Signal Peptide Efficiency: From High-Throughput Data to Prediction and Explanation.

The passage of proteins across biological membranes via the general secretory (Sec) pathway is a universally conserved process with critical functions in cell physiology and important industrial applications. Proteins are directed into the Sec pathway by a signal peptide at their N-terminus. Estimating the impact of physicochemical signal peptide features on protein secretion levels has not been achieved so far, partially due to the extreme sequence variability of signal peptides. To elucidate relevant features of the signal peptide sequence that influence secretion efficiency, an evaluation of ∼12,000 different designed signal peptides was performed using a novel miniaturized high-throughput assay. The results were used to train a machine learning model, and a post-hoc explanation of the model is provided. By describing each signal peptide with a selection of 156 physicochemical features, it is now possible to both quantify feature importance and predict the protein secretion levels directed by each signal peptide. Our analyses allow the detection and explanation of the relevant signal peptide features influencing the efficiency of protein secretion, generating a versatile tool for the de novo design and in silico evaluation of signal peptides.

Safety evaluation of food enzymes produced by a safe strain lineage of Bacillussubtilis.

The safety of microbially-derived food enzymes must be carefully assessed before market introduction. The production strain's safety is central to the assessment. In this paper, we have determined that DSM's Bacillus subtilis strain lineage can be considered safe for food enzyme production. The mutations introduced into this non-pathogenic and non-toxigenic microorganism do not lead to any safety concerns, as ensured by a thorough characterization of the strain lineage. The safety of both targeted and randomly introduced changes into the production strain's genome is confirmed by validating the absence of vector sequences and antibiotic resistance genes in all relevant production strains, and by demonstrating absence of cytotoxic peptide production. Furthermore, three food enzyme preparations produced by strains within this lineage did not show genotoxic potential. 90-day oral toxicity studies performed with the same enzyme preparations did not reveal toxicologically significant adverse effects. These results demonstrate absence of safety concerns from the introduced genetic modifications. Based on the establishment of this safe strain lineage, we postulate that future enzymes produced by current and new strains derived from the lineage can be safely developed without additional genotoxicity and systemic toxicity studies, allowing for a reduction of animal testing without compromising on product safety.

GP4: an integrated Gram-Positive Protein Prediction Pipeline for subcellular localization mimicking bacterial sorting.

Subcellular localization is a critical aspect of protein function and the potential application of proteins either as drugs or drug targets, or in industrial and domestic applications. However, the experimental determination of protein localization is time consuming and expensive. Therefore, various localization predictors have been developed for particular groups of species. Intriguingly, despite their major representation amongst biotechnological cell factories and pathogens, a meta-predictor based on sorting signals and specific for Gram-positive bacteria was still lacking. Here we present GP4, a protein subcellular localization meta-predictor mainly for Firmicutes, but also Actinobacteria, based on the combination of multiple tools, each specific for different sorting signals and compartments. Novelty elements include improved cell-wall protein prediction, including differentiation of the type of interaction, prediction of non-canonical secretion pathway target proteins, separate prediction of lipoproteins and better user experience in terms of parsability and interpretability of the results. GP4 aims at mimicking protein sorting as it would happen in a bacterial cell. As GP4 is not homology based, it has a broad applicability and does not depend on annotated databases with homologous proteins. Non-canonical usage may include little studied or novel species, synthetic and engineered organisms, and even re-use of the prediction data to develop custom prediction algorithms. Our benchmark analysis highlights the improved performance of GP4 compared to other widely used subcellular protein localization predictors. A webserver running GP4 is available at http://gp4.hpc.rug.nl/.

Exploring the Nonconserved Sequence Space of Synthetic Expression Modules in Bacillus subtilis.

Increasing protein expression levels is a key step in the commercial production of enzymes. Predicting promoter activity and translation initiation efficiency based solely on consensus sequences have so far met with mixed results. Here, we addressed this challenge using a "brute-force" approach by designing and synthesizing a large combinatorial library comprising ∼12 000 unique synthetic expression modules (SEMs) for Bacillus subtilis. Using GFP fluorescence as a reporter of gene expression, we obtained a dynamic expression range that spanned 5 orders of magnitude, as well as a maximal 13-fold increase in expression compared with that of the already strong veg expression module. Analyses of the synthetic modules indicated that sequences at the 5'-end of the mRNA were the most important contributing factor to the differences in expression levels, presumably by preventing formation of strong secondary mRNA structures that affect translation initiation. When the gfp coding region was replaced by the coding region of the xynA gene, encoding the industrially relevant B. subtilis xylanase enzyme, only a 3-fold improvement in xylanase production was observed. Moreover, the correlation between GFP and xylanase expression levels was weak. This suggests that the differences in expression levels between the gfp and xynA constructs were due to differences in 5'-end mRNA folding and consequential differences in the rates of translation initiation. Our data show that the use of large libraries of SEMs, in combination with high-throughput technologies, is a powerful approach to improve the production of a specific protein, but that the outcome cannot necessarily be extrapolated to other proteins.

Effect of Genome Position on Heterologous Gene Expression in Bacillus subtilis: An Unbiased Analysis.

A fixed gene copy number is important for the in silico construction of engineered synthetic networks. However, the copy number of integrated genes depends on their genomic location. This gene dosage effect is rarely addressed in synthetic biology. Two studies in Escherichia coli presented conflicting data on the impact of gene dosage. Here, we investigate how genome location and gene orientation influences expression in Bacillus subtilis. An important difference with the E. coli studies is that we used an unbiased genome integration approach mediated by random transposon insertion. We found that there is a strong gene dosage effect in fast growing B. subtilis cells, which can amount to a 5-fold difference in gene expression. In contrast, gene orientation with respect to DNA replication direction does not influence gene expression. Our study shows that gene dosage should be taken into account when designing synthetic circuits in B. subtilis and presumably other bacteria.

Is proteomics a reliable tool to probe the oxidative folding of bacterial membrane proteins?

The oxidative folding of proteins involves disulfide bond formation, which is usually catalyzed by thiol-disulfide oxidoreductases (TDORs). In bacteria, this process takes place in the cytoplasmic membrane and other extracytoplasmic compartments. While it is relatively easy to study oxidative folding of water-soluble proteins on a proteome-wide scale, this has remained a major challenge for membrane proteins due to their high hydrophobicity. Here, we have assessed whether proteomic techniques can be applied to probe the oxidative folding of membrane proteins using the Gram-positive bacterium Bacillus subtilis as a model organism. Specifically, we investigated the membrane proteome of a B. subtilis bdbCD mutant strain, which lacks the primary TDOR pair BdbC and BdbD, by gel-free mass spectrometry. In total, 18 membrane-associated proteins showed differing behavior in the bdbCD mutant and the parental strain. These included the ProA protein involved in osmoprotection. Consistent with the absence of ProA, the bdbCD mutant was found to be sensitive to osmotic shock. We hypothesize that membrane proteomics is a potentially effective approach to profile oxidative folding of bacterial membrane proteins.

Searching for microbial protein over-expression in a complex matrix using automated high throughput MS-based proteomics tools.

In the discovery of new enzymes genomic and cDNA expression libraries containing thousands of differential clones are generated to obtain biodiversity. These libraries need to be screened for the activity of interest. Removing so-called empty and redundant clones significantly reduces the size of these expression libraries and therefore speeds up new enzyme discovery. Here, we present a sensitive, generic workflow for high throughput screening of successful microbial protein over-expression in microtiter plates containing a complex matrix based on mass spectrometry techniques. MALDI-LTQ-Orbitrap screening followed by principal component analysis and peptide mass fingerprinting was developed to obtain a throughput of ∼12,000 samples per week. Alternatively, a UHPLC-MS(2) approach including MS(2) protein identification was developed for microorganisms with a complex protein secretome with a throughput of ∼2000 samples per week. TCA-induced protein precipitation enhanced by addition of bovine serum albumin is used for protein purification prior to MS detection. We show that this generic workflow can effectively reduce large expression libraries from fungi and bacteria to their minimal size by detection of successful protein over-expression using MS.

A distant coilin homologue is required for the formation of cajal bodies in Arabidopsis.

Cajal bodies (CBs) are subnuclear bodies that are widespread in eukaryotes, being found in mammals, many other vertebrates and in all plant species so far examined. They are mobile structures, moving, fusing, and budding within the nucleus. Here we describe a screen for Arabidopsis mutants with altered CBs and describe mutants that have smaller Cajal bodies (ncb-2, ncb-3), lack them altogether (ncb-1), have increased numbers of CBs (pcb) or have flattened CBs (ccb). We have identified the gene affected in the ncb mutants as a distant homolog of the vertebrate gene that encodes coilin (At1g13030) and have termed the resulting protein Atcoilin. A T-DNA insertional mutant in this gene (ncb-4) also lacks Cajal bodies. Overexpression of Atcoilin cDNA in ncb-1 restores Cajal bodies, which recruit U2B'' as in the wild type, but which are, however, much larger than in the wild type. Thus we have shown that At1g13030 is required for Cajal body formation in Arabidopsis, and we hypothesize that the level of its expression is correlated with Cajal body size. The Atcoilin gene is unaffected in pcb and ccb, suggesting that other genes can also affect CBs.