Results by hilen

VIEWS: 7 PAGES: 19

									Identification of a new cancer/testis gene family, CT47, among expressed multi-copy genes on the human X chromosome

Yao-Tseng Chen1,2, Christian Iseli3, Charis A. Venditti2, Lloyd J. Old2, Andrew J. G. Simpson2 and C. Victor Jongeneel3
1

Weill Medical College of Cornell University, New York, NY 10021, USA Ludwig Institute for Cancer Research, New York Branch, Memorial Sloan-Kettering

2

Cancer Center, New York, NY 10021, USA
3

Ludwig Institute for Cancer Research, Office of Information Technology, and Swiss

Institute of Bioinformatics, 1015 Lausanne, Switzerland

1

Abstract
Cancer/Testis (CT) genes are normally expressed in germ cells only, and yet are reactivated and expressed in some tumors. Of approximately 40 CT genes or gene families identified to date, 20 are located on the X chromosome and are present as multi-gene families, many with highly conserved members. This indicates that novel CT gene families may be identified by detecting duplicated expressed genes on chromosome X. By searching for transcript clusters that map to multiple locations on the chromosome, followed by in silico analysis of their gene expression profiles, we identified five novel gene families with testis-specific expression and >98% sequence identity among family members. The expression of these genes in normal tissues and various tumor cell lines and specimens was evaluated by qualitative and quantitative RT-PCR, and a novel CT gene family with at least 13 copies was identified on Xq24, designated as CT47. CT47 mRNA expression was found mainly in testis, with weak expression in placenta. Brain was the only positive somatic tissue tested, with an estimated CT47 transcript level of 0.09% that found in testis. Among tumor specimens tested, CT47 expression was found in ~15% of lung cancer and esophageal cancer specimens, but not in colorectal cancer or breast cancer. The putative CT47 protein consists of 288 amino acid residues, with a Cterminus rich in alanine and glutamic acid. The only species, other than human, in which a gene homologous to CT47 was detected was the chimpanzee, with the predicted protein showing ~80% identity in its carboxy terminal region.

2

Introduction
Cancer/Testis (CT) genes are defined by an expression pattern that is largely restricted to the germ line in normal adults, and re-appears in a subset of cancers (Scanlan, et al. 2002b). Much of the interest in CT genes has focused on their potential as targets for cancer immunotherapy. Initially, individual CT genes were discovered as cancer antigens, containing epitopes against which patients had developed cell-mediated immune responses. The MAGEA1 (CT1.1), BAGE (CT2) and GAGE1 (CT3.1) genes are in this category (Boon and van der Bruggen 1996). In addition, some gene products against which antibodies can be detected in the serum of cancer patients, using the serological expression cloning (SEREX) technique (Sahin, et al. 1995), were also shown to be encoded by CT genes, including NY-ESO-1 (CTAG1, CT6), SSX2 (CT5.2) and SYCP1 (CT8) (Scanlan, et al. 2002b). More recently the search for CT genes has shifted to strategies relying solely on expression patterns, using both experimental techniques (representational difference analysis, differential display, microarrays) (Gure, et al. 2000; Lucas, et al. 2000) and the in silico mining of “digital” expression data (serial analysis of gene expression, expressed sequence tags, and massively parallel signature sequencing) (Chen, et al. 2005; Dong, et al. 2003; Scanlan, et al. 2002a). Overall, over 40 CT genes/gene families have been identified, of which a little less than half have so far been shown to be immunogenic. The characterization of CT genes has revealed that they can be divided into two groups based on chromosomal location. Most of those located on autosomes are single genes, many encoding proteins with known germ line-specific functions. Examples of this group are SYCP1 (CT8) and ADAM2 (CT15). In contrast, CT genes found on 3

chromosome X are often members of extended families that have undergone recent duplication events. Twenty out of 44 CT gene families that were recently reviewed (http://www.cancerimmunity.org/CTdatabase) are located on the X chromosome. These include all of the families with more than two members, most notably MAGEA (CT1), BAGE (CT2), MAGEB (CT3), GAGE (CT4), SSX (CT5), SPANX (CT11) and XAGE (CT12). Benson and colleagues (Warburton, et al. 2004) observed that regions on the X chromosome that contain large highly conserved direct or inverted repeats are enriched for CT genes. The completion of the sequence of the human X chromosome (Ross, et al. 2005) confirmed the phenomenon of CT clustering on the X chromosome and allowed the identification of additional members in several CT families, bringing the total number of CT genes on the X chromosome to at least 99. It was hypothesized that expression of alleles from some of the CT families would have a selective advantage in males, thus favoring their expansion on the X chromosome. This hypothesis is supported by the fact that several CT families have expanded independently in the rodent and primate lineages (Artamonova and Gelfand 2004; Chomez, et al. 2001). The finding that the X chromosome is rich in CT gene families with highly conserved members in direct or inverted repeats indicated that this property might be used to discover new genes with CT expression characteristics. In this paper, we show that screening for expressed loci present at more than one copy on the X chromosome does indeed reveal additional genes preferentially expressed in the testis, and that at least one of these novel families, designated CT47, has a CT expression profile.

4

Materials and Methods
Detection of duplicated genes. The protocol used for generating gene models will be described in detail elsewhere (Iseli, et al. 2002; Iseli et al, unpublished results). Briefly, all publicly available human cDNA sequences (full-length as well as ESTs) were mapped to the genome using megablast (Zhang, et al. 2000), and aligned using a modified version of sim4 (Florea, et al. 1998). cDNA to genome alignments were filtered to eliminate those corresponding to paralogs, pseudogenes or genomic contaminants, and then merged into gene models represented by directed acyclic graphs. Only alignments corresponding to cDNAs matching the genome with >98% identity over >90% of their length were considered. cDNA sequences that participated in the definition of the same gene model were clustered into Human Transcribed Regions (HTRs). We selected the 54 HTRs that mapped to two locations or more in the X chromosome for further study. Thirty-four families were retained after manual curation to remove potential artifacts and to merge closely related families, and are shown in Table 1. The genome coordinates correspond to the NCBI 35 assembly of the human genome. The mapping of HTRs can be visualized by including the DAS source http://tromer.licr.org/cgi-bin/das, DSN = tromer_h, into the Ensembl genome browser (Birney, et al. 2004). The expression patterns of the individual gene families were ascertained in two ways: (1) by examining the sources of the cDNA libraries having contributed to the corresponding HTR; (2) by looking at the distribution of the corresponding MPSS signatures in two different datasets: 32 normal human tissues, and a collection comprising normal testis and placenta, melanocytes and a pool of melanoma biopsies, normal breast epithelial cells and a pool of breast carcinoma biopsies, normal colon and a pool of colon carcinoma biopsies, and the cancer cell lines 5

BT-20 (breast, estrogen receptor negative), MCF-7 (breast, estrogen receptor positive), LC17 (lung) and MEL37 (melanoma).

Tumor Tissues and Cell Lines. Specimens of tumor tissues were obtained from the Departments of Pathology at the Weill Medical College of Cornell University and Memorial Sloan-Kettering Cancer Center, following protocols approved by the institutional review boards. Cell lines were obtained from the cell line bank maintained at the New York Branch of the Ludwig Institute for Cancer Research, New York Branch.

Qualitative RT-PCR. For RT-PCR analysis of normal tissue expression, cDNA was prepared from a panel of commercially obtained total RNA from various human tissues (Clontech-BD Sciences). For tumor cell lines, total RNA was prepared by a standard guanidinium thiocyanate-CsCl gradient method. Two g of total RNA was used for amplification in a 20 l reverse transcription reaction. Two l of resultant cDNA was then used per 25 l PCR reaction. PCR reactions were set up using a commercial master mix (Invitrogen Platinum Taq Supermix), and utilized 35 cycles of amplification, each consisting of 15 sec 94oC, 1 min 60oC, and 1 min 72oC. The PCR products were visualized by 1% agarose gel electrophoresis and ethidium bromide staining.

Quantitative RT-PCR. Quantitative RT-PCR was performed using an ABI PRISM 7000 Sequence Detection System. Normal testis total RNA was obtained commercially (Ambion). Total tumor tissue RNA was prepared using Trizol reagents (Life 6

Technologies). Two g total RNA was used per l reverse transcription reaction, and 2 l cDNA was then used for each 25 l PCR reaction. The reactions were set up in duplicate and the levels of expression were determined as abundance relative to that in the testicular preparation. For this purpose, a standard curve was established for each PCR plate, consisting of testicular cDNA in 4-fold serial dilutions. Forty-five two-step cycles of amplification were performed, each cycle consisting of 15 sec at 95oC and 1 min at 60oC. The RNA quality of the cell lines and tissues was evaluated by separate control amplification of beta-glucuronidase and GAPDH transcripts. All specimens included in the final analysis have Ct values differing by less than four cycles, indicating similar cDNA quality and quantity. For this reason, the Ct values in the experimental group were not normalized against the endogenous control.

Results
Identifying expressed multi-gene families on the X chromosome. The tromer methodology developed in our group includes a genome-wide mapping of individual cDNA sequences, and the generation of high-quality alignments between these cDNAs and the genome. Overlapping alignments produce clusters of transcripts, which are then collapsed into gene models including alternatively spliced forms. In order to identify expressed genes that are repeated on the X chromosome, we searched for transcript clusters that map to multiple locations on this chromosome. Since our mapping criteria are quite stringent (requiring >98% identity between individual cDNA sequences and the

7

genome), multi-gene families with a lower degree of conservation were not detected by the procedure. The results are presented in Table 1. Thirty-four highly conserved gene families were identified on the X chromosome, of which 9 map to the short arm and the rest to the long arm. Thirteen of the 34 correspond to known CT antigens, including the XAGE-1, XAGE-2, GAGE, SPANX, CTAG1 (NY-ESO-1), SSX and MAGE families. Because of the stringency of the mapping, some known families or family members were not detected, and some families were divided into subgroups showing high internal degrees of conservation. For example, only six out of the eight members of the GAGE family (De Backer, et al. 1999) were detected; similarly, a cluster of eight MAGEA family genes encompassing five known genes (MAGEA3, MAGEA6, MAGEA12, MAGEA2B and MAGEA2) was found, but not the rest of the MAGEA family (Chomez, et al. 2001). Two copies of MAGEA9 bracketing MAGEA11 were found as a separate cluster; the additional copy of MAGEA9 was designated MAGEA9B during the annotation of the X chromosome. The SSX family, consisting of nine members with 72% to 91% pairwise identity at the amino acid level (Gure, et al. 2002), was not detected as a repeated family, but two inverted copies each of SSX2 and SSX4 were found; the additional copies correspond to the SSX2B and SSX4B genes identified during the sequencing of the X chromosome (Ross, et al. 2005). The expression patterns of the 21 families that did not correspond to known CT antigens were investigated further by examining the origins of the cDNA libraries within which the corresponding transcripts were identified, and by examining the distribution of the MPSS signatures predicted to be derived from these transcripts among 32 normal 8

human tissues and a collection of cancer biopsies and cell lines (see Materials and Methods). On the basis of their being expressed predominantly in testis among normal tissues, five families encoding candidate CT antigens were identified and selected for further study.

Potential New CT Gene Families. The five new gene families with testis-predominant expression were identified with the following sequences: LOC255313, CXorf2/TEX28, LOC389852, LOC441503, and HTR032426 (Table 1). The LOC255313 transcript, encoding a protein of 288 amino acids and located at Xq24, was found to be derived from a gene family with 13 genes that are arranged as direct tandem repeats over a segment of 60kb (see below). We have designated this gene family as CT47, with the individual members named CT47.1 (most telomeric) through CT47.13 (most centromeric), along the transcribed minus strand (see Figure 4 below). TEX28, a.k.a. CXorf2, encodes a putative protein of 410 amino acids. Three exact copies of this gene were found at Xq28, bracketing two copies of the OPN1MW gene on the opposite strand. LOC389852, a.k.a. PNPK6288, encodes a putative protein of 159 amino acids and is present in two copies at Xp11.23. LOC441503 encodes a protein of 137 amino acids. Two inverted copies of this gene were found at Xq13. HTR032426 (a cluster that contains ESTs AA813325, AA854001, AI015085, AI066759 and AI205897) is located at Xq25-26.3 and appears to have two copies. However, no long open reading frame could be identified in the predicted transcript, and it is therefore not clear whether this transcript is derived from a protein-coding gene.

9

Expression of CT candidate genes. The expression of the five CT candidate genes in normal tissues was examined by RT-PCR (Figure 1). Using a panel of 12 normal tissues, LOC441503 and HTR032426 were found be testis-specific. TEX28 was found to be strongly expressed in testis with marginal expression, estimated at <1% based on the intensity of PCR products, in prostate, spleen, brain, breast, pancreas, and placenta. LOC389852 showed strong expression in testis, but also moderate expression in multiple somatic tissues, including brain, colon, kidney, pancreas, and prostate. CT47 (LOC255313) showed strong expression in testis only, with low expression detected in placenta, and marginal expression in brain. Nine other somatic tissues were negative for CT47 expression. The expression of TEX28, LOC441503, HTR032426 and CT47 was then evaluated in 21 tumor cell lines: seven derived from melanoma (SK-MEL-10, 24, 37, 49, 55, 80, 128), four from small cell lung cancer (NCI-H82, H128, H187, H740), three from non-small cell lung cancer (SK-LC-5, 14, 17), three from colon cancer (SW403, SW480, LS174T), one from renal cancer (SK-RCC-1), one from hepatocellular carcinoma (SKHEP-1), one from bladder cancer (T24), and one from sarcoma (SW982). Each of these cell lines expresses at least one known CT gene (data not shown). LOC441503 and HTR032416 were not detectably expressed in any of the cell lines, and TEX28 showed very low level expression in 8 of the 21 lines. The mRNA levels in the cell lines appeared to be similar to those in the non-testicular normal somatic tissues (see above), estimated at <1% of testicular expression. In contrast, CT47 was expressed in most cell lines, with at least 9 of the 21 showing a moderate quantity of PCR products (Figure 2A, also see qRT-PCR results below). 10

Analysis of CT47 Expression by qRT-PCR. The expression of CT47 was then evaluated by quantitative RT-PCR. Of 12 normal tissues, placenta and brain were the only non-testicular tissues that showed detectable CT47 mRNA, with levels 0.22% and 0.09% that of testicular expression, respectively. This result was consistent with the findings by conventional RT-PCR above. Similarly, qRT-PCR on the 21 cell lines confirmed expression in 20 of 21 cell lines. One of the 20 had mRNA levels >10% that of testis, whereas 5 cell lines were between 1-10%, and 14 had levels between 0.1% and 1% that of testis. All had higher CT47 mRNA levels than normal somatic tissues. The expression of CT47 in various tumor specimens was then analyzed (Figure 3, see also Figure 2B). Using 0.1% of testis expression as cut-off (which roughly correlated with weak to moderate expression defined by conventional RT-PCR), CT47 expression was found in 4 of 28 (14%) lung cancer specimens, 3 of 20 (15%) esophageal cancer specimens, and 2 of 18 endometrial cancer specimens. In comparison, breast cancer (11 cases), colorectal cancer (15) and bladder cancer specimens (14) had no cases with levels of CT47 mRNA >0.1% of that found in testis.

The CT47 Gene Family. The CT47 gene family and the predicted CT47 gene products are depicted in Figure 4. The CT47 transcript sequence (RefSeq identifier NM_173571) corresponds to a full-length cDNA sequence in the Mammalian Gene Collection (Strausberg, et al. 2002). A BLAST search using this sequence as a probe identified matches in two overlapping BAC clones, RP6-166C19 (AL670379) and RP1-321E8 (AC008162). Manual assembly of the AL670379 and AC008162 sequences resulted in 11

the contig shown in Figure 4. This contig differs from the NCBI 35 assembly in that the region of overlap between the 5’ end of AL670379 and the 3’end of AC008162 was found to be 1.5 kb rather than the 6.3 kb used in the NCBI assembly. CT47.12 was mistakenly deleted from the assembly as a result. The region at the junction of AL670379 and AC008162 is composed of multiple tandem 4.86kb repeats, each containing one copy of the CT47 gene; this repeat structure was undoubtedly responsible for the mis-assembly. The possibility that additional copies of the CT47 gene exist in this region cannot be excluded. Available genome sequence data indicate that there are at least 13 copies of CT47, with 11 exact or almost-exact copies (>99% identity, differing by only 1bp in the entire transcript), and two less exact copies (CT47.6 and CT47.13) that are still >95% identical. The gene model based on the NM_173571 sequence, which is confirmed by ESTs, is applicable for CT47.1 through CT47.12, predicting a transcript size of 1286bp excluding the poly(A) tail, and a coding region of 867bp (288 amino acids). Alignment to the genomic sequence shows that the gene is 3.3 kb in size and contains three exons, with the protein-coding sequences residing on the second and third exons. In comparison, the putative CT47.13 transcript, which contains three exons and would encode a protein of 300 amino acids, has not been documented experimentally. Whether this gene is transcribed in vivo thus remains to be proven. The only putative ortholog found for CT47 was Pan troglodytes LOC465839, also located on chromosome X. While there are still too many gaps in the chimpanzee genomic sequence to reliably count the number of copies of the gene, it is clear that LOC465839 is present in multiple copies. The chimpanzee gene model seems to be 12

incomplete, being truncated at the 5’ end. The C-terminal region of the protein, which is rich in alanine and glutamic acid residues, is 80% identical between the two species. The CT47 protein sequence does not contain any signatures for known protein domains, nor does it significantly match any other protein sequences in the public databases. It is noteworthy that no homologs of CT47 were found in any non-primate species, and that no paralogs, outside of the 13 identified CT47 repeats, seem to be present in the human genome.

Discussion
Two recent publications have investigated the distribution of CT genes on the human X chromosome. Warburton et al (Warburton, et al. 2004) identified 24 inverted repeats in the X chromosome sequence with a size larger than 8 kb and sequence conservation of 95% or more, using the Inverted Repeat Finder software developed by their group. They then scanned the repeats for the presence of known genes, and determined from the literature whether these genes were known to have a testis-specific expression pattern. While this approach confirmed that inverted repeats on the X chromosome are enriched for testis-specific genes, it did not uncover any new CT candidates. The annotators of the completed sequence of the human X chromosome (Ross, et al. 2005) used the sequences of CT genes to scan for the presence of previously undocumented members of known families, and identified 28 previously unknown genes. While this significantly increased the number of identified members in several of the known CT families, it did not provide any data on expression, nor could it detect the presence of new CT families.

13

Our approach is distinct in that we have (1) developed our own gene models instead of relying on existing genome annotations, (2) included repeated genes regardless of size but constrained by experimentally validated expression, and (3) evaluated gene expression patterns using multiple criteria, including the recently generated MPSS data. This approach enabled us not only to identify new members of known CT gene families, but also new gene families, such as the CT47 family. The results presented here confirm that many genes that have undergone recent duplications on the X chromosome are expressed preferentially in the testis. Of the 34 gene families that we detected, 14 showed a preferential or exclusive testis-specific expression based on MPSS and EST data. For all five of the novel gene families that were selected for further study, testis-specific expression was confirmed by RT-PCR. One of these families, CT47, was shown to be expressed in a significant proportion of cancers of the lung and esophagus, making it a bona fide CT gene. To evaluate further whether CT47 is an attractive candidate target for cancer immunotherapy, it will be important to examine CT47 protein expression in tumor samples and its immunogenicity in cancer patients. To this end, we have started producing CT47 recombinant protein and anti-CT47 antibodies. It should be emphasized that the approach that was used in this work could not detect all gene families on the X chromosome, or all genes in each family, because of the stringent thresholds set for expression and sequence conservation. This stringency was mandated by our methodology, which was designed to map transcripts and transcript clusters to unique genomic regions whenever possible. Therefore, further work will be required to identify all multi-copy gene families on the X chromosome, including those 14

that were duplicated earlier in evolution (and which are therefore more divergent), and those for which expression data are still incomplete or missing. It is likely that additional CT genes will be discovered in the process.

Acknowledgements
This work was supported by the Ludwig Institute for Cancer Research and by the Cancer Research Institute.

15

References
Artamonova, II, Gelfand MS. 2004. Evolution of the exon-intron structure and alternative splicing of the MAGE-A family of cancer/testis antigens. J Mol Evol 59(5):62031. Birney E, Andrews TD, Bevan P, Caccamo M, Chen Y, Clarke L, Coates G, Cuff J, Curwen V, Cutts T and others. 2004. An overview of Ensembl. Genome Res 14(5):925-8. Boon T, van der Bruggen P. 1996. Human tumor antigens recognized by T lymphocytes. J Exp Med 183(3):725-9. Chen YT, Scanlan MJ, Venditti CA, Chua R, Theiler G, Stevenson BJ, Iseli C, Gure AO, Vasicek T, Strausberg RL and others. 2005. Identification of cancer/testis-antigen genes by massively parallel signature sequencing. Proc Natl Acad Sci U S A 102(22):7940-5. Chomez P, De Backer O, Bertrand M, De Plaen E, Boon T, Lucas S. 2001. An overview of the MAGE gene family with the identification of all human members of the family. Cancer Res 61(14):5544-51. De Backer O, Arden KC, Boretti M, Vantomme V, De Smet C, Czekay S, Viars CS, De Plaen E, Brasseur F, Chomez P and others. 1999. Characterization of the GAGE genes that are expressed in various human cancers and in normal testis. Cancer Res 59(13):3157-65. Dong XY, Su YR, Qian XP, Yang XA, Pang XW, Wu HY, Chen WF. 2003. Identification of two novel CT antigens and their capacity to elicit antibody response in hepatocellular carcinoma patients. Br J Cancer 89(2):291-7. Florea L, Hartzell G, Zhang Z, Rubin GM, Miller W. 1998. A computer program for aligning a cDNA sequence with a genomic DNA sequence. Genome Res 8(9):967-74. Gure AO, Stockert E, Arden KC, Boyer AD, Viars CS, Scanlan MJ, Old LJ, Chen YT. 2000. CT10: a new cancer-testis (CT) antigen homologous to CT7 and the MAGE family, identified by representational-difference analysis. Int J Cancer 85(5):72632. Gure AO, Wei IJ, Old LJ, Chen YT. 2002. The SSX gene family: characterization of 9 complete genes. Int J Cancer 101(5):448-53. Iseli C, Stevenson BJ, de Souza SJ, Samaia HB, Camargo AA, Buetow KH, Strausberg RL, Simpson AJ, Bucher P, Jongeneel CV. 2002. Long-range heterogeneity at the 3' ends of human mRNAs. Genome Res 12(7):1068-74. Lucas S, De Plaen E, Boon T. 2000. MAGE-B5, MAGE-B6, MAGE-C2, and MAGE-C3: four new members of the MAGE family with tumor-specific expression. Int J Cancer 87(1):55-60.

16

Ross MT, Grafham DV, Coffey AJ, Scherer S, McLay K, Muzny D, Platzer M, Howell GR, Burrows C, Bird CP and others. 2005. The DNA sequence of the human X chromosome. Nature 434(7031):325-37. Sahin U, Tureci O, Schmitt H, Cochlovius B, Johannes T, Schmits R, Stenner F, Luo G, Schobert I, Pfreundschuh M. 1995. Human neoplasms elicit multiple specific immune responses in the autologous host. Proc Natl Acad Sci U S A 92(25):11810-3. Scanlan MJ, Gordon CM, Williamson B, Lee SY, Chen YT, Stockert E, Jungbluth A, Ritter G, Jager D, Jager E and others. 2002a. Identification of cancer/testis genes by database mining and mRNA expression analysis. Int J Cancer 98(4):485-92. Scanlan MJ, Gure AO, Jungbluth AA, Old LJ, Chen YT. 2002b. Cancer/testis antigens: an expanding family of targets for cancer immunotherapy. Immunol Rev 188:2232. Searle SM, Gilbert J, Iyer V, Clamp M. 2004. The otter annotation system. Genome Res 14(5):963-70. Strausberg RL, Feingold EA, Grouse LH, Derge JG, Klausner RD, Collins FS, Wagner L, Shenmen CM, Schuler GD, Altschul SF and others. 2002. Generation and initial analysis of more than 15,000 full-length human and mouse cDNA sequences. Proc Natl Acad Sci U S A 99(26):16899-903. Warburton PE, Giordano J, Cheung F, Gelfand Y, Benson G. 2004. Inverted repeat structure of the human genome: the X-chromosome contains a preponderance of large, highly homologous inverted repeats that contain testes genes. Genome Res 14(10A):1861-9. Zhang Z, Schwartz S, Wagner L, Miller W. 2000. A greedy algorithm for aligning DNA sequences. J Comput Biol 7(1-2):203-14.

17

Tables
Table 1 is submitted as a separate Excel spreadsheet.

Figure Legends
Figure 1. mRNA expression of five CT candidate genes in normal tissues by RT-PCR analysis. Tissues tested included: 1. brain, 2. breast, 3. colon, 4. kidney, 5. liver, 6. lung, 7. pancreas. 8. placenta, 9. prostate, 10. skeletal muscle, 11. spleen, and 12. testis.

Figure 2. CT47 mRNA expression in 21 cell lines (panel A) and in 28 primary lung cancer specimens (panel B), using testis total RNA as positive control (+). The cell lines tested were: (1-21): SK-MEL-10, SK-MEL-24, SK-MEL-37, SK-MEL-49, SK-MEL-55, SK-MEL-80, SK-MEL-128, NCI-H82, NCI-H128, NCI-H187, NCI-H740, SK-LC-5, SK-LC-14, SK-LC-17, HCT-15, LS-174T, SW403B, SW982, SK-HEP-1, SK-RCC-1, T24.

Figure 3. Expression of CT47 in tumor specimens. The expression level was determined by quantitative RT-PCR and expressed as a percentage of the testicular expression level. Each open circle represents one sample. Levels > 0.1% roughly correspond to moderate signals detected by qualitative RT-PCR. The four cases of lung cancer above 0.1%, for example, were the four clearly positive cases in Figure 2B. 18

Figure 4. Genomic organization of the CT47 gene family and transcripts. At least 13 copies of CT47 genes are present on Xq24 as direct tandem repeats, transcribed from the minus strand. The sequence of this segment was derived from BAC clones RP6-166C19 and RP1-321E8. In the NCBI 35 assembly of the human genome, the segment marked by dash-line and asterisk (*) was mistakenly deleted due to the highly repetitive nature in this region, and at least one additional CT47 gene (CT47.12) exists in addition to the Vega annotated genes (Searle, et al. 2004) assigned to this region. The CT47 transcript depicted represents all CT47 genes except CT47.13, which has the same 3-exon structure but is slightly different in size.

19


								
To top