Cutaneous T-cell lymphomas (CTCL) comprise a number of neoplasms with primary manifestation in the skin that originate from T lymphocytes. The most common forms of CTCL are mycosis fungoides (MF) and the Sézary syndrome (SS). Both are monoclonal memory T helper lymphomas. MF is characterized by cutaneous patches, plaques, and tumors, SS by erythroderma, lymphadenopathy, and Sézary cells (neoplastic T cells) in the peripheral blood (Burg et al, 1997;Willemze et al, 1997).
There are a number of standard treatment options for CTCL including some new therapeutic drugs, but all of them are still palliative (Dummer et al, 1996;Dippel et al, 2003). Specific immunotherapy like vaccination with peptides or peptide- or tumor-lysate loaded dendritic cells (Rosenberg, 1999;Maier et al, 2003) provides a promising approach for CTCL (Duvic and Cather, 2000). Prerequisite for such strategies is the identification of tumor-specific antigens (Rosenberg, 1999;Rosenberg, 2001). Unfortunately, only very few antigens have been described for CTCL (Eichmüller, 2002).
A new large-scale screening using a newly generated CTCL-derived library and sera of patients suffering from CTCL revealed a new tumor-specific antigen, guanylate-binding protein (GBP)-5ta, which belongs to the family of GBP proteins (Hartmann et al, 2004). The GBP are large GTPases (
70 kDa) being able to bind GTP, GDP, and GMP and to catalyze the hydrolysis of GTP to GDP, as well as GMP (Cheng et al, 1985;Cheng et al, 1991;Schwemmle and Staeheli, 1994). GTPases play an important role in cell proliferation, differentiation, signal transduction, and intracellular protein transportation (Bourne et al, 1991;Boehm et al, 1998). Some members of GTP-binding proteins, including the Ras-family, Rab proteins and Mx proteins, however, do not hydrolyze GTP to GMP. Mutations in ras genes were found in different tumors, suggesting their involvement in the development of specific neoplasms. These mutations result in the loss of GTPase activity and lead to the formation of constitutively active and potentially oncogenic proteins that could cause a deregulation of cell cycle (Weber et al, 2000;Macaluso et al, 2002;Smythe, 2002).
Most known GBP express a CaaX isoprenylation motif at their C-terminus (Nantais et al, 1996) and are interferon-
(IFN-) inducible (Patrone et al, 2002). To date, three human GBP have been published and all mapped to chromosome 1, namely, GBP-1, GBP-2, and GBP-3 (Strehlow et al, 1994;Neun et al, 1996;Luan et al, 2002), but the analysis of their biological functions has only recently been started.Anderson et al (1999) showed that GBP-1 mediates an antiviral effect against vesicular stomatitis virus and encephalomyocarditis virus. Also, GBP-1 counteracts the proliferative effect of inflammatory cytokines like IFN-, interleukin 1-
(IL-1), and tumor necrosis factor-
(TNF-) on endothelial cells (Guenzi et al, 2001). gbp-4 (acc. no. AF288814) and gbp-5 (acc. no. AF288815) are unpublished sequences, which have been submitted to GenBank as open reading frames.
Here we report the identification of three splicing variants of the gbp-5 gene in CTCL biopsies and CTCL-derived cell lines, which translate into two different proteins, GBP-5a/b and GBP-5ta. Furthermore, we characterize the sero-reactivity against these proteins, as well as RNA and protein expression within different tissues.
Results
Screening
Approximately 7
106 recombinant clones of a cDNA library derived from CTCL tissue, cutaneous B-cell lymphoma (CBCL) tissue as well as four CTCL cell lines were screened by sera from patients suffering from CTCL including MF and SS leading to nine reactive antigens (Hartmann et al, 2004). One clone (HD-CL-05) showed high homology to the family of guanylate-binding proteins (GBP) and was therefore named gbp-5ta (gbp-5 tumor antigen; accession no AF328727). gbp-5ta codes for a 490aa protein and contains a guanylate binding site comprised of three motifs, comparable with other GBP members. Although GBP-1 (accession no. M55542;Strehlow et al, 1994) and GBP-2 (accession no. M55543;Cheng et al, 1991) showed only 80% and 77% similarity, respectively, a recently submitted ORF sequence, GBP-5 (accession no. AR035948, unpublished), was identical in the overlapping region. In order to identify more homologous genes we performed a cDNA-library screening of 3105 plaques using digoxigenin-labeled PCR products derived from primer set I Figure 1 as probes and the same CTCL phage library mentioned above. Four clones could be identified: one harbored GBP-2 in reverse orientation, the second was identical to gbp-5ta, and two clones represented gbp-5 splicing variants, gbp-5a and -5b. gbp-5a and -5b possessed in comparison with gbp-5ta two additional exons at the 3' end. gbp-5a lacked the second exon. The deduced proteins of gbp-5a and gbp-5b were identical and are therefore named GBP-5a/b Figure 1.
Figure 1.
Chromosomal alignment of gbp-5ta, gbp-5a, and gbp-5b with chromosome 1. The scheme depicts the chromosomal localization of all three splicing variants and the used primers. All identified variants were 100% homologous to chromosome 1p22.3. Primer set I is specific for gbp-5ta and gbp-5b, whereas primer set II recognized all three variants leading to RT-PCR products with two different sizes.
Full figure and legend (22K)gbp-5 was found on chromosome 1p22.3 (accession no. AC021689; Figure 1) and consists of 12 exons, of which the second is specific for gbp-5ta and gbp-5b. gbp-5ta lacks exons 11 and 12. To confirm the existence of the shorter gbp-5ta without exons 11 and 12 we performed a 3' RACE-PCR using a CTCL RACE-cDNA unraveling two bands. Cloning and sequencing of these bands revealed gbp-5ta for the shorter and gbp-5a/-5b for the longer clone. This indicated that gbp-5ta might be an alternative splicing variant at the 3'-end of gbp-5.
Homology
Blast searches of GenBank and EMBL databases revealed homologous sequences with 67% (591aa), 64% (591aa), and 68% (581aa) homology of GBP-5a/b to the human GBP-1, GBP-2, and GBP-3, respectively. Although gbp-5a and gbp-5b code exactly the same protein (586aa, 67 kDa), GBP-5ta (489aa, 60 kDa) was C-terminal truncated by 97aa.
Serological analysis
Additional sera of patients diagnosed with MF (n=12) or SS (SS) (n=4) and 14 control sera from healthy volunteers were tested using gbp-5a, gbp-5b, and gbp-5ta phage clones in a SEREX approach. Reactivity of CTCL patients' sera tested by SEREX resulted in nine of 16 (56%) positive sera for GBP-5ta and two per 11 (18%) for GBP-5a/b, which has been tested using the GBP-5a coding phage. All tested sera of healthy donors (n=14) were negative.
Expression spectrum of GBP-5 RNAs
Two different primer sets were used to analyze mRNA expression, which could distinguish gbp-5ta/-5b and gbp-5a Figure 1. We found gbp-5ta/-5b in 29% and gbp-5a in 33% control tissues (bone marrow, stomach, placenta, uterus, small intestine, spleen, peripheral blood monocytes; Primer set I was negative for uterus; small band in primer set II PCR was negative in stomach; Table I). Among the investigated tumor tissues, CTCL and CBCL were positive in 32% (primer set I) and 67% (primer set II). We found two (primer set I) and four (primer set II) positive CTCL cell lines, only one positive leukemia cell line (n=5) and six (primer set I) and eight (primer set II) positive melanoma cell lines Table I. Unexpectedly, the results of primer set I did not always fit to the large band of primer set II. Thus, we suppose the existence of yet other splicing variants of gbp-5.
Table I - RT-PCR analyses using antigen-specific primers and cDNA from various kinds of control tissues, tumor tissues, and tumor cell lines.
Full table Expression spectrum of the GBP-5a/b and GBP-5ta protein
We generated polyclonal antibodies K25 and K26 against recombinant GBP-5ta. Antibody K25 detected a 60 kDa protein representing GBP-5ta (Figure 2a, lane 1). Antibody K26 visualized a band of 60 kDa reflecting GBP-5ta and a band of 72 kDa for GBP-5a/b Figure 2b.
Figure 2.
GBP-5ta and GBP-5a/b protein expression in CTCL cell lines and control tissues.A: Antibody K25 detected GBP-5ta, only, which was labeled as a band of 60 kDa in the four CTCL cell lines but not in any of the control tissues. Recombinant GBP-5ta (lane 1), CTCL cell lines (2: MyLa, 3: HuT-78, 4: HH, 5: SeAx), and control tissues (6: testis, 7: fetal liver, 8: uterus, 9: placenta). B: Antibody K26 labeled GBP-5ta and GBP-5a/b as bands of 60 and 72 kDa, respectively, in the four CTCL cell lines (1: MyLa, 2: HuT-78, 3: SeAx, 4: HH). All control tissues (5: bone marrow, 6: testis, 7: stomach) besides PBMC (lane 8) were negative.
Full figure and legend (32K)Western blot analysis using antibodies K25 and K26 was performed using total protein from four CTCL cell lines and control tissues tested positive in RT-PCR (primer set I or primer set II; Table II). The western blot results illustrated that all tested control tissues besides peripheral blood monocytes (PBMC) were not reactive using either antibody. PBMC were positive for GBP-5a/b and a faint band was visible at the size of GBP-5ta. To specify GBP-5ta and GBP-5a/b expression in PBMC, CD4+ and CD8+ cells were separated from the remaining PBMC by magnetic beads and analyzed using K26 Figure 3. GBP-5ta was only moderately expressed in CD4+ and CD8+ cells, whereas GBP-5a/b was expressed in all cell types.
Figure 3.
Expression analysis of GBP-5ta and GBP-5a/b in PBMC. Antibody K26 detected GBP-5a/b in CD4+ (lane 2), CD8+ (lane 3) and in the remaining cells (lane 4) whereas GBP-5ta is only weakly expressed in CD4+ and CD8+ cells and not in the remaining cells. Lane 1: total PBMC.
Full figure and legend (98K)All four CTCL cell lines that were positive in RT-PCR showed one band in the appropriate size of GBP-5ta with antibody K25 Figure 2a. Lysates of the same CTCL cell lines incubated with antibody K26 revealed two bands indicating the presence of both GBP-5ta and GBP-5a/b Figure 2b. GBP-5ta protein was detected in seven of seven CTCL patients by western blot using antibody K26. Results are shown for four patients Figure 4. Tumor tissues of three patients with MF and one with SS showed strong expression of GBP-5ta protein and no expression of GBP-5a/b. The corresponding lymph node of the SS patient showed a lower expression of GBP-5ta than the tumor tissue (Figure 4, SS and SS LN). GBP-5a/b expression was found in total protein of eight of nine melanoma cell lines (negative: SK-Mel023), whereas a weak band for GBP-5ta could be detected in four of nine (e.g., see Figure 5). The observed expression of GBP-5ta in CTCL and melanoma cell lines could be confirmed by antibody K25 in all cases (not shown).
Figure 4.
GBP-5ta protein was detected in CTCL tumor tissues. Representative western blot with protein lysates from CTCL tumor tissue of four different patients (3 MF and 1 SS; patient MF1 suffered additionally from Hodgkin lymphoma) and from a corresponding lymph node (SS LN). Antibody K26 labeled a singular band (60 kDa), only, representing GBP-5ta.
Full figure and legend (27K)Figure 5.
Melanoma cell lines express mainly GBP-5a/b. Six melanoma cell lines stained with K26. GBP-5a/b was visible as a band of 72 kDa in six of six tested melanoma cell lines. GBP-5ta was expressed weakly in four of six (lanes 2, 4, 5, 6). 1: Ma-Mel-17, 2: Ma-Mel-37a, 3: Ma-Mel-05, 4: Ma-Mel-42a, 5: Ma-Mel-04, 6: Ma-Mel-12.
Full figure and legend (21K)Interferon-
induction assay
Due to its homology to the IFN- inducible family of GBP we investigated the IFN- inducibility of GBP-5 proteins. Three CTCL cell lines (HH, HuT-78, and MyLa) were exposed to 10 and 100 U per mL IFN- for 12 h. ICAM-1 was induced in HH, whereas HuT-78 and MyLa were invariable. HLA class I was weakly upregulated only in HuT-78 Figure 6b. The melanoma cell line SK-Mel023 was used as positive control. ICAM-1 and HLA class I could be induced in this cell line by treatment with IFN-, but GBP-5ta or GBP-5a/b were negative before and after IFN- induction. All tested CTCL cell lines were positive for both proteins which remained unchanged by IFN- treatment, as is shown in Figure 6a.
Figure 6.
IFN- did not alter GBP-5 protein expressions. The CTCL cell lines HuT-78 and MyLa were induced with 0, 10, and 100 U per mL IFN- for 12h. A: Antibody K26 detected GBP-5ta as a band of 60 kDa and GBP-5a/b as a band of 72 kDa in both CTCL cell lines, irrespective of the IFN- treatment. B: IFN- induced ICAM-1 in HH (arrow) and HLA class I in HuT-78 (arrow), only.
Discussion
gbp-5 represents a new gene of the GBP-family, whose three splicing variants are transcribed into two different proteins (GBP-5a/b and GBP-5ta). GBP-5ta is a 97aa-truncated variant of GBP-5a/b. More than 50% of CTCL sera, but none of the control sera reacted with GBP-5ta protein, suggesting a strong immunogenicity of this tumor antigen. Interestingly, the truncated GBP-5 protein, GBP-5ta, revealed a stronger antibody response than the full-length protein GBP-5a/b.
gbp-5ta mRNA was found to be expressed in selected normal tissues. RT-PCR detected gbp-5ta/-5b in 26%–58% of CTCL tumor tissues and in 45% of melanomas. By means of two newly generated antibodies we were able to investigate the expression of the two proteins by Western blot. All control tissues tested were negative for both GBP-5a/b and GBP-5ta, with the exception of PBMC, which were positive for GBP-5a/b and weakly positive for GBP-5ta. A detailed analysis of PBMC subpopulations indicated that GBP-5ta was expressed in CD4+ and CD8+ as a faint band only, whereas GBP-5a/b was expressed in all PBMC cell types analyzed. In CTCL-tumor samples only GBP-5ta was detectable, and eight of nine melanoma cell lines expressed GBP-5a/b and four of nine GBP-5ta protein. In contrast, CTCL cell lines were positive for both GBP-5a/b and GBP-5ta. The observed discrepancy between mRNA detection and protein expression has also been reported for other genes: mRNA of the tumor antigens SART-1 and SART-3 have been detected in normal tissue by Northern blot, and the protein was only found in tumor tissues (Shichijo et al, 1998;Kikuchi et al, 1999).
IFN- treatment has been shown to induce the expression of guanylate-binding proteins in both murine and human cells (Strehlow et al, 1994;Nantais et al, 1996;Stickney and Buss, 2000;Gorbacheva et al, 2002). Therefore, we have investigated the IFN- inducibility of GBP-5. The Western blot suggested an unresponsiveness of both GBP-5ta and GBP-5a/b to IFN- as the protein level was not altered which might be explained by IFN- resistance of CTCL derived tumor cells. IFN- resistance is suggested to be a pathway for Th2 tumor cells to avoid inhibition by reactive Th1 lymphocytes (Dummer et al, 2001). Additionally, ICAM-1 and HLA class I were induced only in one of three CTCL cell lines, each. Normally HLA class I molecules and ICAM-1 were found to be upregulated by IFNs (Garbe and Krasagakis, 1993). A melanoma cell line known to be inducible by IFN- (SK-Mel023) was also tested for GBP-5ta and -5a/b expression. We did not detect any GBP-5ta or GBP-5a/b irrespectively whether the cell line was treated by IFN- or not.
A well-known member of the GTPases is the Ras protein with its oncogenic mutant variants. Five domains are essential for Ras activity (Macaluso et al, 2002). Mutations in these regions block the hydrolysis of GTP, leading to the expression of constitutively active protein, which stimulates the cells to uncontrolled proliferation (Scheffzek et al, 1998;Ahmadian et al, 1999). Cystein mutation in the CaaX box prevents farnesylation and Ras function (Hancock et al, 1989;Schafer et al, 1989).
Sequence comparisons suggest that GBP-5a/b may have GTPase and isoprenylation properties similar to those of GBP-1, the best studied GBP family member (Prakash et al, 2000). The CaaX box of GBP-5a/b consists of cystein–valin–leucin–leucin. Post-translational modifications of the CaaX box increase the hydrophobicity by isoprenylation of the cystein, leading to integration into the plasma membrane (Clarke, 1992;Casey, 1995). GBP-5ta lacks 97aa of its C-terminus including the CaaX box, which might result in an inability of membrane integration of GBP-5ta, whereas GBP-5a/b can be isoprenylated.
Murine GBP-5, which is similar to human GBP-5, has been found in two splicing variants (Nguyen et al, 2002). MuGBP-5 contains two GTP-binding motifs and one isoprenylation site at the C-terminus. MuGBP-5a is N-terminally truncated by 112aa and C-terminally extended, but lacks the isoprenylation motif. Thus, both human and murine GBP-5 appear in different splicing variants of which one has lost its isoprenylation site.
GBP-1 was found to have an anti-proliferative activity on endothelial cells independent of its GTPase function and the isoprenylation motif.Guenzi et al (2001) cloned cDNA fragments encoding either the N-terminal or the C-terminal domains to identify the functional component. They demonstrated that the N-terminal domain has nearly no inhibitory effect, whereas the C-terminus is sufficient to inhibit endothelial cell proliferation. Notably, GBP-5ta also has lost its C-terminus, which may have an important impact on the potential proliferation inhibitory function of GBP-5ta.
GBP-5ta and GBP-5a/b might be promising targets for therapy: Both proteins are expressed in CTCL tumor cells, melanoma cells and in PBMC, but in no other control tissues and the expression level of GBP-5ta is higher in CTCL than in normal PBMC. GBP-5ta and GBP-5a/b are recognized by the immune system of many CTCL-patients, but not of healthy volunteers. Most interestingly, gbp-5 derived proteins belong to a well-known family of GTPases involved in cancerogenesis. Ongoing analyses will prove whether these new proteins have similar functions in CTCL and possibly other malignancies.
Material and Methods
Tissue, sera, and cell lines
All human specimens were obtained after informed consent and approval of the local ethical committee. Sera and tumor tissues were acquired during routine diagnostic or therapeutical procedures and were stored at -20°C and -80°C, respectively. Tissue specimens o.btained from 19 CTCL and three CBCL (cutaneous B cell lymphoma) patients served as source for generation of tumor cDNA. The CTCL comprised: 14 MF (stage Ib to IVb, mainly IIb), three SS (stage III), one T-zone lymphoma (stage IVb), and one CD30+CTCL (stage IIb). Additionally, cDNA from the following four CTCL cell lines was generated: MyLa (MF;Kaltoft et al, 1992), SeAx (SS;Kaltoft et al, 1988), HH (SS, ATCC number: CRL-2105), and HuT-78 (SS, ATCC Number: TIB-161). Furthermore, we generated cDNA from six leukemia cell lines (ARA-10, Jurkat, KG1, K562, Nalm-2, and SKW6.4) and from 19 melanoma cell lines. A variety of control cDNAs were used for analyzing tissue distribution within normal tissues, including three panels of commercially available cDNAs (all Clontech, California).
Screening and evaluation of sero-reactivity
GBP-5ta was identified by a SEREX screening using a cutaneous lymphoma tumor library, generated from mRNA of the CTCL cell lines HuT-78, HH, MyLa and SeAx, CTCL tumor tissues (MF and SS) and tumor tissue of a patient suffering from CBCL (Hartmann et al, 2004). The SEREX screening and retesting were performed as described previously. Briefly, pre-absorbed sera from up to 16 CTCL patients and 14 healthy donors were used to determine the immunogenicity of the phage clones in at least two independent SEREX experiments. To identify GBP-5ta homologous genes, a cDNA library screening (HYREX) similar to the SEREX approach was performed as described previously (Usener et al, 2003).
Sequence analysis
Sequencing was done at the central facilities of the German Cancer Research Center. Computational analysis was done using the HUSAR package from the Biocomputing Service Group at the German Cancer Research Center. All sequences were analyzed with prime, map and 2dsweep. Database searches were performed on EMBL and GenBank databases with blast programs for nucleic acids and amino acids.
RT-PCR
Due to limited amounts of RNA, RT-PCR was preferably used for investigating expression patterns of identified sequences within different normal and tumor tissues. RNA was isolated using QIA shredder mini columns and RNeasy mini kit (both Qiagen, Hilden, Germany). The location of primers is given in Figure 1. The primer sequences were (length of PCR-product and annealing temperatures are given in brackets): gbp-5ta, primer set I for: 5'-cgg aca cgc taa ttg ttg tag-3' and rev: 5'-cca tat cca aat tcc ctt ggt gtg ag-3' (364 bp; 63°C); gbp-5ta/b and gbp-5a, primer set II for: 5'-aga agg aag aaa ctc caa aca cat cc-3 and rev: 5'-cca tat cca aat tcc ctt ggt gtg ag –3'(gbp-5ta/b: 515 bp, gbp-5a: 408 bp; 48°C). A standard protocol was used for RT-PCR (95° per 1 min, annealing per 1 min, 72° per 2 min, 35 cycles) using a MJ Research PCT-200 (Biozyme, Oldendorf, Germany) according toEichmüller et al (2001).
Rapid amplification of cDNA ends (RACE)
3' RACE-PCR was performed using a primer located within gbp-5ta (1557 cca gga gct gct gga cct gca cag gac 1583) to verify the different 3' splicing variants. Smart II Oligo labeled 3' cDNA (Smart cDNA amplification kit; BD Clontech, Heidelberg, Germany) was synthesized using the lymphoma library RNA. The proof reading Advantage2 PCR polymerase mix (Clontech) was used according to the manufacturer's protocol. The amplificons were cloned into the pCR TOPO4 vector (Invitrogen, Groningen, Netherlands) and sequenced.
Antibodies
His-tagged GBP-5ta fusion protein was affinity purified on Ni2+-agarose columns (Invitrogen) and separated by a SDS-PAGE. The recombinant GBP-5ta protein was cut out of the gel and the first 13 amino acids at the N-terminal have been sequenced to prove the identity of the protein. The denatured protein was used for generation of two polyclonal rabbit antibodies K25 and K26 (four immunizations on day 1, 7, 14, 28; customized service of BioGenes, Berlin, Germany; following the companies standard protocol). Mouse Anti-actin was purchased from Dianova, Hamburg, Germany.
Antibodies K25 and K26 were tested for specificity in a GST capture ELISA as described bySehr et al (2001). Recombinantly expressed GBP-5ta protein was used as positive control and another antigen, se70-2, was used as negative control. The antibodies were diluted from 1:50 to 1:10,000. At a concentration of at least 1:1000 both antibodies specifically detected recombinant GBP-5ta Figure 7.
Figure 7.
Specificity tests of the antibodies K25 and K26 by ELISA. The ELISA was done using recombinant GST-fusion protein GBP-5ta and the unrelated se70-2 and different dilutions of the antibodies K25 and K26. Both antibodies were specific for recombinantly expressed GBP-5ta at a dilution of 1:1000.
Full figure and legend (14K)Western blot
Western blot was performed using 20 
g protein from total cell lysate and commercial protein medleys of normal tissues (Clontech, Palo Alto, California, USA or BioChain, Hayward, California, USA). Cell pellets of four CTCL cell lines and nine melanoma cell lines (Ma-Mel-04, Ma-Mel-05, Ma-Mel-12, Ma-Mel-17, Ma-Mel-36, Ma-Mel-37a, Ma-Mel-42a, SK-Mel023, and UKRV-Mel-10) were used. Protein was isolated using Triton-X100 and Tris–HCl, separated on SDS-PAGE and blotted onto nitrocellulose membranes (Satorius, Göttingen, Germany). Blots were incubated with a 1:1000 dilution of K25 and K26 antibodies for 1 h at room temperature. Secondary HRP-coupled anti-rabbit antibody (Dianova, Hamburg, Germany) was used at a dilution of 1:10,000. The antibody binding was visualized by ECL-staining (Amersham Bioscience, Buckinghamshire, UK). The specificity of the antibodies was tested by using pre-immune serum of the same animal under the same conditions. All these control western blots were negative. As control for equal amounts of protein a second western blot was done using the same protein lysates and a mouse anti--actin antibody (Dianova, Hamburg, Germany).
Magnetic cell sorting and separation of CD4+ and CD8+ cells
CD4+ and CD8+ cells were isolated from PBMC by magnetic cell sorting (Miltonyi Biotech GmbH, Bergisch Gladbach, Germany) as described by the manufacturer. Twenty g protein extract of the cell pellets were analyzed by western blot.
Interferon-induction assay
CTCL cell lines (HuT-78, HH, MyLa) were treated with 0, 10, and 100 U per mL human IFN- (RD Systems, Wiesbaden, Germany) for 12 h. Protein was extracted (PeqGold TriFast, Peqlab, Erlangen, Germany) and analyzed by western blot using antibody K25 and K26. The melanoma cell line SK-Mel023 was used as positive control for the successful induction of ICAM-1 and HLA class I by treatment with IFN-.
References
| 1. | Ahmadian MR, Zor T, Vogt D, Kabsch W, Selinger Z, Wittinghofer A & Scheffzek K. Guanosine triphosphatase stimulation of oncogenic Ras mutants. Proc Natl Acad Sci USA (1999) 96: 7065–7070. | Article | PubMed | ChemPort | |
| 2. | Anderson SL, Carton JM, Lou J, Xing L & Rubin BY. Interferon-induced guanylate binding protein-1 (GBP-1) mediates an antiviral effect against vesicular stomatitis virus and encephalomyocarditis virus. Virology (1999) 256: 8–14. | Article | PubMed | ISI | ChemPort | |
| 3. | Boehm U, Guethlein L & Klamp T et al. Two families of GTPases dominate the complex cellular response to IFN-gamma. J Immunol (1998) 161: 6715–6723. | PubMed | ISI | ChemPort | |
| 4. | Bourne HR, Sanders DA & McCormick F. The GTPase superfamily: Conserved structure and molecular mechanism. Nature (1991) 349: 117–127. | Article | PubMed | ISI | ChemPort | |
| 5. | Burg G, Kempf W & Häffner A et al. Cutaneous lymphomas. Curr Probl Dermatol (1997) 9: 137–204. |
| 6. | Casey PJ. Mechanisms of protein prenylation and role in G protein function. Biochem Soc Trans (1995) 23: 161–166. | PubMed | ISI | ChemPort | |
| 7. | Cheng YS, Becker-Manley MF, Chow TP & Horan DC. Affinity purification of an interferon-induced human guanylate-binding protein and its characterization. J Biol Chem (1985) 260: 15834–15839. | PubMed | ChemPort | |
| 8. | Cheng Y, Patterson C & Staeheli P. Interferon-induced guanylate-binding proteins lack an N(T)KXD consensus motif and bind GMP in addition to GDP and GTP. Mol Cell Biol (1991) 11: 4717–4725. | PubMed | ISI | ChemPort | |
| 9. | Clarke S. Protein isoprenylation and methylation at carboxyl-terminal cysteine residues. Annu Rev Biochem (1992) 61: 355–386. | Article | PubMed | ISI | ChemPort | |
| 10. | Dippel E, Klemke CD & Goerdt S. Current status of cutaneous T-cell lymphoma: Molecular diagnosis, pathogenesis, therapy and future directions. Onkologie (2003) 26: 477–483. | Article | PubMed | ISI | ChemPort | |
| 11. | Dummer R, Döbbeling U, Geertsen R, Willers J, Burg G & Pavlovic J. Interferon resistance of cutaneous T-cell lymphoma-derived clonal T-helper 2 cells allows selective viral replication. Blood (2001) 97: 523–527. | Article | PubMed | ISI | ChemPort | |
| 12. | Dummer R, Häffner A, Hess M & Burg G. A rational approach to the therapy of cutaneous T-cell lymphomas. Onkologie (1996) 19: 226–230. | ISI | |
| 13. | Duvic M & Cather JC. Emerging new therapies for cutaneous T-cell lymphoma. Dermatol Clin (2000) 18: 147–156. | Article | PubMed | ISI | ChemPort | |
| 14. | Eichmüller S. Towards defining specific antigens for cutaneous lymphomas. Onkologie (2002) 25: 448–454. | Article | PubMed | |
| 15. | Eichmüller S, Usener D, Dummer R, Stein A, Thiel D & Schadendorf D. Serological detection of cutaneous T-cell lymphoma-associated antigens. Proc Natl Acad Sci USA (2001) 98: 629–634. | Article | PubMed | ChemPort | |
| 16. | Garbe C & Krasagakis K. Effects of interferons and cytokines on melanoma cells. J Invest Dermatol (1993) 100: 239S–244S. | Article | PubMed | ChemPort | |
| 17. | Gorbacheva VY, Lindner D, Sen GC & Vestal DJ. The interferon (IFN)-induced GTPase, mGBP-2. Role in IFN-gamma-induced murine fibroblast proliferation. J Biol Chem (2002) 277: 6080–6087. | Article | PubMed | ISI | ChemPort | |
| 18. | Guenzi E, Topolt K & Cornali E et al. The helical domain of GBP-1 mediates the inhibition of endothelial cell proliferation by inflammatory cytokines. EMBO J (2001) 20: 5568–5577. | Article | PubMed | ISI | ChemPort | |
| 19. | Hancock JF, Magee AI, Childs JE & Marshall CJ. All ras proteins are polyisoprenylated but only some are palmitoylated. Cell (1989) 57: 1167–1177. | Article | PubMed | ISI | ChemPort | |
| 20. | Hartmann TB, Thiel D, Dummer R, Schadendorf D & Eichmüller S. SEREX identification of new tumour-associated antigens in cutaneous T-cell lymphoma. Br J Dermatol (2004) 150: 252–258. | Article | PubMed | ISI | ChemPort | |
| 21. | Kaltoft K, Bisballe S, Dyrberg T, Boel E, Rasmussen PB & Thestrup-Pedersen K. Establishment of two continuous T-cell strains from a single plaque of a patient with mycosis fungoides. In Vitro Cell Dev Biol (1992) 28A: 161–167. | PubMed | ISI | ChemPort | |
| 22. | Kaltoft K, Bisballe S, Rasmussen HF, Thestrup-Pedersen K, Boehncke WH, Volker H & Sterry W. C-type particles are inducible in Se-Ax, a continuous T-cell line from a patient with Sezary's syndrome. Arch Dermatol Res (1988) 280: 264–267. | Article | PubMed | ISI | ChemPort | |
| 23. | Kikuchi M, Nakao M, Inoue Y, Matsunaga K, Shichijo S, Yamana H & Itoh K. Identification of a SART-1-derived peptide capable of inducing HLA-A24-restricted and tumor-specific cytotoxic T lymphocytes. Int J Cancer (1999) 81: 459–466. | Article | PubMed | ISI | ChemPort | |
| 24. | Luan Z, Zhang Y, Liu A, Man Y, Cheng L & Hu G. A novel GTP-binding protein hGBP3 interacts with NIK/HGK. FEBS Lett (2002) 530: 233–238. | Article | PubMed | ISI | ChemPort | |
| 25. | Macaluso M, Russo G, Cinti C, Bazan V, Gebbia N & Russo A. Ras family genes: An interesting link between cell cycle and cancer. J Cell Physiol (2002) 192: 125–130. | Article | PubMed | ISI | ChemPort | |
| 26. | Maier T, Tun-Kyi A, Tassis A, Jungius KP, Burg G, Dummer R & Nestle FO. Vaccination of cutaneous T-cell lymphoma patients using intranodal injection of autologous tumor lysate pulsed dendritic cells. Blood (2003) prepublished online. |
| 27. | Nantais DE, Schwemmle M, Stickney JT, Vestal DJ & Buss JE. Prenylation of an interferon-gamma-induced GTP-binding protein: The human guanylate binding protein, huGBP1. J Leukoc Biol (1996) 60: 423–431. | PubMed | ISI | ChemPort | |
| 28. | Neun R, Richter MF, Staeheli P & Schwemmle M. GTPase properties of the interferon-induced human guanylate-binding protein 2. FEBS Lett (1996) 390: 69–72. | Article | PubMed | ISI | ChemPort | |
| 29. | Nguyen TT, Hu Y, Widney DP, Mar RA & Smith JB. Murine GBP-5, a new member of the murine guanylate-binding protein family, is coordinately regulated with other GBPs in vivo and in vitro. J Interferon Cytokine Res (2002) 22: 899–909. | Article | PubMed | ISI | ChemPort | |
| 30. | Patrone L, Damore MA, Lee MB, Malone CS & Wall R. Genes expressed during the IFN gamma-induced maturation of pre-B cells. Mol Immunol (2002) 38: 597–606. | Article | PubMed | ISI | ChemPort | |
| 31. | Prakash B, Renault L, Praefcke GJ, Herrmann C & Wittinghofer A. Triphosphate structure of guanylate-binding protein 1 and implications for nucleotide binding and GTPase mechanism. EMBO J (2000) 19: 4555–4564. | Article | PubMed | ISI | ChemPort | |
| 32. | Rosenberg SA. A new era for cancer immunotherapy based on the genes that encode cancer. Immunity (1999) 10: 281–287. | Article | PubMed | ISI | ChemPort | |
| 33. | Rosenberg SA. Progress in the development of immunotherapy for the treatment of patients with cancer. J Intern Med (2001) 250: 462–475. | Article | PubMed | ISI | ChemPort | |
| 34. | Schafer WR, Kim R, Sterne R, Thorner J, Kim SH & Rine J. Genetic and pharmacological suppression of oncogenic mutations in ras genes of yeast and humans. Science (1989) 245: 379–385. | PubMed | ISI | ChemPort | |
| 35. | Scheffzek K, Ahmadian MR & Wittinghofer A. GTPase-activating proteins: Helping hands to complement an active site. Trends Biochem Sci (1998) 23: 257–262. | Article | PubMed | ISI | ChemPort | |
| 36. | Schwemmle M & Staeheli P. The interferon-induced 67-kDa guanylate-binding protein (hGBP1) is a GTPase that converts GTP to GMP. J Biol Chem (1994) 269: 11299–11305. | PubMed | ISI | ChemPort | |
| 37. | Sehr P, Zumbach K & Pawlita M. A generic capture ELISA for recombinant proteins fused to glutathione S-transferase: Validation for HPV serology. J Immunol Methods (2001) 253: 153–162. | Article | PubMed | ISI | ChemPort | |
| 38. | Shichijo S, Nakao M & Imai Y et al. A gene encoding antigenic peptides of human squamous cell carcinoma recognized by cytotoxic T lymphocytes. J Exp Med (1998) 187: 277–288. | Article | PubMed | ISI | ChemPort | |
| 39. | Smythe E. Direct interactions between rab GTPases and cargo. Mol Cell (2002) 9: 205–206. | Article | PubMed | ISI | ChemPort | |
| 40. | Stickney JT & Buss JE. Murine guanylate-binding protein: Incomplete geranylgeranyl isoprenoid modification of an interferon-gamma-inducible guanosine triphosphate-binding protein. Mol Biol Cell (2000) 11: 2191–2200. | PubMed | ISI | ChemPort | |
| 41. | Strehlow I, Lohmann Matthes ML & Decker T. The interferon-inducible GBP1 gene: Structure and mapping to human chromosome 1. Gene (1994) 144: 295–299. | Article | PubMed | ISI | ChemPort | |
| 42. | Usener D, Gerhardt A, Schadendorf D & Eichmüller S. Sero-reactivity against MAGE-A and LAGE-1 proteins in melanoma patients. Br J Dermatol (2003) 149: 282–288. | Article | PubMed | ISI | ChemPort | |
| 43. | Weber F, Haller O & Kochs G. MxA GTPase blocks reporter gene expression of reconstituted Thogoto virus ribonucleoprotein complexes. J Virol (2000) 74: 560–563. | PubMed | ISI | ChemPort | |
| 44. | Willemze R, Kerl H & Sterry W et al. EORTC classification for primary cutaneous lymphomas: A proposal from the Cutaneous Lymphoma Study Group of the European Organization for Research and Treatment of Cancer. Blood (1997) 90: 354–371. | PubMed | ISI | ChemPort | |
Acknowledgments
We gratefully acknowledge Judith Bartels and Anita Dahlke for excellent technical assistance. We thank Andreas Hunziker and Wolfgang Weinig (DKFZ, Heidelberg, Germany) for sequencing and primer synthesis. This work has been supported in parts by the Cancer Research Institute/Elaine R. Shepard Memorial Investigator Award to SE.
