Assessment of tributyltin contamination based on imposex in Stramonita rustica ( Mollusca : Gastropoda ) along southern Bahia coast , northeastern Brazil

This study investigated the effects of tributyltin (TBT) on the morphology of the genital system of the gastropod Stramonita rustica in southern Bahia, Brazil. For this, 330 specimens were collected during the summer of 2014 at eight sampling points to ascertain whether male sex organs had developed in addition to the complete female genital tract in females (= imposex). The analyses were made under a stereoscopic microscope. Imposex and their associated indexes, and the sterile females, exhibited the highest rates in harbors and shipyards areas. Despite the total ban of TBT in anti-fouling paints on a global scale since 2003, the results of this and other studies indicate the continued use of those paints on the Brazilian coast. This shows the inefficiency of existing legislation and the need to strengthen enforcement of the ban.


Introduction
The organic compound Tributyltin (TBT), used as biocides in antifouling paints since the mid 1960s was considered by Goldberg (1986) to present the highest toxicity among compounds and to be the most harmful anthropogenic contaminant ever produced and deliberately introduced into aquatic ecosystems.Because of the adverse effects of TBT on marine organisms, the International Marine Organization (IMO) imposed a worldwide ban on its use in marine applications in 2003, to effectively force in 2008(Bekri et al., 2006)).In Brazil, the Brazilian Navy (BN) suspended the use of TBT -bases antifouling paints on its vessels in 2003 too.But, small boats continued using antifouling paints with TBT.Because of this, in 2007 the NORMAM 23 (BN) restricted the use of organotin compounds in antifouling products on all Brazilian registered vessels.Despite the prohibitions in national and global level, the TBTs are still being used in many countries, including Brazil (Santos et al., 2011).
TBT can cause negative effects in various taxonomic groups and, among these, mollusks appear to be the invertebrates that are the most sensitive to this compound (Bryan and Gibbs, 1991).Consequently, these organisms are excellent sentinels for monitoring TBT in marine environments (Shim et al., 2002).For example, TBT is the leading cause of defects in bivalve shells (Bryan and Gibbs, 1991) and also acts as an endocrine disrupter in gastropods (Smith, 1971).
Imposex occurs in neogastropods, and is characterized by development of male sex organs (penis and/or vas deferens) in addition to a complete female genital tract (Smith, 1971).Because this syndrome is easy to diagnose, it has been successfully used as a biological effect monitoring system to determine the presence and degree of environmental TBT pollution.Currently, imposex is widely distributed (DeFur et al., 1999), and approximately 260 known species of neogastropods are affected (Titley-O'Neal et al., 2011).
The southern coast of Bahia, in northeastern Brazil, has high levels of boating activity, with the presence of harbors, shipyards, marinas and numerous moorings for small boats.Moreover, it is a geographically diverse region, with the presence of mangroves, coral reefs and sandstone banks along its coast.To date, there is no information about the action of TBT on marine organisms in this area.
The aim of this study was to investigate TBT contamination in this region, using imposex as biomarker in Stramonita rustica (Neogastropoda: Muricidae).

Samples
The samples were taken during the summer of 2014 between latitudes 13 to 16° S, in southern Bahia (Figure 1).Stramonita rustica (n = 330) was collected from environments with hard substrata such as rocks, sandstone banks and coral reefs, all close to harbors, boating activity and marinas with frequent small-boat activity.The sampling points were chosen according to occurrences of the species in representative quantities per catch (n ≥ 30), and also according to the presence of fishing and port activities.Thus, the municipalities of Ilhéus, Maraú, Itacaré and Canavieiras were investigated, but this species was not found at the last two of these places.
This collections were made with permission for activities with scientific purpose issued by Authorization and Information System of Biodiversity-SISBIO/ Chico Mendes Institute for Biodiversity Conservation -ICMBio of Brazil (concessions number: 20912-3.

Laboratory processing
All the snails were measured to the nearest 0.01 mm with calipers and then they were narcotized in a 7% (w/v) magnesium chloride (MgCl 2 ) solution for one hour.Following this, we cracked the shell in a vise.The external morphology of the genital system was analyzed under a stereomicroscope.The penis length was measured to the nearest 0.1 mm with a caliper.The sex ratio (M: F) was calculated as described by Birchenough et al. (2002).We also calculated the percentage of females with sexual morphological changes as I (%).To determine the levels of imposex, we evaluated the mean female (FPL) and male (MPL) penis lengths, by using the relative penis length (RPL) calculation proposed by Gibbs and Bryan (1987).
We also calculated the vas deferens sequence index (VDS) proposed by Gibbs and Bryan (1994).

Statistical analyzes
A matrix using Pearson coefficients was created to examine correlations between snail size and penis length in both female and male specimens at each site; the Kruskal-Wallis test was also used to verify differences in snail size and penis length, in both sexes.All analyses were performed using the Statistica software package.The significance level used was 95%.

Results and Discussion
Imposex was observed at seven of the eight monitoring stations (St) (Figure 2).This syndrome did not occur at St3, where boating activity is virtually absent and where there is high hydrodynamic activity, thereby resulting in stronger circulation and more frequent water renewal than at the other sampling points (Figure 2).Among the total number of females collected (n = 178), 30 specimens (16.39%) exhibited imposex.However, despite low index values, sterile females were observed at both St2 and St5 (vas deferens sequence index = VDS = 4) (Table 1; Figure 2).These stations showed also the highest I% (females with sexual morphological changes).According to Barroso et al. (2002), FPL (mean female penis length), VDS, I% and percentage of sterile females are generally the main indices that have been applied for evaluating imposex intensity in prosobranchs.
According to Bryan et al. (1986), the relative penis length index (RPL) can be used to describe the severity of imposex.We found large differences among the mean values of the RPL (relative penis length) at different sampling stations (Table 1).The stations closest to harbors and to fishing and boat maintenance activities had the lowest average RPL values, while the station farthest from nautical activities showed the highest RPL value (Figure 2).This result should be correlated with different timescale variations of exposure to TBT, and also to its spatial distribution (Smith, 1996;Barroso and Moreira, 1998;Rato et al., 2008).The RPL found in the St7 was one of the largest of northeast.This station is to close to the international exportation harbour of the city and had been lower only of the RPL found by Castro et al. (2008) in Alagoas and Sergipe.Despite of indexes taking penis length as their basis are viewed with reservations, given the possible relationship with testicular maturation and time of year (Rato et al., 2008), we consider that the RPL results of our study are reliable, since all specimens were collected at the same time.
With regard to the total sex ratio, the values were generally higher for females (Table 1).However, the sample stations that showed higher male values were the same ones that had higher frequency of imposex (Figure 2) and corresponded to areas with great nautical activity, which thus confirmed the results obtained from other studies (Miller et al., 1999;Shim et al., 2002).
Imposex in Stramonita rustica on southern Bahia coast Although some authors have considered that imposex causes changes to the growth rate of females due to higher allocation of energy (Terlizzi et al., 1999;Trigui El-Menif et al., 2006), we observed that females were generally larger than males (average = 24.07mm, SD ± 5.51; and 23.97 mm, SD ± 5.52, respectively), but this difference was not significant in relation to length and station (p > 0.05).These findings indicated the presence not only of natural specific variation but also of low environmental contamination by TBT, as was proposed   by Vasconcelos et al. (2010).Regarding penis length, we found a significant morphometric difference between males and imposexed females, but there was no difference in male penis length between the sample stations (Table 2).
This showed that, at least in our study, TBT did not seem to cause changes in males.Imposex induced by TBT are irreversible, i.e., cannot be reversed to the normal morphological situation or reduced to a lower stage, even if the gastropods are subsequently kept under TBT-free conditions (Oehlmann et al., 1998).Regarding S. rustica, the data from the present study showed that there was widespread use of TBT in southern Bahia.In studies conducted by Castro et al. (2007Castro et al. ( , 2008) ) along the northeastern coast of Brazil using S. rustica and S. haemastoma as bioindicators, it could be seen that the first species was less sensitive than S. haemastoma, i.e., S. rustica showed lower RPL and VDS values at the same collection points.These data, together with possibly relatively low TBT contamination in our study area, may explain the low imposex rates in southern Bahia.
In analyzing our results, we can say that the frequencies of imposex tended to be higher in areas with harbors, small-boat maintenance activities and docks.Similar results have been observed worldwide: on the northeastern Brazilian coast, in relation to S. rustica (Castro et al., 2007(Castro et al., , 2008)), in the Mediterranean, in relation to Hexaplex trunculus (L.) (Axiak et al., 1995), and in several harbors in Europe and in Halifax harbor, Canada, in relation to Nucella lapillus (L.) (Sousa et al., 2009).Other studies have also confirmed that areas near harbors and marinas are hotspots for TBT pollution (Barroso et al., 2002;Sousa et al., 2005).
In conclusion, although the TBT contamination levels seem to be bland in southern Bahia, the reproduction of S. rustica was compromised.This suggest the need for compliance and supervision of international and national legislation.

Figure 1 .
Figure 1.Map showing the sampling points along the southern coast of Bahia.

Figure 2 .
Figure 2. Monitored stations on southern Bahia coast indicating sampling stations, locations of main boating activity and sexual alterations, being: Ilhéus region and Maraú region.

Table 1 .
Imposex data for Stramonita rustica on the southern coast of Bahia, Brazil.

Table 2 .
Results from Kruskal-Wallis analysis on the penis length of Stramonita rustica between males and females and between sample stations.