Letters in Applied Microbiology ISSN 0266-8254

NOTE TO THE EDITOR

Dimethyl formamide-free, urea-NaCl fluorescence in situ

hybridization assay for Staphylococcus aureus
T.S. Lawson, R.E. Connally, S. Vemulpad and J.A. Piper
Faculty of Science, Macquarie University, NSW, Australia

Keywords
detection, identification, infection, rapid
methods, staphylococci.
Correspondence
Tom Lawson, Faculty of Science, Macquarie
University, NSW 2109, Australia.
E-mail: tomxlawson@gmail.com
2011 ⁄ 1642: received 27 September 2011,
revised 1 December 2011 and accepted 2
December 2011
doi:10.1111/j.1472-765X.2011.03197.x
Abstract
Aims: To test the feasibility of identifying Staphylococcus aureus with a fluores-
cence in situ hybridization (FISH) assay that uses a single hot-plate and urea-
NaCl reagents.
Methods and Results: Slides spotted with S. aureus and treated with methanol
and lysozyme were incubated with urea-NaCl reagents on a hot-plate with a
precise temperature control and identified with specific DNA probes.
Conclusions: Staphylococcus aureus was detected and differentiated from Staph-
ylococcus epidermidis in 1 h with a novel FISH method that used a single
hot-plate and in the absence of dimethyl formamide.
Significance and Impact of Study: A rapid hot-plate FISH assay with urea-
NaCl and without toxic dimethyl formamide might be useful if FISH is run
infrequently or where resources are limited.

Slide-based fluorescence in situ hybridization (FISH) is a
robust assay for characterizing intact bacteria in clinical
specimens. It is usually run with an incubator and a
water-bath and with dimethyl formamide (referred to as
formamide) and NaCl buffers (Wang 2010). Occasionally
a specialized hot-plate is used for hybridization with DNA
(Poppert et al. 2010) or PNA probes (AC005; AdvanDx,
Wobum, MA) followed by washing with a thermo-mixer
(Poppert et al. 2010) or water-bath (AC006; AdvanDx). A
regular hot-plate has not been used, however, for the
entire FISH assay with DNA probes.
This report identified Staphylococcus aureus with a
novel FISH assay performed on a single hot-plate using
DNA probes. Staphylococcus aureus is a clinically signifi-
cant Gram-positive bacteria (Wang 2010) and if DNA
probes are used, its permeabilization can be more com-
plex than other bacteria (Poppert et al. 2010). The new
assay tested is modified from an assay reported by the
authors (Lawson et al. 2011). Improvements include
marking the slides with a wax pencil, replacing the form-
amide-NaCl buffers with urea-NaCl reagents and using a
hot-plate with a plastic cover instead of 50 ml tubes, an
incubator and a water-bath (Lawson et al. 2011). To
compare the signal strength of the new assay, the previous
assay (Lawson et al. 2011) was run in parallel as a
control.
If a hot-plate is used for incubations, reagents can dry
out and the temperature can fluctuate, both of which can
result in a weaker hybridization signal. Possible solutions
to enhance the signal were tested including using urea as
an alternative denaturing reagent to formamide (Soe et al.
2011). Unlike formamide, urea is non-toxic (Simard et al.
2001), can inhibit RNase degradation (Simard et al. 2001)
and can act as an additional permeabilizer with the result
that the FISH signal could be increased (Huang et al.
2011).
In preparation for FISH, 10 clinical isolates of PBP2-
negative S. aureus and 10 of Staphylococcus epidermidis
previously identified with PCR (Thomas et al. 2007) were
randomly collected at a hospital and cultured in nutrient
broth to an optical density of 1Æ0 at 600 nm (CM0001;
Oxoid, Basingstoke, UK). Cultures were diluted 1 : 10 in
broth, and 10 ll aliquots were spotted to slide-wells
(X1XER308B; Menzel-Glaser, Braunschweig, Germany),
dried at 47C on a hot-plate and fixed with absolute
methanol for 1 min. Wells were marked with a wax pen-
cil to restrain reagents to the wells.
Two FISH assays were tested, an incubator-bath assay
described in Lawson et al. (2011) and a hot-plate FISH
assay (this report). Two sets of hybridization and washing
reagents were tested with each assay. The first used con-
ventional hybridization (formamide-NaCl) and NaCl

ª 2011 The Authors

Letters in Applied Microbiology 54, 263–266 ª 2011 The Society for Applied Microbiology 263
FISH on a hot-plate with urea
washing buffers (Poppert et al. 2010; Lawson et al. 2011).
The second used urea with NaCl (urea-NaCl) for both
hybridization and washing (this report). With all treat-
ments, two DNA probes Staaur (16S69: 5¢-GAAGCAAG-
CTTCTCGTCCG -3¢) and EUB338 probe (16S337: 5¢-
GCTGCCTCCCGTAGGAGT -3¢) conjugated at the 5¢ end
to Alexa Fluor 488 identified S. aureus and eubacteria,
respectively.
The first FISH assay was tested with an incubator-bath
and formamide-NaCl reagents (Poppert et al. 2010; Lawson
et al. 2011). To permeabilize the bacteria, wells were spot-
ted with 30 ll of 15 mg ml)1 lysozyme (pH 7Æ0). The slides
were fitted into 50 ml tubes (210–261; Greiner Bio-One,
Frickenhansen, Germany), incubated at 47C for 30 min
and then rinsed with methanol and dried. For hybridiza-
tion, wells were spotted with 20 ll of formamide-NaCl buf-
fer [30% (v ⁄ v) formamide (A2156; Applichem, Darmstadt,
Germany), 0Æ9 mol l)1 NaCl (S6191; Sigma-Aldrich, St
Louis, MO), 20 lmol l)1 Tris–HCl (pH 7Æ0), 0Æ01% (w ⁄ v)
SDS (4390; Sigma, L4390) in Milli-Q (MQ) water; Milli-
pore, Bedford, MA] containing 1 lmol l)1 of probe.
The slides were fitted into 50 ml tubes and placed in a
47C incubator for 20 min. After hybridization, slides
were rinsed with prewarmed NaCl washing buffer
[0Æ64 mol l)1 NaCl, 20 lmol l)1 Tris–HCl (pH 7Æ0) and
0Æ01% SDS in MQ water] and fitted into 50 ml tubes
containing 25 ml of prewarmed washing buffer and
placed in a gently agitated 47C water-bath for 5 min.
Slides were removed and rinsed with MQ water and
mounted wet with a cover-slip for viewing.
The second FISH assay used a hot-plate and was opti-
mized with urea-NaCl (this report). The permeabilization
step was the same as before except that it was performed on
a 47C hot-plate, and the slides were covered with a clear
plastic lid (78 · 78 mm, 123160; Decor, Melbourne, Aus-
tralia) to minimize temperature change and reagent drying.
The hot-plate was developed by one of the authors (Russell
Connally) and had an accuracy of ±0Æ5C at 47C. A
platinum resistance probe was used with a microcomputer
display for accurate temperature control. Slides were rinsed
with methanol and were dried before 20 ll of urea-NaCl
[1 mol l)1 urea (U6504; Sigma), 0Æ9 mol l)1 NaCl,
20 lmol l)1 Tris-HCl (pH 7Æ0) in MQ water] with 1 lmol
l)1 of probe was spotted to each well.
Slides were incubated on the 47C hot-plate for 20 min
before rinsing twice with prewarmed 250 ll of urea-NaCl
[8 mol l)1 urea, 0Æ9 mol l)1 NaCl, MQ water and
20 lmol l)1 Tris-HCl (pH 7Æ0)]. Slides were placed on
the hot-plate and 30 ll of prewarmed urea-NaCl was
spotted immediately to each well. Slides were incubated
with the plastic cover for 5 min, rinsed twice with urea-
NaCl again before a final rinse with MQ water and
mounting as before.
264 Letters in Applied Microbiology 54, 263–266 ª 2011 The Society for Applied Microbiology
T.S. Lawson et al.
Slides were visualized with an epifluorescence
microscope (BX51; Olympus, Tokyo, Japan) fitted with a
60 objective (UPLFLN; Olympus) and FITC ⁄ DAPI filters
(U-MWU2, U-MWIB2; Olympus). Images were acquired
at a resolution of 1360 · 1024 with an Olympus DP72 cam-
era and software (DP2-BSW v2Æ2; Olympus) set to a gain of
200 ISO and an exposure of 2 s. Images from three experi-
mental runs were analysed with ImageJ (v1Æ43u; NIH,
Bethesda, MD). Summary statistics were compared with
one-way analysis of variance (anova) and a P value of 0Æ05.
A summary of the results for the two FISH assays and
their reagents is listed in Table 1 and shown in Fig. 1(a–
d). Staphylococcus epidermidis was negative for the Staaur
probe (Fig. 1f), and S. aureus and Staph. epidermidis were
both positive for the EUB338 probe. The incubator-bath
assay produced a higher signal than the hot-plate. The
incubation temperature and humidity were observed to
vary more on a hot-plate as it was not sealed. The urea-
NaCl produced a higher signal than the formamide-NaCl.
Urea might be acting as an additional premeabilizer for
S. aureus (Huang et al. 2011). No difference in signal was
observed between the 10 isolates of each bacteria tested.
The one-way anova found a significant difference
(P < 0Æ000) for the four treatments listed in Table 1.
The hot-plate assay was initially developed with the con-
ventional formamide-NaCl hybridization and washing buf-
fers. The signal was observed to be weaker than the same
assay run with an incubator and a water-bath (Table 1). To
increase the signal, urea was tested as a substitute for form-
amide in the hybridization buffer at 0Æ5, 1, 2, 4 and 8 mol l)1
(Kourilsky et al. 1970; Simard et al. 2001; Soe et al. 2011)
with NaCl at 0Æ9 mol l)1 (Poppert et al. 2010; Lawson et al.
2011). The signal was highest at 1 and 2 mol l)1 urea, and
1 mol l)1 was chosen for further testing.
A number of washing treatments were tested with the 1
mol l)1 urea and 0Æ9 mol l)1 NaCl hybridization reagent.
Table 1 A comparison of the Staaur probe signal intensity for each
of the FISH treatments tested with Staphylococcus aureus.
Signal intensity
FISH treatment Mean* CI
Incubator-bath FISH assay
Formamide-NaCl and NaCl buffers 36Æ09 0Æ62
Urea-NaCl reagents 49Æ5 1Æ23
Hot-plate FISH assay§
Formamide-NaCl and NaCl buffers 32Æ7 0Æ52
Urea-NaCl reagents 45Æ2 1Æ13
*Mean signal intensity was measured in 8-bit Gray-scale.
Confidence interval was calculated at 95%.
Incubator-bath FISH protocol (Lawson et al. 2011) as described in
this report.
§Hot-plate FISH protocol as described in this report.
ª 2011 The Authors
T.S. Lawson et al. FISH on a hot-plate with urea

(a) (b)

(c) (d)

(e) (f)

(g)

Figure 1 Staphylococcus aureus visualized with Alexa Fluor 488 after performing an incubator-bath FISH assay with (a) formamide-NaCl (Law-
son et al. 2011) or (b) urea-NaCl reagents (this report). Staphylococcus aureus after performing the hot-plate FISH assay (this report) with (c) form-
amide-NaCl or (d) urea-NaCl reagents. The same S. aureus visualized with (e) DAPI. Staphylococcus epidermidis after performing the hot-plate
assay with urea-NaCl and visualized with (f) Alexa or (g) DAPI. Bar in lower right corner is 5 lm.

A conventional wash containing 0Æ9 mol l)1 NaCl was
applied for 1 min as described previously (Lawson et al.
2011), but the signal was inconsistent. Urea was reported
by Hutton (1977) to reduce melting temperature by
approx. 2Æ25C with each 1 mol l)1 increase in its concen-
tration whereas melting temperature was reduced by
approx. 0Æ6C with each 1% (v ⁄ v) increase of formamide.
To improve the signal and to simplify the assay, the
washing buffer was replaced with the urea-NaCl reagents
used in the hybridization step.
The stringency of this new washing regime was
adjusted with urea rather than with the conventional
NaCl and was set higher than that used for hybridization
so that duplexes with mismatches would be removed. The

ª 2011 The Authors

Letters in Applied Microbiology 54, 263–266 ª 2011 The Society for Applied Microbiology 265
FISH on a hot-plate with urea
urea wash was tested at 0, 1, 2, 4 and 8 mol l)1 with
NaCl at 0Æ9 mol l)1 (Kourilsky et al. 1970; Simard et al.
2001; Soe et al. 2011). The non-specific signal for
Staph. epidermidis was the weakest at 4 and 8 mol l)1
urea. To minimize background signal, the 8 mol l)1 con-
centration was chosen (Kourilsky et al. 1970). Increasing
urea from 1 mol l)1 in the hybridization step to 8 mol l)1
in the washing step decreased the melting temperature by
about 26C (Hutton 1977).
The findings suggest that if urea-NaCl reagents are
used, it is feasible to control hybridization conditions and
produce a sufficient signal with the hot-plate FISH
method. There are, however, some limitations to the find-
ing. The relationship between urea and formamide con-
centrations and urea’s action as a permeabilizer was not
established. Although it was not observed after the
20 min hybridization step, urea can have a slower rate of
denaturation than formamide (Hutton 1977). Other bac-
teria were not tested, and the specificity and sensitivity of
the assay were not assessed.
In conclusion, we described a novel FISH assay that does
not require an incubator, water-bath, formamide, lysosta-
phin or PNA probes (the latter two are expensive). Urea-
NaCl reagents were simple to prepare and unlike formam-
ide, non-toxic. The exclusion of formamide may open up
new applications, such as simplified FISH analysis using cell
sorters or FISH procedures using beacon probes without an
additional washing step. The findings warrant further spec-
ificity and sensitivity testing in a clinical scenario.
Acknowledgements
The authors acknowledge the Australian Research Coun-
cil’s Linkage Projects (LP0775196) for funding this
research and the Australian Proteome Analysis Facility
(APAF) for providing laboratory facilities.
266 Letters in Applied Microbiology 54, 263–266 ª 2011 The Society for Applied Microbiology
T.S. Lawson et al.
References
Huang, E., Talukder, S., Hughes, T.R., Curk, T., Zupan, B.,
Shaulsky, G. and Katoh-Kurasawa, M. (2011) BzpF is a
CREB-like transcription factor that regulates spore
maturation and stability in dictyostelium. Dev Biol 358,
137–146.
Hutton, J.R. (1977) Renaturation kinetics and thermal stability
of DNA in aqueous solutions of formamide and urea.
Nucleic Acids Res 4, 3537–3555.
Kourilsky, P., Manteuil, S., Zamansky, M.H. and Gros, F.
(1970) DNA-RNA hybridization at low temperature in the
presence of urea. Biochem Biophys Res Commun 41, 1080–
1087.
Lawson, T.S., Connally, R.E., Iredell, J.R., Vemulpad, S. and
Piper, J.A. (2011) Detection of Staphylococcus aureus with
a fluorescence in situ hybridization that does not require
lysostaphin. J Clin Lab Anal 25, 142–147.
Poppert, S., Riecker, M., Wellinghausen, N., Frickmann, H.
and Essig, A. (2010) Accelerated identification of Staphylo-
coccus aureus from blood cultures by a modified fluores-
cence in situ hybridization procedure. J Med Microbiol 59,
65–68.
Simard, C., Lemieux, R. and Cote, S. (2001) Urea substitutes
toxic formamide as destabilizing agent in nucleic acid
hybridizations with RNA probes. Electrophoresis 22, 2679–
2683.
Soe, M.J., Moller, T., Dufva, M. and Holmstrom, K. (2011) A
sensitive alternative for MicroRNA in situ hybridizations
using probes of 2¢-O-Methyl RNA+ LNA. J Histochem Cy-
tochem 59, 661–672.
Thomas, L.C., Gidding, H.F., Ginn, A.N., Olma, T. and Iredell,
J. (2007) Development of a real-time Staphylococcus aureus
and MRSA (SAM-) PCR for routine blood culture. J
Microbiol Methods 68, 296–302.
Wang, P. (2010) Simultaneous detection and differentiation of
staphylococcus species in blood cultures using fluorescence
in situ hybridization. Med Princ Pract 19, 218–221.
ª 2011 The Authors