Data mining tools for Salmonella characterization: application to gel-based fingerprinting analysis

Development of a 1-step multiplex PCR assay for the detection of S. Enteritidis, S. Pullorum, S. Typhimurium, and S. Infantis associated with poultry production

Salmonellosis in poultry is detrimental to the advancement of the breeding industry and poses hazards to human health. Approximately 2,600 Salmonella varieties exist, among which S. Enteritidis, S. Pullorum, S. Typhimurium, and S. Infantis are prevalent serotypes in the poultry industry in recent years. They can also infect humans by contaminating poultry eggs and meat. Therefore, identifying these serotypes is crucial for successful preventive and control interventions. The White-Kauffmann-Le Minor scheme is time-consuming and requires expensive reagents. Whole-genome sequencing (WGS) and other molecular biology techniques require skilled technical staff. In comparison, the polymerase chain reaction (PCR) is more accurate, rapid, and inexpensive, thus proving suitable for widespread application in the poultry industry. Here, we selected 4 specific primers: lygD, mdh, ipaJ, and SIN_02055, which correspond to detecting S. Enteritidis, S. Typhimurium, S. Pullorum, and S. Infantis, respectively. They were integrated into a 1-step multiplex PCR method. We optimized the PCR method by utilizing specificity test results to determine the optimal annealing temperature (57°C). The PCR method exhibited excellent sensitivity for genomic DNA and bacterial cultures. We used the developed method to determine 157 clinical Salmonella isolates from various stages of the poultry production chain. The results aligned with serotype data generated via WGS analysis, demonstrating the method's excellent accuracy. In conclusion, this study developed a 1-step multiplex PCR method that simultaneously identifies S. Enteritidis, S. Typhimurium, S. Pullorum, and S. Infantis, allowing routine mass detection in the grass-root poultry industry.

Investigation of the Genes Involved in the Outbreaks of Escherichia coli and Salmonella spp. in the United States

Salmonella spp. and Escherichiacoli (E. coli) are two of the deadliest foodborne pathogens in the US. Genes involved in antimicrobial resistance, virulence, and stress response, enable these pathogens to increase their pathogenicity. This study aims to examine the genes detected in both outbreak and non-outbreak Salmonella spp. and E. coli by analyzing the data from the National Centre for Biotechnology Information (NCBI) Pathogen Detection Isolates Browser database. A multivariate statistical analysis was conducted on the genes detected in isolates of outbreak Salmonella spp., non-outbreak Salmonella spp., outbreak E. coli, and non-outbreak E. coli. The genes from the data were projected onto a two-dimensional space through principal component analysis. Hierarchical clustering was then used to quantify the relationship between the genes in the dataset. Most of the outlier genes identified in E. coli isolates are virulence genes, while outlier genes identified in Salmonella spp. are mainly involved in stress response. Gene epeA, which encodes a high-molecular-weight serine protease autotransporter of Enterobacteriaceae (SPATE) protein, along with subA and subB that encode cytotoxic activity, may contribute to the pathogenesis of outbreak E. coli. The iro operon and ars operon may play a role in the ecological success of the epidemic clones of Salmonella spp. Concurrent relationships between esp and ter operons in E. coli and pco and sil operons in Salmonella spp. are found. Stress-response genes (asr, golT, golS), virulence gene (sinH), and antimicrobial resistance genes (mdsA and mdsB) in Salmonella spp. also show a concurrent relationship. All these findings provide helpful information for experiment design to combat outbreaks of E. coli and Salmonella spp.

Comparison of Conventional Molecular and Whole-Genome Sequencing Methods for Differentiating Salmonella enterica Serovar Schwarzengrund Isolates Obtained from Food and Animal Sources

Over the last decade, Salmonella enterica serovar Schwarzengrund has become more prevalent in Asia, Europe, and the US with the simultaneous emergence of multidrug-resistant isolates. As these pathogens are responsible for many sporadic illnesses and chronic complications, as well as outbreaks over many countries, improved surveillance is urgently needed. For 20 years, pulsed-field gel electrophoresis (PFGE) has been the gold standard for determining bacterial relatedness by targeting genome-wide restriction enzyme polymorphisms. Despite its utility, recent studies have reported that PFGE results correlate poorly with that of closely related outbreak strains and clonally dominant endemic strains. Due to these concerns, alternative amplification-based molecular methods for bacterial strain typing have been developed, including clustered regular interspaced short palindromic repeats (CRISPR) and multilocus sequence typing (MLST). Furthermore, as the cost of sequencing continues to decrease, whole genome sequencing (WGS) is poised to replace other molecular strain typing methods. In this study, we assessed the discriminatory power of PFGE, CRISPR, MLST, and WGS methods to differentiate between 23 epidemiologically unrelated S. enterica serovar Schwarzengrund isolates collected over an 18-year period from distinct locations in Taiwan. The discriminatory index (DI) of each method for different isolates was calculated, resulting in values between 0 (not discriminatory) and 1 (highly discriminatory). Our results showed that WGS has the greatest resolution (DI = 0.982) compared to PFGE (DI = 0.938), CRISPR (DI = 0.906), and MLST (DI = 0.463) methods. In conclusion, the WGS typing approach was shown to be the most sensitive for S. enterica serovar Schwarzengrund fingerprinting.

Characterization of Human Origin Salmonella Serovar 1,4,[5],12:i:- in Eastern China, 2014 to 2018

The prevalence of Salmonella serovar 1,4,[5],12:i:- among diarrhea patients has increased considerably in many countries around the world, including China. However, the characterization of this serovar of human origin has been less reported from China. We characterized 76 isolates of Salmonella 1,4,[5],12:i:- gained from diarrhea patients from 2014 to 2018 in the Jiangsu Province of eastern China. These isolates fell into a single-sequence type (ST34) determined by multilocus sequence typing (MLST), and into 44 pulsed-field gel electrophoresis band patterns, with 1 pattern (JSSMM015) comprising 12 isolates (15.9%). By means of PCR-based assays, the seven prophage located virulence genes were detected in our Salmonella 1,4,[5],12:i:- isolates with a high rate of gipA, gtgB, sspH1, sspH2, sodC1, and gtgE (93.4-97.4%), and with a moderate rate of sopE (42.1%). In contrast, none of the five plasmid-borne virulence genes (spvC, pefA, mig5, rck, and srgA) was identified. We tested the isolates' susceptibility to 18 antibiotics of 9 categories using the VITEK 2 system. A high proportion (89.5%) of the isolates were multidrug resistant (MDR) strains with full resistance to cefazolin, cefotetan, amikacin, gentamycin, and tobramycin, followed by resistance to ampicillin (88.2%) and ampicillin/sulbactam (80.3%). The resistance to piperacillin/tazobactam, ceftazidime, cefepime, and levofloxacin was scarce (2.6-9.2%). Notably, an isolate from 2018 was resistant to carbapenems. blaTEM-1B and aac(6')-Ib-cr were the most common drug resistance genes presented in cephalosporin- and fluoroquinolone-resistant strains. All Salmonella 1,4,[5],12:i:- isolates were capable of forming biofilm, with 13.2% of them having strong ability. However, no association was indicated between the scale of biofilm formation ability and MDR. Our results indicate that the combination of these characteristics may together provide a selective and competitive advantage to those Salmonella 1,4,[5],12:i:- isolates, contributing to their increasing prevalence observed worldwide.

Research Note: Identification and characterization of Salmonella spp. in mechanically deboned chickens using pulsed-field gel electrophoresis

Salmonella is one of the common foodborne bacteria, causing 80.3 million illnesses every year worldwide. This study was conducted to isolate and identify Salmonella enterica serovars from poultry samples responsible for causing foodborne poisoning in the Mississippi area, United States. A total of 55 S. enterica serovars—Enteritidis (6), Oranienburg (1), Schwarzengrund (8), Heidelberg (4), Kentucky (22), 4, [5], 12:i:- (1), Montevideo (2), Infantis (9), and multi serotypes (2)—were isolated from approximately 110 poultry samples. Through pulsed-field gel electrophoresis (PFGE) analysis, 8 to 13 bands were obtained. The profiles showed >90% similarity in strains within the same type. Consequently, PFGE could be a useful tool to determine chromosomal similarity (clonality of strains) that can be used to trace down epidemiologic sources and geographical origins of Salmonella.

Assessment and Comparison of Molecular Subtyping and Characterization Methods for Salmonella

The food industry is facing a major transition regarding methods for confirmation, characterization, and subtyping of Salmonella. Whole-genome sequencing (WGS) is rapidly becoming both the method of choice and the gold standard for Salmonella subtyping; however, routine use of WGS by the food industry is often not feasible due to cost constraints or the need for rapid results. To facilitate selection of subtyping methods by the food industry, we present: (i) a comparison between classical serotyping and selected widely used molecular-based subtyping methods including pulsed-field gel electrophoresis, multilocus sequence typing, and WGS (including WGS-based serovar prediction) and (ii) a scoring system to evaluate and compare Salmonella subtyping assays. This literature-based assessment supports the superior discriminatory power of WGS for source tracking and root cause elimination in food safety incident; however, circumstances in which use of other subtyping methods may be warranted were also identified. This review provides practical guidance for the food industry and presents a starting point for further comparative evaluation of Salmonella characterization and subtyping methods.

Pulsed Field Gel Electrophoresis: Past, present, and future

Pulsed Field Gel Electrophoresis (PFGE) has been considered for many years the 'gold-standard' for characterizing many pathogenic organisms as well as for subtyping bacterial species causing infection outbreaks. This article reviews the basic principles of PFGE and it includes the main advantages and limitations of the different electrode configurations that have been used in PFGE equipment and their influence on the DNA electrophoretic separation. Remarkably, we summarize here the most relevant theoretical and practical aspects that we have learned for more than 20 years developing and using the miniaturized PFGE systems. We also discussed the theoretical aspects related to DNA migration in PFGE agarose gels. It served as the basis for simulating the DNA electrophoretic patterns in CHEF mini gels and mini-chambers during experimental design and optimization. A critical comparison between standard and miniaturized PFGE systems, as well as the enzymatic and non-enzymatic methods for intact immobilized DNA preparation, is provided throughout the review. The PFGE current applications, advantages, limitations and future challenges of the methodology are also discussed.

Correlating exhaled aerosol images to small airway obstructive diseases: A study with dynamic mode decomposition and machine learning

Background

Exhaled aerosols from lungs have unique patterns, and their variation can be correlated to the underlying lung structure and associated abnormities. However, it is challenging to characterize such aerosol patterns and differentiate their difference because of their complexity. This challenge is even greater for small airway diseases, where the disturbance signals are weak.

Objectives and methods

The objective of this study is exploiting different feature extraction algorithms to develop a practical classifier to diagnose obstructive lung diseases using exhaled aerosol images. These include proper orthogonal decomposition (POD), principal component analysis (PCA), dynamic mode decomposition (DMD), and DMD with control (DMDC). Aerosol images were generated via physiology-based simulations in one normal and four diseased airway models in G7-9 bronchioles. The image data were classified using both the support vector machine (SVM) and random forest (RF) algorithms. The effectiveness of different features was evaluated by classification accuracy and misclassification rate.

Findings

Results show a significantly higher performance using dynamic feature extractions (DMD and DMDC) than static algorithms (POD and PCA). Adding the control variables to DMD further improved classification accuracy. Comparing the classification methods, RF persistently outperformed SVM for all types of features considered. While the performance of RF constantly increased with the number of features retained, the performance of SVM peaked at 50 and decreased thereafter. The 5-class classification accuracy was 94.8% using the DMDC-RF model and 93.0% using the DMD-RF model, both of which were higher than 87.0% in the previous study that used fractal dimension features.

Conclusion

Considering that disease progression is inherently a dynamic process, DMD(C)-based feature extraction preserves temporal information and is preferred over POD and PCA. Compared with hand-crafted features like fractals, feature extraction by DMD and DMDC is automatic and more accurate.

Phenotypic and Genotypic Eligible Methods for Salmonella Typhimurium Source Tracking

Salmonellosis is one of the most common causes of foodborne infection and a leading cause of human gastroenteritis. Throughout the last decade, Salmonella enterica serotype Typhimurium (ST) has shown an increase report with the simultaneous emergence of multidrug-resistant isolates, as phage type DT104. Therefore, to successfully control this microorganism, it is important to attribute salmonellosis to the exact source. Studies of Salmonella source attribution have been performed to determine the main food/food-production animals involved, toward which, control efforts should be correctly directed. Hence, the election of a ST subtyping method depends on the particular problem that efforts must be directed, the resources and the data available. Generally, before choosing a molecular subtyping, phenotyping approaches such as serotyping, phage typing, and antimicrobial resistance profiling are implemented as a screening of an investigation, and the results are computed using frequency-matching models (i.e., Dutch, Hald and Asymmetric Island models). Actually, due to the advancement of molecular tools as PFGE, MLVA, MLST, CRISPR, and WGS more precise results have been obtained, but even with these technologies, there are still gaps to be elucidated. To address this issue, an important question needs to be answered: what are the currently suitable subtyping methods to source attribute ST. This review presents the most frequently applied subtyping methods used to characterize ST, analyses the major available microbial subtyping attribution models and ponders the use of conventional phenotyping methods, as well as, the most applied genotypic tools in the context of their potential applicability to investigates ST source tracking.

Multi-resolution classification of exhaled aerosol images to detect obstructive lung diseases in small airways

Exhaled aerosol patterns have been used to detect obstructive respiratory diseases in the upper airways. Signals from small airway diseases are weak and may not manifest themselves in the exhaled aerosol patterns. Therefore, it will be more challenging to detect abnormalities in small airways. The objective of this study is to develop a simulation-based classification model that can accurately classify small airway diseases. The model performance was evaluated in five obstructed models that are located in lung bifurcations G7-9. The exhaled aerosol images were quantified using local fractal dimensions at different sampling resolutions (n × n). The datasets were classified using both the random forest (RF) and support vector machine (SVM) algorithms. Results show that RF performs slightly and persistently better than SVM. The sampling resolution of 12 × 12 gave the optimal classification for both algorithms. Based on the lung models with predefined obstructive levels, the optimal classification accuracy is 87.0% for 5-class classification, and is 92.5% for 4-class classification by regrouping the mislabeled samples. The proposed model with multi-resolution fractal feature extraction and RF algorithm appears to be sensitive enough to accurately distinguish airway abnormalities in small airways beyond G7 with healthy bronchiole diameter <4 mm. This aerosol-based breath test is promising to develop into an alternative or supplemental tool to the low-dose CT scanning for lung cancer screening.