The escalating global concerns over chemical pesticide usage and increasing pest pressures under climate change underscore the urgent need for sustainable biocontrol strategies. Fungal endophytes have emerged as promising agents for managing fungal plant pathogens. However, a comprehensive synthesis specifically focused on fungal endophyte-mediated biocontrol of fungal diseases has been lacking. Here, we analyze 115 studies documenting fungal endophyte-mediated biocontrol across diverse crop diseases. Our analysis identifies Trichoderma, Aspergillus, Fusarium, Penicillium, Colletotrichum, and Alternaria as the most frequently investigated endophyte genera, with wheat, tomato, and grapevine as the most common sources. Herbaceous plants dominate as endophyte sources, while cultivated plants are sampled nearly twice as often as wild plants, revealing significant sampling biases. Root-derived endophytes are most frequently studied, whereas reproductive tissues remain underexplored despite their potential for vertical transmission. We provide a comprehensive overview of seven key biocontrol mechanisms: antibiosis, competition, induced systemic resistance, mycoparasitism, defense signaling modulation, volatile organic compound production, and growth-defense tradeoffs. Notably, 74% of studies report multifactorial mechanisms operating concurrently, underscoring the synergistic nature of endophyte-mediated protection. Critical translational gaps are identified: only 17% of studies have progressed to field validation, and single-strain applications dominate (64% of studies) while mixed consortia remain underoptimized. We also examine factors influencing biocontrol efficacy, including endophyte-host compatibility, environmental conditions, and plant genotype, alongside challenges in evaluation methods, field-scale assessments, long-term monitoring, and economic considerations. This review advances our understanding of fungal endophyte-mediated disease control and provides insights for developing sustainable agricultural technologies.

Abnormal pigment formation during late-stage fungal fermentation poses a significant challenge to industrial product quality. This study established an integrated framework to understand and control pigmentation in Aspergillus niger sodium gluconate fermentation. First, a metabolism-oriented scale-down strategy, simulating industrial glucose consumption trajectories, confirmed that metabolic rate deterioration drives pigment accumulation. Crucially, a rate-driven soft sensing strategy was developed by integrating online Raman spectroscopy, which provided real-time kinetic inputs to significantly enhance pigment prediction accuracy. Guided by this monitoring framework, inorganic salt composition was optimized to stabilize metabolic activity, effectively suppressing pigment formation. Crucially, fermentation validation confirmed that the strategy effectively suppressed abnormal pigmentation, even under existing imperfect oxygen supply conditions. This work presented a transferable engineering paradigm that combines metabolic characterization, intelligent spectral monitoring, and rational nutritional intervention, offering a robust solution for quality control in large-scale fungal fermentations.

The Cre/loxP system is a versatile tool for genome engineering, yet its application in the industrial yeast Komagataella phaffii has been limited primarily to marker recycling. This study reports a comprehensive Cre/loxP toolkit for K. phaffii. The N-terminal SV40 NLS fusion (NLS-Cre) was identified as the optimal expression strategy balancing high recombination efficiency (24.6%) with minimal cellular toxicity. Further, site-specific integration, inversion, and replacement were experimentally validated, demonstrating multiple recombination functionalities of the Cre/loxP systems in this host. High-throughput screening of 60 loxP variants revealed a structure-activity relationship, that is, spacer mutations frequently retained high recombination activity, whereas inverted repeat mutations uniformly abolished function. Therein, 12 variants with efficiency over 65% were identified, including five exceeding 80% (lox2271, loxm52, loxm71, lox514, and lox5171). Orthogonality assessment identified four variants (loxm72, loxm41, lox5272, loxm71) highly orthogonal to the wild-type loxP, offering potential for sequential genome engineering in strains with pre-existing sites. The obtained toolkit provides the first systematic validation of multiple recombination functions and loxP variants in K. phaffii, offering foundational components for genome engineering in this industrial strain.

Single-cell transcriptomics is revolutionizing our understanding of cellular diversity, yet comparing transcriptional programs across the tree of life remains challenging. We developed TranscriptFormer, a family of generative foundation models trained on up to 112 million cells spanning 1.53 billion years of evolution across 12 species. We demonstrate state-of-the-art performance on cell type classification, even for species separated by over 685 million years of evolution, and zero-shot disease state identification in human cells. Developmental trajectories, phylogenetic relationships, and cellular hierarchies emerge naturally in TranscriptFormer’s representations without any explicit training on these annotations. This work establishes a powerful framework for quantitative single-cell analysis and comparative cellular biology, thus demonstrating that universal principles of cellular organization can be learned and predicted across the tree of life.

Crude oil production has been enhanced using microbial-enhanced oil recovery (MEOR) in many oilfields. After over six years of MEOR test and field application in high-salt environments and unconventional oil reservoirs of plateau oilfields, MEOR mechanisms of emulsification, surface tension reduction, petroleum degradation, and reservoir microbial community structure adjustment were developed. Based on the core MEOR mechanisms of microbes, we carried out paraffin wax removal, single-well huff-puff, and microbial flooding in the Qinghai oilfield, thereby increasing oil production and sustaining economic benefits. These results demonstrate that microorganisms can be enhanced and applied in oilfields with large temperature differences, high salinity, and unconventional reservoirs.

Bacillus subtilis was widely used for enzyme production and was gradually engineered to biosynthesize value-added chemicals with the development of genetic parts for it. Compared to other genetic parts for expression, the identified integration sites were fewer, which limits the development of B. subtilis. Here, a library of integration sites was developed for B. subtilis. All candidate sites were selected at the 3′-untranslated region of two opposite nonessential genes and separated by essential genes among genome to avoid the destruction for coding sequence of genes and eliminate the homologously recombined strains with the loss of essential genes. The expression of GFP and cell growth were detected for candidate sites to evaluate the gene expression strength and the influence on cell growth. As a result, 12 loci revealed higher gene expression level and cell growth compared with control site amyE, the highest expression site spxA was 1.89 times as high as amyE. Using the developed integration sites library, threefold gene expression range could be achieved without the replacement of promoter and RBS. When the integration site library was used to construct cell factories, the production of lacto-N-triose II and lycopene was increased by 95% and 83%, respectively. In addition, integration site library was also successfully used to increase the enzymatic activity of secretory β-galactosidase by 101% when the strain using spxA locus compared with that using amyE. The developed integration site library could accelerate the construction of stable and plasmid-free cell factories for B. subtilis in the future.