Development of an RT-PCR for detecting enterococci

To determine quickly whether drinking water is hygienically reliable, it is important – in addition to the available RT-PCR method for E. coli – to have a method for detecting enterococci. This bacterial genus includes many different species. Only some of them are linked to faecal contamination and they can be identified using the standard culture method. The method being developed here will make it possible to determine hygienic reliability within a few hours instead of the two days currently required.

For the purposes of developing a new RT-PCR method to detect enterococci, this study selected at least ten species on the basis of literature research, faecal research and research into the species that are found regularly in Dutch and Belgian drinking water using the standard culture method; the drinking water laboratories conducted the screening. Four promising primer sets were then selected. The results provide an initial indication that the new methods can be used for the sensitive and specific detection of enterococci in drinking water.

Extension of existing fast method

The detection of the faecal indicator organisms E. coli and enterococci is used in particular to determine the hygienic reliability of drinking water. Dutch legislation requires time-consuming (approx. 24 hours for E. coli and 48 hours for enterococci) culture methods (NEN-EN-ISO standards) to be used for this purpose. It is important to develop fast and reliable methods, first of all to minimise the risk of exposure to pathogenic microorganisms quickly. It is also necessary, in the event of an emergency (infection, interventions in the mains network), to be able to determine quickly whether, after the measures have been taken, supplies can be resumed or whether a recommendation to boil water can be suspended. New RT-PCR methods meet these requirements. The genetic material of indicator organisms can be used to determine the hygienic reliability of drinking water within 4 hours.

The RT-PCR method for the fast detection of E. coli has been implemented at all Dutch drinking water laboratories and the Belgian laboratories of De Watergroep and Pidpa. The validation of the method has demonstrated that there are few differences between the analysis results obtained with this new method and those obtained with the standard culture method for drinking water (ISO 9308-1). Given this validation, the Ministry of Infrastructure and Water Management granted the drinking water laboratories permission to use the RT-PCR as an alternative method for the detection of E. coli starting in April 2018. Meanwhile, steps have also been taken to have the method accepted within the European Union. The validation report has been submitted to the European Microbiology Expert Group (EMEG) for assessment.

In addition to a method for the rapid detection of E. coli, a rapid response to calamities also requires a method for the rapid detection of enterococci. A validated method is already available for the detection of four intestinal enterococcal species (Enterococcus faecium, E. faecalis, E. hirae and E. Durans) that have been identified by the WHO as the most faecal-specific species. However, due to the limitation to these four species, differences have been observed with the culture method, which covers a wider range of enterococcal species. Because it cannot be excluded that some of these differences are relevant for the assessment of hygienic reliability, it was decided, after consultation with the RIVM, to extend the current rapid method to detect the same enterococcal species in drinking water as with the culture method. This extension is the subject of this research project.

Steps towards the intended RT-PCR method

The following steps have been taken in this project:

  1. Selection of enterococcal species
    In order to select the relevant enterococci, all the drinking water laboratories participating in the project conducted screening to determine which species are found with the regular culture measurements by typing them up to species level with MALDI-TOF-MS. They identified, in addition to the four ‘WHO species’, six additional enterococcal species that are found in drinking water and may be of faecal origin (see step 2).
  2. Design of primer pairs and probes
    On the basis of the sequence homology with the 16S rRNA gene, the ten selected enterococcal species can be divided into four genetically related groups. To detect them, several potentially applicable primer sets were designed in this research, the reaction conditions were optimised, and sensitivity and specificity were tested on a limited scale. Four promising primer sets were selected from these analyses. They target the 16S ribosomal RNA gene for the ten species found in the previous step:
    a) The four species listed by the WHO: E. faecium, E. faecalis, E. durans and E. hirae.
    b) At least the six additional species from the review: E. casseliflavus, E. mundtii, E. gallinarum, E. moraviensis, E. haemoperoxidus and E. avium.
  3. Optimisation of reaction conditions
    We are continuing with the most promising primer pairs from step 2. The reaction conditions for the RT-PCR analyses are being optimised to establish efficient reactions that detect the target organism only.
  4. Limited validation
    a) Limited validation is taking place to determine whether the selected enterococci can be detected at adequate levels of specificity and sensitivity. If this limited validation shows that the method is expected to be sufficiently specific and sensitive, it will have to be extensively validated in accordance with ISO 16140-2 in a follow-up phase.
    b) Determination of the method’s detection limit for the different target enterococcal species.
    c) Inclusion/exclusion for a limited number of strains
    The inclusiveness of the developed method is being studied for a set of at least 30 strains. For the clearest possible description of detectability, these will be strains belonging both to the ten selected enterococcal species and to other enterococcal species.

The exclusivity of the developed method is being investigated by performing RT-PCR analyses on a limited number of bacterial strains that can be found in water and that are related to enterococci (such as streptococci and aerococci).

Successful RT-PCR development

If the RT-PCR method is applied to dilutions of samples with known species of enterococci, it may be possible to detect the ten selected species with sufficient sensitivity. The fact that the new method identifies the RNA of the selected enterococcal species and not the RNA of bacteria that are not members of the enterococcal genus demonstrates that it is adequately specific. This was also the case with RT-PCR in ten drinking water samples taken from different locations, in which no enterococci were detected using the culture method. These results provide an initial indication that the new methods can be used for the sensitive and specific detection of enterococci in drinking water.

In order to determine the extent to which the results obtained with RT-PCR match the results obtained with the culture method, comprehensive validation will be required in accordance with ISO 16140-2:2016. Before starting this validation process, it will be important to investigate whether the primer sets can be combined. That will make the detection of the ten species in a single reaction (‘multiplex RT-PCR’) possible. This multiplex RT-PCR makes the validation and application of the method easier and less costly. In addition, in order to determine the ‘Relative Limit Of Detection (RLOD)’, it is important to have accurately quantified cell suspensions of enterococci. The development of a robust method is needed to produce those suspensions.