Limits...
Detection of colorectal dysplasia using fluorescently labelled lectins.

Kuo JC, Ibrahim AE, Dawson S, Parashar D, Howat WJ, Guttula K, Miller R, Fearnhead NS, Winton DJ, Neves AA, Brindle KM - Sci Rep (2016)

Bottom Line: Colorectal cancer screening using conventional colonoscopy lacks molecular information and can miss dysplastic lesions.Wheat germ agglutinin (WGA) showed significantly decreased binding to adenomas in the mouse tissue and in sections of human colon from 47 patients.Helix pomatia agglutinin (HGA) distinguished epithelial regions containing NE from regions containing HP, LGD, HGD or C, with 89% sensitivity, 87% specificity and 97% PPV.

View Article: PubMed Central - PubMed

Affiliation: Cancer Research UK Cambridge Institute, University of Cambridge, Li Ka Shing Centre, Cambridge, UK.

ABSTRACT
Colorectal cancer screening using conventional colonoscopy lacks molecular information and can miss dysplastic lesions. We tested here the ability of fluorescently labelled lectins to distinguish dysplasia from normal tissue when sprayed on to the luminal surface epithelium of freshly resected colon tissue from the Apc(min) mouse and when applied to fixed human colorectal tissue sections. Wheat germ agglutinin (WGA) showed significantly decreased binding to adenomas in the mouse tissue and in sections of human colon from 47 patients. Changes in WGA binding to the human surface epithelium allowed regions containing normal epithelium (NE) or hyperplastic polyps (HP) to be distinguished from regions containing low-grade dysplasia (LGD), high-grade dysplasia (HGD) or carcinoma (C), with 81% sensitivity, 87% specificity and 93% positive predictive value (PPV). Helix pomatia agglutinin (HGA) distinguished epithelial regions containing NE from regions containing HP, LGD, HGD or C, with 89% sensitivity, 87% specificity and 97% PPV. The decreased binding of WGA and HPA to the luminal surface epithelium in human dysplasia suggests that these lectins may enable more sensitive detection of disease in the clinic using fluorescence colonoscopy.

No MeSH data available.


Related in: MedlinePlus

Quantitative analysis of acidic mucins on the luminal surface epithelium of the different colorectal pathology classes and correlation with lectin binding.(a) Tissue sections were stained with Alcian blue (AB)–periodic acid Schiff (PAS) combination stain and the resulting colour analysed for the presence of acidic mucins, by using a trained algorithm selective for the “blue” colour produced by AB staining (Fig. 3a, column 2). AB signals were averaged to generate scores (y-axis) for the different pathology classes (x-axis) in each sample in unmatched patient analyses. The P value represents the Jonckheere-Terpstra test for trend. Abbreviations: N, normal; HP, hyperplasia; LGD, low-grade dysplasia; HGD, high-grade dysplasia; C, carcinoma. Linear regression analysis of acidic mucin staining, as a function of WGA (b) and HPA (c) fluorescence, for unmatched patient analyses. WGA, wheat germ agglutinin. HPA, Helix pomatia agglutinin. Dashed lines (b,c) represent the 95% confidence interval hyperbolas for the linear best fits (solid lines).
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4829854&req=5

f4: Quantitative analysis of acidic mucins on the luminal surface epithelium of the different colorectal pathology classes and correlation with lectin binding.(a) Tissue sections were stained with Alcian blue (AB)–periodic acid Schiff (PAS) combination stain and the resulting colour analysed for the presence of acidic mucins, by using a trained algorithm selective for the “blue” colour produced by AB staining (Fig. 3a, column 2). AB signals were averaged to generate scores (y-axis) for the different pathology classes (x-axis) in each sample in unmatched patient analyses. The P value represents the Jonckheere-Terpstra test for trend. Abbreviations: N, normal; HP, hyperplasia; LGD, low-grade dysplasia; HGD, high-grade dysplasia; C, carcinoma. Linear regression analysis of acidic mucin staining, as a function of WGA (b) and HPA (c) fluorescence, for unmatched patient analyses. WGA, wheat germ agglutinin. HPA, Helix pomatia agglutinin. Dashed lines (b,c) represent the 95% confidence interval hyperbolas for the linear best fits (solid lines).

Mentions: Alcian blue (AB)–periodic-acid Schiff (PAS) combination stain (AB-PAS) was used to determine whether WGA and HPA were binding to acidic mucins (sialomucins and sulfomucins) stained blue by AB, or neutral mucins, stained deep-red/magenta by PAS (Fig. 3). The AB staining pattern closely resembled that observed for WGA, but not HPA (Figs 3 and 4). Regions of hyperplasia showed the strongest staining of luminal surface epithelium, with decreased staining in the progression from normal tissue to dysplasia and carcinoma (Fig. 4a) (P < 0.001). There was a strong correlation between WGA binding and AB staining (R = 0.79, P < 0.0001, Pearson product moment correlation; Fig. 4b), consistent with the specificity of WGA for acidic glycans (sialic acids)33. Significant correlations between WGA binding and AB staining were observed for normal epithelium, LGD and HGD but not for HP and carcinoma (Supplementary Table 2a). AB staining, similarly to WGA binding, could distinguish between non-neoplastic epithelium (normal or hyperplasia) and neoplasia with high sensitivity and specificity (Supplementary Table 3). As with WGA binding, normal epithelium showed a wide range of AB staining (Fig. 4a), which again might be explained by loss of mucus during tissue processing. There was no correlation between HPA binding and AB staining, consistent with HPA’s lack of specificity for acidic glycans34 (Fig. 4c and Supplementary Table 2). PAS staining was weak across all the classes and showed no correlation with disease progression (P = 0.525, Supplementary Fig. 3a) or with WGA or HPA binding (Supplementary Fig. 4b,c, Supplementary Table 2b). Deposits of PAS positive material were observed within the lumen of HGD and C (row 4, column 2 in Fig. 3a, Supplementary Fig. 4) and appeared to correlate with glandular regions that showed very strong WGA binding (row 4, column3 in Fig. 3a). These regions occurred mostly deep beneath the luminal surface epithelium and therefore were excluded from analysis. The luminal surface epithelium of HGD and C were mostly devoid AB-PAS staining (column 2, row 3 and 4 in Figs 3a and 4a, Supplementary Fig. 3a).


Detection of colorectal dysplasia using fluorescently labelled lectins.

Kuo JC, Ibrahim AE, Dawson S, Parashar D, Howat WJ, Guttula K, Miller R, Fearnhead NS, Winton DJ, Neves AA, Brindle KM - Sci Rep (2016)

Quantitative analysis of acidic mucins on the luminal surface epithelium of the different colorectal pathology classes and correlation with lectin binding.(a) Tissue sections were stained with Alcian blue (AB)–periodic acid Schiff (PAS) combination stain and the resulting colour analysed for the presence of acidic mucins, by using a trained algorithm selective for the “blue” colour produced by AB staining (Fig. 3a, column 2). AB signals were averaged to generate scores (y-axis) for the different pathology classes (x-axis) in each sample in unmatched patient analyses. The P value represents the Jonckheere-Terpstra test for trend. Abbreviations: N, normal; HP, hyperplasia; LGD, low-grade dysplasia; HGD, high-grade dysplasia; C, carcinoma. Linear regression analysis of acidic mucin staining, as a function of WGA (b) and HPA (c) fluorescence, for unmatched patient analyses. WGA, wheat germ agglutinin. HPA, Helix pomatia agglutinin. Dashed lines (b,c) represent the 95% confidence interval hyperbolas for the linear best fits (solid lines).
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4829854&req=5

f4: Quantitative analysis of acidic mucins on the luminal surface epithelium of the different colorectal pathology classes and correlation with lectin binding.(a) Tissue sections were stained with Alcian blue (AB)–periodic acid Schiff (PAS) combination stain and the resulting colour analysed for the presence of acidic mucins, by using a trained algorithm selective for the “blue” colour produced by AB staining (Fig. 3a, column 2). AB signals were averaged to generate scores (y-axis) for the different pathology classes (x-axis) in each sample in unmatched patient analyses. The P value represents the Jonckheere-Terpstra test for trend. Abbreviations: N, normal; HP, hyperplasia; LGD, low-grade dysplasia; HGD, high-grade dysplasia; C, carcinoma. Linear regression analysis of acidic mucin staining, as a function of WGA (b) and HPA (c) fluorescence, for unmatched patient analyses. WGA, wheat germ agglutinin. HPA, Helix pomatia agglutinin. Dashed lines (b,c) represent the 95% confidence interval hyperbolas for the linear best fits (solid lines).
Mentions: Alcian blue (AB)–periodic-acid Schiff (PAS) combination stain (AB-PAS) was used to determine whether WGA and HPA were binding to acidic mucins (sialomucins and sulfomucins) stained blue by AB, or neutral mucins, stained deep-red/magenta by PAS (Fig. 3). The AB staining pattern closely resembled that observed for WGA, but not HPA (Figs 3 and 4). Regions of hyperplasia showed the strongest staining of luminal surface epithelium, with decreased staining in the progression from normal tissue to dysplasia and carcinoma (Fig. 4a) (P < 0.001). There was a strong correlation between WGA binding and AB staining (R = 0.79, P < 0.0001, Pearson product moment correlation; Fig. 4b), consistent with the specificity of WGA for acidic glycans (sialic acids)33. Significant correlations between WGA binding and AB staining were observed for normal epithelium, LGD and HGD but not for HP and carcinoma (Supplementary Table 2a). AB staining, similarly to WGA binding, could distinguish between non-neoplastic epithelium (normal or hyperplasia) and neoplasia with high sensitivity and specificity (Supplementary Table 3). As with WGA binding, normal epithelium showed a wide range of AB staining (Fig. 4a), which again might be explained by loss of mucus during tissue processing. There was no correlation between HPA binding and AB staining, consistent with HPA’s lack of specificity for acidic glycans34 (Fig. 4c and Supplementary Table 2). PAS staining was weak across all the classes and showed no correlation with disease progression (P = 0.525, Supplementary Fig. 3a) or with WGA or HPA binding (Supplementary Fig. 4b,c, Supplementary Table 2b). Deposits of PAS positive material were observed within the lumen of HGD and C (row 4, column 2 in Fig. 3a, Supplementary Fig. 4) and appeared to correlate with glandular regions that showed very strong WGA binding (row 4, column3 in Fig. 3a). These regions occurred mostly deep beneath the luminal surface epithelium and therefore were excluded from analysis. The luminal surface epithelium of HGD and C were mostly devoid AB-PAS staining (column 2, row 3 and 4 in Figs 3a and 4a, Supplementary Fig. 3a).

Bottom Line: Colorectal cancer screening using conventional colonoscopy lacks molecular information and can miss dysplastic lesions.Wheat germ agglutinin (WGA) showed significantly decreased binding to adenomas in the mouse tissue and in sections of human colon from 47 patients.Helix pomatia agglutinin (HGA) distinguished epithelial regions containing NE from regions containing HP, LGD, HGD or C, with 89% sensitivity, 87% specificity and 97% PPV.

View Article: PubMed Central - PubMed

Affiliation: Cancer Research UK Cambridge Institute, University of Cambridge, Li Ka Shing Centre, Cambridge, UK.

ABSTRACT
Colorectal cancer screening using conventional colonoscopy lacks molecular information and can miss dysplastic lesions. We tested here the ability of fluorescently labelled lectins to distinguish dysplasia from normal tissue when sprayed on to the luminal surface epithelium of freshly resected colon tissue from the Apc(min) mouse and when applied to fixed human colorectal tissue sections. Wheat germ agglutinin (WGA) showed significantly decreased binding to adenomas in the mouse tissue and in sections of human colon from 47 patients. Changes in WGA binding to the human surface epithelium allowed regions containing normal epithelium (NE) or hyperplastic polyps (HP) to be distinguished from regions containing low-grade dysplasia (LGD), high-grade dysplasia (HGD) or carcinoma (C), with 81% sensitivity, 87% specificity and 93% positive predictive value (PPV). Helix pomatia agglutinin (HGA) distinguished epithelial regions containing NE from regions containing HP, LGD, HGD or C, with 89% sensitivity, 87% specificity and 97% PPV. The decreased binding of WGA and HPA to the luminal surface epithelium in human dysplasia suggests that these lectins may enable more sensitive detection of disease in the clinic using fluorescence colonoscopy.

No MeSH data available.


Related in: MedlinePlus