Phenotypic plasticity of the gastropod Melanoides ...

Marine Pollution Bulletin 133 (2018) 701–710

Contents lists available at ScienceDirect

Marine Pollution Bulletin journal homepage: www.elsevier.com/locate/marpolbul

Phenotypic plasticity of the gastropod Melanoides tuberculata in the Nile Delta: A pollution-induced stabilizing selection

T

⁎

Ahmed Awad Abdelhadya,b, , Esraa Abdelrahmanc,d, Ashraf M.T. Elewab, Jiawei Fana, Shengrui Zhanga, Jule Xiaoa,e,f a

Key Laboratory of Cenozoic Geology and Environment, Institute of Geology and Geophysics, Chinese Academy of Sciences, Beijing 100029, China Geology Department, Faculty of Science, Minia University, El–Minia 61519, Egypt c El Mabarra Hospital, General Authority for Health Insurance, El–Minia 61511, Egypt d Zoology Department, Faculty of Science, Minia University, El–Minia 61519, Egypt e CAS Center for Excellence in Life and Paleoenvironment, Beijing 100044, China f College of Earth and Planetary Sciences, University of Chinese Academy of Sciences, Beijing 100049, China b

A R T I C LE I N FO

A B S T R A C T

Keywords: Melanoides tuberculata Phenotypic plasticity Stabilizing selection Geometric morphometric Manzala lagoon Nile Delta pollution

To understand the effect of metal pollution on the speciation process, we conducted comparative analyses of six populations of the gastropod Melanoides tuberculata, which dominated the Manzala lagoon (Nile Delta, Egypt). Geometric morphometric analysis was implemented to quantify the phenotypic plasticity of the species. The results from both Canonical Variate Analysis and Relative Warp indicated an overall decrease in the morphological breadth of M. tuberculata in the polluted sites. The favored phenotypes in the polluted sites have moderate whorl section, moderate ovate aperture, less-prominent radial ornament, and overall moderate-spired shells. Lack of morphological variations and dominance of intermediate phenotypes in the polluted sites indicate that stabilizing selection is driving the morphological pattern of this species. Moreover, analysis by using the partial least square model confirmed that metal pollution is the major predictor of the observed shape variations, whereas other biotic/abiotic traits are a minor predictor.

1. Introduction Understanding the mechanisms underlying the origin and maintenance of biodiversity is a central goal of modern ecological research (Vonlanthen et al., 2012; Tan et al., 2013). Morphological variation has become an interesting topic in evolutionary ecology (Debat and David, 2001; Mcnulty and Vinyard, 2015). Variation in animal morphology is influenced by environmental effects and genetic mutations (Debat and David, 2001). The interactions between genetic variations within a specific population and the environment play a pivotal role in the natural selection process, in which the environment determines the favorable characters that must exist (Debat and David, 2001). Because of the dynamic nature of the environment, genetically variable populations will be able to adapt to these changes compared to those that have no or little genetic variation (Monteiro and Nogueira, 2009). In general, environmental changes are usually accompanied by the appearance of new phenotypes within populations of the same taxon (adaptive radiation; Neubauer et al., 2013; Monteiro and Nogueira, 2009). Coastal ecosystems are vulnerable to both natural and ⁎

anthropogenic hazards (Pérez-Ruzafa et al., 2013). Anthropogenic activities highly control the coastal ecosystems, and this may result in habitat shifts. Therefore, coastal ecosystem taxa can provide the opportunity to understand the consequence of the pollution on the speciation process (Vonlanthen et al., 2012; Abdelhady, 2016; Márquez et al., 2011, 2017). Vonlanthen et al. (2012) found that genetic and functional distinctiveness among fish species was the result of eutrophication, and thus, conservation efforts should not only target threatened species but also identify and protect the ecological processes that generate and maintain these species. The biological processes of speciation and evolution have been continuously assessed by using shape analysis (see Hollander et al., 2006; Frederich et al., 2008; Berner et al., 2008; Monteiro and Nogueira, 2009; Zelditch et al., 2012; Neubauer et al., 2013, 2014). The shell geometry in gastropods is one of the most important biotic factors that determine the ability of the animal to live, feed, and avoid predations. According to laboratory and field observations, abiotic factors such as temperature, current, and salinity have major influences on the shell morphology (see Britton and McMahon, 2004; Rundle et al., 2004; Dillon and Herman, 2009; Márquez et al., 2015; Farani et al., 2015).

Corresponding author at: Key Laboratory of Cenozoic Geology and Environment, Institute of Geology and Geophysics, Chinese Academy of Sciences, Beijing 100029, China. E-mail address: [email protected] (A.A. Abdelhady).

https://doi.org/10.1016/j.marpolbul.2018.06.026 Received 11 January 2018; Received in revised form 11 May 2018; Accepted 8 June 2018 0025-326X/ © 2018 Elsevier Ltd. All rights reserved.


A.A. Abdelhady et al.

Therefore, coastal gastropods have often been used as a model organism for analyzing the phenotypic plasticity and highlighting the role of environmental pollution (e.g., Abdelhady, 2016; Márquez et al., 2011, 2017). Abdelhady (2016) and Márquez et al. (2011) highlighted the role of environmental perturbations on driving or shifting the morphological variation within a given population, a phenomenon that is well generalized in Márquez et al. (2017). Consequently, phenotypic plasticity can provide information on the biodiversity trend in addition to environmental changes. The gastropod Melanoides tuberculata is one of the most dominant freshwater gastropods worldwide. This species has successful invasion strategies and was reported to prevail in different environments (Lau et al., 1998; Vogler et al., 2002; Farani et al., 2015). In addition, qualitative observations demonstrated that this invasive species shows much more morphological variations than others (Pointier et al., 1993). Herein, we aim to quantify the phenotypic variation in M. tuberculata as a model for understanding the ecological consequences of environmental pollution. The main research question to be addressed is how the patterns of variations differ between polluted and unpolluted sites within the Manzala lagoon. Quantification of the morphological variations under stressful conditions in reference to a stable environment should draw a clear idea about the current biodiversity threats, which may result in evolutionary changes. Manzala lagoon, the largest of the Nile Delta, is an important resource for fisheries in the Nile River Delta and accounts for > 30% of all commercial and recreational fish landed and consumed in Egypt. However, Manzala lagoon has been classified as one of the most polluted lakes in Egypt (Wahab and Badawy, 2004; Elkady et al., 2015). Trace metals entering the system are expected to accumulate with time (Gu et al., 2011). Environmental pollution in the Manzala lagoon and its main drainage channels has previously been documented by measuring the concentrations of selected trace metals in fish and sediment samples (Wahab and Badawy, 2004; Gu et al., 2013). Authman et al. (2008) used Oreochromis niloticus as a model for biomonitoring and found that the concentrations of heavy metals exceeded the tolerance level for human consumption. Sources of contaminants in the Manzala lagoon include untreated sewage, atmospheric deposition, as well as agricultural and industrial wastes, which may have an impact on human health when fish are consumed (Elkady et al., 2015).

was located at the north near the Mediterranean outlets (reported as unpolluted; El-Kholy et al., 2012). Conductivity, pH, and water depth were measured at each site during sampling using the Xplorer GLX PS2002 data analysis tool (Table 1). Conductivity was used as an approximation for water salinity (EC/10001.0878∗0.4365).

2. Materials and methods

Many studies adopted landmark-based morphometric approaches for addressing a variety of questions in evolutionary biology (Reyment and Elewa, 2002; Elewa, 2003, 2005; Abdelhady and Elewa, 2010; Mcnulty and Vinyard, 2015). A landmark-based geometric morphometric analysis was implemented for morphological analyses of M. tuberculata, which dominates all the studied sites. Images of each individual were captured using a LEICA stereomicroscope and Nikon D50 digital camera. The procedure same as that described in Abdelhady (2016) for minimizing error associated with capturing 2D image from the 3D object was followed. Twenty-four landmarks were digitized using TPSDig Package (http://life.bio.sunysb.edu/morph/, Rohlf, 2006; Fig. 2). Generalized Procrustes Analysis (GPA) was applied to the coordinate data by the method of Rohlf and Slice (1990). Landmark configurations were translated, scaled, and rotated until the distances among homologous landmarks are minimized according to the partial least square (PLS) criteria. The resulting data were considered with regard to shape only (Appendix A).

2.2. Laboratory analyses In the laboratory, shells that were alive were separated from those that were dead, and specimens with life were identified down to the species level (mainly from Barash and Danin, 1982; Galil, 2004; Lotfy and Lotfy, 2015). Diversity indices (i.e., Dominance, Evenness, and Simpson) were calculated and compared among the study sites. Sediment grain size analysis was conducted by following the method given by Quintino et al. (1989). Sieving was done using an electric shaker for a sediment sample of 100 g (sieve opening size: 2, 1, 0.5, 0.25, 0.125, and 0.063 mm). The percentages of the different size classes were calculated. The percentage of fine-grained sediments (silt and clay) obtained by wet sieving through 0.125- and 0.063-mm sieves was also calculated (Table 1). M. tuberculata shells were picked from the washed and sieved (0.5 mm) samples. The shells with life were separated from those that were dead, and the specimens with life were used for geometric morphometric analyses. Eight major and trace elements (Zn, Cd, Cr, Pb, Cu, Fe, Mn, and Al) were analyzed using inductively coupled plasma optical emission spectrometry (ICP–OES) at the Institute of Geology and Geophysics, Chinese Academy of Sciences (Beijing, China). Sediment samples were dried at room temperature and then ground and screened through a 0.64-mm sieve to obtain the fine-grained proportion. A dried sediment aliquot of 0.2 g was taken in a 50-ml beaker, and then, 3 ml of a concentrated mixture of HNO3 and HCl (1:1) was added. Subsequently, all the treated samples were diluted to 10 ml with ultra-pure water (MilliQ). The samples were then filtrated for 5 min using a centrifuge to split the solid and trace metals. Analytical efficiency was checked using a standard reference material. The analytical values of the reference material were within the range of the certified values (< 5%). Samples were collected in triplicate and analyzed. 2.3. Geometric morphometric analysis

2.1. Sampling and measurements The Manzala lagoon was scanned and sampled during winter 2016. The lagoon is located 170 km north of Cairo between Port Said on the East and Damietta on the West (Fig. 1). The lagoon is connected to the Mediterranean Sea through three outlets, namely, El-Gamil 1, El-Gamil 2, and New El-Boughas. It is also connected to the Red Sea through the Suez Canal by the El-Qabouti Canal (Fig. 1). Recently, intense agriculture runoff has been found to enter the lagoon through many drains (e.g., Bahr El-Baqar, Bahr Hadus, Ramses, Mandesi, El-Sherw, and ElInaniya; see Abdelhady et al., in review; Fig. 1). The depth of the water in the lagoon varies from 4 m to less than half a meter. Eutrophication and pollution are the main threats to the lagoon (El-Kholy et al., 2012). The sediment of post Aswan High Dam of these lakes, occasionally in the Manzala lagoon, is polluted by many metals, especially Mn, Pb, Zn, and Cd, because of the direct human input from neighboring megacities, where the petrochemical industry is thought to be a major source (Gu et al., 2011). Aridity and low precipitation disable the flushing out of these heavy metals to the sea as in the case of humid lakes (Gu et al., 2013; Elkady et al., 2015). Sediments and benthic macrofauna were sampled using a 20 × 20cm Ekman grab sampler. Six sampling sites represented two main transects: the first transect was located at the southernmost part near the agricultural drains (sites 1 and 2; Fig. 1) and the second transect

2.4. Multivariate data analyses The Procrustes residual data were projected to Canonical Variate Analysis (CVA), which maximizes the differences among populations, and Thin plate spline (TPS), which visualizes the shape changes when one specimen is deformed into another one (i.e., displacement/deformation of the landmarks relative to the mean shape; for details see Zelditch et al., 2012; Klingenberg, 2013). The Relative Warp (RW) 702



Fig. 1. Satellite image of the Manzala lagoon and the position of sampling sites. Note the difference in water quality between the stagnant, plant-rich, muddy water of Transect A (sites 1 and 2), and the flowing water of Transect B (Sites 2, 4, 5, and 6). White arrows indicate the agricultural drains, which are the major source of pollution in the lagoon.

site (observations) and the biotic/abiotic parameters (predictors; Appendix B). Then, we correlated the first significant components between the two datasets. The sum of the covariance of the two significant components gives us the total explanatory capacity of the PLS models (Q2, 0 means no correspondence and 100 is complete correspondence). Loadings or the Variable Importance for the Projections (VIPs), which are linear coefficient plots between the particular predictor (herein, the biotic/abiotic variables) and the observations (shell shape) across all the model components, were used to quickly identify the explanatory variables that contribute the most to the models (i.e., the weight of each predictor; for details, see Zelditch et al., 2012). To test whether our data are adequate for quantitative analyses, the dataset was evaluated with rarefaction, a technique to assess species richness (see Sanders, 1968; Hurlbert, 1971). This curve is a plot of the

scores of each specimen are used to map their location on RW axes (Rohlf, 1996; Hammer and Harper, 2006). The plots from ordinations such as CVA and RW in addition to the TPS grid deformation provide a graphical representation of the shape changes and have been widely implemented (Klingenberg, 2013). Previous studies argued that shape variation was due to different biotic factors (e.g., abundance, the challenge for food/space, and genetic diversity) and abiotic factors (e.g., wave energy, substrate type, and water depth; see Abdelhady, 2016). The PLS model was employed to quantify the effect of each environmental/biological parameter. PLS is more reliable than other techniques when identifying relevant variables and their magnitudes of influence, especially in cases of small sample size. Chord transformation has been performed before PLS analysis. The two blocks of the data are the mean shell shape at each 703



Table 1 Species diversity, environmental data, and metal concentrations in the studied sampling sites. Sampling sites

Species

Diversity

Physical parameters

Major and trace metals

Melanoides tuberculata (Müller, 1774) Theodoxus niloticus (Reeve, 1856) Theodoxus michonii (Bourguignat, 1852) Unio terminalis Bourguignat, 1852 Coelatura aegyptiaca (Cailliaud, 1827) Bellamya unicolor (Olivier, 1804) Planorbis planorbis (Locard, 1883) Biomphalaria alexandrina (Ehrenberg, 1831) Haitia acuta (Draparnaud, 1805) Mactra lilacea Lamarck, 1818 Neverita didyma (Röding, 1798) Ceratoderma glaucum (Bruguière, 1789) Fulvia fragilis (Forsskål in Niebuhr, 1775) Pirenella conica Blainville, 1829 Corbicula fluminalis (Müller, 1774) Individuals Species Dominance Simpson Evenness Gravel Coarse sand medium sand Fine sand Silt Clay TDS pH Water depth Pb (μg/Kg) Zn (μg/Kg) Cr (μg/Kg) Cu (μg/Kg) Cd (μg/Kg) Al (g/Kg) Fe (g/Kg) Mn (g/Kg)

1

2

3

4

5

6

261 5 22 19 10 9 23 15 0 9 0 0 0 37 7 417 12 0.4 0.6 0.4 13.0 23.5 9.5 22.5 12.0 19.6 8.7 8.0 2.3 371.49 215.82 56.15 62.24 1.62 25.54 64.68 2.60

127 9 0 5 26 13 17 2 0 11 0 4 0 9 18 318 12 0.3 0.7 0.5 12.0 2.8 10.0 23.0 32.2 20.2 12.7 8.0 2.7 231.07 155.27 12.99 23.16 1.51 26.15 44.44 1.90

51 0 0 30 0 26 22 0 0 0 0 0 0 18 8 155 7 0.2 0.8 0.9 58.8 24.6 2.2 4.7 5.9 3.8 21.3 8.1 3.2 3.00 12.67 5.89 7.72 1.22 24.65 56.41 2.01

118 0 0 60 0 0 34 0 0 17 0 0 42 0 5 238 7 0.3 0.7 0.7 55.7 26.8 0.0 14.9 2.1 0.5 21.3 8.3 3.5 3.00 9.21 4.28 14.05 0.98 11.83 39.63 3.34

104 0 37 0 0 0 0 0 10 0 0 0 11 87 74 323 7 0.2 0.8 0.8 60.4 25.1 1.8 4.5 5.2 3.1 20.2 7.7 3.4 1.52 8.14 3.82 8.93 0.75 11.66 28.54 3.06

52 0 0 0 0 23 0 3 0 52 10 7 105 0 0 252 8 0.3 0.7 0.7 27.0 9.9 36.5 23.0 3.3 0.3 23.0 8.2 2.5 0.85 4.58 2.60 3.05 0.26 5.15 10.90 1.36

Bold indicate that metal concentration is above the Effective Range Low/Median (Long and Morgan, 1990).

distinguished along the CV1 axis, populations from the polluted transect have positive values and are plotted to the left most area, whereas populations from the unpolluted transect loaded negatively and were plotted to the right (Fig. 4). The low variability within the populations from polluted sites compared to that from the unpolluted ones is the most conspicuous feature along the CV1 axis. The population of site 3, which share the same (at least similar) habitat settings with the other population of the same transect (unpolluted), was plotted near populations of the polluted sites. In addition, it has a slight narrower variation/scatter and resembles transect A's populations (Fig. 4). Herein, it is important to refer to the relatively high Cd level (above Effects Range Low [ERL]) of the sediments of this site (Table 1). The second axis (CV2) represents 27.3% of the total variance and arranges the individuals according to specific shape characteristics. Individuals with higher axis two values have a more slender shell shape and rounded aperture, whereas individuals with wider shells and ovate apertures have lower axis two values (Fig. 4).

number of species as a function of the number of specimens. A steep slope of the curve to the right indicates that a large fraction of the species diversity remains to be discovered. A flattening curve to the right that more or less resembles a plateau indicates that a reasonable number of individual samples have been sampled, and hence, more intensive sampling will yield only a few additional species (see Abdelhady and Fürsich, 2014). Rarefaction curves of all the measured sites are flattened to the right (Fig. 3), which means that the sampling is sufficient and additional faunal sampling will have no or little effect on the results. All data were analyzed using PAST version 2.17 (Hammer et al., 2001). 3. Results 3.1. Shape variation 3.1.1. Canonical Variate Analysis We analyzed 300 individuals from six sites. For discrimination among the different populations, the CVA was applied. The CVA plot shows a weak morphological overlap between the two transects and a strong overlap among populations from the same transects (Fig. 4). The CVA plot shows a satisfying shape gradient along the first CV axis (CV1 axis), which explains 54.9% of the total variability among the different populations. Few shell shape attributes can be changed along the CV1 axis. Indeed, positive CV1 values are associated with much wider body whorls, whereas negative CV1 values are associated with narrower body whorls. Although no more specific shell attributes can be

3.1.2. Thin Plate Spline The morphological differences between and within the two transects (interspecific and intraspecific variations) obtained by the CVA agree with the shape analysis of the TPS (Fig. 5), which is expected. Note that the TPS deformation results can be used in statistical analyses like the coordinates obtained by Procrustes and like Bookstein shape coordinates (Zelditch et al., 2012). The deformation/transformation from the original shape of all populations to the shape of a specific population shows a general trend toward increasing variability, where 704



shape of all populations. By contrast, the mean shape of populations from unpolluted sites deviated less from the mean original shape of all populations, and this can be indicated by a minor (site 1) or low (site 2) deformation of the grids (Fig. 5). Furthermore, the deformation (differences) is mainly high for aperture and whorl section (occasionally at the juvenile/mature boundary) but is less cauterized in body whorls. This high deformation is presented by dark red spots on the heat map of the landmarks (Fig. 5). 3.1.3. Relative Warp As the transformation grid of the TPS represents only a method for visualization, which does not represent a complete biological meaning, RW analyses were also implemented (Fig. 6). RW1 (10.6%) explained the variation in individuals that loaded positively with an increase in shell height/shell width ratio (low spired vs. high spired; Fig. 6). The populations from the polluted sites have intermediate values on this axis (Fig. 6). The dense cloud of the specimens of the polluted sites (Fig. 6) indicates limited morphological diversity along both axes of the RW, which confirms the CVA and TPS results. By contrast, the specimens from the second transect are highly scattered (i.e., high morphological variability/diversity). RW2 (7.7%) explained the variation with regard to a combination of morphological characteristics such as aperture roundness, apex (juvenile) length, body whorl width, and whorl section. Positive RW2 values represent individuals with a quadrate whorl section, rounded aperture, wide body whorl, and elongated apex. By contrast, individuals with an elongated whorl section, oval aperture, narrow body whorl, and short apex loaded negatively (Fig. 6). Although the first components explain the low proportion of the total variance, PLS regression (not shown here) indicate that RW1 is the only major evidence for the shell shape variation. The phenotypic variation within and among different populations is well illustrated by the box plot for the RW1 scores (Fig. 7). The two ordination plots indicated that populations from the “polluted” transect A (sites 1 and 2), which plotted mainly on central positions, must have a moderate slender shell shape, moderate apical angle, moderate whorl section, and less prominent radial ornament (low variation). By contrast, populations from the “unpolluted” transect B (sites 3:6) have a characteristic of smooth-to-prominent radial ornament (based on qualitative comparison, see Fig. 7) and high or low (but not moderate) apical angle, quadrate-to-elongate whorls, rounded-toelongated aperture, and low- to high-spired shell shape.

Fig. 2. Shell morphology of Melanoides tuberculata showing the position of the 24 landmarks analyzed.

3.2. Environmental and biological traits Physical parameters (e.g., electric conductivity, pH, and salinity) and substrate type represent the major differences among sites. High water salinity and coarse grain substrates characterize the unpolluted sites of the lagoon, whereas the polluted sites located to the south are characterized by lower salinity and fine-grained substrate (Table 1). In addition, metal concentrations agree with our expectations based on the previous investigations (e.g., Gu et al., 2013; El-Kholy et al., 2012). Metal concentrations indicate unpolluted conditions of the studied metals at the northern sites near the Mediterranean outlets, where the concentrations of the metals fall below the levels of concern (Long and Morgan, 1990; Table 1). By contrast, concentrations of Cd and Zn in the southern sites near the agricultural drains fall above the ERL) and those of Pb fall above Effects Range Median (Long and Morgan, 1990; Table 1). Fifteen mollusk species were identified (11 gastropods and 6 bivalves; Table 1). Species richness at the six sites varied from 7 to 12 species. Abundance also varies from 155 to 417 individuals (Table 1). Although species richness is higher at polluted sites (sites 1 and 2; Fig. 4) than at unpolluted areas, diversity (Simpson) and evenness are low, and dominance is very high in the polluted sites compared to other unpolluted sites (Table 1). M. tuberculata is the only species recorded from all sites. As it occurs with high abundance, it was selected for

Fig. 3. Rarefaction curves show the diversity of the benthic mollusks at the studied sites. All curves flatten to the right, which suggests adequate sampling of the analyzed populations.

the deformation pattern becomes more complicated in populations from unpolluted sites (Fig. 5). The high variation within these populations indicates the occurrence of many phenotypes, and thus, the mean shape of a specific population deviated much more from the mean 705



Fig. 4. The CVA scatter plot of Melanoides tuberculata shows a gradient on the distribution of the populations, where those from the polluted sites were plotted at the right most area. The main findings are the less variability in shape in the populations from the polluted sites.

4. Discussion

further morphological analyses as we mentioned above. The PLS model was employed to determine the effect of the biotic/ abiotic traits on shell morphology. The models are well fitted (the first three components explain 99% of the covariance between the two datasets), and the cumulated Q2 corresponding to this model reaches 40.12% (Fig. 8). Plots of the loadings (VIPs) in Fig. 8 indicate that metal pollution (occasionally Pb and Zn) is the first major predictor of the total variability in shell shape (i.e., a good predictor, 0.48 and 0.46, respectively; Fig. 8). The shale percentage is the second main contributor to the shell shape (0.38). However, the latter can be attributed to the well-known association between metal concentrations and finegrained sediments (i.e., this result is related to the collinearity between the metals and the shale content; Whitney, 1975). Therefore, the shape variation and the stabilizing selection mentioned above can be argued without bias to the metal pollution.

4.1. Ecological remarks M. tuberculata is an epifaunal gastropod that is dominant in shallow slow-flowing water bodies on soft mud or sand substrates (Da Silva et al., 1994). It grazes on algae and diatoms in addition to fine organic materials of the substrate (i.e., generalist/detritivore; Dudgeon, 1989). It exhibits clear phenotypic plasticity, and it has been reported to prevail under a wide range of environmental conditions (Lau et al., 1998; Vogler et al., 2002). M. tuberculata also adapts under different salinity conditions (Farani et al., 2015) and shows high resistance to pollution and other harsh conditions (Pointier et al., 1993). In addition to salinity tolerance, it has resistance to high pH levels and low oxygen saturation levels. The shell is characterized by weak-to-prominent vertical ribs, occasionally in the upper whorls of the shell. The adult shell length

Fig. 5. Shape change in Melanoides tuberculata depicted by thin plate spline. The deformation heat map shows the mean shape of each population when projected against the mean shape of all populations. 706



Fig. 6. Plot of the Relative Warp axes 1 and 2. Deformation grids represent individuals at axis extremes.

populations of M. tuberculata on the polluted sites.

varies between 30 and 36 mm; females have a wider body whorl than males, and both sexual and asexual modes of reproduction were reported (Heller and Farstay, 1989). The juvenile protoconch size is < 4 mm, which is represented by the first two whorls. Low mortality rates, high population densities, and long lifespan in addition to low intraspecific competition enable M. tuberculata to live in stable and harsh environments (Pointier et al., 1991). Consequently, this species has been used continuously as a bioindicator for pollution (see Qarooni et al., 2012). According to Genner et al. (2004b), M. tuberculata may have two different clades. One clade is a recent introduction into Lake Malawi and has Southeast Asian ancestry and the other clade is widespread throughout East Africa and the Middle East (i.e., two cryptic species; Genner et al., 2004a; Dillon, 2014). Sørensen et al. (2005) indicated that three different clades occur in Lake Malawi; one clade consisted of invasive M. tuberculata, another clade of native M. tuberculata, and the third complex one of Melanoides polymorpha. The diversity of phenotypes within the genus Melanoides in Lake Malawi was attributed to the divergence of the genetic clones instead of species differentiation. Finally, we conclude on the basis of these previous investigations that the variation among the studied populations may be generated by the following: (1) sexual dimorphism (males vs. females), (2) reproduction type (i.e., sexual vs. asexual), or (3) cryptic species. However, none of these scenarios can explain the variation observed in the Manzala lagoon. First, all these three possible scenarios (if they exist) should be reproducible and observed at all sites, which is not the case in the Manzala lagoon. Second, all these scenarios adequately produced two or more different, distinctive phenotypes (See Dunithan et al., 2012; Morais et al., 2014; Neubauer et al., 2013, 2014; Abdelhady, 2016; Márquez et al., 2017), which is also not the case herein. Although there is a gradient in shape variability on the morphospace between two extreme phenotypes, this is limited to the unpolluted sites and not seen in any of the polluted ones. Therefore, there must be another scenario, and this should explain the limited morphological variability of the

4.2. Stabilizing versus directional selection M. tuberculata has a large morphological variation in its shell, which ranges from individuals with a less ornament, relatively high spire, ovate aperture, and an elongated whorl section to individuals with a prominent radial rib, rounded aperture, low spire, and a quadrate whorl section. This difference in the shell shape point to the phenotypic plasticity in response to habitat heterogeneity. This morphological diversity may be related to the occurrence within different geographic areas. Svanbäck and Eklöv (2006) found a clear positive correlation between morphological and geographical distances within the same drainage. However, herein, the two transects are at a distance of < 10 km and located within the same environment with minor physicochemical differences. The polluted sites show low variability and thus occupy a restricted intermediate region of the available morphospace (Figs. 4 and 6). These plots show that individuals from both polluted and unpolluted sites (interspecific) can be separated by the shape analysis (separation along ordination axes). The distinct morphological shift does not occur between the polluted and unpolluted sites (i.e., intraspecific) but occurs in the variation among individuals belonging to the same population (i.e., interspecific), where populations from polluted sites has less variability. The latter points to the beginning of the divergence of phenotypes within this microhabitat (distance of < 10 km). Stabilizing selection is indicated when the population of the polluted sites occupies the intermediate position on the plots, have low variability, and deviated less from the mean shape. The extreme phenotypes, which are most common at the unpolluted sites, are decreased significantly at the polluted sites, where the individuals do not vary greatly regarding apical angle, whorl sections, aperture, and overall shape. This is shown as a dense cloud in the CVA (Fig. 4) and RW (Fig. 6). As habitat heterogeneity is expected to 707



argued that the phenotypic variation may be due to the pollution intensity. Although Abdelhady (2016) and Márquez et al. (2017) noted that pollution produces a globular shell shape, an effect that probably allows gastropods to isolate themselves from the external adverse environment, this was not the case in the Manzala lagoon. There is no specific phenotype characterizing the polluted sites; instead, there is a general decrease in the phenotypic diversity. Therefore, stabilizing selection, as the main mechanism of natural selection, rather than phenotypic divergence is indicated. The morphological shift (ecological speciation) in stabilizing selection started with the disappearance of the extreme phenotypes, and the descendant species occupy narrow intermediate peaks corresponding to favorable ecological niches, which is the case in the polluted sites. Note that in stabilizing selection, the dominant mechanism for natural selection, the population mean stabilizes on a particular nonextreme value (Schmalhausen, 1949; Mitchell-Olds et al., 2007), and this resulted in a general decrease in the variation within the population (Lemey et al., 2009). As the pollution intensity represents the major differences among the examined sites (Table 1) and PLS model confirmed that the pollution is the major predictor of shape (Fig. 8), we can conclude that metal pollution has a major influence on the adaptive radiation of the gastropod M. tuberculata. In agreement with our results of the phenotypic plasticity of M. tuberculata in the Manzala lagoon, Gilinsky (1981), based on morphometric data from fossil and modern Archaeogastropoda data, revealed that there has been a progressive reduction in the shell morphologies with time, i.e., a trend toward a decrease in high/low-spired individuals and increase in equidimensional forms. He argued that this macroevolutionary trend is necessary for stabilizing species selection. He added that stabilizing selection has been burdened by the emphasis on differential mortality in microevolution rather than a macroevolutionary change. Similarly, based on shell morphology (mainly height and width), Moreno-Rueda (2009) highlighted the hybrid effect of both disruptive selection, caused by predation, and stabilizing selection, caused by other unknown mortality sources, on gastropod populations with the same strength. He found that the two selective forces acting in opposite directions resulted in an absence of appreciable selection on shell height. Our results also agree with those of Neubauer et al. (2014), who found that the gastropod Microcolpia parreyssii exhibited a tremendous

Fig. 7. Box plot of the RW axis 1 scores for each population. The main phenotypes from different populations illustrating their morphological variability are also presented. The low variability among the populations from polluted sites points to the stabilizing selection mechanism of natural selection.

increase the genetic variations (Debat and David, 2001) and fluctuating environments favor phenotypic plasticity (Svanbäck and Eklöv, 2006), environmental data cannot explain the major decrease in the phenotypic diversity of the populations on the polluted transect; hence, it is

Fig. 8. Plot of the VIPs (Variable Importance for the Projection) of the PLS model between the biotic/abiotic parameters and the shell shape of Melanoides tuberculata. 708



diversity of shapes during the earlier Holocene, although the recent assemblage includes only a single phenotype. They argued that the latter was due to the high eutrophication level (increased input of organic matter) in the small swamp environment, which marginalized the melanopsid gastropod and reduced its population size.

Pleuroceridae) in Indiana, USA. Am. Malacol. Bull. 30, 127–133. Elewa, A.M.T., 2003. Morphometric studies on three ostracod species of the genus Digmocythere Mandelstam from the middle Eocene of Egypt. Palaeontol. Electron. 6 (2), 11 (19 December). Elewa, A.M.T., 2005. Morphological variability and adaptability in Egyptian Eocene ostracod species Paracosta Mokattamensis (Bassiouni) from the Eocene of Egypt. In: Elewa, A.M.T. (Ed.), Migration in Organisms: Climate, Geography, Ecology. SpringerVerlag Publishers, Heidelberg, Germany, pp. 71–84 (August). Elkady, A.A., Sweet, S.T., Wade, T.L., Klein, A.G., 2015. Distribution and assessment of heavy metals in the aquatic environment of Lake Manzala, Egypt. Ecol. Indic. 58, 445–457. El-Kholy, R., El-Saadi, A., Abdin, A., 2012. Pollution and measures towards pollution of the Mediterranean Sea. In: Choukr-Allah, R., Ragab, R., Rodríguez-Clemente, R. (Eds.), Integrated Water Resources Management in the Mediterranean Region: Dialogue towards a New Strategy. Springer, pp. 175–194. Farani, G.L., Nogueira, M.M., Johnsson, R., Neves, E., 2015. The salt tolerance of the freshwater snail Melanoides tuberculata (Mollusca, Gastropoda), a bioinvader gastropod. Pan–American. J. Aquat. Sci. 10 (3), 212–221. Frederich, B., Adriaens, D., Vandewalle, P., 2008. Ontogenetic shape changes in Pomacentridae (Teleostei, Perciformes) and their relationships with feeding strategies: a geometric morphometric approach. Biol. J. Linn. Soc. 95, 92–105. Galil, B.S., 2004. Alien species in the Mediterranean Sea—which, when, where, why? In: Davenport, J., Burnell, G.M., Cross, T., Emmerson, M., McAllen, R., Ramsay, R., Rogan, E. (Eds.), Challenges to Marine Ecosystems. Proceedings of the 41st European Marine Biology Symposiumpp. 105–116. Genner, M.J., Michel, E., Erpenbeck, D., de Voogd, N., Witte, F., Pointier, J.P., 2004a. Camouflaged invasion of Lake Malawi by an Oriental gastropod. Mol. Ecol. 13, 2135–2141. Genner, M.J, Sims, D.W, Wearmouth, V.J, Southall, E.J, Southward, A.J, Henderson, P.A, Hawkins, S.J., 2004b. Regional climatic warming drives long-term community changes of British marine fish. Proc. R. Soc. B 271, 655–661. http://dx.doi.org/10. 1098/rspb.2003.2651. [PMC free article] [PubMed]. Gilinsky, N.L., 1981. Stabilizing species selection in the Archaeogastropoda. Paleobiology 7 (03), 316–331. Gu, J., Chen, Z., Salem, A., 2011. Post–Aswan dam sedimentation rate of lagoons of the Nile Delta, Egypt. Environ. Earth Sci. 64, 1807–1813. Gu, J., Salem, A., Chen, Z., 2013. Lagoons of the Nile delta, Egypt, heavy metal sink: with a special reference to the Yangtze estuary of China. Estuar. Coast. Shelf Sci. 117, 282–292. Hammer, Ø., Harper, D.A.T., 2006. Paleontological Data Analysis. Blackwell, pp. 351. Hammer, Ø., Harper, D.A.T., Ryan, P.D., 2001. PAST: paleontological statistics software package for education and data analysis. Palaeontol. Electron. 4, 1–9. Heller, J., Farstay, V., 1989. A field method to separate males and female of the freshwater snail Melanoides tuberculata. J. Molluscan Stud. 55, 427–429. Hollander, J., Collyer, M.L., Adams, D.C., Johannesson, K., 2006. Henotypic plasticity in two marine snails: constraints superseding life history. J. Evol. Biol. 19, 1861–1872. Hurlbert, S.H., 1971. The nonconcept of species diversity: a critique and alternative parameters. Ecology 52, 577–586. Klingenberg, C.P., 2013. Visualizations in geometric morphometrics: how to read and how to make graphs showing shape changes. Hystrix 24, 15–24. Lau, S., Mohamed, M., Tan, A.C.Y., Su'ut, S., 1998. Accumulation of heavy metals in freshwater molluscs. Sci. Total Environ. 214, 113–121. Lemey, P., Salemi, M., Vandamme, A., 2009. The Phylogenetic Handbook. Cambridge University Press. Long, E.R., Morgan, L.G., 1990. The Potential for Biological Effects of Sediment–sorbed Contaminants Tested in the National Status and Trends Program. NOAA Tech. Memo. NOS OMA 52. US National Oceanic and Atmospheric Administration, Seattle, Washington (175 p). Lotfy, W.M., Lotfy, L.M., 2015. Synopsis of the Egyptian freshwater snail fauna. Folia Malacologica 23 (1), 19–40. Márquez, F., González-Joséa, R., Bigatti, G., 2011. Combined Methods to Detect Pollution Effects on Shell Shape and Structure in Neogastropods Ecological Indicators. 11(2). pp. 248–254. Márquez, F., Vilela, R.A.N., Lozada, M., Bigatti, G., 2015. Morphological and behavioral differences in the gastropod Trophon geversianus associated to distinct environmental conditions, as revealed by a multidisciplinary approach. J. Sea Res. 95, 239–247. Márquez, F., Primost, M.A., Bigatti, G., 2017. Shell shape as a biomarker of marine pollution historic increase. Mar. Pollut. Bull. 114, 816–820. Mcnulty, K.P., Vinyard, C.J., 2015. Morphometry, geometry, function, and the future. Anat. Rec. 298 (1), 328–333. Mitchell-Olds, T., Willis, J.H., Goldstein, D.B., 2007. Which evolutionary processes influence natural genetic variation for phenotypic traits? Nat. Rev. Genet. 8, 845–856. Monteiro, L.R., Nogueira, M.R., 2009. Adaptive radiations, ecological specialization, and the evolutionary integration of complex morphological structures. Evolution 64, 724–743. Morais, P., Rufino, M., Reis, J., Dias, E., Sousa, R., 2014. Assessing the morphological variability of Unio delphinus. J. Molluscan Stud. 80, 17–23. Moreno-Rueda, G., 2009. Disruptive selection by predation offsets stabilizing selection on shell morphology in the land snail. Evol. Ecol. 23, 463–471. Neubauer, T.A., Harzhauser, M., Kroh, A., 2013. Phenotypic evolution in a fossil gastropod species lineage: evidence for adaptive radiation? Palaeogeogr. Palaeoclimatol. Palaeoecol. 370, 117–126. Neubauer, T.A., Harzhauser, M., Georgopoulou, A., Wrozyna, C., 2014. Population bottleneck triggering millennial–scale morphospace shifts in endemic thermal–spring melanopsids. Palaeogeogr. Palaeoclimatol. Palaeoecol. 414, 116–128.

5. Conclusion We used geometric morphometric and multivariate analyses to quantify the phenotypic plasticity of the gastropod M. tuberculata in the Manzala lagoon. The results indicated a decreased morphological breadth in the polluted sites compared to that in the unpolluted ones. Stabilizing selection may favor the intermediate individuals on the morphospace, thereby resulting in a general decrease in the noise around the peak of the favored phenotype. As pollution intensity represents the major difference between the examined sites (Table 1) and PLS confirmed that metal concentrations are the main predictor of shell shape (Fig. 8), the pollution-induced stabilizing selection is responsible for the pattern of low phenotypic diversity of the gastropod M. tuberculata at least in the study area. Future analyses integrating molecular and morphological data are urgently needed to obtain a robust conclusion on the impact of pollution on the genetic variation and the speciation process of the aquatic gastropods. Supplementary data to this article can be found online at https:// doi.org/10.1016/j.marpolbul.2018.06.026. Acknowledgments The authors thank Dr. John Huntley (University of Missouri, Columbia, USA) for providing a literature review as well as valuable comments and suggestions. The authors appreciate Dr. Mahmoud M. Abdel Rahman (Minia University, Egypt) for assisting during the sampling campaign. The authors acknowledge Mr. Shoukri and Sheikh Tarek for being involved in the sampling campaign. The authors are grateful to Dr. Liu Yanhong (Institute of Geology and Geophysics, Chinese Academy of Sciences, Beijing) for valuable guidance in metal analyses. This work was funded by CAS President's International Fellowship Initiative (No. 2017VCC0013). References Abdelhady, A.A., 2016. Phenotypic differentiation of the Red Sea gastropods in response to the environmental deterioration: geometric morphometric approach. J. Afr. Earth Sci. 115, 191–202. Abdelhady, A.A., Elewa, A.M.T., 2010. Evolution of the Upper Cretaceous oysters: Traditional morphometrics approach. In: Elewa, A.M.T. (Ed.), Morphometrics for Nonmorphometricians. Springer–Verlag Publishers, Heidelberg, Germany. Abdelhady, A.A., Fürsich, F.T., 2014. Macroinvertebrate palaeo–communities from the Jurassic succession of Gebel Maghara (Sinai, Egypt). J. Afr. Earth Sci. 97, 173–193. Authman, M.M.N., Bayoumy, E.M., Kenawy, A.M., 2008. Heavy metal concentrations and liver histopathology of Oreochromis niloticus in relation to aquatic pollution. Glob. Vet. 2 (3), 110–116. Barash, A., Danin, Z., 1982. Mediterranean Mollusca of Israel and Sinai: composition and distribution. Isr. J. Zool. 31 (3–4), 86–118. Berner, D., Adams, D.C., Grandchamp, A.C., Hendry, A.P., 2008. Natural selection drives patterns of lake–stream divergence in stickleback foraging morphology. J. Evol. Biol. 21, 1653–1665. Britton, D., McMahon, R., 2004. Environmentally and genetically induced shell–shape variation in the freshwater pond snail Physa (Physella) virgata (Gould, 1855). Am. Malacol. Bull. 19, 93–100. Da Silva, R.E., Melo, A.L., Pereira, L.H., Frederico, L.F., 1994. Levantamento malacologico da bacia hidrografica do Lago Soledade, Ouro Branco (Minas Gerais, Brasil). Rev. Inst. Med. Trop. Sao Paulo 36, 437–444. Debat, V., David, P., 2001. Mapping phenotypes: canalization, plasticity and developmental stability. Trends Ecol. Evol. 16, 555–561. Dillon, R.T., 2014. Cryptic phenotypic plasticity in populations of the North American freshwater gastropod, Pleurocera semicarinata. Zool. Stud. 53, 31. Dillon, R.T., Herman, J., 2009. Genetics, shell morphology, and life history of the freshwater pulmonate limpets Ferrissia rivularis and Ferrissia fraglis. J. Freshw. Ecol. 24, 261–271. Dudgeon, D., 1989. Ecological strategies of Hong Kong Thiaridae (Gastropoda: Prosobranchia). Malacol. Rev. 22, 39–53. Dunithan, A., Jacquemin, S., Pyron, M., 2012. Morphology of Elimia livescens (Mollusca:

709



modifies induced defenses in snails. Proc. R. Soc. Lond. B 271, S67–S70 (special publication). Sanders, H.L., 1968. Marine benthic diversity: a comparative study. Am. Nat. 102, 243–282. Schmalhausen, I., 1949. Factors of evolution. In: The Theory of Stabilizing Selection, (Philadelphia: Blakiston). Sørensen, L. von G., Jørgensen, A., Kristensen, T.K., 2005. Molecular diversity and phylogenetic relationships of the gastropod genus Melanoides in Lake Malawi. Afr. Zool. 40 (2), 179–191. Svanbäck, R., Eklöv, P., 2006. Genetic variation and phenotypic plasticity: causes of morphological and dietary variation in Eurasian perch. Evol. Ecol. Res. 8, 37–49. Tan, J., Kelly, C.K., Jiang, L., 2013. Temporal niche promotes biodiversity during adaptive radiation. Nat. Commun. 4, 2102. Vogler, R.E., Nunez, V., Gutierrez Gregoric, D.E., Beltramino, A.A., Peso, J.G., 2002. Melanoides tuberculata: the history of an invader. In: Hamalainen, E.M., Jarvinen, S. (Eds.), Snails: Biology, Ecology and Conservation. Nova Science Publishers, Inc, New York, pp. 64–84. Vonlanthen, P., Bittner, D., Hudson, A.G., Young, K.A., Muller, R., Lundsgaard-Hansen, B., 2012. Eutrophication causes speciation reversal in whitefish adaptive radiations. Nature 482, 357–362. Wahab, R., Badawy, M.I., 2004. Water quality assessment of the River Nile system: an overview. Biomed. Environ. Sci. 17, 87–100. Whitney, P.R., 1975. Relationship of manganese–iron oxides and associated heavy metals to grain size in stream sediments. J. Geochem. Explor. 4 (2), 251–263. Zelditch, M.L., Swiderski, D.L., Sheets, H.D., 2012. Geometric Morphometrics for Biologists: a Primer. Elsevier Academic Press, London.

Pérez-Ruzafa, A., Marcos, C., Pérez-Ruzafa, I., Pérez-Marcos, M., 2013. Are coastal lagoons physically or biologically controlled ecosystems? Revisiting r vs. K strategies in coastal lagoons and estuaries. Estuar. Coast. Shelf Sci. 132, 17–33. Pointier, J., Toffart, J., Lefévre, M., 1991. Life tables of freshwater snails of the genus Biomphalaria (B. glabrata, B. Alexandrina, B. straminea) and of one of its competitors Melanoides tuberculata under laboratory conditions. Malacologia 33, 43–54. Pointier, J., Theron, A., Borel, G., 1993. Ecology of the introduced snail Melanoides tuberculata (Gastropoda: Thiaridae) in relation to Biomphalaria glabrata in the marshy forest zone of Guadeloupe, French West Indies. J. Molluscan Stud. 59, 421–428. Qarooni, I.H.M., Akber, M.M., Al-Saad, H.T., 2012. Snail Melanoides tuberculata for Shatt Al-Arab river pollution by some heavy metals. In: Proceedings of the 4th Kurdistan Conference on Biological Sciences. University of Duhok. Quintino, V., Rodrigues, A.M., Gentil, F., 1989. Assessment of macrobenthic communities in the lagoon of Óbidos, western coast of Portugal. Sci. Mar. 53, 645–654. Reyment, R.A., Elewa, A.M.T., 2002. Size and shape variation in Egyptian Eocene Loxoconcha (Ostracoda) studied by morphometric methods (a methodological study). In: Thiergärtner, H. (Ed.), Mathematical methods and data bank application in paleontology. Math. Geol. 6. pp. 3–14 Germany (August). Rohlf, 1996. Morphometric spaces, shape components and the effects of linear transformations. In: Marcus, L.F., Corti, M., Loy, A., Naylor, G.J.P., Slice, D.E. (Eds.), Advances in Morphometrics. NATO ASI Series Plenum Press, New York, pp. 117–129. Rohlf, F.J., 2006. tpsDig, Digitize Landmarks and Outlines, Version 2.05. In: Stony Brook. Department of Ecology and Evolution, State University of New York, NY. Rohlf, F.J., Slice, D., 1990. Extensions of the Procrustes method for the optimal superimposition of landmarks. Syst. Zool. 39, 40–59. Rundle, S., Spicer, J., Coleman, R., Vosper, J., Soane, J., 2004. Environmental calcium

710