Review Generation and use of a tailored gene array to investigate vascular biology Amanda L. Evans 1,2 , Andrew S. Sharkey 1,2, *, Samir A. Saidi 1,3, *, Cristin G. Print 1,2, *, Roberto D. Catalano 1,2 , Stephen K. Smith 1,3 & D. Stephen Charnock-Jones 1,3 1 Reproductive Molecular Research Group, 2 Department of Pathology, University of Cambridge, Tennis Court Rd, Cambridge, UK; 3 Department of Obstetrics and Gynaecology, The Rosie Hospital, Robinson Way, Cambridge, UK Received 2 April 2003; accepted in revised form 4 June 2003 Key words: angiogenesis, array, HUVEC, normalisation, transcript Abstract Vasculogenesis, angiogenesis and vascular remodelling are complex processes where the fate of several cell types is determined by different signalling networks. Many of these networks ultimately function by changing the abundance of RNA transcripts within the cells which constitute blood vessel walls. Researchers can now map these transcript abundance changes using gene array technology. In this review, we describe the design, production and use of a gene array specifically tailored to investigate vascular biology. We describe the advantages of tailored gene arrays, and give detailed protocols based on our experience to allow the reader to use such gene arrays to generate meaningful data. We list the issues to consider when choosing and verifying the genes and splice variants included in an array, and describe our use of Arabidopsis sp. RNA spikes for quality control. We present data that illustrates the absolute necessity for both technical and biological replicates to be incorporated in the design of gene array experiments using primary cells such as HUVECS. Finally, we describe methods for the normalisation and interpretation of the data that gene arrays produce. The approach to gene array technology described here is easily within reach of the budget and expertise of most academic research groups. Abbreviations: QC – quality control; HUVEC – human umbilical vein endothelial cells; PCR – polymerase chain reaction Introduction The recent development of DNA array technology has substantially altered the conduct of angiogenesis re- search. This technology allows the abundance of a large number of transcripts within complex RNA populations to be determined simultaneously. For the first time this offers researchers the prospect of understanding the subtle interactions between multiple genes that may underlie complex cellular behaviours. For example, vasculogenesis, angiogenesis and vascular remodelling are complex processes involving interaction between several cell types within and outside the vessel wall [1–3]. We are using gene arrays to determine the transcript abundance regulation that underlies these processes. Gene arrays come in two main types: large generic arrays and small tailored arrays. Large generic gene arrays, are available commercially or by collaboration with dedicated gene array laboratories. One well estab- lished large generic technology is the Affymetrix Gene- chip system (Affymetrix Inc. Santa Clara, California, USA, http://www.affymetrix.com) [4] in which two- dimensional arrays of synthetic oligonucleotides are synthesised using a combination of photolithography and solid phase DNA synthesis [5, 6]. Affymetrix gene chips allow thousands of transcripts to be analysed in a single hybridisation, and are ideal for experiments in which researchers wish to search for unexpectedly regulated genes. While we, and others have shown that this approach can be extremely informative for angio- genesis research [7], it is inappropriate for many experimental designs. Commercial generic arrays are expensive and academically produced generic arrays of limited availability. Therefore, generic gene arrays may be impractical when a large number of experimental replicates are required. In addition, the genes included in most generic arrays are inflexible, and frequently omit genes and splice variants of interest. An alternative approach is for individual research laboratories to construct relatively small tailored gene arrays focussed on their specific research interests. These have the advantage of low cost, which allows researchers to *Authors contributed equally to the work described in this publication. Correspondence to: Dr Stephen Charnock-Jones, Reproductive Mo- lecular Research Group, Department of Obstetrics and Gynaecology, The Rosie Hospital, Cambridge, CB2 2SW, UK. Tel: +44-1223- 336875; Fax:+44-1223-215327; E-mail: dscj1@cam.ac.uk Angiogenesis 6: 93–104, 2003. 93 Ó 2003 Kluwer Academic Publishers. Printed in the Netherlands.