Protein Subcellular Localization Research Articles

A vector system for single and tandem expression of cloned genes and multi-colour fluorescent tagging in Haloferax volcanii.

Archaeal cell biology is an emerging field expected to identify fundamental cellular processes, help resolve the deep evolutionary history of cellular life, and contribute new components and functions in biotechnology and synthetic biology. To facilitate these, we have developed plasmid vectors that allow convenient cloning and production of proteins and fusion proteins with flexible, rigid, or semi-rigid linkers in the model archaeon Haloferax volcanii. For protein subcellular localization studies using fluorescent protein (FP) tags, we created vectors incorporating a range of codon-optimized fluorescent proteins for N- or C-terminal tagging, including GFP, mNeonGreen, mCherry, YPet, mTurquoise2 and mScarlet-I. Obtaining functional fusion proteins can be challenging with proteins involved in multiple interactions, mainly due to steric interference. We demonstrated the use of the new vector system to screen for improved function in cytoskeletal protein FP fusions, and identified FtsZ1-FPs that are functional in cell division and CetZ1-FPs that are functional in motility and rod cell development. Both the type of linker and the type of FP influenced the functionality of the resulting fusions. The vector design also facilitates convenient cloning and tandem expression of two genes or fusion genes, controlled by a modified tryptophan-inducible promoter, and we demonstrated its use for dual-colour imaging of tagged proteins in H. volcanii cells. These tools should promote further development and applications of archaeal molecular and cellular biology and biotechnology.

SCLpred-ECL: Subcellular Localization Prediction by Deep N-to-1 Convolutional Neural Networks.

The subcellular location of a protein provides valuable insights to bioinformaticians in terms of drug designs and discovery, genomics, and various other aspects of medical research. Experimental methods for protein subcellular localization determination are time-consuming and expensive, whereas computational methods, if accurate, would represent a much more efficient alternative. This article introduces an ab initio protein subcellular localization predictor based on an ensemble of Deep N-to-1 Convolutional Neural Networks. Our predictor is trained and tested on strict redundancy-reduced datasets and achieves 63% accuracy for the diverse number of classes. This predictor is a step towards bridging the gap between a protein sequence and the protein's function. It can potentially provide information about protein-protein interaction to facilitate drug design and processes like vaccine production that are essential to disease prevention.

O-GlcNAcylation in ischemic diseases.

Protein glycosylation is an extensively studied field, with the most studied forms being oxygen or nitrogen-linked N-acetylglucosamine (O-GlcNAc or N-GlcNAc) glycosylation. Particular residues on proteins are targeted by O-GlcNAcylation, which is among the most intricate post-translational modifications. Significantly contributing to an organism's proteome, it influences numerous factors affecting protein stability, function, and subcellular localization. It also modifies the cellular function of target proteins that have crucial responsibilities in controlling pathways related to the central nervous system, cardiovascular homeostasis, and other organ functions. Under conditions of acute stress, changes in the levels of O-GlcNAcylation of these proteins may have a defensive function. Nevertheless, deviant O-GlcNAcylation nullifies this safeguard and stimulates the advancement of several ailments, the prognosis of which relies on the cellular milieu. Hence, this review provides a concise overview of the function and comprehension of O-GlcNAcylation in ischemia diseases, aiming to facilitate the discovery of new therapeutic targets for efficient treatment, particularly in patients with diabetes.

Cloning and functional verification of the CmHSP17.9 gene from chrysanthemum.

Small molecular heat shock proteins (sHSPs) belong to the HSP family of molecular chaperones. Under high-temperature stress, they can prevent the aggregation of irreversible proteins and maintain the folding of denatured proteins to enhance heat resistance. In this study, the CmHSP17.9-1 and CmHSP17.9-2 genes, which were cloned from chrysanthemum (Chrysanthemum×morifolium 'Jinba') by homologous cloning, had a complete open reading frame of 480 bp each, encoding 159 amino acids. The protein subcellular localization analysis showed that CmHSP17.9-1 and CmHSP17.9-2 were located in the cytoplasm and mostly aggregated in granules, especially around the nucleus. Real-time quantitative PCR (qRT-PCR) analysis showed that the relative expression level of the CmHSP17.9-1 and CmHSP17.9-2 genes was highest in the terminal buds of the chrysanthemum, followed by the leaves. CmHSP17.9-1 and CmHSP17.9-2 overex-pression vectors were constructed and used to transform the chrysanthemum; overexpression of these genes led to the chrysanthemum phenotypes being less affected by high-temperature, and the antioxidant capacity was enhanced. The results showed that chrysanthemum with overex-pression of the CmHSP17.9-1 and CmHSP17.9-2 genes had stronger tolerance than the wild type chrysanthemum after high-temperature treatment or some degree of heat exercise, and overex-pression of the CmHSP17.9-1 gene led to stronger heat resistance than that of the CmHSP17.9-2 gene, providing an important theoretical basis for the subsequent molecular breeding and pro-duction applications of chrysanthemum.

NRPB1 is a WNK-like pseudokinase that modulates the activity of the WNK4-SPAK/OSR1 pathway with implications in the physiology of the distal convoluted tubule

The pseudokinase NRBP1 and the TSC22D protein family have been implicated in the regulation of cell proliferation and protein turnover. Preliminary observations from one of our laboratories (DA) revealed that, during osmotic stress, NRBP1 and TSC22D proteins interact with WNK1 in vitro. Mouse renal transcriptomic databases show that TSC22D1.1 and TSC22D2 expression is enriched in the distal convoluted tubule (DCT) of the nephron, where WNK kinases play an essential role in the regulation of Na+ reabsorption. WNK4 is the mayor catalytically active WNK in this nephron segment. It is known that WNK’s undergo liquid-liquid phase separation and form biomolecular condensates now called “WNK-bodies” that are observed in the DCT under certain conditions. In this work, our objective was to study the expression and subcellular localization of NRBP1 and TSC22D proteins in the kidney and to investigate their effect on the WNK4-SPAK pathway. We hypothesized that NRBP1 and TSC22D proteins may modulate the activity of the WNK-SPAK pathway. By immunofluorescent staining, we confirmed that TSC22D1.1 and TSC22D2 expression is enriched in the DCT. At baseline, they were localized to the apical membrane and, under conditions in which DCT WNK bodies are observed (mice in low K+ diet or mice with Familial Hyperkalemic Hypertension (FHHt)), they colocalized with WNKs and SPAK in the condensates. NRBP1 was also observed in condensates under these conditions. Increased TSC22D1.1 and TSC22D2 abundance was observed in the FHHt mouse model by immunoblot. In HEK293 transfected cells, the co-expression of NRBP1 and TSC22D1.1, TSC22D2, or TSC22D4 with WNK4 and SPAK increased pSPAK levels, as well as WNK4 T-loop phosphorylation. We then investigated the interactions between NRBP1, TSC22D2, and WNK4. The SPAK-WNK interaction is well characterized and involves RFXV motifs present in WNK kinases and a Conserved C-Terminal (CCT) domains present in SPAK. Long TSC22D proteins possess a conserved RFXV motif, NRBP1 possesses a conserved CCT domain, and WNK kinases possess two CCT domains in addition to the RFXV motifs. Thus, we tested the interaction of WNK4 with TSC22D2 and NRBP1 and observed that mutations on the CCT domain of WNK4 dramatically decreased WNK4’s activity and TSC22D2 binding. Interaction with NRBP1 was not affected. However, the overexpression of TSC22D2 and NRBP1 restored WNK4-CCT mutant activity. In conclusion, TSC22D1.1 and TSC22D2 proteins are enriched in the DCT and, under certain conditions that are known to activate the WNK pathway, colocalize with WNK4, SPAK, and NRPB1 in WNK bodies, demonstrating in vivo interaction of these proteins and suggesting a role in DCT physiology. Increased WNK activation is observed in the presence of NRBP1 and TSC22D1.1/TSC22D2. CONACyT, Mexico. IMIN, Mexico. This is the full abstract presented at the American Physiology Summit 2024 meeting and is only available in HTML format. There are no additional versions or additional content available for this abstract. Physiology was not involved in the peer review process.

O-GlcNAcylation regulates angiotensin II-dependent induction of thrombospondin 1

O-Linked attachment of β-N-acetylglucosamine (O-GlcNAc) on serine and threonine residues of nuclear, cytoplasmic, and mitochondrial proteins is a highly dynamic and ubiquitous post-translational modification that plays an important role in altering the function, activity, subcellular localization, and stability of target proteins. It has been proposed that physiologically, O-GlcNAcylation serves primarily to modulate cellular signaling and transcription regulatory pathways in response to nutrients and stress. However, in the vascular physiology literature, O-GlcNAc has predominantly been shown to be detrimental. Our preliminary data has shown that O-GlcNAcylation can cause aortic stiffness, however, the mechanisms underlying this phenotype remain to be elucidated. The predominant mechanisms by which O-GlcNAcylation contributes to alterations in proteostasis include competition with other post-translational modifications as well as regulation of gene expression. Thrombospondin-1 (TSP1) is a multifunctional matrix glycoprotein that impacts extracellular function (ECM) and cell adhesion, migration, proliferation, and cell-cell interaction. Therefore, we hypothesized that TSP1 could be an important downstream mediator of O-GlcNAcylation in aortic vascular smooth muscle cells (VSMCs). To test this hypothesis, primary VSMCs from the thoracic aorta were isolated from 12-week-old female Wistar rats and submitted to angiotensin (ANG) II treatment (100 nM, 3h). Some cells were also treated in the presence of ANG II type 1 (AT1)-receptor antagonist losartan (10−5 M, 30 min). ANG II was chosen because it is a physiological stressor known to cause arterial stiffening. O-GlcNAcylation (RL2 antibody), O-GlcNAcase (OGA), and TSP1 protein amounts were analyzed by Western blot, and the results were expressed in arbitrary units (AU) and mean ±standard error mean. Unpaired Student's test T was used and p<0.05 was considered statistically significant. Genetic and pharmacological inhibition of O-GlcNAc was achieved through OGT silencing and inhibition with a shRNA (FunGene6) and OSMI-4 (10−5 M, 30 min), respectively. As hypothesized, ANG II significantly increased O-GlcNAc in aortic VSMC (0.3±0.03 vs 0.5±0.03 AU n=8, p<0.05) which was inhibited by losartan (0.1±0.02). Moreover, ANG II significantly decreased OGA expression (0.16±0.04 vs 1.16±0.19 AU n=4, p<0.05), the enzyme responsible for removing O-GlcNAc motifs, supporting the observed increase in O-GlcNAc expression. Importantly, ANG II also increased TSP1 protein expression (0.2±0.04 vs 2.0±0.4 n=4, p<0.05), however, shRNA (0.7±0.05) for OGT and OSMI-4 (0.06±0.08) prevented this increase. Overall, these data indicate that Ang II-AT1R signaling can cause O-GlcNAcylation and that TSP1 could be an important downstream mediator of O-GlcNAc-mediated aortic stiffness. National Institutes of Health (R00HL151889, P20GM109091-Pilot Project, and P20GM103641-Pilot Project). University of South Carolina Offce of Research ASPIRE Award (180950-23-64136). American Heart Association Postdoctoral Fellow (916031). This is the full abstract presented at the American Physiology Summit 2024 meeting and is only available in HTML format. There are no additional versions or additional content available for this abstract. Physiology was not involved in the peer review process.

Identification and functional analysis of a rare variant of gene DHX37 in a patient with 46,XY disorders of sex development.

46,XY sex reversal 11 (SRXY11) [OMIM#273250] is characterized by genital ambiguity that may range from mild male genital defects to gonadal sex reversal in severe cases. DHX37 is an RNA helicase that has recently been reported as a cause of SRXY11. So far, a total of 21 variants in DHX37 have been reported in 58 cases with 46,XY disorders of sex development (DSD). Whole exome sequencing (WES) was conducted to screen for variations in patients with 46,XY DSD. The subcellular localization of mutant DHX37 proteins was detected by immunofluorescence. And the levels of mutant DHX37 proteins were detected via Western blotting. A novel pathogenic variant of DHX37 was identified in a patient with 46,XY DSD c.2012G > C (p.Arg671Thr). Bioinformatics analysis showed that the protein function of the variant was impaired. Compared with the structure of the wild-type DHX37 protein, the number of hydrogen bonds and interacting amino acids of the variant protein were changed to varying degrees. Invitro assays revealed that the variant had no significant effect on the intracellular localization of the protein but significantly reduced the expression level of the protein. Our finding further expands the spectrum of the DHX37 variant and could assist in the molecular diagnosis of 46,XY DSD patients.

An Immunolabeling Technique to Visualize the Subcellular Localization of Proteins

An Immunolabeling Technique to Visualize the Subcellular Localization of Proteins

An Immunolabeling Technique to Visualize the Subcellular Localization of Proteins

An Immunolabeling Technique to Visualize the Subcellular Localization of Proteins

Genome-wide analysis of bZIP transcription factors and their expression patterns in response to methyl jasmonate and low-temperature stresses in Platycodon grandiflorus.

Platycodon grandiflorus belongs to the genus Platycodon and has many pharmacological effects, such as expectorant, antitussive, and anti-tumor properties. Among transcription factor families peculiar to eukaryotes, the basic leucine zipper (bZIP) family is one of the most important, which exists widely in plants and participates in many biological processes, such as plant growth, development, and stress responses. However, genomic analysis of the bZIP gene family and related stress response genes has not yet been reported in P. grandiflorus. P. grandiflorus bZIP (PgbZIP) genes were first identified here, and the phylogenetic relationships and conserved motifs in the PgbZIPs were also performed. Meanwhile, gene structures, conserved domains, and the possible protein subcellular localizations of these PgbZIPs were characterized. Most importantly, the cis-regulatory elements and expression patterns of selected genes exposed to two different stresses were analyzed to provide further information on PgbZIPs potential biological roles in P. grandiflorus upon exposure to environmental stresses. Forty-six PgbZIPs were identified in P. grandiflorus and divided into nine groups, as displayed in the phylogenetic tree. The results of the chromosomal location and the collinearity analysis showed that forty-six PgbZIP genes were distributed on eight chromosomes, with one tandem duplication event and eleven segmental duplication events identified. Most PgbZIPs in the same phylogenetic group have similar conserved motifs, domains, and gene structures. There are cis-regulatory elements related to the methyl jasmonate (MeJA) response, low-temperature response, abscisic acid response, auxin response, and gibberellin response. Ten PgbZIP genes were selected to study their expression patterns upon exposure to low-temperature and MeJA treatments, and all ten genes responded to these stresses. The real-time quantitative polymerase chain reaction (RT-qPCR) results suggest that the expression levels of most PgbZIPs decreased significantly within 6 h and then gradually increased to normal or above normal levels over the 90 h following MeJA treatment. The expression levels of all PgbZIPs were significantly reduced after 3 h of the low-temperature treatment. These results reveal the characteristics of the PgbZIP family genes and provide valuable information for improving P. grandiflorus's ability to cope with environmental stresses during growth and development.

SUMOylation of GMFB regulates its stability and function in retinal pigment epithelial cells under hyperglycemia

BackgroundGlia maturation factor beta (GMFB) is a growth and differentiation factor that acts as an intracellular regulator of signal transduction pathways. The small ubiquitin-related modifier (SUMO) modification, SUMOylation, is a posttranslational modification (PTM) that plays a key role in protein subcellular localization, stability, transcription, and enzymatic activity. Recent studies have highlighted the importance of SUMOylation in the inflammation and progression of numerous diseases. However, the relationship between GMFB and SUMOylation is unclear. ResultsHere, we report for the first time that GMFB and SUMO1 are markedly increased in retinal pigment epithelial (RPE) cells at the early stage of diabetes mellitus (DM) under hyperglycemia. The GMFΒ protein could be mono-SUMOylated by SUMO1 at the K20, K35, K58 or K97 sites. SUMOylation of GMFB led to its increased protein stability and subcellular translocation. Furthermore, deSUMOylation of GMFΒ downregulates multiple signaling pathways, including the Jak-STAT signaling pathway, p38 pathway and NF-kappa B signaling pathway. ConclusionsThis work provides novel insight into the role of SUMOylated GMFB in RPE cells and provides a novel therapeutic target for diabetic retinopathy (DR).

Pooled multicolour tagging for visualizing subcellular protein dynamics

Imaging-based methods are widely used for studying the subcellular localization of proteins in living cells. While routine for individual proteins, global monitoring of protein dynamics following perturbation typically relies on arrayed panels of fluorescently tagged cell lines, limiting throughput and scalability. Here, we describe a strategy that combines high-throughput microscopy, computer vision and machine learning to detect perturbation-induced changes in multicolour tagged visual proteomics cell (vpCell) pools. We use genome-wide and cancer-focused intron-targeting sgRNA libraries to generate vpCell pools and a large, arrayed collection of clones each expressing two different endogenously tagged fluorescent proteins. Individual clones can be identified in vpCell pools by image analysis using the localization patterns and expression level of the tagged proteins as visual barcodes, enabling simultaneous live-cell monitoring of large sets of proteins. To demonstrate broad applicability and scale, we test the effects of antiproliferative compounds on a pool with cancer-related proteins, on which we identify widespread protein localization changes and new inhibitors of the nuclear import/export machinery. The time-resolved characterization of changes in subcellular localization and abundance of proteins upon perturbation in a pooled format highlights the power of the vpCell approach for drug discovery and mechanism-of-action studies.

Abstract P70: HIGH-THROUGHPUT IMAGING-BASED PHENOTYPIC PROFILING OF IMMUNE CELLS FROM BRCA2 VARIANT CARRIERS AND NON-CARRIERS

Abstract Mutated germline alleles in the DNA damage repair (DDR) genes BRCA1/2 are major risk factors for breast and ovarian cancers. Recent studies have found that BRCA1-mutant tumors tend to have higher abundance of immune cells within the tumor microenvironment, but much better response was observed from the BRCA2-mutant tumors when given immune-checkpoint inhibitors, suggesting possible functional and phenotypic effects of germline BRCA2 mutations on the immune cells. However, most existing studies focus on investigating the genomic instability and phenotypes of hereditary mutations in breast and ovarian cancer cells. Thus, the impact of these mutations on immune cells’ phenotypes and functions remains largely unknown. Here, we report a study of immune cells with and without BRCA2 mutations using High-throughput Imaging-based Phenotypic Profiling (HIPP), which can identify changes in cellular morphology and protein subcellular localization patterns that are not obvious from other traditional functional assays. We performed HIPP on immune cells from nine BRCA2 pathogenic variant carriers and eight non-carriers. Automatic quantification of ~200 phenotypic features from each single cell revealed that CD8+ T cells from BRCA2 variant carriers have distinct and quantifiable changes in their morphological features (such as cytoskeleton structures) and immunological features (such as Granzyme B expressions). Moreover, we found that BRCA2-mutant CD8+ T cells proliferate faster than normal CD8+ T cells. Our results suggest that BRCA2 mutations affect the functions and phenotypes of immune cells. Some of these phenotypic features and markers may be used to classify BRCA2 variants and predict the pathogenicity of variants of uncertain significance (VUS). Our study provides new insight into the immune phenotyping of cancer patients at individual level and may help to stratify patients, leading to the development of more effective therapeutic strategies. This method can potentially extend to other germline variants and overcome the challenge of VUS interpretation in clinical management. Citation Format: Guorui Zhong, Oscar Fu, Claresta Yeo, Siao Ting Chong, Sock Hoai Chan, Joanne Ngeow, Lit-Hsin Loo. HIGH-THROUGHPUT IMAGING-BASED PHENOTYPIC PROFILING OF IMMUNE CELLS FROM BRCA2 VARIANT CARRIERS AND NON-CARRIERS [abstract]. In: Proceedings of Frontiers in Cancer Science; 2023 Nov 6-8; Singapore. Philadelphia (PA): AACR; Cancer Res 2024;84(8_Suppl):Abstract nr P70.

Vislocas: Vision transformers for identifying protein subcellular mis-localization signatures of different cancer subtypes from immunohistochemistry images

Proteins must be sorted to specific subcellular compartments to perform their functions. Abnormal protein subcellular localizations are related to many diseases. Although many efforts have been made in predicting protein subcellular localization from various static information, including sequences, structures and interactions, such static information cannot predict protein mis-localization events in diseases. On the contrary, the IHC (immunohistochemistry) images, which have been widely applied in clinical diagnosis, contains information that can be used to find protein mis-localization events in disease states. In this study, we create the Vislocas method, which is capable of finding mis-localized proteins from IHC images as markers of cancer subtypes. By combining CNNs and vision transformer encoders, Vislocas can automatically extract image features at both global and local level. Vislocas can be trained with full-sized IHC images from scratch. It is the first attempt to create an end-to-end IHC image-based protein subcellular location predictor. Vislocas achieved comparable or better performances than state-of-the-art methods. We applied Vislocas to find significant protein mis-localization events in different subtypes of glioma, melanoma and skin cancer. The mis-localized proteins, which were found purely from IHC images by Vislocas, are in consistency with clinical or experimental results in literatures. All codes of Vislocas have been deposited in a Github repository (https://github.com/JingwenWen99/Vislocas). All datasets of Vislocas have been deposited in Zenodo (https://zenodo.org/records/10632698).

DeepLoc 2.1: multi-label membrane protein type prediction using protein language models.

DeepLoc 2.0 is a popular web server for the prediction of protein subcellular localization and sorting signals. Here, we introduce DeepLoc 2.1, which additionally classifies the input proteins into the membrane protein types Transmembrane, Peripheral, Lipid-anchoredand Soluble. Leveraging pre-trained transformer-based protein language models, the server utilizes a three-stage architecture for sequence-based, multi-label predictions. Comparative evaluations with other established tools on a test set of 4933 eukaryotic protein sequences, constructed following stringent homology partitioning, demonstrate state-of-the-art performance. Notably, DeepLoc 2.1 outperforms existing models, with the larger ProtT5 model exhibiting a marginal advantage over the ESM-1B model. The web server is available at https://services.healthtech.dtu.dk/services/DeepLoc-2.1.

Characterization and engineering of peroxisome targeting sequences for compartmentalization engineering in Pichia pastoris.

Peroxisomal compartmentalization has emerged as a highly promising strategy for reconstituting intricate metabolic pathways. In recent years, significant progress has been made in the peroxisomes through harnessing precursor pools, circumventing metabolic crosstalk, and minimizing the cytotoxicity of exogenous pathways. However, it is important to note that in methylotrophic yeasts (e.g. Pichia pastoris), the abundance and protein composition of peroxisomes are highly variable, particularly when peroxisome proliferation is induced by specific carbon sources. The intricate subcellular localization of native proteins, the variability of peroxisomal metabolic pathways, and the lack of systematic characterization of peroxisome targeting signals have limited the applications of peroxisomal compartmentalization in P. pastoris. Accordingly, this study established a high-throughput screening method based on β-carotene biosynthetic pathway to evaluate the targeting efficiency of PTS1s (Peroxisome Targeting Signal Type 1) in P. pastoris. First, 25 putative endogenous PTS1s were characterized and 3 PTS1s with high targeting efficiency were identified. Then, directed evolution of PTS1s was performed by constructing two PTS1 mutant libraries, and a total of 51 PTS1s (29 classical and 22 noncanonical PTS1s) with presumably higher peroxisomal targeting efficiency were identified, part of which were further characterized via confocal microscope. Finally, the newly identified PTS1s were employed for peroxisomal compartmentalization of the geraniol biosynthetic pathway, resulting in more than 30% increase in the titer of monoterpene compared with when the pathway was localized to the cytosol. The present study expands the synthetic biology toolkit and lays a solid foundation for peroxisomal compartmentalization in P. pastoris.

Abnormal protein SUMOylation in liver disease: novel target for therapy.

SUMOylation is an important protein post-translational modification (PTM) process, in which the small ubiquitin-like modifier (SUMO) protein covalently binds to the target protein and regulates stability, subcellular localization, and protein-protein interaction of the target protein. Protein SUMOylation exerts crucial regulatory function in the liver, and its abnormalities are associated with various liver-related disease processes. This review focuses on the biological functions of protein SUMOylation in liver-related diseases in recent years, summarizes the molecular mechanisms of SUMOylation in the replication of hepatitis viruses and the occurrence of hepatocellular carcinoma, and discusses the significance of SUMOylation in liver-related disorders, which is essential for understanding liver biological processes and formulating therapeutic strategies.

Production of ultra-high-molecular-weight poly-γ-glutamic acid by a newly isolated Bacillus subtilis strain and genomic and transcriptomic analyses.

Poly-γ-glutamic acid (γ-PGA) is a microbial-derived polymer with molecular weight (Mw) from 104 to 107 Da, and the high-Mw(>7.0 × 105 Da) or ultra-high-Mw(>5.0 × 106 Da) γ-PGA has important application value as a tissue engineering material, as a flocculant, and as a heavy metal remover. Therefore, how to produce these high-Mw γ-PGAs with low cost and high efficiency has attracted wide attention. In this study, a γ-PGA producer was isolated from the natural environment, and identified and named Bacillus subtilis GXD-20. Then, the ultra-high-Mw(>6.0 × 106 Da) γ-PGA produced by GXD-20 was characterized. Interestingly, GXD-20 could produce γ-PGA at 42°C, and exhibited a γ-PGA titer of up to 22.29±0.59 g L-1 in a 5-L fermenter after optimization of the fermentation process. Comparative genomic analysis indicated that the specific protein sequence and subcellular localization of PgdS (a γ-PGA-degrading enzyme) were closely related to the ultra-high-Mw of γ-PGA. Transcriptomic analysis revealed that the high γ-PGA titer at 42°C was mainly related to the high expression of genes encoding enzymes for sucrose transportation and utilization, nitrogen transportation, endogenous glutamate synthesis, and γ-PGA synthesis. These results provide new insights into the production of ultra-high-Mw γ-PGA by Bacillus at high temperatures.

A Review for Artificial Intelligence Based Protein Subcellular Localization.

Proteins need to be located in appropriate spatiotemporal contexts to carry out their diverse biological functions. Mislocalized proteins may lead to a broad range of diseases, such as cancer and Alzheimer's disease. Knowing where a target protein resides within a cell will give insights into tailored drug design for a disease. As the gold validation standard, the conventional wet lab uses fluorescent microscopy imaging, immunoelectron microscopy, and fluorescent biomarker tags for protein subcellular location identification. However, the booming era of proteomics and high-throughput sequencing generates tons of newly discovered proteins, making protein subcellular localization by wet-lab experiments a mission impossible. To tackle this concern, in the past decades, artificial intelligence (AI) and machine learning (ML), especially deep learning methods, have made significant progress in this research area. In this article, we review the latest advances in AI-based method development in three typical types of approaches, including sequence-based, knowledge-based, and image-based methods. We also elaborately discuss existing challenges and future directions in AI-based method development in this research field.

Protein 3D Graph Structure Learning for Robust Structure-Based Protein Property Prediction

Protein structure-based property prediction has emerged as a promising approach for various biological tasks, such as protein function prediction and sub-cellular location estimation. The existing methods highly rely on experimental protein structure data and fail in scenarios where these data are unavailable. Predicted protein structures from AI tools (e.g., AlphaFold2) were utilized as alternatives. However, we observed that current practices, which simply employ accurately predicted structures during inference, suffer from notable degradation in prediction accuracy. While similar phenomena have been extensively studied in general fields (e.g., Computer Vision) as model robustness, their impact on protein property prediction remains unexplored. In this paper, we first investigate the reason behind the performance decrease when utilizing predicted structures, attributing it to the structure embedding bias from the perspective of structure representation learning. To study this problem, we identify a Protein 3D Graph Structure Learning Problem for Robust Protein Property Prediction (PGSL-RP3), collect benchmark datasets, and present a protein Structure embedding Alignment Optimization framework (SAO) to mitigate the problem of structure embedding bias between the predicted and experimental protein structures. Extensive experiments have shown that our framework is model-agnostic and effective in improving the property prediction of both predicted structures and experimental structures.