By means of the gentamicin HEp-2 cell invasion assay, it was demonstrated that 82% of the Campylobacters tested were cell-invasive, including 83% of isolates from bloody diarrhoea and 80% of isolates from watery diarrhoea. The large number of invasive strains from watery diarrhoea suggests the possible role of invasiveness in the production of watery diarrhoea. Whether this stage can progress further to more severe symptoms such as bloody diarrhoea remains to be elucidated. Whether this progression to bloody diarrhoea occurs as a result of toxin production is still debatable. In Vero cells, invasion was less efficient and intracellular multiplication was not observed.
There are several methods for the detection of haemolytic activity in campylobacters. However, we found the haemolytic effect of campylobacters on conventional blood agar plates to be variable, inconsistent and difficult to interpret. Blood agarose plates showed campylobacter haemolytic activity more clearly. The incubation conditions (temperature and gaseous) appear to be important for the expression of this activity. Ninety four percent of the Campylobacter isolates examined were found to be haemolytic by the microplate assay with minimal haemolytic units that ranged from 1 to 64. Haemolytic activity was detected only from live bacterial cultures and not from any of the 50 bacterial culture supernates, which suggests that campylobacters may possess a cell-associated haemolysin. The identification of such haemolytic activity in a large number of campylobacters (94%) suggests its potential role as a virulence factor in campylobacter gastroenteritis.
Approximately 57% of clinical and 33% of poultry isolates examined produced a cytotoxin. Cytotoxic activity was detected in 25 (50%) isolates of Campylobacter of which 12 were isolated from bloody diarrhea and 9 from watery stools. The cytotoxin titers were low, ranging from 2 to 16. The crude filtrates from 50 Campylobacter isolates showed no cytotoxic effect in Vero cells, no fluid accumulation in suckling mice and no hemolytic activity.
Given that Campylobacter jejuni is recognized as the most common cause of bacterial gastroenteritis worldwide, recent findings showing comparable levels of Campylobacter concisus in patients with gastroenteritis would suggest that this bacterium is clinically important. The prevalence and abundance of Campylobacter concisus in stool samples collected from patients with acute gastroenteritis was examined using quantitative real-time PCR. The associated virulence determinants exotoxin 9 and zonula occludens toxin DNA were detected for Campylobacter concisus-infected samples using real-time PCR. Campylobacter concisus was detected at high prevalence in patients with gastroenteritis (49.7 %), higher than that observed for Campylobacter jejuni (∼5 %). The levels of Campylobacter concisus were putatively classified into clinically relevant and potentially transient subgroups based on a threshold developed using Campylobacter jejuni levels, as the highly sensitive real-time PCR probably detected transient passage of the bacterium from the oral cavity. A total of 18 % of patients were found to have clinically relevant levels of Campylobacter concisus, a significant number of which also had high levels of one of the virulence determinants. Of these patients, 78 % were found to have no other gastrointestinal pathogen identified in the stool, which strongly suggests a role for Campylobacter concisus in the aetiology of gastroenteritis in these patients. These results emphasize the need for diagnostic laboratories to employ identification protocols for emerging Campylobacter species. Clinical follow-up in patients presenting with high levels of Campylobacter concisus in the intestinal tract is needed, given that it has been associated with more chronic sequelae.