Food Spoilage-Associated Leuconostoc, Lactococcus, and Lactobacillus Species Display Different Survival Strategies in Response to Competition

Abstract

Psychrotrophic lactic acid bacteria (LAB) are the prevailing spoilage organisms in packaged cold-stored meat products. Species composition and metabolic activities of such LAB spoilage communities are determined by the nature of the meat product, storage conditions, and interspecies interactions. Our knowledge of system level responses of LAB during such interactions is very limited. To expand it, we studied interactions between three common psychrotrophic spoilage LAB (Leuconostoc gelidum, Lactococcus piscium, and Lactobacillus oligofermentans) by comparing their time course transcriptome profiles obtained during their growth in individual, pairwise, and triple cultures. The study revealed how these LAB employed different strategies to cope with the consequences of interspecies competition. The fastest-growing bacterium, Le. gelidum, attempted to enhance its nutrient-scavenging and growth capabilities in the presence of other LAB through upregulation of carbohydrate catabolic pathways, pyruvate fermentation enzymes, and ribosomal proteins, whereas the slower-growing Lc. piscium and Lb. oligofermentans downregulated these functions. These findings may explain the competitive success and predominance of Le. gelidum in a variety of spoiled foods. Peculiarly, interspecies interactions induced overexpression of prophage genes and restriction modification systems (mechanisms of DNA exchange and protection against it) in Lc. piscium and Lb. oligofermentans but not in Le. gelidum Cocultivation induced also overexpression of the numerous putative adhesins in Lb. oligofermentans These adhesins might contribute to the survival of this slowly growing bacterium in actively growing meat spoilage communities.IMPORTANCE Despite the apparent relevance of LAB for biotechnology and human health, interactions between members of LAB communities are not well known. Knowledge of such interactions is crucial for understanding how these communities function and, consequently, whether there is any possibility to develop new strategies to interfere with their growth and to postpone spoilage of packaged and refrigerated foods. With the help of controlled experiments, detailed regulation events can be observed. This study gives an insight into the system level interactions and the different competition-induced survival strategies related to enhanced uptake and catabolism of carbon sources, overexpression of adhesins and putative bacteriocins, and the induction of exchange of genetic material. Even though this experiment dealt with only three LAB strains in vitro, these findings agreed well with the relative abundance patterns typically reported for these species in natural food microbial communities.

Keywords: Lactobacillus oligofermentans; Lactococcus piscium; Leuconostoc gelidum subsp. gasicomitatum; RNA sequencing; food spoilage; interspecies interactions.

FIG 1

Growth curves based on the optical density (OD₆₀₀) values (A to D) and viable cell counts (CFU/milliliter) (E to H) of the individual and mixed cultures of the three LAB. G, O, P, individual cultures of Le. gelidum, Lb. oligofermentans, and Lc. piscium, respectively; GP, OP, GO, GOP, cocultures (e.g., GP is a coculture of Le. gelidum and Lc. piscium). Panels A and E represent cultures obtained in the first batch, while other panels represent cultures obtained in the second batch (see Materials and Methods). The values are the averages for three biological replicates ± standard errors of the means (SEM). Statistically significant differences between individual cultures of each bacterium are marked: *, P value ≤ 0.05, two-tailed t test.

FIG 2

Growth curves of Le. gelidum (G), Lb. oligofermentans (O), and Lc. piscium (P) in their cocultures (GP, OP, GO, and GOP) based on viable cell counts (CFU/milliliter). G_in_GP indicates cell counts of Le. gelidum in the coculture of Le. gelidum with Lc. piscium. The values are the averages for three biological replicates ± SEM. Statistically significant differences (P value ≤ 0.05, two-tailed t test) between bacterial species in their pairwise cocultures are marked with black asterisks. In the triple culture, red and purple asterisks indicate statistically significant differences of cell counts of Le. gelidum and Lc. piscium, respectively, from the two other bacterial species.

FIG 3

Comparison of relative expression changes (log₂ FC) for 13 genes obtained using droplet digital reverse transcription-PCR (dd-RT-PCR) versus RNA sequencing (RNA-seq). A letter in parentheses after a gene name represents the source organism (G, P, or O represents Le. gelidum, Lc. piscium, or Lb. oligofermentans, respectively). Expression fold changes (FC) represent the ratio of gene expression (averaged across three replicates) between mixed and individual cultures, taken at the same time point. To test gene expression with dd-RT-PCR, the following mixed cultures were used: GP-11 h for nagB(G), glmS(G), pyrP(G), pfl(P), fruK(P), ilvA(P); GOP-11 h for ccpA(G), pdhD(G), mapA(G); OP-3 h for LACPI_0417(P), OP-11 h for ccpA(O), infC(O); GO-11 h for LACOL_1683(O), where, e.g., GP-11 h represents the mixed culture of Le. gelidum and Lc. piscium at 11 h. For dd-RT-PCR, the concentration of the housekeeping gene infB was used to normalize the gene concentrations between samples. For data points with a square marker, the changes in gene expression were statistically significant based on both RNA-seq (DESeq2 adjusted P value < 0.05) and dd-RT-PCR (two-tailed t test P value < 0.05). A full functional description of the genes can be found in Data Set S4.

FIG 4

Differential expression of genes, pathways, and functional groups in each of the three LAB during their cocultivation in the pairwise and triple cultures. For Le. gelidum, the columns “+ P,” “+ O,” and “+ P + O” represent differential expression of its genes/pathways during growth with Lc. piscium, Lb. oligofermentans, and both (Lc. piscium and Lb. oligofermentans), respectively. Upward red arrows indicate upregulation, and downward blue arrows indicate downregulation. The time points (3 h, 5 h, and 11 h) for the differential expression events are marked next to the arrows. Superscript a indicates catabolism and transport of hexoses and di-/oligosaccharides; superscript b indicates that cationic peptides might play the role of bacteriocins; superscript c indicates complete RM systems or their restriction enzymes.

FIG 5

TreeMaps of the GO terms enriched in the DE genes during interspecies interactions of Le. gelidum, Lc. piscium, and Lb. oligofermentans. Only the most representative interactions are shown. The size of the box for each GO term is proportional to the absolute value of its log₁₀ P value. Semantically similar GO terms form clusters distinguished by color. TreeMaps were built using the REVIGO Web server (57).