Synthetic and Systems Biotechnology最新文献

Full conversion of glucose and xylose from lignocellulosic hydrolysates is required for obtaining a high ethanol yield. However, glucose and xylose share flux in the pentose phosphate pathway (PPP) and glycolysis pathway (EMP), with glucose having a competitive advantage in the shared metabolic pathways. In this work, we knocked down ZWF1 to preclude glucose from entering the PPP. This reduced the [NADPH] level and disturbed growth on both glucose or xylose, confirming that the oxidative PPP, which begins with Zwf1p and ultimately leads to CO₂ production, is the primary source of NADPH in both glucose and xylose. Upon glucose depletion, gluconeogenesis is necessary to generate glucose-6-phosphate, the substrate of Zwf1p. We re-established the NADPH regeneration pathway by replacing the endogenous NAD⁺-dependent glyceraldehyde-3-phosphate dehydrogenase (GAPDH) gene TDH3 with heterogenous NADP ⁺ -GAPDH genes GDH, gapB, and GDP1. Among the resulting strains, the strain BZP1 (zwf1Δ, tdh3::GDP1) exhibited a similar xylose consumption rate before glucose depletion, but a 1.6-fold increased xylose consumption rate following glucose depletion compared to the original strain BSGX001, and the ethanol yield for total consumed sugars of BZP1 was 13.5% higher than BSGX001. This suggested that using the EMP instead of PPP to generate NADPH reduces the wasteful metabolic cycle and excess CO₂ release from oxidative PPP. Furthermore, we used a copper-repressing promoter to modulate the expression of ZWF1 and optimize the timing of turning off the ZWF1, therefore, to determine the competitive equilibrium between glucose-xylose co-metabolism. This strategy allowed fast growth in the early stage of fermentation and low waste in the following stages of fermentation.

Descriptors play a pivotal role in enzyme design for the greener synthesis of biochemicals, as they could characterize enzymes and chemicals from the physicochemical and evolutionary perspective. This study examined the effects of various descriptors on the performance of Random Forest model used for enzyme-chemical relationships prediction. We curated activity data of seven specific enzyme families from the literature and developed the pipeline for evaluation the machine learning model performance using 10-fold cross-validation. The influence of protein and chemical descriptors was assessed in three scenarios, which were predicting the activity of unknown relations between known enzymes and known chemicals (new relationship evaluation), predicting the activity of novel enzymes on known chemicals (new enzyme evaluation), and predicting the activity of new chemicals on known enzymes (new chemical evaluation). The results showed that protein descriptors significantly enhanced the classification performance of model on new enzyme evaluation in three out of the seven datasets with the greatest number of enzymes, whereas chemical descriptors appear no effect. A variety of sequence-based and structure-based protein descriptors were constructed, among which the esm-2 descriptor achieved the best results. Using enzyme families as labels showed that descriptors could cluster proteins well, which could explain the contributions of descriptors to the machine learning model. As a counterpart, in the new chemical evaluation, chemical descriptors made significant improvement in four out of the seven datasets, while protein descriptors appear no effect. We attempted to evaluate the generalization ability of the model by correlating the statistics of the datasets with the performance of the models. The results showed that datasets with higher sequence similarity were more likely to get better results in the new enzyme evaluation and datasets with more enzymes were more likely beneficial from the protein descriptor strategy. This work provides guidance for the development of machine learning models for specific enzyme families.

Methanotrophic bacteria are currently used industrially for the bioconversion of methane-rich natural gas and anaerobic digestion-derived biogas to valuable products. These bacteria may also serve to mitigate the negative effects of climate change by capturing atmospheric greenhouse gases. Several genetic tools have previously been developed for genetic and metabolic engineering of methanotrophs. However, the available tools for use in methanotrophs are significantly underdeveloped compared to many other industrially relevant bacteria, which hinders genetic and metabolic engineering of these biocatalysts. As such, expansion of the methanotroph genetic toolbox is needed to further our understanding of methanotrophy and develop biotechnologies that leverage these unique microbes for mitigation and conversion of methane to valuable products. Here, we determined the copy number of three broad-host-range plasmids in Methylococcus capsulatus Bath and Methylosinus trichosporium OB3b, representing phylogenetically diverse Gammaproteobacterial and Alphaproteobacterial methanotrophs, respectively. Further, we show that the commonly used synthetic Anderson series promoters are functional and exhibit similar relative activity in M. capsulatus and M. trichosporium OB3b, but the synthetic series had limited range. Thus, we mutagenized the native M. capsulatus particulate methane monooxygenase promoter and identified variants with activity that expand the activity range of synthetic, constitutive promoters functional not only in M. capsulatus, but also in Escherichia coli. Collectively, the tools developed here advance the methanotroph genetic engineering toolbox and represent additional synthetic genetic parts that may have broad applicability in Pseudomonadota bacteria.

Lactate is an important monomer for the synthesis of poly-lactate (PLA), which is a substitute for the petrochemical plastics. To achieve the goal of high lactate titer, rate, and yield for commercial production, efficient lactate production pathway is needed as well as genetic targets that affect high lactate production and tolerance. In this study, an LldR-based d-lactate biosensor with a broad dynamic range was first applied into Zymomonas mobilis to select mutant strains with strong GFP fluorescence, which could be the mutant strains with increased d-lactate production. Then, LldR-based d-lactate biosensor was combined with a genome-wide CRISPR interference (CRISPRi) library targeting the entire genome to generate thousands of mutants with gRNA targeting different genetic targets across the whole genome. Specifically, two mutant libraries were selected containing 10⁵ and 10⁴ mutants with different interference sites from two rounds of fluorescence-activated cell sorting (FACS), respectively. Two genetic targets of ZMO1323 and ZMO1530 were characterized and confirmed to be associated with the increased d-lactate production, further knockout of ZMO1323 and ZMO1530 resulted in a 15% and 21% increase of d-lactate production, respectively. This work thus not only established a high-throughput approach that combines genome-scale CRISPRi and biosensor-assisted screening to identify genetic targets associated with d-lactate production in Z. mobilis, but also provided a feasible high-throughput screening approach for rapid identification of genetic targets associated with strain performance for other industrial microorganisms.

Precisely controlling gene expression is beneficial for optimizing biosynthetic pathways for improving the production. However, promoters in nonconventional yeasts such as Ogataea polymorpha are always limited, which results in incompatible gene modulation. Here, we expanded the promoter library in O. polymorpha based on transcriptional data, among which 13 constitutive promoters had the strengths ranging from 0–55% of P_GAP, the commonly used strong constitutive promoter, and 2 were growth phase-dependent promoters. Subsequently, 2 hybrid growth phase-dependent promoters were constructed and characterized, which had 2-fold higher activities. Finally, promoter engineering was applied to precisely regulate cellular metabolism for efficient production of β-elemene. The glyceraldehyde-3-phosphate dehydrogenase gene GAP was downregulated to drive more flux into pentose phosphate pathway (PPP) and then to enhance the supply of acetyl-CoA by using phosphoketolase-phosphotransacetylase (PK-PTA) pathway. Coupled with the phase-dependent expression of synthase module (ERG20∼LsLTC2 fusion), the highest titer of 5.24 g/L with a yield of 0.037 g/(g glucose) was achieved in strain YY150U under fed-batch fermentation in shake flasks. This work characterized and engineered a series of promoters, that can be used to fine-tune genes for constructing efficient yeast cell factories.

Aspergillus niger is a highly versatile fungal strain utilized in industrial production. The expression levels of recombinant genes in A. niger can be enhanced by increasing the copy number. Nevertheless, given the prolonged gene editing cycle of A. niger, a “one-step” strategy facilitating the simultaneous integration of recombinant genes into multiple genomic loci would provide a definitive advantage. In our previous study, a visual multigene editing system (VMS) was designed to knock out five genes, employing a tRNA-sgRNA array that includes the pigment gene albA and the target genes. Building upon this system, hybrid donor DNAs (dDNAs) were introduced to establish a clustered regularly interspaced short palindromic repeats (CRISPR)-based multiplex integration toolkit. Firstly, a CRISPR-Cas9 homology-directed repair (CRISPR-HDR) system was constructed in A. niger by co-transforming the CRISPR-Cas9 plasmid (with a highly efficient sgRNA) and the dDNA, resulting in precise integration of recombinant xylanase gene xynA into the target loci (the β-glucosidase gene bgl, the amylase gene amyA, and the acid amylase gene ammA). Subsequently, the length of homology arms in the dDNA was optimized to achieve 100% editing efficiency at each of the three gene loci. To achieve efficient multiplex integration in A. niger, the CRISPR plasmid pLM2 carrying a sgRNA-tRNA array was employed for concurrent double-strand breaks at multiple loci (bgl, amyA, ammA, and albA). Hybrid dDNAs were then employed for repair, including dDNA1-3 (containing xynA expression cassettes without selection markers) and dDNA_albA (for albA knockout). Among the obtained white colonies (RLM2′), 23.5% exhibited concurrent replacement of the bgl, amyA, and ammA genes with xynA (three copies). Notably, the xynA activity obtained by simultaneous insertion into three loci was 48.6% higher compared to that obtained by insertion into only the bgl locus. Furthermore, this multiple integration toolkit successfully enhanced the expression of endogenous pectinase pelA and Candida antarctica lipase CALB. Hence, the combined application of VMS and the CRISPR-HDR system enabled the simultaneous application of multiple selection markers, facilitating the rapid generation in the A. niger cell factories.

Bacteria exhibit a rich repertoire of RNA molecules that intricately regulate gene expression at multiple hierarchical levels, including small RNAs (sRNAs), riboswitches, and antisense RNAs. Notably, the majority of these regulatory RNAs lack or have limited protein-coding capacity but play pivotal roles in orchestrating gene expression by modulating transcription, post-transcription or translation processes. Leveraging and redesigning these regulatory RNA elements have emerged as pivotal strategies in the domains of metabolic engineering and synthetic biology. While previous investigations predominantly focused on delineating the roles of regulatory RNA in Gram-negative bacterial models such as Escherichia coli and Salmonella enterica, this review aims to summarize the mechanisms and functionalities of endogenous regulatory RNAs inherent to typical Gram-positive bacteria, notably Bacillus subtilis. Furthermore, we explore the engineering and practical applications of these regulatory RNA elements in the arena of synthetic biology, employing B. subtilis as a foundational chassis.

The protein inverse folding problem, designing amino acid sequences that fold into desired protein structures, is a critical challenge in biological sciences. Despite numerous data-driven and knowledge-driven methods, there remains a need for a user-friendly toolkit that effectively integrates these approaches for in-silico protein design. In this paper, we present DIProT, an interactive protein design toolkit. DIProT leverages a non-autoregressive deep generative model to solve the inverse folding problem, combined with a protein structure prediction model. This integration allows users to incorporate prior knowledge into the design process, evaluate designs in silico, and form a virtual design loop with human feedback. Our inverse folding model demonstrates competitive performance in terms of effectiveness and efficiency on TS50 and CATH4.2 datasets, with promising sequence recovery and inference time. Case studies further illustrate how DIProT can facilitate user-guided protein design.

The goal of this study was to use statistical optimization to change the nutritional and environmental conditions so that Streptomyces baarensis MH-133 could make more active metabolites. Twelve trials were used to screen for critical variables influencing productivity using the Placket-Burman Design method. S. baarensis MH-133 is significantly influenced by elicitation, yeast extract, inoculum size, and incubation period in terms of antibacterial activity. A total of 27 experimental trials with various combinations of these factors were used to carry out the response surface technique using the Box-Behnken design. The analyses revealed that the model was highly significant (p < 0.001), with a lack-of-fit of 0.212 and a coefficient determination (R2) of 0.9224. Additionally, the model predicted that the response as inhibition zone diameter would reach a value of 27 mm. Under optimal conditions, S. baarensis MH-133 produced 18.0 g of crude extract to each 35L and was purified with column chromatography. The active fraction exhibiting antibacterial activity was characterized using spectroscopic analysis. The MIC and MBC values varied between 37.5 and 300 μg/ml and 75 and 300 μg/ml, respectively. In conclusion, the biostatistical optimization of the active fraction critical variables, including environmental and nutritional conditions, enhances the production of bioactive molecules by Streptomyces species.

Benzyl and phenylpropanoid acids are widely used in organic synthesis of fine chemicals, such as pharmaceuticals and condiments. However, biocatalysis of these acids has received less attention than chemical synthesis. One of the main challenges for biological production is the limited availability of alcohol dehydrogenases and aldehyde dehydrogenases. Environmental microorganisms are potential sources of these enzymes. In this study, 129 alcohol dehydrogenases and 42 aldehyde dehydrogenases from Corynebacterium glutamicum, Pseudomonas aeruginosa, and Bacillus subtilis were identified and explored with various benzyl and phenylpropanoid alcohol and aldehyde substrates, among which four alcohol dehydrogenases and four aldehyde dehydrogenases with broad substrate specificity and high catalytic activity were obtained. Moreover, a cascade whole-cell catalytic system including ADH-90, ALDH-40, and the NAD(P)H oxidase LreNox was established, which showed high efficiency in converting cinnamyl alcohol and p-methylbenzyl alcohol into the respective carboxylic acids. Remarkably, this biocatalytic system can be easily scaled up to gram-level production, facilitating preparation purposes.