The Journal of Molecular Diagnostics, Vol. 22, No.

5, May 2020

jmd.amjpathol.org

Highly Sensitive and Specific Detection and

Serotyping of Five Prevalent Salmonella Serovars
by Multiple Cross-Displacement Amplification
Xiaomei Zhang,* Michael Payne,* Qinning Wang,y Vitali Sintchenko,yz and Ruiting Lan*

From the School of Biotechnology and Biomolecular Sciences,* University of New South Wales, Sydney, New South Wales; the Centre for Infectious Diseases
and MicrobiologyePublic Health,y Institute of Clinical Pathology and Medical Research, NSW Health Pathology, Westmead, New South Wales; and the Marie
Bashir Institute for Infectious Diseases and Biosecurity,z The University of Sydney, Sydney, New South Wales, Australia

Accepted for publication

February 20, 2020.
Address correspondence to
Ruiting Lan, Ph.D., School of
Biotechnology and Biomole-
cular Sciences, University of
New South Wales, Sydney,
New South Wales, 2052,
Australia. E-mail: r.lan@unsw.
edu.au.
Salmonella is a common cause of foodborne disease worldwide, including Australia. More than 85% of
outbreaks of human salmonellosis in Australia were caused by five Salmonella serovars. Rapid, accurate,
and sensitive identification of Salmonella serovars is vital for diagnosis and public health surveillance.
Recently, an isothermal amplification technique, termed multiple cross-displacement amplification
(MCDA), has been employed to detect Salmonella at the species level. Herein, seven MCDA assays were
developed and evaluated for rapid detection and differentiation of the five most common Salmonella
serovars in Australia: Typhimurium, Enteritidis, Virchow, Saintpaul, and Infantis. MCDA primer sets were
designed by targeting seven serovar/lineage-specific gene markers identified through genomic com-
parisons. The sensitivity and specificity of the seven MCDA assays were evaluated using 79 target strains
and 32 nontarget strains. The assays were all highly sensitive and specific to target serovars, with the
sensitivity ranging from 92.9% to 100% and the specificity from 93.3% to 100%. The limit of detection
of the seven MCDA assays was 50 fg per reaction (10 copies) from pure DNA, and positive results were
detected in as little as 8 minutes. These seven MCDA assays offer a rapid, accurate, and sensitive
serotyping method. With further validation in clinically relevant conditions, these assays could be used
for culture-independent serotyping of common Salmonella serovars directly from clinical samples.
(J Mol Diagn 2020, 22: 708e719; https://doi.org/10.1016/j.jmoldx.2020.02.006)

Salmonella enterica is one of the most common causes of
foodborne disease worldwide, including Australia.1,2 The
number of reported human salmonellosis cases in Australia
has increased significantly during recent decades.3 The case
rate is estimated to be 185 per 100,000 population per year,4
with 16,383 cases being notified in 2017, a 30% increase
compared with the mean notifications for the previous 10
years (2007 to 2016; National Notifiable Diseases Surveil-
lance System of Australia). The most prevalent serovar of
Salmonella reported in human infections in Australia has
been Typhimurium,1,5e7 followed by Enteritidis, Virchow,
Saintpaul, and Infantis.1 These five cocirculating Salmonella
serovars are responsible for >85% of outbreaks of human
salmonellosis in Australia.1
Traditional culture-based methods for detection of Sal-
monella pathogens are time-consuming, laborious, and
expensive.8e11 In recent decades, many culture-independent
methods using genetic determinants have offered appealing
alternatives to traditional methods. Among these, PCR-
based techniques (PCR and real-time PCR) and isothermal
amplification techniques, such as loop-mediated isothermal
amplification (LAMP), have been suggested.12e18 In a
recent study, another isothermal amplification technique,
termed multiple cross-displacement amplification (MCDA),
was reported.19,20 MCDA employs 10 primers, instead of
six in LAMP or two in PCR, to recognize 10 distinct
regions, which enhance its specificity and sensitivity.
Supported by the National Health and Medical Research Council of
Australia project grant APP1129713 (R.L. and V.S.).
Disclosures: None declared.

Copyright ª 2020 Association for Molecular Pathology and American Society for Investigative Pathology. Published by Elsevier Inc. All rights reserved.
https://doi.org/10.1016/j.jmoldx.2020.02.006

Comparative genomics leveraging the recent explosion of
publicly available genomic data can be used to identify
molecular markers for pathogen detection. Previously,21 the
use of comparative genomics enabled the identification of a
panel of 15 serovar/lineage-specific gene markers for typing
the 10 most frequent Salmonella serovars in Australia.
When prevalence of Salmonella serovars endemic in
Australia was taken into account, the genes listed in that
panel could be used as markers for laboratory-based typing
with an error rate of <2.4%.
Herein, MCDA assays targeting serovar/lineage-specific
genes were developed and evaluated for rapid, accurate
identification and serotyping of the five Salmonella serovars
Typhimurium, Enteritidis, Virchow, Saintpaul, and Infantis,
which are dominant in Australia and internationally.
Materials and Methods
Bacterial Strains and Genomic DNA Extraction
A total of 111 strains were used, consisting of the following:
16 Salmonella Reference Collection A strains22; 33 Salmo-
nella Reference Collection B (SARB) strains,23 including 9
target serovar strains and 24 nontarget serovar strains; 54
Salmonella strains from New South Wales Enteric Reference
Laboratory, NSW Health Pathology, representing five target
serovars; and 8 non-Salmonella strains, representing eight
different species (Supplemental Table S1). The bacterial
strains were cultured on nutrient agar at 37 C for 18 to 24
hours. Genomic DNA was extracted using phenol-
chloroform method, as described previously.24
Design of MCDA Primers and the Specificities of MCDA
Products
Seven serovar/lineage-specific gene markers21 (STM4494,
SEN1384, R561_RS18155, SESV_RS06060, SES-
PA_RS08460, SeSPB_A1749, and L287_11788) specific to
the five most common Salmonella serovars in Australia and
internationally [Typhimurium, Enteritidis-clade B,
Enteritidis-clade A/C, Virchow, Saintpaul lineage I (Saint-
paul-I), Saintpaul lineage II (Saintpaul-II), and Infantis]
were selected as targets for MCDA primer design. The se-
quences of the seven serovar/lineage-specific gene markers
are listed in Supplemental Table S2.
Each MCDA primer set consisted of two cross primers
(CP1 and CP2), two displacement primers (F1 and F2), and
six amplification primers (D1, C1, R1, D2, C2, and R2).19
Seven MCDA primers sets were designed using Primer3
online software version 0.4.0 (http://bioinfo.ut.ee/primer3-0.
4.0, last accessed April 2018) based on the principle of
MCDA recognizing 10 distinct regions to amplify each
serovar/lineage-specific gene marker. The specificities of
the primers were analyzed by National Center for Biotech-
nology Information BLAST. OligoAnalyzer online software
version 3.1 (Integrated DNA Technologies, Coralville, IA)
The Journal of Molecular Diagnostics - jmd.amjpathol.org
MCDA for Typing of Salmonella Serovars
was used for primer dimer and secondary structure inves-
tigation. The sequences and locations of seven MCDA
primer sets are presented in Table 1 and Figure 1.
The specificity of seven MCDA products (ie, presence
only in the targeted serovars) was also examined in silico by
searching the products against a diverse set of 2258 Sal-
monella species genomes from the identification data group
in a previous study21 using BLASTN with default settings.
The Initial Evaluation of the Seven MCDA Assays
To evaluate the seven MCDA primer sets, Salmonella
speciesespecific gene invA MCDA assay20 was used as
positive control. The MCDA reactions were performed on a
Corbett Rotor-Gene 6000 Real Time PCR Machine with the
WarmStart LAMP DNA Amplification Kit (New England
BioLabs, Melbourne, Victoria, Australia) in a total volume
of 10 mL reaction mixture incubated at 63 C for 60 minutes
and then heated at 95 C for 5 minutes to stop the amplifi-
cation. Real-time LAMP fluorescent dye measurement was
used to monitor the MCDA amplification every minute.
The final 10-mL MCDA reaction mixtures contained the
following primer mixture at concentrations as previously
described20: 2.4 mmol/L each of cross primers CP1 and CP2, 0.4
mmol/L each of displacement primers F1 and F2, 1.2 mmol/L
each of amplification primers R1, R2, D1, and D2, and 0.8 mmol/
L each of amplification primers C1 and C2. The reaction mixture
consisted of 5 mL 2 WarmStart LAMP Master Mix (New
England Biolabs), 0.2 mL 5 fluorescent dye, 1.2 mL MCDA
primer mixture, 2.6 mL Milli-Q water, and 1 mL DNA template.
Evaluation of the LoD of the MCDA Assays in Pure
Culture
Limit of detection (LoD) was defined as the lowest genomic
DNA level where all six replicates were detected by each
MCDA assay on the condition that its detection time was faster
than invA MCDA assay. To demonstrate the efficiency of seven
MCDA assays, each novel MCDA assay together with the
existing Salmonella invA MCDA assay was analyzed for LoD.
The genomic DNA template from seven target strains
belonging to their respective serovars (Salmonella Reference
Collection A 14, Typhimurium; L2376, Enteritidis-clade B;
L2380, Enteritidis-clade A/C; L2349, Virchow; Salmonella
Reference Collection A 28, Saintpaul-I; SARB56, Saintpaul-II;
L2385, Infantis) was serially diluted (5 ng, 500 pg, 50 pg, 5 pg,
500 fg, 50 fg, 25 fg, 12.5 fg, 6.25 fg per microliter) to determine
the LoD. One replicate for the dilutions of 5 ng, 500 pg, 50 pg,
and 5 pg and three replicates for the dilutions of 500, 50, 25,
12.5, and 6.25 fg were tested with two independent runs.
The detection time was defined as the time at which the
fluorescence signal doubled the value of the baseline.
GraphPad Prism software version 7.04 (GraphPad Software
Inc., San Diego, CA) was used to perform statistical ana-
lyses to show the relationship of detection times and di-
lutions between each MCDA assay and invA MCDA assay.
709
Zhang et al

Table 1 Primers Used for Seven Multiple Cross-Displacement Amplification Assays

Primer name* Sequence
Typhimurium-F1 50 -AATCGTCGCTCTTCAATATG-30
Typhimurium-F2 50 -TGTAGCCAGCGTTGTACC-30
Typhimurium-CP1 50 -GATACGTTTACCGCTGAAGAACTGGAACCATGCCCGGTGAATATC-30
Typhimurium-CP2 50 -TCAGGGAATGATCATTCGTTAGATGCTAAACAGCATAATCAGCACCTG-30
Typhimurium-C1 50 -GATACGTTTACCGCTGAAGAACTGG-30
Typhimurium-C2 50 -TCAGGGAATGATCATTCGTTAGATGC-30
Typhimurium-D1 50 -TAGCGTGGCGATCATTTCA-30
Typhimurium-D2 50 -CTTAGCTCCGGCGAACAT-30
Typhimurium-R1 50 -CCTGGATGAATTTCAGCTTC-30
Typhimurium-R2 50 -GCAACGTGTCCTACTGGAT-30
Enteritidis-clade B-F1 50 -ATAACACTTACGGAGCTGAG-30
Enteritidis-clade B-F2 50 -TCGTAACGACGTACCTCAC-30
Enteritidis-clade B-CP1 50 -CCACAACGTTCTGCCTTGTCCAAGGATGACGGGGTTAACCATT-30
Enteritidis-clade B-CP2 50 -GCTTATCGTGCCTGGAAGAAACAGCGTCAGGCAGCTTCCAAATC-30
Enteritidis-clade B-C1 50 -CCACAACGTTCTGCCTTGTCCA-30
Enteritidis-clade B-C2 50 -GCTTATCGTGCCTGGAAGAAACAG-30
Enteritidis-clade B-D1 50 -GTAGTGGCGGGTCGAATA-30
Enteritidis-clade B-D2 50 -GAAAGTGGACGCTGACCT-30
Enteritidis-clade B-R1 50 -TGCCCGCCTGGTACACAT-30
Enteritidis-clade B-R2 50 -GATTTTCCCGTCAGAAGAG-30
Enteritidis-clade A/C-F1 50 -TTTCATTATAGGGCAGGGA-30
Enteritidis-clade A/C-F2 50 -CTGTCACAATCAAATAATGA-30
Enteritidis-clade A/C-CP1 50 -GTGACACGAAATGAATGAGTCCAATCGTCTTGAGATTATAGTTACTCTTG-30
Enteritidis-clade A/C-CP2 50 -TGCGAGTAGGTATTTATAAGGTTGAGTCATGTATATTAAAACTCTGGTC-30
Enteritidis-clade A/C-C1 50 -GTGACACGAAATGAATGAGTCCAATC-30
Enteritidis-clade A/C-C2 50 -TGCGAGTAGGTATTTATAAGGTTGAG-30
Enteritidis-clade A/C-D1 50 -CGAAAATCCGAATTCACTCC-30
Enteritidis-clade A/C-D2 50 -ACTCATCTTATCTGGAATGG-30
Enteritidis-clade A/C-R1 50 -CAACAGATCACCTTCATCA-30
Enteritidis-clade A/C-R2 50 -TGTTGGGTGAGCAAAAAGG-30
Virchow-F1 50 -TCATTATTAGACCAATCTGC-30
Virchow-F2 50 -TTCGTTTGCTGATTCCATG-30
Virchow-CP1 50 -GTGCTGAAACTTTTATTTATGCTTGGGAATTGACCAGTCGGTTAAGGC-30
Virchow-CP2 50 -GCCAGCACAAATGAATACTGTATGGCAACGGGATCCTATTC-30
Virchow-C1 50 -GTGCTGAAACTTTTATTTATGCTTGGG-30
Virchow-C2 50 -GCCAGCACAAATGAATACTGT-30
Virchow-D1 50 -AACTTTCGCGTTGTGAGCT-30
Virchow-D2 50 -CTGGATCTTAAATAGTCATC-30
Virchow-R1 50 -ATTTTAGGTGGCACCCATC-30
Virchow-R2 50 -TATGTTGTGGCATATGATGG-30
Saintpaul-I-F1 50 -TCAGACTGAAGACCAGCTT-30
Saintpaul-I-F2 50 -TAGCATCTTTAGTACCAGC-30
Saintpaul-I-CP1 50 -TCCACTGAGCGGAAAAATGCCAGAAAGCTAAAAGGATATACGGG-30
Saintpaul-I-CP2 50 -CTGGATGGCTCTCTGGTGCTTCCGTAGCTTGCAGCGTTTC-30
Saintpaul-I-C1 50 -TCCACTGAGCGGAAAAATGCCAG-30
Saintpaul-I-C2 50 -CTGGATGGCTCTCTGGTGCT-30
Saintpaul-I-D1 50 -GCGACATTGGGTGTAATC-30
Saintpaul-I-D2 50 -GATAAAATAACGTGGCTGG-30
Saintpaul-I-R1 50 -GCGAATAGCGAAACTCACT-30
Saintpaul-I-R2 50 -TGAGCGGGATAGTAAGAAG-30
Saintpaul-II-F1 50 -TTATTACCAGTGCCGCGAT-30
Saintpaul-II-F2 50 -TGTAGCCAGCGTTGTACC-30
Saintpaul-II-CP1 50 -CACCACGTTTTTAGGGCTGATGAAGCGGGCTCTTTTAATGCTAAGT-30
Saintpaul-II-CP2 50 -GACATTTCCTCACCTTCCAGGGCTTCCGTATCAAGGTTATGGG-30
Saintpaul-II-C1 50 -CACCACGTTTTTAGGGCTGATGAAG-30
Saintpaul-II-C2 50 -GACATTTCCTCACCTTCCAGGG-30
(table continues)

710 jmd.amjpathol.org - The Journal of Molecular Diagnostics

 MCDA for Typing of Salmonella Serovars

Table 1 (continued )
Primer name* Sequence
Saintpaul-II-D1 50 -GAAATTTTCCTGGAGCCAGT-30
Saintpaul-II-D2 50 -TGAAGGGATCCTGTTTTCTG-30
Saintpaul-II-R1 50 -CATAAACAATGCTTTTGTTGCC-30
Saintpaul-II-R2 50 -TACCTGATCGATGACACTC-30
Infantis-F1 50 -TTATGGCTGACAACGAGAG-30
Infantis-F2 50 -ATCCAGGTCAAACGCTTGC-30
Infantis-CP1 50 -TCCGACTCTGCGTTTAACGATGCTATTCATCCTGATGTCGCTC-30
Infantis-CP2 50 -TCAAGGCATCGAAAACCTGATCCTGACTGTAGAAAGCACAACACC-30
Infantis-C1 50 -TCCGACTCTGCGTTTAACGATG-30
Infantis-C2 50 -TCAAGGCATCGAAAACCTGATCCT-30
Infantis-D1 50 -ACGACCTCATTTTCTGCC-30
Infantis-D2 50 -TACCGGTGTGACTACCAG-30
Infantis-R1 50 -GTTCGGTAAACGAGAAAGC-30
Infantis-R2 50 -TGAGATGATCCTTCGTGC-30
*Typhimurium, STM4494; Enteritidis-clade B, SEN1384; Enteritidis-clade A/C, R561_RS18155; Virchow, SESV_RS06060; Saintpaul-I, SESPA_RS08460;
Saintpaul-II, SeSPB_A1749; Infantis, L287_11788.
Saintpaul-I, Saintpaul lineage I; Saintpaul-II, Saintpaul lineage II.

Evaluation of the Sensitivity and Specificity of Seven
MCDA Assays in Pure Culture
The sensitivity of each MCDA assay was defined as the
percentage of strains of a targeted serovar being detected as
positive. The sensitivity was evaluated with genomic DNA
templates from 79 target strains (Typhimurium, n Z 11;
Enteritidis, n Z 24; Virchow, n Z 10; Saintpaul, n Z 20;
Infantis, n Z 14) under the same conditions described
above. Specificity was defined as the percentage of strains
from nontargeted serovars being detected as negative and
ideally should be 100%. A panel of 30 strains, including 24
nontarget strains from SARB collection representing 24
serovars and target strains from the other 6 assays, were
used to analyze the specificity of MCDA assays within
Salmonella. An additional eight non-Salmonella strains
were tested for specificity of seven MCDA assays with other
species.
All strains were tested in duplicate at 500 pg/mL level
with two independent runs. Both replicates with a kinetic
graph within 30-minute incubation time were considered as
positive amplification.
Phylogenetic Analysis
Whole-genome sequencing of four SARB Enteritidis
strains was performed by Illumina NextSeq (Illumina,
Scoresby, VIC, Australia). DNA libraries were con-
structed using Nextera XT Sample preparation kit (Illu-
mina Inc., San Diego, CA) and sequenced using the
NextSeq sequencer (Illumina Inc.). FASTQ sequences
were deposited in the National Center for Biotechnology
Information Short Read Archive under the BioProject
PRJNA552918. Raw reads for these strains were de novo
assembled using SPADES version 3.10.1 assembler with
default settings.25 The serovar of the four assembled
SARB Enteritidis genomes was predicted by Salmonella
in Silico Typing Resource26 and SeqSero.27 The se-
quences of flagellin gene fliC were extracted from Seq-
Sero27 database. The genome of Infantis SARB27 strain
was downloaded from the National Center for Biotech-
nology Information GenBank (RefSeq assembly accession
number GCF_000230875.1).
To investigate the phylogeny of observed false-negative
strains and false-positive strains with related serovars,
Parsnp version 1.228 with default parameters was used to
generate phylogenetic trees from genomes. The tree was
visualized using Figtree version 1.4.3.29 A phylogenetic tree
of the flagellin encoding fliC gene sequence was constructed
using Mega X with default parameters by maximum parsi-
mony method with 500 bootstrap replicates.30
Results
Selection of Serovar-Specific Genes for MCDA Products
Targeting Five Serovars
One marker each for monophyletic serovars was selected.
STM4494, SESV_RS06060, and L287_11788 were
selected for Typhimurium, Virchow, and Infantis,
respectively, which were identified and confirmed as
serovar-specific markers previously.21 For polyphyletic
serovar Saintpaul (Saintpaul-I and Saintpaul-II) and par-
aphyletic serovar Enteritidis (Enteritidis-clade B and
Enteritidis-clade A/C), more than one marker was required
to identify the different lineages of the serovars. One
marker each was selected for Saintpaul-I and Saintpaul-II
(SESPA_RS08460 and SeSPB_A1749, respectively).
SEN1384 was selected for Enteritidis-clade B, and
R561_RS18155 was selected for Enteritidis-clade A/C. A

The Journal of Molecular Diagnostics - jmd.amjpathol.org 711

Zhang et al
A B
TGAATCGTCGCTCTTCAATATGAAGGCAACCATGCCCGGTGAATATCCTGAAATGATCGC
>>>>>>>>F1>>>>>>>>>> >>>>>>>>>P1>>>>>>>>> <<<<<<<<D1<<
CACGCTAAACCAGTTCTTCAGCGGTAAACGTATCGAAGCTGAAATTCATCCAGGACAGCG
<<<<<<< <<<<<<<<<C1<<<<<<<<<<<<<<<<<<<<<<<<<R1<<<<<<<
GCAACGTGTCCTACTGGATTCAGGGAATGATCATTCGTTAGATGCGCTTAGCTCCGGCGA
>>>>>>>R2>>>>>>>>>>>>>>>>>>>>>C2>>>>>>>>>>>>> >>>>>>>D2>>>>>
ACATCAGGTGCTGATTATGCTGTTTACGGTACAACGCTGGCTACAGCCCGGCGGTGTTGT
>>>><<<<<<<<<<<P2<<<<<<<<< <<<<<<<<F2<<<<<<<<
C D
ATTGATTTTCATTATAGGGCAGGGAGTCTTGAGATTATAGTTACTCTTGATGGAGTGAAT
>>>>>>>>>>F1>>>>>>>>>>>>>>>>>>P1>>>>>>>>>>> <<<<<<<<<
TCGGATTTTCGATTTTTTTTTGATTGGACTCATTCATTTCGTGTCACTGATGAAGGTGAT
<<D1<<<<<<< <<<<<<<<<C1<<<<<<<<<<<<<<<<<<<<<<R1<<<<
CTGTTGAAAATGTTGGGTGAGCAAAAAGGAAAAATGCGAGTAGGTATTTATAAGGTTGAG
<<<<<< >>>>>>>R2>>>>>>>>>> >>>>>>>>>>>C2>>>>>>>>>>>>>
GACTCATCTTATCTGGAATGGTTTAATGACCAGAGTTTTAATATACATGAAAAAGAGAAA
>>>D2>>>>>>>>>>>>>>> <<<<<<<<<<<P2<<<<<<<<<<
ATTATTCATTATTTGATTGTGACAGTAAATGATATCATTGATGTTTTGTCCTCAGAGTCT
<<<<<<<<<<F2<<<<<<<<
E F
ATGCTCCCGAATCGAATGGTACTTAGCCGTCAGACTGAAGACCAGCTTAAAAAGCTAAAA
>>>>>>>>>F1>>>>>>>> >>>>>>>>>>
GGATATACGGGGATTACACCCAATGTCGCGGCTCGGCTGGCATTTTTCCGCTCAGTGGAG
>F1>>>>>>>><<<<<<<D1<<<<<<<<< <<<<<<<<<<<C1<<<<<<<<<<
AGTGAGTTTCGCTATTCGCCTGAGCGGGATAGTAAGAAGCTGGATGGCTCTCTGGTGCTG
<<<<<<<R1<<<<<<<<<< >>>>>>>>>>R2>>>>>>>>>>>>>>>>C2>>>>>>>>>
GATAAAATAACGTGGCTGGGGGAAACGCTGCAAGCTACGGAGCTGGTACTAAAGATGCTA
>>>>>>>>D2>>>>>>>>> <<<<<<<<P2<<<<<<<<<<<<<<<<<<F2<<<<<<<<<
TATCCGCAGCTAGAGCAGAAAATACTAATTAAAGCATGGGCAGCACATGTTGATGATGGA
G Figure 1 Nucleotide sequence and location of seven multiple cross-
AATGGAATTTTTATGGCTGACAACGAGAGCACGGCTATTCATCCTGATGTCGCTCCGGCA
>>>>>>>>F1>>>>>>>>> >>>>>>>>>P1>>>>>>>>>> <<<<
GAAAATGAGGTCGTCATCGTTAAACGCAGAGTCGGAGCTTTCTCGTTTACCGAACTTGAG
<<D1<<<<<<<<<<<<<<<<<<<<C1<<<<<<<<<<<<<<<<<<<R1<<<<<<<< >>>>
ATGATCCTTCGTGCTCAAGGCATCGAAAACCTGATCCTTACCGGTGTGACTACCAGTGGT
>>R2>>>>>>>>>>>>>>>>>>>>C2>>>>>>>>>>>>>>>>>>>>>>D2>>>>>> <<<
GTTGTGCTTTCTACAGTCGGGCAAGCGTTTGACCTGGATTACCGCCTCATCGTTGTGAGT
<<<<<P2<<<<<<<<<<< <<<<<<<<F2<<<<<<<<<
total of seven MCDA assays based on these markers were
designed to detect these five serovars.
The seven MCDA products were examined in silico using
BLASTN against 2258 genomes used previously21 to
confirm the specificities of each MCDA product. The
BLASTN results showed that each MCDA product was
found in the same genomes as the respective genes with no
additional false positives, in agreement with the previous
study (Supplemental Table S3).21 In all cases, false-positive
genomes were from serovars that were rare in human in-
fections and are not expected to be major limitations on the
applicability of these assays in a clinical setting.
712
ATGAACTCCGGCCTGATAACACTTACGGAGCTGAGGAGGATGACGGGGTTAACCATTTAT
>>>>>>>F1>>>>>>>>>>> >>>>>>>>>P1>>>>>>>>>><<<
TCGACCCGCCACTACCTGGACAAGGCAGAACGTTGTGGGGATGTGTACCAGGCGGGCAGA
<<D1<<<<<<<<<<< <<<<<<<<<<<C1<<<<<<<<< <<<<<<<<<R1<<<<<<<
AGAGGGGGGATTTTCCCGTCAGAAGAGGCTTATCGTGCCTGGAAGAAACAGGCGAAAGTG
>>>>>>>>>R2>>>>>>>>>>>>>>>>>>>C2>>>>>>>>>>> >>>>>>>
GACGCTGACCTGATTTGGAAGCTGCCTGACGGTGAGGTACGTCGTTACGACAGGCACCAC
>>>>>>D2>>><<<<<<<<P2<<<<<<<<<<<<<<<F2<<<<<<<<<<<<
GAAGTAAAATCATTATTAGACCAATCTGCAATTGACCAGTCGGTTAAGGCAGCTCACAAC
>>>>>>>>>F1>>>>>>>>>>>>>>>>>P1>>>>>>>>>>><<<<<<D1<<
GCGAAAGTTTTAAACCCAAGCATAAATAAAAGTTTCAGCACTGGATGGGTGCCACCTAAA
<<<<<<<<< <<<<<<<<<<<<<C1<<<<<<<<<<<< <<<<<<<<R1<<<<<<<
ATTATAAATTATGTTGTGGCATATGATGGGCCAGCACAAATGAATACTGTGTATAACTGG
<< >>>>>>>>>R2>>>>>>>>>>>>>>>>>>C2>>>>>>>>>> >>>>
ATCTTAAATAGTCATCAAACGAATAGGATCCCGTTGCCATCATGGAATCAGCAAACGAAA
>>>D2>>>>>>>>>>> <<<<<<<<<P2<<<<<<<<<<<<<<<<<<<F2<<<<<<<
TATCAAACACCAGGTACAGCACTTGATGGTGTGGTTTTATTAAATAAAGGATTTATAAAG
CGATACGAAATATTTATTACCAGTGCCGCGATGGACTGGCCGGGCTCTTTTAATGCTAAG
>>>>>>>>>F1>>>>>>>> >>>>>>>>>P1>>>>>>>>>
TATCACTGGCTCCAGGAAAATTTCCCCTTCATCAGCCCTAAAAACGTGGTGTTTTGTGGC
> <<<<<<<<<<D1<<<<<<<< <<<<<<<<<<<<C1<<<<<<<<<<< <<<
AACAAAAGCATTGTTTATGCTGACTACCTGATCGATGACACTCCGCGACATTTCCTCACC
<<<<<<R1<<<<<<<<<<< >>>>>>>>>R2>>>>>>>> >>>>>>>>>C2>>>
TTCCAGGGTGAAGGGATCCTGTTTTCTGCCCCCCATAACCTTGATACGGAAGGTTACCGC
>>>>>>>>>>>>>>>>>D2>>>>>>>>> <<<<<<<<<<<P2<<<<<<<<
AGGGTTAATAGCTGGCTGGATGTGGAAACGCTCTTCCTTTCATAA
<<<<<<<<F2<<<<<<<<<
displacement amplification (MCDA) primer sets. The nucleotide se-
quences of the primer set are shown in the context of the target gene. A:
Typhimurium. B: Enteritidis-clade B. C: Enteritidis-clade A/C. D: Virchow.
E: Saintpaul lineage I. F: Saintpaul lineage II. G: Infantis. Left arrowheads
and right arrowheads show complementary and sense sequences,
respectively. The locations of primers are highlighted: red, displacement
primers F1 and F2; green, primers P1 and P2; purple, amplification primers
C1 and C2; yellow, amplification primers D1 and D2; and blue, amplification
primers R1 and R2.19
Evaluation of LoD of the Seven MCDA Assays in Pure
Culture
The LoD of each of the seven MCDA assays designed in
this study and the species identification invA MCDA assay20
was performed on serially diluted genomic DNA. The
amplification curves of the eight MCDA assays are shown
in Figure 2. The seven serovar-specific target strains can be
detected within 8 minutes incubation time at highest con-
centration tested (5 ng/mL) in all seven MCDA assays. The
detection time of invA MCDA assay at this same concen-
tration was 12 minutes.
jmd.amjpathol.org - The Journal of Molecular Diagnostics
 MCDA for Typing of Salmonella Serovars

A B C
110 110 110
100 100 100

Relative fluorescence
Relative fluorescence

Relative fluorescence
90 90 90
80 80 80
70 70 70
60 60 60
50 50 50
40 40 40
30 30 30
20 20 20
10 10 10
0 0 0
0 5 10 15 20 25 30 0 5 10 15 20 25 30 0 5 10 15 20 25 30
Time (minutes) Time (minutes) Time (minutes)

D E F
110 110 110
100 100 100

Relative fluorescence
90 Relative fluorescence 90 90
Relative fluorescence

80 80 80
70 70 70
60 60 60
50 50 50
40 40 40
30 30 30
20 20 20
10 10 10
0 0 0
0 5 10 15 20 25 30 0 5 10 15 20 25 30 0 5 10 15 20 25 30
Time (minutes) Time (minutes) Time (minutes)

G H
110 110
100 100
Relative fluorescence

25 fg
Relative fluorescence

90 90 5 ng
80 80 500 pg 12.5 fg
70 70
60 60 50 pg 6.25 fg
50 50 NTC
5 pg
40 40
30 30 500 fg NEG
20 20 50 fg
10 10
0 0
0 5 10 15 20 25 30 0 5 10 15 20 25 30
Time (minutes) Time (minutes)

Figure 2 Limit of detection (LoD) amplification curves of seven multiple cross-displacement amplification (MCDA) assays. LoD amplification curves of
seven MCDA assays generated with GraphPad Prism software version 7.04 are listed. A: invA. B: Typhimurium. C: Enteritidis-clade B. D: Enteritidis-clade A/C. E:
Virchow. F: Saintpaul lineage I (Saintpaul-I). G: Saintpaul lineage II (Saintpaul-II). H: Infantis. Curves for each concentration of DNA are marked in the figure.
The concentration started with 5 ng per reaction and decreased from left to right as per the color key. Each point on each line is the average of relative
fluorescence of at least six replicates. LoD was defined as the lowest concentration at which all replicates produced final relative fluorescence of >90%. LoD for
each of the assays was as follows: 50 fg (10 copies): Enteritidis-clade B, Enteritidis-clade A/C, Virchow, and Infantis; 25 fg (5 copies): Saintpaul-I and
Saintpaul-II; 12.5 fg (2.5 copies): Typhimurium. NEG, Escherichia coli K12; NTC, no template control.

Enteritidis-clade B, Enteritidis-clade A/C, Virchow, and MCDA assay positive results were observed after the
Infantis MCDA assays had LoD of 50 fg/mL (10 copies). serovar-specific result (Figure 3). Therefore, invA assay
The LoD for Saintpaul MCDA assays (Saintpaul-I and acted as a benchmark for serovar-specific MCDA assay. If
Saintpaul-II) was 25 fg/mL (5 copies), and the LoD of the invA assay provides a positive result after the serovar-
Typhimurium MCDA assay was 12.5 fg/mL (2.5 copies). specific result, the serovar-specific results can be trusted.
Successful detection of targets was also observed at even If the invA assay is positive before the serovar-specific
lower concentrations for seven MCDA assays not for all assay or negative completely, the result is likely to be a
six replicates. For example, Enteritidis-clade A/C with false positive.
R561_RS18155 MCDA assay detected positive amplifi-
cations at 25 and 12.5 fg/mL DNA level with five of the Evaluation of the Sensitivity of Seven MCDA Assays in
six replicates and three of the six replicates, respectively. Pure Culture
To distinguish low copy number positive results from
false positives, the LoD curves of the invA MCDA assay The sensitivity of each MCDA assay was determined
and the seven MCDA assays designed herein were against target strains listed in Supplemental Table S1. All
compared. In all assays and concentrations, the invA target strains were successfully amplified by respective

The Journal of Molecular Diagnostics - jmd.amjpathol.org 713

Zhang et al

MCDA assays, except for Infantis strain SARB27 and four (SARB16, SARB17, and SARB19) lacked both Enteritidis
Enteritidis stains (SARB16, SARB17, SARB18, and gene markers, whereas SARB18 contained Enteritidis-clade
SARB19). Among these five target strains, Infantis strain A/C gene marker R561_RS18155.
SARB27 was negative in the Infantis MCDA assay. For the To confirm the serovar’s identity, serovar prediction of the
four Enteritidis strains, only SARB18 was detected by four assembled SARB Enteritidis genomes was performed
Enteritidis-clade A/C MCDA assay. using Salmonella in Silico Typing Resource26 and SeqSero.27
BLASTN searches of the Infantis-specific gene marker SARB16 and SARB19 were assigned to serovars Duisburg
was performed against SARB27 genome sequence down- and Emek, respectively. SARB17 and SARB18 were predicted
loaded from National Center for Biotechnology as serovar Enteritidis. Phylogenetic tree of genomes con-
Information, and BLASTN results indicated that SARB27 structed using Parsnp revealed that SARB18 was clustered with
lacked Infantis-specific gene marker. BLASTN searches of Enteritidis-clade A/C, whereas SARB16, SARB17, and
the two Enteritidis-specific gene markers were also con- SARB19 were grouped with other serovars (Supplemental
ducted against the four SARB Enteritidis genomes Figure S1). A single-nucleotide polymorphismebased phylo-
sequenced in this study. Three SARB Enteritidis strains genetic tree of fliC (Supplemental Figure S2) indicated that

A B C
30 30 30

R² = 0.8674 R² = 0.8416 R² = 0.7405

Time (minutes)

Time (minutes)

Time (minutes)
20 20 20

10 10 10

0 0 0
10 -5 10 -4 10 -3 10 -2 10 -1 10 0 10 1 10 -5 10 -4 10 -3 10 -2 10 -1 10 0 10 1 10 -5 10 -4 10 -3 10 -2 10 -1 10 0 10 1
Concentration (ng) Concentration (ng) Concentration (ng)

D E F
30 30 30

R² = 0.6646 R² = 0.9055 R² = 0.7133

Time (minutes)

Time (minutes)

Time (minutes)

20 20 20

10 10 10

0 0 0
10 -5 10 -4 10 -3 10 -2 10 -1 10 0 10 1 10 -5 10 -4 10 -3 10 -2 10 -1 10 0 10 1 10 -5 10 -4 10 -3 10 -2 10 -1 10 0 10 1
Concentration (ng) Concentration (ng) Concentration (ng)

G
30

R² = 0.9176
Serovar/Lineage-Specific Genes
Time (minutes)

20
invA
10 invA Average R² = 0.9055

0
10 -5 10 -4 10 -3 10 -2 10 -1 10 0 10 1
Concentration (ng)

Figure 3 Standard curves of seven multiple cross-displacement amplification (MCDA) assays based on average detection times and serial dilutions.
Standard curves of average detection times based on serial dilutions for each of the seven MCDA assays. A: Typhimurium. B: Enteritidis-clade B. C: Enteritidis-
clade A/C. D: Virchow. E: Saintpaul lineage I. F: Saintpaul lineage II. G: Infantis. Closed circle indicates seven MCDA assays, and open circle indicates invA
MCDA assay. Error bars represent the range of all replicates at each concentration. The average detection times for each MCDA assay were linear along the
dilution gradient and had a significant correlation (P < 0.001) with linear standard curves. Parallel lines between each MCDA assay and invA assay were
observed. In all assays and concentrations, the invA positive results were observed after the serovar-specific results.

714 jmd.amjpathol.org - The Journal of Molecular Diagnostics

 MCDA for Typing of Salmonella Serovars

Enteritidis-clade A/C and Enteritidis-clade B were identical. In
contrast, SARB17 fliC was not grouped with Enteritidis fliC,
and they differed by three single-nucleotide polymorphisms,
with one being a nonsynonymous single-nucleotide poly-
morphism. It was concluded that SARB17 does not represent
Salmonella Enteritidis.
Combined with the serovar prediction results and
phylogenetic analysis, SARB18 was assigned to Enteritidis-
clade A/C, whereas SARB16, SARB17, and SARB19 were
assigned to other serovars. Therefore, SARB16, SARB17,
and SARB19 were excluded from the target strains of
Enteritidis-clade A/C and Enteritidis-clade B. The sensi-
tivity of each MCDA assay is defined as positive rate of
targeted serovar strains. The sensitivity of seven MCDA
assays varied from 92.9% to 100% (Table 2).
Evaluation of the Specificity of Seven MCDA Assays in
Pure Culture
The specificity of each MCDA assay was evaluated against
eight non-Salmonella strains and 30 Salmonella strains
(Supplemental Table S1). The specificity of seven MCDA
assays with other species was 100%, whereas the specificity
of seven MCDA assays with nontargeted Salmonella strains
varied from 93.3% to 100% (Table 2). Typhimurium and
Enteritidis-clade A/C MCDA assays produced two false-
positive results each. One Derby strain (SARB9) and one
Infantis strain (SARB27) were detected by Typhimurium
MCDA assay, which led to the specificity of 93.3% for
Typhimurium MCDA assay. One strain each for Dublin and
Gallinarum (SARB13 and SARB21, respectively) were
amplified by Enteritidis-clade A/C MCDA assay; therefore,
the specificity of Enteritidis-clade A/C MCDA assay was
93.3%.
BLASTN result showed that Infantis SARB27 contained
Typhimurium gene marker STM4494. Phylogenetic tree
constructed using Parsnp indicated that SARB27 did not
cluster with Typhimurium and was not part of the main
Infantis clade (Supplemental Figure S3).
Discussion
Salmonella is a common cause of foodborne illness in
Australia.1 Rapid, accurate, and sensitive detection and
identification of Salmonella serovars have been essential for
clinical diagnosis and public health surveillance. In recent
decades, PCR-based and real-time PCR techniques have
frequently been used to detect and differentiate Salmonella
serovars.10,16,31e38 LAMP was also used as an alternative
method for rapid and sensitive detection of specific gene
targets for identification of Salmonella.13,39e41 A few
LAMP assays can also differentiate Typhimurium and
Enteritidis from other Salmonella serovars.12,14,15,40,42e46
The limit of detection for the reported Salmonella LAMP
assays ranged from 5 fg to 5.6 ng genomic DNA per reac-
tion in pure culture, and the comparison between LAMP and
PCR or real-time PCR performed in the same study showed
that LAMP was 10-fold to 10,000-fold more sensitive.47
Another isothermal amplification technique, MCDA, was
developed in 2015 with at least 160-fold and 16-fold higher
analytical sensitivity than PCR and LAMP, respectively.19
MCDA has been employed to detect Salmonella at the
species level but not the serovar level. MCDA can detect
Salmonella at 6.25 fg genomic DNA per reaction, at least
400-fold more sensitive than real-time PCR.20
With the increasing uptake of culture-independent direct
testing for the diagnosis of Salmonella enterocolitis, culture-
independent serovar detection and identification targeting
serovar-specific gene markers would be useful as current
serotyping is mostly performed on DNA from purified iso-
lates. Several genes [STM4493, STM4495, STM4497, typh,
lygD (SEN1383), sdfI, safA, prot6E, and sefA] have been
used to develop PCR-based methods and LAMP assays for

Table 2 The Sensitivity and Specificity of the Seven MCDA Assays

Nontarget
Specific gene Target strains Non-Salmonella Sensitivity, Specificity Specificity
Target serovars marker (locus tag)21 strains (n Z 30) strains (n Z 8) %* (within Salmonella), %* (non-Salmonella), %*
Typhimurium STM4494 11/11 2/30y 0 100 93.3 100
Enteritidis-clade B SEN1384 9/9 0 0 100 100 100
Enteritidis-clade R561_RS18155 12/12 2/30z 0 100 93.3 100
A/C
Virchow SESV_RS6060 10/10 0 0 100 100 100
Saintpaul-I SeSPB_A1352 18/18 0 0 100 100 100
Saintpaul-II SeSPA_A1749 2/2 0 0 100 100 100
Infantis L287_11788 13/14x 0 0 92.9 100 100
*Sensitivity: (number of positive of target strains)/(number of total target strains). Specificity: 1  (number of positive of nontarget strains)/(number of
total nontarget strains).
y
False-positive strains: SARB9 (Derby) and SRAB27 (Infantis).
z
False-positive strains: SRAB13 (Dublin) and SARB21 (Gallinarum).
x
Infantis strain SARB27 was negative to Infantis MCDA assay.
MCDA, multiple cross-displacement amplification; Saintpaul-I, Saintpaul lineage I; Saintpaul-II, Saintpaul lineage II.

The Journal of Molecular Diagnostics - jmd.amjpathol.org 715

Zhang et al
typing Typhimurium and Enteritidis.12,14,36,42,46,48 Previ-
ously,21 genes STM4493, STM4494, and STM4497 were
identified as potential Typhimurium gene markers but were
present in Infantis SARB27. However, with genomic data of
2258 genomes, the specificities of gene markers STM4493
and STM4497 were 93.22% and 92.56%, respectively,
whereas the specificity of gene marker STM4494 was
94.11%, slightly higher than STM4493 and STM4497. A
novel PCR was also developed for identification of Infantis
based on flagellin fljB gene,49 and Virchow-specific primers
were designed for the detection of Virchow.50
These results indicate that a set of MCDA assays tar-
geting seven serovar/lineage-specific gene markers can
rapidly detect and serotype the five most common and
clinically relevant Salmonella serovars. Seven serovar/
lineage-specific gene markers, STM4494, SEN1384/
R561_RS18155, SESV_RS06060, SeSPB_A1749/SeS-
PA_A1352, and L287_11788,21 were used to develop
Typhimurium, Enteritidis-clade B/Enteritidis-clade A/C,
Virchow, Saintpaul-I/Saintpaul-II, and Infantis MCDA as-
says, respectively. The initial evaluation of the seven
MCDA assays was accomplished by using Salmonella
speciesespecific gene invA MCDA assay20 as positive
control. Multiple means can be used to display the correct
amplification of MCDA.19 In this study, real-time fluores-
cence measurement was used to detect the seven MCDA
products. The seven MCDA assays were sensitive, with an
LoD of 50 fg (10 copies) with pure DNA, and the assays
were rapid, with a result detectable within 8 minutes at the
highest concentration tested. The assays also were highly
specific to target serovars, and the positive reactions were
monitored in a real-time format. These assays provided a
unified approach using serovar-specific gene markers ob-
tained from extensive in silico genome analysis21 and a
common MCDA assay platform.19 These gene markers can
also be used to develop assays on other platforms.
The speed and LoD of the MCDA assay compared
favorably with published LAMP assays. Previous
studies12,14 evaluated LAMP assays for detection of
Typhimurium and Enteritidis by targeting STM4497 and
safA genes, respectively. The fastest detection time in these
studies was approximately 24 minutes, and the LoDs were
4.38 pg/mL for Typhimurium and 1.44 pg/mL for Enteritidis
in pure culture. By comparison, the Typhimurium MCDA
targeting STM4494 and Enteritidis-clade B MCDA target-
ing SEN1384 were nearly 15 minutes faster than LAMP. In
addition, these results were produced with 50 fg/mL of pure
DNA, at least 87-fold more sensitive than LAMP for
Typhimurium MCDA assay and 29-fold more sensitive than
LAMP for Enteritidis-clade B MCDA assay.
MCDA assays could also produce positive results at even
lower concentrations of 6.25 fg/mL (1.25 copies) genomic
DNA within 28 minutes, although these results were not as
consistent. This inconsistency may be due to the extremely
low copy number of the sample, resulting in no template
being present by chance during template sampling.
716
Inconsistent amplification with few copies of the template
within 28 minutes indicated that any amplification later than
28 minutes incubation time was unreliable. Therefore, a
30-minute incubation time was used to set the cutoff value
for evaluation of sensitivity and specificity of the seven
MCDA assays.
In some reactions, amplification can occur at time points
that most likely contained few copies of the template. These
amplifications may be due to a real (true-positive) detection
of dilute target DNA or may be caused by inefficient
amplification of nontarget DNA (false positive). To differ-
entiate between these scenarios, the invA MCDA assay can
be used as an outer limit on the amplification time of a true-
positive result. The sample will only be considered suc-
cessfully serotyped by the newly designed seven MCDA
assays when it is positive to both invA and the serovar-
specific target and the amplification of target occurs before
invA.
Three strains (SARB16, SARB17, and SARB19) from
the Salmonella reference collection B, which was assem-
bled during the early 1980s,23 were serotyped as Enter-
itidis. However, SARB16 and SARB19 were assigned as
other serovars by whole-genome sequencingebased sero-
var prediction methods Salmonella in Silico Typing
Resource and SeqSero. These three strains were not
closely related to Enteritidis on a genome-based phylo-
genetic tree. Previous studies also showed that SARB17
and SARB19 were distantly related to the vast majority of
Enteritidis isolates on phylogenetic trees.51,52 Phylogenetic
analysis of flagellin gene fliC indicated that the SARB17
fliC gene was distinct from Enteritidis. Consequently,
SARB16, SARB17, and SARB19 were excluded as
target strains for Enteritidis. The genomic signatures that
were targeted by the MCDA assays can provide more
useful serovar identification than traditional serotyping,
especially for those strains with the same serovar but
different evolutionary history.
All seven MCDA assays had high overall sensitivity,
ranging from 92.9% to 100%. A false-negative result only
occurred in SARB27 in the Infantis MCDA assay. Infantis
SARB27, which was isolated in Senegal, was distantly
related to the globally distributed Infantis.23 All pure culture
from clinical strains was correctly detected and identified by
the Infantis MCDA assay.
Only two strains were used for evaluating the sensitivity
of Saintpaul-II MCDA assay. More strains were not tested
as Saintpaul-II is a rare minor lineage, making strain
acquisition difficult. From previous genomic analysis21
and a further analysis of 291 genomes, the lineage has
low diversity and the Saintpaul-II specific gene has 100%
in silico specificity and sensitivity. Therefore, the assay
most likely will be effective in detecting Saintpaul-II
isolates.
The false-positive Derby strain in the Typhimurium
assay is expected on the basis of previous genomic anal-
ysis in the previous study, which showed that Derby was a
jmd.amjpathol.org - The Journal of Molecular Diagnostics

potential false positive of the Typhimurium gene marker
STM4494.21 An Australian potential false-positive rate
was estimated for Derby detection in the Typhimurium
assay by combining the Derby frequency in Australian
human infections and the rate of false positives from
Derby genomes. This potential false-positive rate was
<0.41% in human infections in Australia.21 Therefore,
Derby should not be a major limitation to the application
of this assay to human clinical samples in culture-
independent Salmonella serotyping. For veterinary and
agricultural samples, the false-positive rate may increase,
depending on the prevalence of Derby in the source animal
population. The prevalence (11.2%) of Derby in pigs and
raw pork products in New South Wales53 would result in a
potential false-positive rate of 3.1%. However, the routine
surveillance of Salmonella infections in animal population
is limited in Australia,53 which makes accurate determi-
nation of likely false-positive rates difficult.
The false positives in the Enteritidis-clade A/C MCDA
assay from serovars Dublin and Gallinarum were also ex-
pected from previous genomic analysis.21 Dublin is one of
the most prevalent serovars in cattle.54 However, the fre-
quency of Dublin in human infections is <1.5% in
Australia. The rate of genome-based false positives of
Dublin was 2.78% for Enteritidis-clade A/C gene marker
R561_RS18155.21 The Australian potential false-positive
rate for Dublin in the Enteritidis-clade A/C MCDA assay
would be <0.04% in human infections. Among cattle and
raw beef products (New South Wales prevalence of
33.4%), the potential false-positive rate would be 0.93%.
Gallinarum is restricted to poultry reservoir in many
developing countries37,55e58 and remains rare in Australia.
The potential false-positive rate with Gallinarum in
Enteritidis-clade A/C MCDA assay of human samples is
therefore negligible.
In conclusion, seven MCDA assays were developed to
amplify target serovar gene markers successfully from the
five most frequent Salmonella serovars in Australia. These
assays demonstrated an LoD of 50 fg per reaction (10 copies
of target DNA) from pure culture and were specific to the
target serovars. The assay is time efficient, is isothermal,
and can provide test results in as little as 8 minutes. The
MCDA assays developed offer a rapid, accurate, and sen-
sitive serotyping method, which will be useful also for
culture-independent serotyping of common Salmonella
serovars directly from clinical samples. The performance of
the MCDA assays warrants further validation on clinical
and environmental samples.
Acknowledgment
We thank Peter Howard (New South Wales Enteric Refer-
ence Laboratory, NSW Health Pathology) for providing
access to examples of clinical strains.
The Journal of Molecular Diagnostics - jmd.amjpathol.org
MCDA for Typing of Salmonella Serovars
Supplemental Data
Supplemental material for this article can be found at
https://doi.org/10.1016/j.jmoldx.2020.02.006.
References
1. Ford L, Moffatt CRM, Fearnley E, Miller M, Gregory J, Sloan-
Gardner TS, Polkinghorne BG, Bell R, Franklin N, Williamson DA,
Glass K, Kirk MD: The epidemiology of Salmonella enterica out-
breaks in Australia, 2001e2016. Front Sustain Food Syst 2018, 2:86
2. Majowicz SE, Musto J, Scallan E, Angulo FJ, Kirk M, O’brien SJ,
Jones TF, Fazil A, Hoekstra RM; International Collaboration on
Enteric Disease “Burden of Illness” Studies: The global burden of
nontyphoidal Salmonella gastroenteritis. Clin Infect Dis 2010, 50:
882e889
3. Ford L, Glass K, Veitch M, Wardell R, Polkinghorne B, Dobbins T,
Lal A, Kirk MD: Increasing incidence of Salmonella in Australia,
2000e2013. PLoS One 2016, 11:e0163989
4. Kirk M, Ford L, Glass K, Hall G: Foodborne illness, Australia, circa
2000 and circa 2010. Emerg Infect Dis 2014, 20:1857e1864
5. OzFoodNet-Working-Group: Monitoring the incidence and causes of
diseases potentially transmitted by food in Australia: annual report of
the OzFoodNet network, 2009. Commun Dis Intell Q Rep 2010, 34:
396e426
6. OzFoodNet-Working-Group: Monitoring the incidence and causes of
diseases potentially transmitted by food in Australia: annual report of
the OzFoodNet network, 2010. Commun Dis Intell Q Rep 2012, 36:
E213eE241
7. OzFoodNet-Working-Group: Monitoring the incidence and causes of
diseases potentially transmitted by food in Australia: annual report of
the OzFoodNet network, 2011. Commun Dis Intell Q Rep 2015, 39:
E236eE264
8. ISO-1: Microbiology of the Food Chain-Horizontal Method for the
Detection, Enumeration and Serotyping of Salmonella-Part 1: Detec-
tion of Salmonella spp. Geneva, Switzerland: International Organiza-
tion for Standardization, 2017
9. Andrews WH, Jacobson A, Hammack T: Bacteriological Analytical
Manual (BAM). Washington, DC: U.S. Food and Drug Administration,
2011
10. Hein I, Flekna G, Krassnig M, Wagner M: Real-time PCR for the
detection of Salmonella spp. in food: an alternative approach to a
conventional PCR system suggested by the FOOD-PCR project. J
Microbiol Methods 2006, 66:538e547
11. Josefsen MH, Krause M, Hansen F, Hoorfar J: Optimization of a
12-hour TaqMan PCR-based method for detection of Salmonella
bacteria in meat. Appl Environ Microbiol 2007, 73:3040e3048
12. Azinheiro S, Carvalho J, Prado M, Garrido-Maestu A: Evaluation of
different genetic targets for Salmonella enterica serovar Enteriditis
and Typhimurium, using Loop-Mediated Isothermal AMPlification
for detection in food samples. Front Sustain Food Syst 2018, 2:5
13. Domesle KJ, Yang Q, Hammack TS, Ge B: Validation of a Salmo-
nella loop-mediated isothermal amplification assay in animal food. Int
J Food Microbiol 2018, 264:63e76
14. Garrido-Maestu A, Fuciños P, Azinheiro S, Carvalho J, Prado M:
Systematic loop-mediated isothermal amplification assays for rapid
detection and characterization of Salmonella spp., Enteritidis and
Typhimurium in food samples. Food Contr 2017, 80:297e306
15. Hu L, Ma LM, Zheng S, He X, Hammack TS, Brown EW, Zhang G:
Development of a novel loop-mediated isothermal amplification
(LAMP) assay for the detection of Salmonella ser. Enteritidis from
egg products. Food Contr 2018, 88:190e197
16. Kasturi KN, Drgon T: Real-time PCR method for detection of Sal-
monella spp. in environmental samples. Appl Environ Microbiol
2017, 83:e00644
717
Zhang et al
17. Kim H, Park S, Lee T, Nahm B, Chung Y, Seo K, Kim H: Identifi-
cation of Salmonella enterica serovar Typhimurium using specific
PCR primers obtained by comparative genomics in Salmonella
serovars. J Food Prot 2006, 69:1653e1661
18. Siala M, Barbana A, Smaoui S, Hachicha S, Marouane C,
Kammoun S, Gdoura R, Messadi-Akrout F: Screening and detecting
Salmonella in different food matrices in Southern Tunisia using a
combined enrichment/real-time PCR method: correlation with con-
ventional culture method. Front Microbiol 2017, 8:2416
19. Wang Y, Wang Y, Ma A-J, Li D-X, Luo L-J, Liu D-X, Jin D, Liu K,
Ye C-Y: Rapid and sensitive isothermal detection of nucleic-acid
sequence by multiple cross displacement amplification. Sci Rep
2015, 5:11902
20. Wang Y, Wang Y, Zhang L, Liu D, Luo L, Li H, Cao X, Liu K, Xu J,
Ye C: Multiplex, rapid, and sensitive isothermal detection of nucleic-
acid sequence by endonuclease restriction-mediated real-time multi-
ple cross displacement amplification. Front Microbiol 2016, 7:753
21. Zhang X, Payne M, Lan R: In silico identification of serovar-specific
genes for Salmonella serotyping. Front Microbiol 2019, 10:835
22. Beltran P, Plock SA, Smith NH, Whittam TS, Old DC, Selander RK:
Reference collection of strains of the Salmonella typhimurium com-
plex from natural populations. J Gen Microbiol 1991, 137:601e606
23. Boyd EF, Wang F-S, Beltran P, Plock SA, Nelson K, Selander RK:
Salmonella reference collection B (SARB): strains of 37 serovars of
subspecies I. J Gen Microbiol 1993, 139:1125e1132
24. Pang S, Octavia S, Feng L, Liu B, Reeves PR, Lan R, Wang L:
Genomic diversity and adaptation of Salmonella enterica serovar
Typhimurium from analysis of six genomes of different phage types.
BMC Genomics 2013, 14:718
25. Bankevich A, Nurk S, Antipov D, Gurevich AA, Dvorkin M,
Kulikov AS, Lesin VM, Nikolenko SI, Pham S, Prjibelski AD:
SPAdes: a new genome assembly algorithm and its applications to
single-cell sequencing. J Comput Biol 2012, 19:455e477
26. Yoshida CE, Kruczkiewicz P, Laing CR, Lingohr EJ, Gannon VP,
Nash JH, Taboada EN: The Salmonella in silico typing resource
(SISTR): an open web-accessible tool for rapidly typing and sub-
typing draft Salmonella genome assemblies. PLoS One 2016, 11:
e0147101
27. Zhang S, Yin Y, Jones MB, Zhang Z, Kaiser BLD, Dinsmore BA,
Fitzgerald C, Fields PI, Deng X: Salmonella serotype determination
utilizing high-throughput genome sequencing data. J Clin Microbiol
2015, 53:1685e1692
28. Treangen TJ, Ondov BD, Koren S, Phillippy AM: The Harvest suite
for rapid core-genome alignment and visualization of thousands of
intraspecific microbial genomes. Genome Biol 2014, 15:524
29. Schneider S, Roessli D, Excoffier L: Arlequin: a software for popu-
lation genetics data analysis. Evol Bioinform Online 2000, 2:
2496e2497
30. Kumar S, Stecher G, Li M, Knyaz C, Tamura K: MEGA X: molecular
evolutionary genetics analysis across computing platforms. Mol Biol
Evol 2018, 35:1547e1549
31. Akiba M, Kusumoto M, Iwata T: Rapid identification of Salmonella
enterica serovars, Typhimurium, Choleraesuis, Infantis, Hadar,
Enteritidis, Dublin and Gallinarum, by multiplex PCR. J Microbiol
Methods 2011, 85:9e15
32. Bugarel M, Tudor A, Loneragan GH, Nightingale KK: Molecular
detection assay of five Salmonella serotypes of public interest:
Typhimurium, Enteritidis, Newport, Heidelberg, and Hadar. J
Microbiol Methods 2017, 134:14e20
33. Kim S, Frye JG, Hu J, Fedorka-Cray PJ, Gautom R, Boyle DS:
Multiplex PCR-based method for identification of common clinical
serotypes of Salmonella enterica subsp. enterica. J Clin Microbiol
2006, 44:3608e3615
34. Park SH, Kim HJ, Cho WH, Kim JH, Oh MH, Kim SH, Lee BK,
Ricke SC, Kim HY: Identification of Salmonella enterica subspecies
I, Salmonella enterica serovars Typhimurium, Enteritidis and Typhi
using multiplex PCR. FEMS Microbiol Lett 2009, 301:137e146
718
35. Rodrıguez-Lázaro D, Hernández M, Esteve T, Hoorfar J, Pla M: A
rapid and direct real time PCR-based method for identification of
Salmonella spp. J Microbiol Methods 2003, 54:381e390
36. Shanmugasundaram M, Radhika M, Murali H, Batra H: Biotech-
nology: detection of Salmonella enterica serovar Typhimurium by
selective amplification of fliC, fljB, iroB, invA, rfbJ, STM2755,
STM4497 genes by polymerase chain reaction in a monoplex and
multiplex format. World J Microbiol Biotechnol 2009, 25:
1385e1394
37. Xiong D, Song L, Pan Z, Jiao X: Identification and discrimination of
Salmonella enterica serovar Gallinarum biovars Pullorum and Galli-
narum based on a one-step multiplex PCR assay. Front Microbiol
2018, 9:1718
38. Xiong D, Song L, Tao J, Zheng H, Zhou Z, Geng S, Pan Z, Jiao X:
An efficient multiplex PCR-based assay as a novel tool for accurate
inter-serovar discrimination of Salmonella Enteritidis, S. Pull-
orum/Gallinarum and S. Dublin. Front Microbiol 2017, 8:420
39. Ohtsuka K, Yanagawa K, Takatori K, Hara-Kudo Y: Detection of
Salmonella enterica in naturally contaminated liquid eggs by loop-
mediated isothermal amplification, and characterization of Salmo-
nella isolates. Appl Environ Microbiol 2005, 71:6730e6735
40. Okamura M, Ohba Y, Kikuchi S, Suzuki A, Tachizaki H,
Takehara K, Ikedo M, Kojima T, Nakamura M: Loop-mediated
isothermal amplification for the rapid, sensitive, and specific detection
of the O9 group of Salmonella in chickens. Vet Microbiol 2008, 132:
197e204
́
41. Ziros PG, Kokkinos PA, Papanotas K, Vantarakis A: Loop-mediated
isothermal amplification (LAMP) for the detection of Salmonella spp.
isolated from different food types. J Microbiol Biotechnol Food Sci
2012, 2:152e161
42. Gong J, Zhuang L, Zhu C, Shi S, Zhang D, Zhang L, Yu Y, Dou X,
Xu B, Wang C: Loop-mediated isothermal amplification of the sefA
gene for rapid detection of Salmonella Enteritidis and Salmonella
Gallinarum in chickens. Foodborne Pathog Dis 2016, 13:177e181
43. Okamura M, Ohba Y, Kikuchi S, Takehara K, Ikedo M, Kojima T,
Nakamura M: Rapid, sensitive, and specific detection of the O4 group
of Salmonella enterica by loop-mediated isothermal amplification.
Avian Dis 2009, 53:216e221
44. Pavan Kumar P, Agarwal R, Thomas P, Sailo B, Prasannavadhana A,
Kumar A, Kataria J, Singh D: Rapid detection of Salmonella enterica
subspecies enterica serovar Typhimurium by loop mediated
isothermal amplification (LAMP) test from field chicken meat sam-
ples. Food Biotechnol 2014, 28:50e62
45. Ravan H, Yazdanparast R: Development of a new loop-mediated
isothermal amplification assay for prt (rfbS) gene to improve the
identification of Salmonella serogroup D. World J Microbiol Bio-
technol 2012, 28:2101e2106
46. Yang JL, Ma GP, Yang R, Yang SQ, Fu LZ, Cheng AC, Wang MS,
Zhang SH, Shen KF, Jia RY, Deng SX, Xu ZY: Simple and rapid
detection of Salmonella serovar Enteritidis under field conditions by
loop-mediated isothermal amplification. J Appl Microbiol 2010, 109:
1715e1723
47. Yang Q, Domesle KJ, Ge B: Loop-mediated isothermal amplification
for Salmonella detection in food and feed: current applications and
future directions. Foodborne Pathog Dis 2018, 15:309e331
48. Chen Z, Zhang K, Yin H, Li Q, Wang L, Liu Z: Detection of
Salmonella and several common Salmonella serotypes in food by
loop-mediated isothermal amplification method. Food Sci Human
Wellness 2015, 4:75e79
49. Kardos G, Farkas T, Antal M, Nogrady N, Kiss I: Novel PCR assay
for identification of Salmonella enterica serovar Infantis. Lett Appl
Microbiol 2007, 45:421e425
50. Chiang Y-C, Wang H-H, Ramireddy L, Chen H-Y, Shih C-M, Lin C-
K, Tsen H-Y: Designing a biochip following multiplex polymerase
chain reaction for the detection of Salmonella serovars Typhimurium,
Enteritidis, Infantis, Hadar, and Virchow in poultry products. J Food
Drug Anal 2018, 26:58e66
jmd.amjpathol.org - The Journal of Molecular Diagnostics

51. Deng X, Desai PT, den Bakker HC, Mikoleit M, Tolar B, Trees E,
Hendriksen RS, Frye JG, Porwollik S, Weimer BC: Genomic
epidemiology of Salmonella enterica serotype Enteritidis based on
population structure of prevalent lineages. Emerg Infect Dis 2014, 20:
1481e1489
52. Zou Q-H, Li R-Q, Liu G-R, Liu S-L: Genotyping of Salmonella with
lineage-specific genes: correlation with serotyping. Int J Infect Dis
2016, 49:134e140
53. Simpson KM, Hill-Cawthorne GA, Ward MP, Mor SM: Diversity of
Salmonella serotypes from humans, food, domestic animals and
wildlife in New South Wales, Australia. BMC Infect Dis 2018, 18:623
54. Selander RK, Smith N, Li J, Beltran P, Ferris K, Kopecko D,
Rubin F: Molecular evolutionary genetics of the cattle-adapted
serovar Salmonella dublin. J Bacteriol 1992, 174:3587e3592
55. Jones MA, Wigley P, Page KL, Hulme SD, Barrow PA: Salmonella
enterica serovar Gallinarum requires the Salmonella pathogenicity
The Journal of Molecular Diagnostics - jmd.amjpathol.org
MCDA for Typing of Salmonella Serovars
island 2 type III secretion system but not the Salmonella pathoge-
nicity island 1 type III secretion system for virulence in chickens.
Infect Immun 2001, 69:5471e5476
56. Kang MS, Kwon YK, Jung BY, Kim A, Lee KM, An BK, Song EA,
Kwon JH, Chung GS: Differential identification of Salmonella
enterica subsp. enterica serovar Gallinarum biovars Gallinarum and
Pullorum based on polymorphic regions of glgC and speC genes. Vet
Microbiol 2011, 147:181e185
57. Pal S, Dey S, Batabyal K, Banerjee A, Joardar SN, Samanta I,
Isore DP: Characterization of Salmonella Gallinarum isolates from
backyard poultry by polymerase chain reaction detection of invasion
(invA) and Salmonella plasmid virulence (spvC) genes. Vet World
2017, 10:814e817
58. Uzzau S, Brown DJ, Wallis T, Rubino S, Leori G, Bernard S,
Casadesús J, Platt DJ, Olsen JE: Host adapted serotypes of Salmo-
nella enterica. Epidemiol Infect 2000, 125:229e255
719
Supplemental Figure S1 Phylogenetic relationship of four Salmonella Reference Collection B (SARB) Enteritidis strains. The single-nucleotide
polymorphismebased phylogenetic tree shows the genetic relationship of four SARB Enteritidis strains with Enteritidis strains. SARB18 was grouped with
Enteritidis-clade A/C isolates, whereas SARB16, SARB17, and SARB19 were grouped with other serovars.

Supplemental Figure S2 Phylogenetic relationship of Salmonella Reference Collection B (SARB) 17, Enteritidis, and other serovars. The single-
nucleotide polymorphismebased phylogenetic tree constructed in MEGA X by maximum parsimony method for flagellin gene fliC of SARB17, Enteritidis-
clade A/C, Enteritidis-clade B, and other closely related to Enteritidis serovars. Tree lengths are indicated at the nodes. Enteritidis-clade B and
Enteritidis-clade A/C are identical. SARB17 is not related to Enteritidis and is the same distance or further away than four other serovars.

Supplemental Figure S3 Phylogenetic relationship of Infantis strain Salmonella Reference Collection B (SARB) 27. The genome-based phylogenetic tree
constructed by Parsnp shows that SARB27 does not cluster with either Typhimurium or Infantis. The genomes used to build this tree were from a previous
study,21 except Infantis SARB27. The genome of Infantis SARB27 (RefSeq assembly accession number GCF_000230875.1) was downloaded from National Center
for Biotechnology Information (https://www.ncbi.nlm.nih.gov).