1. Volume 123, Issue 1, Pages 152-162 (July 2002)
Isolation and sequencing of a novel tropomyosin isoform preferentially associated with colon cancer 
Jenny L.–C. Lin, Xin Geng, Sangeeta Das Bhattacharya, Jae–Ran Yu, Rebecca S. Reiter, Bhagyalakshmi Sastri, Kenneth D. Glazier, Zafar K. Mirza, Kenneth K. Wang, Peter S. Amenta, Kiron M. Das, Jim J.–C. Lin 
Gastroenterology 
Volume 123, Issue 1, Pages (July 2002)
DOI: /gast
Copyright © 2002 American Gastroenterological Association Terms and Conditions

2. Fig. 1
Comparisons of amino acid sequence at the last coding exon and nucleotide sequence at the 3'-noncoding region of TC22 isoform with sequences found in other known nonmuscle tropomyosins. The last coding exon (amino acid residues from 222 to 247) of TC22 tropomyosin is compared with that of (A) known human fibroblast tropomyosins and (B) rat and chick brain tropomyosin, as well as a sequence from gene of Leishmania major. hTM5, hTM4, hTM5a, and hTM1, human fibroblast tropomyosin isoform 5,25 4,26 5a,4 and 1,27,28 respectively. In addition, human fibroblast tropomyosin isoform hTM5a, hTM5b, hTM2, hTM3, and hTMsmα have identical sequences in this region.4,29,30 rNM4, rat cochlea nonmuscle tropomyosin NM-422; rTMBr and cTMBr, rat23 and chick,24 respectively, brain tropomyosin isoforms; LM, amino acid residues from 38 to 63 of Leishmania major gene (accession number AAF73087). Dashes indicate identities, and nonidentities are indicated by the single-letter amino acid code relative to TC22. (C) Comparison of the 3'-noncoding sequence found in TC22 and rat cochlea nonmuscle tropomyosin NM-4 (rNM4).22 The middle row designates identities between 2 sequences. Dots in the top and bottom rows indicate gaps.
Gastroenterology  , DOI: ( /gast )
Copyright © 2002 American Gastroenterological Association Terms and Conditions

3. Fig. 2
Dot blot and Northern blot analyses. (A and B) The dot blot results with labeled REN29 probe and TC22-specific probe, respectively, to demonstrate the TC22 probe specificity. Plasmid containing full-length cDNA insert of hTM1 (lane 1), hTM2 (lane 2), hTM3 (lane 3), hTM4 (lane 4), hTM5 (lane 5), TC22 (lane 6), hTMsmα (lane 7), or hTM5a (lane 8) was 5-fold serially diluted and spotted onto the nitrocellulose membrane from 20 to 0.16 ng. (A) The hybridization with labeled REN29 probe, which recognized all known hTM isoforms,4 was used to show the cDNA loading in the blot. (B) However, the hybridization with labeled TC22-specific probe showed that this probe recognized only TC22 but not hTM1, hTM2, hTM3, hTM5, hTMsmα, and hTM5a. C shows the Northern blot results when TC22-specific probe was used as probe on the blot membrane containing total RNAs isolated from LS180 (lane 1), T84 (lane 2), DLD-1 (lane 3), and EJ (lane 4) cells. The exposure time for this autoradiogram was 2.5 days. (D) The Northern blot results when various tropomyosin isoform–specific probes and β-actin control probe were used as probe on the same or similar blot membrane as in C. The description and specificity of these tropomyosin isoform–specific probes were reported previously.4 Only the portion of the blot containing hybridized band with corresponding RNA size was shown here. The exposure times for these autoradiograms were 16 hours for hTM5-, hTM4-, and hTM1-specific probe; 7 hours for hTM2- and hTM3-specific probe; and 45 minutes for β-actin control probe. (E) The dot blot results with TC22-specific probe to examine the expression level in normal human tissues. The tissue position and amount of poly(A)+ RNA applied to each dot is listed below. A, colon, 177 ng; B, small intestine, 275 ng; C, stomach, 293 ng; D, brain, 257 ng.
Gastroenterology  , DOI: ( /gast )
Copyright © 2002 American Gastroenterological Association Terms and Conditions

4. Fig. 3
Northern blot analyses of TC22 tropomyosin isoform expression in normal and transformed human cell lines as well as in normal and tumor tissues. (A) Northern blot results when TC22-specific probe (top panel) or γ-actin control probe (bottom panel) was used as probe on the blot membrane containing total RNA samples isolated from normal (KD and WI-38) and transformed (HuT-11 and WI-38 VA13) human cell lines. A total of 8 μg of total RNAs was loaded in each lane. Lanes 1–4 were loaded as follows: lane 1, HuT-11; lane 2, KD; lane 3, WI-38 VA13; lane 4, WI-38. B shows Northern blot results when TC22-specific probe was used as probe on the blot membrane containing total RNAs from normal and tumor tissues (top panel). The bottom panel shows the total RNA loading as visualized by ethidium bromide staining. Lanes 1–6 were loaded as follows: lane 1, stomach tumor; lane 2, normal stomach; lane 3, colon tumor; lane 4, normal colon; lane 5, rectal tumor; lane 6, normal rectum.
Gastroenterology  , DOI: ( /gast )
Copyright © 2002 American Gastroenterological Association Terms and Conditions

5. Fig. 4
Western blot analyses of purified recombinant hTM isoforms, total cell extract from a colon cancer cell line T84, and different amounts of hTM5 and TC22. In A, approximately 1–2 μg of purified recombinant tropomyosin proteins, including hTM1 (lane 1), hTM2 (lane 2), hTM3 (lane 3), hTM4 (lane 4), hTM5 (lane 5), and TC22 (lane 6), were loaded in each lane and separated by 12.5% sodium dodecyl sulfate/polyacrylamide gel electrophoresis. The left panel shows the Coomassie blue–stained protein profile, whereas the right panel shows the Western blot results with TC22-4 antibody. The TC22-4 antibody recognized only TC22 tropomyosin isoform. In B, T84 total cell extract was separated by 12.5% sodium dodecyl sulfate/polyacrylamide gel electrophoresis and either stained with Coomassie blue dye (lane 1) or transferred and blotted with TC22-4 antibody (lane 2). Western blot result showed that a protein band with a molecular mass of 32 kilodaltons was specifically recognized by TC22-4 mAb. (C) Purified hTM5 and TC22 proteins were dotted on the membrane in 2 different amounts and then processed for Western blot analysis with LC1 antibody (lane a) or with TC22-4 mAb (lane b). 1 and 2, 1 μg and 0.05 μg, respectively, of hTM5; 3 and 4, 1 μg and 0.05 μg, respectively, of TC22.
Gastroenterology  , DOI: ( /gast )
Copyright © 2002 American Gastroenterological Association Terms and Conditions

6. Fig. 5
Western blot analyses of various mucosal protein extracts with (A) TC22-4 mAb and (B) LC1 antibody. Tropomyosin-enriched fractions were prepared from colonic and jejunal mucosa as described previously.9 Colonic mucosas were obtained from a colon cancer segment (lane 2) and its normal counterpart (lane 1) as well as a normal colon (spastic colon syndrome) (lane 4). Jejunal mucosa was obtained from a patient who underwent gastric bypass surgery for obesity (lane 3). A total of 10 μg of each tropomyosin-enriched fraction was loaded per lane and used in Western blot analysis. Recombinant TC22 and hTM5 proteins were used as controls (lanes 5 and 6, respectively). The reactivity against TC22-4 mAb is evident in the colon cancer tissue but not in the normal segment from a patient with colon cancer, normal colon tissue, and normal jejunum. However, LC1 reacted with each of these samples examined in parallel because hTM5 is present in all epithelial tissues. As expected, LC1 reacted with both TC22 (lane 5) and hTM5 (lane 6).
Gastroenterology  , DOI: ( /gast )
Copyright © 2002 American Gastroenterological Association Terms and Conditions

7. Fig. 6
Immunoperoxidase staining of a moderately differentiated adenocarcinoma with TC22-4 mAb. A shows H&E staining, and B and C show lower and higher magnification of the serial sections of the same tumor stained with TC22-4 mAb. Diffuse cytoplasmic staining is clearly evident in the tumor tissue. In addition, there are strong dot-like structures in the apical areas (C, arrow). Similar dot-like reactivity can also be seen in GFP-TC22–expressing cells (data not shown). The intervening stroma is completely nonreactive (B and C). (D) Normal colon tissue did not react with TC22-4 mAb. Serial sections of a hyperplastic polyp stained with (E) H&E and (F) TC22-4 mAb. TC22-4 did not react with the hyperplastic polyp.
Gastroenterology  , DOI: ( /gast )
Copyright © 2002 American Gastroenterological Association Terms and Conditions

8. Fig. 7
Immunoperoxidase staining with TC22-4 mAb of (A and B) a benign tubular adenoma, (C and D) another adenomatous polyp with adjacent normal colonic mucosa, and (E and F) a tubular adenoma with dysplasia. Serial sections stained with H&E are shown on the left for each polyp in A, C, and E. In the benign adenomatous polyps, the immunoreactivity is localized in the cytoplasm of the adenomatous glands in a more punctate manner (arrows) than the more diffuse pattern observed in adenocarcinoma (Figure 6B and C). In the polyp with adjacent normal tissue (large arrow in C), normal colonic glands did not react (asterisks in D). In the polyp with dysplasia, the reactivity is relatively stronger in the cytoplasm with intense intracellular punctate staining (F).
Gastroenterology  , DOI: ( /gast )
Copyright © 2002 American Gastroenterological Association Terms and Conditions