Growth Kinetics of Parent and Green Fluorescent Protein-Producing Strains of Salmonella

1. Growth Kinetics of Parent and Green Fluorescent Protein-Producing Strains of Salmonella Thomas P. Oscar, Agricultural Research Service, USDA, 1124 Trigg Hall, UMES, Princess Anne, MD 21853 410-651-6062; 410-651-6568 (fax); toscar@mail.umes.edu Introduction The green fluorescent protein (GFP) is a small polypeptide (27 kDa) from the jellyfish Aequora victoria that has been cloned and expressed in both prokaryotic and eukaryotic cells. Colonies of bacterial cells expressing GFP can be easily detected and counted by illuminating viable cell count plates with ultraviolet light. This is a desirable characteristic for predictive model development because it allows the automated counting of large numbers of plates without the need for addition of exogenous substrates. A number of studies with GFP-producing strains of bacteria indicate that GFP expression does not alter the biochemical, morphological or growth characteristics of the bacterium. However, only anecdotal or limited (i.e., at one temperature) data regarding the effects of GFP expression on microbial growth are provided in these studies. Objective To conduct a systematic comparison of the growth kinetics of parent and GFP-producing strains of Salmonella over a broad range of temperature. Hypothesis The hypothesis tested was that the GFP strains have growth kinetics that are not different from the parent strains and thus, would be suitable marker strains for constructing predictive models with naturally contaminated food. Experimental Approach Parent strains of Salmonella Typhimurium, Enteritidis and Dublin were transformed with a high copy plasmid encoding wild type GFP under the control of the lacZ promoter. Growth curves were obtained using cooked chicken burgers incubated at temperatures from 8 to 48C. Kinetic data were fit to a three phase linear model to determine lag time (LT), specific growth rate (SGR) and maximum population density (MPD) at each temperature. Secondary models for the growth parameters as a function of temperature were generated and compared among the parent and GFP strain pairs. Results The effects of GFP on LT were significant and slightly different among the serotypes of Salmonella. Whether GFP increased, decreased or did not alter LT depended on the incubation temperature and serotype (Figure A to C). GFP reduced SGR in the three serotypes tested. The magnitude of the reduction in SGR was dependent on the incubation temperature and serotype (Figure D to F). The most consistent effect was that GFP reduced the optimum SGR by 0.17 to 0.2 log CFU per h. constitutively express GFP. Such a high level of marker protein expression could slow growth and decrease maximum population density by creating a competition for and eventual deficiency of essential nutrients. Although the results of this study indicated that the GFP strains tested displayed different growth kinetics than the parent strains and thus, would not be good strains for developing predictive models in naturally contaminated food, it should be possible to construct marker strains of Salmonella that do not over-express GFP and grow in a manner similar to the parent strains. For example, by placing gfp under the control of a different promoter that requires an inducer not found in food, the expression of GFP could be repressed during the growth of the pathogen on the food but then induced during growth of the pathogen on the viable cell count plate by including the inducer in the agar medium. Not all of the growth curves exhibited three-phases of growth. In some instances, sampling was not extended for enough time to detect the stationary phase. Consequently, MPD data were not obtained for all incubation temperatures. Nonetheless, GFP decreased MPD on the chicken burgers (Figure G to I). Likewise, GFP reduced MPD by 1 to 1.5 log cycles in the starter cultures used to inoculate the burgers (results not shown). Discussion The failure of the three GFP strains tested to display similar growth kinetics as the parent strains may have resulted from over-expression of GFP. The plasmid encoding GFP in the current study was a high copy plasmid in which gfp was under the control of the lacZ promoter for which most Salmonella do not have a lacI gene encoding for the lac repressor protein. It has been reported that GFP accounts for up to 75% of total cellular protein in bacteria that