2.6. MUG direct assay using the hand-held detector
Borosilicate glass test tubes (size 10 75 mm) were prepared
with 100 ml of 50 mM MUG in PBS pH 6.8. Aliquots (900 ml) of
freshly collected river water samples (not filtered or otherwise
treated) were added to the tubes, and after mixing, the tubes
were incubated for 30 min, after which the reaction was
stopped by adding 100 ml of 1 M NaOH. The tubes were individually
read with the hand-held confocal fluorescence
detector FLUO SENS SD (ESE GmbH) which was aligned into
a prototype sample holder (Fig. S1). This holder was designed
to allow use of disposable glass tubes and can also be readily
adapted to house cuvettes. Each sample was measured with
25 excitation-detection cycles in 1 s intervals (excitation
wavelength 360 nm, emission wavelength 460 nm). The
sensor was controlled by and the data were stored on a FLUO
LOG Handheld (ESE GmbH) control unit. The instrument with
the coupled control unit works independently allowing
portable use and on-site measurements. The control unit was
connected to a Laptop/PC via USB and the data exported into
Microsoft Excel for further analysis. Each water sample was
measured in triplicate. The two controls consisted of water
sample without MUG substrate and MUG substrate with
ultrapure water. The sensitivity of the detector was adjusted
and a reference curve recorded with dilutions of 4MU in PBS;
NaOH was added and the fluorescence measured in the same
manner than for the water samples. Control experiments
were carried out by adding solutions of varied concentrations
of b-D-glucuronidase to the tubes and analysing the fluorescence
in the same manner than for the water samples.
Samples of serial diluted E. coli cultures (see above) were also
analysed for b-D-glucuronidase activity following the same
protocol.
2.7. Optimisation of detection conditions for the
hand-held detector
The FLUO SENS SD confocal fluorescence detector includes
a software controlled setup for detector sensitivity as well as
a correction for background signal and blank sample. Detector
sensitivity was adjusted initially with a 4MU standard curve
with the scope to achieve linearity over a range of 103
magnitudes and reproducible signal to background ratios for
the lowest concentrations. 4MU concentrations used for this
step were from 10 nM to 20 mM. The settings where then
further adjusted with the MUG-GUD assay and finally with E.
coli spiked water samples. Once the sensitivity setup had been
optimised it was stored in the software file and remained
unchanged for the analyses of river water samples. Sterile
Ringer’s solution was used as a blank to zero the instrument.