Fig 2: Western blot analysis using an anti-EpCAM mAb. Cell lysates (10 µg) of CHO-K1, CHO/EpCAM, SAS, and HSC-2 were electrophoresed and transferred onto PVDF membranes. The membranes were incubated with 5 µg/ml of EpMab-16 and 1 µg/ml of anti-ß-actin, followed by peroxidase-conjugated anti-mouse immunoglobulins. The red arrow indicates EpCAM. The blue arrows indicate non-specific bands, which were detected in both CHO-K1 and CHO/EpCAM cells. EpCAM, epithelial cell adhesion molecule; mAb, monoclonal antibody; PVDF, polyvinylidene difluoride.

Fig 3: ?H2AX positivity correlates with extent of chronic inflammation. (A) Tissues were classified according to extent of chronic inflammatory cell infiltration as positive (n=6) or negative (n=3) and subjected to multiplex immunohistochemistry staining with antibodies against EpCAM/pan-Cytokeratin (CK), MCM2 and ?H2AX. Immunofluorescence images of MCM2 or ?H2AX with EpCAM/pan-Cytokeratin (CK) are shown. Scoring map depicts epithelial cells positive for ?H2AX (magenta), MCM2 (green), ?H2AX /MCM2 double-positive (DP) (red) and ?H2AX/MCM2 double-negative (grey). (B) Plot shows percent MCM2 positivity in the chronic inflammation positive and negative groups. (C) Plot shows percent ?H2AX positivity in the chronic inflammation positive and negative groups. Line within the vertical scatter plot shows the mean value. Error bar shows ±SD. Statistical significance was analysed by the Mann-Whitney non-parametric test. * p<0.05, n.s-not significant.

Fig 4: Modest increase in ?H2AX positivity during GC progression. (A) TMA slide (ST1001) was stained with antibodies against EpCAM/pan-Cytokeratin and ?H2AX. Seven cases of normal, eight cases of NAG and six cases of IM were analysed. (B) TMA slide (IC00011B) was stained with antibodies against EpCAM/pan-Cytokeratin (CK) and ?H2AX. Twenty-six cases of IM and 14 cases of IGCs were analysed. The ?H2AX staining shown with pseudo-brown colour has been overlaid on the digital H&E image. Areas marked by the yellow rectangle within the fluorescent images have been zoomed by two-fold under the images marked as ‘inset’. (C) Plot shows percent ?H2AX positivity across the three stages, normal, NAG and IM. (D) Plot shows percent ?H2AX positivity across IM and IGCs. Line within the vertical scatter plot shows the mean value. Error bar shows ±SD. Statistical significance was analysed by the Mann-Whitney non-parametric test. * p<0.05, **p<0.01. GC, gastric cancer; IGC, intestinal-type gastric cancer; IM, intestinal metaplasia;NAG, non-atrophic gastritis; TMA, tissue microarrays.

Fig 5: MCM2 positivity increases during GC progression. (A) TMA slide (ST1001) was stained with antibodies against EpCAM/pan-Cytokeratin (CK) and MCM2. Six cases each of normal gastric tissues, NAG and IM were analysed. (B) TMA slide (IC00011B) was stained with antibodies against EpCAM/pan-Cytokeratin (CK) and MCM2. Twenty-six cases of IM and 14 cases of IGCs were analysed. The panel labelled as digital H&E is a simulated bright field image obtained by using Hoechst 33342 and tissue autofluorescence as masks. The MCM2 staining pseudo-coloured as brown, has been overlaid on the the digital H&E image. Areas marked by the yellow rectangle within the fluorescent images have been zoomed by two-fold under the images marked as ‘inset’. (C) Plot shows percent MCM2 positivity across the three stages, normal, NAG and IM. (D) Plot shows percent MCM2 positivity across IM and adenocarcinoma. Line within the vertical scatter plot shows the mean value. Error bar shows ±SD. Statistical significance was analysed by the Mann-Whitney non-parametric test. **p<0.01. GC, gastric cancer; IM, intestinal metaplasia; NAG, non-atrophic gastritis; TMA, tissue microarrays.