QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

First published online 16 December 2004
doi: 10.1242/dev.01579

This Article Summary Full Text Full Text (PDF) Supplementary Material Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Lepourcelet, M. Articles by Shivdasani, R. A. Search for Related Content PubMed PubMed Citation Articles by Lepourcelet, M. Articles by Shivdasani, R. A.

Insights into developmental mechanisms and cancers in the mammalian intestine derived from serial analysis of gene expression and study of the hepatoma-derived growth factor (HDGF)

Maina Lepourcelet^1,2,*, Liqiang Tou^1,2,*, Li Cai¹, Jun-ichi Sawada^1,3, Alexander J. F. Lazar⁴, Jonathan N. Glickman⁴, Jessica A. Williamson¹, Allen D. Everett⁵, Mark Redston⁴, Edward A. Fox^1,6, Yoshihiro Nakatani^1,3 and Ramesh A. Shivdasani^1,2,

¹ Dana-Farber Cancer Institute, 44 Binney Street, Boston, MA 02115, USA
² Department of Medicine, Harvard Medical School, Boston, MA 02115, USA
³ Department of Biological Chemistry and Molecular Pharmacology, Harvard Medical School, Boston, MA 02115, USA
⁴ Department of Pathology, Brigham & Women's Hospital, 75 Francis Street, Boston, MA 02115, USA
⁵ Department of Pediatric Cardiology, Johns Hopkins University School of Medicine, Baltimore, MD 21287, USA
⁶ Department of Pediatrics, Harvard Medical School, Boston, MA 02115, USA

View larger version (66K): [in a new window]   Fig. 1. Tissue and molecular anatomy of mouse intestine development. (A) Histology of the mid-gestation small intestine revealed in Hematoxylin- and Eosin-stained transverse sections. Gut endoderm (arrows) transforms into a simple villous epithelium between E13 and E15. Also shown are temporal clusters of a subset of SAGE tags representing 703 genes that vary significantly (P<0.015) over the E12-E15 interval. Each colored line corresponds to one SAGE tag and denotes its relative frequency among the three SAGE libraries. (B) Validation and extension of SAGE results for representative transcripts from three classes of temporally regulated genes [increasing: apolipoprotein 1A (Apoa1) and apoptosis-specific protein with CARD domain (Asc; Pycard – Mouse Genome Informatics); decreasing: parathyroid hormone receptor (Pthr1) and Gli1; and dynamically regulated: zinc-finger regulator of apoptosis and cell-cycle arrest (Zac1; Plagl1 – Mouse Genome Informatics)]. RT-PCR analysis of gut tissue isolated at four fetal stages is compared with E13 and E15 SAGE data (tag numbers). Glyceraldehyde 3-phosphate dehydrogenase (Gapd) serves as a loading control and the lack of PCR products in samples not treated with reverse transcriptase (RT–) indicates absence of DNA in the RNA preparations. (C) RNA in situ hybridization analysis to distinguish predominantly epithelial (e.g. CD151) from stromal (e.g. growth receptor-binding protein 10, Grb10) transcripts. Sense probes (not shown) gave negligible staining. (D) Characterization of one gene product (proline-rich acidic protein 1, Prap1) suggested in SAGE to represent a marker of intestine differentiation. RT-PCR and in situ hybridization analysis confirm onset of epithelial expression after E13. In adults, Prap1 expression is concentrated in rostral segments, duodenum (D) and jejunum (J), compared with the ileum (I) or colon (C). (E) Northern analysis confirms SAGE data that a group of genes traditionally associated with the cellular stress response peaks in expression around E13, coincident with the villous epithelial transition. Sample results are shown for peptidylprolyl isomerase C (Ppic), calreticulin (Calr), FK506-binding protein 9 (Fkbp9) and stress-induced phosphoprotein 1 (Stip1). N.T., not tested. In situ hybridization localized expression of these transcripts to the epithelial compartment, as shown for Calr.

View larger version (57K): [in a new window]   Fig. 2. Expression of developmentally regulated genes in the adult mouse intestine. E11-E17 fetal gut and adult (12-week) duodenum (D), jejunum (J), ileum (I) and colon (C) were analyzed by RT-PCR for selected transcripts that display developmentally regulated expression in SAGE analysis. Representative examples are shown for three categories, defined according to the relationship among expression levels at E13, E15 and E17, i.e. Increasing, Dynamically regulated or Decreasing; numbers in parentheses refer to the number of transcripts investigated in each class.

View larger version (59K): [in a new window]   Fig. 3. Re-expression of developmentally attenuated genes in human colorectal cancers. (A) Normalization of cDNAs prepared from five of seven independent pairs of tumors (T) and matched normal (N) colonic tissue by RT-PCR analysis of ß-amyloid mRNA, which is known to be expressed to similar levels in cancerous and normal human colon. Labels refer to serial dilution of cDNA before PCR. (B) Expression (RT-PCR) analysis for the seven out of 27 transcripts that show increased expression in two or more of seven T-N pairs, and illustrative results for two transcripts that show minimal change or inconsistent differences between T and N. (C) Results were quantified by gel densitometry, as shown here for one transcript, which encodes a 68 kDa TGF-ß-induced protein. (D) Immunohistochemistry reveals the representative oncofetal marker Trp53 to be expressed in tumor cells and not in the surrounding stroma.

View larger version (88K): [in a new window]   Fig. 4. Expression of the hepatoma-derived growth factor (HDGF) in development and neoplasia. (A) Relative expression levels of HDGF mRNA (left) and protein (right) in the mouse small intestine at the indicated fetal stages, assessed by semi-quantitative RT-PCR and immunoblot analysis, respectively. (B) Localization of HDGF mRNA (top) and protein (bottom) in the fetal mouse small intestine at the indicated developmental stages, as judged by in situ hybridization and immunohistochemistry, respectively. (C) Localization of HDGF in the adult (12-week-old) mouse small intestine, assessed by immunohistochemistry (left) and immunofluorescence (right). Exposure times were increased over fetal samples to permit localization of weaker signals. (D,E) Strong expression of HDGF in human colon tumors (T) relative to adjacent normal (N) mucosa, shown here in representative low- (D, 100x) and high- (E, 600x) magnification photomicrographs. (F) Relative HDGF staining signals in human colon cancer specimens according to the status of DNA mismatch repair (MMR) in the tumor, which was determined separately by the presence or absence of MMR gene (Mlh1 and Msh2) expression.

View larger version (30K): [in a new window]   Fig. 5. HDGF properties and associations revealed through sequence similarities and isolation of a nuclear multi-protein complex. (A) Rapid release of nuclear HDGF but not histone 2Az protein following treatment of cultured HeLa cells with low concentrations of NP-40 detergent. This behavior parallels that of the related protein HMGB1 (Scaffidi et al., 2002) and may explain why HDGF previously was found intact in conditioned culture media. (B) Silver-stained gel of HDGF immunoprecipitates from mock-transfected or epitope-tagged HDGF-overexpressing HeLa cells resolved by SDS-PAGE. The marked specific bands were extracted and identified by mass spectrometric analysis to contain the indicated proteins: two novel factors (Fig. S1), hnRNPs K and I, and TLS/Fus, also known as mouse pigpen. (C) Immunoblot analysis of HeLa cell (top) and E13 mouse intestine (bottom) nuclear fractions resolved over a glycerol gradient and probed for the presence of HDGF, hnRNPK and TLS/Fus proteins. Resolution of protein complexes is similar in the two cell sources and indicates that a significant proportion of nuclear HDGF may be complexed with hnRNPK (red) and TLS/Fus (green), which in turn may associate independently with each other and with other factors in bulkier protein complexes. (D) Immunofluorescence analysis of subcellular localization of HDGF and TLS/Fus (similar results for hnRNPK are not shown) reveals the abundance of each in tiny nuclear dots, although the resolution cannot unambiguously define protein association in this context. (E) mRNA (RT-PCR, left) and protein (immunoblot, right) levels of the putative HDFG-associated factors TLS/Fus and hnRNPs K and I are downregulated in tandem with HDGF during mouse intestine development. These results reinforce the possibility of the four proteins functioning within a common cellular pathway that regulates epithelial differentiation and cancer. N.T., not tested.

View larger version (64K): [in a new window]   Fig. 6. Overexpression of HDGF in mouse fetal gut explants retards epithelial differentiation. (A) UV micrograph showing expression of GFP-tagged HDGF in a sample E14 explant 18 hours after plasmid electroporation. (B) Immunoblot confirming that full-length and mutant (m) HDGF were expressed to similar levels. Duplicate samples are shown for each; the prominent faster-migrating protein band represents a non-specific immunoblot signal and surrogate loading control. (C) RT-PCR analysis of cultured intestinal explants for molecular markers of gut epithelial differentiation after forced expression of GFP (CTL) or GFP-tagged wild-type or mutant (m) HDGF. Explant RNA was isolated 1 and 2 days after transfection and analyzed by RT-PCR for transcript levels of differentiation markers: apolipoprotein A1 (Apo1a), liver (Fabpl) and intestinal (Fabpi) fatty acid-binding proteins, proline-rich acidic protein 1 (Prap1), villin and metallothionein 2 (Mt2). The results represent five independent experiments with intact and two separate studies with the mutant form of HDGF. Gapd, loading control. RT+ and RT– refer to mRNA samples treated with and without reverse transcriptase, respectively.