BULLETIN OF MARINE SCIENCE. 87(3):421–435. 2011
http://dx.doi.org/10.5343/bms.2010.1093
The Relationship between Imposex and
Tributyltin (TBT) Concentration in Strombus
gigas from the British Virgin Islands
Cassander P Titley-O’Neal, Bruce A MacDonald,
Émilien Pelletier, Richard Saint-Louis, and Orville S Phillip
ABSTRACT
The queen conch, Strombus gigas (Linnaeus, 1758), was collected from sites
with substantial boating activity and adjacent reference sites from the British
Virgin Islands in 2008–2009 to determine the incidence and severity of imposex.
Comparisons were made to a previous study by Phillip (2000) in the same area to
determine if there were any changes in imposex over the last decade. Imposex was
present in S. gigas at all sites with boating activity, but not found in individuals
collected at reference sites. Tributyltin (TBT) and its degradative product, dibutyltin
(DBT), were found in S. gigas, turtle grass, Thalassia testidium (Banks ex König),
and marine algae from sites with known point sources of TBT. We established a
relationship between the highest incidence and severity of imposex and TBT
concentrations in the tissues of S. gigas from sites with elevated boating activity
(Road Harbour > Nanny Cay > Trellis Bay), suggesting that TBT, a known inducer of
imposex in female prosobranch gastropods, is the causative agent. Strombus gigas
is the second most important commercial fishery species in the Caribbean and
any factors that could affect reproduction and population ecology are important
in maintaining a viable population and managing a sustainable fishery. We
suggest using S. gigas as a sentinel of TBT pollution despite not meeting all of the
recommended criterion, such as lack of non-dispersive larvae.
The coastal environment of the Caribbean is comprised of mangroves, seagrass
meadows, and coral reefs. Even though each habitat can be viewed as a single entity, they are interconnected and the demise of one will affect the environment as a
whole. Environmental managers of these habitats are faced with the daunting task of
balancing development and conservation of these valuable natural resources, which
are impacted by both natural and anthropogenic influences. Some notable anthropogenic impacts on the Caribbean coast include eutrophication, coastline development
for tourism related activities such as hotels, marinas, villas, cruise ship piers, and the
dredging of harbors for marine traffic. Linton and Warner (2003) suggested the use
of biomarkers as a tool to give early warning signs of ecosystems under stress in the
Caribbean where natural resources are severely threatened and human resources are
limited. Two widely used biomarkers are imposex and intersex, which are known to
affect female gastropods. Imposex and intersex have been used globally as bioindicators of butyltin pollution, tributyltin (TBT) and triphenlyltin (TPT), thought to
be primarily released through their use in antifouling paint. In females affected by
imposex, the entire female genital system is conserved, but superimposed by male
organs, i.e., penis and/or vas deferens (Matthiessen et al. 1999), while in intersex
females, there is a modification of the female pallial organs that eventually supplants
the corresponding male formation (prostate gland, Matthiessen et al. 1999).
Bulletin of Marine Science
© 2011 Rosenstiel School of Marine and Atmospheric Science
of the University of Miami
421
422
BULLETIN OF MARINE SCIENCE. VOL 87, NO 3. 2011
The majority of published literature on imposex and intersex focuses on countries
from Europe and other temperate areas. When it comes to tropical environments,
there is a paucity of published imposex literature from countries in Latin America
(Central and South America and the Caribbean) when compared to Europe, Asia,
and North America (Titley-O’Neal 2011). This trend appears to be changing, as
there has been a surge in imposex studies from South America and the Caribbean
within the last decade, especially from countries bordering the Atlantic Ocean,
where the occurrence of imposex has been attributed to boating activity and TBT
input from antifouling paint. Imposex has been recently linked to TBT in Argentina
(Penchaszadeh et al. 2001, Goldberg et al. 2004, Biggatti and Penchaszadeh 2005,
Bigatti et al. 2009), Brazil (Caetano and Absãlo 2002, Camillo et al. 2004, Limaverde
et al. 2007), Chile (Osorio and Huaquín 2003, Huaquín et al. 2004), Colombia
(Hernández and Stotz 2004), Venezuela (Miloslavich et al. 2007), the British Virgin
Islands (BVI, Phillip 2000), and the United States Virgin Islands (USVI, Strand et al.
2009).
While a variety of gastropods are the focus of many imposex studies, Strombids
are rarely used. Strombids possessing both an egg groove and pseudopenis have
been observed in two species from the Pacific Ocean, Strombus luhuanus (Linnaeus,
1758) (Kuwamura et al. 1983) and Strombus carnarium (Schumacher, 1817) (Cob et
al. 2008), and two species from the Atlantic Ocean, Strombus gigas (Linnaeus, 1758)
(Reed 1993, Phillip 2000) and Strombus pugilis (Linnaeus, 1758) (Reed 1993). In the
present study, we refer to female Strombus spp. with an egg groove and pseudopenis
as imposex in agreement with Cob et al. (2008), and as described by Matthiessen et
al. (1999) where the female genital system is conserved, but superimposed by male
organs, a pseudopenis or vas deferens.
Of the four studies that reported imposex in strombid females globally, only the
study reported by Phillip (2000) performed chemical analysis. Butyltin analysis of
imposex specimens from areas adjacent to boating activity in the BVI recorded elevated TBT levels in affected females, and laboratory induction experiments further
confirmed imposex induction in S. gigas by TBT (Phillip 2000). Further, imposex
females from field sites were found to have the highest TBT levels in the gills > gonad
> digestive gland > mantle > foot, and TBT levels in the gills ranged from < 3 to 110
ng TBT Sn g−1 dry weight (Phillip 2000).
Strombus gigas is a tropical gastropod and member of the Strombidae family that
inhabits shallow seagrass beds (Randall 1964) within its geographic range encompassing the western Atlantic Ocean (Randall 1964, Reed 1993). All Strombus species
are gonochoristic (i.e., separate sexes) exhibiting sexual dimorphism where males
have a penis (termed “verge”) and females have an egg groove (Reed 1993). Sexual
maturity in S. gigas is attained when the shell lip is fully developed at 3–4 yrs of
age (Davis et al. 1984, Appeldoorn 1988), and S. gigas demonstrates seasonal reproduction which can vary throughout its geographic range (Brownwell 1977). During
copulation, the male extends his verge and deposits sperm within the female (Orr
and Berg 1987), where internal fertilization occurs (Davis and Shawl 2005). Fertilized eggs are laid on the seafloor between March and October (Davis 2000), but peak
months for egg laying are generally when water temperatures are ~28–30 °C, which
is usually between July and September (Davis et al. 1984, Weil and Laughlin 1984). In
an undisturbed population, S. gigas maintains a 1:1 sex ratio (Robertson 1959, Davis
et al. 1984).
titley-o’neal et al.: imposex and tbt in strombus gigas
423
Strombus gigas was the species of choice in our study because: (1) females with an
egg groove and pseudopenis have been previously reported within its geographic
range (Reed 1993, Phillip 2000); (2) S. gigas is a commercially lucrative species in the
Caribbean and ranked as the second most valuable fishery after the spiny lobster,
Panuliris argus (Latreille, 1804) (Brownell and Stevely 1981); (3) S. gigas is an endangered species that is on Appendix II of the Convention on International Trade in Endangered Species (CITES) list (Braütigam 1992); and (4) TBT antifouling paint is still
available for sale in the BVI, so the potential for TBT pollution by antifouling paints
exists. The objectives of the study were to: (1) assess imposex incidence and severity
in areas of boating activity and surrounding reference sites in the BVI; (2) correlate
imposex incidence and severity with TBT body concentration; (3) evaluate temporal
changes in imposex occurrence by comparing our 2008–2009 results with a 2000
survey at the same sites; (4) discuss the implications of using S. gigas as bioindicator
of TBT pollution; and (5) use marine flora to monitor TBT levels in the environment.
Material and Methods
Sampling.—The BVI are situated at 18°30´N and 64°30´W, roughly 100 km east of Puerto
Rico (Fig. 1), and are bordered by the Atlantic Ocean to the north and the Caribbean Sea to
the south. Field collection was completed on August 19–21, 2008, and August 19–28, 2009,
at sites previously surveyed by Phillip (2000) and O’Neal (2001). At each site, depth (m), associated flora and fauna, and type of boating activity were recorded. Boating activity was
designated as commercial harbor, cruise ship pier, ferry terminal, marina, mooring, navigational passage, and shipyard and/or drydock facilities (Table 1). Two sites were used solely
as navigational passages (Bluffy Bay and Guana Island), and they were classified as reference
sites because there were no stationary boats (e.g., docks, marinas, and/or moorings) in their
vicinity for any period of time, and there were no shipyards. Qualitatively, boating intensity
at all sites was ranked Road Harbour > Nanny Cay > Trellis Bay > Bluff Bay ≥ Guana Island >
Anegada based on the number of activities at each site and local knowledge of the study area
(CP Titley-O’Neal and OS Phillip, pers obs). A survey of maritime chandleries operating in
the BVI revealed that antifouling paint containing TBT was still available for retail primarily
from two manufacturers, SeaHawk® and Pettit®.
Adult S. gigas (n = 10–30) were collected from all sites using snorkel and/or SCUBA, and
kept alive en route to the Marine Centre at the H Lavity Stoutt Community College (HLSCC).
At each site, five turtle grass, Thalassia testidium (Banks ex König), or marine algae samples were collected, placed in labelled plastic bags, and stored on ice in ice boxes en route to
HLSCC. The species of marine algae collected included: Acanthophora spicifera (Lamouroux,
1813), Avrainvillea longicaulis (Murray and Boodle, 1889), Dictyota linearis (Lamouroux,
1809), and Ceramium spp. In 2008, T. testidium was initially collected because: (1) seagrasses
are sensitive to different types of pollution (Macauley et al. 1990); (2) high tin levels were
recorded at the same sites where imposex in S. gigas was found (O’Neal 2001); and S. gigas is
a herbivore known to feed on T. testidium, which could potentially act as a source for TBT
(Levine et al. 1990). Butyltin analysis of T. testidium samples in autumn 2008 revealed that
butyltin distribution was patchy within sites, so marine algae commonly associated with seagrass beds were collected in 2009.
Imposex Analysis.—Imposex analyses were performed at the HLSCC Marine Centre on
the same day as sample collection. For each specimen, shell length and penis length (mm)
were measured using digital stainless steel calipers. Sexually mature females were differentiated from males by presence of an egg groove (Reed 1993), and if sex could not be determined,
the animal was not used in the analysis. The degree of imposex in affected females was ranked
based on a classification system modified from Andersen (2004). The classification system
424
BULLETIN OF MARINE SCIENCE. VOL 87, NO 3. 2011
Figure 1. Map of the British Virgin Islands situated in the northern Caribbean depicting sites
sampled in the summers of 2008 and 2009 where ● delineates boating activity sites and ♦
reference sites. Site A = Nanny Cay, B = Road Harbour, C = Trellis Bay, D = Bluff Bay, E =
Guana Island, and F = Anegada.
consisted of four stages: normal female without a penis (Grade 0); penis length < 0.5 mm, i.e.,
area is raised, but unable to be measured (Grade 1); penis length 0.5 to < 5.0 mm (Grade 2);
and penis length ≥ 5.0 mm (Grade 3). Imposex incidence and severity was assessed at each site
using: mean male penis length (mMPL), mean female penis length (mFPL), percentage of females in each population with imposex (%I), male:female ratio (M:F), and relative penis length
index (RPLI: mFPL/mMPL × 100; Oehlmann et al. 1992). RPLI was used to compare imposex
intensity and severity between populations (Barroso and Mora 1998).
Butyltin Analysis.—All samples used for butyltin analysis were freeze dried for 48–72
hrs at −80 °C in a LABCONO® Freezezone® freeze drier. Freeze dried samples were kept at
−20 °C until they were analyzed for the presence of TBT, dibutyltin (DBT), and monobutyltin
(MBT) at the Institut des Sciences de la mer de Rimouski (ISMER) at the University of Québec, in Rimouski (Canada). For S. gigas samples, the gills and digestive gland of two animals
of each designation (male, female, and imposex female) were pooled since Phillip (2000) had
previously shown that they had the highest butyltin concentration. Determination of butyltins in tissue samples was adapted from Chau et al. (1997) with some modifications.
Approximately 0.25 g of freeze dried sample was digested in tetramethyl ammonium hydroxide (TMAH, 25% H2O). Samples were mixed on a vortex mixer, then soniccated for 1
hr at 60 °C in an ultrasonic bath. Certified reference material (CRM) BCR CRM 462, mussel
tissue (European Commission), and a procedural blank were treated in similar fashion. After
soniccation, samples were buffered (pH 4.5) with sodium acetate, NaCl solution, and NaCl
crystals. Butyltins were derivatized to an ethylated neutral compound by addition of sodium
tetraethylborate (NaBET4, 2% H2O). Extraction of ethylated butyltins was accomplished by
recovery of two separate, but successive additions of 1:1 hexane:toluene mixture.
One gram of Florisil® was added directly to the recovered organic layer (~6 ml), in a first
clean-up step, centrifuged (3000 rpm, 10 min, IEC HN-SII centrifuge), and the organic layer
Harbour
None
Boating activity
Year
2008
2009
Bluff Bay
NP
2008
2009
Guana Island NP
2008
2009
Nanny Cay
MA, SH
2008
2009
Road Harbour CH, CSP, FT, MA, MO, SH 2008
2009
Trellis Bay
MO
2008
2009
Site type
Site
Reference Anegada
nTotal
n/a
28
10
25
24
26
13
30
17
24
n/a
18
No. of ♀
n/a
15
4
17
15
16
6
13
7
4
n/a
8
No. of ♂
n/a
13
6
8
9
10
7
17
10
20
n/a
10
%I
n/a
0.0
0.0
0.0
0.0
0.0
50.0
46.2
28.6
100.0
n/a
50.0
mFPL (mm)
n/a
0.0
0.0
0.0
0.0
0.0
12.60 ± 13.60
7.07 ± 6.49
20.60 ± 7.71
9.75 ± 8.90
n/a
3.45 ± 1.38
mMPL (mm)
n/a
5.42 ± 8.26
1.97 ± 4.25
0.73 ± 0.35
13.80 ± 15.10
5.36 ± 13.99
18.90 ± 17.60
8.67 ± 5.83
15.86 ± 7.61
18.74 ± 12.36
n/a
22.80 ± 20.60
RPLI
n/a
0.0
0.0
0.0
0.0
0.0
66.8
81.6
129.6
52.0
n/a
15.1
Mean
imposex
grade
n/a
0.0
0.0
0.0
0.0
0.0
1.2
1.4
0.9
2.5
n/a
0.8
Table 1. Site type, boating activity, sample size (nTotal), imposex index data: % Imposex (%I), mean female penis length (mFPL) ± standard deviation (SD), mean
male penis length (mMPL) ± SD, relative penis length index (RPLI) and mean imposex grade for Strombus gigas collected in the British Virgin Islands during
August 2008 and 2009. Boating activity recorded at sites as: commercial harbor (CH); cruise ship pier (CSP); ferry terminal (FT); marina (MA); moorings (MO);
navigational passage (NP); and shipyard (SH). n/a = not applicable because not sampled.
titley-o’neal et al.: imposex and tbt in strombus gigas
425
426
BULLETIN OF MARINE SCIENCE. VOL 87, NO 3. 2011
recovered. A second clean-up step consisted of cleaning the second recovered organic layer
on a silica gel micro-column with anhydrous sodium sulfate. The organic fraction was eluted
with a 1:1 hexane:toluene mixture to a final volume of roughly 10 ml. Tetrapropyltin (100 μl)
was added as the internal standard, and the final volume was reduced to 200 μl under a gentle
stream of N2 gas at room temperature. The concentrated evaporates was analyzed by gas chromatography linked to an ion trap mass spectrometer (GC-MS, Trace GC-Polaris Q Thermo
Finnigan®) equipped with a Valcobond® VB-5 GC column using helium (He) as the carrier gas.
Butyltin identification was based on the respective mass spectrum and GC retention time.
Quantification was performed with reference to the internal standard by Xcalibur®, a software
program which manages the Polaris Q® Trace GC. Recovery of butyltin species was determined with successive extractions (n = 5) of standard CRM 462. Percent recovery for DBT
and TBT (± standard deviation) was 55% (± 2.9) and 72% (± 1.6), respectively. Due to a low
recovery yield, MBT was not analyzed in the present study. Butyltin concentrations were corrected for 100% recovery and are reported as tin (ng Sn g−1 dry weight). Quality assurance for
the analytical component was verified by: (1) calibration of GC, (2) use of CRM for every five
tissue samples, and (3) use of procedural blanks for every five samples. The limit of detection
expressed as tin (Sn) atom, with one microliter injected, was 0.1 ng Sn g−1 for DBT and TBT.
Statistical Analysis.—All statistical analyses were performed using SPSS version 17.
Where the assumptions of parametric testing were violated, the non-parametric equivalent
was used (α = 0.05). Wilcoxon sign ranks test was used to assess temporal and spatial changes
in established imposex indices by year and site. Spearman Rank correlation was used to correlate imposex incidence (%I, mFPL, and RPLI) with log10 TBT concentration in all females (normal and imposex) from polluted and reference sites. For statistical purposes, tissue samples
where butyltins were not detected were assigned a value of 0.10 ng Sn g−1, which was the limit
of detection.
Results
Imposex Incidence and Severity
Present Study.—Imposex was observed in S. gigas collected from all boating activity sites, but not found at any of the reference sites (Table 1). Percent imposex (%I) at
boating sites ranged from 28.6% to 100% for 2008–2009, with the lowest and highest
incidence of imposex observed at the same site, Road Harbour. At boating activity
sites, mFPL and RPLI ranged from 3.45 mm ± 1.38 to 20.6 mm ± 7.71 and 15.1 to
129.6, respectively (Table 1), and the highest incidence of each occurred at Road Harbour. In several females with imposex, the penis was often bi-lobed and split at its extremity. Mean imposex grade for S. gigas in this our ranged from 0 to 2.5 for all sites
surveyed (Table 1). All reference sites had a mean score of 0. Of the boating activity
sites, the highest mean imposex score (2.5) was observed at Road Harbour while
the lowest mean imposex score (0.80) was at Trellis Bay. A comparison of imposex
severity between sites for 2009, when data were collected from all six sites, showed a
significant difference between all sites for %I (z = −3.734, n = 24, P < 0.001), mMPL
(Wilcoxon Sign Ranks test: z = -4.256, n = 24, P < 0.001), RPLI (Wilcoxon Sign Ranks
test: z = −3.734, n = 24, P < 0.001), and M:F (Wilcoxon Sign Ranks test: z = −2.287, n
= 24, P < 0.05) . No significant among-site difference was found for mFPL (Wilcoxon
Sign Ranks test: z = −0.889, n = 24, P > 0.05).
The highest mean TBT (30.0 ng TBT Sn g−1 dry weight) and mean DBT (295.0 ng
DBT Sn g−1 dry weight) concentrations in any sex were observed in males from Road
Harbour (Table 2). Anegada was the only site of any type (reference or boating activity) where no butyltins were found in any sex analyzed (Table 2). However, butyltins
titley-o’neal et al.: imposex and tbt in strombus gigas
427
Table 2. Site type and mean butyltin tissue concentration (n = 2, except for Anegada and Trellis
Bay) for normal female, imposex female, and male Strombus gigas collected from the British
Virgin Islands during August 2008 and 2009. Tissue concentration reported as ng Sn g−1 dry
weight. < DL = below detection limit, n/a = not applicable.
Site Type Site
Reference Anegada
Harbour
Sex
Female (♀)
Male (♂)
Bluff Bay
Female (♀)
Male (♂)
Guana Island Female (♀)
Male (♂)
Nanny Cay
Female (♀)
Imposex Female (♀♂)
Male (♂)
Road Harbour Female (♀)
Imposex Female (♀♂)
Male (♂)
Trellis Bay
Female (♀)
Imposex Female (♀♂)
Male (♂)
2008
TBT DBT
n/a
n/a
n/a
n/a
< DL < DL
< DL
3.1
< DL < DL
< DL
6.8
< DL
7.1
0.6
12.0
< DL
2.2
< DL
0.3
10.0
47.0
25.0 286.0
n/a
n/a
n/a
n/a
n/a
n/a
2009
TBT DBT
< DL < DL
< DL < DL
< DL
1.4
< DL 10
< DL < DL
18.0
10.0
< DL
7.7
0.9
13.0
< DL
3.9
< DL
0.7
16.0
87.0
34.0 304.0
< DL < DL
5.6
63.0
< DL
3.9
Mean
TBT DBT
< DL < DL
< DL < DL
< DL
0.7
< DL
6.6
< DL < DL
9.0
8.4
< DL
7.4
0.8
13.0
< DL
3.1
< DL
0.5
13.0
67.0
30.0 295.0
< DL < DL
5.6
63.0
< DL
3.9
were found in males from the remaining reference sites, Bluff Bay and Guana Island,
in 2008 and 2009, and females from Bluff Bay in 2009 (Table 2), although no imposex
was observed (Table 1). TBT tissue concentrations in imposex females from boating
activity sites were quantitatively ranked Road Harbour > Trellis Bay > Nanny Cay
with mean TBT tissue concentrations of 13.0, 5.6, and 0.8 ng TBT Sn g−1 dry weight,
respectively (Table 2). Spearman rho statistical analysis confirmed a significant onetailed correlation between mFPL (rs = 0.932, n = 59, P < 0.01), and %I (rs = 0.300, n =
59, P < 0.05) and RPLI (rs = 0.285, n = 59, P < 0.05), to a lesser degree.
Turtle grass (n = 20) and marine algae (n = 30) collected in 2008 and 2009, respectively, were used to determine environment levels of butyltins in the BVI. Of
the 20 T. testidium samples collected in 2008, TBT was the only butyltin detected
in a single sample (5%) collected from Road Harbour (36 ng TBT Sn g−1 dry weight).
The remaining seagrass samples were below the detection limit, suggesting a patchy
to non-existent distribution of butyltins for 2008 in T. testidium. Butyltin analysis
of marine algae samples in 2009 yielded similar results as T. testidium in 2008. No
butyltins were detected at any of the reference sites, but they were detected in four
samples (13.3%) collected from boating activity sites. TBT was detected in two samples from Road Harbour (13 and 18 ng TBT Sn g−1 dry weight), and one sample from
Trellis Bay (7.4 ng TBT Sn g−1 dry weight). No TBT was detected at Nanny Cay, but
DBT, one of TBT’s degradative products, was detected in one sample (4.3 ng DBT Sn
g−1 dry weight).
Comparison with 2000 Imposex Survey.—The first imposex study reported in the
BVI was completed by Phillip (2000) using the same species and sites used in our study,
with the exception of Anegada. The highest incidence of imposex (%I) occurred in
Road Harbour in 2000 and 2009, while the highest recorded mFPL occurred in 2008
and at Road Harbour (Fig. 2). Recalculated RPLI values for 2000 were 9.6, 26.7, and
428
BULLETIN OF MARINE SCIENCE. VOL 87, NO 3. 2011
Figure 2. (A) Percent imposex (%I) and (B) mean female penis length (mFPL) ± SD for Strombus
gigas collected in the British Virgin Islands in 2000 (Phillip 2000) and 2008–2009 (present
study). * = significant differences (P < 0.05) among years.
6.1 for Nanny Cay, Road Harbour, and Trellis Bay, respectively. In 2008, RPLI values
for Nanny Cay (RPLI = 66.8) and Road Harbour (RPLI = 129.6) were calculated, but
no values for Trellis Bay could be determined as no samples were collected during
that time (Table 2). RPLI values for 2009 for boating activity sites were 81.6, 52.0, and
15.1 for Nanny Cay, Road Harbour, and Trellis Bay, correspondingly.
Between 2000 and 2009, %I at Nanny Cay decreased (Wilcoxon Sign Ranks test: z
= −2.232, n = 6, P < 0.05) over time, while at Trellis Bay %I increased, but not significantly (Wilcoxon Sign Ranks test: z = −1.890, n = 5, P > 0.05; Fig. 2). At Road Harbour,
%I differed significantly over time, being significantly lower in 2008 than in 2000
and 2009 (Wilcoxon Sign Ranks test: z = −2.232, n = 6, P < 0.05; Fig. 2). Mean FPL
at Nanny Cay and Road Harbour was significantly higher in 2008 than in either of
the other years (Wilcoxon Sign Ranks test: z = −2.232, n = 6, P < 0.05), while differ-
titley-o’neal et al.: imposex and tbt in strombus gigas
429
ences at Trellis Bay among years were not significant (Wilcoxon Sign Ranks test: z =
−1.890, n = 5, P > 0.05). Data from the present study, as well as from Phillip (2000),
suggest that sites should be ranked Road Harbour > Nanny Cay > Trellis Bay for %I
and mFPL (Fig. 2).
Discussion
To the best of our knowledge, this is the third imposex study reported from the
Caribbean. Phillip (2000) used S. gigas collected from BVI field sites in 2000. Strand
et al. (2009) investigated four harbor sites and seven coastal reference sites in the
USVI based on data collected in 2007. Strand et al. (2009) observed imposex in
Purpura patula (Linnaeus, 1758), Thais deltoidea (Lamarck, 1822), and Thais rustica
(Lamarck, 1822). In comparing our BVI study with that of the neighbouring USVI,
the highest incidence of imposex (%I), the highest reported tissue concentrations,
and the highest severity of imposex [RPLI or vas deferens sequence index (VDSI)]
were all recorded at major ports of entry. However, in the Strand et al. (2009) study,
imposex was recorded in six of the seven reference sites in the USVI, whereas we did
not find imposex in any of the reference sites in the BVI.
In the present study, the mean imposex grade for all reference sites was 0, indicative of a site with normal females. However, at boating activity sites, imposex
grade ranged from 0.8 to 2.5, reflecting the presence of females with some degree of
imposex. The prevalence of imposex at all boating activity sites, and the absence of
imposex at all reference sites suggests that TBT input via antifouling paint is potentially responsible. This is consistent with chemical analysis of all females (normal and
imposexed), with tissue concentrations correlating with mFPL, and to a lesser degree
with RPLI and %I.
In the present study, TBT and DBT were detected in T. testidium and marine algae,
but only rarely at concentrations above the detection limit. Hence, our findings show
that butyltin distribution in marine flora appears to be very “patchy,” and not likely
to be a significant source of TBT for S. gigas. TBT-induced imposex in S. gigas in our
field study is most likely due to exposure from TBT in the water column, although
this has not been quantified. Further, the prevalence of DBT over TBT in the tissues
of S. gigas may reflect a low biotransformation rate. The presence of TBT and DBT
in sessile marine flora suggests that the release of butyltins into the marine environment is suspected point sources related to boating activity.
Some of the imposex-affected S. gigas females had a bi-lobed penis at their extremity and appeared deformed. Females with lobed penises were not unique to our
study as bi- and tri-lobed penises have been observed by others, e.g., S. gigas (Reed
1993, 1994), Hexaplex trunculus (Linnaeus, 1758) (Terlizzi et al. 1999, Lahbib et al.
2008, Abidli et al. 2009), and Bolinus brandaris (Linnaeus, 1758) (Abidli et al. 2009).
Strombus gigas is a lucrative species within its geographic range, so a more in depth
knowledge of factors that could affect its reproduction and population ecology is
important for understanding its population dynamics and managing a viable fishery.
Since imposex females have female and male sexual organs (egg groove and pseudopenis), it would be important to know if the presence of a pseudopenis affects the
female’s ability to reproduce; however, we did not examine this here. Histological and
hatchability data from other studies suggest that imposex in S. gigas does not impair
430
BULLETIN OF MARINE SCIENCE. VOL 87, NO 3. 2011
its ability to successfully reproduce (Reed 1993), but it would be interesting to know
if TBT affects S. gigas later on in its life cycle or on a molecular level.
The presence of butyltins in S. gigas from two of the reference sites (Guana Island
and Bluff Bay) was unexpected, but considering that those two sites are used as navigational passages, the release of TBT from passing vessels is a likely source. There
are not enough data to conclude why TBT and DBT were present predominantly
in males from these two reference sites, but it may be possible that strombid males
are more susceptible to bioaccumulate TBT than their female counterparts. When
sites are ranked based on the level of TBT tissue concentration of all samples (conch,
seagrass, and marine algae), %I, mFPL, and RPLI from highest to lowest, rankings for
each parameter in our study always followed the order: Road Harbour > Trellis Bay >
Nanny Cay, consistent with decreasing levels of boating activity.
When determining if a species should be used as a model bioindicator for TBT pollution, Gibbs and Bryan (1987) suggested the following criteria: (1) wide geographic
range; (2) conspicuous and easily identifiable; (3) limited potential for dispersal; (4)
durable egg capsules as a convenient marker of breeding activity; (5) hardy species
suitable for experimentation; and (6) highly sensitive to TBT. Strombus gigas has
a wide geographic range encompassing the western Atlantic Ocean (Randall 1964,
Reed 1993), and it is easily identifiable from the other five strombids in its home
range. Egg capsules, a sign of breeding activity, are mixed with sand in a crescent
shape (Davis and Shawl 2005), and these were commonly seen during the time of
sample collection. Further, laboratory induction studies by Phillip (2000) have demonstrated the sensitivity of S. gigas to TBT, as well as its hardiness for experimentation. A review of the recommended criteria needed for a model bioindicator indicates
that the use of S. gigas as a bioindicator would comply with all criteria with the exception of a limited potential for dispersal. This is due in part to the reproductive
cycle of S. gigas where offspring hatch from an egg mass into planktotrophic veligers
drifting in the water column for ~3 wks (Davis and Shawl 2005). On the other hand,
the dogwhelk, Nucella lapillus (Linnaeus, 1758), which is arguably the best documented species used in imposex studies, meets all the criteria as suggested by Gibbs
and Bryan (1987).
We recommend the use of S. gigas as sentinel of TBT pollution for three reasons.
Firstly, a positive relationship between tissue concentration and two established imposex indices (mFPL and RPLI) has been established for S. gigas in the present study,
and this relationship has already been confirmed for other well documented female
gastropods affected by imposex (Oehlmann et al. 1992, Solé et al. 1998, Smith et al.
2006) and intersex (Bauer et al. 1995). Secondly, a model species used in intersex
studies already exists that violates the criterion of having a limited dispersal. The
periwinkle, Littorina littorea (Linnaeus, 1758), has a pelagic larval veliger stage (Reid
1996), and has been reported as having intersex as a consequence of TBT pollution in
the United Kingdom (Matthiessen et al. 1995), France (Fioroni et al. 1991), Germany
(Oehlmann et al. 1994), and eastern Canada (Covert 1997). Lastly, the Oslo/Paris
Convention (OSPAR) for the protection of the marine environment in the NorthEast Atlantic recommends the use of L. littorea as one of their three model species
for TBT pollution studies (Davies et al. 1997).
Even though S. gigas does not meet all the recommended criteria for a model indicator of TBT pollution as suggested by Gibbs and Bryan (1987), we recommend its use
as a sentinel for TBT pollution within its geographic range when taking the following
titley-o’neal et al.: imposex and tbt in strombus gigas
431
into consideration: (1) the potential limitation of acquiring an adequate sample size
and (2) S. gigas is an endangered species on appendix II of CITES (Braütigam 1992).
At the majority of sites in our study, the sample size was < 30, whereas the literature
recommends an adequate sample size of 30 (Stroben et al. 1995, Schulte-Oehlmann
et al. 1998, Bech 2002, Zhengyan 2005) or 40 (OSPAR 2008), ideally with an equal
proportion of animals to adequately assess spatial and temporal trends in imposex
studies. For our study, every effort was made to attain a minimum sample size of 30,
but obtaining animals with a flared shell was not always possible, and if the sex of the
animal could not be clearly determined it was discarded from imposex analysis. The
issue of limited sample size to properly assess spatial and temporal trends was quite
evident when assessing changes in %I and mFPL from 2000 to 2009. For example, in
2009 at Road Harbour, data from 24 animals were obtained, but there was a disproportionate number of males (n = 20) to females (n = 4), and all of the sampled females
were affected by imposex.
With the limited sample size used throughout our study, no conclusion can be
drawn on spatial or temporal trends. Laboratory studies in the BVI have confirmed
TBT-induced imposex in S.gigas (Phillip 2000), and given the fact that imposex is
irreversible (Gibbs and Bryan 1996, Oehlmann 1998, Foale 1993, Matthiessen et al.
1999), it was not surprising to find a significant correlation between mFPL and log10
TBT concentration in our study. The confirmation of TBT-induced imposex in S.
gigas and the positive correlation between mFPL and log10 TBT concentration provide supporting evidence for S. gigas as a sentinel of TBT pollution in the BVI and
throughout its geographic range.
In summary, imposex was observed in female queen conch, S. gigas, collected in
the BVI during the summers of 2008–2009 from areas adjacent to boating activity, but imposex was absent from adjacent non-impacted reference sites. Data suggest that TBT input from its use in antifouling paint could be the main cause; this
is consistent with the presence of TBT and DBT in the tissues of conch, seagrass,
and marine algae. In addition, a positive correlation between mFPL, and to a lesser
degree, RPLI, with TBT concentration in females is further evidence for TBT being
the causative agent of imposex in S. gigas. In our study, the incidence and severity
of imposex was the highest in Road Harbour, which also has the highest diversity of
boating activity, and similar results were observed at the same location 9 yrs earlier
(Phillip 2000).
While there is no clear pattern for %I or mFPL for all sites when imposex severity
is measured by RPLI, the data suggest an increase over time. While S. gigas meets all
the requirements proposed for a model bioindicator species of TBT pollution except
for one, a limited potential for dispersal, we recommend its use as a sentinel of TBT
pollution within its geographic range.
Acknowledgments
This work was funded by NSERC (Natural Sciences and Engineering Research Council)
of Canada (B MacDonald), the Canadian Research Chair in Molecular Ecotoxicology (É
Pelletier), CBIE (Canadian Bureau for International Education), H Lavity Stoutt Community
College, and the Government of the Virgin Islands via the office of the Premier, RT O’Neal.
We thank R Frazer, G Brooks, and L Lewis for their assistance with field work.
432
BULLETIN OF MARINE SCIENCE. VOL 87, NO 3. 2011
Literature Cited
Abidli S, Lahbib Y, El Menif NT. 2009. Imposex and genital tract malformation in Hexaplex
trunculus and Bolinus brandaris collected in the Gulf of Tunis. Bull Mar Sci. 85:11–25.
Andersen LE. 2004. Imposex: a biological effect of TBT contamination in Port Curtis,
Queensland. Aus J Ecotox. 10:105–113.
Appeldoorn RS. 1988. Age determination, growth, mortality, and age of first reproduction in
adult queen conch, Strombus gigas, off Puerto Rico. Fish Res. 6:363–378. http://dx.doi.
org/10.1016/0165-7836(88)90005-7
Barroso CM, Mora MH. 1998. Reproductive cycle of Nassarius reticulatus in the Ria De Aveiro,
Portugal: implications for imposex studies. J Mar Biol Assoc UK. 78:1233–1246. http://
dx.doi.org/10.1017/S0025315400044453
Bauer B, Fioroni P, Ide I, Liebe S, Oehlmannn J, Stroben E, Watermann B. 1995. TBT effects on
the female genital system of Littorina littorea: a possible indicator of tributyltin pollution.
Hydrobiologia. 309:15–27. http://dx.doi.org/10.1007/BF00014468
Bech M. 2002. A survey of imposex in Muricids from 1996 to 2000 and identification of optimal
indicators of tributyltin contamination along the east coast of Phuket Island, Thailand. Mar
Pollut Bull. 44:887–896. http://dx.doi.org/10.1016/S0025-326X(02)00115-7
Biggatti G, Penchaszadeh PE. 2005. Imposex in Odontyocymbola magellancia (Caenogastropoda: Volutide) in Patagonia. Comunicaciones de la Sociedad Malacologia del Uruguay.
9(88):371–375.
Biggatti G, Primost MA, Cledón M, Averbuj A, Theobald N, Gerwinski W, Arntz W, Morriconi
E, Penchaszadeh PE. 2009. Biomonitoring of TBT contamination and imposex incidence
along 4700 km of Argentinean shoreline (SW Atlantic: from 38S to 54S). Mar Pollut Bull.
58:695–701. PMid:19211114. http://dx.doi.org/10.1016/j.marpolbul.2009.01.001
Braütigam A. 1992. CITES: a conservation tool. 3rd ed. Center for Marine Conservation,
Washington, DC.
Brownell WN. 1977. Reproduction, laboratory culture, and growth of Strombus gigas, S. costatus, and S. pugilus in Los Roques, Venezuela. Bull Mar Sci. 2:668–680.
Brownell WN, Stevely J. 1981. The biology, fisheries and management of the queen conch,
Strombus gigas. US National Marine Fisheries Services, Marine Fisheries Review. 43(7):1–
12.
Caetano CHS, Absalão RS. 2002. Imposex in Olivancillaria vsica vesica (Gmelin) Gastropoda,
Olividae from a southeastern Brazilian sandy beach. Rev Bras Zool. 19:215–218. http://
dx.doi.org/10.1590/S0101-81752002000600022
Camillo E, Quadros J, de Castro IB, Fernandez M. 2004. Imposex in Thais rustica (Mollusca:
Neogastropoda) (Lamark, 1822) as an indicator of organotin compounds pollution at Maceio coast (north-eastern Brazil). Braz J Oceanogr. 52:101–105. http://dx.doi.org/10.1590/
S1679-87592004000200002
Chau YK, Yang F, Brown M. 1997. Evaluation of derivatization techniques for the analysis
of organotin compounds in biological tissue. Anal Chim Acta. 338:51–55. http://dx.doi.
org/10.1016/S0003-2670(96)00427-8
Cob ZC, Arshad A, Idris MH, Bujang JS, Ghaffar MA. 2008. Sexual polymorphism in a population of Strombus canarium Linnaeus, 1758 (Mollusca: Gastropoda) at Merambong Shoal,
Malaysia. Zool Stud. 47:318–325.
Covert TJ. 1997. Masculinization of dogwhelks (Nucella lapillus) and periwinkles (Littorina
littorea) in Halifax Harbour: monitoring of Tributyltin. BSc honors thesis. Saint Mary’s
University, Nova Scotia, Canada.
Davies IM, Minchin A, Harding A. 1997. OSPAR working group on concentrations, trends
and effects of substances in the marine environment (SIME) – report on the TBT training
workshop (September 24–26, 1997). Aberdeen Marine Laboratory Report, No. 9/07. 36 p.
titley-o’neal et al.: imposex and tbt in strombus gigas
433
Davis M. 2000. Queen conch (Strombus gigas) culture techniques for research, stock enhancement and growout markets. In: Fingermna M, Nagahushanam R, editors. Marine Biotechnology. Science Publishers, Inc., USA. p. 127–159.
Davis M, Mitchell BA, Brown JL. 1984. Breeding behavior of the queen conch Strombus gigas
Linne held in a natural enclosed habitat. J Shellfish Res. 4:17–21.
Davis M, Shawl AL. 2005. A guide for culturing queen conch, Strombus gigas. Manual of fish
culture. Am Fish Soc Symp. 46:125–142.
Fioroni P, Oehlmannn J, Stroben E. 1991. The pseudohermaphroditism of prosobranchs: morphological aspects. Zool Anz. 226:1–26.
Foale S. 1993. An evaluation of the potential of gastropod imposex as a bioindicator of tributyltin pollution in Port Phillip Bay, Victoria. Mar Pollut Bull. 26:546–552. http://dx.doi.
org/10.1016/0025-326X(93)90404-8
Gibbs PE, Bryan GW. 1987. TBT paints and the demise of the dog-whelk, Nucella lapillus
(Gastropoda). In: Proceedings of oceans ’87: the ocean—an international workplace, 28
September–1 October, 1987, Halifax, Nova Scotia. p. 1482–1487.
Gibbs PE, Bryan GW. 1996. Reproductive failure in the gastropod Nucella lapillus associated
with imposex caused by tributyltin pollution: a review. In: Champ MA, Seligman PF, editors. Organotin. Chapman and Hall, London. p. 259–280.
Golberg RN, Averbuj A, Cledó M, Luzzatto D, Sbarbati N. 2004. Search for triorganotins along
the Mar del Plata (Argentina) marine coast: finding of tributyltin in egg capsules of a snail
Adelomelon brasiliana (Lamarck, 1822) population showing imposex effects. Appl Organomet Chem. 18:117–123. http://dx.doi.org/10.1002/aoc.590
Hernández S, Stotz WB. 2004. Reproductive biology of the “Copey” snail Melongena melongena
(Linnaaeus, 1758) in Cispatá Bay on the Caribbean coast of Colombia. J Shellfish Res.
23:849–854.
Huaquín LG, Osorio C, Verdugo R, Collado G. 2004. Morphological changes in the reproductive system of females Acanthina monodon (Pallas, 1774) (Gastropoda: Muricidae) affected
by imposex from the coast of central Chile. Invert Reprod Dev. 46:111–117. http://dx.doi.
org/10.1080/07924259.2004.9652614
Kuwamura T, Fukao R, Nishida M, Wada K, Yanagisawa Y. 1983. Reproductive biology of gastropod Strombus luhuanus (Strombidae). Publ Seto Mar Biol Lab. 28:433–443.
Lahbib Y, Abidli S, El Menif NT. 2008. Imposex and penis malformation in Hexaplex trunculus
from the Tunisian coast. Am Malacol Bull. 24:79–89. http://dx.doi.org/10.4003/0740-278324.1.79
Levine SN, Rudnick DT, Kelly JR, Morton RD, Buttle LA. 1990. Pollutant dynamics as influenced by seagrass beds: experiments with tributyltin in Thalassia microcosms. Mar Environ Res. 30:297–322. http://dx.doi.org/10.1016/0141-1136(90)90004-8
Limaverde AM, Wagener A, Fernandez MA, Scofield ADL, Coutinho R. 2007. Stramonita haemastoma as a bioindicators for organotin contamination in coastal environments. Mar Environ Res. 64:384–398. PMid:17459467. http://dx.doi.org/10.1016/j.marenvres.2007.03.001
Linton DM, Warner GF. 2003. Biological indicators in the Caribbean coastal zone and their
role in integrated coastal management. Ocean Coast Manage. 46:261–276. http://dx.doi.
org/10.1016/S0964-5691(03)00007-3
Macauley JM, Clark JR, Pitts AR. 1990. Use of Thalassia and its epiphytes for toxicity assessment: effects of a drilling fluid and tributyltin. In: Wang W, Gorsuch JW, Lower WR, editors. Plants for toxicity assessment. American Society for Testing and Materials, Philadelphia. p. 255–266. http://dx.doi.org/10.1520/STP19068S
Matthiessen P, Reynoldson T, Billinghurst Z, Brassard DW, Cameron P, Chandler GT, Davies
IM, Horiguchi T, Mount DR, Oehlmannn J, et al. 1999. In: deFur PL, Crane M, Ingersol
CG, Tattersfield LJ, editors. Endocrine disruption in invertebrates: endocrinology, testing,
and assessment. Society of Environmental Toxicology and Chemistry (SETAC). p. 199–279.
Matthiessen P, Waldock R, Thain JE, Waite ME, Scrope-Howe S. 1995. Changes in periwinkle
(Littorina littorea) populations following the ban on TBT-based antifoulings on small boats
434
BULLETIN OF MARINE SCIENCE. VOL 87, NO 3. 2011
in the United Kingdom. Ecotox Environ Safe. 30:180–194. PMid:7539373. http://dx.doi.
org/10.1006/eesa.1995.1023
Miloslavich P, Penchaszadeh PE, Bigatti G. 2007. Imposex in Neogastropods from Venezuela.
Cienc Mar. 33:319–324.
Oehlmann J. 1998. Untersuchungen zum Einsatz von Pathomorphosen der ableitenden Geschlechtswege von Vorderkiemerschnecken (Gastropod a: Prosobranchia) für ein biologisches TBT-Effektmonitoring. Habilitation thesis. International Graduate School Zittau,
Zittau, Germany.
Oehlmann J, Liebe S, Watermann B, Stroben E, Fioroni P, Deutsch U. 1994. New perspectives
on sensitivity of littorinids to TBT pollution. Cah Biol Mar. 35:254–255.
Oehlmann J, Stroben E, Fioroni P. 1992. The rough tingle Ocenebra erinacea (Neogastropoda:
Muricidae): an exhibitor of imposex in comparison to Nucella lapillus. Helgoland Mar Res.
46:311–328.
O’Neal CP. 2001. Assessment of environmental levels of organotins in sediments, seawater,
and the seagrass, Thalassia testudinum, in coastal waters of the British Virgin Islands. MSc
thesis, University of Plymouth, Plymouth, United Kingdom.
Orr KS, Berg Jr. CJ. 1987. The queen conch. Winward Publishing, Inc. Miami, Florida.
Osorio CR, Huaquín LGM. 2003. Alteración de la sexualidad de Acanthina monodon (Pallas,
1774) (Gastropoda: Muricide) en el litoral de Chile Central, inducica pr compuestos organoestañosos. Cienc Technol Mar. 26(2):97–107. In Spanish with English abstract.
OSPAR. 2008. JAMP Guidelines for Contaminant-Specific Biological Effects (OSPAR Agreement 2008-09). In: OSPAR Convention for the Protection of the Marine Environment of
the North-East Atlantic.
Penchaszadeh PE, Averbuj A, Cledón M. 2001. Imposex in gastropods from Argentina
(south-western Atlantic). Mar Pollut Bull. 42:790–791. http://dx.doi.org/10.1016/S0025326X(01)00098-4
Phillip OS. 2000. The relationship between masculinized female conch (Strombus gigas) and
tributyltin-based antifouling paints in coastal waters of the British Virgin Islands. MSc thesis, University of Plymouth, Plymouth, United Kingdom.
Randall JE. 1964. Contributions to the biology of the “queen conch” Strombus gigas. Bull Mar
Sci. 14:246–295.
Reed SE. 1993. Gonadal comparison of masculinized females and androgynous males to normal
males and females in Strombus (Mesogastropoda: Strombidae). J Shellfish Res. 12:71–75.
Reed SE. 1994. Masculinized females in the genus Strombus: aspects of their biology and possible advantages for mariculture of conch. In: Appeldoorn RS, Rodríguez B, editors. Queen
conch biology, fisheries, and mariculture. Fundación Científica Los Roques, Caracas (Venezuela). p. 213–221.
Reid DG. 1996. Systematics and evolution of Littorina. The Ray Society, London.
Robertson R. 1959. Observations on the spawn and veligers of conch (Strombus) in The Bahamas. Proc Malacol Soc Lond. 33:164–172.
Schulte-Oehlmann U, Oehlmann J, Bauer B, Fioroni P, Leffler US. 1998. Toxico-kinetic and
dynamic aspects of TBT-induced imposex in Hydrobia ulvae compared with intersex in
Littorina littorea (Gastropoda, Prosobranchia). Hydrobiologia. 378:215–225. http://dx.doi.
org/10.1023/A:1003270528688
Smith AJ, Thain JE, Bary J. 2006. Exploring the use of caged Nucella lapillus to monitor changes
to TBT hotspot areas: a trial in the River Tyne estuary (UK). Mar Environ Res. 62:149–163.
PMid:16690115. http://dx.doi.org/10.1016/j.marenvres.2006.03.008
Solé M, Morcillo Y, Porte C. 1998. Imposex in the commercial snail Bolinus brandaris in the
northwestern Mediterranean. Environ Pollut. 99:241–246. http://dx.doi.org/10.1016/
S0269-7491(97)00186-3
Strand J, Jørgensen A, Tairova Z. 2009. TBT pollution in molluscs at US Virgin Islands, Caribbean Sea. Environ Int. 35:707–711. PMid:19201027. http://dx.doi.org/10.1016/j.envint.2009.01.007
titley-o’neAl et Al.: iMpoSex And tbt in StroMBuS gigaS
435
stroben e, schulte-oehlmann u, fioroni p, oehlmann J. 1995. a comparative method for easy
assessment of coastal TbT pollution by the degree of imposex in prosobranch species. haliotis. 24:1–12.
Terlizzi a, geraci s, gibbs pe. 1999. Tributyltin (TbT)-induced imposex in the neogastropod
Hexaplex trunculus in italian coastal waters: morphological aspects and ecological implications. ital J Zool. 66:141–146. http://dx.doi.org/10.1080/11250009909356248
Titley-o’neal Cp. 2011. across latitudes: using molluscs as biomonitors to assess spatial and
temporal distribution of tributyltin (TbT) in eastern Canada and the british Virgin islands.
phd thesis, university of new brunswick, saint John, Canada.
weil em, laughlin Rg. 1984. biology, population dynamics, and reproduction of the queen
conch, Strombus gigas linné, in the archipelago de los Roques national park. J shellfish
Res. 4:45–62.
Zhengyan li. 2005. seasonal variation in imposex intensity of Thais clavigera. Chin J oceanol
limnol. 23(2):163–168. http://dx.doi.org/10.1007/bf02894233
date suBmitted: 2 november, 2010.
date accepted: 30 march, 2011.
availaBle online: 18 april, 2011.
addresses: (CpT-o, bam) Department of biology, university of New brunswick, Saint John,
New brunswick, Canada, E2L 4L5. (ep, Rs-l) institut des sciences de la mer de rimouski
(iSmEr), université du Québec à rimouski, Québec, Canada, g5L 3a1. (CpT-o, osp) Natural
Sciences Department, H. Lavity Stoutt Community College, Paraquita bay, tortola, british
Virgin islands, Vg1110. Corresponding author: (CpT-o) E-mail: <[email protected]>.