Multiplex-PCR for simultaneous detection of 3
bacterial fish pathogens, Flavobacterium columnare,
Edwardsiella ictaluri, and Aeromonas hydrophila
Victor S. Panangala1,*, Craig A. Shoemaker1, Vicky L. van Santen2, Kevin Dybvig3,
Phillip H. Klesius1
1Aquatic Animal Health Research Unit, US Department of Agriculture, Agricultural Research Service,
Aquatic Animal Health Research Unit, PO Box 952, Auburn, Alabama 36831-0952, USA
2Department of Pathobiology, College of Veterinary Medicine, Auburn University, Auburn, Alabama 36830, USA
3Department of Genetics, University of Alabama at Birmingham, Birmingham, Alabama 35294, USA
ABSTRACT: A multiplex PCR (m-PCR) method was developed for simultaneous detection of 3 important
fish pathogens in warm water aquaculture. The m-PCR to amplify target DNA fragments from
Flavobacterium columnare (504 bp), Edwardsiella ictaluri (407 bp) and Aeromonas hydrophila (209
bp) was optimized by adjustment of reaction buffers and a touchdown protocol. The lower detection
limit for each of the 3 bacteria was 20 pg of nucleic acid template from each bacteria per m-PCR reaction
mixture. The sensitivity threshold for detection of the 3 bacteria in tissues ranged between 3.4 ×
102 and 2.5 × 105 cells g–1 of tissue (channel catfish Ictalurus punctatus Rafinesque). The diagnostic
sensitivity and specificity of the m-PCR was evaluated with 10 representative isolates of each of the
3 bacteria and 11 other Gram-negative and 2 Gram-positive bacteria that are taxonomically related
or ubiquitous in the aquatic environment. Except for a single species (A. salmonicida subsp. salmonicida),
each set of primers specifically amplified the target DNA of the cognate species of bacteria.
m-PCR was compared with bacteriological culture for identification of bacteria in experimentally
infected fish. The m-PCR appears promising for the rapid, sensitive and simultaneous detection of
Flavobacterium columnare, E. ictaluri and A. hydrophila in infected fish compared to the timeconsuming
traditional bacteriological culture techniques.
KEY WORDS: Multiplex-PCR · Fish · Edwardsiella · Flavobacterium · Aeromonas
Resale or republication not permitted without written consent of the publisher
Dis Aquat Org 74: 199–208, 2007
most important bacterial diseases affecting the aquaculture
industry in the USA (United States Department
of Agriculture 2003). The prevalence and impact of
septicemic disease in fish caused by Aeromonas
hydrophila (referred to as motile aeromonad septicemia,
MSA), is less known. However, A. hydrophila
is ubiquitous in aquatic ecosystems (Holmes et al.
1996) and recent studies have shown that catfish
latently infected with A. hydrophila (carriers) could
shed the organism when co-infected with Edwardsiella
ictaluri (Nusbaum & Morrison 2002). Since all 3 species
of bacteria (Flavobacterium columnare, E. ictaluri, and
A. hydrophila) are ubiquitous in the aquatic environment
and fish are reared in extensive pond acreages
with high stocking densities (20 000 to 30 000 fish ha–1),
the occurrence of coinfections or multiple infections in
the same host should not be overlooked (Jack et al.
1992). Traditional methods of diagnosing bacterial
infections using culture techniques require several
days to arrive at a definitive diagnosis, resulting in
delayed implementation of control measures and
increased potential for spreading of disease. Because
PCR can target unique genetic sequences of microorganisms,
PCR-based nucleic acid amplification techniques
have gained recognition as rapid, sensitive, and
specific methods for detection of disease-causing
pathogens in various aquaculture species (Vantarakis
et al. 2000, Del Cero et al. 2002, Bader et al. 2003,
Bilodeau et al. 2003). When multiple bacterial
pathogens are likely to occur, as in the aquatic environment,
amplification of multiple target genes in a
single reaction mixture is possible with the multiplex
PCR (m-PCR) method (Brasher et al. 1998, Del Cero et
al. 2002, Panicker et al. 2004), thus reducing cost, time
and effort without compromising the test utility. In this
study, we developed an m-PCR method for simultaneous
identification of F. columnare, E. ictaluri, and
A. hydrophila, 3 of the most important bacterial
pathogens causing extensive losses in the channel catfish
aquaculture industry.
MATERIALS AND METHODS
Bacteria isolates and culture conditions. The bacterial
isolates used in this study are listed in Table 1. The
majority of the fish-pathogenic bacteria were originally
isolated from diseased fish, characterized into
their respective genera and species using standard
methods (Arias et al. 2004, Panangala et al. 2005), and
maintained in the archived culture repository of the
Aquatic Animal Health Research Unit, Auburn,
Alabama, USA. Additional cultures were kindly provided
by John M. Grizzle (Department of Fisheries and
Allied Aquaculture, Auburn University, Alabama,
USA), Andrew E. Goodwin (Department of Aquaculture/
Fisheries, University of Arkansas, Pine Bluff,
Arkansas, USA) and Ronald D. Schultz (Department of
Pathobiology, University of Wisconsin, Madison, Wisconsin,
USA). The bacterial isolates were cultured
directly from glycerol stocks on brain heart infusion
agar or blood agar plates (tryptic soy agar with 5% v/v
defibrinated sheep blood: Difco) for Edwardsiella
ictaluri and Aeromonas hydrophila, and Shieh agar
(Shieh 1980) for Flavobacterium columnare. Single
colonies picked after incubation for 36 h were transferred
to 10 ml brain heart infusion broth and cultured
to log-phase growth at 28°C in a shaker water bath.
For isolation of bacteria from tissues of experimentally
infected fish, Shieh medium modified by replacing
peptone with tryptone (Difco) was used for selective
isolation of F. columnare. For isolation of E. ictaluri,
Shotts selective medium (Shotts & Waltman 1990) was
used, and for isolation of A. hydrophila, blood agar
plates were used. Results were recorded after 24 to
48 h of incubation at 28°C. Isolates were distinguished
as F. columnare, E. ictaluri and A. hydrophila on the
basis of their morphological appearance on selective
media and Gram-stain characteristics. Select isolates
were characterized biochemically to species level
using API 20E test strips (bioMerieux) and additional
biochemical tests (when necessary) according to established
criteria (Decostere et al. 1998, Altwegg 1999,
Farmer 2003). Occasionally, contaminating organisms
were also isolated on culture media, but these organisms
were readily distinguished from F. columnare,
E. ictaluri and A. hydrophila on the basis of morphological
and biochemical characteristics.
Infection of fish and sample collection. Fingerling
(~12 to 14 g) channel catfish Ictalurus punctatus
Rafinesque, National Warmwater Aquaculture Center
Strain 103, obtained from disease-free stock reared at
the Aquatic Animal Health Research Unit, were used
for experimental infection. The fish were maintained in
58 l glass aquaria with flow-through dechlorinated tap
water, constant aeration, a water temperature of 26 ±
2°C and a 12:12 h light:dark photoperiod. A commercial
diet (Aquamax Grower 400 Brentwood) was fed
daily to satiation. Initially, fish were randomized into 3
groups, with 7 fish per group. Fish in Group 1 were
injected intraperitoneally with 0.1 ml of a log-phase
broth culture of Flavobacterium columnare ARS-1,
containing ~1 × 105 CFU ml–1, Group 2 fish were similarly
injected with Edwardsiella ictaluri (American
Type Culture Collection: ATCC-33202), and Group 3
fish were injected with ~102 CFU ml–1 of Aeromonas
hydrophila K106K. Subsequently, 3 other groups of 7
fish per group (Groups 4 to 6) were injected with a
combination of the organisms to simulate mixed infections
as follows: Group 4, F. columnare + E. ictaluri
200
Panangala et al.: Multiplex-PCR for detection of 3 fish pathogens
(~1.3 × 105 CFU ml–1 of each organism); Group 5, E.
ictaluri + A. hydrophila (~ 1.4 × 105 CFU ml–1 of each
organism); Group 6, F. columnare + E. ictaluri + A.
hydrophila (~1 × 105 CFU ml–1 of each organism). An
uninfected group (Group 7) of 7 fish were maintained
as controls. Fish were monitored following infection
and dead or moribund fish were promptly removed for
sample collection. Blood was collected into 1 ml hypodermic
syringes by direct cardiac puncture, and the
kidney and gill tissues were surgically removed.
Duplicate sample sets were collected to provide for
nucleic acid extraction and for bacteriological culture.
In addition to samples from infected fish, tissues (gills,
kidney and blood) obtained from euthanized (using
200 mg l–1 tricaine methanesulfonate; Western Chemical)
naïve fish were spiked with F. columnare, E.
201
Species Isolate Source Origin
Edwardsiella ictaluri ATCC-33202 Channel catfish Ictalurus punctatus Rafinesque ATCC
E. ictaluri AL-93-75 Channel catfish Alabama
E. ictaluri ALG-03-58 Channel catfish Alabama
E. ictaluri ALG-03-161 Channel catfish Alabama
E. ictaluri 013-S99-1908 Channel catfish Mississippi
E. ictaluri 016-S99-1911 Channel catfish Mississippi
E. ictaluri 017-S99-1914 Channel catfish Mississippi
E. ictaluri 003-S99-1760 Channel catfish Mississippi
E. ictaluri IA-30-NJ#1 Tadpole madtom Noturus gyrinus Mitchell New Jersey
E. ictaluri EILO Walking catfish Clarius batrachus L. Thailand
Flavobacterium columnare ATCC-23463 Chinook salmon Oncorhynchus tshawytscha Walbaum ATCC
F. columnare ARS-1 Channel catfish Alabama
F. columnare ALG-00-530 Channel catfish Alabama
F. columnare ALG-00-522 Channel catfish Alabama
F. columnare ALG-02-036 Largemouth bass Micropterus salmoides Lacepede Alabama
F. columnare BioMed Channel catfish Alabama
F. columnare HS Channel catfish Alabama
F. columnare LSU Channel catfish Louisiana
F. columnare MS-02-463 Channel catfish Mississippi
F. columnare IR Common carp Cyprinus carpio L. Israel
Aeromonas hydrophila CECT-839 Milk CECT
A. hydrophila CIB Channel catfish Alabama
A. hydrophila K106K Channel catfish Alabama
A. hydrophila