Exploring the secretome of Corynebacterium glutamicum ATCC 13032

Abstract

The demand for alternative sources of food proteins is increasing due to the limitations and challenges associated with conventional food production. Advances in biotechnology have enabled the production of proteins using microorganisms, thus prompting the exploration of attractive microbial hosts capable of producing functional proteins in high titers. Corynebacterium glutamicum is widely used in industry for the production of amino acids and has many advantages as a host organism for recombinant protein production. However, its performance in this area is limited by low yields of target proteins and high levels of native protein secretion. Despite representing a challenge for heterologous protein production, the C. glutamicum secretome has not been fully characterized. In this study, state-of-the-art mass spectrometry-based proteomics was used to identify and analyze the proteins secreted by C. glutamicum. Both the wild-type strain and a strain that produced and secreted a recombinant β-lactoglobulin protein were analyzed. A total of 427 proteins were identified in the culture supernatants, with 148 predicted to possess a secretion signal peptide. MS-based proteomics on the secretome enabled a comprehensive characterization and quantification (based on abundance) of the secreted proteins through label-free quantification (LFQ). The top 12 most abundant proteins accounted for almost 80% of the secretome. These are uncharacterized proteins of unknown function, resuscitation promoting factors, protein PS1, Porin B, ABC-type transporter protein and hypothetical membrane protein. The data can be leveraged for protein production by, e.g., utilizing the signal peptides of the most abundant proteins to improve secretion of heterologous proteins. In addition, secretory stress can potentially be alleviated by inactivating non-essential secreted proteins. Here we provide targets by identifying the most abundant, secreted proteins of which majority are of unknown function. The data from this study can thus provide valuable insight for researchers looking to improve protein secretion and optimize C. glutamicum as a host for secretory protein production.

Keywords: Corynebacterium glutamicum; recombinant protein production; secretome analysis; α-lactalbumin; β-lactoglobulin.

FIGURE 1

Secretome analysis of C. glutamicum ATCC 13032 (A). Genes encoding the whey proteins were inserted in pEC-XC99E vector and used to transform C. glutamicum ATCC 13032. (B) Growth profiles of the whey protein secreting strains and the wildtype strain without the vector were monitored and cells were assessed for their viability at 2 h intervals. (C) Supernatant fractions of the wild type, and whey protein secreting strains were analyzed using SDS-PAGE and Western blot. (D) Supernatant fractions of the wild type, and whey protein secreting strains were precipitated and analyzed using LC-MS/MS.

FIGURE 2

Production of whey proteins: (A) The plasmid map of pEC-XC99E with the LALBA or LGB gene inserted. The expression cassette with the translational fusion of signal peptide, LALBA/LGB gene and 6xHis-tag under control of trc promoter and Shine-Dalgarno (SD) sequence is indicated. (B) Western blot using anti-His antibody of samples from the α-lactalbumin-producing (LA) and β-lactoglobulin-producing (LG) strains harvested at 14–24 h. (C) SDS-polyacrylamide gel of secreted proteins from the wild type (WT), and the α-lactalbumin-producing (LA) and β-lactoglobulin-producing (LG) strains. Samples harvested at 14 and 24 h post-induction are shown.

FIGURE 3

(A) Growth profile of the wild type C. glutamicum ATCC 13032 strain and its β-lactoglobulin-producing derivative (LGB). (A) Growth curve of the wild type (Green) and LGB (Red) strains cultured in CGXII 0.9% BHI medium at 30°C. The averages of three biological replicates are displayed. The arrows indicate the sampling times. (B) Fluorescence microscopy of live/dead-stained wild type cells sampled at 8, 14, and 24 h.

FIGURE 4

Secreted proteins identified in this and other studies. The Venn diagram illustrates the C. glutamicum ATCC 13032 proteins identified in this study and those of Hermann et al., 2001; Hansmeier et al., 2006. Total numbers identified and in parenthesis those containing a signal peptide are shown.

FIGURE 5

(A) Volcano plot of secreted proteins in the secretome of C. glutamicum strains WT and LGB at 14 and 24 h. (A–D)– Comparison of proteins from within and amongst the WT and LGB strains at 14 and 24 h. The dotted lines correspond to a fold-change cutoff of two-fold (Log₂ ≤ −1 or ≥1) and a p-value of 0.05. (B) Protein abundance of secreted proteins. (A–D)- Comparison of protein iBAQ values (a proxy for protein abundance) for supernatant at 14 and 24 h for the wildtype and the strain secreting LGB, and between the two wildtype and LGB secreting strains. Each data point represents a protein. The red data point represents β-lactoglobulin. The blue and yellow data points correspond to proteins with and without a signal peptide, respectively.

FIGURE 6

(A,B) Pathway analysis for proteins without a signal peptide. (A) Fold enrichment analysis of the genes without a secretion signal identified in the secretome. Genes are grouped into their respective pathways. (B) A hierarchical clustering tree summarizes the correlation among significant pathways listed in the enrichment figure (A). (C, D) Pathway analysis for proteins with a signal peptide. (C) Fold enrichment analysis of the genes without a secretion signal identified in the secretome. Genes are grouped into their respective pathways. (D) A hierarchical clustering tree summarizes the correlation among significant pathways listed in the enrichment figure (C). Pathways with many shared genes are clustered together. Bigger dots indicate more significant p-values.