Global expression profiling of RNA isolated from laser microdissected cells allows one to profile a specific set of cells allowing for enhanced sensitivity and for cell- or site-specific patterns of expression to emerge. In Chapter 19, we detail our optimized methods of tissue preparation, laser microdissection (LMD), and RNA isolation of cells at the site of Golovinomyces orontii infection of mature Arabidopsis leaves. Here, we describe (1) amplification of the RNA to obtain sufficient starting material for microarray analysis, (2) microarray hybridization and associated quality control assessments. As tissue preparation, LMD, and/or RNA amplification could impact mRNA quality, distribution, and/or microarray processing and output, it is important to include quality control assessments at every step of the protocol to ensure that the final data is a reproducible and accurate readout of the biological source material. The collection of parallel samples to evaluate these components of the experimental protocol allows one to determine their impact on mRNA quality and distribution (described in Chapter 19) and on microarray output (discussed here). In addition, one likely wants to compare similarly processed whole leaf samples to LMD-isolated samples in order to identify genes and processes specifically impacted or more highly impacted at the infection site compared with the whole leaf. Using the procedures described herein to profile cells specifically at the site of powdery mildew infection of Arabidopsis (Chandran et al., Proc Natl Acad Sci U S A 107(1):460–465, 2010), we determined that our site-specific global expression data was a highly reproducible, sensitive, and accurate readout of the infection site. Furthermore, this site-specific analysis allowed us to identify novel processes (e.g., endore�duplication), regulators (e.g., MYB3R4), and process components associated with the sustained growth and reproduction of the powdery mildew G. orontii on Arabidopsis thaliana at 5 days postinfection that were hidden in whole leaf analyses (Chandran et al., Proc Natl Acad Sci U S A 107(1):460–465, 2010).