Protein-protein interaction and site prediction using transfer learning.

The advanced language models have enabled us to recognize protein-protein interactions (PPIs) and interaction sites using protein sequences or structures. Here, we trained the MindSpore ProteinBERT (MP-BERT) model, a Bidirectional Encoder Representation from Transformers, using protein pairs as inputs, making it suitable for identifying PPIs and their respective interaction sites. The pretrained model (MP-BERT) was fine-tuned as MPB-PPI (MP-BERT on PPI) and demonstrated its superiority over the state-of-the-art models on diverse benchmark datasets for predicting PPIs. Moreover, the model's capability to recognize PPIs among various organisms was evaluated on multiple organisms. An amalgamated organism model was designed, exhibiting a high level of generalization across the majority of organisms and attaining an accuracy of 92.65%. The model was also customized to predict interaction site propensity by fine-tuning it with PPI site data as MPB-PPISP. Our method facilitates the prediction of both PPIs and their interaction sites, thereby illustrating the potency of transfer learning in dealing with the protein pair task.

Mining and rational design of psychrophilic catalases using metagenomics and deep learning models.

A complete catalase-encoding gene, designated soiCat1, was obtained from soil samples via metagenomic sequencing, assembly, and gene prediction. soiCat1 showed 73% identity to a catalase-encoding gene of Mucilaginibacter rubeus strain P1, and the amino acid sequence of soiCAT1 showed 99% similarity to the catalase of a psychrophilic bacterium, Pedobacter cryoconitis. soiCAT1 was identified as a psychrophilic enzyme due to the low optimum temperature predicted by the deep learning model Preoptem, which was subsequently validated through analysis of enzymatic properties. Experimental results showed that soiCAT1 has a very narrow range of optimum temperature, with maximal specific activity occurring at the lowest test temperature (4 °C) and decreasing with increasing reaction temperature from 4 to 50 °C. To rationally design soiCAT1 with an improved temperature range, soiCAT1 was engineered through site-directed mutagenesis based on molecular evolution data analyzed through position-specific amino acid possibility calculation. Compared with the wild type, one mutant, soiCAT1S205K, exhibited an extended range of optimum temperature ranging from 4 to 20 °C. The strategies used in this study may shed light on the mining of genes of interest and rational design of desirable proteins. KEY POINTS: • Numerous putative catalases were mined from soil samples via metagenomics. • A complete sequence encoding a psychrophilic catalase was obtained. • A mutant psychrophilic catalase with an extended range of optimum temperature was engineered through site-directed mutagenesis.

Combined assembly of long and short sequencing reads improve the efficiency of exploring the soil metagenome.

BACKGROUND: Advances in DNA sequencing technologies have transformed our capacity to perform life science research, decipher the dynamics of complex soil microbial communities and exploit them for plant disease management. However, soil is a complex conglomerate, which makes functional metagenomics studies very challenging. RESULTS: Metagenomes were assembled by long-read (PacBio, PB), short-read (Illumina, IL), and mixture of PB and IL (PI) sequencing of soil DNA samples were compared. Ortholog analyses and functional annotation revealed that the PI approach significantly increased the contig length of the metagenomic sequences compared to IL and enlarged the gene pool compared to PB. The PI approach also offered comparable or higher species abundance than either PB or IL alone, and showed significant advantages for studying natural product biosynthetic genes in the soil microbiomes. CONCLUSION: Our results provide an effective strategy for combining long and short-read DNA sequencing data to explore and distill the maximum information out of soil metagenomics.

Rational redesign of thermophilic PET hydrolase LCCICCG to enhance hydrolysis of high crystallinity polyethylene terephthalates.

Polyethylene terephthalate (PET)-degrading enzymes represent a promising solution to the plastic pollution. However, PET-degrading enzymes, even thermophilic PETase, can effectively degrade low-crystallinity (∼8%) PETs, but exhibit weak depolymerization of more common, high-crystallinity (30-50%) PETs. Here, based on the thermophilic PETase, LCCICCG, we proposed two strategies for rational redesign of LCCICCG using the machine learning tool, Preoptem, combined with evolutionary analysis. Six single-point mutants (S32L, D18T, S98R, T157P, E173Q, N213P) were obtained that exhibit higher catalytic efficiency towards PET powder than wild-type LCCICCG at 75 °C. Additionally, the optimal temperature for degrading 39.07% crystalline PET increased from 65 °C in the wild-type LCCICCG to between 75 and 80 °C in the LCCICCG_I6M mutant that carries all six single-point mutations. Especially, the LCCICCG_I6M mutant has a significantly higher degradation effect on some commonly used bottle-grade plastic powders at 75-80 °C than that of wild type. The enzymatic digestion of ground 31.30% crystalline PET water bottles by LCCICCG_I6M yielded 31.91 ± 0.99 mM soluble products in 24 h, which was 3.64 times that of LCCICCG (8.77 ± 1.52 mM). Overall, this study provides a feasible route for engineering thermostable enzymes that can degrade high-crystallinity PET plastic.

MPEPE, a predictive approach to improve protein expression in E. coli based on deep learning.

The expression of proteins in Escherichia coli is often essential for their characterization, modification, and subsequent application. Gene sequence is the major factor contributing expression. In this study, we used the expression data from 6438 heterologous proteins under the same expression condition in E. coli to construct a deep learning classifier for screening high- and low-expression proteins. In conjunction with conserved residue analysis to minimize functional disruption, a mutation predictor for enhanced protein expression (MPEPE) was proposed to identify mutations conducive to protein expression. MPEPE identified mutation sites in laccase 13B22 and the glucose dehydrogenase FAD-AtGDH, that significantly increased both expression levels and activity of these proteins. Additionally, a significant correlation of 0.46 between the predicted high level expression propensity with the constructed models and the protein abundance of endogenous genes in E. coli was also been detected. Therefore, the study provides foundational insights into the relationship between specific amino acid usage, codon usage, and protein expression, and is essential for research and industrial applications.

A novel thermophilic chitinase directly mined from the marine metagenome using the deep learning tool Preoptem.

Chitin is abundant in nature and its degradation products are highly valuable for numerous applications. Thermophilic chitinases are increasingly appreciated for their capacity to biodegrade chitin at high temperatures and prolonged enzyme stability. Here, using deep learning approaches, we developed a prediction tool, Preoptem, to screen thermophilic proteins. A novel thermophilic chitinase, Chi304, was mined directly from the marine metagenome. Chi304 showed maximum activity at 85 â„ƒ, its Tm reached 89.65 ± 0.22℃, and exhibited excellent thermal stability at 80 and 90 °C. Chi304 had both endo- and exo-chitinase activities, and the (GlcNAc)2 was the main hydrolysis product of chitin-related substrates. The product yields of colloidal chitin degradation reached 97% within 80 min, and 20% over 4 days of reaction with crude chitin powder. This study thus provides a method to mine the novel thermophilic chitinase for efficient chitin biodegradation.

Protein engineering of stable IsPETase for PET plastic degradation by Premuse.

Poly(ethylene terephthalate) (PET) is used widely by human beings, but is very difficult to degrade. Up to now, the PET degradation effect of PETase from Ideonella sakaiensis 201-F6 (IsPETase) variants with low stability and activity was not ideal. In this study, a mutation design tool, Premuse, was developed to integrate the sequence alignment and quantitative selection of the preferred mutations based on natural sequence evolution. Ten single point mutants were selected from 1486 homologous sequences using Premuse, and then two mutations (W159H and F229Y) with improved stability were screened from them. The derived double point mutant, W159H/F229Y, exhibited a strikingly enhanced enzymatic performance. Its Tm and catalytic efficiency values (kcat/Km) respectively increased by 10.4 °C and 2.0-fold using p-NPP as the substrate compared with wild type. The degradation activity for amorphous PET was increased by almost 40-fold in comparison with wild type at 40 °C in 24 h. Additionally, the variant could catalyze biodegradation of PET bottle preform at a mean rate of 23.4 mgPET/h/mgenzyme. This study allowed us to design the mutation more efficiently, and provides a tool for achieving biodegradation of PET pollution under mild natural environments.

Metformin ameliorates ionizing irradiation-induced long-term hematopoietic stem cell injury in mice.

Exposure to ionizing radiation (IR) increases the production of reactive oxygen species (ROS) not only by the radiolysis of water but also through IR-induced perturbation of the cellular metabolism and disturbance of the balance of reduction/oxidation reactions. Our recent studies showed that the increased production of intracellular ROS induced by IR contributes to IR-induced late effects, particularly in the hematopoietic system, because inhibition of ROS production with an antioxidant after IR exposure can mitigate IR-induced long-term bone marrow (BM) injury. Metformin is a widely used drug for the treatment of type 2 diabetes. Metformin also has the ability to regulate cellular metabolism and ROS production by activating AMP-activated protein kinase. Therefore, we examined whether metformin can ameliorate IR-induced long-term BM injury in a total-body irradiation (TBI) mouse model. Our results showed that the administration of metformin significantly attenuated TBI-induced increases in ROS production and DNA damage and upregulation of NADPH oxidase 4 expression in BM hematopoietic stem cells (HSCs). These changes were associated with a significant increase in BM HSC frequency, a considerable improvement in in vitro and in vivo HSC function, and complete inhibition of upregulation of p16(Ink4a) in HSCs after TBI. These findings demonstrate that metformin can attenuate TBI-induced long-term BM injury at least in part by inhibiting the induction of chronic oxidative stress in HSCs and HSC senescence. Therefore, metformin has the potential to be used as a novel radioprotectant to ameliorate TBI-induced long-term BM injury.

Identification of specific gene modules in mouse lung tissue exposed to cigarette smoke.

BACKGROUND: Exposure to cigarette may affect human health and increase risk of a wide range of diseases including pulmonary diseases, such as chronic obstructive pulmonary disease (COPD), asthma, lung fibrosis and lung cancer. However, the molecular mechanisms of pathogenesis induced by cigarettes still remain obscure even with extensive studies. With systemic view, we attempted to identify the specific gene modules that might relate to injury caused by cigarette smoke and identify hub genes for potential therapeutic targets or biomarkers from specific gene modules. MATERIALS AND METHODS: The dataset GSE18344 was downloaded from the Gene Expression Omnibus (GEO) and divided into mouse cigarette smoke exposure and control groups. Subsequently, weighted gene co-expression network analysis (WGCNA) was used to construct a gene co-expression network for each group and detected specific gene modules of cigarette smoke exposure by comparison. RESULTS: A total of ten specific gene modules were identified only in the cigarette smoke exposure group but not in the control group. Seven hub genes were identified as well, including Fip1l1, Anp32a, Acsl4, Evl, Sdc1, Arap3 and Cd52. CONCLUSIONS: Specific gene modules may provide better understanding of molecular mechanisms, and hub genes are potential candidates of therapeutic targets that may possible improve development of novel treatment approaches.

Multi-allergen challenge stimulates steriod-resistant airway inflammation via NF-κB-mediated IL-8 expression.

Neutrophilic airway inflammation associated with multiple allergens has been related to steroid-resistant asthma. However, most animal models use only one allergen, which cannot simulate asthma closely as seen in patients. To determine the mechanism of inflammatory process involved in this severe condition, BALB/c mice were repetitively challenged with the pooled extract of dust mite, ragweed, and Aspergillus species (DRA). We found that DRA increased interleukin (IL)-10 and TGF-ß levels and neutrophil recruitment in bronchial alveolar lavage fluid. We also found that although dexamethasone suppressed the release of these two cytokines, mast cells recruitment, and mucus hypersecretion, it actually increased neutrophil infiltration and the level of keratinocyte-derived chemokine (mKC), a functional homolog of human IL-8. Treatment of human lung alveolar A549 cells with Der p1, an extract of house dust mite Dermatophagoides pteronyssinus, increased the expression of IL-8 and activity of NF-κB. The elevated IL-8 level was suppressed by BAY11-7082, a selective NF-κB inhibitor, but not by dexamethasone. These results suggest that increased IL-8 (mKC) levels may be involved in steroid-resistant neutrophilic airway inflammation through an NF-κB-dependent pathway.