Lab on a Chip
View Article Online
PAPER
Published on 17 February 2015. Downloaded by DTU Library on 26/02/2015 10:34:45.
Cite this: DOI: 10.1039/c4lc01459f
View Journal
A lab-on-a-chip system with integrated sample
preparation and loop-mediated isothermal
amplification for rapid and quantitative detection
of Salmonella spp. in food samples
Yi Sun,†a Than Linh Quyen,†b Tran Quang Hung,†b Wai Hoe Chin,†b Anders Wolff†a
and Dang Duong Bang†*b
Foodborne disease is a major public health threat worldwide. Salmonellosis, an infectious disease caused
by Salmonella spp., is one of the most common foodborne diseases. Isolation and identification of
Salmonella by conventional bacterial culture or molecular-based methods are time consuming and usually
take a few hours to days to complete. In response to the demand for rapid on line or on site detection of
pathogens, in this study, we describe for the first time an eight-chamber lab-on-a-chip (LOC) system with
integrated magnetic bead-based sample preparation and loop-mediated isothermal amplification (LAMP)
for rapid and quantitative detection of Salmonella spp. in food samples. The whole diagnostic procedures
including DNA isolation, isothermal amplification, and real-time detection were accomplished in a single
Received 12th December 2014,
Accepted 12th February 2015
DOI: 10.1039/c4lc01459f
www.rsc.org/loc
chamber. Up to eight samples could be handled simultaneously and the system was capable to detect
Salmonella at concentration of 50 cells per test within 40 min. The simple design, together with high level
of integration, isothermal amplification, and quantitative analysis of multiple samples in short time, will
greatly enhance the practical applicability of the LOC system for rapid on-site screening of Salmonella for
applications in food safety control, environmental surveillance, and clinical diagnostics.
Introduction
Food safety remains a major public health threat in both
developing and developed countries. Salmonellosis, an infectious disease caused by Salmonella spp., is one of the most
common foodborne diseases worldwide. It is estimated that
there are at least 100 000 salmonellosis cases per year in the
EU countries,1 and approximately 1 million cases in the
United States.2 Symptoms of Salmonella infection may vary
from gastroenteritis and diarrhea to severe life-threatening
systemic diseases such as typhoid fever. Salmonellosis has an
impact on society in terms of morbidity, healthcare costs, and
loss of productivities. As Salmonella is frequently found in
large mammals, avian, reptiles and also in raw food products
that come from animals, monitoring and surveillance control
a
Department of Micro and Nanotechnology (DTU-Nanotech), Technical University
of Denmark, Ørsteds Plads, DK-2800 Kgs. Lyngby, Denmark
b
Division of Food Microbiology, National Food Institute (DTU-Food), Technical
University of Denmark, Mørkhøj Bygade 19, DK-2860 Søborg, Denmark.
E-mail: ddba@food.dtu.dk; Fax: +45 35 88 70 01; Tel: +45 38 88 73 82
† Authors' contributions: Dang Duong Bang, Yi Sun, and Anders Wolff designed
the research plan; Yi Sun, Than Linh Quyen, Tran Quang Hung and Wai Hoe
Chin performed the experiments; Yi Sun, Than Linh Quyen and Dang Duong
Bang wrote the paper.
This journal is © The Royal Society of Chemistry 2015
of Salmonella in food and food production lines are extremely
important for food safety.3
To date, the conventional bacterial culture method is still
used as the main workhorse and reference method for food
safety control.4 The methods involve several steps to obtain
the final result: pre-enrichment, selective enrichment, isolation on selective agar, and serological and biochemical confirmation. The whole process usually takes 5 to 7 days to be
completed. In the last few years, molecular diagnostic
methods such as polymerase chain reaction (PCR) have been
developed for the detection of Salmonella in much shorter
time.5 PCR allows for an accurate and unambiguous identification of target nucleic acid sequences. The detection limit of
the method is 10 to 100 copies of purified genomic DNA and
103 to 104 CFU mL−1 for cultured samples.6 Nevertheless, the
use of PCR is limited by the thermal constraints. The PCR
process needs an electrically powered thermal cycler with precise three stage temperature control and a fast transition
between stages, which is usually accomplished by a bulky
and power-intensive apparatus. Moreover, the whole procedures
are costly, laborious, and prone to errors and contaminants.7
Recently, an alternative approach the so called isothermal
amplification has been developed to overcome the drawbacks
of PCR.8 The use of other polymerases allows nucleic acid
Lab Chip
View Article Online
Published on 17 February 2015. Downloaded by DTU Library on 26/02/2015 10:34:45.
Paper
amplification at constant and low temperature. There are a
number of different isothermal amplification techniques,
and among them, loop-mediated isothermal amplification
(LAMP) has attracted considerable interest due to its rapid
amplification, simple operation, and high sensitivity and
specificity.9 LAMP is characterized by the use of four different
primers, specifically designed to recognize six distinct
regions on the target gene, which makes it highly specific for
its target even in the presence of high concentration of nontarget DNA. The process is performed at a constant temperature (60–65 °C) using a strand displacement reaction, thus
obviating the demand for sophisticated thermal control. Fast
and efficient amplification can be achieved since there is no
time required for temperature ramping during the LAMP process. Owing to these advantages, the LAMP technology has
been widely applied for rapid detection of different pathogens such as Salmonella spp., Campylobacter spp., Escherichia
coli, Staphylococcus aureus, etc.10 A LAMP assay developed to
detect Salmonella with sensitivity of 60 CFU per test has
recently been reported.11
In order to comply with the demands of consumers for
safe, pathogen-free food, there is an urgent need for development of a rapid and reliable method for on line or on site
Salmonella detection. Adapting LAMP to lab-on-a-chip (LOC)
systems seems to be a very promising approach for monitoring pathogen at point-of-care (POC) settings.12,13 Tourlousse
et al. fabricated a chip containing 15 interconnected reaction
wells with dehydrated primers for LAMP. The chip is
designed to handle genomic DNA and could detect multiple
pathogens, including Salmonella, Campylobacter, Shigella and
Vibrio cholerae, in less than 20 min.14 A sensitivity of 10–100
gene copies per test was achieved by using real-time fluorescence detection. Fang et al. reported another micro-LAMP
system which adapted LAMP to a microfluidic chip for
Pseudorabies virus analysis.15 The readout was obtained via
absorbance measurement by an optic sensor, giving a detection limit of 10 fg DNA μl−1. The two systems demonstrated
the feasibility of integrating LAMP on microfluidic chips for
quantitative detection of pathogens. However, for these systems, both tedious sample preparation and nucleic acid purification processes were performed off-chip prior to the LAMP
reaction. Wang et al. reported a new diagnostic assay by combining nucleic acid extraction and LAMP in a magnetic beadbased microfluidic system for detection of Staphylococcus
aureus.16 The LAMP amplified products were analyzed by a
spectrophotometer and the limit of detection was approximately 10 CFU μl−1. Although in this work the sample preparation was integrated on chip, the complexity of system
design and operation posed difficulties for the system to process more than one samples. The chip consisted of multilayer PDMS structures and a glass substrate with metal
electrodes, and required plasma treatment for bonding. The
fabrication process was complicated and costly, therefore not
suitable for mass production. Moreover, absorption measurement at end point was inadequate for accurate quantitative
analysis.
Lab Chip
Lab on a Chip
Despite the fact that various LAMP-based LOC devices
have been developed for rapid detection and identification of
foodborne pathogens, portable platforms for on-site testing
are still in the infancy stage.17 Fully integrated LOC systems
for multiple sample detection have seldom been reported. In
this paper, we describe for the first time, a LOC system with
integrated sample preparation and LAMP for rapid and realtime detection of Salmonella spp. in multiple samples. All of
the steps from sample preparation, DNA purification to
LAMP amplification were performed in one micro-chamber,
which significantly simplified chip design and thus allowed
for parallel analysis. By interfacing with an eight-channel
peristaltic micropump, the developed microchip was capable
of handling eight samples simultaneously. The real-time
detection of LAMP products were achieved by using an appropriate fluorescence dye, which provided accurate quantification of Salmonella presented in different samples. The whole
process took less than 40 min, and the detection limit of the
system was 10 cells μl−1. The microchip was made in low-cost
polymeric material by injection molding which can be upscaled for industrial production. The developed LOC system
is suitable for on-site rapid detection of pathogens in food,
environmental or clinical samples.
Experimental
Design and fabrication of the LOC system
The LOC system consisted of a disposable microfluidic cyclic
olefin copolymer (COC) chip with eight chambers, and a
reusable actuation unit with magnets, an eight-channel
pump, a heater for LAMP amplification, and a small ESE log
detector for real-time fluorescence measurement. Fig. 1
shows the details of the device. The microchip consisted of
eight chambers in a microscopy slide format and was inhouse fabricated by injection molding (ENGEL, PA, USA). The
master was made in aluminum using computer controlled
milling (Folken Ind, Glendale, California, USA). Each chamber
had dimensions of 5 mm (length) × 4 mm (width) × 0.5 mm
(height), corresponding to a volume of 10 μL. Microfluidic
channels with widths of 500 μm and depths of 200 μm led
from the chambers and connected these to 0.8 mm diameter
inlet and outlet through holes. The COC slide was lidded
with a 200 μm-thin COC film using a bonding press
IJP/O/Weber, Remshalden, Germany) at 120 °C and 2 bar for
10 min.
To handle multiple samples simultaneously, the COC
microchip was interfaced to the eight-channel MainSTREAM
micropump (Fig. 1a). The details of micropump fabrication
were described elsewhere.18 Briefly, the peristaltic micropump
had three core components: a monolithic PDMS pumping
inlay (PI) containing 8 integrated channels, a rotor bed (RB),
and a multi-roller (MR). The PI was placed in the RB complementary in shape to the MR. The MR was driven by a stepper
motor (Lego A/S, Denmark) which was originally designed
for Lego Mindstorms robot. It comes with a programmable
This journal is © The Royal Society of Chemistry 2015
View Article Online
Lab on a Chip
Paper
loaded to the micro-chambers through the plastic tubes
connected at the inlets of the chamber chip. During LAMP
amplification, the plastic tubes were removed and the holes
were sealed with PCR tape. In this system, only the eightchamber microchip and the inlet plastic tubing were disposable, while the PDMS inlay and the chip containing waste reservoirs were re-usable.
Published on 17 February 2015. Downloaded by DTU Library on 26/02/2015 10:34:45.
Chemicals
All chemicals and reagents used in this study were of analytical grade and purchased from Pierce Inc., USA or SigmaAldrich, USA unless otherwise specified.
Bacterial strains and culture conditions
Salmonella enterica reference strain CCUG – 32352 originated
from the Culture Collection, University of Göteborg Sweden
was provided by the National Food Institute, Technical University of Denmark (DTU-Food). The strain was resuscitated
and selected on Xylose Lysine Deoxycholate Agar (XLD) and
grown overnight on Blood Agar (BA, 40 g l−1 blood agar base
no. 2 (CM271, Oxoid, Basingstoke, UK) supplemented with
5% calf blood at 37 °C before use).
DNA preparation
Fig. 1 (a) Integration of the COC microchip with the MainSTREAM
micropump. (b) System overview. The LOC system consists of an
eight-chamber microfluidic biochip made by injection molded on a
COC slide, an external heater element was mounted under the microchip and a plastic frame containing eight magnetic elements was
clamped on top of the chips for on chip sample preparation and LAMP
amplification. An eight-channel peristaltic micropump was connected
to the eight-chamber microfluidic chip. The samples and reagents
were loaded to the micro-chambers through the plastic tubes
connected at the inlets. The reservoirs at outlets were used to store
waste. Mixtures of BPW enriched pork meat sample and magnetic
beads were pumped into each chamber at a flow rate of 50 μl per
minute.
brick containing control software, USB communication port
and rechargeable battery. As the MR rotated, it contacted and
compressed the PI channels and pushed fluid in the direction of rotation. The micropump could also function as a
microvalve when MR was kept static. The micropump/valve
system was connected to the microfluidic chip through the
PDMS inlay to address all the steps required for multiplesample processing, including DNA purification and LAMP
amplification. The other end of the PDMS inlay was coupled
to waste reservoirs milled in a PMMA chip.
The system provided very easy world-to-chip interfaces
(Fig. 1b). The samples, washing fluid and LAMP reaction mixture were simply placed in three rows of eight PCR tubes and
This journal is © The Royal Society of Chemistry 2015
The chromosomal DNA from S. enterica used for the sensitivity study was isolated using DNeasy Blood and Tissue kit
(Qiagen, Germany). The DNA concentration was determined
by Nano drop (Thermo Scientific, USA). Two ng μl−1 S.
enterica genomic DNA was prepared to test the performance
of LAMP using a bench-top thermal cycler.
Spiked BPW enriched pork meat samples
Pig meat samples (1 : 10) enriched in Buffered Peptone Water
(BPW) were collected from slaughters. At the slaughter, the
samples were tested for the presence of Salmonella using
both conventional culture and molecular methods as
described in the literature.19 On arrival, the Salmonella status
of the BPW enriched pork meat samples was confirmed by
TaqMan real-time PCR.
Briefly, 1 ml of the top phase of the BPW was collected
and centrifuged at 3000 × g for 5 minutes and the supernatant was discarded. Next, the pellet was re-suspended in
200 μl of 1× Tris-EDTA buffer and was heated at 96 °C for
10 minutes for DNA extraction. TaqMan real-time PCR
targeting at ttrRSBCA locus was used to detect Salmonella
spp. All the negative samples were collected and stored at
4 °C for later use.
To prepare Salmonella-spiked samples, a serial 10-fold
dilution of Salmonella in 1 ml of saline water (0.9% NaCl)
was prepared from a S. enterica CCUG 32352 stock (OD600 =
0.8, corresponding to 108 cells ml−1). Ten μl of Salmonella of
each dilution was added to 90 μl of the enriched pork meat
juice to give a final concentration of Salmonella ranging from
100 to 104 cells μl−1. The Salmonella-spiked BPW enriched
Lab Chip
View Article Online
Paper
pork meat samples were used to test and evaluate the performance of the LOC system. A non-spiked sample was also
included in the experiment as a negative control.
Published on 17 February 2015. Downloaded by DTU Library on 26/02/2015 10:34:45.
LAMP using a conventional thermal cycler
In-tube LAMP reactions were carried out using a conventional
benchtop real-time PCR thermal cycler (ThermoFisher Scientific, MA, USA) at a constant temperature of 65 °C for
30 minutes. The reactions were terminated by heating up to
80 °C for 2 minutes. Loopamp Salmonella Detection Kit from
Eiken Chemical Co. Ltd (Tokyo, Japan) was used for the
experiments. The kit consists of a set of primers specifically
designed for Salmonella invasion gene invA (Eiken Chemical
Co. Ltd, Tokyo, Japan).
The LAMP master mixture contained 9.5 μl of Reaction
Mix. Sal and 0.5 μl of Bst DNA polymerase. One μl of Salmonella DNA (2 ng μl−1) was mixed with the LAMP master mixture. In initial experiments, different DNA intercalating dyes,
such as Eva Green, SYBR Green I, SYTO-26, SYTO-62 and
SYTO-82, were added to LAMP reactions to test the possibility
to use the dyes for real-time LAMP detection. All dyes were
purchased from Invitrogen (CA, USA). One μl of fluorescent
dye (5 μM) was added to the LAMP mixture and fluorescence
signals were recorded with 1 min intervals. End-point analysis of LAMP products was done by using 2% agarose gel
electrophoresis with 1× TAE buffer.
Study of the inhibitory effect of magnetic beads on LAMP
Total volumes of 20 μl, 50 μl, 80 μl, 100 μl, 150 μl and 200 μl
of magnetic bead solution were taken from Dynabeads® DNA
DIRECT™ Universal kit (d = 2.8 μm, approximately 2 × 109
beads ml−1, ThermoFisher Scientific, MA, USA) and transferred to PCR tubes. The respective tubes were centrifuged at
5000 × g for 5 min and the lysis solution was discarded.
Beads in all tubes were washed three times with sterile water.
After washing, 10 μl of LAMP master mixture and 1 μl of 2 ng μl−1
Salmonella DNA were added into the tubes and LAMP was
performed under the above mentioned conditions. The inhibition
effect of the magnetic beads was assessed by gel electrophoresis.
On-chip detection of Salmonella using spiked samples
The feasibility of the LOC system for rapid and parallel detection of Salmonella was demonstrated using 10-fold serial dilutions of the Salmonella spiked BPW enriched pork meat samples. By integrating sample preparation and real-time LAMP
(rtLAMP) amplification in one chamber, the LOC was able to
handle eight samples at one time. The procedures are as follows. Five μl of spiked samples were mixed with 20 μl of magnetic beads suspended in lysis buffer (Dynabeads® DNA
DIRECT™ Universal kit, ThermoFisher Scientific, MA, USA).
The mixtures were transferred to the sample reservoirs on the
microchip and pumped through the microchambers using
the eight-channel micropump with a flow rate of 50 μl min−1.
The DNA bound magnetic beads were then captured by the
magnets situated below the microchip. Right after, 200 μl of
Lab Chip
Lab on a Chip
washing buffer was pumped through each microchamber to
remove any residual contaminants and potential rtLAMP
inhibitors in the BPW enriched pork meat samples. Next the
micro-chambers were dried by pumping air in for 1 min. The
chambers were then filled up with the LAMP reaction mixture
(10 μl LAMP master mixture and 1 μl SYTO-62), and LAMP
amplification was carried out by heating up the microchambers to 65 °C for 40 min. Real-time monitoring of eight
samples was accomplished by sequentially moving the ESE
log detector (Qiagen, Germany) onto each microchamber.
Fluorescence signals at 670 nm were recorded with 1 min
intervals. For quantification, a concept of threshold time (Tt),
which is similar to the threshold cycle (Ct) in rtPCR, was
applied. The Tt was defined as the interpolated time at which
the baseline-subtracted signal was equal to 10 times the standard deviations of the baseline signal, which is equivalent to
a signal-to-noise ratio (SNR) of ten.14 Standard curves were
plotted for the dilution series.
Results and discussion
Comparison of multiple DNA intercalating dyes
for real-time LAMP amplification
Several studies have reported the use of LAMP, and various detection methods have been developed to identify the
amplified LAMP products. In most cases, LAMP was applied
for pathogen screening, and a “yes-or-no” answer was adequate. For such applications, the simplest method for judging a positive or negative LAMP reaction is to assess the turbidity of the solution by the naked eye.20 For better visibility,
a dye such as SYBR green21 is added to the solution after the
reaction is complete, and a colour change is observed under
UV light for a positive LAMP reaction. These methods are
very cost-effective, however, they are limited by the low sensitivity. In order to accurately quantify the amount of the pathogen, the reaction should be followed in real-time. This is
possible by using a real-time turbidimeter22 or by measuring
the signals from DNA produced via fluorescent intercalating
dyes.21,23 Fluorescence detection is by far the most sensitive
method. However, in a previous study, the intercalating dyes
were shown to partially or completely inhibit PCR-based
amplification, which compromised the advantages of using
fluorescence measurement.24 To our best knowledge, only a
limited number of intercalating dyes such as SYTO 9,21 SYTO
82 (ref. 14) or SYBR Green21 have been used in rtLAMP. However, in all these studies, the inhibitory effects of the dyes on
LAMP have not been evaluated. In order to select suitable dye
candidates for rtLAMP detection, we screened six DNA intercalating dyes from different dye families and with a wide
range of optical properties.
The inhibition effects of the respective dyes were investigated by carrying out LAMP using a conventional real-time
PCR machine. Fig. 2 compares the performance of various
intercalating dyes. From the real-time amplification curves
(Fig. 2a), the Tt value was determined and used as an
This journal is © The Royal Society of Chemistry 2015
View Article Online
Lab on a Chip
Paper
Published on 17 February 2015. Downloaded by DTU Library on 26/02/2015 10:34:45.
amplified product was observed. An appropriate intercalating
dye is a fairly important component of real-time LAMP
assays, as it could increase the sensitivity and reduce the
false positive and false negative rates for pathogen detection.
Our results strongly suggest that Eva Green, SYTO-24, SYBR
Green and SYBR Safe are not suitable for use in rtLAMP.
SYTO-62 and SYTO-82 are the good candidates, but care must
be taken for the high background noise of SYTO-82. SYTO-62
was therefore used for rtLAMP for the rest of the study in this
paper.
Investigation of the inhibitory effect of magnetic
beads on LAMP
Fig. 2 Comparison of the effect of DNA intercalating dyes on LAMP
efficiency. The reaction mixture consisted of 10 μl of LAMP master
mixture, 1 μl of Salmonella DNA template (2 ng μl−1) and 1 μl of each
fluorescence dye (5 μM). LAMP was performed on a conventional realtime PCR machine at 65 °C for 40 min. (a) Amplification curves in real
time. (b) End point analysis of LAMP products using 2% agarose gel
electrophoresis at 100 V for 45 min. Lane 1: negative control; lane 2:
positive control (without addition of dye to the LAMP reaction media);
lane 3: SYTO 24; lane 4: SYTO 62; lane 5: SYTO 82; lane 6: SYBR
Green; lane 7: Eva Green; lane 8: SYBR Safe.
indicator of inhibitory effects on LAMP efficiency, since significant decreases in efficiency would result in increased Tt
values. The dyes could be divided into three classes according
to the degree of inhibition. SYTO-62 and SYTO-82 had a very
similar Tt value of 13 min, indicating that the dyes had the
least effect on LAMP efficiency. Eva Green showed an
increased Tt value of 16 min, suggesting that it partially
inhibited LAMP reaction. In contrast, SYTO-24, SYBR Green I
and SYBR Safe showed clear negative effects on LAMP as no
Tt could be determined. The effects of the respective dyes on
LAMP were further confirmed by gel electrophoresis (Fig. 2b).
Compared to the positive control, SYTO-62 demonstrated no
change in the LAMP product, whereas SYTO-82 showed very
high background noise, suggesting that lower dye concentration should be used. Eva Green showed a large decrease in
the amount of LAMP product, while for other dyes with high
inhibition effect (SYTO-24, SYBR Green and SYBR Safe), no
This journal is © The Royal Society of Chemistry 2015
Recently, a few LOCs have been developed for pathogen
detection and some involved complex operations including
total sample preparation processes, amplification, and finally
detection.25 However, to address the multiple steps involved
in nucleic acid extraction and to avoid possible contamination to the downstream amplification, most of the published
systems tended to perform DNA purification and amplification in two separate chambers.26,27 As a consequence, multiple pumps and valves were required to precisely control fluid
transfer, which greatly increased the complexity of system
design and operation and hence limited the practical applicability of the LOC systems. Moreover, such a two-chamber system also poses difficulties in processing more than one sample due to the constraints of the size and the cost of the
control elements. One efficient way to avoid these limitations
is to combine sample preparation and LAMP amplification in
a single chamber, and this requires a simple and LAMPfriendly DNA extraction method. Magnetic beads have proven
to be effective in obtaining high sensitivity and selectivity
when used to extract DNA from biological samples,28 however
their potential inhibitory effect on LAMP amplification has
not been investigated. In order to adapt the magnetic beadbased sample preparation method to our LOC system, we
tested the efficiency of LAMP in the presence of magnetic
beads.
The LAMP amplification was carried out with magnetic
beads at various concentrations. Fig. 3 shows the electrophoresis gel image of the LAMP products. The beads demonstrated no inhibitory effect at all on LAMP even at concentration of 4 × 1010 beads ml−1 (corresponding to 200 μl
Dynabeads® solution). According to the manufacture's manual, 20 μl of Dynabeads® can isolate at least 20 ng of high
quality genomic DNA, thus 20 μl of beads should be sufficient to extract DNA from 5 μl of spiked BPW enriched pork
meat samples.
As magnetic beads are completely compatible with LAMP
reaction, there is no need to elute the DNA and transfer it to
a second chamber. After capturing the DNA–magnetic bead
complex and washing, the LAMP master mixture could be
pumped directly into the chamber to perform the amplification. The integration of sample preparation and LAMP amplification into a single chamber greatly reduced the complexity
Lab Chip
View Article Online
Published on 17 February 2015. Downloaded by DTU Library on 26/02/2015 10:34:45.
Paper
Lab on a Chip
Fig. 3 Test of inhibitory effect of magnetic beads on LAMP reaction.
20 μl, 50 μl, 80 μl, 100 μl, 150 μl and 200 μl of Dynabeads® solution
were centrifuged at 5000 × g for 5 min and washed three times with
sterile water. After washing, 10 μl of LAMP master mixture and 1 μl of 2
ng μl−1 Salmonella DNA were added into the tubes and LAMP was
performed at 65 °C for 40 min and terminated the reaction at 80 °C
for 2 min. L: ladder; lane 1: 200 μl; lane 2: 150 μl; lane 3: 100 μl; lane
4: 80 μl; lane 5: 50 μl; lane 6: 20 μl; P: positive control; N: negative
control.
of the microchip as well as the microfluidic control elements,
which in turn increased the applicability of the system for
on-line assays.
Real-time detection of Salmonella using the
integrated LOC system
The feasibility of the integrated LOC system was demonstrated by detecting Salmonella spp. in multiple Salmonella
spiked BPW enriched pork meat samples. Five spiked samples with concentrations of S. enterica CCUG 32352 cells
ranging from 100 to 104 cells μl−1 and one negative control
were processed simultaneously on the eight-chamber microchip. The real-time amplification curves for the six assays are
shown in Fig. 4(a). The expected positive and negative signals
were observed for all assays, implying that the chip was suitable for parallel analysis of several samples, without interference among assays. By setting the threshold at 10 times the
standard deviation, the LOD system was able to detect as low
as 50 cells per test, or 10 cells μl−1 before sample preparation.
Good linearity was observed between the Tt and logarithm of
the amount of Salmonella cells per test (Fig. 4b), which
showed that real-time fluorescence monitoring could be utilized for not only detection but also quantification of pathogens. The analytical sensitivity of the on-chip LAMP Salmonella assay was comparable to the conventional PCR methods
which can typically detect 1 to 10 cells μl−1 of cultured Salmonella cells;6 however LAMP is superior to PCR in terms of
simple thermal conditions and short reaction time.
The successful detection of Salmonella spp. directly from
spiked BPW enriched pork meat samples using the eightchamber LOC system was attributed to a number of factors.
Firstly, the magnetic bead-based method was adopted for onchip sample preparation. Salmonella nucleic acids were effectively captured by the magnetic beads; and with the freedom
of particle manipulation by the magnetic fields, up to eight
Lab Chip
Fig. 4 Real-time detection of Salmonella on the integrated LOC
system. Five spiked samples with concentrations of S. enterica CCUG
32352 cells ranging from 100 to 104 cells μl−1 and one negative control
were processed simultaneously on the eight-chamber microchip.
(a) The real-time amplification curves for the six assays. The threshold
was set at 10 times the standard deviation. The LOD was 50 cells per
test. (b) The standard curve. It was determined based on the Tt values
for different sample concentrations. The experiments were repeated
three times.
samples could be preceded at one time and downscaled to
small sample volumes suitable for on-chip operation. Secondly, the magnetic beads were biocompatible with the
LAMP polymerase. Since the beads showed no noticeable
inhibition on LAMP, sample preparation and amplification
were well integrated into a single chamber, which made it
realistic for accomplishing parallel multi-sample analysis.
Moreover, the efficient intercalating dye SYTO-62 was
selected for monitoring the LAMP amplification in real time.
The low background and inhibition effects of the dye ensured
high sensitivity of the on-chip LAMP assay.
Conclusions
In this study, we have developed a multichannel microfluidic
system for parallel detection of Salmonella spp. in multiple
This journal is © The Royal Society of Chemistry 2015
View Article Online
Published on 17 February 2015. Downloaded by DTU Library on 26/02/2015 10:34:45.
Lab on a Chip
samples. To our best knowledge, this was the first LOC system that integrated both sample preparation and LAMP for
real-time detection of Salmonella spp. in food matrices. In
contrast to conventional PCR assays which required 2–3
hours for DNA purification and amplification, the whole process using the LOC system described here could be accomplished within less than 40 min. The detection limit was 50
cells per test, comparable with the standard real-time PCR
method. Of particular note, the stepper motor and the control unit from Lego were used for multi-channel pumping,
which made the system small in size and easy to use. The
simple design, low reaction temperature, and quantitative
analysis of multiple samples in a short time will eventually
facilitate the use of this novel LOC system as a point-of-care
device not only for fast screening of Salmonella or for other
foodborne pathogens in food and animal production but also
for clinical diagnostics and other applications in life sciences.
Competing interests
The authors declare no competing financial interest.
Acknowledgements
This work is co-supported by Forskningsrådet Teknologi og
Produktion (FTP), Sagsnr. 09066487, and HTF SMARTDETECT
project, grant no. 118-2012-3. We thank Kirsten Kirkeby from
Danish Crown Herning, Denmark for providing the BPW
enriched pork meat samples, Dr. Martin Dufva from DTU
Nanotech for providing the MainSTREAM micropump, and
Pia Englishman from DTU Food for technical assistant.
Notes and references
1 http://www.efsa.europa.eu/en/topics/topic/salmonella.htm, 2014.
2 http://www.cdc.gov/salmonella/, 2014.
3 http://eur-lex.europa.eu/legal-content/EN/TXT/?qid=
1416401613707&uri=CELEX:32012R0200.
4 A. Bhagwat, J. Patel, T. Chua, A. Chan, S. Cruz and G.
Aguilar, Methods Mol. Biol., 2008, 429, 33–43.
5 S. Koyuncu, M. G. Andersson and P. Haggblom, Appl.
Environ. Microbiol., 2010, 76, 2815–2822.
6 R. S. Tebbs, L. Y. Wong, P. Brzoska and O. V. Petrauskene,
in Salmonella: Distribution, Adaptation, Control Measures and
Molecular Technologies, ed. B. A. Annous and J. B. Gurtler,
InTeOp, 2012, pp. 481–504.
7 J. J. Maurer, Annu. Rev. Food Sci. Technol., 2011, 2, 259–279.
8 L. Yan, J. Zhou, Y. Zheng, A. S. Gamson, B. T. Roembke, S.
Nakayama and H. O. Sintim, Mol. BioSyst., 2014, 10,
970–1003.
This journal is © The Royal Society of Chemistry 2015
Paper
9 M. Parida, S. Sannarangaiah, P. K. Dash, P. V. L. Rao and K.
Morita, Rev. Med. Virol., 2008, 18, 407–421.
10 L. Niessen, J. Luo, C. Denschlag and R. F. Vogel, Food
Microbiol., 2013, 36, 191–206.
11 Salmonella Detection Kit. http://www.bio-protech.com.tw/
databank/datasheet/pathodiag/lmp601_salmonella_110704.
pdf, 2006.
12 P. J. Asiello and A. J. Baeumner, Lab Chip, 2011, 11,
1420–1430.
13 L. M. Zanoli and G. Spoto, Biosensors, 2013, 3, 18–43.
14 D. M. Tourlousse, F. Ahmad, R. D. Stedtfeld, G. Seyrig, J. M.
Tiedje and S. A. Hashsham, Biomed. Microdevices, 2012, 14,
769–778.
15 X. Fang, Y. Liu, J. Kong and X. Jiang, Anal. Chem., 2010, 82,
3002–3006.
16 C.-H. Wang, K.-Y. Lien, J.-J. Wu and G.-B. Lee, Lab Chip,
2011, 11, 1521–1531.
17 V. Velusamy, K. Arshak, O. Korostynska, K. Oliwa and C.
Adley, Biotechnol. Adv., 2010, 28, 232–254.
18 D. Sabourin, D. Snakenborg, P. Skafte-Pedersen, J. P. Kutter
and M. Dufva, in 14th International Conference on
Miniaturized Systems for Chemistry and Life Science,
Groningen, The Netherlands, 2010, pp. 1433–1435.
19 M. H. Josefsen, M. Krause, F. Hansen and J. Hoorfar, Appl.
Environ. Microbiol., 2007, 73, 3040–3048.
20 N. Tomita, Y. Mori, T. Notomi and K. Nagamine, Biochem.
Biophys. Res. Commun., 2001, 289, 150–154.
21 A. S. J. Mikosza, N. U. Mathu, Z. K. Njiru, J. C. Enyaru, T.
Armstrong and A. R. C. Thompson, PLoS Neglected Trop.
Dis., 2008, 2, e147.
22 Y. Mori, M. Kitao, N. Tomita and T. Notomi, J. Biochem.
Biophys. Methods, 2004, 59, 145–157.
23 J. C. Patel, J. Oberstaller, M. Xayavong, J. Narayanan, J. D.
DeBarry, G. Srinivasamoorthy, L. Villegas, A. A. Escalante, A.
DaSilva, D. S. Peterson, J. W. Barnwell, J. C. Kissinger, V.
Udhayakumar and N. W. Lucchi, PLoS One, 2013, 8, e54986.
24 H. Gudnason, M. Dufva, D. D. Bang and A. Wolff, Nucleic
Acids Res., 2007, 35, e127.
25 A. M. Foudeh, T. F. Didar, T. Veres and M. Tabrizian, Lab
Chip, 2012, 12, 3249–3266.
26 M. Bu, I. R. Perch-Nielsen, K. S. Sørensen, J. Skov, S. Yi,
D. D. Bang, M. E. Pedersen, M. F. Hansen and A. Wolff,
J. Micromech. Microeng., 2013, 23, 074002.
27 Y. Sun, J. Høgberg, T. Christine, L. Florian, L. G. Monsalve,
S. Rodriguez, C. Cao, A. Wolff, J. M. Ruano-Lopez and D. D.
Bang, Lab Chip, 2013, 13, 1509–1514.
28 R. Dhumpa, K. Handberg, P. H. Jøgensen, Y. Sun, A. Wolff
and D. D. Bang, Diagn. Microbiol. Infect. Dis., 2011, 69,
258–265.
Lab Chip