Assessment of the northern distribution range of selected Perkinsus species in eastern oysters (Crassostrea virginica)and hard clams (Mercenaria mercenaria) using PCR-based detection assays.

Abstract

Perkinsus species are protistan parasites of molluscs. In Chesapeake Bay, P. marinus, P. chesapeaki, and P. andrewsi are sympatric, infecting oysters and clams. Although P. marinus is a pathogen for Crassostrea virginica, it remains unknown whether P. andrewsi and P. chesapeaki are equally pathogenic. Perkinsus species have been reported in C. virginica as far north as Maine, sometimes associated with high prevalence, but low mortality. Thus, we hypothesized that, in addition to P. marinus, Perkinsus species with little or no pathogenicity for C. virginica, may be present. Accordingly, we investigated the distribution of Perkinsus species in C. virginica and Mercenaria mercenaria, collected from Maine to Virginia, by applying PCR-based assays specific for the genus Perkinsus, the species P. marinus, P. andrewsi, and a Perkinsus sp., isolated from M. mercenaria. DNA samples of M. mercenaria possessed potent PCR inhibitory activity, which was overcome by the addition of 1 mg/ml BSA and 5% (v/v) DMSO to the PCR reaction mixture. All 3 Perkinsus species were found in both host species throughout the study area. Interestingly, the prevalence of P. marinus in M. mercenaria was significantly lower than in C. virginica, suggesting that M. mercenaria is not an optimal host for P. marinus.

Full text

410
J. Parasitol., 94(2), 2008, pp. 410–422
䉷
American Society of Parasitologists 2008
ASSESSMENT OF THE NORTHERN DISTRIBUTION RANGE OF SELECTED PERKINSUS
SPECIES IN EASTERN OYSTERS (CRASSOSTREA VIRGINICA) AND HARD CLAMS
(MERCENARIA MERCENARIA) WITH THE USE OF PCR-BASED DETECTION ASSAYS
Wolf T. Pecher, Mohammad R. Alavi, Eric J. Schott, Jose´ A. Fernandez-Robledo, Laura Roth*, Sean T. Berg*, and
Gerardo R. Vasta
Center of Marine Biotechnology, University of Maryland Biotechnology Institute, Baltimore, Maryland 21202. e-mail: vasta@umbi.umd.edu
ABSTRACT
: Perkinsus species are protistan parasites of molluscs. In Chesapeake Bay, Perkinsus marinus, Perkinsus chesapeaki,
and Perkinsus andrewsi are sympatric, infecting oysters and clams. Although P. marinus is a pathogen for Crassostrea virginica,
it remains unknown whether P. andrewsi and P. chesapeaki are equally pathogenic. Perkinsus species have been reported in C.
virginica as far north as Maine, sometimes associated with high prevalence, but low mortality. Thus, we hypothesized that, in
addition to P. marinus, Perkinsus species with little or no pathogenicity for C. virginica may be present. Accordingly, we
investigated the distribution of Perkinsus species in C. virginica and Mercenaria mercenaria, collected from Maine to Virginia,
by applying PCR-based assays specific for P. marinus, P. andrewsi, and a Perkinsus sp. isolated from M. mercenaria. DNA
samples of M. mercenaria possessed potent PCR inhibitory activity, which was overcome by the addition of 1 mg/ml BSA and
5% (v/v) DMSO to the PCR reaction mixture. All 3 Perkinsus species were found in both host species throughout the study
area. Interestingly, the prevalence of P. marinus in M. mercenaria was significantly lower than in C. virginica, suggesting that
M. mercenaria is not an optimal host for P. marinus.
Perkinsus species (Perkinsozoa, Alveolata) are the causative
agent of perkinsosis in a variety of mollusc species. For some
host species, such as the eastern oyster C. virginica, Perkinsus
species infections cause widespread mortality in both natural
and farmed oyster populations, resulting in severe economic
loss for the shellfishery, and detrimental effects on the environ-
ment (Andrews, 1988; Ford, 1996; Villalba et al., 2004). Cur-
rently, 3 Perkinsus species are recognized along the Atlantic
coast of the United States, i.e., P. marinus, isolated from the
eastern oyster C. virginica (Mackin et al., 1950), P. chesapeaki
from the soft shell clam Mya arenaria (McLaughlin et al.,
2000), and P. andrewsi from the Baltic clam Macoma balthica
(Coss, Robledo, and Vasta, 2001). In addition, various Perkin-
sus isolates have been reported, including an isolate from the
hard clam M. mercenaria [hereafter referred to as Perkinsus sp.
(M. mercenaria)] that appears to be closely related to P. an-
drewsi (Andrews, 1955; Perkins, 1988; Coss, 2000). It is not
yet clear whether P. chesapeaki and P. andrewsi are different
species and, although their synonymization has been proposed
(Burreson et al., 2005), because of the limited evidence avail-
able at present time, we consider P. andrewsi as a distinct Per-
kinsus species in the present study.
The standard diagnostic method for Perkinsus spp. infections
has been the fluid thioglycollate medium (FTM) assay (Ray,
1952, 1966), which is considered to be more sensitive than
histological diagnosis (McLaughlin and Faisal, 1999). However,
neither method is able to discriminate among Perkinsus species
(reviewed in Villalba et al., 2004); diagnostic assays based on
anti-Perkinsus sp. antibodies (Choi et al., 1991; Dungan and
Robertson, 1993; Ottinger et al., 2001; Montes et al., 2002)
have not been rigorously validated, and may exhibit cross-re-
activity with dinoflagellates (Dungan et al., 1993; Bushek et al.,
2002; Villalba et al., 2004).
The development of culture methods for Perkinsus species
Received 2 April 2007; revised 8 August 2007; accepted 6 August
2007.
* L. Roth and S. T. Berg participated in the project as student interns
through the Technology Magnet Program, Howard County Public
School System, Ellicott City, Maryland, and the Ingenuity Research
Practicum, Baltimore Polytechnic Institute, Baltimore, Maryland.
(Gauthier and Vasta, 1993; Kleinschuster and Swink, 1993; La
Peyre et al., 1993) greatly facilitated the development of spe-
cific PCR-based diagnostic assays. The first PCR-based assay
was developed for P. marinus and was species-specific and
more sensitive than the FTM assay (Marsh et al., 1995; Robledo
et al., 1998). Subsequently, PCR-based assays specific for Per-
kinsus olseni (de la Herra´n et al., 2000; Robledo et al., 2000),
P. andrewsi (Coss, Robledo, Ruiz, and Vasta, 2001), and for
other species of Perkinsus (Robledo et al., 2002) were also
developed. Quantitative PCR assays for P. marinus (Yarnall et
al., 2000; Gauthier et al., 2006), a multiplex PCR assay de-
tecting P. marinus and Haplosporidium species (Penna et al.,
2001), and modified PCR-based assays have been developed
that can distinguish between P. marinus, P. olseni, Perkinsus
mediterraneus, and P. andrewsi/P. chesapeaki or P. marinus
and P. olseni, respectively (Elandalloussi et al., 2004; Abollo
et al., 2006).
Prior to 2000, all surveys for Perkinsus species were con-
ducted with the use of histology or FTM-based assays, and all
Perkinsus infections observed in C. virginica were attributed to
P. marinus, the only Perkinsus species described along the At-
lantic coast of the Americas at that time. By 2001, 2 new spe-
cies, P. chesapeaki and P. andrewsi, were described from clams
(M. arenaria and M. balthica, respectively) in Chesapeake Bay
(Coss, 2000; McLaughlin et al., 2000; Coss, Robledo, Ruiz, and
Vasta, 2001). However, in addition to its type host, P. andrewsi
was also found in C. virginica (the type host of P. marinus)
and in the clams M. mercenaria and M. mitchelli (Coss, 2000;
Coss, Robledo, Ruiz, and Vasta, 2001). Conversely, P. marinus
was detected in M. arenaria, M. balthica, and Macoma mitch-
elli (Kotob et al., 1999; Coss, 2000; McLaughlin et al., 2000;
Coss, Robledo, Ruiz, and Vasta, 2001), suggesting a broad host
range for these Perkinsus species.
Perkinsus species infections have been observed in oysters
from Tabasco, Mexico, to Maine (reviewed in Burreson and
Ragone Calvo, 1996; Ford, 1996; Soniat, 1996). In some areas
of the northeastern United States, mortalities in oyster popula-
tions were low to moderate, despite high prevalence and infec-
tion densities of Perkinsus species (Ford, 1996; Karolus et al.,
2000). This observation led us to hypothesize that, in addition

PECHER ET AL.—DISTRIBUTION OF SELECTED PERKINSUS SPECIES 411
to P. marinus, other Perkinsus species are present in the north-
eastern regions that may be less virulent towards C. virginica.
We therefore surveyed C. virginica and M. mercenaria obtained
from selected sites from Maine to Virginia for the presence of
Perkinsus species, and specifically for P. marinus, P. andrewsi,
and Perkinsus sp. (M. mercenaria) using specific PCR-based
assays. This is the first study that assesses the distribution of
sympatric Perkinsus species in 2 economically important mol-
luscan hosts.
MATERIALS AND METHODS
Collection of tissue specimens and DNA extraction
Crassostrea virginica (size: 44–142 mm) and M. mercenaria (size:
40–73 mm) specimens, collected monthly from June 2002 to September
2002, were obtained from shellfish farmers and academic institutions
from 8 sites along the Atlantic coast of the United States as follows:
C. virginica were obtained from Walpole (Maine), Martha’s Vineyard
(Massachusetts), Narragansett Bay (Rhode Island), Oyster Bay (New
York), Delaware Bay (New Jersey), and Sandy Point (Maryland); M.
mercenaria were obtained from Eliot (Maine), Martha’s Vineyard (Mas-
sachusetts), and Cheriton (Virginia) (Fig. 1). Upon arrival, the shellfish
were stored up to 72 hr at 4 C until further processing.
Eighteen to 60 specimens from each sampling site and collection date
were individually dissected. From each individual, gut, gill, and mantle
tissues were collected and pooled (50–100-mg wet weight of total tis-
sue/pool), and DNA was extracted with the use of a commercially avail-
able kit (DNeasy, 96-well format, QIAGEN, Valencia, California). DNA
concentration and purity were estimated by spectrophotometry at wave-
lengths of 260 and 280 nm. The DNA samples were stored at
⫺
20 C
until testing.
PCR assays
PCR-based assays specific for the genus Perkinsus, and the species
P. marinus, P. andrewsi, and P. olseni (syn. Perkinsus atlanticus), were
used according to Marsh et al. (1995), Coss, Robledo, Ruiz, and Vasta
(2001), and Robledo et al. (2000, 2002).
Development of a PCR-based diagnostic assay specific for the
Perkinsus species isolated from Mercenaria mercenaria
Primers designated M6L (sense, 5
⬘
-GCGGCGCAAATTCATCACT
TG AG-3
⬘
) and M5 (antisense, 5
⬘
-AACCATCCCGACTACCATCT
GG-3
⬘
) were designed based on the intergenic spacer of the rRNA gene
cluster of Perkinsus sp. (M. mercenaria) with the use of an Oligo Cal-
culator v3.07 (http://www.basic.northwestern.edu/biotools/oligocalc.
html). Thermocycler conditions were 94 C for 4 min, 35 cycles of 94
C for 1 min, 65 C for 30 sec with an extension of 1 sec per cycle, 72
C for 1 min, with a final extension at 72 C for 7 min.
Three different PCR reaction mixtures (A–C) were used. PCR reac-
tion mixture A consisted of 1
⫻
QIAGEN PCR Master mix (contains
Taq DNA Polymerase (250 mU/
␮
l), KCl, Tris-Cl, (NH
4
)
2
SO
4
, 1.5 mM
MgCl, and 200
␮
M of each dNTP) (QIAGEN), and 40 nM of each
primer. To obtain PCR reaction mixture B, heat-treated bovine serum
albumin (BSA) (New England Biolabs, Ipswich, Massachusetts) and
dimethyl sulfoxide (DMSO) (Sigma-Aldrich, St. Louis, Missouri) were
added to a final concentration of 1 mg/ml BSA and 5% (v/v) DMSO
to PCR reaction mixture A. PCR reaction mixture C consisted of
TaKaRa Ex Taq
娂
DNA Polymerase (250 mU/
␮
l) (TaKaRa Bio, Inc.,
Otsu, Shiga, Japan), 1
⫻
of the proprietary Ex Taq reaction buffer (con-
tains 2 mM MgCl
2
), 200
␮
M of each dNTP (TaKaRa Bio), 1 mg/ml
BSA, 5% (v/v) DMSO, and 40 nM of each primer.
Assessment of the specificity and sensitivity of the PCR assays
To assess the specificity of each PCR assay, 50 ng of DNA from P.
marinus (ATCC 50489), P. andrewsi (ATCC 50807), and Perkinsus sp.
(M. mercenaria) were used as templates in the PCR reactions. Sensitiv-
ities of the species-specific assays were assessed with the use of de-
creasing amounts of genomic DNA (100 pg to 1 fg) from the respective
Perkinsus species. For the genus-specific assay, the sensitivity was as-
sessed with the use of P. marinus, P. andrewsi, and Perkinsus sp. (M.
mercenaria) genomic DNA, and for the Perkinsus sp. (M. mercenaria)–
specific assay genomic DNA of Perkinsus sp. (M. mercenaria) and P.
andrewsi genomic DNA were used. Assay sensitivities were assessed
in the presence or absence of 500 ng of host (C. virginica or M. mer-
cenaria) genomic DNA. Sensitivity of the PCR assays in the presence
of C. virginica genomic DNA was assessed with the use of PCR re-
action mixture A. For assessment of the sensitivity in the presence of
M. mercenaria genomic DNA, PCR reaction mixtures A and C were
used. Negative controls contained similar PCR reaction mixtures, except
that the template was replaced by sterile double-distilled H
2
O.
PCR-based detection of selected Perkinsus species in oyster and
clam samples
Five hundred nanograms of DNA extracted from C. virginica and M.
mercenaria were tested for Perkinsus species, P. marinus, P. andrewsi,
and Perkinsus sp. (M. mercenaria). For C. virginica DNA samples, PCR
reaction mixture A were used and for M. mercenaria DNA samples,
PCR reaction mixture C. Positive controls consisted of similar PCR
reaction mixtures, except that 1 ng of genomic DNA extracted from
cultured Perkinsus species was used as a template. In negative PCR
controls, the DNA template was substituted by sterile double-distilled
H
2
O.
To minimize false negatives, the small subunit rRNA gene (SSU)
was amplified from all M. mercenaria samples with the use of the
universal primers UPRA and UPRB from Medlin et al. (1988), which
are designed to amplify the SSU of all eukaryotes. For C. virginica,45
of 226 samples that were negative for the presence of Perkinsus species
were tested for the amplifiability of the SSU. PCR reaction mixtures
were identical to those used to detect Perkinsus infections in C. virgin-
ica and M. mercenaria. Positive PCR amplification controls consisted
of similar PCR reaction mixtures, except that 500 ng of genomic DNA
extracted from either C. virginica or M. mercenaria was used that was
known to be amplifiable. Cycling conditions were 94 C for 4 min, 35
cycles of 94 C for 1 min, 50 C for 1 min, 72 C for 2 min, with a final
extension at 72 C for 7 min.
Attenuation of inhibitory effects on the PCR amplification
To establish and optimize PCR conditions that would attenuate the
observed inhibition of PCR amplification, experiments were conducted
by spiking M. mercenaria genomic DNA (500 ng) with 10 pg and 1
pg P. olseni genomic DNA. These mixtures were tested for P. olseni
as described elsewhere (Robledo et al., 2000), with the use of the PCR
mixtures A, B, and C. PCR amplification controls consisted of PCR
reaction mixtures containing only P. olseni genomic DNA. Perkinsus
olseni DNA was used because infections with this species have not been
reported in the United States. Therefore, it is unlikely that the clams or
oysters carried P. olseni, allowing us to control accurately for the
amount of target DNA added to the sample.
Sequencing
Forty amplicons generated by the Perkinsus genus–specific PCR as-
say of samples that tested negative with all of the Perkinsus species–
specific assays were cored from agarose gels and reamplified with the
use of the genus-specific assay as described above. The amplified prod-
ucts were separated on 1.5% agarose gels, purified from the gels (QIA-
quick, QIAGEN) and sequenced from both directions with the primers
PER1 and PER2 (Robledo et al., 2002). Sequencing services were pro-
vided by the Bioanalytical Services Laboratory at the Center of Marine
Biotechnology, Baltimore, Maryland. Fragment assembly was per-
formed with the use of the Staden Package v1.6.0 on a Mac OS X
(Apple Computer, Inc., Cupertino, California) or Linux Fedora
娂
Core
5 (Red Hat, Inc., Raleigh, North Carolina)–based computer.
Ribosomal RNA sequences for assay design and sequence
comparisons
Sequences of rRNA genes and intergenic spacers of the rRNA se-
quences of P. andrewsi (Genbank AF102171 and AY305326), P. mar-
inus (AF497479), P. olseni (syn. P. atlanticus, AF140295), and Perkin-
sus sp. (M. mercenaria) (deposited as Perkinsus sp. CCA2001,
AF252288) were obtained from GenBank
娂
. Sequence alignments were

412 THE JOURNAL OF PARASITOLOGY, VOL. 94, NO. 2, APRIL 2008
F
IGURE
1. Sample site locations. Crassostrea virginica (
䢇
) and Mercenaria mercenaria ( ) specimens were received from shellfish providers
each month from June 2002 to September 2002. 1: Walpole (Maine); 2: Eliot (Maine); 3: Martha’s Vineyard (Massachusetts); 4: Narragansett
Bay (Rhode Island); 5: Oyster Bay (New York); 6: Delaware Bay (New Jersey); 7: Sandy Point (Maryland); 8: Cheriton (Virginia). The map was
generated with the Generic Mapping Tools, v.4.1.4 (Smith and Wessel, 1990; Wessel and Smith, 1998).

PECHER ET AL.—DISTRIBUTION OF SELECTED PERKINSUS SPECIES 413
F
IGURE
2. Specificity of the PCR-based assays. Fifty nanograms of genomic DNA from Perkinsus marinus, Perkinsus andrewsi, and Perkinsus
sp. (Mercenaria mercenaria) were tested with PCR-based assays specific for the Perkinsus genus, P. marinus, P. andrewsi, and Perkinsus sp. (M.
mercenaria) with the use of the PCR reaction mixtures A. Perkinsus sp.: Perkinsus sp. (M. mercenaria); H
2
O: negative control.
performed with the use of the Needleman–Wunsch global alignment
algorithm within EMBOSS (Rice et al., 2000). Sequence alignments
were used to design the Perkinsus sp. (M. mercenaria)–specific assay
and to identify possible new Perkinsus sp. strains in the study area.
Statistical analysis
The main focus of this article was to assess and compare infection
frequencies of Perkinsus species collected from 2 hosts at several sam-
pling sites over a relatively short sampling period (4 mo). Therefore,
the
␹
2
test and, for pairwise comparison of the sampling site and the 2
host species, the Fisher’s exact test were used. Statistical analyses were
performed with the R software suite (R Development Core Team, 2006).
Climatologic data
Temperature and precipitation data were obtained from COOP Data/
Record of Climatological Observations Forms from selected weather
stations made available to the public online by the National Climatic
Data Center, U.S. Department of Commerce (http://www7.ncdc.noaa.
gov/IPS).
RESULTS
Specificity of the diagnostic assays
To assess the specificity of the 4 PCR diagnostic assays used
in this study, 50-ng genomic DNA from clonal cultures of P.
marinus, P. andrewsi, and Perkinsus sp. (M. mercenaria) were
tested. The genus-specific assay amplified a fragment of ap-
proximately 300 bp from each DNA preparation, whereas the
assays designed to be specific for P. marinus and P. andrewsi
amplified fragments of expected size (approximately 300 bp)
only from genomic DNA preparations of the respective Perkin-
sus species (Fig. 2). The assay designed for P. andrewsi showed
intense low molecular bands when P. marinus and Perkinsus
sp. (M. mercenaria) were used as templates. Because this band
also appeared with lesser intensity in the negative PCR control,
they may represent primer dimers. The PCR assay designed for
Perkinsus sp. (M. mercenaria) amplified a fragment of expected
size (approximately 300 bp) from Perkinsus sp. (M. mercen-
aria). However, it also amplified a 300-bp fragment from P.
andrewsi (Fig. 2).
Sensitivity of the diagnostic assays
The sensitivity of the Perkinsus genus– and species–specific
assays was assessed by performing the respective assays on
serially diluted genomic DNA (100 pg to 1 fg) with the use of
the standard PCR reaction mixture A. The Perkinsus genus–
specific assay amplified 100 fg of P. marinus and 1 pg of P.
andrewsi and Perkinsus sp. (M. mercenaria) genomic DNA
(Fig. 3A). The P. marinus– and the P. andrewsi–specific assays
amplified 1 pg of P. marinus and P. andrewsi genomic DNA
(Fig. 3B, C). The assay designed for the Perkinsus sp. (M. mer-
cenaria) amplified 100 fg Perkinsus sp. (M. mercenaria) and 1
pg of P. andrewsi genomic DNA (Fig. 3D). In the genus-spe-
cific assay, the addition of 500 ng C. virginica DNA had no
effect on the detection limit of P. andrewsi and Perkinsus sp.
(M. mercenaria), but reduced the sensitivity by about 10-fold
for P. marinus. No effects of 500 ng C. virginica DNA were
observed on the P. marinus–, and P. andrewsi–specific assays
(Fig. 3B, C). In the Perkinsus sp. (M. mercenaria)–specific as-
say, although the 500 ng of C. virginica DNA had no effect on
the detection limit of P. andrewsi, it reduced the sensitivity by
about 10-fold for Perkinsus sp. (M. mercenaria) (Fig. 3D). The
addition of 500 ng of M. mercenaria genomic DNA to the PCR
reactions reduced the sensitivity by at least 1,000-fold in all 4

414 THE JOURNAL OF PARASITOLOGY, VOL. 94, NO. 2, APRIL 2008
F
IGURE
3. Sensitivity of the PCR-based assays with the use of standard PCR reaction conditions (PCR reaction mixture A) and PCR reaction
conditions optimized for Mercenaria mercenaria (PCR reaction mixture C). Decreasing amounts of genomic DNA (100 pg to 1 fg) from Perkinsus
marinus, Perkinsus andrewsi, and Perkinsus sp. (M. mercenaria) were tested with the respective PCR-based assays using the PCR reaction mixture
A. Assays were performed in the absence or presence of 500 ng Crassostrea virginica (C.v.) or M. mercenaria (M.m.) genomic DNA. (A)
Sensitivity of the genus-specific assay. (B) Sensitivity of the P. marinus–specific assay. (C) Sensitivity of the P. andrewsi–specific assay. (D)
Sensitivity of the Perkinsus sp. (M. mercenaria)–specific assay. C.v.: C. virginica; M.m.: M. mercenaria; Perkinsus sp.: Perkinsus sp. (M.
mercenaria); H
2
O: negative control; PCR mix: PCR reaction mixture.
assays for all Perkinsus species tested (Fig. 3A–D), suggesting
that the M. mercenaria genomic DNA preparations possessed
potent PCR inhibitory activity.
Attenuation of the inhibition of the PCR amplification
To obtain PCR conditions that attenuate the inhibition of the
PCR amplification, 500 ng of C. virginica and M. mercenaria
DNA were spiked with 10 pg and 1 pg P. olseni genomic DNA.
These mixtures were tested for P. olseni with the use of PCR
mixtures A, B, and C. PCR reaction mixture A did not amplify
P. olseni in the presence of M. mercencaria genomic DNA,
confirming PCR-amplification inhibition by DNA extractions
from M. mercenaria (Fig. 4). The use of PCR reaction mixture
B, which contains BSA (1 mg/ml) and DMSO (5% v/v), alle-

PECHER ET AL.—DISTRIBUTION OF SELECTED PERKINSUS SPECIES 415
F
IGURE
4. Attenuation of PCR amplification inhibition. Five hundred nanograms genomic DNA from 1 individual of Crassostrea virginica
and 2 Mercenaria mercenaria specimens were spiked with 10 pg and 1 pg Perkinsus olseni genomic DNA. Samples were tested with a P. olseni–
specific PCR-based assay with the use of PCR reaction mixtures A, B, and C. In the positive control, host DNA was omitted. H
2
O: negative
control.
viated PCR inhibitory effects in most clam DNA preparations
(Fig. 4). However, DNA extracted from some individual clams
was not amplified even in mixture B. With the use of PCR
reaction mixture C (containing TaKaRa Ex Taq
娂
,1
⫻
of the
proprietary Ex Taq reaction buffer, 1 mg/ml BSA, and 5% (v/
v) DMSO), no PCR amplification inhibition was observed in
any of DNA extractions tested (Fig. 4). Under these conditions,
the detection limits of the genus- and the species-specific assays
were 1 pg to 10 fg, respectively (Fig. 3).
False-negative PCR results analysis
To exclude false-negative PCR results from our analysis, the
SSU of DNA samples was amplified with the use of primers
that anneal in conserved regions of the SSU (Medlin et al.,
1988). Forty-five of 226 C. virginica that tested negative with
the diagnostic PCR assays were examined. In all samples tested,
the SSU was amplified (data not shown). With the use of PCR
reaction mixture C, the SSU in 225 out of 244 M. mercenaria
samples could also be amplified (data not shown). The 19 M.
mercenaria samples (7.8%) for which no amplification was ob-
served were excluded from further analysis.
Prevalence of Perkinsus species in Crassostrea virginica
In total, 625 C. virginica collected monthly from June to
September 2002 from Walpole (Maine), Martha’s Vineyard
(Massachusetts), Narragansett Bay (Rhode Island), Oyster Bay
(New York), Delaware Bay (New Jersey), and Sandy Point
(Maryland) (Fig. 1) were tested for the presence of P. marinus,
P. andrewsi, and Perkinsus sp. (M. mercenaria) (Table I). Over-
all, by using the genus-specific assay, 449 (66.5%) C. virginica
tested positive for Perkinsus species as far north as Maine. The
differences in prevalence between the sampling sites were sta-
tistically significant (P
⬍
0.001). Generally, prevalences of Per-
kinsus species infections increased from north (Walpole; 10.3%)
to south (Sandy Point; 96.5%), with the exceptions of Martha’s
Vineyard, which had a significantly higher prevalence com-
pared to Narragansett Bay (Fisher’s exact test, P
⬍
0.001), and
Oyster Bay, which had lower prevalence compared to Delaware
Bay (Fisher’s exact test, P
⬍
0.001) and Narragansett Bay
(Fisher’s exact test, P
⬍
0.001) (Table I). There were no sig-
nificant differences between prevalences at Sandy Point and
Martha’s Vineyard, and Sandy Point and Delaware Bay.
A seasonal trend in Perkinsus sp. infection prevalences was
observed in Narragansett Bay, Oyster Bay, and Delaware Bay,
where infections were lower in early summer (June), as com-
pared to mid- (August; Delaware Bay, Narragansett Bay) or late
summer (September; Oyster Bay). In Walpole, prevalences re-
mained low (0–19.4%; Fisher’s exact test, P
ⱖ
0.237) during
the sampling period. In Martha’s Vineyard, prevalences were
high (86.7–100%; Fisher’s exact test, P
ⱖ
0.173) throughout

416 THE JOURNAL OF PARASITOLOGY, VOL. 94, NO. 2, APRIL 2008
T
ABLE
I. Percent prevalence of Perkinsus species infections in Crassostrea virginica collected from June 2002 to September 2002. 1: Walpole,
Maine; 3: Martha’s Vineyard, Massachusetts; 4: Narragansett Bay, Rhode Island; 5: Oyster Bay, New York; 6: Delaware Bay, New Jersey; 7:
Sandy Point, Maryland; N: number of individuals; n: number of infected individuals; (%): prevalence in percent; P. m.: Perkinsus marinus; P. a.:
Perkinsus andrewsi; P. sp.: Perkinsus sp. (Mercenaria mercenaria); P. spp.: Perkinsus infections detected with the generic PCR-based assay.
Site N
P. spp.
n(%)
P. m .
n(%)
P. a .
n(%)
P. sp.
n(%)
P. m. and P. a.
n(%)
P. m. and P. sp.
n(%)
P. spp. only
n(%)
1 116 12 10.3 1 0.9 3 2.6 0 0.0 0 0.0 0 0.0 8 6.9
3 180 173 96.1 173 96.1 2 1.1 7 3.9 2 1.1 7 3.9 0 0.0
4 112 81 72.3 64 57.1 0 0.0 0 0.0 0 0.0 0 0.0 17 15.2
5 99 30 30.3 18 18.2 0 0.0 0 0.0 0 0.0 0 0.0 12 12.1
6 111 98 88.3 84 75.7 4 3.6 2 1.8 4 3.6 2 1.8 14 12.6
7 57 55 96.5 54 94.7 0 0.0 6 10.5 0 0.0 6 10.5 1 1.8
All 675 449 66.5 394 58.4 9 1.3 15 2.2 6 0.9 15 2.2 52 7.7
the entire sampling period. Samples from Sandy Point were not
available for the months of June and September. Prevalences at
this site did not differ between July and August (Fisher’s exact
test, P
⫽
1) (Fig. 5A).
The 3 Perkinsus species, P. marinus, P. andrewsi, and Per-
kinsus sp. (M. mercenaria), were detected in C. virginica from
Maryland to Maine. Perkinsus marinus was the dominant spe-
cies, with 394 (58.4%) C. virginica testing positive. In contrast,
only 15 (2.2%) and 9 (1.3%) C. virginica tested positive for
Perkinsus sp. (M. mercenaria) and P. andrewsi, respectively.
This trend holds true for all sampling sites with the exception
of the sampling site in Walpole, where P. marinus and P. an-
drewsi were both found at low prevalences (Table I). Coinfec-
tions with P. marinus and P. andrewsi or P. marinus and Per-
kinsus sp. (M. mercenaria) species were also observed (Table
I). However, none of the oysters that tested positive for P. an-
drewsi was positive for Perkinsus sp. (M. mercenaria).
Fifty-two oysters that tested positive for infection with a Per-
kinsus species were negative for P. marinus, P. andrewsi,or
Perkinsus sp. (M. mercenaria) (Table I). Sequence analysis of
amplicons obtained by the genus-specific PCR assay from 22
of the 52 oysters suggests that 13 oysters carried P. marinus,3
P. andrewsi, and 2 Perkinsus sp. (M. mercenaria). Four sam-
ples showed extensive sequence ambiguities, possibly due to
infections with more than 1 Perkinsus species.
Prevalence of Perkinsus species in
Mercenaria mercenaria
To assess the prevalence of Perkinsus species infections in
the hard clam M. mercenaria, 225 specimens were tested with
the PCR-based diagnostic assays described above. The speci-
mens tested were collected monthly from June 2002 to August
2002 from Eliot (Maine), Martha’s Vineyard (Massachusetts),
and July 2002 to September 2002 from Cheriton (Virginia) (Fig.
1).
Overall, by using the genus-specific PCR-based assay, a total
of 72 (32%) specimens tested positive for a Perkinsus species
(Table II). Infection prevalences differed significantly between
sites (
␹
2
test, P
⬍
0.001), increasing from north (Elliot) to south
(Cheriton) (Table II). A seasonal trend in infection prevalence
was only observed in Cheriton, where prevalence was lowest
in July and increased over the sampling period. In Martha’s
Vineyard, prevalence peaked in July. In Eliot, Perkinsus species
infections were not observed in June. Prevalence observed in
July and August did not differ considerably (Fisher’s exact test,
P
⫽
0.765) (Fig. 5B).
Overall, 10 (4.4%) clam specimens tested positive for P.
marinus, and 3 (1.3%) and 17 (7.6%) tested positive for P.
andrewsi and Perkinsus sp. (M. mercenaria), respectively. Sim-
ilar to infections in C. virginica, coinfections in individual host
specimens with P. marinus and Perkinsus sp. (M. mercenaria)
were observed (Table II). None of the clams that tested positive
for P. andrewsi was positive for Perkinsus sp. (M. mercenaria).
Perkinsus marinus infections were observed at all 3 sampling
sites. Perkinsus andrewsi infections were observed solely in
Martha’s Vineyard, and Perkinsus sp. (M. mercenaria) were
observed in Eliot and Cheriton (Table II).
Forty-four specimens tested positive for a Perkinsus species
infection, but negative for any of the Perkinsus species or iso-
lates tested. Eighteen of the 44 amplicons generated by the Per-
kinsus genus–specific PCR were sequenced. Fourteen of the ob-
tained sequences were highly similar or identical to the se-
quence of P. marinus and 3 sequences to P. andrewsi. One
sequence showed extensive ambiguities, suggesting an infection
with more than 1 Perkinsus species.
Comparison of Perkinsus species prevalences in
Crassostrea virginica and Mercenaria mercenaria
Prevalence of Perkinsus species and P. marinus in M. mer-
cenaria was significantly lower compared to prevalence in C.
virginica (Fisher’s exact test, P
⬍
0.001). Perkinsus andrewsi
prevalence was similar in both host species (Fisher’s exact test,
P
⫽
1). Prevalence of Perkinsus sp. (M. mercenaria) was sig-
nificantly higher in M. mercenaria (Fisher’s exact test, P
⬍
0.001).
Martha’s Vineyard provides a very useful side-by-side com-
parison of prevalence in both clams and oysters because both
hosts were collected from proximal locations. Here, 96.1% of
the C. virginica were infected with a Perkinsus species, com-
pared to 26.2% of the M. mercenaria (Fig. 6; Tables I, II). All
infected C. virginica specimens carried P. marinus, and P. an-
drewsi infected 1.1% and Perkinsus sp. (M. mercenaria) 3.9%
of C. virginica. Each of the oysters infected with P. andrewsi
and Perkinsus sp. (M. mercenaria) were dually infected with
P. marinus.InM. mercenaria, P. marinus and P. andrewsi were
present in low prevalences (3.3% and 4.9%, respectively). Per-

PECHER ET AL.—DISTRIBUTION OF SELECTED PERKINSUS SPECIES 417
F
IGURE
5. Monthly Perkinsus infection prevalence in Crassostrea virginica and Mercenaria mercenaria. Percent prevalences of Perkinsus
species (detected by the genus-specific assay) (
䡵
), Perkinsus marinus (
u
), Perkinsus andrewsi (
q
), Perkinsus sp. (M. mercenaria)(
r
), dual
infections with P. marinus and P. andrewsi (
t
), as well as P. marinus and Perkinsus sp. (M. mercenaria)( ), and infections with Perkinsus
species only (
䡺
) are shown for all sampling sites. (A) Percent prevalence in C. virginica.(B) Percent prevalence in M. mercenaria. Jun: June;
Jul: July; Aug: August; Sep: September.

418 THE JOURNAL OF PARASITOLOGY, VOL. 94, NO. 2, APRIL 2008
T
ABLE
II. Percent prevalence of Perkinsus species infections in Mercenaria mercenaria collected from June 2002 to September 2002. 2: Eliot,
Maine; 3: Martha’s Vineyard, Massachusetts; 8: Cheriton, Virginia; N: number of individuals; n: number of infected individuals; (%): prevalence
in percent; P. m.: Perkinsus marinus; P. a.: Perkinsus andrewsi; P. sp.: Perkinsus sp. (M. mercenaria); P. spp.: Perkinsus infections detected with
the generic PCR-based assay.
Site N
P. spp.
N(%)
P. m .
n(%)
P. a .
n(%)
P. sp.
n(%)
P. m. and P. a.
n(%)
P. m. and P. sp.
n(%)
P. spp. only
n(%)
2 79 15 19.0 4 5.1 0 0.0 3 3.8 0 0.0 2 2.5 10 12.7
3 61 16 26.2 2 3.3 3 4.9 0 0.0 0 0.0 0 0.0 11 18.0
8 85 41 48.2 4 4.7 0 0.0 14 16.5 0 0.0 0 0.0 23 27.1
All 225 72 32.0 10 4.4 3 1.3 17 7.6 0 0.0 2 0.9 44 19.6
F
IGURE
6. Comparison of the Perkinsus infection prevalence in
Crassostrea virginica and Mercenaria mercenaria. Percent prevalences
of Perkinsus species, Perkinsus marinus, Perkinsus andrewsi, and Per-
kinsus sp. (M. mercenaria)inC. virginica (
䡵
) and M. mercenaria (
u
)
collected from June 2002 to September 2002 in Martha’s Vineyard,
Massachusetts are shown. Perkinsus sp.: Perkinsus sp. (M. mercenaria).
kinsus sp. (M. mercenaria) was not detected (Fig. 6; Tables I,
II). Differences in prevalence between C. virginica and M. mer-
cenaria were statistically significant for Perkinsus species and
P. marinus infections (Fisher’s exact test, P
⬍
0.001), but not
for P. andrewsi (Fisher’s exact test, P
⫽
0.105) or Perkinsus
sp. (M. mercenaria) (Fisher’s exact test, P
⫽
0.208).
Water temperature and precipitation
Water temperature and salinity are the main environmental
factors affecting Perkinsus species infections (Andrews, 1988;
Burreson and Ragone Calvo, 1996). Air temperature is often
used as a substitute for water temperature, because it is more
frequently recorded and correlates with the temperature of near-
by water bodies (Jeffries and Johnson, 1976; Sauriau, 1991).
Salinity of coastal water bodies is influenced by freshwater in-
flux from major rivers and precipitation. Thus, to compare prev-
alences to an earlier study on the distribution of P. marinus,
and 2 Haplosporidium species in C. virginica collected in 2000
(Russell et al., 2004), temperature and precipitation data were
obtained for May to August 2000 and 2002, from Mineola
(New York) (COOP ID 305377), a weather station near Oyster
Bay (New York). Mean monthly air temperatures in 2000 were
16.9, 20.4, 21.2, and 21.9 C in May, June, July, and August,
respectively. Monthly precipitation was 108.2, 111.3, 137.7,
and 61.2 mm over the same time period, respectively. In 2002,
mean monthly temperatures were 15.2, 21.0, 25.9, and 24.8 C
from May to August. Monthly precipitation (May–August) was
93.0, 95.8, 22.9, and 183.9 mm.
DISCUSSION
The aim of the present study was to assess the presence of
P. marinus, P. andrewsi, and Perkinsus sp. (M. mercenaria)in
2 economically important bivalves, C. virginica and M. mer-
cenaria, with the use of PCR-based assays. Along the Gulf of
Mexico and Atlantic coast of the United States, 3 Perkinsus
species have been described, i.e., P. marinus (Mackin et al.,
1950; type host C. virginica), P. chesapeaki (McLaughlin et al.,
2000; type host M. arenaria), and P. andrewsi (Coss, Robledo,
Ruiz, and Vasta, 2001; type host M. balthica).
The heterospecificity of P. andrewsi and P. chesapeaki has
been controversial and synonymization has been suggested
(Burreson et al., 2005). According to the original description,
P. chesapeaki is a distinct morphotype (McLaughlin et al.,
2000). In contrast, P. andrewsi cannot be distinguished from
other Perkinsus species based on morphology, but sequences of
the rRNA genes and intergenic regions of P. andrewsi differ
from other Perkinsus species (Coss, Robledo, and Vasta, 2001;
Coss, Robledo, Ruiz, and Vasta, 2001). The Perkinsus isolate
that was analyzed to clarify the relationship of P. andrewsi and
P. chesapeaki has been designated as P. chesapeaki because it
has been isolated from the appropriate type host. However, this
isolate appears to be morphologically identical to P. andrewsi
and, thus, may not be the P. chesapeaki originally described.
Therefore, until additional evidence is obtained, we retain the
P. andrewsi designation for the present study.
The standard diagnostic assay for Perkinsus species that is
based on the FTM method does not distinguish between the
sympatric Perkinsus species found along the Atlantic coast
(Ray, 1952; Coss, 2000; McLaughlin et al., 2000; Coss, Rob-
ledo, Ruiz, and Vasta, 2001). However, several Perkinsus spe-
cies–specific assays are available (Marsh et al., 1995; Yarnall
et al., 2000; Coss, Robledo, Ruiz, and Vasta, 2001; Robledo et
al., 2002). No PCR-based assay has been developed for P. ches-
apeaki, mainly due to the lack of a bona fide type culture that
would allow design and validation of such an assay.
For the purpose of the present study, we used available PCR-
based assays specific for the genus Perkinsus, and the species
P. marinus and P. andrewsi (Marsh et al., 1995; Coss, Robledo,
Ruiz, and Vasta, 2001; Robledo et al., 2002), and developed a
new PCR-based assay for a Perkinsus species isolated from M.

PECHER ET AL.—DISTRIBUTION OF SELECTED PERKINSUS SPECIES 419
mercenaria [Perkinsus sp. (M. mercenaria)]. The species-spe-
cific assays are based on sequence differences within the inter-
genic spacer (IGS) of the rRNA gene locus of Perkinsus species
that links the 5S and SSU genes, whereas the genus-specific
assay is based on a conserved region at the 3
⬘
end of the IGS
(Marsh et al., 1995; Coss, Robledo, Ruiz, and Vasta, 2001; Rob-
ledo et al., 2002). As expected, the genus-specific assay de-
tected all Perkinsus species tested in this study, including Per-
kinus sp. (M. mercenaria). Perkinsus sp. (M. mercenaria) was
not detected by the P. marinus–specific or the P. andrewsi–
specific assays, demonstrating the capacity of the genus-specific
Perkinus assay to detect new Perkinsus species or strains.
The assay designed for Perkinsus sp. (M. mercenaria) also
amplified P. andrewsi genomic DNA extracted from the P. an-
drewsi type culture. This is to be expected, for we have pre-
viously shown that P. andrewsi has 2 distinct rRNA gene units
(types A and B) (Pecher et al., 2004). Sequence analysis of the
rRNA gene unit of the Perkinsus sp. (M. mercenaria) revealed
only 1 rRNA gene unit that is very similar in sequence to the
rRNA-B gene unit of the P. andrewsi type culture (W. T. Pecher
and G. R. Vasta, unpubl. obs.). In particular, the IGS of Per-
kinsus sp. (M. mercenaria) is 98.9% identical to the IGS of the
P. andrewsi rRNA-B gene unit (W. T. Pecher and G. R. Vasta,
unpubl. obs.), explaining the cross amplification. On the other
hand, the P. andrewsi–specific assay does not detect Perkinsus
sp. (M. mercenaria), because it has been developed based on
the IGS of the rRNA-A gene unit that is only 71.3% identical
to the IGS of Perkinsus sp. (M. mercenaria).
The sensitivity of each species-specific assay observed in our
study (0.1–1-pg genomic DNA of the respective Perkinsus spe-
cies) is similar to the sensitivities for the P. marinus and P.
andrewsi diagnostic assays reported by Marsh et al. (1995) and
Coss, Robledo, Ruiz, and Vasta (2001). Our data suggest that
the sensitivity of the genus-specific assay (10 fg to 1 pg ge-
nomic Perkinsus species DNA) is equal, or greater, compared
to the respective species-specific assays, allowing us to identify
low-intensity Perkinsus infections.
Inhibition of PCR amplification is frequently observed in en-
vironmental and biological samples (reviewed by Wilson,
1997). Inhibitory substances include organic and phenolic com-
pounds, humic acids, heavy metals, fats, and polysaccharides.
In molluscs, PCR inhibition has been attributed to glycogen
(Hill et al., 1991; Andersen and Omiecinski, 1992; Atmar et
al., 1993). Modified DNA extraction protocols (Atmar et al.,
1993), and inclusion of additives in the PCR reaction mixture
is commonly used to attenuate the effects of the interfering
substances (reviewed in Wilson, 1997). In addition, commer-
cially available kits have been developed that can be used to
extract DNA from plants, animals, and fungi from complex
sources such as the soil and other environmental samples.
We did not observe PCR inhibition in C. virginica DNA ex-
tracts. However, PCR inhibition was dramatic in DNA from M.
mercenaria, and has been observed in scallop DNA (Argopec-
ten irradians) extracted with a commercial tissue kit (W. T.
Pecher and G. R. Vasta, unpubl. obs.). We succeeded in atten-
uating the PCR amplification inhibition in M. mercenaria sam-
ples by adding 1 mg/ml BSA in combination with 5% (v/v)
DMSO to the PCR reaction mixture. While using regular Taq
DNA polymerase with a standard PCR buffer, we were able to
PCR amplify the SSU from 70% of all M. mercenaria samples
(data not shown). The use of a specialty
Taq DNA polymerase
with its optimized buffer system designed to amplify large
DNA fragments increased the success rate to 92%. However,
similar reaction conditions failed to amplify scallop DNA sam-
ples (W. T. Pecher and G. R. Vasta, unpubl. obs.). These find-
ings underline the importance of validation of PCR conditions
for each sample type used, including template ‘‘amplifiability’’
vis-a`-vis inhibition. Once DNA is extracted from samples and
PCR conditions are optimized, the PCR-based assays enable
detection of any Perkinsus species and different Perkinsus spe-
cies in the same sample. Thus, the application of the genus-
and species-specific assays presents a valuable alternative to the
FTM assay.
Studies based on the FTM assay documented the distribution
of Perkinsus species in oysters from the Yucata´n Peninsula,
Mexico, to Maine (Burreson et al., 1994; Ford, 1996; Soniat,
1996). These infections have been attributed to P. marinus.
However, the discovery of additional Perkinsus species and the
development of specific PCR assays for them have provided
tools to test this assumption. In the present study, commercially
harvested C. virginica and M. mercenaria populations were
tested for the presence of Perkinsus species, P. marinus, P.
andrewsi, and a Perkinsus sp. isolated from M. mercenaria with
the use of specific PCR-based assays.
In accordance with the studies identified above, with the use
of the genus-specific assay in both bivalve host species, Per-
kinsus species, infections were observed as far north as Maine.
The intensity of the amplicons obtained by the PCR-based as-
says suggested, in the majority of the positive samples, the pres-
ence of 10 pg or more of Perkinsus spp. DNA (data not shown),
which is roughly equivalent to more than 100 Perkinsus spp.
cells (see below). However, it cannot be ruled out that in some
specimens that yielded low-intensity amplicons, these actually
reveal only the presence of parasite rather than true infections.
With the use of the genus-specific assay, prevalences of Per-
kinsus species in C. virginica appeared lower in early summer
(June) than in mid- (August) or late summer (September) in
Delaware Bay, Oyster Bay, and Narragansett Bay. Similar
trends were observed in the Chesapeake Bay and other regions.
When compared to C. virginica, significantly fewer M. mercen-
aria specimens tested positive for the genus Perkinsus and P.
marinus. Prevalences of Perkinsus species (as assessed by the
genus-specific method) differed from site to site in both host
species. Generally, prevalences of Perkinsus species decreased
from south to north, with the exception of Martha’s Vineyard
(Massachusetts) and Oyster Bay (New York) in C. virginica.In
Martha’s Vineyard, prevalences were surprisingly high (86.7–
100.0% over the 4-mo study period) compared to those in Nar-
ragansett Bay and Oyster Bay and are in contrast to observa-
tions by Russell et al. (2004). These authors did not observe P.
marinus infections in C. virginica specimens that were collected
in September 2000 and tested by a multiplex PCR-based assay,
suggesting that Martha’s Vineyard may have experienced a lo-
calized Perkinsus epizootic in 2002.
Prevalences in Oyster Bay were significantly lower compared
to Narragansett Bay. However, the observed prevalences in
Oyster Bay in 2002 were higher than those reported by Russell
et al. (2004), who observed 0% and 3% P. marinus prevalence
in C. virginica collected in June and August 2000, respectively,
from a site in Oyster Bay. Similarly, we observed no P. marinus

420 THE JOURNAL OF PARASITOLOGY, VOL. 94, NO. 2, APRIL 2008
infections in Oyster Bay in June 2002. In August 2002, how-
ever, 17% of C. virginica were infected with P. marinus. These
differences may be due to higher temperatures and drier con-
ditions in 2002 compared to 2000, as judged by monthly mean
air surface temperatures and precipitation recorded by a nearby
weather station (Mineola, New York, COOP ID 305377). High-
er temperatures and drier conditions may result in higher water
temperatures and higher salinity, both conditions favorable to
P. marinus infections (Andrews, 1988; Burreson et al., 1996).
Furthermore, differences in the sensitivities of the PCR assays
may have contributed to the observed differences. The PCR
assay used in the current study detects 1 pg genomic DNA of
P. marinus. Based on the following calculation, the PCR-based
assay used in the present study is about 2.5-fold more sensitive
than the one used by Russell et al. (2004). Based on available
sequences, P. marinus has an estimated genome size between
70 and 80 Mb (N. M. El-Sayed, J. A. Fernandez-Robledo, and
G. R. Vasta, unpubl. obs.; http://www.tigr.org/tdb/e2k1/pmg/),
and a single rRNA gene unit is approximately 7 kb long (J. A.
Fernandez-Robledo and G. R. Vasta, unpubl. obs.). If a genome
size of 80 Mb is assumed, then 1 pg genomic of DNA repre-
sents 13 genome equivalents. If it is furthermore assumed that,
similar to P. olseni (syn. P. atlanticus) (de la Herra´n et al.,
2000), P. marinus rRNA genes are encoded by 5% of its ge-
nome, then about 570 copies of a single rRNA gene unit are
present per genome equivalent. Thus, the PCR assay used in
the present study would detect 7,500 copies of the rRNA gene
unit, or, as each gene unit contains 1 IGS, roughly 7,500 copies
of the IGS. The multiplex PCR assay used by Russell et al.
(2004) detects 100 fg of plasmid DNA containing the P. mari-
nus IGS (Russell et al. 2004). Because the plasmid with the
IGS sequence is approximately 4.5 kb, 100 fg plasmid DNA
represents about 20,000 copies of the IGS.
Our data enabled us to extend the observed range of P. an-
drewsi and Perkinsus sp. (M. mercenaria) from Chesapeake
Bay to Maine. It is noteworthy that the prevalence of Perkinsus
sp. (M. mercenaria) was highest in the south (Cheriton, Virgin-
ia), suggesting that this Perkinsus isolate may prefer warmer
waters. However, until investigations on the southern distribu-
tion range of this Perkinsus isolate are conducted, this remains
speculative. Interestingly, we did not detect P. andrewsi in
farmed oysters in Chesapeake Bay. This finding is in contrast
to reports of Coss, Robledo, Ruiz, and Vasta (2001) that indi-
cate 65% of 125 C. virginica collected from natural populations
throughout Chesapeake Bay were infected with P. andrewsi,
64% with P. marinus, and 34% with both. In our study, 94.7%
of oysters were infected with P. marinus and 10.5% with Per-
kinsus sp. (M. mercenaria). A possible explanation is that the
conditions at the particular location were favorable for a P.
marinus infection. Alternatively, and not mutually exclusively,
it is conceivable that once a P. marinus infection has been es-
tablished in an oyster population, it may outgrow other Perkin-
sus infections. To address these questions, further investigations
need to be conducted on the infection dynamics of different
Perkinsus species.
Perkinsus marinus appears to be the most prevalent Perkin-
sus species in C. virginica. Therefore, discrepancies between
high infection density, prevalence, and low mortality observed
in other studies in C. virginica (Ford, 1996; Karolus et al.,
2000) cannot be attributed to the presence of a different Per-
kinsus species with less pathogenicity toward the oyster. Alter-
native hypotheses that will require further study include the
notion of variable P. marinus strains with different pathogenic-
ities, rather than different Perkinsus species, and, perhaps, dif-
ferences in environmental factors such as cooler summer peak
temperatures that could influence the outcome of an infection
(Ford, 1996). Of course, the presence of host populations with
different genetic backgrounds could be another component
leading to lower host mortality. This is exemplified by the ob-
servation that in M. mercenaria Perkinsus
sp. (M. mercenaria)
was the most prevalent Perkinsus species.
Because of sequence similarities of the second rRNA gene
unit (rRNA-B) of the P. andrewsi hapantotype to the Perkinsus
sp. isolated from M. mercenaria (W. T. Pecher and G. R. Vasta,
unpubl. obs.), the P. andrewsi hapantotype is detected by the
Perkinsus sp. (M. mercenaria)–specific assay with a 10-fold
lower sensitivity. In our study, none of the clams and oysters
that tested positive for P. andrewsi also tested positive for Per-
kinsus sp. (M. mercenaria). Certainly, the lower sensitivity of
the Perkinsus sp. (M. mercenaria) toward P. andrewsi may par-
tially explain this observation, but the presence of P. andrewsi
isolates that contain only the rRNA-A gene unit and thus are
not detected by the Perkinsus sp. (M. mercenaria) assay cannot
be ruled out. Because the rRNA unit of Perkinsus sp. (M. mer-
cenaria) and the P. andrewsi rRNA-B unit share high sequence
similarities, Perkinsus sp. (M. mercenaria) could be considered
a variant of P. andrewsi that possesses only the rRNA-B gene
unit. Similar observations have been reported for Trypanosoma
cruzi, where isolates have been identified that possess either 2
distinct rRNA gene units (rRNA unit 1 and 2), or one of the 2
rRNA gene units (Souto et al., 1996; Zingales et al., 1999; Stolf
et al., 2003).
Application of the genus-specific PCR-based assay to both
C. virginica and M. mercenaria specimens resulted in the de-
tection of Perkinsus infections that could not be attributed to
P. marinus, P. andrewsi,orPerkinsus sp. (M. mercenaria)by
the species-specific PCR assays. However, sequence analysis of
selected amplicons generated by the genus-specific PCR failed
to reveal novel sequences, which could suggest the presence of
yet-undescribed Perkinsus species or strains. Further, it rather
suggested that the density of Perkinsus infections in these spec-
imens was below the detection limit of the species-specific as-
says, but high enough to be detected by the genus-specific as-
say.
Although Perkinsus species appear to lack strict host speci-
ficity, they may have adapted best to the hostile environment
of one particular host species. Studies on the effects of plasma
of different mollusc species on the in vitro proliferation of P.
marinus show that parasite growth is reduced by plasma or sera
from bivalve molluscs (Andara ovalis, Geukensia demissa, M.
mercenaria
, and Mytilus edulis) that are naturally exposed to
the parasite as compared to plasma or sera from C. virginica
(Anderson, 2001; Gauthier and Vasta, 2002). These observa-
tions suggest a preference of P. marinus for its type host C.
virginica. Results from our study provide further evidence for
a possible host preference of P. marinus. Although P. marinus
was detected in M. mercenaria and C. virginica, the prevalence
in M. mercenaria was significantly lower. Further studies aimed
at elucidating the molecular mechanisms behind this host pref-
erence are ongoing.

PECHER ET AL.—DISTRIBUTION OF SELECTED PERKINSUS SPECIES 421
ACKNOWLEDGMENTS
This study was supported by grants 556804 from NRAC, USDA, and
SA7528035-D and SA7528068-H from the Maryland Sea Grant,
NOAA. The authors are indebted to Gary Arnold (Mook Sea Farm Inc.,
Walpole, Maine), Kenneth LaValley (Spinney Creek Shellfish Inc., El-
iot, Maine), Richard C. Karney (Martha’s Vineyard Shellfish Group,
Oaks Bluff, Massachusetts), Robert B. Rheault (Moonstone Oysters,
Wakefield, Rhode Island), Braun Seafood Co. (Cutchogue, New York),
Susan E. Ford (Rutgers University, Haskins Shellfish Research Labo-
ratory, Port Norris, New Jersey), Donald W. Merritt (University of
Maryland Center for Environmental Science, Horn Point Laboratory,
Cambridge, Maryland), and Mike Peirson (Cherrystone Aquafarms,
Cheriton, Virginia) for providing oysters (C. virginica) and clams (M.
mercenaria), and to Harmony A. Hancock for assistance with shellfish
dissection.
LITERATURE CITED
A
BOLLO
, E., S. M. C
ASAS
,G.C
ESCHIA
,
AND
A. V
ILLALBA
. 2006. Differ-
ential diagnosis of Perkinsus species by polymerase chain reaction–
restriction fragment length polymorphism assay. Molecular and
Cellular Probes 20: 323–329.
A
NDERSEN
, M. R.,
AND
C. J. O
MIECINSKI
. 1992. Direct extraction of
bacterial plasmids from food for polymerase chain reaction ampli-
fication. Applied and Environmental Microbiology 58: 4080–4082.
A
NDERSON
, R. A. 2001. A comparative study of anti–Perkinsus marinus
activity in bivalve sera. Journal of Shellfish Research 20: 1011–
1017.
A
NDREWS
, J. D. 1955. Notes on fungus parasites of bivalve mollusks in
Chesapeake Bay. Proceedings of the National Shellfisheries Asso-
ciation 45: 157–163.
———. 1988. Epizootiology of the disease caused by the oyster path-
ogen Perkinsus marinus and its effect on the oyster industry. Amer-
ican Fisheries Society Special Publication 18: 47–63.
A
TMAR
, R. L., T. G. M
ETCALF
,F.H.N
EILL
,
AND
M. K. E
STES
. 1993.
Detection of enteric viruses in oysters by using the polymerase
chain reaction. Applied and Environmental Microbiology 59: 631–
635.
B
URRESON
, E. M., R. S. A
LVAREZ
,V.V.M
ARTINEZ
,
AND
L. A. M
ACEDO
.
1994. Perkinsus marinus (Apicomplexa) as a potential source of
oyster Crassostrea virginica mortality in coastal lagoons of Tabas-
co, Mexico. Diseases of Aquatic Organisms 20: 77–82.
———,
AND
L. M. R
AGONE
C
ALVO
. 1996. Epizootiology of Perkinsus
marinus disease of oysters in Chesapeake Bay, with emphasis on
data since 1985. Journal of Shellfish Research 15: 17–34.
———, K. S. R
EECE
,
AND
C. F. D
UNGAN
. 2005. Molecular, morpholog-
ical, and experimental evidence support the synonymy of Perkinsus
chesapeaki and Perkinsus andrewsi. Journal of Eukaryotic Micro-
biology 52: 258–270.
B
USHEK
, D., C. F. D
UNGAN
,
AND
A. J. L
EWITUS
. 2002. Serological affin-
ities of the oyster pathogen Perkinsus marinus (Apicomplexa) with
some dinoflagellates (Dinophyceae). Journal of Eukaryotic Micro-
biology 49: 11–16.
C
HOI
, K. S., D. H. L
EWIS
,E.N.P
OWELL
,P.F.F
RELIER
,
AND
S. M. R
AY
.
1991. A polyclonal antibody developed from Perkinsus marinus
hypnospores fails to cross react with other life stages of P. marinus
in oyster (Crassostrea virginica) tissues. Journal of Shellfish Re-
search 10: 411–415.
C
OSS
, C. A. 2000. Investigation of Perkinsus species from clams sym-
patric to oysters with emphasis on infections in the Baltic clam
Macoma balthica of Chesapeake bay. Ph. D. Thesis. George Wash-
ington University, Washington, DC, 247 p.
———, J. A. F. R
OBLEDO
,G.M.R
UIZ
,
AND
G. R. V
ASTA
. 2001. De-
scription of Perkinsus andrewsi n. sp. isolated from the Baltic clam
(Macoma balthica) by characterization of the ribosomal RNA lo-
cus, and development of a species-specific PCR-based diagnostic
assay. Journal of Eukaryotic Microbiology 48: 52–61.
———, ———,
AND
G. R. V
ASTA
. 2001. Fine structure of clonally
propagated in vitro life stages of a Perkinsus sp. isolated from the
Baltic clam Macoma balthica. Journal of Eukaryotic Microbiology
48: 38–51.
DE LA
H
ERRA
´
N
, R., M. A. G
ARRIDO
-R
AMOS
,J.I.N
AVAS
,C.R
UIZ
R
EJO
´
N
,
AND
M. R
UIZ
R
EJO
´
N
. 2000. Molecular characterization of the ribo-
somal RNA gene region of Perkinsus atlanticus: Its use in phylo-
genetic analysis and as a target for a molecular diagnosis. Parasi-
tology 120: 345–353.
D
UNGAN
,C.F.,
AND
B. S. R
OBERTSON
. 1993. Binding specificities of
mono- and polyclonal antibodies to the protozoan oyster pathogen
Perkinsus marinus. Diseases of Aquatic Organisms 15: 9–22.
E
LANDALLOUSSI
,L.M.,R.M.L
EITE
,R.A
FONSO
,P.A.N
UNES
,J.A.
R
OBLEDO
,G.R.V
ASTA
,
AND
M. L. C
ANCELA
. 2004. Development
of a PCR-ELISA assay for diagnosis of Perkinsus marinus and
Perkinsus atlanticus infections in bivalve molluscs. Molecular and
Cellular Probes 18: 89–96.
F
ORD
, S. E. 1996. Range extension by the oyster parasite Perkinsus
marinus into the northeastern United States: Response to climate
change? Journal of Shellfish Research 15: 45–56.
G
AUTHIER
, J. D., C. R. M
ILLER
,
AND
A. E. W
ILBUR
. 2006. TAQ-
MAN
㛮
MCG real-time PCR approach to quantification of Perkinsus
marinus and Perkinsus spp. in oysters. Journal of Shellfish Re-
search 25: 619–624.
———,
AND
G. R. V
ASTA
. 1993. Continuous in vitro culture of the
eastern oyster parasite Perkinsus marinus. Journal of Invertebrate
Pathology 62: 321–323.
———,
AND
———. 2002. Effects of plasma from bivalve mollusk
species on the in vitro proliferation of the protistan parasite Per-
kinsus marinus. Journal of Experimental Zoology 292: 221–230.
H
ILL
, W. E., S. P. K
EASLER
,M.W.T
RUCKSESS
,P.F
ENG
,C.A.K
AYSNER
,
AND
K. A. L
AMPEL
. 1991. Polymerase chain reaction identification
of Vibrio vulnificus in artificially contaminated oysters. Applied and
Environmental Microbiology 57: 707–711.
J
EFFRIES
,H.P.,
AND
W. C. I. J
OHNSON
. 1976. Petroleum, temperature,
and toxicants: Examples of suspected responses by plankton and
benthos on the continental shelf. In Effects of energy-related activ-
ities on the Atlantic continental shelf, B. Manowitz (ed.). Brook-
haven National Laboratory, New York, New York, p. 96–108.
K
AROLUS
, J., I. S
UNILA
,S.S
PEAR
,
AND
J. V
OLK
. 2000. Prevalence of
Perkinsus marinus (Dermo) in Crassostrea virginica along the
Connecticut shoreline. Aquaculture 183: 215–221.
K
LEINSCHUSTER
, S. J.,
AND
S. L. S
WINK
. 1993. A simple method for the
in vitro culture of Perkinsus marinus. Nautilus 107: 76–78.
K
OTOB
, S. I., S. M. M
C
L
AUGHLIN
,P.V
AN
B
ERKUM
,
AND
M. F
AISAL
. 1999.
Discrimination between two Perkinus spp. isolated from the soft-
shell clam, Mya arenaria, by sequence analysis of two internal
transcribed spacer regions and the 5.8S ribosomal RNA gene. Par-
asitology 199: 363–368.
L
A
P
EYRE
,J.F.,M.F
AISAL
,
AND
E. M. B
URRESON
. 1993. In vitro prop-
agation of the protozoan Perkinsus marinus, a pathogen of the east-
ern oyster, Crassostrea virginica. Journal of Eukaryotic Microbi-
ology 40: 304–310.
M
ACKIN
, J. G., H. M. O
WEN
,
AND
A. C
OLLIER
. 1950. Preliminary note
on the occurrence of a new protistan parasite, Dermocystidium mar-
inum n. sp. in Crassostrea virginica (Gmelin). Science 111: 328–
329.
M
ARSH
, A. G., J. D. G
AUTHIER
,
AND
G. R. V
ASTA
. 1995. A semiquan-
titative PCR assay for assessing Perkinsus marinus infections in
the eastern oyster, Crassostrea virginica. Journal of Parasitology
81: 577–583.
M
C
L
AUGHLIN
,S.M.,
AND
M. F
AISAL
. 1999. A comparison of the diag-
nostic assays for detection of Perkinsus spp. in the softshell clam
Mya arenaria. Aquaculture 172: 197–204.
———, B. D. T
ALL
,A.S
HAHEEN
,E.E.E
LSAYED
,
AND
M. F
AISAL
. 2000.
Zoosporulation of a new Perkinus species isolated from the gills
of the softshell clam Mya arenaria. Parasite 7: 115–122.
M
EDLIN
, L., H. J. E
LWOOD
,S.S
TICKEL
,
AND
M. L. S
OGIN
. 1988. The
characterization of enzymatically amplified eukaryotic 16S-like
rRNA-coding regions. Gene 71: 491–499.
M
ONTES
,J.F.,M.D
URFORT
,A.L
LADO
´
,
AND
J. G
ARC
I
´
A
V
ALERO
. 2002.
Characterization and immunolocalization of a main proteinaceous
component of the cell wall of the protozoan parasite Perkinsus
atlanticus. Parasitology 124: 477–484.
O
TTINGER
, C. A., T. D. L
EWIS
,D.A.S
HAPIRO
,M.F
AISAL
,
AND
S. L.
K
AATTARI
. 2001. Detection of Perkinsus marinus extracellular pro-
teins in tissues of the eastern oyster Crassostrea virginica: Potential

422 THE JOURNAL OF PARASITOLOGY, VOL. 94, NO. 2, APRIL 2008
use in diagnostic assays. Journal of Aquatic Animal Health 13:
133–141.
P
ECHER
, W. T., J. A. F. R
OBLEDO
,
AND
G. R. V
ASTA
. 2004. Identification
of a second rRNA gene unit in the Perkinsus andrewsi genome.
Journal of Eukaryotic Microbiology 51: 234–245.
P
ENNA
, M. S., M. K
HAN
,
AND
R. A. F
RENCH
. 2001. Development of a
multiplex PCR for the detection of Haplosporidium nelsoni, Hap-
losporidium costale and Perkinsus marinus in the eastern oyster
(Crassostrea virginica, Gmelin, 1971). Molecular and Cellular
Probes 15: 385–390.
P
ERKINS
, F. O. 1988. Parasite morphology, strategy, and evolution.
American Fisheries Society Special Publication 18: 93–111.
RD
EVELOPER
C
ORE
T
EAM
. 2006. R: A language and environment for
statistical computing. R Foundation for Statistical Computing, Vi-
enna, Austria.
R
AY
, S. M. 1952. A culture technique for the diagnosis of infection
with Dermocystidium marinum in oysters. Science 116: 360–361.
———. 1966. A review of the culture method for detecting Dermocys-
tidium marinum, with suggested modifications and precautions.
Proceedings of the National Shellfisheries Association 54: 55–69.
R
ICE
, P., I. L
ONGDEN
,
AND
A. B
LEASBY
. 2000. EMBOSS: The European
Molecular Biology Open Software Suite. Trends in Genetics 16:
276–277.
R
OBLEDO
,J.A.F.,C.A.C
OSS
,
AND
G. R. V
ASTA
. 2000. Characterization
of the ribosomal RNA locus of Perkinsus atlanticus and develop-
ment of a polymerase chain reaction–based diagnostic assay. Jour-
nal of Parasitology 86: 972–978.
———, J. D. G
AUTHIER
,C.A.C
OSS
,A.C.W
RIGHT
,
AND
G. R. V
ASTA
.
1998. Species-specificity and sensitivity of a PCR-based assay for
Perkinsus marinus in the eastern oyster, Crassostrea virginica:A
comparison with the fluid thioglycollate assay. Journal of Parasi-
tology 84: 1237–1244.
———, P. A. N
UNES
,M.L.C
ANCELA
,
AND
G. R. V
ASTA
. 2002. Devel-
opment of an in vitro clonal culture and characterization of the
rRNA gene cluster of Perkinsus atlanticus, a protistan parasite of
the clam Tapes decussatus. Journal of Eukaryotic Microbiology 49:
414–422.
R
USSELL
, S., S. F
RASCA
,J
R
., I. S
UNILA
,
AND
R. A. F
RENCH
. 2004. Ap-
plication of a multiplex PCR for the detection of protozoan path-
ogens of the eastern oyster Crassostrea virginica in field samples.
Diseases of Aquatic Organisms 59: 85–91.
S
AURIAU
, P.-G. 1991. Spread of Cyclope neritea (Mollusca: Gastropoda)
along the north-eastern Atlantic coasts in relation to oyster culture
and to climatic fluctuations. Marine Biology 109: 299–309.
S
MITH
,W.H.F.,
AND
P. W
ESSEL
. 1990. Gridding with continuous cur-
vature splines in tension. Geophysics 55: 293–305.
S
ONIAT
, T. M. 1996. Epizootiology of Perkinsus marinus disease of
eastern oysters in the Gulf of Mexico. Journal of Shellfish Research
15: 35–43.
S
OUTO
, R. P., O. F
ERNANDES
,A.M.M
ACEDO
,D.A.C
AMPBELL
,
AND
B.
Z
INGALES
. 1996. DNA markers define two major phylogenetic lin-
eages of Trypanosoma cruzi. Molecular and Biochemical Parasi-
tology 83: 141–152.
S
TOLF
, B. S., R. P. S
OUTO
,A.P
EDROSO
,
AND
B. Z
INGALES
. 2003. Two
types of ribosomal RNA genes in hybrid Trypanosoma cruzi
strains. Molecular and Biochemical Parasitology 126: 73–80.
V
ILLALBA
, A., K. S. R
EECE
,M.C.O
RDA
´
S
,S.M.C
ASAS
,
AND
A. F
I
-
GUERAS
. 2004. Perkinsosis in molluscs: A review. Aquatic Living
Resources 17: 411–432.
W
ESSEL
,P.,
AND
W. H . F. S
MITH
. 1998. New, improved versions of the
Generic Mapping Tools released. Eos Transactions of the American
Geophysical Union 79: 579.
W
ILSON
, I. G. 1997. Inhibition and facilitation of nucleic acid amplifi-
cation. Applied and Environmental Microbiology 63: 3741–3751.
Y
ARNALL
, H. A., K. S. R
EECE
,N.A.S
TOKES
,
AND
E. M. B
URRESON
.
2000. A quantitative competitive polymerase chain reaction assay
for the oyster pathogen Perkinsus marinus. Journal of Parasitology
86: 827–837.
Z
INGALES
, B., B. S. S
TOLF
,R.P.S
OUTO
,O.F
ERNANDES
,
AND
M. R.
B
RIONES
. 1999. Epidemiology, biochemistry and evolution of Try-
panosoma cruzi lineages based on ribosomal RNA sequences. Me-
mo´rias do Instituto Oswaldo Cruz 94(Suppl. 1): 159–164.