Carbohydrate starvation causes a metabolically active but nonculturable state in Lactococcus lactis

Abstract

This study characterized the ability of lactococci to become nonculturable under carbohydrate starvation while maintaining metabolic activity. We determined the changes in physiological parameters and extracellular substrate levels of multiple lactococcal strains under a number of environmental conditions along with whole-genome expression profiles. Three distinct phases were observed, logarithmic growth, sugar exhaustion, and nonculturability. Shortly after carbohydrate starvation, each lactococcal strain lost the ability to form colonies on solid media but maintained an intact cell membrane and metabolic activity for over 3.5 years. ML3, a strain that metabolized lactose rapidly, reached nonculturability within 1 week. Strains that metabolized lactose slowly (SK11) or not at all (IL1403) required 1 to 3 months to become nonculturable. In all cases, the cells contained at least 100 pM of intracellular ATP after 6 months of starvation and remained at that level for the remainder of the study. Aminopeptidase and lipase/esterase activities decreased below detection limits during the nonculturable phase. During sugar exhaustion and entry into nonculturability, serine and methionine were produced, while glutamine and arginine were depleted from the medium. The cells retained the ability to transport amino acids via proton motive force and peptides via ATP-driven translocation. The addition of branched-chain amino acids to the culture medium resulted in increased intracellular ATP levels and new metabolic products, indicating that branched-chain amino acid catabolism resulted in energy and metabolic products to support survival during starvation. Gene expression analysis showed that the genes responsible for sugar metabolism were repressed as the cells entered nonculturability. The genes responsible for cell division were repressed, while autolysis and cell wall metabolism genes were induced neither at starvation nor during nonculturability. Taken together, these observations verify that carbohydrate-starved lactococci attain a nonculturable state wherein sugar metabolism, cell division, and autolysis are repressed, allowing the cells to maintain transcription, metabolic activity, and energy production during a state that produces new metabolites not associated with logarithmic growth.

FIG. 1.

Cell counts during growth and carbohydrate starvation in buffered CDM. Plate counts of ML3 in buffered CDM containing 0.1% lactose at pH 7.0 (squares) (A) and plate counts of ML3 (squares), SK11 (diamonds), and IL1403 (circles) in CDM at pH 7.2 (B) and within the first 15 days of starvation (B1) are shown. Plate counts of ML3 (squares), SK11 (diamonds), and IL1403 (circles) in CDM at pH 5.2 (C) and within the first 15 days of starvation (C1) are also shown. ML3 data are shown for up to 2 years, but ML3 was NC for up to 3.5 years. The coefficient of variation ranged between 0.1 and 9% for all strains at each time point and under each pH condition.

FIG. 2.

Lactose utilization during growth and nonculturability in buffered CDM and intracellular ATP concentrations of cells. The initial lactose level was 0.2 to 0.25% for all media. (A) Lactose levels during long-term starvation of ML3 at pH 7.2 (filled squares) and pH 5.2 (open squares), SK11 at pH 7.2 (filled diamonds) and pH 5.2 (open diamonds), and IL1403 at pH 7.2 (filled circles) and pH 5.2 (open circles). (B) Lactose levels of ML3 at pH 7.0 during short-term starvation. Data are shown for 93 days, while lactose levels were maintained for 3.5 years. ATP levels of ML3 (squares), SK11 (diamonds), and IL1403 (circles) in CDM at pH 7.2 (C) and within the first 7 days of starvation (C1) are shown. ATP levels of ML3 (squares), SK11 (diamonds), and IL1403 (circles) in CDM at pH 5.2 (D) and within the first 7 days of starvation (D1) are also shown. ML3 data are shown for up to 2 years, but ML3 contained 100 pM ATP for up to 3.5 years. The coefficient of variation ranged between 0.8 and 11% for all strains at all time points and under all pH conditions tested.

FIG. 3.

Extracellular amino acid profile for ML3 grown in CDM with 0.1% lactose. Serine concentrations are depicted on the yy axis and all other amino acid concentrations on the y axis. The coefficient of variation ranged between 1 and 8% over all time points.

FIG. 4.

Schematic representation of gene expression of the lactococcal PEP-dependent PTS for various PTS sugars with connections to glycolytic intermediates. The symbol indicates gene repression. NAG, N-acetylglucosamine.