High-resolution YAC fragmentation map of 1q21

SELENBP1 inhibits progression of colorectal cancer by suppressing epithelial–mesenchymal transition

Selenium-binding protein 1 (SELENBP1) is frequently dysregulated in various malignancies including colorectal cancer (CRC); however, its roles in progression of CRCs and the underlying mechanism remain to be elucidated. In this study, we compared the expression of SELENBP1 between CRCs and colorectal normal tissues (NTs), as well as between primary and metastatic CRCs; we determined the association between SELENBP1 expression and CRC patient prognoses; we conducted both in vitro and in vivo experiments to explore the functional roles of SELENBP1 in CRC progression; and we characterized the potential underlying mechanisms associated with SELENBP1 activities. We found that the expression of SELENBP1 was significantly and consistently decreased in CRCs than that in adjacent NTs, while significantly and frequently decreased in metastatic than primary CRCs. High expression of SELENBP1 was an independent predictor of favorable prognoses in CRC patients. Overexpression of SELENBP1 suppressed, while silencing of SELENBP1 promoted cell proliferation, migration and invasion, and in vivo tumorigenesis of CRC. Mechanically, SELENBP1 may suppress CRC progression by inhibiting the epithelial–mesenchymal transition.

Selenium binding protein 1 inhibits tumor angiogenesis in colorectal cancers by blocking the Delta-like ligand 4/Notch1 signaling pathway

•

SELENBP1 localizes to vessels and is suppressed in tumor vessels.

•

SELENBP1 inhibits in vivo and in vitro angiogenesis.

•

SELENBP1 antagonizes tumor angiogenesis by blocking the DLL4/Notch1 signaling pathway.

•

SELENBP1 is a candidate target to treat bevacizumab-resistance in colorectal cancer.

Background

Selenium binding protein 1 (SELENBP1) is frequently downregulated in malignancies such as colorectal cancer (CRC), however, whether it is involved in tumor angiogenesis is still unknown.

Methods

We analyzed the expression and localization of SELENBP1 in vessels from CRC and neighboring tissues. We investigated the in vitro and in vivo activity of SELENBP1 in angiogenesis and explored the underlying mechanism.

Results

SELENBP1 was localized to endothelial cells in addition to glandular cells, while its vascular expression was decreased in tumor vessels compared to that in vessels from neighboring non-tumor tissues. Gain-of-function and loss-of-function experiments demonstrated that SELENBP1 inhibited angiogenesis in vitro, and blocked communications between HUVECs and CRC cells. Overexpression of SELENBP1 in CRC cells inhibited tumor growth and angiogenesis, and enhanced bevacizumab-sensitivity in a mouse subcutaneous xenograft model. Mechanic analyses revealed that SELENBP1 may suppress tumor angiogenesis by binding with Delta-like ligand 4 (DLL4) and antagonizing the DLL4/Notch1 signaling pathway. The inhibitory effects of SELENBP1 on in vitro angiogenesis could largely be rescued by DLL4.

Conclusion

These results revealed a novel role of SELENBP1 as a potential tumor suppressor that antagonizes tumor angiogenesis in CRC by intervening the DLL4/Notch1 signaling pathway.

Downregulation of S100 Calcium Binding Protein A9 in Esophageal Squamous Cell Carcinoma

The development of esophageal squamous cell carcinoma (ESCC) is poorly understood and the major regulatory molecules involved in the process of tumorigenesis have not yet been identified. We had previously employed a quantitative proteomic approach to identify differentially expressed proteins in ESCC tumors. A total of 238 differentially expressed proteins were identified in that study including S100 calcium binding protein A9 (S100A9) as one of the major downregulated proteins. In the present study, we carried out immunohistochemical validation of S100A9 in a large cohort of ESCC patients to determine the expression and subcellular localization of S100A9 in tumors and adjacent normal esophageal epithelia. Downregulation of S100A9 was observed in 67% (n = 192) of 288 different ESCC tumors, with the most dramatic downregulation observed in the poorly differentiated tumors (99/111). Expression of S100A9 was restricted to the prickle and functional layers of normal esophageal mucosa and localized predominantly in the cytoplasm and nucleus whereas virtually no expression was observed in the tumor and stromal cells. This suggests the important role that S100A9 plays in maintaining the differentiated state of epithelium and suggests that its downregulation may be associated with increased susceptibility to tumor formation.

Downregulation of cornulin in esophageal squamous cell carcinoma

Early events in the development of esophageal squamous cell carcinoma (ESCC) are poorly understood and many of the key molecules involved have not yet been identified. We previously used isobaric tags for a relative and absolute quantitation (iTRAQ)-based quantitative proteomics approach to identify differentially expressed proteins in ESCC tissue as compared to the adjacent normal mucosa. Cornulin was identified as one of the major downregulated molecules in ESCC. Cornulin is a member of the S100 fused-type protein family, which has an EF-hand calcium binding motif and multiple tandem repeats of specific peptide motifs. Cornulin was 5-fold downregulated in ESCC as compared to normal epithelium mirroring our previous findings in a gene expression study of ESCC. In the present study, we performed immunohistochemical validation of cornulin (CRNN) in a larger set of patients with ESCC. Downregulation of cornulin was observed in 89% (n=239) of 266 different ESCC tissues arrayed on tissue microarrays (TMAs). Expression of cornulin was observed in the prickle and functional cell layers of normal esophageal mucosa, localized predominantly in the cytoplasm and perinuclear region. The large majority of ESCC cases had little or no expression of cornulin in the carcinoma or stroma. These findings suggest that cornulin is an important molecule in normal esophageal pathology and is likely lost during the conversion of normal to neoplastic epithelium.

Expression and clinical significance of S100A2 and p63 in esophageal carcinoma

AIM: To investigate the expression and clinical significance of S100A2 mRNA and protein, p63 protein in esophageal squamous cell carcinoma (ESCC) and their roles in carcinogenesis and progression of esophageal carcinoma (EC).

METHODS: Immunohistochemical staining (S-P method) for S100A2 and p63 protein were performed in 40 samples of ESCC and 40 samples of normal esophageal mucosa. In situ hybridization (ISH) was used to detect the expression of S100A2 mRNA.

RESULTS: Expression of S100A2 mRNA in ESCC was positive in 77.5% of samples, which was lower than that in normal mucosa (100%) by ISH (P = 0.002). The expression level of S100A2 mRNA was closely related to differentiation and and node-metastasis (P = 0.012, P = 0.008). Expression of S100A2 protein was positive in 72.5% of ESCC samples and expression of p63 protein was positive in 37.5% of ESCC samples, and was lower than that in normal mucosa (100%) (P = 0.000). The expression of S100A2 protein was correlated with the differentiation and node-metastasis (P = 0.007, P = 0.001), but no relationship was observed between the expression of p63 protein and clinical pathological manifestations. S100A2 protein was positively correlated with the expression of S100A2 mRNA, and negatively associated with the expression of p63 protein (P = 0.000, P = 0.002).

CONCLUSION: S100A2 and p63 protein both play important roles in the carcinogenesis of ESCC. An investigation into the combined expression of S100A2 and p63 may be helpful in early diagnosis and in evaluating the prognosis of ESCC.

Identification of regulatory elements by gene family footprinting and in vivo analysis

Gene families of recently duplicated but subsequently diverged genes provide an unique opportunity for comparative analysis of regulatory elements. We have studied the human SPRR gene family of small proline rich proteins involved in barrier function of stratified squamous epithelia. These genes are all expressed in normal human keratinocytes, but respond differently to environmental insults. Comparisons of the functional promoter regions allows the rapid identification of both conserved and of novel regulatory elements that appeared after gene duplication. Competitive electrophoretic mobility shift assays can be used to confirm their presence. Here we show the power of gene family footprinting by the identification of two novel elements in the SPRR3 promoter, not present in SPRR1A and SPRR2A. One of these elements binds a protein similar to GAAP-1, a pro-apoptotic activator of IRF-1 and p53. In vivo analysis shows that this element functions as an inhibitor of SPRR3 transcription. The second novel element functions as an activator of promoter activity and is characterized by its A/T rich sequence. The latter interacting protein indeed binds through contacts in the minor groove, and strikingly, depends on the presence of calcium for DNA interaction.

S100A2, a potential marker for early recurrence in early-stage oral cancer

Early-stage oral cancer patients may have distinct clinical outcomes and respond differently to the same treatment. Up to now, there is still no individual marker to identify such patients with poor outcome. Down-regulation of a tumor suppressor gene, S100A2, in oral cancer cells was identified by mRNA profiling analysis then confirmed by RT-PCR and Southern blotting. The expression of nuclear S100A2 protein examined by immunohistochemistry was not significantly associated with any patient characteristic among the 70 early-stage oral squamous cell carcinoma (SCC) patients. Intriguingly, the loss of nuclear S100A2 positivity was significantly associated with shorter disease-free survival (p=0.019) while having no effect on the overall survival of these patients. Cox regression analysis with backward elimination identified S100A2 (p=0.006), tobacco smoking (p=0.013), and betel quid chewing (p=0.019) as independent predictors of disease-free survival. This is a first study to demonstrate that loss of nuclear S100A2 may serve as an independent prognostic marker for early-stage oral cancer patients at high risk of recurrence. A more aggressive treatment modality and intensive follow-up may be recommended for the patients with reduced expression of S100A2 in tumor cell nuclei.

Cornulin, a New Member of the “Fused Gene” Family, Is Expressed During Epidermal Differentiation

The protein encoded by the C1orf10 gene was described to be esophageal-specific and a marker for cancer development. This protein, however, has the previously unreported structural features of the "fused gene" family combining sequences and structural similarities of both the S100 proteins and precursor proteins of the cornified cell envelope as in profilaggrin, trichohyalin, and repetin. Since all members of this family are expressed in keratinocytes, we suspected a role in epidermal differentiation and named the protein cornulin. Here, we report that human cornulin mRNA is expressed primarily in the upper layers of differentiated squamous tissues including the epidermis. Using polyclonal peptide antibodies, we show that cornulin is expressed in the granular and lower cornified cell layers of scalp epidermis and foreskin, as well as in calcium-induced differentiated cultured keratinocytes. Ca(2+)-overlay assay indicated that EF-hand domains of cornulin are functional and bind calcium. In HeLa cells, cornulin, co-transfected with transglutaminase 1, was diffusely distributed throughout the cytoplasm in contrast to small proline-rich 4, which localized to the cell periphery. We conclude that cornulin is a new member of the "fused gene" family, does not appear to be a precursor of the cornified cell envelope by itself, and is a marker of late epidermal differentiation.

Seeking candidate mutations that affect iron homeostasis

Hereditary hemochromatosis is characterized by marked variation of expression of the defect: very few homozygotes with the C282Y/C282Y HFE genotype have full-blown clinical disease, a larger number show biochemical stigmata of iron overload, and some seem normal biochemically. The following candidate genes have been examined in detail to determine whether polymorphisms in them may be responsible for this variation: transferrin, transferrin receptor 1, transferrin receptor 2, ferritin-L, ferritin-H, IRP1, IRP2, HFE, beta(2) microglobulin, mobilferrin/calreticulin, ceruloplasmin, ferroportin, NRAMP1, NRAMP2 (DMT1), haptoglobin, heme oxygenase-1, heme oxygenase-2, hepcidin, USF2, ZIRTL, duodenal cytochrome b ferric reductase (DCYTB), TNFalpha, keratin 8, and keratin 18. The coding sequence, exon-intron junctions, and promoters of each of these genes was sequenced in DNA from 20 subjects: 5 HFE C282Y/C282Y with clinical disease, 5 HFE C282Y/C282Y with normal/low ferritin levels and no disease, 5 wt/wt with high ferritin and transferrin saturation, and 5 wt/wt normal controls. When coding or promoter polymorphisms were encountered, DNA from large numbers of ethnically defined subjects was examined for these polymorphisms and a relationship between their existence and abnormalities of iron homeostasis was sought. Only in the case of one transferrin mutation did we find a strong relationship between the polymorphism and iron deficiency anemia. The putative genes that affect the expression of HFE mutations remain elusive.

A combinatorial approach for fast, high-resolution mapping

High-resolution physical maps can be used as a scaffold for several subsequent studies, such as sequencing projects and positional cloning of disease genes and genetic elements that regulate gene expression. Here we describe a method for fast, high-resolution physical mapping on stretched DNA molecules, based on a combinatorial multi-FISH approach. Fluorescent labels are assigned to a binary code and probes are identified by a binary tag according to their labeling. To validate the approach, we have mapped eight probes covering a region of about 300 kb on human chromosome 11 with three hybridization assays. This approach enables one to determine the structural organization of a large region by means of the order of its clones, without ambiguities. The structure established in a control cell constitutes a reference for further studies, to detect rearrangements displayed by disease cells and to find differences shown by different cell types and organisms.

A common element in epidermal expression?

Members of an epidermally expressed gene cluster on human chromosome 21 each contain a short sequence similar to an element that can drive ectoderm-specific gene expression.

Positional cloning identifies a novel cyclophilin as a candidate amplified oncogene in 1q21

Gains of 1q21-q23 have been associated with metastasis and chemotherapy response, particularly in bladder cancer, hepatocellular carcinomas and sarcomas. By positional cloning of amplified genes by yeast artificial chromosome-mediated cDNA capture using magnetic beads, we have identified three candidate genes (COAS1, -2 and -3) in the amplified region in sarcomas. COAS1 and -2 showed higher amplification levels than COAS3. Most notably, amplification was very common in osteosarcomas, where in particular COAS2 was highly expressed. COAS1 has multiple repeats and shows no homology to previously described genes, whereas COAS2 is a novel member of the cyclosporin-binding peptidyl-prolyl isomerase family, very similar to cyclophilin A. COAS2 was overexpressed almost exclusively in aggressive metastatic or chemotherapy resistant tumours. Although COAS2 was generally more amplified than COAS1, it was not expressed in well-differentiated liposarcomas, where amplification of this region is very common. All three genes were found to be amplified and over-expressed also in breast carcinomas. The complex nature of the 1q21-23 amplicons and close proximity of the genes make unequivocal determination of the gene responsible difficult. Quite likely, the different genes may give selective advantages to different subsets of tumours.

Subchromosomal positioning of the epidermal differentiation complex (EDC) in keratinocyte and lymphoblast interphase nuclei

The epidermal differentiation complex (EDC) at 1q21 is host to many structurally and functionally related genes coding for proteins involved in the differentiation process of keratinocytes. The grouping together of these genes which share spatial and temporal expression and interrelated functions is a remarkable genomic feature which has led to suggestions that the region may have a coordinated transcription control mechanism. With the growing awareness that the organization of the genome within the interphase nucleus is relevant to transcriptional activity, we have investigated the spatial organization of the EDC in the nuclei of keratinocytes, where the EDC genes are highly expressed, and lymphoblasts, where they are silent. Using 2D and 3D FISH we find that in keratinocyte nuclei the EDC is frequently positioned external to the chromosome 1 territory compared to lymphoblasts where the EDC more often adopts a peripheral or internal location. It has been previously shown that the MHC region can extend from the chromosome 6 territory in relation to transcriptional activity. This study of the EDC thus provides a further example of a gene-dense complex capable of assuming extraterritorial positioning in relation to cell type/transcription status.

Novel evidence of a role for chromosome 1 pericentric heterochromatin in the pathogenesis of B-cell lymphoma and multiple myeloma

1q rearrangement is a remarkably frequent secondary chromosomal change in both non-Hodgkin's lymphoma (NHL) and multiple myeloma (MM), where it is associated with tumor progression. To gain insight into 1q rearrangement-associated disease mechanisms, we used fluorescence in situ hybridization (FISH) to search for recurring 1q breaks in 35 lymphoma samples (31 NHL patients and 4 lymphoma-derived cell lines) as well as 22 MM patients with cytogenetically determined 1q abnormalities. Strikingly, dual-color FISH analysis with chromosome 1 centromere and 1q12-specific probes identified constitutive heterochromatin band 1q12 as the single most frequent breakpoint site in both NHL and MM (39% and 89% of 1q breaks, respectively). These rearrangements consistently generated aberrant heterochromatin/euchromatin junctions and gain of 1q12 material. A further 30% of NHL 1q breaks specifically involved two other novel, closely spaced sites (clusters I and II) within a 2.5 Mb region of proximal 1q21 (D1S3620 to D1S3623). A possible association between these sites and NHL subtype was evident; the cluster I rearrangement was frequent in follicular and diffuse large cell lymphoma, whereas the cluster II rearrangement was more frequently observed in diffuse small-cell lymphoma (2/2 marginal zone lymphomas, 1/2 atypical chronic lymphocytic leukemias, and 1 lymphoplasmacytic lymphoma in this series). Candidate oncogenes bordering this interval (BCL9 and AF1Q) were not rearranged in any patient except one (AF1Q). This study provides the first evidence of involvement of 1q12 constitutive heterochromatin in the pathogenesis of NHL and MM and indicates proximal 1q21 to be of specific pathological significance in NHL.

Identification of human epidermal differentiation complex (EDC)-encoded genes by subtractive hybridization of entire YACs to a gridded keratinocyte cDNA library

The epidermal differentiation complex (EDC) comprises a large number of genes that are of crucial importance for the maturation of the human epidermis. So far, 27 genes of 3 related families encoding structural as well as regulatory proteins have been mapped within a 2-Mb region on chromosome 1q21. Here we report on the identification of 10 additional EDC genes by a powerful subtractive hybridization method using entire YACs (950_e_2 and 986_e_10) to screen a gridded human keratinocyte cDNA library. Localization of the detected cDNA clones has been established on a long-range restriction map covering more than 5 Mb of this genomic region. The genes encode cytoskeletal tropomyosin TM30nm (TPM3), HS1-binding protein Hax-1 (HAX1), RNA-specific adenosine deaminase (ADAR1), the 34/67-kD laminin receptor (LAMRL6), and the 26S proteasome subunit p31 (PSMD8L), as well as five hitherto uncharacterized proteins (NICE-1, NICE-2, NICE-3, NICE-4, and NICE-5). The nucleotide sequences and putative ORFs of the EDC genes identified here revealed no homology with any of the established EDC gene families. Whereas database searches revealed that NICE-3, NICE-4, and NICE-5 were expressed in many tissues, no EST or gene-specific sequence was found for NICE-2. Expression of NICE-1 was up-regulated in differentiated keratinocytes, pointing to its relevance for the terminal differentiation of the epidermis. The newly identified EDC genes are likely to provide further insights into epidermal differentiation and they are potential candidates to be involved in skin diseases and carcinogenesis that are associated with this region of chromosome 1. Moreover, the extended integrated map of the EDC, including the polymorphic sequence tag site (STS) markers D1S1664, D1S2346, and D1S305, will serve as a valuable tool for linkage analyses.

Exclusion of ZIRTL as candidate gene of juvenile hemochromatosis and refinement of the critical interval on 1q21

Hemochromatosis type 2 (HFE2) or juvenile hemochromatosis (JH) is a rare recessive disorder that causes iron overload, characterized by early onset and severe clinical course. The JH locus maps to chromosome 1q, in a 4-cM region encompassing markers D1S442 and D1S2347. Recently a gene named ZIRTL has been characterized and mapped to 1q21. This gene belongs to a family of divalent metal ion-transporting genes that encode for proteins involved in transport of different metals, including iron. Thus, the ZIRTL gene represents a positional and functional candidate for JH. Here we further restrict the candidate region through segregation analysis of two new polymorphic markers and haplotype analysis in JH families. Furthermore, we exclude ZIRTL as a JH candidate gene showing that it maps outside the critical interval and that its genomic sequence is normal in three patients.

Isolation and characterization of human and mouse ZIRTL, a member of the IRT1 family of transporters, mapping within the epidermal differentiation complex

We report the precise mapping and characterization of ZIRTL (zinc-iron regulated transporter-like) gene, the first mammalian member of an extensive family of divalent metal ion transporters, comprising IRT1 and ZIP1, ZIP2, ZIP3, and ZIP4 in plants and ZRT1 and ZRT2 in yeast. The human gene maps at the telomeric end of the epidermal differentiation complex (EDC), within chromosomal band 1q21, while the mouse gene maps within the mouse EDC, on mouse chromosome 3, between S100A9 and S100A13. The structure of the human gene has been determined, and message was detected in most adult and fetal tissues including the epidermis. The mouse gene is developmentally regulated and found expressed in fetal and adult suprabasal epidermis, osteoblasts, small intestine, and salivary gland.

Human epidermal differentiation complex in a single 2.5 Mbp long continuum of overlapping DNA cloned in bacteria integrating physical and transcript maps

Terminal differentiation of keratinocytes involves the sequential expression of several major proteins which can be identified in distinct cellular layers within the mammalian epidermis and are characteristic for the maturation state of the keratinocyte. Many of the corresponding genes are clustered in one specific human chromosomal region 1q21. It is rare in the genome to find in such close proximity the genes belonging to at least three structurally different families, yet sharing spatial and temporal expression specificity, as well as interdependent functional features. This DNA segment, termed the epidermal differentiation complex, contains 27 genes, 14 of which are specifically expressed during calcium-dependent terminal differentiation of keratinocytes (the majority being structural protein precursors of the cornified envelope) and the other 13 belong to the S100 family of calcium binding proteins with possible signal transduction roles in the differentiation of epidermis and other tissues. In order to provide a bacterial clone resource that will enable further studies of genomic structure, transcriptional regulation, function and evolution of the epidermal differentiation complex, as well as the identification of novel genes, we have constructed a single 2.45 Mbp long continuum of genomic DNA cloned as 45 p1 artificial chromosomes, three bacterial artificial chromosomes, and 34 cosmid clones. The map encompasses all of the 27 genes so far assigned to the epidermal differentiation complex, and integrates the physical localization of these genes at a high resolution on a complete NotI and SalI, and a partial EcoRI restriction map. This map will be the starting resource for the large-scale genomic sequencing of this region by The Sanger Center, Hinxton, U.K.

JTB: a novel membrane protein gene at 1q21 rearranged in a jumping translocation

1q21 is frequently involved in different types of translocation in many types of cancers. Jumping translocation (JT) is an unbalanced translocation that comprises amplified chromosomal segments jumping to various telomeres. In this study, we identified a novel gene human JTB (Jumping Translocation Breakpoint) at 1q21, which fused with the telomeric repeats of acceptor telomeres in a case of JT. hJTB (human JTB) encodes a trans-membrane protein that is highly conserved among divergent eukaryotic species. JT results in a hJTB truncation, which potentially produces an hJTB product devoid of the trans-membrane domain. hJTB is located in a gene-rich region at 1q21, called EDC (Epidermal Differentiation Complex). This is the first report identifying the gene involved in unbalanced translocations at 1q21.

YAC fragmentation with repetitive and single-copy sequences: detailed physical mapping of the presenilin 1 gene on chromosome 14

We constructed new LYS2 fragmentation vectors that allow direct acentric and centric fragmentation of yeast artificial chromosomes (YACs) and selection of fragmented YACs in yeast strain AB1380. The fragmentation vectors were used efficiently with repetitive (e.g., Alu), low-copy (e.g., CA-repeats) and single-copy (e.g., exons) sequences. High recombination efficiencies were obtained in fragmenting two different CEPH YACs with the Alu consensus sequence as target sequences for homologous recombination. Analysis of the acentric Alu fragmentation panel of 788H12, containing the presenilin 1 (PSEN1) gene for familial Alzheimer's disease (AD), indicated that high-resolution YAC fragmentation panels covering the entire parent YAC are obtained. Also, marker content analysis of the fragmentation panel indicated that fragmented YACs were propagated stably without rearrangements. The same fragmentation vectors were used efficiently for fragmentation of 788H12 with unique sequences, i.e., exons 3 and 12 of PSEN1 and D14S77, a polymorphic CA repeat, as target sequences. Together, our YAC fragmentation data of 788H12 provided a size estimate for the coding region of PSEN1 of 60kb and a more precise localization of D14S77 at 25kb upstream of PSEN1.

A Sequence-Ready BAC Clone Contig of a 2.2-Mb Segment of Human Chromosome 1q24

Human chromosomal region 1q24 encodes two cloned disease genes and lies within large genetic inclusion intervals for several disease genes that have yet to be identified. We have constructed a single bacterial artificial chromosome (BAC) clone contig that spans over 2 Mb of 1q24 and consists of 78 clones connected by 100 STSs. The average density of mapped STSs is one of the highest described for a multimegabase region of the human genome. The contig was efficiently constructed by generating STSs from clone ends, followed by library walking. Distance information was added by determining the insert sizes of all clones, and expressed sequence tags (ESTs) and genes were incorporated to create a partial transcript map of the region, providing candidate genes for local disease loci. The gene order and content of the region provide insight into ancient duplication events that have occurred on proximal 1q. The stage is now set for further elucidation of this interesting region through large-scale sequencing.

[The sequence data described in this paper have been submitted to GenBank under accession nos. G42259–G42312 and G42330–G42335.]