2 C PDF

RESEARCH ARTICLE
Highly-sensitive detection of Salmonella typhi

in clinical blood samples by magnetic
nanoparticle-based enrichment and in-situ
measurement of isothermal amplification of
nucleic acids
Avinash Kaur1, Arti Kapil2, Ravikrishnan Elangovan3, Sandeep Jha1,4,
Dinesh Kalyanasundaram1,4*
a1111111111
1 Centre for Biomedical Engineering, Indian Institute of Technology Delhi, New Delhi, India, 2 Department of
a1111111111
Microbiology, All India Institute of Medical Sciences, New Delhi, India, 3 Department of Biochemical
a1111111111 Engineering and Biotechnology, Indian Institute of Technology Delhi, New Delhi, India, 4 Department of
a1111111111 Biomedical Engineering, All India Institute of Medical Sciences, New Delhi, India
a1111111111
* [email protected], [email protected]
OPEN ACCESS
Abstract
Citation: Kaur A, Kapil A, Elangovan R, Jha S, Enteric fever continues to be a major cause of mortality and morbidity globally, particularly in
Kalyanasundaram D (2018) Highly-sensitive poor resource settings. Lack of rapid diagnostic assays is a major driving factor for the
detection of Salmonella typhi in clinical blood
empirical treatment of enteric fever. In this work, a rapid and sensitive method ‘Miod’ ‘has
samples by magnetic nanoparticle-based
enrichment and in-situ measurement of isothermal been developed. Miod includes a magnetic nanoparticle-based enrichment of target bacte-
amplification of nucleic acids. PLoS ONE 13(3): rial cells, followed by cell lysis and loop-mediated isothermal amplification (LAMP) of nucleic
e0194817. https://doi.org/10.1371/journal. acids for signal augmentation along with concurrent measurement of signal via an in–situ
pone.0194817
optical detection system. To identify positive/negative enteric fever infections in clinical
Editor: Ruslan Kalendar, University of Helsinki, blood samples, the samples were processed using Miod at time = 0 hours and time = 4
FINLAND
hours post-incubation in blood culture media. Primers specific for the STY2879 gene were
Received: November 14, 2017 used to amplify the nucleic acids isolated from S. typhi cells. A limit of detection of 5 CFU/
Accepted: March 10, 2018 mL was achieved. No cross-reactivity of the primers were observed against 106 CFU/mL of
Published: March 28, 2018 common pathogenic bacterial species found in blood such as E. coli, P. aeruginosa, S.
aureus, A. baumanni, E. faecalis, S. Paratyphi A and K. pneumonia. Miod was tested on 28
Copyright: © 2018 Kaur et al. This is an open
access article distributed under the terms of the human clinical blood samples. The detection of both pre-and post-four-hours incubation
Creative Commons Attribution License, which confirmed the presence of viable S. typhi cells and allowed clinical correlation of infection.
permits unrestricted use, distribution, and The positive and negative samples were successfully detected in less than 6 hours with
reproduction in any medium, provided the original
100% sensitivity and specificity.
author and source are credited.
Data Availability Statement: For additional data,

the readers/reviewers can contact the
corresponding author at [email protected].
Funding: The funding agencies are: Naval research Introduction

board (NRB/4003/PG/359); Department of Science
and Technology (YSS/2014/000880); Indo-German
Salmonella typhi causes the enteric fever that is a major public health problem, both in develop-
Science and Technology (IGSTC/Call 2014/ ing as well as developed economies [1,2]. About 21 million cases and 222,000 deaths per year
Sound4All/24/2015-16). The funders had no role in are caused worldwide by Salmonella typhi [3]. Antibiotics such as ampicillin, chloramphenicol,
PLOS ONE | https://doi.org/10.1371/journal.pone.0194817 March 28, 2018 1 / 14

Highly-sensitive detection of Salmonella typhi in clinical blood samples
study design, data collection and analysis, decision cotrimoxazole, fluoroquinolones and 3rd generation cephalosporin are the antibiotics of choice
to publish, or preparation of the manuscript. for the treatment of enteric fever. However, typhoidal Salmonella species have increasingly
Competing interests: The authors have declared become resistant to conventional antibiotics such as ampicillin, chloramphenicol, cotrimoxa-
that no competing interests exist. zole, and fluoroquinolones in developing countries [4]. The death rate in enteric fever is
expected to increase by 30% without appropriate diagnosis and effective antibiotic therapy [5].
Blood culture remains the gold standard test for diagnosis of enteric fever till today. The
organism once isolated from culture assay is further identified by biochemical tests [6,7]. Sero-
logical methods such as Widal test are regularly employed in many healthcare settings [8].
However, these serological methods have low sensitivity and specificity and are inconclusive
[9]. So far, confirmation of enteric fever depends on isolation of Salmonella typhi from the clin-
ical specimens such as urine, bone marrow, rose spot extracts, duodenal aspirates and stool
[10]. Blood culture based diagnosis of enteric fever demands (a) multiple time-consuming pro-
tocols (b) skilled labour and (c) numerous instruments and related infrastructure. Further,
only 45 to 70% of true positive cases are identified by this method [11]. These challenges in S.
typhi diagnosis are amplified multifold especially in resource-poor settings. Also, the lack of
specific, rapid and affordable diagnostic assays lead to inappropriate use of antibiotics in all
fevers cases, majority of which are viral or malaria. In addition to the above facts, approxi-
mately 20 to 30 mL of blood is required for detection of blood-related infections (both aerobic
and anaerobic bacterial cultures). Requirement of such volumes of blood pose a challenge for
geriatric and neonatal patients [12]. A diagnostic method requiring minimal blood volumes
with rapid and accurate detection is therefore required.
Some of the recently published methods for the detection of S. typhi in blood, water sam-
ples, and food are, lateral flow immunoassay (LFIA) using antibody-coated gold nanoparticles
[13], uniplex and, multiplex PCR [14], reverse transcriptase multiplex PCR (RT-MPCR) [15]
and gel electrophoresis [16]. The limit of detection (LOD) of these above methods ranges
between 500 to 104 CFU/mL. Besides, these methods are time-consuming, require numerous
equipments including expensive thermocyclers and trained labour. Henceforth, there is a
necessity for new methods that offer rapid, specific, and sensitive detection.
Isothermal amplification of nucleic acids is a promising option for quick and effective am-
plification eliminating the need for multiple cycles of rapid heating and cooling as demanded
in thermocycling PCR [17]. This feature greatly reduces the complexity of the device and
therefore the cost. Hence, isothermal based techniques have the potential for easy implementa-
tion in developing economies. Various isothermal amplification methods have been intro-
duced over the last decade, such as, nucleic acid sequence-based amplification (NASBA) [18],
loop-mediated amplification (LAMP) [19,20], strand displacement amplification [21], heli-
case-dependent amplification (HDA) [17], rolling circle amplification (RCA) [22], recombi-
nase polymerase amplification (RPA) [23], and multiple displacement amplification (MDA)
[24]. Of the above techniques, LAMP assay utilizes four to six primers along with strand dis-
placing DNA polymerase and amplifies target sequences in a rapid manner besides achieving
high specificity. LAMP assay can be utilized for many applications including detection of path-
ogens in food products, environmental samples, genetic testing, and point-of-care testing [25].
Cost-effective devices have been intended to enhance the portability of the LAMP assay for
field applications.
Ravan and Yazdanparast reported LAMP–ELISA assay for the detection of enteric fever in
spiked samples. The spiked samples were incubated up to 24 hours. The hybridization of
probe and amplification were performed simultaneously. The spiked samples were detected in
2 hours 40 minutes by measuring absorbance at 450 nm using microplate ELISA reader. The
LOD of this assay was reported as 10 CFU/mL [26]. Bozorgmehr et al. used LAMP based non-
crosslinking gold nanoprobes for the detection of S. typhi DNA. The team used surface

plasmon resonance (SPR) for end point-detection [27]. However, in the two methods de-
scribed above, no clinical blood samples were reported. Abdullah et al. developed an in-house
LAMP assay for the detection of S. typhi using three sets of primers designed for PapD gene.
LAMP reaction was performed using heating block (at 63˚C for 60 minutes) followed by detec-
tion using colorimetry. The LAMP method was compared against the gold standard of culture
method and polymerase chain reaction (PCR). The team reported a LOD of 104 CFU/mL or
20 CFU/reaction while that of PCR was 200 CFU/reaction [28].
The sensitivity of LAMP assay can be improved considerably by an enrichment protocol.
Immunomagnetic concentration and separation of target bacteria offers an advantage over
labour- intensive conventional pathogen enrichment methods (such as biological or serologi-
cal confirmation, plating and enrichment methods) [29]. Though LAMP is sensitive at lower
copies of nucleic acids, however, it is prone to inhibition and contaminants that are often
found in clinical samples. Immunomagnetic isolation helps to remove contaminants that
might interfere with the subsequent LAMP amplification detection assay [30,31]. In other
words, the enrichment protocol enhances the LAMP process by (a) eliminating of the contam-
inants (thereby reducing the incidence of false positive and false negative test results [29]) and
by (b) concentrating the target cells in a smaller volume. The utilization of 100 nm magnetic
nanoparticles (MNPs) enhances the binding kinetics in a diffusion-limited process [32]. Like-
wise, a higher surface to volume proportion enhances the signal by bringing down the steric
obstruction during ligand-receptor binding due to bigger molecule surface curvature [33].
Miod was developed in the present study that employs a magnetic nanoparticle-based enrich-
ment protocol followed by isothermal amplification of S. typhi nucleic acids and encompasses
an in-situ optical detection system with high specificity. In the preceding work, the protocol
for magnetic nanoparticle-based enrichment of S. typhi was optimized. The capture efficiency
of S. typhi cells using antibody coated MNPs was found to be above 65% [29]. In the current
work, isothermal amplification and detection were augmented to create a stand-alone system.
For isothermal amplification or signal augmentation, the STY2879 gene was amplified using 4
primary primers (pair of inner primers (FIP and BIP) and pair of outer primers (F3 and B3))
along with 2 additional loop primers (LF and LB). Gene STY2879 encodes for reverse tran-
scriptase protein in all S. typhi isolates. The specificity and cross-reactivity were tested against
other bacterial isolates such as Escherichia coli, Staphylococcus aureus, Pseudomonas aerugi-
nosa, Acinetobacter baumanni, Enterococcus faecalis, Salmonella Paratyphi A and Klebsiella
pneumonia. Miod was evaluated against 28 human clinical blood samples of suspected enteric
fever patients.
Materials and methods

Institute ethical approval
This study was approved by ethics committee of All India Institute of Medical Sciences, New
Delhi (document number IEC-307 dated 07th June 2016). Blood samples were obtained from
all participants with a written consent form, at the department of microbiology of All India
Institute of Medical Sciences, New Delhi. The blood samples were spiked in blood culture
media immediately.
Materials used
Dextran-coated magnetic nanoparticles (MNPs) of 100 nm with free carboxyl groups at the sur-
face (FluidMAG-CMX, Chemicell, Germany) conjugated with the polyclonal antibody against
Salmonella spp. (Rabbit antisera, Difco™, BD) were used for the enrichment of S. typhi cells. Iso-
thermal LAMP master mix (Optigene, UK), and primers (Integrated DNA Technologies, US)

were used for the nucleic acid amplification. Reagents such as phosphate buffer saline (PBS),
tryptone soy broth (TSB), bovine serum albumin (BSA), tween 20, 100X Tris-EDTA (TE)
buffer, N-(3-dimethylaminopropyl)-N0 -ethylcarbodiimide hydrochloride (EDC), N-hydroxy-
succinimide (NHS) were obtained from Sigma-Aldrich, USA. Neodymium magnets with
approximately 300 mT surface field were procured from a local hardware store. S. typhi Ty2
(ATCC 19430) strain, available with the department of microbiology at All India Institute of
Medical Sciences was used for standardization.
Bacterial culture for standardization

S. typhi was grown on TSB for 12 hours at 37˚C. The cells were collected in the mid-log phase
of the culture (O.D. ~ 0.6, approximately corresponding to 5×108 CFU/mL). The correlation
between optical density and CFU/mL was obtained via serial dilution and plating method. The
cells were spiked in sterile blood culture media and diluted serially to obtain 500, 400, 300,
200, 100, 50, 10 and 5 CFU/mL. All serial dilutions were carried out using sterile blood culture
media. The mixtures were vortexed for 30 seconds for homogenous mixing. The sterile blood
culture media without spiking of S. typhi was used as negative control. One milliliter solution
of each concentration was used for obtaining the first standardization curve. The spiked blood
cultures were incubated for 4 hours at 37˚C for growth of the S. typhi cells. Post incubation,
one milliliter of a solution of each concentration was used for obtaining the second standardi-
zation curve. To obtain the concentration of S. typhi for pre- and post-four-hour incubation,
20 μL of each dilution was plated on MacConkey1 agar plates followed by colony counts after
24 hours.
Conjugation of magnetic nanoparticles with S. typhi specific antibodies

Antibody conjugated magnetic nanoparticles (MNP-Ab) were prepared as described elsewhere
[29]. In brief, MNPs were washed twice with 10 mM PBS, followed by incubation with 20 mg/
mL of EDC and NHS for 15 minutes at room temperature (25˚C). MNPs were washed again
twice with PBS and then incubated with 50 μg of antibodies (Rabbit antisera, Difco™, BD) for
12 hours at room temperature. To remove non-specifically bound antibodies, the MNPs were
washed with 1% BSA and 0.1% Tween 20 for 1 hour at room temperature and stored at 4˚C.
MNP based enrichment, cell lysis and DNA isolation

S. typhi cells were enriched using MNPs-Ab in the ratio of 20:1 (i.e. 1 mL of solution to 50 μL
of MNPs-Ab). After 30 minutes of incubation at room temperature, target bacteria from the
blood culture media bound to the MNPs via the antibodies, were isolated using magnets held
on the outside of the vial. The remaining contents in the samples were eliminated by washing
twice with nuclease-free water (NFW). The enrichment process helps to eliminate the con-
taminants that hinder the LAMP amplification process. The concentrated cells were finally
resuspended in 50 μL of NFW. The suspension was heated at 65˚C for 45 minutes, to allow
detachment of MNPs from target S. typhi cells. The supernatant containing the cells were heat-
treated at 100˚C for 5 minutes for cell lysis. Post cell lysis, the suspension was centrifuged at
12,000 RPM for 5 minutes. Five microlitres of the supernatant were collected and used for the
LAMP assay. The protocol is shown in Fig 1.
Nucleic acid amplification

LAMP based amplification mandates four primers; forward outer primer (F3), backward outer
primer (B3), forward inner primer (FIP), backward inner primer (BIP) along with two

Fig 1. Flowchart showing detail protocol for conventional method and DNA based detection using LAMP.
https://doi.org/10.1371/journal.pone.0194817.g001
additional primers; loop forward (LF) and loop backward (LB) primers. Loop primers enhance
the rate of amplification. Detailed LAMP method of amplification is described elsewhere [34].
LAMP assay reaction was performed in 25 μL (total volume) reaction mixture containing
15 μL of LAMP master mix, 5 μL of primer mix (0.2 μM of F3 and B3 primers, 0.8 μM of FIP
and BIP, 0.4 μM of LF and LB primers) and 5 μL of target or template DNA. The LAMP master
mix contains Geobacillus species DNA polymerase (Optigene, UK), optimized reaction buffer
containing Mg2Cl2, deoxynucleotide triphosphates and ds-DNA binding dye (Optigene, UK).
The FIP, BIP, F3 and B3 primers for the STY2879 gene were obtained from Fan et al.[35]. The
loop primers LF and LB were designed using a commercial software LAMP designer software
version 1.10 (Optigene, UK).
Standardization of concentration versus intensity

LAMP reactions were carried out for different dilutions of S. typhi cells spiked in sterile blood
culture media from 500 CFU/mL down to 5 CFU/mL, pre—and post-incubation (4 hours).
DNA from each dilution was pre-concentrated using MNPs-Ab and lysed using heat treatment
method. Five microlitres of the supernatant were collected and used for the LAMP assay.
LAMP reaction was carried out at 65˚C for 15 minutes. The fluorescence intensity during
LAMP reaction mix increased over time, as detected by the fiber optic reflectance probe and a
fiber optic spectrometer (HR 2000+ES, Ocean optics, USA). The intensity corresponding to
each concentration was measured at different time intervals of 5, 10, 15, 20, 25 and 30 minutes.
The intensity grew clearly distinguishable between various concentrations at 15 minutes. An
increase in the intensity between pre—and post incubated samples confirms cell viability. The
intensity at 520 nm of wavelength was obtained for the pre-and post—incubated samples for
various concentration at 15 minutes to plot the calibration curve (or standard curve). All
experiments were carried out in triplicates.
Cross-reactivity and specificity with common pathogens/nuclei acids

Common pathogenic bacterial species including E. coli, S. aureus, P. aeruginosa, A. baumanni,
E. faecalis, S. paratyphi A and K. pneumonia are also typically found in blood samples of enteric

fever patients. Hence, experiments on the cross-reactivity of the primers along with master
mix and 106 CFU/mL concentration of the above bacterial species were carried out. DNA was
isolated using the protocol as described earlier. All experiments were carried out in triplicates.
LAMP assay from clinical samples

Human clinical blood samples from 28 suspected enteric fever patients were collected at the
medical institute. Upon collection of blood from the patient, the blood is immediately mixed
with culture media in the blood culture bottles. The blood collection was carried out using
standard protocol as described elsewhere [36,37]. Briefly, the site of venipuncture was disin-
fected using 70% alcohol, 10% povidone iodine and again 70% alcohol. Five milliliters of blood
were drawn from the patient. It was then spiked in blood culture bottles containing 45 mL of
brain heart infusion (BHI) media (BD Difco, USA). The bottle was incubated at 37˚C for cul-
ture. DNA was isolated after enrichment of cells using MNPs-Ab from 1 mL of media before
and incubation. The cells were then lysed using heat treatment method as described in the
MNP based Enrichment, cell lysis and DNA isolation section. Five microlitres of the superna-
tant were collected and used for the LAMP assay. LAMP assay was run for both pre- and post-
four hour incubated samples.
Comparison with conventional method

The results of Miod were validated by conventional methods. First, the blood culture bottle
samples were incubated at 37˚C for the culture of the bacterial organisms for 24 hours and 48
hours until observation of turbidity. The samples were further sub-cultured on a MacConkey
agar plates (BioMesrieux, France) at 37˚C for 24 hours. The formation of pale transparent col-
onies indicates the possible presence of S. typhi as per the standard protocol discussed else-
where [36,37]. Confirmatory biochemical test such as motility test, triple sugar iron (TSI) test,
urease test, citrate test and slide agglutination test were performed post sub-culturing on agar
plate. Motility test was used to determine the motility of S. typhi cells. Triple sugar iron agar
test was used to determine whether S. typhi cells utilize glucose, lactose or sucrose and produce
hydrogen sulfide (H2S) gas. Citrate utilization test was used to determine the ability of S. typhi
cells to utilize sodium citrate as a carbon source and inorganic NH4H2PO4 as a source of ni-
trogen. Urease test was used to determine whether S. typhi cells produce urease enzyme, an
indication of viable cells. A slide agglutination test using Antisera (Statens serum institute,
Copenhagen) was used to detect S. typhi O9, poly O and H antigens in blood. A brief descrip-
tion of each of the biochemical tests are included in the supplementary section S1.2 and shown
in S1 Fig. The conventional method in total consumes a minimum of 72 hours as compared to
Miod’s protocol of less than 6 hours, as given in S1 Table.
Statistical analysis. Based on the outcomes, the clinical sensitivity, specificity, positive
predictive value (PPV), and negative predictive value (NPV) were calculated to estimate the
effectiveness and efficiency of Miod [38].
Results and discussions

Design of device
An aluminum heating block of dimensions (length 5 cm, breadth 3 cm, height 2.5 cm) was
machined with three wells to accommodate conical PCR tubes of 200 μL. The depth and diam-
eter of the well was 10 mm and 5.5 mm. Three horizontal bores of diameters 6 mm and length
50 mm were drilled to seat two cartridge type heaters (Pratik heat products Pvt. Ltd., Mumbai,
India) and one thermocouple (Pixsys electronics, Italy). The thermocouple was placed

underneath the wells of PCR tubes to monitor the temperature of the wells during the reaction.
The cartridge heaters were placed parallel to both sides of the wells. Electrical power to the
heaters and thermocouple were supplied by two 1-Ampere transformers connected in parallel.
Thermocouple was linked to corresponding integral derivative (PID) controller (Selec, Mum-
bai) through solid state relay (SSR) (FUTEK, USA) to control the temperature of the heating
block. LAMP assay was carried out at 65˚ C for 15 minutes. The reaction was then terminated
by denaturing the polymerase in the master mix (Optigene, UK) by heating to 80˚ C for 2 min-
utes. The setup along with the instrumentation is shown in Fig 2. The current dimension of
the device is 30 cm × 30 cm × 8 cm. Y-type optical probes were placed on top of the vials in the
heating block. The optical fiber transmits white light from the source (Ocean optics, USA) into
the vials. The fluorescence light intensity (at a shorter band of wavelength) was observed at the
flip side of the optic fiber utilizing a spectrometer (HR 2000+ES, Ocean optics, USA) and spec-
tra suite software.
Standardization of intensity versus concentration

The intensity of fluorescence was measured in-situ during LAMP reaction. At 15 minutes and
at 520 nm of wavelength, the intensity for pre-incubation and post-four-hour incubation are
shown in Fig 3(A) and 3(B) respectively. The R2 value for pre-incubation was 0.948 and for
post-incubation (4 hours) was 0.983. The response of the detector for the negative control was
found to be 45 counts and 57 counts for pre-incubation and post-four-hour incubation respec-
tively. The theoretical limit of detection (LOD) calculated at 3 times the sensor reading of nega-
tive control are 135 counts (for pre-incubation), and 171 counts (for post-four incubation) for
reliable detection of the signal. Hence, based on S/N ratio 3, the LOD of pre-incubated sam-
ples was 200 CFU/mL (corresponding to 327 counts) while in the case of post-four-hour incu-
bation, the LOD was 5 CFU/mL (corresponding to 221 counts). The response of the detector
was linear at lower concentrations (5 to 500 CFU/mL) and non-linear at higher concentrations
Fig 2. (a) Schematic and (b) actual view: isothermal amplification and detection assembly (c) signal at detector at
various bacterial concentrations.

Fig 3. Calibration curve: Intensity at 15 minutes (a) LAMP assay versus initial concentration from 5 to 500 CFU/mL
along with (b) post-four-hour incubation. For post-incubated samples, the approximate CFU/mL count was obtained
by a count of cultured bacteria on agar plates as shown by red lines. The linear regime is shown by the dotted lines.
as shown in Fig 3(A) and 3(B) respectively. Linear curve fitting was used to fit the response for
pre-incubated samples. The equation of best-line-fit was x = [(y-38.20)/1.331] where, x repre-
sents the concentration in CFU/mL and y represents the intensity in number of counts. The
Boltzmann equation for post-four-hour incubated samples was [x = 39.70 ln ((y+5.28E6)/
(754.84-y))–354.20] respectively. This represents nonlinear curve fitting. An unknown concen-
tration of the cells in a given sample can be determined by using the equation of the standard
curve. The results of the bacterial count were obtained by plating (S2 Table and S2 Fig).
Our protocol involves experimenting both pre-incubated and post-incubated samples to
ascertain the viability of the cells in addition to high sensitivity. The initial concentration of 5
CFU/mL, yielded an intensity of 221 counts post-four-hour incubation. Hence, we conclude
that our theoretical LOD is 5 CFU/mL. The saturation for the post-incubation curve is due to
the limited number of fluorescent molecules. Since the experiment is designed to detect one
lakh S. typhi cells as a limit of detection shown in Fig 3. In the linear regime, the slope of the
calibration curve for pre-incubation sample was 1.331 counts per CFU/mL which, represents
the sensitivity of the sensor.
Cross-reactivity tests for primers

The cross-reactivity of the primers for the STY2879 gene was examined against K. pneumonia,
E. faecalis, A. baumanni, E. coli, P. aeruginosa, S. aureus and S. paratyphi A. Five microlitres of
the DNA template was used for LAMP reaction. The intensity was measured using optical
detection system at 520 nm wavelength and 15 minutes. Fig 4(A) represents the cross-reactiv-
ity reactions against the other bacterial species as mentioned above. It can be observed that no
cross-reactivity was observed even at high concentration of 106 CFU/mL of the pathogenic
bacterial species. Fig 4(B) depicts both pre-and post-four-hour incubation of S. typhi at various
concentrations. Though 5 and 50 CFU/mL of S. typhi shows equivalent signal to negative con-
trol and 106 CFU/mL of other bacterial species at pre-incubation, there is a significant increase
in the signal for S. typhi at post-four-hour incubation. In other words, the S. typhi signals are
higher than thrice the signal of negative control post-four-hour incubation.
Evaluation of clinical samples by Miod and conventional method

Assessment of 28 clinical samples by Miod and cross-validation with conventional method
were performed. In the conventional method, the blood culture bottles were incubated for 24
hours for the given clinical samples at 37˚C followed by sub-culture on MacConkey1 agar. S.

Fig 4. (a) Cross-reactivity of primers against other bacterial species spiked in sterile blood culture media (final concentration of 106 CFU/mL) (b)
positive (S. typhi) and negative control at pre-and post-four-hour incubation.
typhi was confirmed by standard bacteriology methods as shown in S1 Table. The minimum
time taken for the conventional method was 72 hours.
Six representative samples including three number of positive and negative samples were
identified by Miod is shown in Fig 5. The same samples were tested by the conventional
method and the detailed observations along with results are given in Table 1. It can be seen
that there is a 100% concurrence between the two methods. For the remaining 22 samples, the
results of the tests conducted by both Miod and the conventional method are given in the S3
Table of the supplementary section. Compared with the conventional method, Miod’s clinical
sensitivity, clinical specificity, PPV and NPV was 100% (Table 2). Hence, it can be concluded
Fig 5. Detection of S. typhi by Miod from human blood samples in the medical institute. The corresponding data
are shown only for representative six clinical samples.

Table 1. Results of culture and biochemical tests of the representative six clinical samples.
Clinical sample/ Negative BR256 OM812 RS704 AT959 LV529 SJ587
Negative control control
Bacterial Culture No colonies No No No Formation of pale Formation of pale Formation of pale
colonies colonies colonies transparent colonies transparent colonies transparent colonies
Motility test No change No No No Formation of Red turbidity Formation of Red turbidity Formation of Red turbidity
change change change
Triple sugar iron test No change No No No Red slant and yellow butt Red slant and yellow butt Red slant and yellow butt
Citrate test No change No No No No change in color No change in color No change in color
Urease test No change No No No No change in color No change in color No change in color
Slide agglutination test No clumps No No No Formation of visible Formation of visible Formation of visible
clumps clumps clumps clumps clumps clumps
Confirmation of No No No No Yes Yes Yes
S. typhi
https://doi.org/10.1371/journal.pone.0194817.t001
Table 2. A comparative analysis of conventional method and Miod detection of S. typhi.

Miod Conventional method Positive Conventional method Negative Total Sensitivity (%) Specificity (%) PPV (%) NPV (%)
Positive 3 0 3 100 100 100 100
Negative 0 25 25
Total 3 25 28
Table 3. Comparison of Miod with newly developed LAMP based amplification techniques.
Proposed by Sample Volume of Incubation and Time for sample Confirmation Sensitivity Tested on Confirms
sample processing time preparation, technique of S. clinical blood viability of
required amplification and typhi samples cells
detection
Abdullah et al. Human blood 100 μL 1 hour 1 hour Gel electrophoresis 104 CFU/mL
[28] in blood culture Yes No
media
Miod Human blood 2 mL 4 hours 1 hour 40 minutes Optical detection 5 CFU/mL
in blood culture Yes Yes
media
Ravan and Spiked samples 1 mL 2–24 hours 2 hours 40 minutes ELISA 10 CFU/mL
Yazdanparast in blood No No
[26]
Bozorgmehr Bacterial strains 200 μL 1 hour 30 minutes SPR 20 CFU/mL
et al. [27] in buffer No No
Fan et al. [35] simulated 2 mL, 500 μL 10 hour 65 minutes Real-time turbidity 20 CFU/mL
blood and stool (blood), 200 No No
samples CFU/g (stool)
that Miod is quite efficient at prediction given its reduced time for diagnosis and at lower LOD.
Miod confirms the presence of a viable pathogen in the patient within 6 hours as compared to
the conventional method of 72 hours that many nuclei acid based devices fail to confirm.
Recently, a few techniques have been published on using either optical or colorimetric
detection of S. typhi from spiked blood culture media and human blood samples, after amplifi-
cation of nucleic acids by LAMP methods. A comparison with of Miod with these recently
published methods in the literature is given in Table 3. It can be noticed that Miod has the

lowest LOD of 5 CFU/mL in clinical samples and has comparable processing time. The syner-
getic effects of magnetic enrichment combined with four-hour incubation and LAMP based
amplification protocol helps us in achieving the lowest LOD possible for the given sample. The
enrichment using MNP-Ab removes many inhibiting materials and thereby reduces the inci-
dence of false positive and false negative test results. In Table 3, many of the referenced tech-
niques have neither commented on the detection from clinical samples nor on the detection of
cell viability. Therefore, it can be observed that Miod has the potential for clinical use due to its
low LOD, ability to detect viable cells and in a quicker turnaround time. The current device is
a proof-of-concept for establishing the ‘Miod” protocol. Miod has the potential to further be
developed into a hand-held, portable and an economical device, however not in the current
form. The miniaturization of the device to suit resource-poor settings is the future scope of
work.
Conclusions
In this work, a rapid and highly sensitive detection of S. typhi was demonstrated. Miod employs
a magnetic nanoparticle-based pathogen enrichment protocol, followed by loop-mediated
nucleic acid amplification and simultaneous detection by an in-situ optical system. The system
detected the presence of S. typhi pre-and post-four-hour incubation in culture media that con-
firmed cell viability. LOD of the proposed Miod was 5 CFU/mL in spiked samples. In the eval-
uation of human clinical blood samples, Miod detected all 28 clinical samples including both
positive and negative samples with 100% sensitivity, specificity, PPV and NPV against the con-
ventional methods. The Miod protocol including detection was completed in less than 6 hours
while the conventional method involving cell culture followed by biochemical confirmatory
tests demanded 72 hours or more (S1 Table). Hence, Miod has the potential for clinical use
due to its low LOD, ability to detect viable cells and in a quicker turnaround time, a combina-
tion of significant factors that are not provided by other nucleic acid-based methods. The pro-
posed system be used to detect other pathogens with modification of the primers and minimal
modification of the lysis protocol. An offshoot application for the proposed system is the food
industry where rapid, on-site testing is necessary to rapidly detect potential sources of contam-
ination and infection. Miod shall be miniaturized into a hand-held form factor to adapt to
requirements of resource-poor settings. As dehydrated polymerase and primers are stable at
room temperature [39], Miod platform can be made into a field test device. Due to isothermal
amplification, power requirements are minimal and shall be met with portable batteries.
Supporting information
S1 Fig. Typical results observed in biochemical tests: (a) motility test, (b) TSI test, (c) citrate
test, (d) urease test. Shown only for comparison between clinical samples (indicated by ‘+’)
and control samples (indicated by ‘-‘).
(TIF)
S2 Fig. Calibration curve of cell concentration (CFU/mL) versus optical density for S. typhi
bacteria.
(TIF)
S1 Table. Protocol of conventional method. ^^ indicates tests run in parallel to each other.
(DOCX)
S2 Table. Plating of different dilutions of S. typhi cells spiked in blood culture bottles.
20 μL volume of inoculum was used for the plating.
(DOCX)

S3 Table. Results from conventional and proposed Miod for 28 clinical samples.
(DOCX)
S1 File. Supplementary 21st March.docx.
(DOCX)
Author Contributions
Conceptualization: Arti Kapil.
Formal analysis: Ravikrishnan Elangovan, Sandeep Jha.
Investigation: Avinash Kaur, Dinesh Kalyanasundaram.
Methodology: Avinash Kaur, Dinesh Kalyanasundaram.
Project administration: Dinesh Kalyanasundaram.
Resources: Arti Kapil, Sandeep Jha, Dinesh Kalyanasundaram.
Supervision: Arti Kapil, Sandeep Jha, Dinesh Kalyanasundaram.
Validation: Ravikrishnan Elangovan, Sandeep Jha.
Writing – original draft: Avinash Kaur, Dinesh Kalyanasundaram.
Writing – review & editing: Ravikrishnan Elangovan, Sandeep Jha, Dinesh
Kalyanasundaram.
References
1. Kumar S, Balakrishna K, Singh GP, Batra H V. Rapid detection of Salmonella typhi in foods by combina-
tion of immunomagnetic separation and polymerase chain reaction. World J Microbiol Biotechnol. 2005;
21: 625–628. https://doi.org/10.1007/sl1274-004-3553-x
2. Alba S, Bakker MI, Hatta M, Scheelbeek PFD, Dwiyanti R, Usman R, et al. Risk factors of typhoid infec-
tion in the Indonesian archipelago. PLoS One. 2016;11. https://doi.org/10.1371/journal.pone.0155286
PMID: 27281311
3. WHO. Immunization, Vaccines and Biologicals Typhoid. World Heal Organ. 2015; 1–2.
4. Dahiya S, Sharma P, Kumari B, Pandey S, Malik R, Manral N, et al. Characterisation of antimicrobial
resistance in Salmonellae during 2014–2015 from four centres across India: An ICMR antimicrobial
resistance surveillance network report. Indian J Med Microbiol. Medknow; 2017; 35: 61. https://doi.org/
10.4103/ijmm.IJMM_16_382 PMID: 28303820
5. Dutta S, Das S, Mitra U, Jain P, Roy I, Ganguly SS, et al. Antimicrobial Resistance, Virulence Profiles
and Molecular Subtypes of Salmonella enterica Serovars Typhi and Paratyphi A Blood Isolates from
Kolkata, India during 2009–2013. PLoS One. 2014;9. https://doi.org/10.1371/journal.pone.0101347
PMID: 25098613
6. Martiny D, Debaugnies F, Gateff D, Gérard M, Aoun M, Martin C, et al. Impact of rapid microbial identifi-
cation directly from positive blood cultures using matrix-assisted laser desorption/ionization time-of-
flight mass spectrometry on patient management. Clin Microbiol Infect. 2013; 19: E568–E581. https://
doi.org/10.1111/1469-0691.12282 PMID: 23890423
7. Paolucci M, Landini MP, Sambri V. Conventional and molecular techniques for the early diagnosis of
bacteraemia. Int J Antimicrob Agents. 2010; 36: S6–S16. https://doi.org/10.1016/j.ijantimicag.2010.11.
010 PMID: 21129933
8. Mogasale V, Ramani E, Mogasale V V, Park J. What proportion of Salmonella Typhi cases are detected
by blood culture? A systematic literature review. Ann Clin Microbiol Antimicrob. BioMed Central; 2016;
15: 32. https://doi.org/10.1186/s12941-016-0147-z PMID: 27188991
9. Darton TC, Zhou L, Blohmke CJ, Jones C, Waddington CS, Baker S, et al. Blood culture-PCR to opti-
mise typhoid fever diagnosis after controlled human infection identifies frequent asymptomatic cases
and evidence of primary bacteraemia. J Infect. 2017; 74: 358–366. https://doi.org/10.1016/j.jinf.2017.
01.006 PMID: 28130144

10. Bhutta Z. Current concepts in the diagnosis and treatment of typhoid fever. BMJ. 2006; 333: 78–82.
https://doi.org/10.1136/bmj.333.7558.78 PMID: 16825230
11. Parry CM, Wijedoru L, Arjyal A, Baker S. The utility of diagnostic tests for enteric fever in endemic loca-
tions. Expert Rev Anti Infect Ther. 2011; 9: 711–725. https://doi.org/10.1586/eri.11.47 PMID: 21692675
12. Mancini N, Carletti S, Ghidoli N, Cichero P, Burioni R, Clementi M. The Era of Molecular and Other Non-
Culture-Based Methods in Diagnosis of Sepsis. Clin Microbiol Rev. 2010; 23: 235–251. https://doi.org/
10.1128/CMR.00043-09 PMID: 20065332
13. Singh J, Sharma S, Nara S. Nanogold based lateral flow assay for the detection of Salmonella typhi in
environmental water samples. Anal Methods. Royal Society of Chemistry; 2015; 7: 9281–9288. https://
doi.org/10.1039/c5ay02271a
14. Ranjbar R, Mortazavi SM, Mehrabi Tavana A, Sarshar M, Najafi A, Soruri Zanjani R. Simultaneous
Molecular Detection of Salmonella enterica Serovars Typhi, Enteritidis, Infantis, and Typhimurium. Iran
J Public Health. Iran; 2017; 46: 103–111. PMID: 28451535
15. Maan S, Maan NS, Belaganahalli MN, Potgieter AC, Kumar V, Batra K, et al. Development and evalua-
tion of real time RT-PCR assays for detection and typing of bluetongue virus. PLoS One. 2016;11.
https://doi.org/10.1371/journal.pone.0163014 PMID: 27661614
16. Phumkhachorn P, Rattanachaikunsopon P. Detection of viable Salmonella Typhi by reverse transcrip-
tion-multiplex polymerase chain reaction. Emirates J Food Agric. 2017; 29: 1. https://doi.org/10.9755/
ejfa.2016-12-1867
17. Kivlehan F, Mavré F, Talini L, Limoges B, Marchal D. Real-time electrochemical monitoring of isother-
mal helicase-dependent amplification of nucleic acids. Analyst. 2011; 136: 3635–3642. https://doi.org/
10.1039/c1an15289k PMID: 21792448
18. Kim J, Easley CJ. Isothermal DNA amplification in bioanalysis: strategies and applications. Bioanalysis.
2011; 3: 227–39. https://doi.org/10.4155/bio.10.172 PMID: 21250850
19. Perera RS, Ding XC, Tully F, Oliver J, Bright N, Bell D, et al. Development and clinical performance of
high throughput loop-mediated isothermal amplification for detection of malaria. PLoS One. 2017;12.
https://doi.org/10.1371/journal.pone.0171126 PMID: 28166235
20. Kokkinos PA, Ziros PG, Bellou M, Vantarakis A. Loop-Mediated Isothermal Amplification (LAMP) for the
Detection of Salmonella in Food. Food Analytical Methods. 2014. pp. 512–526. https://doi.org/10.1007/
s12161-013-9748-8
21. Toley BJ, Covelli I, Belousov Y, Ramachandran S, Kline E, Scarr N, et al. Isothermal strand displace-
ment amplification (iSDA): a rapid and sensitive method of nucleic acid amplification for point-of-care
diagnosis. Analyst. Royal Society of Chemistry; 2015; 140: 7540–7549. https://doi.org/10.1039/
C5AN01632K PMID: 26393240
22. Da Silva FRC, Cibulski SP, Daudt C, Weber MN, Guimarães LLB, Streck AF, et al. Novel bovine papillo-
mavirus type discovered by rolling-circle amplification coupled with next-generation sequencing. PLoS
One. 2016;11. https://doi.org/10.1371/journal.pone.0162345 PMID: 27606703
23. Wang J, Liu L, Wang J, Sun X, Yuan W. Recombinase polymerase amplification assay-a simple, fast
and cost-effective alternative to real time PCR for specific detection of feline herpesvirus-1. PLoS One.
2017;12. https://doi.org/10.1371/journal.pone.0166903 PMID: 28045956
24. Yan L, Zhou J, Zheng Y, Gamson AS, Roembke BT, Nakayama S, et al. Isothermal amplified detection
of DNA and RNA. Mol Biosyst. 2014; 10: 970. https://doi.org/10.1039/c3mb70304e PMID: 24643211
25. Notomi T, Mori Y, Tomita N, Kanda H. Loop-mediated isothermal amplification (LAMP): principle, fea-
tures, and future prospects. Journal of microbiology (Seoul, Korea). 2015. pp. 1–5. https://doi.org/10.
1007/s12275-015-4656-9 PMID: 25557475
26. Ravan H, Yazdanparast R. Loop region-specific oligonucleotide probes for loop-mediated isothermal
amplification-enzyme-linked immunosorbent assay truly minimize the instrument needed for detection
process. Anal Biochem. Elsevier Inc.; 2013; 439: 102–108. https://doi.org/10.1016/j.ab.2013.04.014
PMID: 23624347
27. Bozorgmehr A, Yazdanparast R, Hamidreza M. Non-crosslinking gold nanoprobe-LAMP for simple, col-
orimetric, and specific detection of Salmonella typhi. J Nanoparticle Res. 2016; 351.
28. Abdullah J, Saffie N, Sjasri FAR, Husin A, Abdul-Rahman Z, Ismail A, et al. Rapid detection of Salmo-
nella Typhi by loop-mediated isothermal amplification (LAMP) method. Brazilian J Microbiol. 2014; 45:
1385–1391.
29. Singh S, Upadhyay M, Sharma J, Gupta S, Vivekanandan P, Elangovan R. A portable immunomagnetic
cell capture system to accelerate culture diagnosis of bacterial infections. Royal Society of Chemistry;
2016; 3358–3366. https://doi.org/10.1039/c6an00291a PMID: 27118505
30. Cudjoe K. Immunomagnetic separation of Salmonella from foods and their detection using immuno-
magnetic particle (IMP)-ELISA. Int J Food Microbiol. 1995; 27: 11–25. https://doi.org/10.1016/0168-
1605(94)00134-R PMID: 8527325

31. Liu Y, Che Y, Li Y. Rapid detection of Salmonella typhimurium using immunomagnetic separation and
immuno-optical sensing method. Sensors Actuators B Chem. 2001; 72: 214–218. https://doi.org/10.
1016/S0925-4005(00)00663-8
32. Kang JH, Um E, Diaz A, Driscoll H, Rodas MJ, Domansky K, et al. Optimization of pathogen capture in
flowing fluids with magnetic nanoparticles. Small. Wiley Online Library; 2015; 11: 5657–5666. https://
doi.org/10.1002/smll.201501820 PMID: 26389806
33. Hejazian M, Li W, Nguyen N-T. Lab on a chip for continuous-flow magnetic cell separation. Lab Chip.
Royal Society of Chemistry; 2015; 15: 959–970.
34. Nagamine K, Hase T, Notomi T. Accelerated reaction by loop-mediated isothermal amplification using
loop primers. Mol Cell Probes. 2002; 16: 223–229. PMID: 12144774
35. Fan F, Du P, Kan B, Yan M. The development and evaluation of a loop-mediated isothermal amplifica-
tion method for the rapid detection of Salmonella enterica serovar Typhi. PLoS One. 2015;10. https://
doi.org/10.1371/journal.pone.0124507 PMID: 25910059
36. Crichton PB. Enterobacteriaceae: Escherichia, Klebsiella, proteus and other genera. Mackie McCart-
ney Pract Med Microbiol. Churchill Livingston, New York; 1996; 14: 361–384.
37. DL K, PH G. Manual of Clinical Microbiology. 9th editio. Murray PR, Baron EJ, Jorgensen JH, Pfaller
MA YR, editor. American Society for Microbiology Press; 2007.
38. Lee D, Kim EJ, Kilgore PE, Kim SA, Takahashi H, Ohnishi M, et al. Clinical Evaluation of a Loop-Medi-
ated Isothermal Amplification (LAMP) Assay for Rapid Detection of Neisseria meningitidis in Cerebro-
spinal Fluid. PLoS One. 2015; 10: e0122922. https://doi.org/10.1371/journal.pone.0122922 PMID:
25853422
39. Carter C, Akrami K, Hall D, Smith D, Aronoff-Spencer E. Lyophilized visually readable loop-mediated
isothermal reverse transcriptase nucleic acid amplification test for detection Ebola Zaire RNA. J Virol
Methods. 2017; 244: 32–38. https://doi.org/10.1016/j.jviromet.2017.02.013 PMID: 28242293

2 C PDF

Uploaded by

Copyright:

Available Formats

2 C PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

2 C PDF

Uploaded by

Copyright:

Available Formats

RESEARCH ARTICLE

Highly-sensitive detection of Salmonella typhi

Data Availability Statement: For additional data,

Funding: The funding agencies are: Naval research Introduction

PLOS ONE | https://doi.org/10.1371/journal.pone.0194817 March 28, 2018 1 / 14

PLOS ONE | https://doi.org/10.1371/journal.pone.0194817 March 28, 2018 2 / 14

Materials and methods

PLOS ONE | https://doi.org/10.1371/journal.pone.0194817 March 28, 2018 3 / 14

Bacterial culture for standardization

Conjugation of magnetic nanoparticles with S. typhi specific antibodies

MNP based enrichment, cell lysis and DNA isolation

Nucleic acid amplification

PLOS ONE | https://doi.org/10.1371/journal.pone.0194817 March 28, 2018 4 / 14

Standardization of concentration versus intensity

Cross-reactivity and specificity with common pathogens/nuclei acids

PLOS ONE | https://doi.org/10.1371/journal.pone.0194817 March 28, 2018 5 / 14

LAMP assay from clinical samples

Comparison with conventional method

Results and discussions

PLOS ONE | https://doi.org/10.1371/journal.pone.0194817 March 28, 2018 6 / 14

Standardization of intensity versus concentration

PLOS ONE | https://doi.org/10.1371/journal.pone.0194817 March 28, 2018 7 / 14

Cross-reactivity tests for primers

Evaluation of clinical samples by Miod and conventional method

PLOS ONE | https://doi.org/10.1371/journal.pone.0194817 March 28, 2018 8 / 14

PLOS ONE | https://doi.org/10.1371/journal.pone.0194817 March 28, 2018 9 / 14

Table 2. A comparative analysis of conventional method and Miod detection of S. typhi.

PLOS ONE | https://doi.org/10.1371/journal.pone.0194817 March 28, 2018 10 / 14

PLOS ONE | https://doi.org/10.1371/journal.pone.0194817 March 28, 2018 11 / 14

PLOS ONE | https://doi.org/10.1371/journal.pone.0194817 March 28, 2018 12 / 14

PLOS ONE | https://doi.org/10.1371/journal.pone.0194817 March 28, 2018 13 / 14

PLOS ONE | https://doi.org/10.1371/journal.pone.0194817 March 28, 2018 14 / 14

You might also like