In order to unravel the genetic architecture underlying plant response to drought, we adopted an integrated approach, combining transcript profiling and quantitative trait loci (QTL) mapping. In fact, improving plant tolerance to water stress is an important, but, at the same time, a difficult task, since plant tolerance is the result of many complex mechanisms acting at different levels of plant organization, and its genetic basis is largely unknown. The phenotypic data, concerning yield components and flowering time, of a population of 142 maize Recombinant Inbred Lines (RILs), grown under well watered conditions or under water stress, were submitted to linkage analysis to detect drought-tolerance QTLs. Thirty genomic regions containing 50 significant QTLs distributed on nine chromosomes were identified. At the same time, a customized targeted oligoarray was used to monitor the expression levels of 1,000 genes, representative of the immature maize kernel transcriptome. Using this DNA array we compared transcripts from 10 days after pollination kernels of two susceptible and two drought tolerant genotypes (extracted from our RILs) grown under control and water stress field conditions. Two hundred and fifty-two genes were significantly affected by stress in at least one genotype. From a set of these, 49 new molecular markers were developed. By mapping most of them and by in silico mapping other regulated sequences, 88 differentially expressed genes were localized onto our linkage map, which, added to the existing 186 markers, brought their total number on the map to 274. Twenty-two of the 88 differentially expressed genes mapped in the same chromosomal segments harbouring QTLs for tolerance, thus representing candidate genes for further functional studies.