BW25113 hha mutant vs wild type biofilm cells 4hr LB and 4,15,24hr LB glu

E. coli K-12 BW25113 mutant strain hha expression in biofilm cells relative to E. coli wild-type strain expression in biofilm cells. Samples were cultured in LB with glasswool at 37C for 4 hours and in LB glu with glasswool at 37C for 4, 15 and 24 hours. Hha is a temperature- and osmolarity-dependent modulator of gene expression that is induced 30-fold in Escherichia coli biofilms. Here we show through whole-transcriptome analysis that Hha decreases biofilm formation in both LB and LB glu media by (i) repressing fliC encoding the main structural flagellar protein flagellin, (ii) by repressing fimA encoding the major structural subunit of type I fimbriae, (iii) by repressing ihfA encoding a subunit of the transcriptional regulator IHF that induces the transcription of type I fimbriae genes, (iv) by regulating tnaA encoding tryptophanase that inhibits biofilms, and (v) by repressing ybaJ that forms an operon with hha. Corroborating the microarray data, hha deletion increased motility 3.2 ± 0.1-fold, decreased extracellular indole concentrations 12 ± 2-fold, and decreased type I fimbriae (as measured by yeast agglutination). Biofilm tests using single and double mutants of fimA and ihfA and transcriptional studies of the fimA, ihfA and ybaJ-hha promoters confirmed that Hha represses biofilm by inhibiting type I fimbriae production and that it negatively regulates its own transcription and that of ybaJ. Nickel-enhanced DNA microarrays to determine in vivo Hha binding sites confirmed that Hha binds the ybaJ-hha promoter, that it binds fimZ, a positive regulator of fimA, and that it binds to the rare codon tRNAs argU, ileXY, and proL. Sequence analysis of fimZ, fimB, fimE, and the type I fimbriae gene cluster fimAICDFGH revealed a high bias for the rare codons of arginine, isoleucine, proline, leucine, and threonine, and overexpressing Hha leads to cell death. Therefore, it appears Hha decreases biofilms by decreasing type I fimbriae production as a result of inhibiting synthesis of tRNAs for rare codons. Keywords: effect of hha deletion in biofilm formation 4 hr LB and 4,15 and 24 hr LB glu Strains: E.coli K-12 BW25113 wild-type, mutant hha Medium:LB and LB glu Biofilm grown on glass wool Time: 4 hr for LB, 4,15 and 24 hr for LB glu Cell type: biofilm RNA isolation and DNA microarrays. Pre-cultures of E. coli BW25113 and BW25113 hha mutant were done for 15 h in LB and LB with kanamycin (50 μg/mL), respectively. Biofilm cells of each strain were prepared as described before (Zhang et al., 2007), and then the total RNA was isolated from biofilm cells as described in (Ren et al., 2004). The E. coli Genechip antisense genome array chips (P/N 900381, Affymetrix, Santa Clara, CA) c were used to analyze the complete E. coli transcriptome as described before (Zhang et al., 2007). All the procedures were done according to the Gene Expression Technical Manual (Affymetrix) and the DNA microarray data analysis was done using the GeneChip operating software (Affymetrix). The data quality was confirmed following the manufacturer's specifications (GeneChip Expression Analysis: Data Analysis Fundamentals; Affymetrix) and based on the expected signals of E. coli BW25113 and the hha mutant genotypes (e.g., signals of the deleted genes, araA and rhaA were low for both BW25113 and BW25113 hha mutant, while signal of hha were low for hha mutant). A gene was considered differentially expressed only when the P value was lower than 0.05 and the expression ratio was higher than: 2-fold for the 4 h LB and LB glu microarrays, higher than 4-fold for the LB 15 h microarray and higher than 2.46-fold for the 24 h LB glu microarray since the standard deviation for the expression ratio for all of the genes were: 1.4 for the 4 h LB, 1.38 for the 4 h LB glu, 3.29 for the 15 h LB glu and 1.63 for the 24 h LB glu microarray. The gene functions were obtained from the National Center for Biotechnology Information database (http://www.ncbi.nlm.nih.gov/) (Edgar et al., 2002), and the SRI International Institute for Genomic Research, University of California at San Diego, and UNAM database (http://ecocyc.org/) (Keseler et al., 2005).