Isolation and characterisation of genes affecting low temperature growth in Serratia plymuthica RVH1

Refrigerated foods are one of the most rapidly growing segments in the food industry. Refrigerated storage is necessary because these foods are perishable and can support growth of microorganisms that can cause spoilage or foodborne disease. Psychrotrophy, i.e. the ability to grow at temperatures below 4 °C, is widespread among bacteria of diverse phylogenetic lineages, but often varies at the genus, species or even strain level. The adaptations that allow bacteria to grow at low temperature are of interest from a basic scientific viewpoint, but also because they may lead to improved strategies to control undesirable psychrotrophs in the food production chain. Studies on the molecular and cellular basis of psychrotrophs have mainly focussed on Gram-positive bacteria, Listeria monocytogenes in particular. Other psychrotrophic food-related bacteria particularly those belonging to the Enterobacteriaceae family, have been much less studied. Therefore, the current work was intended to improve our understanding of the mechanisms that are required in enterobacteria to grow under cold stress. Serratia plymuthica RVH1, a psychrotrophic strain isolated from a catering kitchen that has been studied extensively in our laboratory, was chosen for this work, and a genome-wide mutational approach was chosen as the initial strategy.

A transposon-based mutant library of S. plymuthica was constructed and screened for mutants showing impaired growth at low temperature but normal growth at optimal temperature [30 °C]. Four putative psychrotrophy mutants displaying consistent growth impairment at 4 and 10 °C were isolated and genetically analysed. The transposons in these four mutants were localised in open reading frames that were putatively identified as plsC [1-acyl-sn-glycerol-3 phosphate acyltransferase], mnmA [tRNA-specific 2-thiouridylase], an unnamed ORF [polysaccharide pyruvyl transferase], and ubiB [2-octaprenylphenol hydroxylase]. None of these genes, except for fatty acid biosynthesis [plsC], and none of the pathways in which these genes are involved have been previously related to psychrotrophy. Since plsC and mnmA mutants showed almost no and severe impaired growth, respectively, at 10 and 4 °C, these were selected for detailed analysis.

The mutant with transposon insertion in the upstream region of plsC gene encoding lysophosphatidic acid acyltansferase involved in biosynthesis of phosphatidic acid [PA, a primary intermediate in membrane glycerolipid biosynthesis], showed a six to sevenfold reduced ratio of palmitoleic acid to oleic acid [C16:1 / C18:1] though the ratio of saturated to unsaturated fatty acid was unaffected. Low temperature growth defect and fatty acid composition were mostly restored by introduction of a complementation plasmid overexpressing plsC. Low temperature growth was also partially restored by supplementation of C16:1 to the growth medium, indicating that a shift from C18:1 to C16:1 was required for psychrotrophy.The mutant was also significantly more susceptible to pressure treatment at 250 MPa but not at higher pressure, and its growth was reduced at low pH but not at elevated NaCl concentration. These results provided novel information on the role of fatty acid composition on bacterial stress tolerance. The impact of the observed fatty acid shift on cold adaptation is in line with the well-known homeoviscous adaptation principle, but how knock-out or modulation of PlsC activity in the mutant leads to this shift remains unclear. This mutant may prove useful in further studies addressing the precise function of PlsC in S. plymuthica.

A second mutant that was subjected to a more detail analysis harboured a transposon insertion in mnmA, a gene encoding a tRNA-specific 2-thiouridylase involved in 2-thiouridine modification [s2] at the wobble position in the anticodon stem loop. In a parallel screening in the context of another project in our research group, a mutant of another tRNA modification gene, mnmE, was identified to

support growth of E. coli MG1655 in mildly acidic conditions. These observations from independent screenings suggested a more general role for tRNA modification in stress management, and therefore we constructed single knockout of both genes in S. plymuthica, and analysed their tolerance to temperature, osmotic, acid, protein synthesis inhibitors, and oxidative stresses. While S. plymuthica required the MnmA-mediated modification for normal growth at low temperature, this modification affected growth at supra-optimal temperature in E. coli. Interestingly, the effects of MnmA on growth under temperature stress in both bacteria disappeared when the bacteria were grown at an elevated level [2 – 3 % w/v] of sodium chloride in the growth media. The MnmE-mediated modification had no influence on growth under temperature stress in S. plymuthica or E. coli. Nonetheless, both MnmA and MnmE were indispensable for normal growth of E. coli under mildly acid conditions while this role in S. plymuthica was less pronounced. Further, MnmA supported survival of a lethal tert-butyl hydroperoxide challenge. Lastly, the sensitivity to antibiotics inducing translation infidelity was also influenced by MnmA- and MnmE-mediated modification in strain-dependent manner confirming their involvement in protein biosynthesis. Modifications of uridine tRNA at the wobble position of ASL [Anticodon Stem Loop] are important for the efficiency and fidelity of the translation. Lacking these modifications causes translational inefficiency leading to physiological consequences that are magnified during stress conditions. This can be explained by two proposed models of the role of tRNA modification as reported by others. The first model assumes that translation of stress response proteins depends more strongly on modified tRNAs than translation of other proteins. This is because stress response transcripts are rich in codons that are decoded by modified tRNA’s. If the demand for such tRNAs cannot be fulfilled, the translation of the stress response transcripts decreases and that may lead to incapability of the cell to overcome stress. Second, the role of tRNA modifications on maintaining proteome homeostasis. In this model, the absence of modified tRNA’s in stress conditions induces ribosome pausing eliciting protein aggregates and thus introducing acute proteotoxic insults. Consequently, proteome integrity is perturbed.

In conclusion, this work has contributed to a better understanding of some genes and cellular functions required for psychrotrophy in S. plymuthica. Some of these functions had not been previously related to psychrotrophy. Fatty acid composition and tRNA modification have been studied in more detail in particular, and appear to play a role in tolerance to other stresses as well.  The knowledge of the cellular process and pathways involved in low temperature growth may be used for discovery of novel preservatives specifically targeting psychrotrophic bacteria in refrigerated foods, thus increasing the safety and stability of the foods.