LB, ∆fur, and Stationary Phase

Lit Review
Published

October 3, 2025

Dr. Yep noticed how ∆fur stalled earlier than the other strains. It’s not necessarily a problem, but I’d like to understand the cause. The paper should also explain this, so we’ll need the rationale anyway.

∆fur entered stationary phase early in Lysogeny Broth (LB). This didn’t happen with any of the other strains or media. Why? I’ll attempt to explain this in a few ways.

ROS Damage

Since this phenomenon was absent in all other media, I think the early stalling most likely reflects a specific interaction between the high intracellular iron characteristic of ∆fur, and some inherent property of LB.

There appears to be commercial variance in background ROS in LB. LB (Sigma) tends to have more H2O2 than LB (Difco), which is generally gentler (I’m fairly confident we use Sigma). Together with constitutively elevated intracellular iron (Fenton reaction), ∆fur may also be exposed to higher levels of peroxide, causing ROS damage. Components such as riboflavin, reducing sugars, and transition metal ions—present in LB—can generate ROS depending on storage conditions, especially with light exposure (Ezraty et al. 2014).

Note that ∆fur is easily pushed into oxidative stress as well. Deletion of fur indirectly represses sodB, which encodes iron superoxide dismutase. This enzyme acts as an antioxidant, preventing ROS damage. But in ∆fur, this antioxidant is repressed. Cells must then rely on the product of sodA (manganese superoxide dismutase), and, unfortunately, if Mn is lacking in the environment, there isn’t much the cell can do to prevent ROS damage (Chareyre and Mandin 2018).

Side note: although it wasn’t exactly stationary phase, there was some growth inhibition with ∆fur in swarming medium as well. Both LB and swarming medium contain yeast extract, which may include components necessary for ROS generation. Of course, this perceived symmetry may be coincidental, and some other intrinsic characteristic attributable to the mutation may be at play; I’m only speculating.

pH Threshold

The effect seen in ∆fur is most likely due to various reasons in combination, including pH damage. One other thing I suspect is the pH threshold of E. coli since LB isn’t really buffered, so left alone, growth can take the media up to a pH of 9 without glucose (swarming media has glucose, which may be why CFT doesn’t top out in swarming media.) Excretion of excess ammonia is to blame for the alkalinization (Sezonov, Joseleau-Petit, and D’Ari 2007).

Carbon Exhaustion

This one is pretty simple. Since LB is pretty limited terms of its nutrient content, lack of carbon sources in combination with the preceding reasons may account for the early stalling observed for ∆fur in LB (Sezonov, Joseleau-Petit, and D’Ari 2007).

References

Chareyre, Sylvia, and Pierre Mandin. 2018. “Bacterial Iron Homeostasis Regulation by sRNAs.” Microbiology Spectrum 6 (2): 10.1128/microbiolspec.rwr-0010-2017. https://doi.org/10.1128/microbiolspec.rwr-0010-2017.
Ezraty, Benjamin, Camille Henry, Marion Hérisse, Erick Denamur, and Frédéric Barras. 2014. “Commercial Lysogeny Broth Culture Media and Oxidative Stress: A Cautious Tale.” Free Radical Biology and Medicine 74 (September): 245–51. https://doi.org/10.1016/j.freeradbiomed.2014.07.010.
Sezonov, Guennadi, Danièle Joseleau-Petit, and Richard D’Ari. 2007. “Escherichia Coli Physiology in Luria-Bertani Broth.” Journal of Bacteriology 189 (23): 8746–49. https://doi.org/10.1128/jb.01368-07.