-mesenchymal transition. Cancer Res . 2009;69:2638–2646. [CrossRef] [PubMed]
Nagakubo D Taira T Kitaura H DJ-1, a novel oncogene which transforms mouse NIH3T3 cells in cooperation with ras. Biochem Biophys Res Commun . 1997;231:509–513. [CrossRef] [PubMed]
Yuen HF Chan YP Law S DJ-1 could predict worse prognosis in esophageal squamous cell carcinoma. Cancer Epidemiol Biomarkers Prev . 2008;17:3593–3602. [CrossRef] [PubMed]
Bande MF Santiago M Blanco MJ Serum DJ-1/PARK 7 is a potential biomarker of choroidal nevi transformation. Invest Ophthalmol Vis Sci . 2012;53:62–67. [CrossRef] [PubMed]
Footnotes
 Supported by a research fellowship from the German Research Foundation (GZ Li1900/1-1) (ALi) and funding from Royal Victoria Eye and Ear Hospital Research Foundation; proteomics equipment provided by Science Foundation Ireland and the Higher Education Authority Programme for Research and Third Level Institutions.
Footnotes
 Disclosure: A. Linge, None; S. Kennedy, None; D. O'Flynn, None; S. Beatty, None; P. Moriarty, None; M. Henry, None; M. Clynes, None; A. Larkin, None; P. Meleady, None
Footnotes
2  These authors contributed equally to this study and should be considered joint first authors.
Footnotes
6  These authors contributed equally to this study and should be considered equivalent authors.
Figure 1. 
 
Representative 2D DIGE gel image of Cy2-labeled pool of protein lysates from primary uveal melanomas that were subsequently found to have metastasized and the primary uveal melanomas that did not metastasize. Differentially expressed proteins that have been successfully identified by LC/MS-MS are represented on the gel. Proteins are numbered for clarity and are outlined in Table 2.
Figure 1. 
 
Representative 2D DIGE gel image of Cy2-labeled pool of protein lysates from primary uveal melanomas that were subsequently found to have metastasized and the primary uveal melanomas that did not metastasize. Differentially expressed proteins that have been successfully identified by LC/MS-MS are represented on the gel. Proteins are numbered for clarity and are outlined in Table 2.
Figure 2. 
 
Progenesis SameSpots software output showing two representative differentially expressed protein spots. Enlarged regions of 2D DIGE representative gels of protein lysates from primary uveal melanomas that did not metastasize (left) and primary uveal melanomas that were found to have subsequently metastasized (right). (A) Spot no. 6 was identified as F-actin capping protein subunit alpha-1 (CAPZA1). Average ratio = 1.3; t-test = 0.028. (B) Spot no. 15 was identified as fatty acid–binding protein, heart type (FABP3). Average ratio = 2.2; t-test = 0.00035.
Figure 2. 
 
Progenesis SameSpots software output showing two representative differentially expressed protein spots. Enlarged regions of 2D DIGE representative gels of protein lysates from primary uveal melanomas that did not metastasize (left) and primary uveal melanomas that were found to have subsequently metastasized (right). (A) Spot no. 6 was identified as F-actin capping protein subunit alpha-1 (CAPZA1). Average ratio = 1.3; t-test = 0.028. (B) Spot no. 15 was identified as fatty acid–binding protein, heart type (FABP3). Average ratio = 2.2; t-test = 0.00035.
Figure 3. 
 
Representative images of the immunohistochemical analysis of (A) FABP3 and (B) TPI1 in primary uveal melanomas that did not metastasize (left) and in primary uveal melanomas that have subsequently been found to have metastasized (right). Magnification ×400. Scale bar = 100 μm.
Figure 3. 
 
Representative images of the immunohistochemical analysis of (A) FABP3 and (B) TPI1 in primary uveal melanomas that did not metastasize (left) and in primary uveal melanomas that have subsequently been found to have metastasized (right). Magnification ×400. Scale bar = 100 μm.
Figure 4. 
 
Representative images of the immunohistochemical analysis of (A) CAPZA1 expression in primary uveal melanomas that did not metastasize and of (B) CAPZA1 expression in primary uveal melanomas that have subsequently been found to have metastasized (magnification ×400; scale bar = 100 μm). Serial uveal melanoma sections stained with (C) CAPZA1 and (D) CD68 indicate that CAPZA1 is expressed in tumor cells and in a number of tumor-infiltrating macrophages. Magnification ×200. Scale bar = 200 μm.
Figure 4. 
 
Representative images of the immunohistochemical analysis of (A) CAPZA1 expression in primary uveal melanomas that did not metastasize and of (B) CAPZA1 expression in primary uveal melanomas that have subsequently been found to have metastasized (magnification ×400; scale bar = 100 μm). Serial uveal melanoma sections stained with (C) CAPZA1 and (D) CD68 indicate that CAPZA1 is expressed in tumor cells and in a number of tumor-infiltrating macrophages. Magnification ×200. Scale bar = 200 μm.
Figure 5. 
 
Representative images of the immunohistochemical analysis of (A) PDIA3, (B) SELENBP1, and (C) PARK7 in primary uveal melanomas that did not metastasize. Magnification ×400. Scale bar = 100 μm.
Figure 5. 
 
Representative images of the immunohistochemical analysis of (A) PDIA3, (B) SELENBP1, and (C) PARK7 in primary uveal melanomas that did not metastasize. Magnification ×400. Scale bar = 100 μm.
Figure 6. 
 
siRNA knockdown of (A) FABP3 and (B) TPI1 decreases invasion and motility. (i) Western blot analysis showing the reduction of (A) FABP3 and (B) TPI1 following siRNA knockdown. Control cells (untreated), transfection reagent treated cells (Lipo), scrambled siRNA treated cells, siRNA #1 and siRNA #2 specific for (A) FABP3 or (B) TPI1 treated cells were lysed 72 hours after transfection and subjected to Western blot analysis. GAPDH served as an internal loading control. (A, B, ii) Invasion assays of 92.1 cells following siRNA transfection. The total number of 92.1 cells invading through membrane of invasion chambers after siRNA transfection is shown. (A, B, iii) Motility assays of 92.1 cells following siRNA transfection. The total number of 92.1 cells migrating through uncoated membrane after siRNA transfection is shown. One representative result out of three independent experiments is shown, and error bars represent standard deviation from data obtained from two technical repeats. *P < 0.05; **P < 0.01 compared with scrambled siRNA treated controls.
Figure 6. 
 
siRNA knockdown of (A) FABP3 and (B) TPI1 decreases invasion and motility. (i) Western blot analysis showing the reduction of (A) FABP3 and (B) TPI1 following siRNA knockdown. Control cells (untreated), transfection reagent treated cells (Lipo), scrambled siRNA treated cells, siRNA #1 and siRNA #2 specific for (A) FABP3 or (B) TPI1 treated cells were lysed 72 hours after transfection and subjected to Western blot analysis. GAPDH served as an internal loading control. (A, B, ii) Invasion assays of 92.1 cells following siRNA transfection. The total number of 92.1 cells invading through membrane of invasion chambers after siRNA transfection is shown. (A, B, iii) Motility assays of 92.1 cells following siRNA transfection. The total number of 92.1 cells migrating through uncoated membrane after siRNA transfection is shown. One representative result out of three independent experiments is shown, and error bars represent standard deviation from data obtained from two technical repeats. *P < 0.05; **P < 0.01 compared with scrambled siRNA treated controls.
Table 1. 
 
Clinical and Histopathological Features of the Patients Included in the 2D DIGE and Immunohistochemical Studies
Table 1. 
 
Clinical and Histopathological Features of the Patients Included in the 2D DIGE and Immunohistochemical Studies
Sample Sex Age at Diagnosis (y) Follow-up Time (y)* Clinical Characteristics Histopathological Characteristics Study
Metastatic Sites Ciliary Body Involvement Extrascleral Extension Cell Type LTD (mm) No. of Mitotic Cells (per 40 HPF) Chromosome 3 Status
1 F 49 NA Kidney N N S 12 4 Monosomy DIGE, IHC
2 F 51 1 Liver N N E 15 4 Monosomy DIGE, IHC
3 F 73 1 Liver Y N E 10 1 Monosomy DIGE, IHC
4 F 58 5 Liver Y N M 15 0 Monosomy DIGE, IHC
5 F 32 6 Liver Y Y S 8 12 Disomy IHC
6 F 49 5 Liver, skin N Y S 9 1 Monosomy DIGE, IHC
7 M 71 2 Lung N N M 11 16 Monosomy DIGE, IHC
8 M 64 10 Lung Y N M 10 12 Disomy DIGE, IHC
9 F 69 11 Liver N N S 12 16 Monosomy DIGE
10 M 71 7 Liver Y Y E 20 6 Monosomy DIGE
11 F 52 12 N N Y M 10 2 Disomy DIGE, IHC
12 F 55 12 N N N E 20 0 Disomy DIGE, IHC
13 F 76 12 N N N M 8 4 Disomy DIGE, IHC
14 M 53 12 N N N M 10 8 Trisomy DIGE, IHC
15 F 42 11 N N N S 17 1 Disomy DIGE, IHC
16 M 68 9 N Y N S 20 18 Disomy DIGE, IHC
17 M 68 11 N N N E 12 20 Trisomy DIGE, IHC
18 F 64 10 N N N E 8 8 Disomy IHC
19 F 46 11 N N N E 23 4 Disomy DIGE
20 M 34 17 N N N M 15 4 Disomy DIGE
21 F 75 11 N N N S 22 4 Disomy DIGE
22 M 52 10 N N N S 19 0 Disomy DIGE
23 M 50 11 N N N M 17 4 Disomy DIGE
24 M 86 9 N N N S 12 0 NA DIGE
25 M 62 9 N N N S 8 4 NA DIGE
26 F 70 9 N N N M 9 4 Disomy DIGE
27 F 74 7 N N N M 14 1 NA DIGE
Table 2. 
 
List of Differentially Regulated Proteins from the Comparison of Primary Uveal Melanomas That Were Found To Have Subsequently Metastasized versus Primary Uveal Melanomas That Did Not Metastasize
Table 2. 
 
List of Differentially Regulated Proteins from the Comparison of Primary Uveal Melanomas That Were Found To Have Subsequently Metastasized versus Primary Uveal Melanomas That Did Not Metastasize
Spot No.* Protein Accession No.† Gene Name Protein Name MWt (Da) pI % Coverage MASCOT Score Average Ratio‡ t-test Previously Published§
1 P30101 PDIA3 Protein disulfide-isomerase A3 precursor 57,146 5.98 37 3396 1.5 0.011
2 P08670 VIM Vimentin 53,676 5.06 50 1445 1.8 0.007 32
P06865 HEXA Beta-hexosaminidase subunit alpha 61,106 5.04 24 525 1.8 0.007 18
3 Q13228 SELENBP1 Selenium-binding protein 1 52,907 6.13 39 2426 1.3 0.044
4 P06733 ENO1 Alpha-enolase 47,481 7.01 53 4944 1.4 0.007 18
5 P05198 EIF2S1 Eukaryotic translation initiation factor 2 subunit 1 36,374 5.02 57 1648 −1.6 0.035
6 P52907 CAPZA1 F-actin capping protein subunit alpha-1 33,073 5.45 50 2522 1.3 0.028
7 P25788 PSMA3 Proteasome subunit alpha type 3 28,643 5.19 26 1267 −1.2 0.026
8 P08865 RPSA 40S ribosomal protein SA 32,947 4.79 30 861 −1.4 0.023 20
9 P30040 ERP29 Endoplasmic reticulum protein ERp29 29,032 6.77 34 2592 1.4 0.040
10 P60174 TPI1 Triosephosphate isomerase 26,938 6.45 61 2987 1.6 0.00057
11 P60174 TPI1 Triosephosphate isomerase 26,938 6.45 30 317 1.7 0.00009
12 Q99497 PARK7 Protein DJ-1 20,050 6.33 34 2037 1.2 0.018
13 P07437 TUBB Tubulin beta chain 50,095 4.78 10 587 −1.7 0.017
14 P68363 TUBA1B Tubulin alpha-1B chain 50,804 4.94 13 553 −1.9 0.006
15 P05413 FABP3 Fatty acid-binding protein, heart 14,906 6.29 51 1790 2.2 0.00035
Table 3.  
 
Expression Pattern of Proteins in the Immunohistochemical Study
Table 3.  
 
Expression Pattern of Proteins in the Immunohistochemical Study
Sample CAPZA1 FABP3 PARK7 PDIA3 SELENBP1 TPI1
Pos Int Score Pos Int Score Pos Int Score Pos Int Score Pos Int Score Pos Int Score
1 90 1 4 30 1 2 70 1 3 55 2 6 40 1 2 40 2 4
2 75 2 8 70 2 6 95 2 8 20 2 2 100 1 4 100 3 12
3 25 1 2 90 3 12 95 1 4 60 2 6 30 3 6 80 2 8
4 40 1 2 90 3 12 95 1 4 35 2 4 0 0 0 95 1 4
5 75 1 4 0 0 0 80 3 12 95 3 12 95 2 8 80 3 12
6 10 2 2 80 3 12 25 1 2 60 2 6 0 0 0 80 3 12
7 40 2 4 100 3 12 90 2 8 30 2 4 0 0 0 45 3 6
8 90 2 8 90 3 12 95 3 12 35 1 2 20 3 3 75 3 12
11 75 3 12 50 1 3 55 1 3 95 3 12 10 3 3 90 3 12
12 0 0 0 5 2 2 40 1 2 95 3 12 0 0 0 20 2 2
13 75 3 12 95 2 8 95 2 8 80 1 4 20 1 1 0 0 0
14 90 3 12 40 1 2 95 2 8 95 2 8 90 2 8 90 3 12
15 90 2 8 90 3 12 80 2 8 80 2 8 0 0 0 80 2 8
16 90 3 12 80 2 8 60 1 3 70 2 6 25 3 6 100 2 8
17 10 1 1 5 3 3 0 0 0 75 2 8 0 0 0 0 0 0
18 75 2 8 50 1 3 80 2 8 35 1 2 0 0 0 100 2 8
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×