Protein engineering via Bayesian optimization-guided evolutionary algorithm and robotic experiments.

Directed protein evolution applies repeated rounds of genetic mutagenesis and phenotypic screening and is often limited by experimental throughput. Through in silico prioritization of mutant sequences, machine learning has been applied to reduce wet lab burden to a level practical for human researchers. On the other hand, robotics permits large batches and rapid iterations for protein engineering cycles, but such capacities have not been well exploited in existing machine learning-assisted directed evolution approaches. Here, we report a scalable and batched method, Bayesian Optimization-guided EVOlutionary (BO-EVO) algorithm, to guide multiple rounds of robotic experiments to explore protein fitness landscapes of combinatorial mutagenesis libraries. We first examined various design specifications based on an empirical landscape of protein G domain B1. Then, BO-EVO was successfully generalized to another empirical landscape of an Escherichia coli kinase PhoQ, as well as simulated NK landscapes with up to moderate epistasis. This approach was then applied to guide robotic library creation and screening to engineer enzyme specificity of RhlA, a key biosynthetic enzyme for rhamnolipid biosurfactants. A 4.8-fold improvement in producing a target rhamnolipid congener was achieved after examining less than 1% of all possible mutants after four iterations. Overall, BO-EVO proves to be an efficient and general approach to guide combinatorial protein engineering without prior knowledge.

Identification of a unique 1,4-ß-D-glucan glucohydrolase of glycoside hydrolase family 9 from Cytophaga hutchinsonii.

Cytophaga hutchinsonii is an aerobic cellulolytic soil bacterium that rapidly digests crystalline cellulose. The predicted mechanism by which C. hutchinsonii digests cellulose differs from that of other known cellulolytic bacteria and fungi. The genome of C. hutchinsonii contains 22 glycoside hydrolase (GH) genes, which may be involved in cellulose degradation. One predicted GH with uncertain specificity, CHU_0961, is a modular enzyme with several modules. In this study, phylogenetic tree of the catalytic modules of the GH9 enzymes showed that CHU_0961 and its homologues formed a new group (group C) of GH9 enzymes. The catalytic module of CHU_0961 (CHU_0961B) was identified as a 1,4-ß-D-glucan glucohydrolase (EC 3.2.1.74) that has unique properties compared with known GH9 cellulases. CHU_0961B showed highest activity against barley glucan, but low activity against other polysaccharides. Interestingly, CHU_0961B showed similar activity against ρ-nitrophenyl ß-D-cellobioside (ρ-NPC) and ρ-nitrophenyl ß-D-glucopyranoside. CHU_0961B released glucose from the nonreducing end of cello-oligosaccharides, ρ-NPC, and barley glucan in a nonprocessive exo-type mode. CHU_0961B also showed same hydrolysis mode against deacetyl-chitooligosaccharides as against cello-oligosaccharides. The kcat/Km values for CHU_0961B against cello-oligosaccharides increased as the degree of polymerization increased, and its kcat/Km for cellohexose was 750 times higher than that for cellobiose. Site-directed mutagenesis showed that threonine 321 in CHU_0961 played a role in hydrolyzing cellobiose to glucose. CHU_0961 may act synergistically with other cellulases to convert cellulose to glucose on the bacterial cell surface. The end product, glucose, may initiate cellulose degradation to provide nutrients for bacterial proliferation in the early stage of C. hutchinsonii growth. KEY POINTS: • CHU_0961 and its homologues formed a novel group (group C) of GH9 enzymes. • CHU_0961 was identified as a 1,4-ß-d-glucan glucohydrolase with unique properties. • CHU_0961 may play an important role in the early stage of C. hutchinsonii growth.

Transcription Factor Atf1 Regulates Expression of Cellulase and Xylanase Genes during Solid-State Fermentation of Ascomycetes.

Transcriptional regulation of cellulolytic and xylolytic genes in ascomycete fungi is controlled by specific carbon sources in different external environments. Here, comparative transcriptomic analyses of Penicillium oxalicum grown on wheat bran (WB), WB plus rice straw (WR), or WB plus Avicel (WA) as the sole carbon source under solid-state fermentation (SSF) revealed that most of the differentially expressed genes (DEGs) were involved in metabolism, specifically, carbohydrate metabolism. Of the DEGs, the basic core carbohydrate-active enzyme-encoding genes which responded to the plant biomass resources were identified in P. oxalicum, and their transcriptional levels changed to various extents depending on the different carbon sources. Moreover, this study found that three deletion mutants of genes encoding putative transcription factors showed significant alterations in filter paper cellulase production compared with that of a parental P. oxalicum strain with a deletion of Ku70 (ΔPoxKu70 strain) when grown on WR under SSF. Importantly, the ΔPoxAtf1 mutant (with a deletion of P. oxalicumAtf1, also called POX03016) displayed 46.1 to 183.2% more cellulase and xylanase production than a ΔPoxKu70 mutant after 2 days of growth on WR. RNA sequencing and quantitative reverse transcription-PCR revealed that PoxAtf1 dynamically regulated the expression of major cellulase and xylanase genes under SSF. PoxAtf1 bound to the promoter regions of the key cellulase and xylanase genes in vitro This study provides novel insights into the regulatory mechanism of fungal cellulase and xylanase gene expression under SSF.IMPORTANCE The transition to a more environmentally friendly economy encourages studies involving the high-value-added utilization of lignocellulosic biomass. Solid-state fermentation (SSF), that simulates the natural habitat of soil microorganisms, is used for a variety of applications such as biomass biorefinery. Prior to the current study, our understanding of genome-wide gene expression and of the regulation of gene expression of lignocellulose-degrading enzymes in ascomycete fungi during SSF was limited. Here, we employed RNA sequencing and genetic analyses to investigate transcriptomes of Penicillium oxalicum strain EU2101 cultured on medium containing different carbon sources and to identify and characterize transcription factors for regulating the expression of cellulase and xylanase genes during SSF. The results generated will provide novel insights into genetic engineering of filamentous fungi to further increase enzyme production.

Purification and characterization of an endo-xylanase from Trichoderma sp., with xylobiose as the main product from xylan hydrolysis.

Fungal endo-ß-1,4-xylanases (endo-xylanases) can hydrolyze xylan into xylooligosaccharides (XOS), and have potential biotechnological applications for the exploitation of natural renewable polysaccharides. In the current study, we aimed to screen and characterize an efficient fungal endo-xylanase from 100 natural humus-rich soil samples collected in Guizhou Province, China, using extracted sugarcane bagasse xylan (SBX) as the sole carbon source. Initially, 182 fungal isolates producing xylanases were selected, among which Trichoderma sp. strain TP3-36 was identified as showing the highest xylanase activity of 295 U/mL with xylobiose (X2) as the main product when beechwood xylan was used as substrate. Subsequently, a glycoside hydrolase family 11 endo-xylanase, TXyn11A, was purified from strain TP3-36, and its optimal pH and temperature for activity against beechwood xylan were identified to be 5.0 and 55 °C, respectively. TXyn11A was stable across a broad pH range (3.0-10.0), and exhibited strict substrate specificity, including xylan from beechwood, wheat, rye, and sugarcane bagasse, with Km and Vmax values of 5 mg/mL and 1250 µmol/mg min, respectively, toward beechwood xylan. Intriguingly, the main product obtained from hydrolysis of beechwood xylan by TXyn11A was xylobiose, whereas SBX hydrolysis resulted in both X2 and xylotriose. Overall, these characteristics of the endo-xylanase TXyn11A indicate several potential industrial applications.

Characterization of novel roles of a HMG-box protein PoxHmbB in biomass-degrading enzyme production by Penicillium oxalicum.

High-mobility group (HMG)-box proteins are involved in chromatin organization in eukaryotes, especially in sex determination and regulation of mitochondrial DNA compaction. Although a novel HMG-box protein, PoxHmbB, had been initially identified to be required for filter paper cellulase activity by Penicillium oxalicum, the biological roles of HMG-box proteins in biomass-degrading enzyme production have not been systematically explored. The P. oxalicum mutant ∆PoxHmbB lost 34.7-86.5% of cellulase (endoglucanase, p-nitrophenyl-ß-cellobiosidase, and p-nitrophenyl-ß-glucopyranosidase) activities and 60.3% of xylanase activity following Avicel induction, whereas it exhibited about onefold increase in amylase activity following soluble corn starch induction. Furthermore, ∆PoxHmbB presented delayed conidiation and hyphae growth. Transcriptomic profiling and real-time quantitative reverse transcription-PCR revealed that PoxHmbB regulated the expression of major genes encoding plant biomass-degrading enzymes such as PoxCel7A-2, PoxCel5B, PoxBgl3A, PoxXyn11B, and PoxGA15A, as well as those involved in conidiation such as PoxBrlA. In vitro binding experiments further confirmed that PoxHmbB directly binds to the promoter regions of these major genes. These results further indicate the diversity of the biological functions of HMG-box proteins and provide a novel and promising engineering target for improving plant biomass-degrading enzyme production in filamentous fungi.

Deep Mutational Scanning of an Oxygen-Independent Fluorescent Protein CreiLOV for Comprehensive Profiling of Mutational and Epistatic Effects.

Oxygen-independent, flavin mononucleotide-based fluorescent proteins (FbFPs) are promising alternatives to green fluorescent protein in anaerobic contexts. Deep mutational scanning performs systematic profiling of protein sequence-function relationships but has not been applied to FbFPs. Focusing on CreiLOV from Chlamydomonas reinhardtii, we created and analyzed two comprehensive mutant collections: (1) single-residue, site-saturation mutagenesis libraries covering all 118 residues; and (2) a full combinatorial metagenesis library among 20 mutations at 15 residues, where mutation and residue selection was based on single-site mutagenesis results. Notably, the second type of library is indispensable to study higher-order epistasis but underrepresented in the literature. Using optimized FACS-seq assays, 2,185 (>92.5%) out of 2,360 possible single-site mutants and 165,428 (>89.7%) out of 184,320 possible combinatorial mutants were reliably assigned with fitness values. We constructed statistical and machine-learning models to analyze the CreiLOV data set, enabling accurate fitness prediction of higher-order mutants using lower-order mutagenesis data. In addition, we successfully isolated CreiLOV variants with improved fluorescence quantum yield and thermostability. This work provides new empirical data and design rules to engineer combinatorial protein variants.

Towards one sample per second for mass spectrometric screening of engineered microbial strains.

Microbial cell factories convert renewable feedstocks into desirable chemicals and materials. Due to the lack of predictive modeling, high-throughput screening remains essential for microbial strain engineering. Mass spectrometry (MS) is a label-free modality with superior sensitivity and chemical specificity. Critical advances in improving the throughput of MS assays on complex microbial samples include massively parallel cultivation, robotic sample preparation, and chromatography-free instrumentation. Here, we review the recent development and application of rapid MS assays in screening microbial libraries, achieving or approaching a rate of one sample per second. We conclude with unique challenges associated with MS screening of strain libraries and discuss future solutions.

Robotic Construction and Screening of Lanthipeptide Variant Libraries in Escherichia coli.

Lanthipeptides are a major class of ribosomally synthesized and post-translationally modified peptides (RiPPs) characterized by thioether cross-links called lanthionine (Lan) and methyllanthionine (MeLan). Previously, we developed a method to produce mature lanthipeptides in recombinant Escherichia coli, but manual steps hinder large-scale analogue screening. Here we devised an automated workflow for creating and screening variant libraries of haloduracin, a two-component class II lanthipeptide. An integrated work cell of a synthetic biology foundry was programmed to robotically execute DNA library construction, host transformation, peptide production, mass spectrometry analysis, and activity screening by agar diffusion assay. For recombinantly produced Halα peptides, the sequence-activity relationship of 380 single-residue variants and >1300 triple-residue combinatorial variants were rapidly analyzed in microplates within weeks. The peptide expression levels in E. coli were also visualized via robotic creation and analysis of GFP-lanthipeptide fusions for select peptide mutants. Following shake-flask fermentation and purification, one Halα mutant was confirmed with enhanced specific antimicrobial activity relative to the wild-type peptide. Overall, this approach may be generally applicable for the high-throughput characterization and engineering of RiPP natural products.

Accelerating strain engineering in biofuel research via build and test automation of synthetic biology.

Biofuels are a type of sustainable and renewable energy. However, for the economical production of bulk-volume biofuels, biosystems design is particularly challenging to achieve sufficient yield, titer, and productivity. Because of the lack of predictive modeling, high-throughput screening remains essential. Recently established biofoundries provide an emerging infrastructure to accelerate biological design-build-test-learn (DBTL) cycles through the integration of robotics, synthetic biology, and informatics. In this review, we first introduce the technical advances of build and test automation in synthetic biology, focusing on the use of industry-standard microplates for DNA assembly, chassis engineering, and enzyme and strain screening. Proof-of-concept studies on prototypes of automated foundries are then discussed, for improving biomass deconstruction, metabolic conversion, and host robustness. We conclude with future challenges and opportunities in creating a flexible, versatile, and data-driven framework to support biofuel research and development in biofoundries.

Identification of an essential regulator controlling the production of raw-starch-digesting glucoamylase in Penicillium oxalicum.

BACKGROUND: Raw-starch-digesting glucoamylases (RSDGs) from filamentous fungi have great commercial values in starch processing; however, the regulatory mechanisms associated with their production in filamentous fungi remain unknown. Penicillium oxalicum HP7-1 isolated by our laboratory secretes RSDG with suitable properties but at low production levels. Here, we screened and identified novel regulators of RSDG gene expression in P. oxalicum through transcriptional profiling and genetic analyses. RESULTS: Penicillium oxalicum HP7-1 transcriptomes in the presence of glucose and starch, respectively, used as the sole carbon source were comparatively analyzed, resulting in screening of 23 candidate genes regulating the expression of RSDG genes. Following deletion of 15 of the candidate genes in the parental P. oxalicum strain ∆PoxKu70, enzymatic assays revealed five mutants exhibiting significant reduction in the production of raw-starch-digesting enzymes (RSDEs). The deleted genes (POX01907, POX03446, POX06509, POX07078, and POX09752), were the first report to regulate RSDE production of P. oxalicum. Further analysis revealed that ∆POX01907 lost the most RSDE production (83.4%), and that POX01907 regulated the expression of major amylase genes, including the RSDG gene POX01356/PoxGA15A, a glucoamylase gene POX02412, and the α-amylase gene POX09352/Amy13A, during the late-stage growth of P. oxalicum. CONCLUSION: Our results revealed a novel essential regulatory gene POX01907 encoding a transcription factor in controlling the production of RSDE, regulating the expression of an important RSDG gene POX01356/PoxGA15A, in P. oxalicum. These results provide insight into the regulatory mechanism of fungal amylolytic enzyme production.