REGULAR ARTICLE Comparative proteomic analysis using samples obtained with laser microdissection and saturation dye labelling Kate E. Wilson 1 , Rita Marouga 2 , John E. Prime 3 , D. Paul Pashby 3 , Paul R. Orange 3 , Steven Crosier 1 , Alexander B. Keith 1 , Richard Lathe 4 , John Mullins 4 , Peter Estibeiro 5 , Helene Bergling 2 , Edward Hawkins 3 and Christopher M. Morris 1,6 1 MRC Building, Newcastle General Hospital, Newcastle upon Tyne, UK 2 GE Healthcare (formerly Amersham Biosciences), Uppsala, Sweden 3 GE Healthcare (formerly Amersham Biosciences), Chalfont St. Giles, Bucks, UK 4 Centre for Genome Research, University of Edinburgh, Edinburgh, UK 5 ExpressOn BioSystems Ltd., Roslin BioCentre, Roslin, Edinburgh, UK 6 Institute of Human Genetics, International Centre for Life, Newcastle upon Tyne, UK Comparative proteomic methods are rapidly being applied to many different biological systems including complex tissues. One pitfall of these methods is that in some cases, such as oncology and neuroscience, tissue complexity requires isolation of specific cell types and sample is limited. Laser microdissection (LMD) is commonly used for obtaining such samples for proteomic stud- ies. We have combined LMD with sensitive thiol-reactive saturation dye labelling of protein samples and 2-D DIGE to identify protein changes in a test system, the isolated CA1 pyramidal neurone layer of a transgenic (Tg) rat carrying a human amyloid precursor protein transgene. Saturation dye labelling proved to be extremely sensitive with a spot map of over 5,000 proteins being readily produced from 5 mg total protein, with over 100 proteins being significantly altered at p , 0.0005. Of the proteins identified, all showed coherent changes associated with transgene expression. It was, however, difficult to identify significantly different proteins using PMF and MALDI-TOF on gels containing less than 500 mg total protein. The use of saturation dye labelling of limiting samples will therefore require the use of highly sensitive MS techniques to identify the significantly altered proteins isolated using methods such as LMD. Received: September 9, 2004 Revised: December 22, 2004 Accepted: January 17, 2005 Keywords: Alzheimer’s disease / 2-DE / Laser microdissection / Transgenics Proteomics 2005, 5, 3851–3858 3851 1 Introduction 2-DE is one of the most efficient methods for separating thousands of proteins from a single sample and as such is a frequently used technique in proteomics. Following IEF and SDS-PAGE, gels are traditionally stained to reveal the spot maps which are then compared between gels to identify changes in abundance between test and control samples, for example. However, gel-to-gel variation makes this system poorly reproducible, and to obtain the most accurate results several replicate experiments have to be run. More recently, 2-D DIGE has been developed, whereby protein samples are labelled with CyDye  fluors prior to separation [1, 2]. The CyDye DIGE fluors, Cy  2, Cy3 and Cy5, are all size and charge matched and therefore the same protein labelled with a particular CyDye will migrate to the same point on a gel regardless of which fluor is bound. These dyes are all spec- trally resolvable, with different excitation and emission Correspondence: K.E. Wilson, MRC Building, Newcastle General Hospital, Westgate Road, Newcastle upon Tyne, NE4 6BE, UK E-mail: k.e.wilson@ncl.ac.uk Fax: 144-191-444-4402 Abbreviations: AD, Alzheimer’s disease; APP , amyloid precursor protein; HSP , heat shock protein; LMD, laser microdissection; Prx, peroxiredoxin; Tg, transgenic  2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.proteomics-journal.de DOI 10.1002/pmic.200401255