| The DNA microarray technology was applied to detect differential transcription profiles in Escherichia coli. In the first stage of this effort, over one hundred metabolic and regulatory genes were cloned on a plasmid and arrayed in the UCLA Microarray Core Facility. In the second stage, 96% of all E. coli open reading frames (ORFs), 4129 genes, were PCR-amplified and arrayed in our laboratory. When RNA was purified from E. coli strains, labeled with fluorescent dyes, and hybridized to the microarrays, the fluorescence level of each spot represents the transcript level of the gene. To obtain reliable data, several repeats of experiments as well as statistical validation were required. Therefore, a rigorous statistical method to analyze microarray data was developed. This method consists of quality filtering to remove outlying spots, a rank-invariant method to construct the normalization curves, and a Markov Chain Monte Carlo computation method for assessing gene expression levels with confidence intervals. The method considers many sources of error and takes them into account during the analysis. The expression profile of E. coli grown with different carbon source, acetate, glycerol, and glucose, was monitored. In addition, the response of metabolic genes to the protein overproduction was also studied. In particular, the comparison of E. coli grown in glucose and in acetate confirmed many known features of its metabolism such as the induction of glyoxylate pathway, TCA cycle, and gluconeogenic genes and repression of glycolytic and glucose phosphotransferase genes in acetate. It also provided many previously unknown features, including induction of malic enzymes, ppsA, and glycolate pathway in acetate. The carbon flux delivered from the malic enzyme and phosphoenolpyruvate synthase (PpsA) pathway in acetate was further confirmed by deletion mutations. In general, the gene expression profile dramatically resembles the preferred metabolic pathways, and may serve as a predictor for gene function and metabolic flux distribution. |
