Fluorescence for fluorophobesVirtual fluorescence using chromagens and histochemical

Fluorescence for fluorophobes--Virtual fluorescence using chromagens and histochemical stains Robert Dunstan, Luke Jandreski and the Comparative Pathology Laboratory

Lecture outline 1. Introduction 2. Using “non-fluorophores” for fluorescence • Histochemical stains • Use of red IHC chromagens for fluorescence 3. Use of registration/co-registration 4. Conclusions 2 November 13, 2003

Vision for virtual imaging and analysis “Increase signal, decrease noise with consistency” 3 November 13, 2003

Vision for virtual imaging and analysis Brightfield Pro --Morphologic assessment Con --Lower signal: noise --Non-linear expression of chromagens --Colocalization difficult 4 November 13, 2003 Fluorescence Pro --High signal to noise --Linear expression of fluorophores --Colocalization easy Con --Morphologic assessment difficult

Vision for virtual imaging and analysis Brightfield Pro --Morphologic assessment Con --Lower signal: noise --Non-linear expression of chromagens --Colocalization difficult 5 November 13, 2003 Fluorescence --Virtual microscopy --“Non-fluorophores” for fluorescence --Co-registration of images Pro --High signal to noise --Linear expression of chromagens --Colocalization Con --Morphologic assessment difficult

Using non-fluorophores for fluorescence-histochemical stains Example of histochemical stains that fluoresce • • • Eosin > hematoxylin Gentian Violet Neutral Red 6 November 13, 2003 • Thioflavin S and T and Congo Red • Toluidine Blue O

Using non-fluorophores for fluorescence-histochemical stains What determines which stains fluoresce? The number of conjugated bonds “Of the dyes with conjugated bond numbers (CBNs) of 29 or less, 90% showed fluorescence; 70% of the dyes whose CBNs exceeded 29 did not. . . “ Juarranz et al, Histochem ’ 86 7 November 13, 2003

Using non-fluorophores for fluorescence-histochemical stains Histochemical Stains Eosin fluoresces! 8 November 13, 2003

Using non-fluorophores for fluorescence-histochemical stains Histochemical Stains--Thioflavin S Brain (TG-2576 mouse with amyloid plaques) 9 November 13, 2003

Using non-fluorophores for fluorescence—Red IHC chromagens for fluorescence Not perfect but very good Good • Can assess morphology and fluorescence • Will colocalize: structures > cells > regions within cells • Higher signal: noise with fluorescence than bright field • Improved morphometry 10 November 13, 2003 Bad • Not as specific as standard antibody-bound fluorophores • Some red chromagens will smear

Using non-fluorophores for fluorescence—Red IHC chromagens for fluorescence Comparing fluorescence from a chromagen with fluorescence from a fluorophore Avidin-biotin complex with alkaline phosphatase Vector Fast Red Biotin 2 ndry Ab 1 o Ab Epitope 11 November 13, 2003

Using non-fluorophores for fluorescence—Red IHC chromagens for fluorescence Comparing fluorescence from a chromagen with fluorescence from a fluorophore CD-138 Red chromagen CD-138 Alexofluor 488 Human Lymph Node 12 November 13, 2003

Using non-fluorophores for fluorescence—Red IHC chromagens for fluorescence CD 31 (endothelial cell) Smooth muscle actin (smooth muscle) CD 31 (endothelial cell) Human to mouse xenograft 13 November 13, 2003

Using non-fluorophores for fluorescence—Red IHC chromagens for fluorescence Is anything gained? Fluorescent CD 31 (endothelial cell) Deconvolution of brightfield image Human to mouse xenograft 14 November 13, 2003

Using non-fluorophores for fluorescence—Red IHC chromagens for fluorescence CD 138 (plasma Cell) CD 138 (Plasma cell) VS 38 C (plasma Cell) Human lymph node 15 November 13, 2003

Using non-fluorophores for fluorescence—Red IHC chromagens for fluorescence Is anything gained? CD 138 (Plasma cell) Fluorescent Deconvolution of brightfield image Human lymph node 16 November 13, 2003

Using non-fluorophores for fluorescence—Red IHC chromagens for fluorescence B 220 (B cell) B 220 (B Cell) F 480 (macrophage) Mouse spleen 17 November 13, 2003

Using non-fluorophores for fluorescence—Red IHC chromagens for fluorescence Is anything gained? Fluorescent B 220 (B Cell) Deconvolution of brightfield image Mouse spleen 18 November 13, 2003

Using non-fluorophores for fluorescence—Red IHC chromagens for fluorescence B 220 (B Cell) F 480 (macrophage) Mouse liver 19 November 13, 2003

Using non-fluorophores for fluorescence—Red IHC chromagens for fluorescence Is anything gained? B 220 (B Cell) Fluorescent Deconvolution of brightfield image Mouse liver 20 November 13, 2003

Registration of the same image Image registration--the process of transforming 2 or more related images into one coordinate system Fluorescent Bright field Merged 21 November 13, 2003

Registration of the same image B 220 (B cell) B 220 (B Cell) F 480 (macrophage) Mouse spleen 22 November 13, 2003

Registration of the same image Cleaved caspase 3/B 220 23 November 13, 2003 CD 31/Smooth muscle actin B 220/F 480

Registration of virtual images Using AE 1 and AE 3 (pancytokeratin) as a mask on TMAs Ki-67 +’ve cell in tumor cluster Ki-67 +’ve cell Brightfield Fluorescent 24 November 13, 2003 Registration

Registration of different images Chromagen 1 3 um step sections Chromagen 2 Pseudo color Merged 25 November 13, 2003

Registration of different images CD 31 3 um apart CD 31 26 November 13, 2003

Registration of different images CD 31 pseudocolored 3 um apart CD 31 pseudocolored 27 November 13, 2003

Registration of different images CD 31 pseudocolored 28 November 13, 2003 + CD 31 pseudocolored Registered

Registration of different images CD 31 3 um apart Smooth muscle actin (SMA) 29 November 13, 2003

Registration of different images CD 31 pseudocolored 3 um apart SMA pseudocolored 30 November 13, 2003

Registration of different images CD 31 pseudocolored 31 November 13, 2003 + SMA pseudocolored Registered

Registration of virtual images Automatic section alignment Michael Grunkin, Visopharm 32 November 13, 2003 Final result

Registration of virtual images Michael Grunkin, Visopharm 33 November 13, 2003

Registration of different images Chromagen 1 3 um step sections Fluorophore Fluorescent) Bright field Merged 34 November 13, 2003

Registration of different images Imaging Mass Spectrometry Paul Kowalski, Bruker Daltonics 35 November 13, 2003

Conclusions • Virtual microscopy is just beginning to meet its potential as a tool to analyze morphologic changes • Advances in virtual fluorescence, fluorophores, image registration and evolving image analysis programs will make image analysis easier and more accurate than ever • Increase signal, decrease noise in a consistent manner 36 November 13, 2003