Unlike substrate-requiring bioluminescent or fluorescent-based reporters, the bacterial luciferase (lux) reporter system is capable of synthesizing and scavenging its substrates from within its host cell environment, allowing for an autobioluminescent phenotype when expressed exogenously. Here we leverage this phenotype to determine if changes in the autobioluminescent
production of cultured human cells can serve as a near real-time bioreporter for bacterial contamination.
Autobioluminescent human cells (HEK293) stably expressing luxCDABEfrp genes were plated into 24 well plates at 1 × 106 cells/ml in 1 ml of DMEM supplemented with 10% FBS and lacking all antibiotic components. Cells were imaged over 24 h with a 10 min photon counting integration every 15 min. After the first 3 h, cells were treated with doses of either virulent or nonvirulent E. coli O157:H7 ranging from 1 to 6 × 106 in a volume of 100 μl LB media. As a vehicle control, autobioluminescent cells were similarly inoculated with 100 μl LB without bacteria.
The dynamics of autobioluminescent production were similar between groups of cells treated with virulent or non-virulent E. coli strains. This pattern remained across all inoculation dose ranges, but underwent shifts in the time to peak output and time to diminishment of signal compared to control. A shift in the autobioluminescent pattern was also observed among cells treated with equivalent inoculation doses of the virulent and non-virulent E. coli strains, with virulent strain treatment leading to diminishment below positive control output levels (p < 0.05) and complete extinction of bioluminescent production (p > 0.05 compared to cell free controls) 0.5 h prior to cells treated with non-virulent strains. The time until autobioluminescent production diminished below unexposed controls correlated strongly with the initial inoculation dose of toxic E. coli strains (R2 > 0.999) across all doses tested.
Because of their constitutive autobioluminescent phenotype, human cells expressing lux genes present a means to accurately and inexpensively determine the magnitude of an E. coli contamination event in cell culture. Virulent and non-virulent strains can be distinguished by the length of time required for cessation of the autobioluminescent signal, but cannot provide
information about virulence on a cellular level.
D.M. Close1,2, J.D. Webb3, S.A. Ripp2,4, and G.S. Sayler1,2,3,4
1Joint Institute for Biological Sciences, Oak Ridge National Laboratory
2490 BioTech, Inc., Knoxville, TN
3Department of Ecology and Evolutionary Biology, The University of Tennessee, Knoxville
4Center for Environmental Biotechnology, University of Tennessee, Knoxville