A POPULATION STUDY OF E. COLI DOUBLING TIMES: EVIDENCE OF AN ANOMALOUS BIMODAL DISTRIBUTION BELOW A THRESHOLD CELL CONCENTRATION

Authors: Irwin, Peter; Paoli, George; Chen, Chinyi; Nguyen, Ly Huong

Submitted to: BMC Microbiology
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 6/30/2010
Publication Date: 10/1/2010
Citation: Irwin, P.L., Paoli, G., Chen, C., Nguyen, L.T. 2010. A POPULATION STUDY OF E. COLI DOUBLING TIMES: EVIDENCE OF AN ANOMALOUS BIMODAL DISTRIBUTION BELOW A THRESHOLD CELL CONCENTRATION. BMC Microbiology. 10:207.

Interpretive Summary: The contamination of foods with pathogenic bacteria (e.g., Salmonella or E. coli O157:H7) may lead to substantial food poisoning epidemics. In order to sensitively detect food poisoning bacteria, certain isolation methods, such as the well-known immunomagnetic bead (IMB) technique for separating and concentrating these bacteria, have been developed. Unfortunately, the level of pathogenic bacteria in foods is very low relative to the concentration of benign bacteria (such as non-pathogenic E. coli). Thus, IMB methods for extracting pathogenic bacteria from foods also capture, and concentrate, some of the benign bacteria which interferes with the detection of pathogens. In this work we report that a well-known problematic benign E. coli isolate, which adheres to IMBs nearly as well as their intended target organism, has growth rates which are dependent upon its own initial cell concentration. This is an significant finding since it seems to imply that at least some food-borne (background) organisms chemically communicate amongst themselves in order to, presumably, coordinate growth under appropriate conditions. This information is useful to microbiologists in the detection of pathogenic bacteria from foods using IMB-based protocols.

Technical Abstract: Working with a native, food-borne, E. coli isolate, we found that doubling times (tau) derived from two fundamentally different microplate techniques (i.e., optical density as a function of growth time {OD[t]} or time to ½-maximal OD, tm, as a function of initial cell density {tm[Ci]}) were in excellent agreement with this same parameter acquired from total aerobic plate counting when using either liquid broth-based (Luria-Bertani, LB) or defined (minimal medium, MM) media at 37 deg C: tau = 17-18 min (LB) or 51-54 (MM); s = 2.31 (randomized complete block design); Ci > 1000 CFU/mL. Making use of truncated OD[t] data, fit to the sigmoidal Boltzmann equation, we were able to collected many observations of tau as a function of starting cell concentration (Ci). We noticed that tau appeared to be distributed in two distinct populations (e.g., bimodal) at low initial cell densities (Ci < 100-1000 CFU/mL; < 27-270 cells/well). When Ci < 100 CFU/mL (stationary phase cells diluted in LB), we found that about 48% of the observed tau values were normally distributed around a mean (mu_tau1) of 18 ± 1 min (± sigmal_tau1) and 52% with mu_tau2 = 20 ± 2 min (n = 479 observations). However, at higher starting cell densities (Ci > 100 CFU/mL), the tau values were clearly distributed unimodally (mu_tau = 18 ± 1 min; n =174). When dilutions were made from individual microplate wells (Ci ~ 1 cell/well; 21% negative growth) displaying large tau values (ca. 25-28 min; n = 2) and the growth experiments repeated, large deviations in tau were observed (ca. 15-29 min; Ci < 1000 CFU/mL; n = 148). This finding argues that the bimodal observation was physiologically-based. Comparable bimodal tau distribution results were also observed using E. coli cells diluted from mid-log phase cultures (Ci < 1000, 35% with mu_tau1 = 18 ± 1 min and 65% with mu_tau2 = 20 ± 2 min; n = 829). Population doubling time studies using MM (Ci < 300) as the diluent did not reverse the effect: 30% with mu_tau1 = 51 ± 2 min and 70% with mu_tau2 = 57 ± 8 min (n = 258). Similar results were obtained when using either an E. coli O157:H7 (Ci < 1000, 40% with mu_tau1 ~ 19 ± 1 min and 60% with mu_tau2 ~ 20 ± 2 min; 309 observations) or Citrobacter strain (Ci < 1000, 60% with mu_tau1 ~ 42 ± 4 min and 40% with mu_tau2 ~ 51 ± 6 min; 163 observations). When sterile-filtered LB media, which formerly contained relatively high concentrations of bacteria (> 1000/mL), was employed as a diluent, there was an evident shift of the two populations, albeit the bimodal effect was still apparent using either stationary or log phase cells.