Aims: In this study, we evaluated the impact of different real-time reverse-transcription PCR (RT-PCR) data normalization methods on the interpretation of stationary-phase and nutrient-starved Lactobacillus plantarum WCFS1 gene expression levels. Methods and Results: Lactobacillus plantarum WCFS1 culture characteristics and housekeeping gene transcripts were measured during stationary phase in standard growth medium and during extreme nutrient starvation. These conditions differentially affected L. plantarum viability and RNA/DNA ratios. Real-time RT-PCR gene expression data were normalized according to three different methods: (i) total RNA amounts added to the reactions; (ii) the comparative 2(-Delta Delta Ct) method using recA as a reference; and (iii) the geNorm approach based on the average expression values of several housekeeping genes. Each of these methods revealed differences in the abundance of housekeeping gene transcripts between L. plantarum in the exponential phase of growth and in stationary phase or undergoing nutrient starvation. Conclusions: Real-time RT-PCR data analysis with a normalization factor comprised of several of the most stably expressed housekeeping genes best accounted for the expected activity levels of the cells contained in the different cultures. Significance and Impact of the Study: The relative normalization of real-time RT-PCR data using multiple housekeeping reference genes should be useful for the quantification of bacterial gene expression levels in nonoptimal growth conditions in situ.