Study explains why bacteria that contaminate food do not always cause serious disease in Brazil

Researchers have revealed the molecular mechanisms that make the Brazilian strain of Shiga toxin-producing E. coli less virulent and pathogenic than strains of the bacterium found in other countries. Their findings can be used to formulate epidemiological surveillance strategies.

By Elton Alisson  |  Agência FAPESP – Some strains of Escherichia coli bacteria, which normally live in the digestive tract of humans and other animals, produce Shiga toxin, one of the most potent bacterial toxins known, and are therefore known as STEC, short for Shiga toxin-producing E. coli. 

Microbiologists have been intrigued by the fact that consumption of STEC-contaminated beef is associated with hemolytic-uremic syndrome (HUS) in Australia, the United States, Japan and Argentina; however, in Brazil, for example, a certain serotype of STEC has been found in beef and cattle dung – a natural reservoir for these bacteria – but never in patients diagnosed with HUS. 

HUS is a severe blood disorder that can cause kidney failure and mainly occurs in children under the age of five.

Scientists at the University of São Paulo (USP) in Brazil, working with colleagues at Butantan Institute and the Federal University of São Paulo (UNIFESP), have now discovered molecular mechanisms that make “foreign” STEC more virulent and pathogenic than the Brazilian STEC strain. 

The findings, produced as part of a project supported by FAPESP, are published in the journal Microorganisms, which highlights the study on its cover. 

“The results of the study can help identify new molecular markers of virulence and pathogenicity for STEC, and these markers will be useful to design epidemiological surveillance strategies. The risk is that the Brazilian strain could undergo a genetic mutation that also makes it capable of causing disease,” Carlos Alberto Moreira Filho, the principal investigator of the project, told Agência FAPESP. 

According to Moreira Filho, a professor in the Pediatrics Department of USP’s Medical School (FM-USP), HUS-causing STEC survives in the intestine for longer than less-virulent STEC strains and gradually releases Shiga toxin, which crosses the intestinal barrier, enters the bloodstream and causes kidney failure. 

To understand the difference between this microorganism and the strains that do not cause the disease, the researchers conducted a study in which they compared the genome of a strain isolated from an Australian patient with HUS (EH41) with a Brazilian strain isolated from cattle dung (Ec472/01). The study was published in 2017 in the journal PLOS ONE.

The results showed that 15 genes expressed by EH41, including the virulence regulator dicA, were not active in Ec472/01.

“In this study, we succeeded in clarifying the genetic and genomic differences between the Australian and Brazilian strains,” Moreira Filho said.

Next, to analyze the response of cells in the inner wall of the small intestine to these two strains, the researchers performed an experiment using the Caco-2 cell line, which is widely used as an in vitro model for studying the human intestinal epithelial barrier.

“This cell line very closely mimics the layer of epithelial cells that line the inside of the small intestine [enterocytes] and to which STEC adheres to create colonies,” Moreira Filho explained.

Bacteria of the strains EH41 and Ec472/01 were placed on Caco-2 cells for three hours. A systems biology approach was used to analyze changes in their gene expression profiles during the three-hour period at intervals of 7.5 minutes. 

Morphological alterations in the Caco-2 cells were assessed on the basis of scanning electron microscopy at intervals of one, two and three hours. 

The results showed that the strain associated with HUS expressed a set of virulence genes that induced a distinct response in human enterocytes, which could help explain its more severe pathogenicity.

Activation of these genes in the HUS-causing strain prevents enterocytes from expressing certain defense reactions; as a result, the pathogen is able to remain in the intestine for longer, the researchers surmised. While there, the pathogen slowly produces Shiga toxin, which enters the bloodstream and damages the kidneys.

“The enterocytes, which are also natural immune cells, fail to carry out their mission of stopping the bacteria, allowing it to remain in tissue and release Shiga toxin,” Moreira Filho said.

Network analysis

The researchers used a methodology called gene coexpression network analysis to investigate enterocyte gene expression and interaction over time, as induced by exposure to bacterial strains. To this end, they deployed 3D visualization software developed by colleagues at USP’s São Carlos Physics Institute (IFSC) and mathematical models developed by a team led by Roberto Marcondes César Junior, a professor at USP’s Mathematics and Statistics Institute (IME), via the project supported by FAPESP.

The mathematical models identify variations in network topology – changes in the hierarchical level of a gene in a network in terms of its influence on the expression of other genes.

“The algorithms look at the genes’ functioning and assemble networks, helping to identify groups or modules that explain why certain genes are connected in one situation and not in others. The methodology also enables us to simulate the effects on a network of the inactivation of a specific gene,” César Junior said.

The analysis showed that HUS-associated bacteria caused a sharp change in Caco-2 cell gene coexpression network topology during the three-hour interaction.

“The bacteria caused a loss of connectivity among the cells’ genes, leaving the cells with fewer links,” Moreira Filho said.

Gene group analysis showed that the HUS-causing strain induced Caco-2 cell inflammation and death in the first hour of exposure, whereas the Brazilian strain caused fewer gene alterations, especially in relation to cell structure (cytoskeleton) and immune response.

Scanning electron microscope analysis revealed that EH41 caused greater destruction of Caco-2 cells than Ec472/01, especially loss of microvilli, which are typical of healthy enterocytes and are crucial to the proper functioning of the intestine.

“These findings help us understand the mechanisms that trigger immune response and may lead to serious diseases, such as HUS,” Moreira Filho said.

In his opinion, the study points to the need for further research to investigate the influence of environmental factors on STEC virulence, including the passage of these bacteria through the digestive tract in humans and cattle.

“We need to find out whether the passage of these bacteria through the digestive system of Brazilian cattle, which feed differently from cattle elsewhere, changes the expression of their virulence genes or even switches them off,” Moreira Filho said.

The Microorganisms article “Dynamic gene network analysis of Caco-2 cell response to Shiga toxin-producing Escherichia coli-associated hemolytic-uremic syndrome” (doi: 10.3390/microorganisms7070195) by Silvia Y. Bando, Priscila Iamashita, Filipi N. Silva, Luciano da F. Costa, Cecilia M. Abe, Fernanda B. Bertonha, Beatriz E. C. Guth, André Fujita and Carlos A. Moreira-Filho can be read at: www.mdpi.com/2076-2607/7/7/195. 

The PLOS ONE article “A hemolytic-uremic syndrome-associated strain O113:H21 Shiga toxin-producing Escherichia coli specifically expresses a transcriptional module containing dicA and is related to gene network dysregulation in Caco-2 cells” (doi: 10.1371/journal.pone.0189613) by Silvia Yumi Bando, Priscila Iamashita, Beatriz E. Guth, Luis F. dos Santos, André Fujita, Cecilia M. Abe, Leandro R. Ferreira and Carlos Alberto Moreira-Filho can be read at: journals.plos.org/plosone/article?id=10.1371/journal.pone.0189613.