Colorectal Cancer in Mice Genetically Deficient in the Mucin Muc2 Anna Velcich, WanCai Yang, Joerg Heyer, Alessandra Fragale, Courtney Nicholas, Stephanie Viani, Raju Kucherlapati, Martin Lipkin, Kan Yang, and Leonard Augenlicht

Generation of targeting vector. A 2.7 kilobase (kb) Xho II fragment encompassing the region upstream of the Muc2 mRNA start site up to the beginning of exon 2 (homology 1) was cloned into the Bgl II site of pKO901. A 6.5 kb Eco RI Muc2 genomic fragment, which encompassed the end of Muc2 exon4 and extended further downstream, was cloned into the Eco RI site of pKO901-hom1 vector. An Asc I fragment from the pKO select Neo plasmid, containing the Neo gene under the control of the PGK promoter was cloned into the AscI site of the pKO901-hom1/hom2 plasmid. The transcriptional orientation of the Neo cassette was opposite to that of the Muc2 gene. The cloning into pKO901-hom1/hom2/Neo of the Herpes thymidine kinase gene, as a Rsr II fragment from the pKO SelectTK plasmid, provided negative selection. All plasmids were from Lexicon Genetics. The resulting vector, linearized, was electroporated into E14-1 ES cells; cells were selected with G418 and gancyclovir as previously described (1). DNA was isolated from 120 selected clones and analyzed by nested PCR utilizing primers in the Muc2 sequence upstream of the recombination site and downstream primers in the Neo cassette. The primers were: forward 5'GTTCGTATCAGGTCATTCAG; 5'CTCCATCTAGGGAGGATCTAC (nested PCR). Reverse 5' TGGGAAGACAATAGCAGG; 5' TTCTGAGGCGGAAGGAAC (nested PCR). The nested PCR product was 2865 nucleotides long. Two independently isolated Muc2^+/- ES cell clones were microinjected into C57Bl/6 blastocysts to generate chimeras. Chimeric male mice derived from each cell line were mated with C57BL/6J females, and germline transmission of mutant Muc2 allele was obtained from both lines. The resulting F₁ progeny was intercrossed to obtain F₂ mice. Routine genotype analysis of the F₂ mice was performed by PCR assay on DNA purified from tail biopsies. The wild type allele [nucleotide 280 (nt 280)] and the mutant allele (nt 328) were detected utilizing a common primer in intron 1 (5' TCCACATTATCACCTTGAC), a primer in the Neo gene (5'GATTGGGAAGACAATAG), and a primer from exon2 which was retained only in the wild type allele (5'AGGGAATCGGTAGACATC).

Total RNA isolation and analysis. Total RNA was isolated from snap frozen segments of the small and large intestine by homogenization in Trizol (Gibco-BRL) and analyzed by Northern blot, as described (2).

The following probes were used: for Muc2 PH666 (3); for Muc5ac a cDNA comprising two tandem repeats and unique 3'sequence (4); for Muc3 an RT-PCR generated probe comprising nt 637-1365 of the published sequence (5); for Muc13 an RT-PCR generated probe spanning nt 177-943 of the published sequence (6); for Itf an RT-PCR generated probe encompassing nt 211-429 of the published sequence (7)

Mice. Mice were maintained in a barrier facility. However, both wild-type and Muc2^-/- mice harbored Helicobacter hepaticus, which is endemic in animal facilities and generally produces no phenotype. Mice did not show clinical and histological evidence of inflammatory bowel disease (IBD) that has been linked to H. hepaticus colonization in immunocompromised animals (8, 9). For tumor analysis, mice were sacrificed at 6 months or 1 year of age ± 1 month.

Histology and immunohistochemistry. Mice were sacrificed by CO₂ overdose and rapidly dissected. Briefly, segments of the gastrointestinal tract were fixed in 10% buffered formalin and paraffin embedded. For tumor evaluation the intestine was opened longitudinally, formalin fixed and analyzed under a dissecting microscope for the presence of tumors (10). Tumors were dissected and paraffin embedded. Sections (5 m) were cut and stained with haematoxylin and eosin to assess morphology and with Alcian blue to identify goblet cells.

For immunohistochemical analysis sections were stained with the following antibodies: PH497, rabbit polyclonal anti-rat Muc2 (1:3000 (11), a gift of G. Hansson, Goteborg University, Sweeden). ITF, rabbit polyclonal anti-human ITF (1: 800, a gift of K. Tomasetto, INSERM, Strasbourg, France). Anti--catenin, mouse-monoclonal, (1:50, Transduction Laboratories, Lexington, KY, (12). C-MYC, rabbit polyclonal (1:250, Santa Cruz Biotechnology, Santa Cruz, CA). Prior to incubation with the antibodies, antigen retrieval was performed as follow: for -catenin, sections were heated at 95 to 97°C in 0.01M sodium citrate pH6 for 15 minutes. For c-Myc, sections were heated at 95°C in 0.01M EDTA pH 7.5 for 15 minutes (Cattoretti, personal communication). Endogenous peroxidase activity was quenched by treatment with 1.5-3 % hydrogen peroxide in water for 10 minutes at room temperature (RT); sections were then incubated with either 10 % normal goat serum in PBS or blocking buffer [2.5% nonfat dry milk, 2% BSA in TBS-T (1X TBS + 0.05% Tween-20)] for 40 min at RT, followed by incubation with the primary antibody, diluted in blocking buffer, over night at 4°C. For the -catenin antibody incubation time was 2 hours at RT. Following incubation with the secondary antibody, the Vectastain ABC kit (Vector Laboratories, Burlingame, CA) was used for detection with diaminobenzidine tetrahydrochloride (Sigma) as the substrate. Sections were counterstained with hematoxylin. Negative control sections were incubated with pre-immune serum or without primary antibody.

Apoptotic cells were detected by TUNEL assay using an in situ Apoptosis Detection Kit (Trevigen, Inc., Gaithersburg, MD) as described (13). We studied nine mice/genotype. The number of cells per crypt was determined by counting at least 50 crypt columns per mouse. Crypts had to be well oriented, visible from bottom to top, and show an open intracrypt lumen. Cells synthesizing DNA were labeled by injecting mice (three mice/genotype/time point; these mice were an experimentally distinct group from that for tumor determination) intraperitoneally with a mixture of 5'-bromo-2'-deoxyuridine (BrdU, 120 mg/kg) and 5'-fluoro-2'-deoxyuridine (12 mg/kg) dissolved in normal saline solution. Mice were sacrificed 2 hours post BrdU injection for analysis of cell proliferation, and 24 and 48 hours later for analysis of cell migration. Formalin fixed sections were prepared and cells, which had incorporated BrdU, were visualized using a BrdU staining Kit from Oncogene Research Products.

Supplemental Figure 1. Targeted disruption of the Muc2 gene. (A) Structure and partial restriction enzyme map of the genomic segment of Muc2 utilized to generate the targeting vector. Muc2 exons and introns are indicated by filled boxes and lines, respectively. The vector's sequence is depicted by a dotted line. Also indicated are the segments corresponding to homology1 (Hom1) and homology 2 (Hom2). In the inactivated allele a portion of the Muc2 gene from exon 2 to exon 4 is replaced by the Neo cassette, which is inserted in the opposite transcriptional orientation of that of the Muc2 gene, as indicated by the arrows. Restriction enzymes are as follow: Bam HI (B); Sma I (S); Hind III (H); Eco RI (RI); Xho II (X). (B) Southern blot analysis of DNA isolated from ES clones and F₂ mice (lanes 1-11). DNA was digested with Bam HI and Kpn I and hybridized with probe P1, shown in (A). The 5.1 and 3.5 kb bands correspond to the wild type and mutant allele, respectively. Relevant DNA markers, in kilobases, are also marked. (C) Muc2 mRNA levels in total RNA isolated from the indicated tissues from wild-type and Muc2^-/- mice were analyzed by Northern blot (1, stomach; 2, duodenum; 3, jejunum; 4, ileum; 5, cecum; 6, proximal colon; 7, distal colon; 8, liver); the probe was PH666, a partial rat Muc2 cDNA that hybridizes to sequences located 3' of the recombination event.

Medium version | Full size version

Supplemental Figure 2. Absence of Muc2 apomucin and alteration of cell maturation in the Muc2^-/- intestine. Sections from the indicated portions of the intestine of wild-type and Muc2^-/- mice were stained in (A), with the polyclonal antibody PH497 that recognizes Muc2; in (B), with Alcian blue to identify goblet cells; and in (C), with a polyclonal antibody anti Itf. Bars for (A) and (C), 25 m; for (B), 100 m (duodenum), and 25 m in distal colon.

Medium version | Full size version

Supplemental Figure 3. Structure of the hybrid Muc2-Neo mRNA. The structure of the Muc2 mutant allele and of the corresponding Muc2-Neo hybrid mRNA is shown. The DNA sequence of the RT-PCR product generated using primers AV114/AV116 is also shown. Indicated are: the initiation of transcription, Muc2 intron 1 (small letters), the junction between the portion of Muc2 exon 2 of the mutant allele, in capital letters, and the polylinker sequence in the vector (smaller, capital letters). The new acceptor site for the hybrid mRNA in the mutant allele is also shown. The deduced aminoacid sequence of the putative Muc2-neo hybrid polypeptide is shown. Capital letters are the first 26 amino acid residues derived from the Muc2 sequence.

Medium version | Full size version

Supplemental Figure 4. Levels of expression of Muc5ac, Muc3, Muc13 and Itf mRNAs in the intestine of wild type and mutant mice. Total RNA was isolated from the indicated tissues from wild-type and Muc2 ^-/- mice (1, stomach; 2, duodenum; 3, jejunum; 4, ileum; 5, cecum; 6, proximal colon; 7, distal colon; 8, liver) and hybridized with the following probes: (A) Muc5ac, a probe which recognizes Muc5ac, gastric mucin. (B) Muc3, a RT-PCR generated fragment which recognizes the mouse Muc3 mRNA. (C) Muc13, a RT-PCR generated fragment which detects Muc13 mRNA (3.2 kb). (D) Itf, a probe which detects Itf, a gene expressed specifically in goblet cells. The heterogeneous hybridization observed in the blots is a specific feature of several mucin mRNAs which are preferential targets for degradation and shearing due to the large size of these mRNAs. The apparent increased expression of Muc3 and Muc13 in the blots with Muc2^-/- mRNA is due to unequal loading, as judged by GAPDH reprobing (data not shown).

Medium version | Full size version

Supplemental Figure 5. Immunohistochemical analysis of Muc5ac and Muc3 in the intestine of wild type and mutant mice. Sections from the indicated portions of the intestinal tract were stained for the expression, in (A), of Muc5ac using Ho3 antibody against gastric mucin; in (B), for Muc3 apoprotein using anti mouse Muc3 antibody. Bars, 25 m.

Medium version | Full size version

Supplemental Figure 6. Elevated c-Myc expression in tumors in Muc2^-/- mice. (A and B) immunohistochemical detection of -catenin and c-Myc in sections of an Apc1638 tumor illustrating -catenin nuclear localization (A) and c-Myc expression (B). C, hematoxylin & eosin stained section of an adenoma in the duodenum of Muc2^-/- mouse. (D through G) immunohistochemical analysis of -catenin (D) through (E) and c-Myc (F) and (G) in sections of the adenoma shown in (C), illustrating absence of nuclear staining of -catenin and strong nuclear staining for c-Myc. Bars for (A), (B), (E), and (G), 25 m; for (C), 250 m; and for (D) and (F), 100 m;

Medium version | Full size version

Supplemental Figure 7. Pattern of expression of -catenin and c-Myc in normal tissue of wild-type and Muc2^-/- mice. (A and B), sections from the normal mucosa of wild-type and Muc2^-/- mice were stained for the expression of -catenin (A) and c-Myc (B). (C) Western blot analysis of -catenin and c-Myc expression in cell extracts from tumor and normal tissue pairs of Muc2^-/- animals and tissue derived from wild-type mice, showing elevated expression of c-Myc and no changes in -catenin levels in tumors compared to normal tissues. The blots were reprobed with an anti -actin antibody to control for equal loading and transfer. Bars, 25 m.

Medium version | Full size version

References

1. R. Fodde et al., Proc. Natl. Acad. Sci. U.S.A. 91, 8969-8973 (1994).

2. A. Velcich, L. Augenlicht, J. Biol. Chem. 268, 13956-13961 (1993).

3. G. Hansson, D. Baeckstrom, I. Carlstedt, K. Klinga-Levan, Biochem. Biophys. Res. Comm. 198, 181-190 (1994).

4. L. Shekels et al., Biochem J. 311, 775-785 (1995).

5. L. L. Shekels et al., Biochem.J 330, 1301-1308 (1998).

6. S. Williams et al., J. Biol. Chem, 276, 18327-18336 (2001).

7. M. Tomita et al., Biochem J. 311, 293-297 (1995).

8. C. Foltz et al., Helicobacter 3, 69-78 (1998).

9. A. Burich et al., Am. J. Physiol. 281, G764-G778 (2001).

10. K. Yang et al., J. Exp. Zool. 277, 245-254 (1997).

11. N. Karlsson et al., Glycoconj. J. 13, 823-831 (1996).

12. M. Iwamoto, D. Ahnen, W. Franklin, T. Maltzman, Carcinogenesis 21, 1935-1940 (2000).

13. L. Augenlicht et al., Cancer Res. 59, 6005-6009 (1999).