RESUMO
In the analysis of water samples, the type of filtration membrane material can influence the recovery of Legionella species, although this issue has been poorly investigated. Filtration membranes (0.45 µm) from different materials and manufacturers (numbered as 1, 2, 3, 4, and 5) were compared: mixed cellulose esters (MCEs), nitrocellulose (NC), and polyethersulfone (PES). After membrane filtration of samples, filters were placed directly onto GVPC agar and incubated at 36 ± 2 °C. The highest mean counts of colony-forming units and colony sizes for Legionella pneumophila and Legionella anisa were obtained with PES filters (p < 0.001). All membranes placed on GVPC agar totally inhibited Escherichia coli and Enterococcus faecalis ATCC 19443 and ATCC 29212, whereas only the PES filter from manufacturer 3 (3-PES) totally inhibited Pseudomonas aeruginosa. PES membrane performance also differed according to the manufacturer, with 3-PES providing the best productivity and selectivity. In real water samples, 3-PES also produced a higher Legionella recovery and better inhibition of interfering microorganisms. These results support the use of PES membranes in methods where the filter is placed directly on the culture media and not only in procedures where membrane filtration is followed by a washing step (according to ISO 11731:2017).
RESUMO
Glycine-vancomycin-polymyxin-cycloheximide agar (GVPC) is a recommended medium for the detection of Legionella spp. in water samples. However, its quality could be improved in terms of recovery of Legionella spp. and selectivity properties. Modifications were introduced in GVPC manufacture: autoclaving conditions (115°C, 15 min) and atmosphere during component-stirring (removal of oxygen and N2 injection). The use of softer autoclaving conditions (115°C, 15 min) improved the growth of Legionella anisa by the spiral method and Legionella pneumophila after membrane filtration. The medium manufactured with O2 removal and autoclaving for 15 min at 115°C allowed a faster growth of L. pneumophila (colonies visible at day 2) and a notable increase of L. anisa growth (colonies appearing at day 3, and statistically significant numbers of CFU at day 5). After 3 to 5 days of incubation, the improved media showed higher selectivity properties, particularly for Enterococcus faecalis ATCC 29212 and Pseudomonas aeruginosa ATCC 9027. A further improvement was achieved by the addition of N2 during ingredient stirring, leading to a statistically significant faster growth of L. pneumophila at days 2 and 3 and L. anisa at day 3. Selectivity properties were also enhanced, resulting in the complete inhibition of both E. faecalis strains and Escherichia coli and complete-partial inhibition of P. aeruginosa. Oxygen removal during GVPC manufacture using a vacuum pump system promotes the growth of L. pneumophila and L. anisa, and markedly inhibits the growth of E. coli, P. aeruginosa, and E. faecalis. IMPORTANCE Currently, GVPC is a recommended medium for the detection of Legionella spp. in water samples. However, recovery of Legionella spp. and selectivity properties can be improved. GVPC medium manufactured without oxygen improved the growth of Legionella pneumophila and Legionella anisa. Oxygen removal during GVPC manufacture also improved selectivity properties. A further improvement was achieved by the addition of N2 during ingredient stirring, leading to a faster growth of L. pneumophila at days 2 and 3 and L. anisa at day 3 and enhancement of selectivity properties. The introduction of the modified GVPC medium in routine practice can allow a better detection of Legionella spp. in water samples.