Computational prediction of secreted proteins in gram-negative bacteria

T4SEpp: A pipeline integrating protein language models to predict bacterial type IV secreted effectors

Many pathogenic bacteria use type IV secretion systems (T4SSs) to deliver effectors (T4SEs) into the cytoplasm of eukaryotic cells, causing diseases. The identification of effectors is a crucial step in understanding the mechanisms of bacterial pathogenicity, but this remains a major challenge. In this study, we used the full-length embedding features generated by six pre-trained protein language models to train classifiers predicting T4SEs and compared their performance. We integrated three modules into a model called T4SEpp. The first module searched for full-length homologs of known T4SEs, signal sequences, and effector domains; the second module fine-tuned a machine learning model using data for a signal sequence feature; and the third module used the three best-performing pre-trained protein language models. T4SEpp outperformed other state-of-the-art (SOTA) software tools, achieving ∼0.98 accuracy at a high specificity of ∼0.99, based on the assessment of an independent validation dataset. T4SEpp predicted 13 T4SEs from Helicobacter pylori, including the well-known CagA and 12 other potential ones, among which eleven could potentially interact with human proteins. This suggests that these potential T4SEs may be associated with the pathogenicity of H. pylori. Overall, T4SEpp provides a better solution to assist in the identification of bacterial T4SEs and facilitates studies of bacterial pathogenicity. T4SEpp is freely accessible at https://bis.zju.edu.cn/T4SEpp.

Distribution patterns and influential factors of pathogenic bacteria in freshwater aquaculture sediments

Pathogenic bacteria, the major causative agents of aquaculture diseases, are a serious impediment to the aquaculture industry. However, the bioinformatics of pathogenic bacteria and virulence factors (VFs) in sediments, an important component of freshwater aquaculture ecosystems, are not well characterized. In this study, 20 sediment samples were collected from fish pond sediments (FPS), shrimp field sediments (SFS), fish pond sediment control (FPSC), and shrimp field sediment control (SFSC). Molecular biological information was obtained on a total of 173 pathogenic bacteria, 1093 virulence factors (VFs), and 8475 mobile genetic elements (MGEs) from these samples. The results indicated that (1) aquaculture patterns and sediment characteristics can affect the distribution of pathogenic bacteria. According to the results of the Kruskal-Wallis H test, except for Mycobacterium gilvum, there were significant differences (P < 0.05) among the four sediment types in the average abundance of major pathogenic bacteria (top 30 in abundance), and the average abundance of major pathogenic bacteria in the four sediment types followed the following pattern: FPS > SFS > FPSC > SFSC. (2) Pathogenic bacteria are able to implement a variety of complex pathogenic mechanisms such as adhesion, invasion, immune evasion, and metabolic regulation in the host because they carry a variety of VFs such as type IV pili, HSI-I, Alginate, Colibactin, and Capsule. According to the primary classification of the Virulence Factor Database (VFDB), the abundance of VFs in all four types of sediments showed the following pattern: offensive VFs > non-specific VFs > defensive VFs > regulation of virulence-related genes. (3) Total organic carbon (TOC), total phosphorus (TP), available phosphorus (AP), nitrite, and nitrate were mostly only weakly positively correlated with the major pathogenic bacteria and could promote the growth of pathogenic bacteria to some extent, whereas ammonia was significantly positively correlated with most of the major pathogenic bacteria and could play an important role in promoting the growth and reproduction of pathogenic bacteria. (4) Meanwhile, there was also a significant positive correlation between CAZyme genes and major pathogenic bacteria (0.62 ≤ R ≤ 0.89, P < 0.05). This suggests that these pathogenic bacteria could be the main carriers of CAZyme genes and, to some extent, gained a higher level of metabolic activity by degrading organic matter in the sediments to maintain their competitive advantage. (5) Worryingly, the results of correlation analyses indicated that MGEs in aquaculture sediments could play an important role in the spread of VFs (R = 0.82, P < 0.01), and in particular, plasmids (R = 0.75, P < 0.01) and integrative and conjugative elements (ICEs, R = 0.65, P < 0.05) could be these major vectors of VFs. The results of this study contribute to a comprehensive understanding of the health of freshwater aquaculture sediments and provide a scientific basis for aquaculture management and conservation.

A computational and secretome analysis approach reveals exclusive and shared candidate type six secretion system substrates in Pectobacterium brasiliense 1692

The type 6 secretion system (T6SS) of Gram-negative bacteria (GNB) has implications for bacterial competition, virulence, and survival. For the broad host range pathogen, Pectobacterium brasiliense 1692, T6SS-mediated competition occurs in a tissue-specific manner. However, no other roles have been investigated. The aim of this study was to identify T6SS-associated proteins under virulence inducing conditions. We used Bastion tools to predict 1479 Pbr1692 secreted proteins. Sixteen percent of these overlap between type 1-4 secretion systems (T1SS-T4SS) and T6SS. Using label-free quantitative mass spectrometry of Pbr1692 T6SS active and T6SS inactive strains' secretomes cultured in minimal media supplemented with host extract, 49 T6SS-associated proteins with varied gene ontology predicted functions were identified. We report 19 and 30 T6SS primary substrates and differentially secreted proteins, respectively, in T6SS mutants versus wild type strains. Of the total 49 T6SS-associated proteins presented in this study, 25 were also predicted using the BastionX platform as T6SS exclusive and shared substrates with T1SS-T4SS. This work provides a list of Pbr1692 T6SS secreted effector candidates. These include a potential antibacterial toxin HNH endonuclease and several predicted virulence proteins, including plant cell wall degrading enzymes. A preliminary basis for potential crosstalk between GNB secretion systems is also highlighted.

Protein Sorting Prediction

Many computational methods are available for predicting protein sorting in bacteria. When comparing them, it is important to know that they can be grouped into three fundamentally different approaches: signal-based, global property-based, and homology-based prediction. In this chapter, the strengths and drawbacks of each of these approaches are described through many examples of methods that predict secretion, integration into membranes, or subcellular locations in general. The aim of this chapter is to provide a user-level introduction to the field with a minimum of computational theory.

De novo identification of bacterial antigens of a clinical isolate by combining use of proteosurfaceomics, secretomics, and BacScan technologies

Background

Emerging infectious diseases pose a significant threat to both human and animal populations. Rapid de novo identification of protective antigens from a clinical isolate and development of an antigen-matched vaccine is a golden strategy to prevent the spread of emerging novel pathogens.

Methods

Here, we focused on Actinobacillus pleuropneumoniae, which poses a serious threat to the pig industry, and developed a general workflow by integrating proteosurfaceomics, secretomics, and BacScan technologies for the rapid de novo identification of bacterial protective proteins from a clinical isolate.

Results

As a proof of concept, we identified 3 novel protective proteins of A. pleuropneumoniae. Using the protective protein HBS1_14 and toxin proteins, we have developed a promising multivalent subunit vaccine against A. pleuropneumoniae.

Discussion

We believe that our strategy can be applied to any bacterial pathogen and has the potential to significantly accelerate the development of antigen-matched vaccines to prevent the spread of an emerging novel bacterial pathogen.

Features and algorithms: facilitating investigation of secreted effectors in Gram-negative bacteria

Gram-negative bacteria deliver effector proteins through type III, IV, or VI secretion systems (T3SSs, T4SSs, and T6SSs) into host cells, causing infections and diseases. In general, effector proteins for each of these distinct secretion systems lack homology and are difficult to identify. Sequence analysis has disclosed many common features, helping us to understand the evolution, function, and secretion mechanisms of the effectors. In combination with various algorithms, the known common features have facilitated accurate prediction of new effectors. Ensemblers or integrated pipelines achieve a better prediction of performance, which combines multiple computational models or modules with multidimensional features. Natural language processing (NLP) models also show the merits, which could enable discovery of novel features and, in turn, facilitate more precise effector prediction, extending our knowledge about each secretion mechanism.

DeepSecE: A Deep-Learning-Based Framework for Multiclass Prediction of Secreted Proteins in Gram-Negative Bacteria

Proteins secreted by Gram-negative bacteria are tightly linked to the virulence and adaptability of these microbes to environmental changes. Accurate identification of such secreted proteins can facilitate the investigations of infections and diseases caused by these bacterial pathogens. However, current bioinformatic methods for predicting bacterial secreted substrate proteins have limited computational efficiency and application scope on a genome-wide scale. Here, we propose a novel deep-learning-based framework-DeepSecE-for the simultaneous inference of multiple distinct groups of secreted proteins produced by Gram-negative bacteria. DeepSecE remarkably improves their classification from nonsecreted proteins using a pretrained protein language model and transformer, achieving a macro-average accuracy of 0.883 on 5-fold cross-validation. Performance benchmarking suggests that DeepSecE achieves competitive performance with the state-of-the-art binary predictors specialized for individual types of secreted substrates. The attention mechanism corroborates salient patterns and motifs at the N or C termini of the protein sequences. Using this pipeline, we further investigate the genome-wide prediction of novel secreted proteins and their taxonomic distribution across ~1,000 Gram-negative bacterial genomes. The present analysis demonstrates that DeepSecE has major potential for the discovery of disease-associated secreted proteins in a diverse range of Gram-negative bacteria. An online web server of DeepSecE is also publicly available to predict and explore various secreted substrate proteins via the input of bacterial genome sequences.

Molecular epidemiology and clinical characteristics of the type VI secretion system in Klebsiella pneumoniae causing abscesses

Purpose

The type VI system (T6SS) has the potential to be a new virulence factor for hypervirulent Klebsiella pneumoniae (hvKp) strains. This study aimed to characterize the molecular and clinical features of T6SS-positive and T6SS-negative K. pneumoniae isolates that cause abscesses.

Patients and methods

A total of 169 non-duplicate K. pneumoniae strains were isolated from patients with abscesses in a tertiary hospital in China from January 2018 to June 2022, and clinical data were collected. For all isolates, capsular serotypes, T6SS genes, virulence, and drug resistance genes, antimicrobial susceptibility testing, and biofilm formation assays were assessed. Multilocus sequence typing was used to analyze the genotypes of hvKp. T6SS-positive hvKp, T6SS-negative hvKp, T6SS-positive cKP, and T6SS-negative cKP (n = 4 strains for each group) were chosen for the in vivo Galleria mellonella infection model and in vitro competition experiments to further explore the microbiological characteristics of T6SS-positive K. pneumoniae isolates.

Results

The positive detection rate for T6SS was 36.1%. The rates of hvKp, seven virulence genes, K1 capsular serotype, and ST23 in T6SS-positive strains were all higher than those in T6SS-negative strains (p < 0.05). Multivariate logistic regression analysis indicated that the carriage of aerobactin (OR 0.01) and wcaG (OR 33.53) were independent risk factors for T6SS-positive strains (p < 0.05). The T6SS-positive strains had a stronger biofilm-forming ability than T6SS-negative strains (p < 0.05). The T6SS-positive and T6SS-negative strains showed no significant differences in competitive ability (p = 0.06). In the in vivo G. mellonella infection model, the T6SS(+)/hvKP group had the worst prognosis. Except for cefazolin and tegacyclin, T6SS-positive isolates displayed a lower rate of antimicrobial resistance to other drugs (p < 0.05). The T6SS-positive isolates were more likely to be acquired from community infections (p < 0.05).

Conclusion

Klebsiella pneumoniae isolates causing abscesses have a high prevalence of T6SS genes. T6SS-positive K. pneumoniae isolates are associated with virulence, and the T6SS genes may be involved in the hvKp virulence mechanism.

Herbivorous Fish Microbiome Adaptations to Sulfated Dietary Polysaccharides

Marine herbivorous fish that feed primarily on macroalgae, such as those from the genus Kyphosus, are essential for maintaining coral health and abundance on tropical reefs. Here, deep metagenomic sequencing and assembly of gut compartment-specific samples from three sympatric, macroalgivorous Hawaiian kyphosid species have been used to connect host gut microbial taxa with predicted protein functional capacities likely to contribute to efficient macroalgal digestion. Bacterial community compositions, algal dietary sources, and predicted enzyme functionalities were analyzed in parallel for 16 metagenomes spanning the mid- and hindgut digestive regions of wild-caught fishes. Gene colocalization patterns of expanded carbohydrate (CAZy) and sulfatase (SulfAtlas) digestive enzyme families on assembled contigs were used to identify likely polysaccharide utilization locus associations and to visualize potential cooperative networks of extracellularly exported proteins targeting complex sulfated polysaccharides. These insights into the gut microbiota of herbivorous marine fish and their functional capabilities improve our understanding of the enzymes and microorganisms involved in digesting complex macroalgal sulfated polysaccharides. IMPORTANCE This work connects specific uncultured bacterial taxa with distinct polysaccharide digestion capabilities lacking in their marine vertebrate hosts, providing fresh insights into poorly understood processes for deconstructing complex sulfated polysaccharides and potential evolutionary mechanisms for microbial acquisition of expanded macroalgal utilization gene functions. Several thousand new marine-specific candidate enzyme sequences for polysaccharide utilization have been identified. These data provide foundational resources for future investigations into suppression of coral reef macroalgal overgrowth, fish host physiology, the use of macroalgal feedstocks in terrestrial and aquaculture animal feeds, and the bioconversion of macroalgae biomass into value-added commercial fuel and chemical products.

Bacterial secretion system functions: evidence of interactions and downstream implications

Unprecedented insights into the biology and functions of bacteria have been and continue to be gained through studying bacterial secretion systems in isolation. This method, however, results in our understanding of the systems being primarily based on the idea that they operate independently, ignoring the subtleties of downstream interconnections. Gram-negative bacteria are naturally able to adapt to and navigate their frequently varied and dynamic surroundings, mostly because of the covert connections between secretion systems. Therefore, to comprehend some of the linked downstream repercussions for organisms that follow this discourse, it is vital to have mechanistic insights into how the intersecretion system functions in bacterial rivalry, virulence, and survival, among other things. To that purpose, this paper discusses a few key instances of molecular antagonistic and interdependent relationships between bacterial secretion systems and their produced functional products.

Redefining the bacterial Type I protein secretion system

Type I secretion systems (T1SS) are versatile molecular machines for protein transport across the Gram-negative cell envelope. The archetypal Type I system mediates secretion of the Escherichia coli hemolysin, HlyA. This system has remained the pre-eminent model of T1SS research since its discovery. The classic description of a T1SS is composed of three proteins: an inner membrane ABC transporter, a periplasmic adaptor protein and an outer membrane factor. According to this model, these components assemble to form a continuous channel across the cell envelope, an unfolded substrate molecule is then transported in a one-step mechanism, directly from the cytosol to the extracellular milieu. However, this model does not encapsulate the diversity of T1SS that have been characterized to date. In this review, we provide an updated definition of a T1SS, and propose the subdivision of this system into five subgroups. These subgroups are categorized as T1SSa for RTX proteins, T1SSb for non-RTX Ca²⁺-binding proteins, T1SSc for non-RTX proteins, T1SSd for class II microcins, and T1SSe for lipoprotein secretion. Although often overlooked in the literature, these alternative mechanisms of Type I protein secretion offer many avenues for biotechnological discovery and application.

Recruitment of heterologous substrates by bacterial secretion systems for transkingdom translocation

Bacterial secretion systems mediate the selective exchange of macromolecules between bacteria and their environment, playing a pivotal role in processes such as horizontal gene transfer or virulence. Among the different families of secretion systems, Type III, IV and VI (T3SS, T4SS and T6SS) share the ability to inject their substrates into human cells, opening up the possibility of using them as customized injectors. For this to happen, it is necessary to understand how substrates are recruited and to be able to engineer secretion signals, so that the transmembrane machineries can recognize and translocate the desired substrates in place of their own. Other factors, such as recruiting proteins, chaperones, and the degree of unfolding required to cross through the secretion channel, may also affect transport. Advances in the knowledge of the secretion mechanism have allowed heterologous substrate engineering to accomplish translocation by T3SS, and to a lesser extent, T4SS and T6SS into human cells. In the case of T4SS, transport of nucleoprotein complexes adds a bonus to its biotechnological potential. Here, we review the current knowledge on substrate recognition by these secretion systems, the many examples of heterologous substrate translocation by engineering of secretion signals, and the current and future biotechnological and biomedical applications derived from this approach.

A serralysin-like protein of Candidatus Liberibacter asiaticus modulates components of the bacterial extracellular matrix

Huanglongbing (HLB), the current major threat for Citrus species, is caused by intracellular alphaproteobacteria of the genus Candidatus Liberibacter (CaL), with CaL asiaticus (CLas) being the most prevalent species. This bacterium inhabits phloem cells and is transmitted by the psyllid Diaphorina citri. A gene encoding a putative serralysin-like metalloprotease (CLIBASIA_01345) was identified in the CLas genome. The expression levels of this gene were found to be higher in citrus leaves than in psyllids, suggesting a function for this protease in adaptation to the plant environment. Here, we study the putative role of CLas-serralysin (Las1345) as virulence factor. We first assayed whether Las1345 could be secreted by two different surrogate bacteria, Rhizobium leguminosarum bv. viciae A34 (A34) and Serratia marcescens. The protein was detected only in the cellular fraction of A34 and S. marcescens expressing Las1345, and increased protease activity of those bacteria by 2.55 and 4.25-fold, respectively. In contrast, Las1345 expressed in Nicotiana benthamiana leaves did not show protease activity nor alterations in the cell membrane, suggesting that Las1345 do not function as a protease in the plant cell. Las1345 expression negatively regulated cell motility, exopolysaccharide production, and biofilm formation in Xanthomonas campestris pv. campestris (Xcc). This bacterial phenotype was correlated with reduced growth and survival on leaf surfaces as well as reduced disease symptoms in N. benthamiana and Arabidopsis. These results support a model where Las1345 could modify extracellular components to adapt bacterial shape and appendages to the phloem environment, thus contributing to virulence.

Host-microbiome protein-protein interactions capture disease-relevant pathways

Background

Host-microbe interactions are crucial for normal physiological and immune system development and are implicated in a variety of diseases, including inflammatory bowel disease (IBD), colorectal cancer (CRC), obesity, and type 2 diabetes (T2D). Despite large-scale case-control studies aimed at identifying microbial taxa or genes involved in pathogeneses, the mechanisms linking them to disease have thus far remained elusive.

Results

To identify potential pathways through which human-associated bacteria impact host health, we leverage publicly-available interspecies protein-protein interaction (PPI) data to find clusters of microbiome-derived proteins with high sequence identity to known human-protein interactors. We observe differential targeting of putative human-interacting bacterial genes in nine independent metagenomic studies, finding evidence that the microbiome broadly targets human proteins involved in immune, oncogenic, apoptotic, and endocrine signaling pathways in relation to IBD, CRC, obesity, and T2D diagnoses.

Conclusions

This host-centric analysis provides a mechanistic hypothesis-generating platform and extensively adds human functional annotation to commensal bacterial proteins.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13059-022-02643-9.

T1SEstacker: A Tri-Layer Stacking Model Effectively Predicts Bacterial Type 1 Secreted Proteins Based on C-Terminal Non-repeats-in-Toxin-Motif Sequence Features

Type 1 secretion systems play important roles in pathogenicity of Gram-negative bacteria. However, the substrate secretion mechanism remains largely unknown. In this research, we observed the sequence features of repeats-in-toxin (RTX) proteins, a major class of type 1 secreted effectors (T1SEs). We found striking non-RTX-motif amino acid composition patterns at the C termini, most typically exemplified by the enriched “[FLI][VAI]” at the most C-terminal two positions. Machine-learning models, including deep-learning ones, were trained using these sequence-based non-RTX-motif features and further combined into a tri-layer stacking model, T1SEstacker, which predicted the RTX proteins accurately, with a fivefold cross-validated sensitivity of ∼0.89 at the specificity of ∼0.94. Besides substrates with RTX motifs, T1SEstacker can also well distinguish non-RTX-motif T1SEs, further suggesting their potential existence of common secretion signals. T1SEstacker was applied to predict T1SEs from the genomes of representative Salmonella strains, and we found that both the number and composition of T1SEs varied among strains. The number of T1SEs is estimated to reach 100 or more in each strain, much larger than what we expected. In summary, we made comprehensive sequence analysis on the type 1 secreted RTX proteins, identified common sequence-based features at the C termini, and developed a stacking model that can predict type 1 secreted proteins accurately.

DeepT3 2.0: improving type III secreted effector predictions by an integrative deep learning framework

Type III secretion systems (T3SSs) are bacterial membrane-embedded nanomachines that allow a number of humans, plant and animal pathogens to inject virulence factors directly into the cytoplasm of eukaryotic cells. Export of effectors through T3SSs is critical for motility and virulence of most Gram-negative pathogens. Current computational methods can predict type III secreted effectors (T3SEs) from amino acid sequences, but due to algorithmic constraints, reliable and large-scale prediction of T3SEs in Gram-negative bacteria remains a challenge. Here, we present DeepT3 2.0 (http://advintbioinforlab.com/deept3/), a novel web server that integrates different deep learning models for genome-wide predicting T3SEs from a bacterium of interest. DeepT3 2.0 combines various deep learning architectures including convolutional, recurrent, convolutional-recurrent and multilayer neural networks to learn N-terminal representations of proteins specifically for T3SE prediction. Outcomes from the different models are processed and integrated for discriminating T3SEs and non-T3SEs. Because it leverages diverse models and an integrative deep learning framework, DeepT3 2.0 outperforms existing methods in validation datasets. In addition, the features learned from networks are analyzed and visualized to explain how models make their predictions. We propose DeepT3 2.0 as an integrated and accurate tool for the discovery of T3SEs.