, 2001, 2003). FlhD by itself, independent of FlhC, has also been reported to regulate cell division in E. coli (Prüß

& Matsumura, 1996). Cells in flhD mutant cultures were observed to continue dividing for several generations after cells in the flhD+ parental culture had stopped growing and entered the stationary DNA Damage inhibitor phase. This work is frequently cited as evidence that FlhD regulates cell division (Kaper & Sperandio, 2005; Umehara et al., 2007; Cui et al., 2008; Hatt & Rather, 2008; Isalan et al., 2008); however, our data indicate that this is not the case. We re-examined the effects of flhD mutations on entry to the stationary phase and found that the previously observed phenotype is not due to the flhD locus. Here, we show that the difference in the final cell number is due to the thyA mutation in the parental flhD+ strain, which had apparently reverted in the flhD− mutant strain used in the study. When the strains being compared have the same thyA allele (wild type or mutant), flhD mutations have no effect on growth. The E. coli K-12 strains and phage used in this study are listed in Table 1. λWM7 (Mao & Siegele, 1998) is a derivative of λRS45 (Simons et al., 1987) that carries an operon fusion between the mcb operon promoter (positions

−344 to +79) and the lac operon. Strains lysogenic for λWM7 were isolated by infecting YK410 and YK4131 with λWM7 and screening survivors on medium containing X-Gal Trichostatin A where lysogens form blue colonies. Monolysogens were identified by measuring β-galactosidase activity in several independent PI-1840 isolates of

each lysogen. Transductions with P1vir were performed as described by Miller (1972). Hfr mapping was performed as described (Singer et al., 1989) using the Hfr strains described in that paper as donors. To facilitate the exchange of flhD alleles, derivatives of YK410 (λPmcb-lacZ) and YK4131 (λPmcb-lacZ) were constructed that carry the linked uvrC279∷Tn10 mutation and retain their original flhD allele. These are strains DS507 and DS511, respectively, which were used as the donor strains in all subsequent strain constructions. Motility assays (described below) were used to determine whether transductants carried the wild type or the mutant flhD allele. Introduction of the uvrC279∷Tn10 mutation did not affect the expression of the Pmcb-lacZ fusion (Table 2 and data not shown). For β-galactosidase assays, cultures were grown in TB medium [1% Bacto tryptone, 0.5% NaCl (Arber et al., 1983)] supplemented with MgSO4 (10 mM), thymidine (10 μg mL−1), and thiamine (2 μg mL−1). For plates, 1.3% Bacto agar (Difco Laboratories) was included.