immunosorbent assay (GM-EIA) was accepted as an
initial screening tool for invasive aspergillosis. Various
schemes of nucleic acid amplifi cation have been
suggested by a number of studies. Our group has devel-
oped a method to detect 18S ribosomal RNA (rRNA)
using nucleic acid sequence based amplifi cation using
molecular beacon(MB-NASBA) and have reported a
cut-off value for the clinical diagnosis of IA [4,5].
Recently, we have improved the NASBA method in two
ways. First, we have designed a new set of primers and probes
for Aspergillus 18S rRNA different from the original ones in
that their sequences do not overlap with those of human 18S
rRNA. Second, we used LightCycler (Roche Diagnostics) as
a platform for the detection of NASBA products. In this
study, we attempted to use fl uorescence resonance energy
transfer technology (FRET-NASBA) instead of molecular
beacon technology. We also performed MB- and FRET-
NASBA on clinical specimens, and compared the results
obtained with each to evaluate the correlation between the
Received 15 March 2010; Received in fi nal revised form 4 June 2010;
Accepted 7 July 2010
1 These authors contributed equally to this work.
Correspondence: Jin-Hong Yoo, Department of Internal Medicine,
Bucheon St. Mary ’ s Hospital, 2-Sosa-Dong, Wonmi-Gu, Bucheon, 420-
717, Kyonggi-Province, South Korea. Tel: ? 82 32 340 7014; fax: ? 82
32 340 2669. E-mail: firstname.lastname@example.org
Evaluation of nucleic acid sequence based amplifi cation
using fl uorescence resonance energy transfer
(FRET -NASBA) in quantitative detection of Aspergillus
CHULMIN PARK * ,1 , EUN-YOUNG KWON * ,1 , NA-YOUNG SHIN * , SU-MI CHOI † , SI-HYUN KIM † ,
SUN HEE PARK † , DONG-GUN LEE † , JUNG-HYUN CHOI † & JIN-HONG YOO †
* Catholic Research Institutes of Medical Science, and † Department of Internal Medicine, The Catholic University of Korea,
College of Medicine, Seoul, South Korea
We attempted to apply fl uorescence resonance energy transfer technology to nucleic
acid sequence-based amplifi cation (FRET-NASBA) on the platform of the LightCycler
system to detect Aspergillus species. Primers and probes for the Aspergillus 18S rRNA
were newly designed to avoid overlapping with homologous sequences of human 18s
rRNA. NASBA using molecular beacon (MB) showed non-specifi c results which have
been frequently observed from controls, although it showed higher sensitivity (10 ?2
amol) than the FRET. FRET-NASBA showed a sensitivity of 10 ?1 amol and a high
fi delity of reproducibility from controls. As FRET technology was successfully applied
to the NASBA assay, it could contribute to diverse development of the NASBA assay.
These results suggest that FRET-NASBA could replace previous NASBA techniques in
the detection of Aspergillus .
Keywords Aspergillus , NASBA , FRET
Invasive aspergillosis (IA) is one of the major infectious
complications in patients with hematologic malignancy
who received cytotoxic chemotherapy or transplantation
. Because of its high fatality, early detection is the
key to successful treatment . However, the use of con-
ventional blood culture is almost always negative for
Aspergillus species [2,3]. Therefore, non-culture based
detection methods could become the main diagnostic
tools for IA. Several methods have been evaluated
and used to detect nucleic acids or antigens from
Aspergillus species. For example, galactomannan enzyme
© 2011 ISHAM DOI: 10.3109/13693786.2010.507604
Medical Mycology January 2011, 49, 73–79
© 2011 ISHAM, Medical Mycology, 49, 73–79
Park et al.
Materials and methods
Fungal culture and nucleic acids
A. fumigatus ATCC16424, IFO30870, A. fl avus KCTC6984,
A. terreus ATCC10690, and A. niger ATCC 9029 were
cultured on Sabouraud ’ s dextrose agar at 37 ° C for 2 or
The total RNA and genomic DNA from each strain were
used to construct control RNA employing in vitro tran-
scription and to evaluate the newly designed NASBA. The
genomic DNA and total RNA of each strain were extracted
using MasterPure Yeast DNA purifi cation kit and Master-
Pure RNA purifi cation kit (Epicentre, Wisconsin, USA),
respectively, following the manufacturer ’ s instruction. For
the calculation of cut-off value, nucleic acids were extracted
from 20 blood samples of healthy volunteers using RNeasy
mini kit (Qiagen, GmbH, Germany).
In silico study and the design of primers
The nucleotide sequences of Aspergillus 18S rRNA are
homologous to those of human 18S rRNA. The comparison
of the alignment between them was performed with Clust-
alW2 (http://www.ebi.ac.uk/Tools/clustalw2)  in order
to design primers and probes for Aspergillus 18S rRNA
which could avoid non-specifi c interaction with human
18S rRNA. We designed new primers and probes for
FRET-NASBA of Aspergillus 18S rRNA with Light Cycler
probe design software 2.0 (Roche, Mannheim, Germany)
(Table 1). The nonspecifi c interaction with human 18S
rRNA of newly designed primers for FRET-NASBA and
the previous primers  was tested using in silico PCR
(http://genome.ucsc.edu/) and BLAST (http://blast.ncbi.
In vitro transcription of control RNA from the genomic DNA
of Aspergillus strains
The target DNA template was amplifi ed by PCR with
NASBA primers (Table 1) from the genomic DNA of each
strain. The conditions of PCR were as follows: 5 min at
94 ° C; 30 cycles of 1 min at 94 ° C, 30 sec at 55 ° C, and 30
sec at 72 ° C; and 10 min at 72 ° C. The size of each amplicon
was 161 bps. The amplicons were purifi ed with QIAquick
PCR purifi cation kit (Qiagen, GmbH, Germany). The purity
was examined by agarose gel electrophoresis and spectro-
photometer. The purifi ed amplicons were used as a template
for T7 RNA polymerase. RNA polymerization was per-
formed using MAXIxcript kit (Ambion, Austin, TX, USA).
Briefl y, 1 ug of DNA was added to the reaction mixture
containing 10X buffer and rNTPs and incubated at 37 ° C for
1 h. Then, 1 ul of DNaseI was added to the mixture and
incubated at 37 ° C for 15 min. The reaction was stopped by
adding 1 ul of 0.5M EDTA. Each in vitro transcript was
purifi ed by NucAway Spin Columns (Ambion, Austin, TX,
USA) and each A260 value was calculated based on the
copy number of the transcripts. Samples were diluted ten-
fold, ranging from 1 fmol/ μ l to 0.0001 amol/ μ l (represent-
ing 6 ? 10 8 to 6 ? 10 copies/ μ l).
FRET-NASBA on LightCycler 480 system (LC480)
Except for the manufacturer ’ s enzyme diluent, NASBA
was primarily performed using the Nuclisens Basic Kit
(BioM é rieux Korea, Seoul, Korea), real-time monitored
with the LC480 system (Roche, Mannheim, Germany).
Instead of the manufacturer ’ s enzyme diluent, an in-
house enzyme diluent of 1.5 M Sorbitol and 10 mM
Tris-HCl (pH 8.3) was used in this study. Briefl y, 2 ul
of target RNA was added to each 10 ul of reagent/KCl
(80 mM)/primers and the probes mixture (Table 1;
hybridization probes or molecular beacons, 200 nM
respectively). After that, the reaction mixture was incu-
bated at 65 ° C for 2 min to denature the secondary struc-
ture of RNA and immediately cooled to 41 ° C for 2 min
for annealing primers. Then, 5 ul of enzyme mixture
containing avian myeloblastosis virus (AMV) retrotran-
scriptase, RNase H and T7 RNA polymerase was added
to the mixture. In the case of 2 step NASBA, additional
reverse transcription reactions were performed by
adding 3U of AMV-RT (Takara Korea Seoul, Korea)
Table 1 Primers and probes for NASBA used in this study.
Primers and probes
LC Red 640-TCTTAACCATAAACTATGCCGACTAGGGATCGG-Phosphate
Loeffl er et al.
Loeffl er et al.
Yoo et al.
aRelative to accession no. AB008401.
*The underlined indicates T7 promoter sequences.
© 2011 ISHAM, Medical Mycology, 49, 73–79
FRET-NASBA for the detection of Aspergillus 75
Assessing the previous Aspergillus 18S
rRNA MB-NASBA on the platform of the
The previous MB-NASBA  was tried on the platform of
the LC480 system. The cut-off value for TTP of MB-
NASBA was determined to be 1.571 on the LC 480 system.
The coeffi cient of determination ( R 2 ) of the linear relation-
ship between TTP and the transcript using 10 3 to 10 8 cop-
ies was 0.940 (Fig. 2A). MB-NASBA showed 10 -2 amol/ μ l
(6 ? 10 3 copies/ μ l) of the 100% sensitivity. These results
suggest that quantitative NASBA (Q-NASBA) could work
on the LightCycler system on which diverse wavelengths
could be detected.
The non-specifi c reaction of MB-and FRET-NASBA on the
MB-NASBA with negative controls (35 samples) and healthy
volunteer blood samples (20 samples) showed fl uorescence
values with a wide range (0.564 – 8.881, mean of 1.990, and
standard deviation of 2.108) (Fig. 2B). These suggest that
MB-NASBA may cause non-specifi c reactions or may be
responsible for great variability between replicate samples.
FRET-NASBA with negative controls and healthy volunteer
blood samples (total 55 samples) showed fl uorescence val-
ues with a narrower range (0.104 – 0.210, mean of 0.128, and
standard deviation of 0.025) than MB-NASBA (Fig. 2C).
Assessing new Aspergillus 18S rRNA FRET-NASBA on the
platform of LC480 system
We designed a new Aspergillus 18S rRNA NASBA moni-
tored by FRET with hybridization probes to increase the
accuracy or decrease non-specifi c results. The primers and
probes in this assay were designed to avoid possible binding
to human transcripts (Table 1). For real-time monitoring, it
to primers-templates mixture of NASBA at 41 ° C to
increase the sensitivity before the NASBA reaction
(65 ° C reaction). The reaction was performed at 41 ° C on
a LightCycler 480 system (Roche, Mannheim, Germany)
using 90 cycles with 1 min acquisition at a 610-nm
The cut-off values of MB- and FRET-NASBA were
determined to be 1.671 times and 2.019 times the mean of
the negative controls, respectively, according to the manu-
facturer ’ s instructions. Time to positivity (TTP) was defi ned
as the time at which the target and IC amplifi cation curves
passed the cut-off values.
Comparison of RNA copies by FRET-NASBA with those by
MB-NASBA on clinical specimen
Seventeen specimens (whole blood) obtained from patients
with probable invasive aspergillosis were tested for the
FRET- and MB-NASBA. RNA copies were calculated and
correlation coeffi cient between these two methods was
The analysis of primers and probes in the previous
Aspergillus 18S rRNA NASBA with in silico study
The alignment between the nucleotide sequences of human
and Aspergillus 18s rRNA by ClustalW2, showed 75.6%
of homology. But Asp2.1 and Asp1.2 primers  were
highly homologous to those of human 18s rRNA, with
identical 3 ′ ends (Fig. 1, ref|NR_003286.1|). Moreover, in
silico PCR results (http://genome.ucsc.edu/) showed that
they could be highly homologous to those of other parts of
the human genome (Fig. 1. chromosome Y and 2). Asper-
gillus 18S rRNA NASBA has been specially designed for
the diagnosis of aspergillosis.
Fig. 1 The homology of the previous NASBA primers to the nucleotide sequences of Aspergillus fumigatus (AB008401), human genome (chrY:10628845-
10629047 and chr2:132846672-132846880), and transcript (ref|NR_003286.1|) from in silico studies. It was performed with the comprehensive analysis
of ClustalW2 (http://www.ebi.ac.uk/Tools/clustalw2) in silico PCR (http://genome.ucsc.edu/) and BLAST (http://blast.ncbi.nlm.nih.gov).
© 2011 ISHAM, Medical Mycology, 49, 73–79
Park et al.
was assayed on the LC480 system (Roche) which has a
wide fl uorescence spectrum and a high sensitivity. The con-
centration of in vitro 18s rRNA transcripts was approxi-
mately 1.5 ? 10 12 copies/ μ l. Diluted transcripts (10 8 – 10 1
copies/ μ l) were used as templates for FRET-NASBA. The
cut-off value for the time to positivity (TTP) of FRET-
NASBA was determined to be 0.260. The R 2 of the linear
relationship between TTP and the transcript using 10 4
to 10 8 copies was 0.934 (Fig. 3A). FRET-NASBA showed
10 -1 amol/ μ l (6 ? 10 4 copies/ μ l) of the 100% sensitivity.
A standard curve for the conidia number (10 4 – 10 8 conidia)
of A. fumigatus showed a good fi t with R 2 values of 0.9551
(Fig. 3B) and R 2 of standard curves of A. fl avus , A. terreus ,
and A. niger were 0.7798, 0.9539, and 0.8695, respectively
(Fig. 3C, 3D, and 3E).
Correlation between MB- and FRET-NASBA
RNA copies were calculated from each TTP values
using both methods. As Fig. 4 showed, there was a good
correlation (r ? 0.825, P ? 0.01) between MB- and
NASBA is an isothermal assay which selectively
amplifi es target RNA through the action of multiple
enzymes (avian myeloblastosis virus reverse transcriptase,
AMV-RT, T7 RNA polymerase, and RNase H) . NASBA
has been used with end-point detection, by enhanced
chemiluminescence (ECL), or real-time detection by
molecular beacon (MB), which can be used as a quantifi ca-
tion assay [7,9 – 13].
Quantitative Aspergillus 18S rRNA NASBA has usually
employed MBs which are hairpin stem-loop probes labeled
with a fl uorophore and a quencher at either end. When an
MB hybridizes with a target nucleic acid, the fl uorophore
separates from the quencher and emits fl uorescence.
In other probe systems, the fl uorescence resonance
energy transfer (FRET) technology using hybridization
Fig. 2 The standard curve analysis of MB-NASBA using in vitro transcripts (A) and the fl uorescence values of negative controls (n ? 20) and healthy
volunteers (n ? 35) blood samples using MB-NASBA (B); Fluorescence values of negative controls (n ? 20) and healthy volunteer (n ? 35) blood
samples using FRET-NASBA (C).
© 2011 ISHAM, Medical Mycology, 49, 73–79
FRET-NASBA for the detection of Aspergillus 77
probes usually relies on two fl uorescence probes, a donor
and an acceptor. When they hybridize together to targets
in the strict distance (1 – 5 bases), an energy transfer occurs
between the fl uorescent dyes, leading to an emission of
fl uorescent light from the acceptor probe. As hybridization
probes can hybridize more sequences than single probe
systems, they could theoretically be expected to increase
the specifi city of the test.
Therefore, we attempted to apply FRET technology to
NASBA (FRET-NASBA), and compare it to real time
monitoring by MBs (MB-NASBA). Also, we used the
LightCycler system (Roche Diagnostics, Mannheim,
Germany) as the universal platform for NASBA.
There has been a previous trial for the application of
NASBA real-time monitoring by MBs on the LightCycler
system . As the LightCycler system is able to adopt
diverse probes and is used in many laboratories, we inves-
tigated whether the system could be employed universally
as a platform for Aspergillus Q-NASBA.
The variability or unexpected results of Q-NASBA
between replicate samples is often observed to be greater
than that seen in PCR-based assays. So, for precision and
Fig. 3 The standard curve analysis of FRET-NASBA using in vitro transcripts (A), nucleic acids from Aspergillus fumigatus ATCC16424 conidia (B),
Aspergillus fl avus KCTC6984 (C), Aspergillus terreus ATCC10690 (D), and A. niger ATCC 9029 (E).
© 2011 ISHAM, Medical Mycology, 49, 73–79
Park et al.
accuracy in predicting unknown concentrations of target
RNA, the quantifi cation of NASBA has typically been per-
formed with an external known standard curve , or
with the TTP ratio using internal controls . But, our
previous study involving one tube NASBA assay with
internal controls showed that it is diffi cult to apply to the
previous Aspergillus 18s rRNA NASBA in which target
NASBA reactions interfered with the reaction of internal
controls . Hence, we adopted NASBA assay with
external standard curve analysis in this study.
For Aspergillu s diagnostics using NASBA, MBs have
been employed in real time monitoring [5,17]. Despite the
usefulness of MBs, it has also been reported that they could
be disrupted due to degradation by nucleases, which could
in turn result in false positive signals due to non-specifi c
interactions [18,19]. T7 RNA polymerase was also demon-
strated to possess nuclease activity against the oligomer of
DNA and RNA . This suggests that real time monitor-
ing with MBs in NASBA might carry the risk of false
positive results in clinical diagnostics. For the results of
MB-NASBA with negative controls (Fig. 2B), we suggest
that the false positivity (or the possibility of artifi cially
increased fl uorescence at the end of the test) may be due
to the above-mentioned risk of MBs, i.e., they could be
easily disrupted by inherent nuclease activity [18,19].
Aspergillus 18S rRNA NASBA has been specifi cally
designed for the diagnosis of aspergillosis. As clinical
specimens usually contain human RNA, if primers or
probes of NASBA are homologous to human mRNA or
other transcripts, reaction interference could occur. It could
result in false negativity, especially at low concentration as
in most of clinical specimens. The lower annealing tem-
perature of NASBA than that of PCR could also cause the
false negativity. Hence, we have designed a new set of
primers and probe which could avoid nonspecifi c interac-
tion with human 18S rRNA as shown in Table 1.
FRET-NASBA could provide some benefi ts in the
development of other NASBA diagnostics, e.g., for viral
and fungal diseases which are diffi cult to diagnose [9 – 15].
To increase the performance of NASBA, other probes such
as 2 ′ - O -Methyl MBs resistant to linear nuclease  or
simple (one) probe systems using FRET will also be fur-
ther analyzed by NASBA assay.
Although the comparison of RNA copies by FRET- with
MB-NASBA showed a good correlation in this study, it
was rather a minimum evaluation of clinical specimens. To
determine the usefulness of FRET-NASBA as an alterna-
tive in the diagnosis of invasive aspergillosis, further study
using more clinical specimens is necessary. Hence, we are
planning to perform a further study on clinical cases to
evaluate whether FRET-NASBA could be practically used
to predict clinical outcomes and be applied to the clinical
diagnostics of invasive aspergillosis.
We thought that our results expanded the practical spec-
trum of NASBA through the utilization of another probe
system like FRET. Moreover, the performance of NASBA
for Aspergillus detection on the LightCycler system could
be more practical for many laboratories which are not
equipped with specialized NASBA instruments.
This study was supported by a grant (A080696) of the
Korea Healthcare technology R&D Project, Ministry for
Health, Welfare & Family Affairs, Republic of Korea.
Declaration of interest: The authors report no confl ict of
interest. The authors alone are responsible for the content
and writing of the paper.
1 van den Bergh MF, Verweij PE, Voss A. Epidemiology of nosocomial
fungal infections: invasive aspergillosis and the environment. Diagn
Microbiol Infect Dis 1999; 34 : 221 – 227.
Denning DW. Invasive aspergillosis. 2
781 – 805.
Yeo SF, Wong B. Current status of nonculture methods for diagnosis 3
of invasive fungal infections. Clin Microbiol Rev 2002; 15 : 465 – 484.
Yoo JH, Choi JH, Choi SM, 4
et al . Application of nucleic acid
sequence-based amplifi cation for diagnosis of and monitoring the
clinical course of invasive aspergillosis in patients with hematologic
diseases. Clin Infect Dis 2005; 40 :392 – 398.
Yoo JH, Choi SM, Lee DG, 5
et al . Comparison of the real-time nucleic
acid sequence-based amplifi cation (RTi-NASBA) with conventional
NASBA, and galactomannan assay for the diagnosis of invasive
aspergillosis. J Kor Med Sci 2007; 22 : 672 – 676.
Larkin MA, Blackshields G, Brown NP, 6
version 2. Bioinformatics 2007; 23 :2947 – 2948.
Loeffl er J, Hebart H, Cox P, 7
et al . Nucleic acid sequence-based ampli-
fi cation of Aspergillus RNA in blood samples. J Clin Microbiol 2001;
39 : 1626 – 1629.
Clin Infect Dis 1998; 26 :
et al . ClustalW2 and ClustalX
Fig. 4 Comparison of RNA copies measured by FRET-NASBA with
those by MB-NASBA performed on clinical specimens.
© 2011 ISHAM, Medical Mycology, 49, 73–79 Download full-text
FRET-NASBA for the detection of Aspergillus 79
Compton J. Nucleic acid sequence-based amplifi cation.
350 : 91 – 92.
Casper ET, Paul JH, Smith MC, Gray M. Detection and quantifi cation
of the red tide dinofl agellate Karenia brevis by real-time NASBA.
Appl Environ Microbiol 2004; 70 : 4727 – 4732.
Hibbits S, Rahman A, John R, Westmoreland D, Fox JD. Development
and evaluation of Nuclisens basic kit NASBA for diagnosis of parain-
fl uenza virus infection with end-point and real-time detection. J Virol
Methods 2003; 108 : 145 – 155.
Lanicotti RS, Kerst AJ. Nucleic acid sequence-based amplifi cation
assays for rapid detection of West Nile and St Louis Encephalitis
viruses. J Clin Microbiol 2001; 39 : 4506 – 4513.
Leone G, van Schijndel H, van Gemen B, Kramer FR, Schoen CD.
Molecular beacon probes combined with amplifi cation by NASBA
enable homogeneous, real-time detection of RNA. Nucl Acids Res
1998; 26 : 2150 – 2155.
Widjojoatmodjo MN, Borst A, Schukkink RA,
sequence-based analysis (NASBA) detection of medically important
Candida species. J Microbiol Methods 1999; 38 : 81 – 90.
Costa AM, Lamb D, Garland SM, Tabrizi SN. Evaluation of Light-
Cycler as a platform for nucleic acid sequence-based amplifi cation
(NASBA) in real-time detection of enteroviruses. Curr Microbiol
2008; 56 : 80 – 83.
et al . Nucleic acid
Polstra AM, Goudsmit J, Cornelissen M. Development of real-time
NASBA assays with molecular beacon detection to quantify mRNA
coding for HHV-8 lytic and latent genes. BMC Infect Dis 2002; 2 : 18.
Patterson SS, Casper ET, Garcia-Rubio L, Smith MC, Paul JH
3rd. Increased precision of microbial RNA quantifi cation using
NASBA with an internal control . J Microbiol Methods 2005; 60 :
343 – 352.
Yoo JH, Choi SM, Choi JH,
et al . Construction of internal control for
the quantitative assay of Aspergillus fumigatus using real-time nucleic
acid sequence-based amplifi cation. Diagn Microbiol Infect Dis 2008;
60 : 121 – 124.
Li JJ, Geyer R, Tan W. Using molecular beacons as a sensitive fl uo-
rescence assay for enzymatic cleavage of single-stranded DNA. Nucl
Acids Res 2000; 28 : e52.
Nilsson M, Gullberg M, Dahl F, Szuhai K, Raap AK. Real-time moni-
toring of rolling-circle amplifi cation using a modifi ed molecular bea-
con design. Nucl Acids Res 2002; 30 : e66 .
Sastry SS, Ross BM. Nuclease activity of T7 RNA polymerase and
the heterogeneity of transcription elongation complexes. J Biol Chem
1997; 272 : 8644 – 8652.
Chen AK, Behlke MA, Tsourkas A. Avoiding false-positive signals
with nuclease-vulnerable molecular beacons in single living cells.
Nucl Acids Res 2007; 35 : e105.
This paper was fi rst published online on Early Online on 20 August