Environmental Microbiology (2020) 00(00), 00–00 doi:10.1111/1462-2920.

14948

A comparison between farmed oysters using floating

cages and oysters grown on-bottom reveals more
potentially human pathogenic Vibrio in the on-bottom
oysters

Rachel Canty, Denene Blackwood, Rachel Noble* and Introduction

Brett Froelich
There is a $14B+ United States (US) seafood trade defi-
Institute of Marine Sciences, The University of North
cit (NOAA Fisheries, 2020). To close this gap, a nation-
Carolina at Chapel Hill, Morehead City, NC, 28557, USA.
wide expansion of shellfish aquaculture has occurred,
with the goals of expanding commercial oyster markets
Summary and streamlining the permitting process for growers in
most coastal states. For example, in 2018, North Carolina
Eating raw oysters can come with serious health
prepared a Shellfish Mariculture Plan that provides a
risks, as oysters can potentially contain bacteria of
roadmap for increasing shellfish aquaculture production
the Vibrio genus that cause food-borne infections.
Vibrio bacteria are concentrated by oysters and, in NC by an order of magnitude by 2030. This type of
when consumed, infections can result with severe expansion is mirrored by many coastal states, with
symptoms such as diarrhoea, lesions on the extremi- increases in available shellfish leasing areas, marketing
ties, or even death. Vibrio spp. concentrations are of boutique shellfish products, and improvements in food
strongly affected by season, location, and other fac- safety regulations for shellfish production. During this
tors such as temperature and salinity. Previous period of rapid mariculture expansion, the health of the
research in North Carolina oysters has been con- shellfish consumer and the reputation of growers must be
ducted on wild and farmed oysters but not at the protected by minimizing the health risks that come from
same time. Farmed, or aquaculture raised, oysters consuming raw or undercooked oysters.
are considerably different from wild oysters and Eating raw oysters comes with serious risks as many
could possibly pose different health risks. Farmed oysters contain naturally occurring harmful bacteria,
oysters are handled, raised from seed, and often including strains of Vibrio vulnificus and Vibrio para-
grown using suspended grow-out systems called haemolyticus that can cause severe illness, or even death,
‘floating cages’. Therefore, farmed oysters can be when consumed. Commercially, oysters can be harvested
grown at the surface of the estuary, while wild oys- in two ways; wild-caught or grown as part of aquaculture
ters typically grow at the bottom of the water column. or farming activities. Many states have regulations that
This project compared the concentrations of Vibrio limit the size, quantity, area, season, and times of collec-
spp. in suspended, farm-grown oysters and wild oys- tion for wild oysters in order to promote native shellfish
ters at three sites, using a paired approach with reefs, but these restrictions are not in place for farmed
farmed and wild oysters sampled in proximity. An oysters. Most of the expansion in shellfish commercial har-
important part of this comparison was identifying vest will be achieved by shifts in oyster farming practices,
pathogenicity of the bacteria isolated from the sam- such as the use of suspended farming system, usually
ples. Distinction was made between off- and on- referred to as ‘floating cages’ for oyster grow-out.
bottom farming. Interestingly, on-bottom oysters had Both V. vulnificus and V. parahaemolyticus occur nat-
more pathogenic V. vulnificus than off-bottom urally in estuarine waters worldwide. The vector to
oysters. humans is primarily raw/undercooked oysters, which
account for 93% of ingestion cases (Oliver, 2013). Vib-
rio vulnificus is the single most fatal foodborne patho-
gen in the United States and possibly in the world
Received 13 January, 2020; accepted 18 February, 2020. *For corre-
spondence. E-mail [email protected]; Tel. (+00) 252 726 6841; (Baker-Austin and Oliver, 2018). It accounts for 95% of
Fax (+00) 252 726 2426. all U.S. seafood-related deaths and has a fatality rate of
© 2020 Society for Applied Microbiology and John Wiley & Sons Ltd.
2 R. Canty et al.
ca. 50%, greatly exceeding that of other foodborne developed over the course of the study to rapidly type iso-
pathogens such as Salmonella (0.6%), Escherichia coli lates selected from culture-based analyses.
(3%–5%), and Clostridium botulinum (<8%), although
the number of cases are lower (Mead et al., 1999;
Experimental procedures
Jones and Oliver, 2009). Symptoms of a V. vulnificus
infection include fever, chills, nausea, hypotension, and Oyster collection
secondary lesions, which develop on the extremities.
Both wild and farmed oysters (Crassostrea virginica)
The incubation period is very short, and death can
were collected by hand from Cedar Island Bay, Jarrett
occur in 24–72 h after eating a single oyster (Jones and Bay, and the Newport River Estuary (Fig. 1). Oyster sam-
Oliver, 2009). Vibrio parahaemolyticus infections, unlike pling occurred between late July 2018 and September
those caused by V. vulnificus, are usually not fatal but 2018. Each site contained a wild location and a farm
instead come with symptoms that include diarrhoea with location, and they were within no more than 1000 m dis-
abdominal cramps, nausea, vomiting, headache, chills, tance and within 3‰ salinity difference, except during a
and low-grade fever (Yeung and Boor, 2004). While single extreme rainfall condition (Fig. 1 and Supporting
foodborne Vibrio infections are sometimes self-limiting Information Table S1). Cedar Island Bay was sampled on
and typically last 3 days, fatal cases of septicemia may 22 July and 13 August, Jarret Bay was sampled on
occur in immunocompromised patients or those with 3 August and 24 August, and the Newport River Estuary
pre-existing medical conditions. In the United States, an was sampled on 7 August and 4 September. The farm
estimated 84 000 people contract foodborne Vibrio location and its corresponding wild location from each
infections each year (Scallan et al., 2011). Unfortu- site were sampled on the same day at or within 3 h of
nately, in contrast to other leading foodborne bacteria, low tide, with oysters harvested typically within an hour of
numbers of Vibrio cases are increasing (Centers for each other. Cages were removed from water to retrieve
Disease Control and Prevention, 2007). samples. Each site was visited twice within the 2-month
There is a critical need for information about pathogenic period, with 16 samples occurring at each visit (eight from
Vibrio in oysters, particularly studies on various farming farm and eight from wild). All oysters were transported to
practices. Farmed oysters have generally not been the laboratory on ice and processed within 5 h.
wellstudied in the context of growing practices and Vibrio Two sites were sampled comparing off-bottom farmed
abundance. Furthermore, there are no studies in NC com- oysters and nearby wild oysters, while the third site was
paring Vibrio concentrations in farmed and wild oysters. on-bottom farmed oysters and wild oysters, which served
When compared to wild oysters, farmed oysters are often as a control (Fig. 2). On each sampling date, 48 oysters
grown in floating cages, which mean that these oysters were collected from the wild site and 48 oysters were col-
experience vastly different conditions than wild oysters, lected from the farmed site. Each site was sampled on
which grow along the benthos of the estuary. Some of two separate occasions.
these differences in growth conditions include air exposure,
UV irradiation, temperature, agitation, water column height,
Oyster processing
and/or handling. Because these oysters experience such
disparate conditions, it was hypothesized that there could Shellfish were rinsed with cold water, shucked, aseptically,
be a difference in the concentration or type of human Vibrio and the hemolymph drained. Any internal sediment was
pathogens found in suspended farmed versus on-bottom rinsed with phosphate buffered saline (PBS). Meats of six
wild oysters, even when these oysters are harvested from oysters were pooled, weighed, and diluted with PBS at a
matching estuarine locations under with very similar envi- 1:1 (weight/volume) ratio. Eight samples of six oysters each
ronmental conditions. The objectives of this project were to were collected from each site (farmed and wild) on each
examine and compare suspended, farm-grown off-bottom sampling day. Shellfish meats were blended for 10 min in a
oysters and wild on-bottom oysters at three sites, with paddle blender (Fisher Scientific, Waltham, MA) at
farmed and wild oysters collected at paired sites in close 280 rpm. Homogenates were diluted 1:10 in PBS, and
proximity to one another. Over the course of the study, total 100 μl aliquots of both diluted and undiluted homogenates
Vibrio, V. vulnificus and V. parahaemolyticus concentra- were plated on media as described in the next section.
tions were determined in composite oyster samples using
both culture-based methods and molecular confirmation of
Colony growth and isolation
pathogenicity. Attention was paid to ensure that samples
were collected from paired wild and farmed oyster sites For all samples, 100 μl of both undiluted and diluted
under similar environmental conditions, with sampling con- homogenate were spread plated on thiosulfate-citrate-bile
ducted within 2 h. Molecular confirmation methods were salts-sucrose (TCBS) and CHROMagar Vibrio (CAV)
© 2020 Society for Applied Microbiology and John Wiley & Sons Ltd., Environmental Microbiology
 More pathogenic Vibrio in on-bottom oysters than off-bottom 3

Fig. 1. A. Study area in eastern NC. Sampling sites included Cedar Island Bay (CIB), Jarrett Bay (JB), and Newport River (NR).
B and C. Farmed/surface (FS) and wild/bottom sites (WB).
D. Farmed/bottom site (FB) and wild/bottom site (WB).

Fig. 2. Sample design of short-term

experiment. The study had three
sites, with the Newport River site
acting as a control (with on-bottom
farmed oysters).

(CHROMagar, Paris, France). TCBS plates, prepared as CAV plates, prepared as instructed, were used to isolate
instructed (Himedia), were used to enumerate total Vibrio. presumptive colonies of V. parahaemolyticus (purple) and
Green and yellow colonies were counted, as described by V. vulnificus (blue). All plates were incubated at 37
C for
Froelich et al. (2015), and values were summed to deter- 24 h. After incubation, colonies on plates were counted
mine total Vibrio abundance per unit shellfish tissue mass. and the data were converted to CFU per gram of oyster.
© 2020 Society for Applied Microbiology and John Wiley & Sons Ltd., Environmental Microbiology
4 R. Canty et al.
From each plate 10 colonies each (or as many as possible date (ANOVA, P > 0.05, Table 2,B). There was no differ-
if less than 10 were present) of presumed V. vulnificus ence in concentrations of total Vibrio between suspended
and V. parahaemolyticus were isolated using sterile and on-bottom oysters (t-test, P = 1.00), nor at the control
approaches and placed into 100 μl of heart-infusion site with both farmed and wild oysters being grown on
(HI) broth and incubated at room temperature overnight. bottom (t-test, P = 0.52, Fig. 3). There was also no differ-
Following this, cells were lysed to release DNA by incuba- ence in total V. parahaemolyticus concentrations
tion at 100
C for 10 min. Centrifugation at 10 000× g for between farmed or wild oysters at both the experimental
10 min separated the aqueous DNA from cellular material. (off vs on bottom) and control (both on bottom) sites
Supernatants, to be used as PCR templates, were diluted (two-way ANOVA, P > 0.05, Fig. 4). Analysis of patho-
in nuclease-free water and the undiluted, 10-1, and 10-2 genic V. parahaemolyticus was not performed due to too
isolate subaliquots were stored at −80
C until they were few samples containing these bacteria.
prepared for PCR amplification as described below. A significant difference was observed in total V. vulnificus
concentrations, shown in Fig. 5, with off-bottom farmed oys-
ters having fewer total V. vulnificus than wild oysters (t-test,
Molecular confirmation of isolates and determination of
P = 0.0334). This difference was not mirrored in the control
potential pathogenicity
site with both on-bottom farmed and on-bottom wild oysters
Molecular species identification of both V. vulnificus (vvhA) (t-test, P = 0.8379). Vibrio vulnificus was found in 87.5% of
and V. parahaemolyticus (toxR), and subsequent potential samples in this study, with 91.7% of bottom grown and
for pathogenicity (vgcC for V. vulnificus, and tdh/trh for 81.3% of off-bottom farmed oyster samples containing the
V. parahaemolyticus) was performed via PCR amplification bacteria. Ten samples were devoid of confirmed
on the BioRad CFX96™ Real-Time System (BioRad) using V. vulnificus, four from on-bottom oysters and six from off-
the PowerUp™ SYBR® Green Master Mix (ThermoFisher bottom oysters. Half of the off-bottom oyster samples that
Scientific, Pittsburg, PA). Following SYBR® Green-PCR, a were devoid of V. vulnificus came from sampling at the
melt curve was generated in order to confirm amplification of Jarret Bay farmed site on 24 August 2018, meaning that
only the target amplicon and only those peaks that matched three of the eight off-bottom oyster samples from that date
the positive control (see below) were considered positive for did not have any confirmed V. vulnificus. The
the corresponding gene. Primers are listed in Table 1. corresponding on-bottom site at Jarret Bay had confirmed
Quantification of V. vulnificus and V. parahaemolyticus V. vulnificus in 10 out of 10 oyster samples for that date.
was performed as described by Froelich et al. (2015) Oyster samples taken from waters with salinities lower than
where concentrations in CFU/g obtained from culture 20 ppt all had confirmed V. vulnificus.
data were multiplied by the percentage of vvhA-positive Of the 266 confirmed V. vulnificus (vvhA-positive) iso-
and toxR-positive (respectively) isolates. The same pro- lates throughout the entire study, 44 contained the virulence
cess was used in quantifying abundance and percent correlated gene, vcgC, constituting 16.5% of the sample
potentially pathogenic V. vulnificus (vcgC-positive) and population. When analysed according to growing approach,
V. parahaemolyticus (tdh/trh-positive). i.e. by off-bottom and on-bottom oysters, however, 20.1%
of on-bottom oysters were vcgC-positive and only 10.3% of
off-bottom oysters were potentially pathogenic.
Results
Like confirmed V. vulnificus, potentially pathogenic
Total Vibrio concentrations between farmed and wild oys- V. vulnificus and percent potentially pathogenic
ters did not vary statistically from site to site (ANOVA, V. vulnificus were also lower in off-bottom farmed oysters
P > 0.05, Table 2,A) nor from sampling date to sampling than on-bottom wild oysters at the experimental sites

Table 1. Primer sequences.

Gene target Direction Sequence (50 -30 ) Amplicon size (bp) Source

vvhA F TTCCAACTTCAAACCGAACTATGAC 205 Panicker and Bej (Panicker and Bej, 2005)
R ATTCCAGTCGATGCGAATACGTTG
vcgC F AAAACTCATTGARCAGTAACGAAA 146 Warner and Oliver, (Warner and Oliver, 2008)
R AGCTGGATCTAAKCCCAATGC
toxR F GTCTTCTGACGCAATCGTTG 368 Kim et al. (Kim et al., 1999)
R ATACGAGTGGTTGCTGTCATG
tdh F GTAAAGGTCTCTGACTTTTGGAC 269 Bej et al. (Bej et al., 1999)
R TGGAATAGAACCTTCATCTTCACC
trh F TTGGCTTCGATATTTTCAGTATCT 500 Bej et al. (Bej et al., 1999)
R CATAACAAACATATGCCCATTTCCG

© 2020 Society for Applied Microbiology and John Wiley & Sons Ltd., Environmental Microbiology
 More pathogenic Vibrio in on-bottom oysters than off-bottom 5
Table 2. Log total Vibrio concentrations in farmed and wild oysters.
Total Vibrio concentrations were separated by site (A) and by date of
harvest (B). Total Vibrio concentrations were obtained from culture-
based analyses data.

Site Average total Vibrio

(A) W/F S/B log(CFU/g)

CIB-FS F S 4.3 ± 1.1

CIB-WB W B 4.2 ± 0.9
JB-FS F S 4.1 ± 0.9
JB-WB W B 4.2 ± 1.1
NR-FB F B 3.5 ± 1.3
NR-WB W B 3.1 ± 1.1

Date of harvest Average total Vibrio

(B) Site W/F S/B log(CFU/g)

22 July 2018 CIB-FS F S 3.9 ± 1.2 Fig. 4. Comparison of total V. parahaemolyticus in farmed and wild
CIB-WB W B 4.2 ± 1.3 oysters. Error bars are standard error of the mean. There were no
3 August 2018 JB-FS F S 4.1 ± 1.3 significant differences (P > 0.05).
JB-WB W B 4.3 ± 1.6
7 August 2018 NR-FB F B 3.5 ± 1.3
NR-WB W B 3.1 ± 1.1
13 August 2018 CIB-FS F S 4.6 ± 1.6
CIB-WB W B 4.2 ± 1.3
24 August 2018 JB-FS F S 4.2 ± 1.4
JB-WB W B 4.0 ± 1.3
4 September 2018 NR-FB F B 3.7 ± 1.3
NR-WB W B 3.7 ± 1.3

CIB, Cedar Island Bay; JB, Jarret Bay; NR, Newport River; FS,
Farmed-Surface; FB, Farmed-Bottom; WB, Wild-Bottom; ‘F’ indicates
farmed oysters and ‘W’ indicates wild oysters.

[t-test, P = 0.0366 (Fig. 6) and t-test, P = 0.0342 (not

shown) respectively]. Again, this was not demonstrated at
the control site [t-test, P = 0.7832 (Fig. 6) and t-test,
Fig. 5. Comparison of total V. vulnificus in farmed and wild oysters.
P = 0.8924 (not shown) respectively]. Potentially patho- Error bars are standard error of the mean. Asterisk indicates signifi-
genic V. vulnificus was found in 35.0% of the oyster sam- cant difference in mean (P = 0.0334).
ples in this study: 41.7% of on-bottom oysters contained
vcgC-positive V. vulnificus and 25% of off-bottom oysters same sample site and day (Newport River-Farmed bottom
contained vcgC-positive V. vulnificus. Two samples con- and Newport River-Wild bottom on 8 July 2018). The salin-
tained 100% vcgC-positive V. vulnificus, both from the ity was 23 ppt and daily air temperature was 28
C.

Fig. 3. Comparison of total Vibrio in farmed and wild oysters. Error Fig. 6. Comparison of pathogenic V. vulnificus in farmed and wild
bars are standard error of the mean. There was no significant differ- oysters. Error bars are standard error of the mean. Asterisk indicates
ent between farmed and wild oysters by group (P > 0.05). significant difference in mean (P = 0.0366).

© 2020 Society for Applied Microbiology and John Wiley & Sons Ltd., Environmental Microbiology
6 R. Canty et al.
Daily air temperatures during this time period averaged in eastern NC waters previously (Froelich et al., 2019).
at 27
C, with a range of 24
C–29
C. Throughout this The distinction between farmed and off-bottom oysters is
study period, temperature and salinity exhibited very important, as the control site with farmed oysters grown
weak correlations with total Vibrio concentrations. on-bottom showed no differences with wild oysters
July 2018 brought heavy rainfall. In Carteret County, (Figs. 3–6). Thus, it is less likely that the handling and
NC, rainfall total for the month of July was ca 12 in., mak- other aquaculture procedures that occur with farming are
ing it the wettest July on record; 24 July 2018, alone had influencing the concentration of pathogenic Vibrio but
3.51 in. of rain (Supporting Information Table S2). Rain- rather the use of floating cages that is the important
fall had differing effects on off- and on-bottom oysters. factor.
Specifically, 24-h rainfall correlated positively with on- Vibrio parahaemolyticus was confirmed in nearly all
bottom oyster total Vibrio populations (r2 = 0.329). About oyster samples (97.9% on-bottom and 100% of off-
3- and 7-day antecedent rainfall, however, has a negative bottom oysters). Variances between off and on-bottom
impact on total Vibrio in oysters (r2 = −0.618 and −0.439 V. parahaemolyticus concentrations were not observed in
respectively). Daily wind speeds negatively correlated this study, which is in contrast to a study by Cole et al.
with total Vibrio, V. parahaemolyticus, and V. vulnificus in (2015) that, like this study, focused on the effects of off-
surface oysters (r2 = −0.617, −0.649 and −0.625 respec- bottom farming on Vibrio populations in a shallow, estua-
tively). There was no significant correlation with wind and rine location. Cole et al. (2015) found higher total
Vibrio populations in on-bottom oysters. V. parahaemolyticus concentrations in off-bottom oysters.
Unlike this study, the Cole et al. (2015) study was con-
ducted in the Gulf Coast, over a longer time-scale
Discussion
(1 year) and deployed their own oysters for on-bottom
Oysters from farmed and wild sites were collected and oysters (in cages), indicating that they did not use wild
analysed for human pathogenic Vibrio species, including oysters for their on-bottom studies. Our study was in
V. parahaemolyticus and V. vulnificus. This experiment agreement with Cole et al. (2015) on V. vulnificus dynam-
controlled for confounding factors by using three sites ics, which indicated a higher trend in on-bottom
that were chosen because wild oysters were found in populations of these bacteria. Cole et al. (2015)
close proximity to farmed oysters. Additionally, the oys- suggested that off- and on-bottom Vibrio discrepancies in
ters were harvested together, within a short time frame. oysters could be due to the higher concentrations of Vib-
The proximity and simultaneous collection ensured that rio in sediments (Johnson et al., 2012). Thus, oysters that
most environmental effects were controlled for. Addition- are on bottom, and closer to sediments, are exposed to
ally, a robust sampling scheme was performed, with each higher concentrations of bacteria. This theory is
site being sampled on two separate dates. Each sam- supported by research previously conducted by Fries
pling day consisted of sampling both the off and on- et al. (2008) in the Neuse River Estuary, in which Vibrio
bottom locations with eight individual samples of six oys- spp. that were attached to sediment were a prominent
ters each. This resulted in 96 oysters being collected at proportion of the total Vibrio population. In that study, it
each sampling date, 48 from both the farmed and wild was demonstrated that resuspension events could drive
paired locations. The third site, which served as a control, higher concentrations of Vibrio into the upper water col-
contained farmed oysters that were grown on bottom, umn. In the current study, it was found that total Vibrio
while at the other two sites oysters were grown in floating was negatively correlated with daily wind speed, an indi-
cages, off-bottom. This design allowed for control of cator of potential sediment resuspension events. How-
aquaculture methodology. ever, the factors explaining these patterns may be more
There was no observable difference in the number of complex than just daily wind speed (i.e. sustained wind
total Vibrio in farmed or wild oysters, regardless of aqua- speed, gust speed, sustained wind direction, water col-
culture practice (Table 2 and Fig. 3). This appears to indi- umn depth) and the timing of the data pairing may be
cate that Vibrio will, either by uptake or by replication, inappropriate.
reach some maximum carrying capacity in an oyster. Yet,
wild oysters contained significantly more V. vulnificus,
including pathogenic forms (Figs. 5 and 6). This suggests
that oysters growing on the surface versus on-bottom Acknowledgements
can contain differing concentrations of specific Vibrio The authors would like to thank Tom Kiffney, Matt Price, Vic-
species, even when the total number of Vibrio is nearly toria Pruente, and Kelsey Jesser for assistance in sample
identical. Because each species has its own separate collection and processing. The authors thank the owners of
conditions in which it will survive or thrive, this is not the aquaculture facilities for donating time, space, and sam-
unexpected, and a similar phenomenon has been seen ples. Funding for this project was provided by NC Sea Grant

© 2020 Society for Applied Microbiology and John Wiley & Sons Ltd., Environmental Microbiology
 More pathogenic Vibrio in on-bottom oysters than off-bottom 7
(15-5473) and by the Interstate Shellfish Sanitation Kim, Y.B., Okuda, J., Matsumoto, C., Takahashi, N.,
Conference. Hashimoto, S., and Nishibuchi, M. (1999) Identification of
Vibrio parahaemolyticus strains at the species level by
PCR targeted to the toxR gene. J Clin Microbiol 37:
References 1173–1177.
NOAA Fisheries (2020) U.S. Aquaculture
Baker-Austin, C., and Oliver, J.D. (2018) Vibrio vulnificus:
Oliver, J.D. (2013) Vibrio vulnificus: death on the half Shell.
new insights into a deadly opportunistic pathogen. Environ
A personal journey with the pathogen and its ecology.
Microbiol 20: 423–430.
Microb Ecol 65: 793–799.
Bej, A.K., Patterson, D.P., Brasher, C.W., Vickery, M.C.L.,
Mead, P.S., Slusker, L., Dietz, V., McCaig, L.F., Bresee, J.
Jones, D.D., and Kaysner, C.A. (1999) Detection of total
S., Shapiro, C., et al. (1999) Food-related illness and
and hemolysin-producing Vibrio parahaemolyticus in
death in the United States. Emerg Infect Dis 5: 607–625.
shellfish using multiplex PCR amplification of tl, tdh and
Panicker, G., and Bej, A.K. (2005) Real-time PCR detection
trh. J Microbiol Methods 36: 215–225.
of Vibrio vulnificus in oysters: comparison of oligonucleo-
Centers for Disease Control and Prevention (2007) Sum-
tide primers and probes targeting vvhA. Appl Env
mary of human Vibrio cases reported to CDC. Cholera
Microbiol 71: 5702–5709.
IVibrio IIllness Surveill Syst.
Scallan, E., Hoekstra, R.M., Angulo, F.J., Tauxe, R.V.,
Cole, K.M., Supan, J., Ramirez, A., and Johnson, C.N.
Widdowson, M.-A., Roy, S.L., et al. (2011) Foodborne ill-
(2015) Suspension of oysters reduces the populations of
ness acquired in the United States—major pathogens.
Vibrio parahaemolyticus and Vibrio vulnificus. Lett Appl
Emerg Infect Dis 17: 7–15.
Microbiol 61: 209–213.
Warner, E.B., and Oliver, J.D. (2008) Multiplex PCR assay
Fries, J.S., Characklis, G.W., and Noble, R.T. (2008) Sedi-
for detection and simultaneous differentiation of genotypes
ment–water exchange of Vibrio sp. and fecal indicator
of Vibrio vulnificus biotype 1. Foodborne Pathog Dis 5:
bacteria: implications for persistence and transport in the
691–693.
Neuse River estuary, North Carolina, USA. Water Res 42:
Yeung, P.S.M., and Boor, K.J. (2004) Epidemiology, patho-
941–950.
genesis, and prevention of foodborne Vibrio para-
Froelich, B., Gonzalez, R., Blackwood, D., Lauer, K., and
haemolyticus infections. Foodborne Pathog Dis 1: 74–88.
Noble, R. (2019) Decadal monitoring reveals an increase
in vibrio spp. concentrations in the Neuse River estuary,
North Carolina, USA. PLOS ONE 14: e0215254.
Supporting Information
Froelich, B.A., Ayrapetyan, M., Fowler, P., Oliver, J.D., and
Noble, R.T. (2015) Development of a matrix tool for the Additional Supporting Information may be found in the online
prediction of vibrio species in oysters harvested from version of this article at the publisher’s web-site:
North Carolina. Appl Environ Microbiol 81: 1111–1119.
Johnson, C.N., Bowers, J.C., Griffitt, K.J., Molina, V., Table S1 Harvest site locations, salinities, and oyster types.
Clostio, R.W., Pei, S., et al. (2012) Ecology of Vibrio para- CIB has two wild locations because oyster clusters were
haemolyticus and Vibrio vulnificus in the coastal and estu- scarce in the original wild area we chose to sample. Salin-
arine waters of Louisiana, Maryland, Mississippi, and ities were within 3 ppt of each other except during a single
Washington (United States). Appl Environ Microbiol 78: extreme rainfall condition (JB, August 3, 2018). ‘W’ = wild,
7249–7257. ‘F’ = farmed, ‘S’ = suspended, ‘B’ = on-bottom.
Jones, M.K., and Oliver, J.D. (2009) Vibrio vulnificus: dis- Table S2 Daily wind speed, wind directions, termperature
ease and pathogenesis. Infect Immun 77: 1723–1733. and precipitation

© 2020 Society for Applied Microbiology and John Wiley & Sons Ltd., Environmental Microbiology