Differential response to Cadmium exposure by

0 downloads 0 Views 2MB Size Report
Aug 7, 2018 - depression and high anaerobic capacity during exposure to air and ..... MODELLER package with two different strategies: (1) by defining rigid ...... Goddard, T.D., Huang, C.C., Couch, G.S., Greenblatt, D.M., Meng, E.C., Ferrin,.
Science of the Total Environment 648 (2019) 561–571

Contents lists available at ScienceDirect

Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv

Differential response to Cadmium exposure by expression of a two and a three-domain metallothionein isoform in the land winkle Pomatias elegans: Valuating the marine heritage of a land snail Lara Schmielau a,1, Martin Dvorak a,1, Michael Niederwanger a, Nicole Dobieszewski a, Veronika Pedrini-Martha a, Peter Ladurner a, Jaime Rodríguez-Guerra Pedregal b, Jean-Didier Maréchal b, Reinhard Dallinger a,⁎ a b

Department of Zoology and Center of Molecular Biosciences Innsbruck, University of Innsbruck, Technikerstraße 25, 6020 Innsbruck, Austria Insilichem, Departament de Química, Universitat Autònoma de Barcelona, 08193 Bellaterra, Barcelona, Spain

H I G H L I G H T S

G R A P H I C A L

A B S T R A C T

• Pomatias elegans is a land snail closely related to the marine periwinkle, Littorina littorea. • Under Cd stress, the two MT genes of Pomatias elegans are upregulated differentially, expressing in the midgut gland a two and a three-domain metallothionein, respectively • Primary sequence analysis of MTs of both species suggest origin from a common marine ancestor • However, the three-domain MTs of both species have evolved independently • The MT system of Pomatias elegans is suggested as a valuable biomarker tool in ecotoxicology

a r t i c l e

i n f o

Article history: Received 28 March 2018 Received in revised form 16 July 2018 Accepted 30 July 2018 Available online 7 August 2018 Editor: Henner Hollert Keywords: Pomatias elegans Littorina littorea Cadmium detoxification Metallothionein upregulation Biomarker Evolution

a b s t r a c t Through evolution, marine snails have adapted several times independently to terrestrial life. A prime example for such transitions is the adaptation to terrestrial conditions in members of the gastropod clade of Littorinoidea (Caenogastropoda). Some species of this lineage like the periwinkle (Littorina littorea), live in intertidal habitats, where they are intermittently exposed to semi-terrestrial conditions. Pomatias elegans is a close relative of Littorina littorea that has successfully colonized terrestrial habitats. Evolutionary transitions from marine to terrestrial conditions have often been fostered in marine ancestors by acquisition of physiological pre-adaptations to terrestrial life. Such pre-adaptations are based, among others, on the optimization of a wide repertoire of stress resistance mechanisms, such as the expression of metal inactivating metallothioneins (MTs). The objective of our study was to explore the Cd handling strategy in the terrestrial snail Pomatias elegans in comparison to that observed previously in Littorina littorea. After Cd exposure, the metal is accumulated mainly in the midgut gland of Pomatias elegans, in a similar way as in its marine relative. Upon Cd exposure, Pomatias elegans expresses Cd-specific MTs, as also described from Littorina littorea. In contrast to the latter species, however, the detoxification of Cd in Pomatias elegans is mediated by two different MT isoforms, one two-domain and one three-domain MT. Although the MT proteins of both species are homologous and clearly originate from one

⁎ Corresponding author at: Department of Zoology, Universität Innsbruck, Technikerstraße 25, 6020 Innsbruck, Austria. E-mail address: [email protected] (R. Dallinger). 1 Both authors contributed equally to this study.

https://doi.org/10.1016/j.scitotenv.2018.07.426 0048-9697/© 2018 Elsevier B.V. All rights reserved.

562

L. Schmielau et al. / Science of the Total Environment 648 (2019) 561–571

common ancestor, the three-domain MT isoform of Pomatias elegans has evolved independently from the threedomain MT of its marine counterpart, probably by addition of a third domain to the pre-existing two-domain MT. Obviously, the occurrence of homologous MT structures in both species is a hereditary character, whereas the differentiation into two distinct MT isoforms with different upregulation capacities in Pomatias elegans is an adaptive feature that probably emerged upon transition to life on land. © 2018 Elsevier B.V. All rights reserved.

1. Introduction Through the history of evolution, marine snails from different clades of Gastropoda have adapted several times independently to life on land (Rosenberg, 1996; Barker, 2001). Occasionally, such transitions may have been fostered in marine ancestors by acquisition of physiological pre-adaptations to intertidal and terrestrial conditions (Little, 1989; Marshall and McQuaid, 1991). Pomatias elegans (the mouth-rounded snail) is a little South and Western European land snail living in moist terrestrial habitats of mainly calcareous soils (Kerney and Cameron, 1979; Pfenninger, 2002). In contrast to most other terrestrial snails, Pomatias elegans belongs to the gastropod class of Caenogastropoda and specifically, to the clade of Littorinoidea (Barker, 2001). It is considered as a land winkle closely related to the North Atlantic littoral periwinkle Littorina littorea, both of them being included into the superfamily of Littorinacea (Jordaens et al., 2001). Littorina littorea is optimally adapted to life in the intertidal zone, being alternately exposed to marine and terrestrial conditions. In contrast, Pomatias elegans and its close relatives of the Pomatiidae family have succeeded to become true terrestrial snails. Their emergence goes probably back to the cretaceous age, when they originated from an ancestral clade that gave rise to both, the modern periwinkles and the terrestrial Pomatiidae (Barker, 2001). Like in other gastropod lineages that made the transition from marine to terrestrial life (Marshall and McQuaid, 1991), specific physiological preadaptations may have facilitated the colonization of terrestrial habitats by ancestral Littorinoidea. Such pre-adaptations may include metabolic depression and high anaerobic capacity during exposure to air and under freezing conditions (Churchill and Storey, 1996; Storey et al., 2013), or increased capacity for osmoregulation (Klekowski, 1963; Taylor and Andrews, 1988). Consistently, adaptations of gastropods to life on land have also required the acquisition of particular adjustments at the biochemical and molecular levels. In Littorina littorea, hypoxic conditions due to the interruption of oxygen uptake by gills during aerial exposure can trigger transcriptional arrest (Larade and Storey, 2007) and regulation of gene expression by involvement of microRNAs (Biggar et al., 2012). At the same time, activation of protective mechanisms may prevent or mitigate adverse effects during intertidal phases. The expression of a metallothionein (MT) in the foot and the midgut gland due to freezing and anoxic impact, for example, may increase the stress resistance of air-exposed periwinkles (English and Storey, 2003). The respective MT has been shown, moreover, to be a Cdselective protein (Palacios et al., 2017). Its three-dimensional structure was recently disclosed by solution NMR, revealing a novel threedomain structure, by which this protein is able to bind 9 Cd2+ ions per MT molecule (Baumann et al., 2017). Hence, apart from possible, other functions (English and Storey, 2003), the MT of Littorina littorea seems to be primarily involved in Cd detoxification. In fact, this metal is strongly accumulated in the snail midgut gland, where the MT gene is readily induced due to metal exposure in a concentration- and time dependent manner (Benito et al., 2017). As a consequence, most of the Cd accumulated in this organ is bound to the expressed MT protein (Bebianno et al., 1992). At the same time, virtually no Cd excretion can be detected, which may be attributed to the extremely low half time of MT degradation in the periwinkle's midgut gland (Bebianno and Langston, 1998). Apparently, the strategy of Littorina littorea with respect to the non-essential trace element Cd is a process of macro-

concentration, a phenomenon that was previously described for some other terrestrial gastropods, too (Dallinger, 1993). Against this background, the central objective of our study was to explore the Cd handling strategy of Pomatias elegans, a close relative of Littorina littorea, but with an exclusive terrestrial lifestyle. First, we were interested in the question whether, after Cd exposure, the toxic metal would be accumulated and retained mainly in the midgut gland of Pomatias elegans in a similar way as previously observed in its marine relative, Littorina littorea. Second, we wanted to explore if the handling and detoxification of Cd in Pomatias elegans is predominantly mediated by MT or MT isoforms, as in Littorina littorea. Thirdly, we wanted to find out whether the MT system of Pomatias elegans was homologous to that of the intertidal periwinkle, in terms of primary and tertiary structure and gene expression capacity in response to environmental Cd exposure. Fourthly, we wanted to understand if potential ecophysiological similarities and/or differences in the functioning of the MT system between the two closely related species could rather be explained by their common ancestry or, instead, by their adaptation to strongly different lifestyles (marine versus terrestrial). Finally, we tried to evaluate the ecotoxicological implications and biomarker potential of the MT system of Pomatias elegans. 2. Material and Methods 2.1. Animals and rearing conditions Individuals of Pomatias elegans were collected in summer 2014 on the island of Brač (Croatia), in a field site near Sumartin (43.2906522N, 16.8740650E). They were kindly provided to our lab by Dr. Lucie Juřičková (Department of Zoology, Charles University Praha, Czech Republic). During the winter season of 2014/15, the snails hibernated buried in dry soil substrate within a covered glass terrarium in our laboratory at room temperature (18–23 °C) and an average ambient air humidity of 36%. The soil substrate consisted of coarse earth, wood particles and small lime stone fragments, covered with a layer of mixed decaying leaf litter to account for food preferences of Pomatias elegans and a sufficient degree of habitat structuration (Triebskorn, 1987; Pfenninger, 2002; De Oliveira et al., 2010). In spring of 2015, snails in the terrarium were awaked and activated by moistening the substrate until reaching a relative air humidity of 95%, along with an increase of temperature to 26 °C. 2.2. Cd exposure Adult, active animals were sampled from the rearing terrarium. They were kept separately in small plastic Petri dishes (Sarstedt, Nürnbrecht, Germany) placed in a box with a layer of wet absorbent paper on the bottom and covered with perforated parafilm above to ensure an air humidity of ±95%. This box was stored in a climate chamber with an adjusted temperature of 26° and a day-night rhythm of 12 h from 6 am to 6 pm. Snails were fed on uncontaminated or Cd-loaded agar plates made up with commercial vegetable juice (“Gemüsesaft erntefrisch”, Alnatura Produktions- und Handesls GmbH, Bickenbach, Germany). 100 ml of food plates were produced by mixing 2 g agar (Agar-Agar Kobe I, pulverized; density 0.55; Carl Roth GmbH & Co. KG, Karlsruhe, Germany) with 0.66 g CaCO3 powder (Merck Chemicals, Vienna, Austria) and

L. Schmielau et al. / Science of the Total Environment 648 (2019) 561–571

50 ml Milli-Q water (Merck). The mixture was boiled shortly until the ingredients were completely dissolved. The vegetable juice was admixed to the hot solution and filled up to a volume of 100 ml. For Cd exposure, 651 or 1302 μl of a Cd standard solution (1000 mg/l) were added to achieve a final nominal Cd concentration in the agar plates of 100 μg/g or 200 μg/g (dry weight) corresponding to effective Cd concentrations of 94.4 ± 22.6 μg/g and 184.8 ± 46.3 μg/g dry weight, (n = 10), respectively. After cooling, circular agar plates with a diameter of 2 cm and a thickness of 3 mm were stamped out. The circular shape and semisolid consistence of agar plates allowed controlling the daily feeding activity and dietary intake of experimental snails. At first, a pilot test with agar plates of both nominal Cd concentrations (100 or 200 μg/g dry wt.) was carried out using 15 individuals each, in order to assess the tolerable Cd concentration of snails over a period of 14 days. Both concentrations tested were sub-lethal, and no mortality of exposed snails was detected. The lower nominal Cd concentration of 100 μg/g was chosen for the following exposure experiment. This concentration is sufficiently low to prevent effects that might affect the snail's activity and on the other hand, 100 μg/g Cd is in a concentration range that may casually also be encountered in heavily contaminated soils in the field, e.g. for mining sites (Ma and Rao, 1997; Nahmani and Rossi, 2003). Hence, in the main experiment one group of 30 snails was fed with agar plates containing a nominal concentration of 100 μg Cd/g (dry weight) over a period of 14 days. Another 25 snails were fed on Cdfree agar plates and served as a control group. Cd concentration in control agar plates and in the components serving as base materials for agar plate production was below detection limits of analysis. Each individual was placed in a single Petri dish (see above) and offered one agar plate (uncontaminated or Cd-loaded) once a day. There was no significant difference in the feeding behavior between Cd-exposed and control animals. As an average, control snails fed 0.418 ± 0.32 g fresh weight (0.276 ± 0.21 g dry weight) of agar plates per day, while Cd-exposed snails consumed 0.416 ± 0.14 g fresh weight per day (0.275 ± 0.01 g dry weight). Of the 30 individuals in the Cd-exposed group, only one snail died, whereas 22 snails (out of 25) survived in the control group. Hence, 4–6 snails (depending on the number of surviving individuals at the respective day) were sampled on days 0, 1, 3, 5, 8 and 14 of exposure. 2.3. Tissue dissection After breaking the shell, snails were sacrificed by decapitation. Three tissue pools were extracted: a) midgut gland (MGG), b) head, foot and mantle tissue (HFM) and c) the visceral mass (VM), containing the remaining soft tissues after removing the midgut gland, foot, mantle and head. The tissue weight was determined wet and after 7 d drying at 70 °C. Two small aliquots of midgut gland per individual were additionally taken for RNA isolation and stored in Ambion's RNAlater® (Fisher Scientific, Vienna, Austria) at −80 °C. 2.4. Sample digestion and metal quantification Dried tissue samples and small agar plate aliquots (~5–30 mg dry wt.) were transferred each into 2 ml Safe Seal Eppendorf reaction tubes (Sarstedt, Nürnbrecht, Germany). After addition of 500 μl of 65% nitric acid (suprapure, Merck, Darmstadt, Germany) and Milli Q water (Millipore, Molsheim, France) (1:1), the samples were digested for 7 days at 70 °C under pressure in a heating aluminum block. Heat pressure was achieved by fixing the closed reaction tubes with a heavy aluminum lid into the wells of the heating block. Manual release of formed digestion gases from heated Eppendorf tubes occurred 5 min, 25 min, 120 min and then every 12 h after starting the digestion by lifting the heavy aluminum lid from the block and carefully opening the tightly closed tubes for a few seconds. Eppendorf tubes with 20–40 mg of oven-dried standard reference material (Tort-2 Lobster Hepatopancreas

563

powder, NRC, Ottawa, Canada) were processed in the same way as the samples. The clear solution remaining after digestion was made up to a volume of 1.5 ml with Milli Q water (Merck). Cd concentrations were assessed by graphite furnace atomic absorption spectrophotometry (Z-8200 Polarized Zeeman Atomic Absorption Spectrophotometer with SSC-300 Auto Sampler; Hitachi, Üdern, Germany), using Pd(NO3) 2 as a matrix modifier and diluted Titrisol standard solutions (Merck) with 5% nitric acid (suprapure, Merck) for calibration. The accuracy of measurement was validated with Tort-2 Lobster Hepatopancreas (see above). The Cd concentrations assessed in the reference samples amounted to 28.39 ± 1.46 μg/g, corresponding to an average Cd recovery of 106.4% of the concentrations verified in the standard reference material. 2.5. RNA isolation and cDNA synthesis After tissue homogenization with a Precellys® homogenizer (Bertin Instruments, Montigny-le-Bretonneux, France), midgut gland RNA from each individual was isolated with the RNeasy Plant Mini Kit (Qiagen, Hilden, Germany). RNA integrity was controlled by electrophoresis, and the RNA concentration measured with the Quant-iT™ RiboGreen® RNA Assay Kit (Life Technologies Corporation, Carlsbad, CA, USA) using the Victor™ X4 2030 Multilable Reader (Perkin Elmer, Waltham, MA, USA). cDNA was synthesized from 250 ng total RNA applied to a volume of 50 μl, using a random hexamer primer and Revert-Aid™ H Minus MMLV reverse transcriptase (Fermentas, St. Leon-Rot, Germany). 2.6. Transcriptome generation and MT sequence blasting For transcriptome generation, isolated RNA from one individual exposed to Cd through 14 days was sent to the Duke Center for Genomic and Computational Biology (GBC, Duke University, Durham, NC, USA). The RNA was subjected to 125 bp paired-end Illumina sequencing. Raw data were assembled at the institute of Zoology of the University Innsbruck using Trinity (GitHub Inc., San Francisco, US) (Grabherr et al., 2011) and provided for analysis on a local TBLAST page. Raw reads will be provided by NCBI Acc. No. SRP156830. cDNA sequences coding for diverse snail MTs and in particular, for the CdMT of the closest related species Littorina littorea (Baumann et al., 2017; Palacios et al., 2017) were blasted against the transcriptome data set of Pomatias elegans using TBlast. The purpose was to find transcriptomic sequences with highest similarities to nucleotide sequences coding for snail MTs and, possibly, for the CdMT of Littorina littorea. Nucleotide sequence hits found in the transcriptome were translated to protein using CLC Main workbench 6.9 (Quiagen, Aarhus, Denmark) and the online alignment tool Clustal Omega (http://www. ebi.ac.uk/Tools/msa/clustalo/), in order to identify respective reading frames and characteristic MT cysteine motifs. 2.7. Confirmation by PCR amplification and Sanger sequencing Since transcriptomic analysis suggested the existence of two different MT mRNAs in Pomatias elegans (encoding for a three-domain PeMT1 and a two-domain PeMT2 isoform, respectively), the appropriate transcriptomic hits were independently confirmed by PCR amplification and sequencing from additional midgut gland RNA extracts of Pomatias elegans. After cDNA synthesis, primers for both MT mRNAs were designed from transcriptomic sequences and applied for amplification as follows. PeMT1 primers, sense: 5′-GTA CTG AGT TTT TCC CGC-3′; antisense: 5′-AGA ACT TTG CTG TTA CTG C-3′. MT2 primers, sense: 5′-CAA AAG ACA GAA ACA GTG C-3′; antisense: 5′-CTG GCA CGT CAT CAG TTC-3′. For amplification, PCR conditions were chosen as follows. Denaturation at 95 °C for 1 min, followed by 30 cycles at 95 °C for 30 s, 52 °C for 30 s, and 68 °C for 40 s; and a final extension step at 70 °C for 10 min. Because of some missing nucleotides in the stop codon, the full length of PeMT1 cDNA sequence was confirmed by

564

L. Schmielau et al. / Science of the Total Environment 648 (2019) 561–571

using the primer, 5′-GTT GTA AGG ATG ACT GCC AGT G-3′, applied in a 3′end RACE protocol as described in the SMARTer™ RACE cDNA Amplification Kit manual (Clontech, Shimogyo-ku, Kyoto, Japan). The amplicon fragments were separated by size on a 1.5% agarose gel by electrophoresis and correct bands were excised for DNA purification with the QIAquick™ Gel Extraction Kit (Qiagen, Hilden, Germany). Subsequent cloning was performed with the pCR™4-TOPO®vector of the TOPO® TA Cloning® Kit for Sequencing (Invitrogen, Thermo Fisher Scientific, Waltham, MA, USA) and TOP10 electrochemical competent E. coli cells. The insertion of the DNA fragments into the plasmids was checked by toothpick PCR and electrophoresis. The plasmids were purified using the Plasmid Mini Preparation Kit (Biozym Scientific GmbH, Hessisch Oldendorf, Germany) and sent for Sanger sequencing to Microysnth AG, Balgach, Switzerland. Both sequences were submitted to GenBank (NCBI) and are available under the accession numbers KY646305.1 (PeMT1) and KY646306.1 (PeMT2). 2.8. Quantitative Real Time PCR For expression studies of Pomatias elegans MT isoform genes (PeMT1 and PeMT2, respectively), quantitative Real Time (qRT) primers were designed by applying the Primer Express 3.0 software (Applied Biosystems, Foster City, CA, USA) as follows. PeMT1: sense, 5′-GAG CCC ATG TGG CTG TAA A-3; antisense, 5′-TGC AAC CAG CAC CGT ATT-3. PeMT2: sense, 5′-GGA TGA CTG CCA GTG CAC AA-3′; antisense, 5′-CAA TTG CAA GTT CCT GCA CAT T-3′. The primer binding and the size of amplicons were checked by PCR in a temperature gradient from 50 to 65 °C and subsequent gel electrophoresis. A primer matrix was generated to find out the optimum primer concentrations (900 nM for all primers). The respective PCR products were used for generation of calibrations curves for PeMT1 (y = −4,0183x + 42,106) and PeMT2 (y = −3,7673x + 41,552) to determine the Δct values for both genes. The DNA concentration of the PCR products was determined with the Quant-iT™ PicoGreen® dsDNA Assay Kit (Thermo Fisher Scientific, Waltham, USA) using the Victor™ X4 2030 Multilable Reader (Perkin Elmer). The expression of both genes from all individuals of the different time points of exposure was measured in triplicates by qRT PCR at the institute of Genetic Epidemiology (Medical University of Innsbruck) with the QuantStudio™ 6 Flex Real-Time PCR System (Applied Biosystems, Thermo Fisher Scientific, Waltham, MA, USA), using Power SYBR Green (Applied Biosystems). A four-stage temperature program was applied as follows. 1), one cycle at 50 °C for 2 min; 2), one cycle at 95 °C for 10 min; 3), 40 cycles with denaturation at 95 °C for 15 s, followed by annealing and extension combined at 60 °C for 1 min; 4), one final cycle at 95 °C for 15 s, 60 °C for 1 min and 95 °C for 15 min. mRNA copy numbers were referred to total RNA (Egg et al., 2014; Höckner et al., 2009). For comparison of mRNA expression of the two genes the differing primer efficiencies (77% for PeMT1 versus 84% for PeMT2) were included by normalizing the respective ct-values to a theoretical PCR efficiency of 100% as described in Pérez et al. (2013). 2.9. Tertiary structure modeling For modeling of tertiary structures of the two MT isoforms of Pomatias elegans (PeMT1 and PeMT2), software packages and bioinformatic tools available from online platforms were utilized. A preliminary structure identity prediction for PeMT1 and PeMT2 was attained using the “Protein Homology/analogY Recognition Engine V2.0” (Phyre2) (http://www.sbg.bio.ic.ac.uk/~phyre2/html/page.cgi?id=index) (Structural Bioinformatics Group, Imperial College, London, UK) (Kelley et al., 2015). After converting primary alignments into PIR format, calculation of 3D-structures of PeMT1 and PeMT2 was achieved using UCSF Chimera 1.12 (Resource for Biocomputing, Visualization, and Informatics) (RBVI, University of California, USA) (graphics and calculation set up) (Pettersen et al., 2004), MODELLER 9.19 (homology modeling) (Webb and Sali, 2014) and Python 2.7.14 (scripting and analysis)

(Departments of Biopharmaceutical Sciences and Pharmaceutical Chemistry, and California Institute for Quantitative Biomedical Research, Mission Bay Byers Hall, University of California San Francisco, USA). The NMR structure with pdb code 5ML1 (Littorina littorea MT, NMR 3D-structure) (Baumann et al., 2017) was downloaded and the first of the 20 NMR conformers was used as a template. The modeling was performed in such a way that the metal and its first coordination sphere was directly taken into account while building the 3Dstructure. Three types of runs were performed to generate the 3D models. An initial attempt used the default configuration of the MODELLER interface from the UCSF Chimera 1.12 package, which does not consider the metal coordination restraints (also in scripts/ model_simple.py). This simulation confirmed that the template structure is suitable for homology modeling, but more refined models could be potentially obtained if the metal coordination was taken into account. This was achieved using the Python 2.7 API provided by the same MODELLER package with two different strategies: (1) by defining rigid blocks in the alignment file to mimic the Cd2+ ions and running the calculation using the automodel routines (scripts/model_blocks.py), and (2) by subclassing the automodel routines to manually define the distance restraints between Cd2+ ions and Cys sulfur (named SG) atoms within 4.0 Å, as suggested in the PDB file ‘REMARK 620’ comments (scripts/model_restrained.py). Quantitative evaluation of the three methods was assessed with the molpdf and DOPE scoring functions built in MODELLER, which account for global satisfaction of restraints, and by measuring the RMSD of the Cd2+ ions coordination spheres. While the Cd-free models (initial attempt) provided better DOPE scores, the coordination spheres were better reproduced in the restrained models, specifically in the manually restrained one. The rigid block automated restrained mode in MODELLER is meant for covalent bonds that, while adequate for some of the coordination interactions, are too short for others. As a result, the automated algorithm, which is based on distance threshold, only considered some coordination bonds. The two approaches that consider coordination interaction in the building of the models lead to the same outcome for the two protein structures of PeMT1 and PeMT2, respectively. 2.10. Statistical analysis The Shapiro-Wilk normality test and the equal variance test failed for most tissue types (p N 0.05). Therefore, non-parametric statistical methods were applied. The data of the Cd measurement and of qRT PCR were tested for outliers with the online Grubbs test (www. graphpad.com; GraphPad Software, San Diego, CA, USA). A two-way Analysis of Variance (ANOVA) was performed in Sigmaplot 12.5 (SYSTAT software, San Jose, CA, USA) applying the Holm-Šídák method for pairwise and multiple comparisons. The overall significance level was set at p ≤ 0.05. Occasionally, differences in significance levels were indicated as “significant” (*) (p ≤ 0.05), “very significant” (**) (p ≤ 0.01) or “extremely significant” (***) (p ≤ 0.001). Regression analysis was applied for calculation of slope differences in mRNA transcription rates between the two genes PeMT1 and PeMT2 from days 0 to 8 of Cd exposure. All charts were designed with Sigmaplot 12.5 (SYSTAT software) and the images were edited with Photoshop CS4. 3. Results and discussion 3.1. The midgut gland of Pomatias elegans as a central organ of Cd accumulation As shown in Fig. 1, Cd accumulation by Pomatias elegans occurred in a tissue-specific manner similar to findings in other mollusc species (Boshoff et al., 2013; Gundacker, 1999; Sokolova et al., 2005). Clearly, the organ reaching the highest accumulation levels was the midgut gland, with an increase from 28.48 μg Cd/g (dry wt.) at the start of the

L. Schmielau et al. / Science of the Total Environment 648 (2019) 561–571

200 Midgut Gland

μg Cd / g dry wt.

A 150

100

***

***

***

50

0

μg Cd / g dry wt.

B

Visceral Mass

150

100

50

*

**

* 0

C

Head / Foot / Mantle

μg Cd / g dry wt.

40 30

***

20

***

***

10

* 0 0

2

4

6

8

10

12

14

Days Fig. 1. Cd concentrations (μg/g dry wt.) in tissue compartments of Pomatias elegans, showing the time course of Cd accumulation through 14 days in the midgut gland (A), the visceral mass (B) and the head/foot/mantle tissue (C) in control snails (curves with empty circle symbols) and of Cd-exposed individuals (curves with black circle symbols). Means and standard deviations are shown (n = 4–6). While all Cd accumulation curves of tissue compartments from metal-exposed snails (A, B and C) were significant upon ANOVA, the single values of Cd exposed tissues at given time points differed from the respective control values with increasing levels of significance, with p ≤ 0.05 (*), p ≤ 0.01 (**), or p ≤ 0.001 (***).

exposure experiment to a maximal concentration of 119.77 μg Cd/g at day 14 (Fig. 1A). In contrast, Cd concentrations in the midgut gland of control snails remained at a persistent lower level (24.68 μg Cd/g dry wt. as an average) throughout the course of the experiment. Through the 14 days of exposure, the Cd accumulation in the midgut gland accounted for 64.49% of the total metal absorbed by the snail. Apart from the midgut gland, the second highest Cd accumulation was assessed in the visceral mass, where the metal concentrations increased from 10.16 μg Cd/g dry wt. in control animals at day one up to 44.39 μg/g in Cd-exposed snails towards the end of the experiment (Fig. 1B). The visceral mass contains a considerable proportion of the stomach and the midgut. Its elevated Cd concentration may, therefore, be the consequence of Cd uptake via the feed through the epithelia of the intestine, as shown, for example, in the terrestrial copse snail, Arianta arbustorum (Berger and Dallinger, 1989). Much lower concentrations of Cd were observed in the head, foot and mantle tissue compartment of Pomatias elegans, where the metal concentrations in Cd-exposed snails increased from 0.50 μg Cd/g dry wt. at day 1 up to 21.56 μg Cd/g at day 14 (Fig. 1C). In contrast, no significant increase of metal concentration was detected in the respective tissues of control snails. A comparison of Cd

565

accumulation patterns between the three tissue compartments reveals, moreover, that Cd concentrations increased steadily over time in the head/foot/mantle compartment (Fig. 1C). In contrast, there seems to be a delayed increase of Cd accumulation in the midgut gland and visceral mass of metal-treated snails (Fig. 1A and B), where Cd concentrations became significantly different from respective control values after five days of exposure at the earliest. A similar pattern of Cd increase with a delayed uptake was also observed in the midgut gland of metal-exposed individuals of Littorina littorea, the close marine relative of Pomatias elegans (Benito et al., 2017). In fact, the Cd accumulation capacity of Littorina littorea has been known for a long time (Marigómez et al., 1990; Nott et al., 1993). Pomatias elegans, although being a terrestrial snail of the Littorinoidea clade shares this huge accumulation capacity for Cd not only with its closely related marine periwinkle, but also with a number of more distantly related snails. In particular, gastropod species from the clade of Stylommatophora, such as the Roman snail (Helix pomatia) or the garden snail (Cornu aspersum) can reach very high Cd concentrations in their midgut gland (Dallinger and Wieser, 1984; Hispard et al., 2008). Even species from the distantly related freshwater clade of Hygrophila, can take up Cd from the water and enrich the metal in their midgut gland, as recently shown for the freshwater snail, Biomphalaria glabrata (Niederwanger et al., 2017). In addition, the gastropod midgut gland retains its central role as a Cd-accumulating organ irrespective of the main uptake pathways through gills as in aquatic and marine snails, or through the alimentary tract or even in aerosolized form through the lungs, as in terrestrial gastropods (Marigómez et al., 2002; Sturba et al., 2018). Thus, it seems that through evolutionary diversification of gastropods, the central role of the midgut gland as a metabolic turntable and a central site of metal accumulation have been conserved. From an evolutionary point of view, this is consistent with the hypothesis that the midgut gland as an absorbing digestive organ has evolved early in mollusc evolution, being shared by all species of Conchifera (shell-bearing molluscs), independent of midgut gland evolution in the shell-less caudofoveate lineage (Haszprunar, 1992). Hence, it is comprehensible that the midgut gland shares a homologous morphology among all gastropod species, consisting of mainly two or three cell types (digestive cells, excretory cells and calcium cells) which form an epithelium that delimits the lumen of digestive tubuli (Sumner, 1965). Overall, the midgut gland cells are characterized by their huge adaptive plasticity (Triebskorn and Köhler, 1992; Marigómez et al., 1993; Porcel et al., 1996; Lobo-da-Cunha, 1999; Taïeb and Vicente, 1999). In response to environmental stimuli, the midgut gland tissue of gastropods can thus shortly change its structure (Cajaraville et al., 1990; Benito et al., 2017). In addition, cell types of the midgut gland can modify their shape and morphology (Manzl et al., 2004; Hamed et al., 2007; Sawasdee et al., 2011). They can be replaced and regenerated within a short-term cycling (Zhu et al., 2011; Benito et al., 2017). This has also been shown for midgut gland cells of both, Pomatias elegans (Triebskorn, 1987) and Littorina littorea (Vega et al., 1989; Zaldibar et al., 2007). In Cd-stressed Littorina littorea, cellular reorganization of midgut gland is accompanied by the transcriptional up-regulation of the LilMT gene. The expressed Cd-specific LilMT protein readily binds the majority of cytoplasmic Cd (Baumann et al., 2017; Palacios et al., 2017), thereby protecting the proliferating midgut gland cells from Cd2+ ions absorbed from the extracellular lumen or leaking out from damaged intracellular structures (Benito et al., 2017). 3.2. Identification of two MT isoform genes in Pomatias elegans Screening by TBLAST of transcriptomic data from Pomatias elegans with nucleotide sequences of the MT from Littorina littorea, combined with confirmation of identified sequences of RNA tissue extracts by PCR, revealed the presence, of two different MT isoform genes, called PeMT1 and PeMT2, in Pomatias elegans. The respective cDNA and the deduced translated amino acid sequences of the two proteins are reported

566

L. Schmielau et al. / Science of the Total Environment 648 (2019) 561–571

in Fig. 2. Interestingly, the two cDNA sequences differed among each other in their chain length (Fig. 2A), suggesting the existence of two MT proteins (PeMT1 and PeMT2) diverging essentially in their molecular size. The translated amino acid sequences of PeMT1 and PeMT2 are shown in Fig. 2B, along with the primary sequence of the recently characterized MT of Littorina littorea (LilMT), the close marine relative of Pomatias elegans (Benito et al., 2017). Indeed, the sequence similarity between the MT of Littorina littorea and the two MT isoforms of Pomatias elegans is evident. PeMT1 in particular, resembles very much the MT of Littorina littorea, since both proteins share, apart from their sequence similarity, a three-domain partition with two homologous N-terminal domains and one distinct C-terminal domain (called “α1” and “α2” or “β”, respectively, in the recently published MT structure of Littorina littorea) (Baumann et al., 2017). Apart from the three-domain partition shared by LilMT and PeMT1, however, the degree of reciprocal intraspecific sequence similarity between PeMT1 and PeMT2 within Pomatias elegans is much higher than their similarity with LilMT of Littorina littorea (compare Fig. 2B and C). In addition, the sequence similarity between the two N-terminal domains (“α1” and “α2”) within each of the two three-domain MTs (LilMT and PeMT1) is higher than the similarity between the respective domains across the two species (see Supplementary Table 1). This suggests that the three-domain partition shared by PeMT1 and LilMT may have evolved in the two proteins independently (Fig. 2D), rather than being inherited from a common three-domain precursor. It is assumed, therefore, that the MTs of Littorina littorea and Pomatias elegans may have originated from a two-domain MT in a marine ancestor. This original MT may have undergone a duplication of the N-terminal domain in Littorina littorea, giving rise to the emergence of the three-domain LilMT. In Pomatias elegans, by contrast, a descendent two- domain MT is still present in the form of PeMT2. We suggest that the three-domain PeMT1 may have arisen after duplication of PeMT2, by addition of a supplementary N-terminal

domain, independent of the domain duplication of LilMT in Littorina littorea (Fig. 2D). Overall, duplications of protein domains are common in animal genomes, considering that about 14% of eukaryotic proteins contain one or more domain tandem repetitions (Marcotte et al., 1998). They occur particularly in proteins composed of independent units or domains contributing autonomously to a particular function such as, for example, ligand binding (Miller et al., 1985; Björklund et al., 2006). In such cases, domain duplications may additively improve the binding capacity of a protein, as recently shown for the LilMT of Littorina littorea, whose Cd binding capacity has been increased from 6 to 9 Cd2+ equivalents per protein molecule through duplication of the N-terminal binding domain (Baumann et al., 2017; Palacios et al., 2017). In Pomatias elegans too, the three-domain PeMT1 is expected to have a higher Cd binding capacity (9 Cd2+ equivalents per mole of protein) compared to its two-domain PeMT2 counterpart with 6 Cd2+ equivalents bound per protein molecule. In many cases, domain duplications may have been achieved by exon shuffling during recombination (Kolkman and Stemmer, 2001; Van Rijk and Bloemendal, 2003; Björklund et al., 2006). In the MT superfamily modular amplification of metal binding domains has so far been reported from a number of species of very different taxa, including ciliates (Gutierrez et al., 2011), unicellular fungi (Espart, 2015), annelids (Schmitt-Wrede et al., 2004), and molluscs (Jenny et al., 2016; Niederwanger et al., 2016; Baumann et al., 2017). Domain duplications are assumed to contribute significantly to the evolution of new proteins by introduction of rapidly evolving protein sequences that may allow faster adaptation to changing environments (Marcotte et al., 1998). It seems that in molluscs, particularly, the MT family is actually right in the middle of evolutionary diversification. This is confirmed by the present study, suggesting that in Pomatias elegans, the emergence of a three-domain MT from an intra-specific allied two-domain protein is a rather recent evolutionary event that

A

PeMT1 5´-ATGTCTACTTCAGGAGCTAATGTAATCTATGGTGCTGGTTGCACAGGCACATGCAAGCAGAGCCCATGTGGCTGTAAAAATTCAGCTGCAGGCTGCCGTTGTAAGGATGACTGCCAGTGCCC AGCCTGCGCAAAATACGGTGCTGGTTGCACAGGAACATGCAAGCAGAGCCCATGTGGCTGTAAAAATTCAGCTGCAGGTTGCGGTTGCAAGGATGACTGCCGGTGCCCTGCGTGCGCAAAATCTT GCAAATGCGGAACATGCAATTGTGGCAAAGGCTGCACTGGACCCAGCAACTGCAAGTGTGATGATGGATGCTCTTGCAAATAG-3´ PeMT2 5´-ATGTCTTCTTCAGGAGCTAATGCAACTGGTGCTGGCTGCACAGAAACATGCAAGGAGAGTCCCTGTGGTTGCAAAAATTCAGCTGCAGGCTGCAAGTGCAAGGATGACTGCCAGTGCACAAC GTGCGCAAAATCTTGCAAATGTGCAGGAACTTGCAATTGTGGCAAAGGCTGCACTGGACCCAACAGCTGCAAATGTGACGGTGGATGCCCTTGCAAATAG-3´

B LilMT PeMT1 PeMT2

C

PeMT1 N1 PeMT1 N2 PeMT2 N1

N1 (“α1”)

N2 (“α2”)

C (“β”)

MSSVF-----GAGCTDVCKQTPCGCATS--GCNCTDDCKCQSC-KYGAGCTDTCKQTPCGCG-S--GCNCKEDCRCQSCSTACKCAAGSCKCGKGCTGPDSCKCDRSCSCK 100 * ***** *** **** * ** * *** * * ******* **** **** * ** ** **** * *** * * ******** **** **** MSTSGANVIYGAGCTGTCKQSPCGCKNSAAGCRCKDDCQCPACAKYGAGCTGTCKQSPCGCKNSAAGCGCKDDCRCPACAKSCKC--GTCNCGKGCTGPSNCKCDDGCSCK 108 * **** ***** *** ************ ******* *** *** ************ **** ** ** MSSSGANAT-GAGCTETCKESPCGCKNSAAGCKCKDDCQCTTCAKS------------------------------------CKCA-GTCNCGKGCTGPNSCKCDGGCPCK 72

MSTSGANVIYGAGCTGTCKQSPCGCKNSAAGCRCKDDCQCPACAK *********************** ***** ****** * **** YGAGCTGTCKQSPCGCKNSAAGCGCKDDCRCPACAK ***** *** ************ ***** * *** MSSSGANAT-GAGCTETCKESPCGCKNSAAGCKCKDDCQCTTCAK

Littorina littorea

D

N1 N

C

Marine Ancestor

N2

C

LilMT

Pomatias elegans

N

C

N1

N2

PeMT2

C

PeMT1

Fig. 2. A: Nucleotide sequences of coding cDNA regions of PeMT1 and PeMT2 from Pomatias elegans, showing start (underlaid in green) and stop codons (underlaid in grey). B: Alignment of the amino acid sequences of LilMT from Littorina littorea (underlaid in blue) and the two MT isoforms PeMT1 and PeMT2 from Pomatias elegans, showing the starter methionine marked in green, and conserved Cys positions marked in pink. Identical positions between single sequences are indicated by star symbols. C: Alignment of N-terminal amino acid sequences between N-terminal domains of PeMT1 (N1) and PeMT2 (N1 and N2). Color marks and star symbols as in B. D: Origin of the three-domain MT system of Littorina littorea (LilMT) and the two- and three-domain MTs of Pomatais elegans (PeMT2 and PeMT1) from a common littorinoid marine ancestor, showing independent evolution of three-domain MT structures in Littorina littorea (LilMT) and Pomatias elegans (PeMT1). Abbreviations: N, N1, N2 … N-terminal domains; C, C-terminal domain. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

L. Schmielau et al. / Science of the Total Environment 648 (2019) 561–571

567

occurred independently from domain duplications in related other littorinoid MTs. 3.3. Tertiary structure models of PeMT1 and PeMT2 The apparent homology between PeMT1 and PeMT2 from Pomatias elegans and LilMT from Littorina littorea suggests that the high similarity at the amino acid sequence level between these MT isoforms (see Fig. 2) may also be reflected by respective concordance of their tertiary structures. This applies especially to the three-domain MTs of the two species. Indeed, a screening of the primary structure of PeMT1 against potentially similar 3D model structures by Phyre2 (Kelley et al., 2015) suggested LilMT as one of the most appropriate templates for structure prediction of PeMT1 (see Supplementary Table 2). The predicted 3D-structure of PeMT1 reveals a three-domain partition of the protein, that closely mimics the template structure of LilMT (Baumann et al., 2017) (Fig. 3A). It appears that each of the domains includes a tightly packed metal cluster with three Cd2+ ions and nine sulfur atoms. Cd2+ ions in animal MTs were shown to be tetrahedrally coordinated by four sulfur atoms (Kägi, 1993). According to the coordination model suggested for the three-metal clusters in the paradigmatic CdMT of the terrestrial gastropod Helix pomatia, this means that each cluster forms a cyclohexane-like ring structure consisting of three Cd2 + ions held together by three bridging and six terminal sulfur atoms (Dallinger et al., 2001). Owing to the primary structure similarity, it is expected that the three-dimensional configuration of PeMT1 may be very similar by analogy to the spatial model of LilMT, even in its structural details. This is confirmed by a superposition of the NMR structure of LilMT with the predicted model structure of PeMT1 (Fig. 3B), suggesting that the two MTs share, apart from their overall shape, even most of their secondary structural elements, including the metal cluster configurations. The absence of an amino acid stretch representing a complete domain in the primary structure of PeMT2 (see Fig. 2) implicates that the three dimensional configuration of this MT must in principle be similar to other animal two-domain MT structures, which mostly exhibit a dumbbell-shaped architecture (Kägi, 1993; Fischer and Davie, 1998; Capdevila et al., 2012). Also in this case, the most accurate structure was obtained using spatial restraints between the metal and the predicted metal-coordinating amino acid. (Fig. 3C). Other examples for this kind of configuration were provided by structures of rabbit MT (Davis et al., 1998), mouse MT-1 (Zangger et al., 1999) and rat MT-2 (Dolderer et al., 2007), although all these MTs contain one additional metal ion, all of them being unequally distributed between the two metal clusters (four and three divalent metal ions in the clusters of domains α and β, respectively). A two domain structure was also disclosed for the MT-1 of the blue crab, Callinectes sapidus, which in contrast to mammalian MTs is built of two clusters with three divalent metal ions and nine Cys sulfur atoms each (Narula et al., 1995), thus equivalent in their sulfur/metal stoichiometry to that observed in the two domains of PeMT2. So, although the model of the PeMT2 structure resembles the known dumbbell-shaped organization of vertebrate MTs (Fig. 3C), its sulfur/metal stoichiometry is the same as that observed in the two domains of MT-1 from Callinectes sapidus (Narula et al., 1995). 3.4. Transcriptional upregulation of PeMT1 and PeMT2 genes upon Cd exposure In most Cd-specific gastropod MTs studied so far, Cd binding specificity at the protein level is functionally complemented by a distinct responsiveness at the molecular level, reflected by transcriptional upregulation of the respective genes upon Cd exposure (Dallinger et al., 2004a; Höckner et al., 2011; Palacios et al., 2011; Pedrini-Martha et al., 2016). This is also true for PeMT1 and PeMT2 from the midgut gland of Pomatias elegans (Fig. 4), although the levels and patterns of transcriptional upregulation during exposure differ between the two

Fig. 3. 3D-structure models of PeMT1 and PeMT2 from Pomatias elegans and comparison of structures with LilMT from Littorina littorea. A: 3D model of PeMT1 of Pomatias elegans with secondary structure elements, showing the three domains with their respective metal clusters, all of them built of three Cd2+ ions (large yellow spheres) and nine sulfur atoms (yellow sticks) for metal coordination each. B: 3D structure superposition of PeMT1 of Pomatias elegans (yellow) and LilMT of Littorina littorea (blue), showing a very high degree of congruence. Also shown are the Cd2+ ions (yellow spheres) and their coordination geometry (dashed lines), along with metal binding sulfur atoms (yellow sticks). C: 3D model of PeMT2 with secondary structure elements, showing two domains with their respective metal clusters, both of them built of three Cd2+ ions (yellow spheres) and nine sulfur atoms (yellow sticks) for metal coordination. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

A

2.0x107 PeMT1 1.5x107

107

**

5.0x106

1.6x106 1.4x106 1.2x106

PeMT1

106

PeMT2

8.0x10

5

6.0x105 4.0x105 2.0x105 0

1

3 5 Days of exposure

8

14

B

2.0x107

***

PeMT2 1.5x107

***

* 107

5.0x106

2.0x10 7 1.8x10 7 1.6x10 7 1.4x10 7 1.2x10 7

PeM T2

10 7 8.0x10 6 6.0x10 6 4.0x10 6

PeM T1

2.0x10 6 0 0

0 0

1

3 5 Days of exposure

8

14

Fig. 4. Quantitative Real-time PCR of PeMT1 and PeMT2 from Pomatias elegans during a feeding experiment through 14 days. A: Time course of mRNA transcription (copy numbers/10 ng of total RNA) of the three-domain MT gene (PeMT1) showing control values (black circle symbols) versus values of Cd-exposed snails (empty circle symbols). B: Time course of mRNA transcription (copy numbers/10 ng of total RNA) of the two domain MT gene (PeMT2) showing control values (black circle symbols) versus values of Cd-exposed snails (empty circle symbols). Single mean values (n = 3 to 5) and standard deviations (bars) are shown. Significant differences between single values of each exposure day (Holm-Šídák method for pairwise and multiple comparisons) are indicated with star symbols, symbolizing “significant” (*) (p ≤ 0.05), “very significant” (**) (p ≤ 0.01) or “extremely significant” (***) (p ≤ 0.001) differences.

genes. As seen in Fig. 4A, the transcription of PeMT1 in the snail midgut gland started to increase only after a lag time of three days, similar to the time pattern observed for Cd accumulation in the midgut gland and visceral mass of Cd-exposed snails (see Fig. 1). From a basal expression level of about 600,000 copies/10 ng RNA at day 0, the expression of the PeMT1 gene reached its maximal value on day 8 of exposure, with about 1.7 million copies/10 ng total RNA (Fig. 4A). A possible explanation for the delayed upregulation of PeMT1 in the midgut gland of Cdexposed Pomatias elegans could be that it may take some time for the Cd-contaminated feed to reach the midgut gland via the gut and the midgut gland tubules, before being absorbed by the PeMT1-expressing cells in the midgut gland epithelium. The PeMT2 transcription increased steadily from a basal expression of about 400,000 copies/10 ng total RNA at day 0 up to about 12 million copies/10 ng total RNA at day 8 of Cd exposure, remaining at this high level until the end of the metal treatment. The response of both genes to Cd exposure is apparently of transient nature, as suggested by the fact that their transcriptional upregulation reached its maximal peak on day 8 of exposure, with a slight decrease after this time point until the end of the exposure. The constitutive expression of both, PeMT1 and PeMT2 is high, but does not significantly differ between the two genes (Fig. 5A). However, there is a significant difference in the mRNA upregulation intensity between PeMT1 and PeMT2 upon Cd exposure (see Fig. 4A and B). As

1

2

3

4

5

6

7

8

Days of exposure 18

C

16

PeMT1

PeMT2

14

-fold induction

Copy numbers / 10 ng total RNA

***

1.8x106

0 0

B

***

Copy numbers / 10 ng total RNA

Copy numbers / 10 ng total RNA

A

L. Schmielau et al. / Science of the Total Environment 648 (2019) 561–571

Copy numbers / 10 ng total RNA

568

12 10 8 6 4 2 0

0

1

3

5

8 14

0

1

3

5

8 14

Days of exposure Fig. 5. -Fold induction figures of PeMT1 and PeMT2 of Pomatias elegans. A: Constitutive expression of the PeMT1 and PeMT2 genes (Copy numbers/10 ng total RNA), representing for each of the two genes the mean values of all single days of the experiment (black bar: PeMT1; grey bar: PeMT2) with standard deviations. B: Regression slopes of mRNA transcription between PeMT1 (black circle symbols) and PeMT2 (empty circle symbols) during the first 8 days of Cd exposure, showing regression lines (full lines) with 95% confidence limits (dashed lines). For regression equation parameters and statistics, see Supplementary Table 2. C: -fold transcriptional induction of PeMT1 (blue bars) and PeMT2 (green bars) from days 0 to 14 of Cd exposure, compared to the respective basal expression values, normalized to 1 (black bars). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

shown in Fig. 5B, the time course-dependent regression slopes of incremental mRNA transcription differed significantly between the two genes until day 8 of exposure, with a much flatter rise of mRNA transcription for PeMT1 in contrast to PeMT2. Consequently, the two genes also differed with respect to their transcriptional upregulation towards Cd referred to control levels, with a maximal 16-fold induction of PeMT2

L. Schmielau et al. / Science of the Total Environment 648 (2019) 561–571

upon day 8 of exposure, compared with an 6-fold induction of PeMT1 (Fig. 5C). Interestingly, the key figures of Cd-dependent transcription dynamics of PeMT1 from Pomatias elegans are comparable to those of LilMT from Littorina littorea. In both species, the gene of the three-domain MT variant responds to Cd exposure with some delay and with a relatively weak intensity of upregulation, reaching an only 3.5-fold induction in Littorina littorea (Benito et al., 2017) and a 6-fold induction in Pomatias elegans (see Fig. 5C). 3.5. A functional interpretation of the MT system of Pomatias elegans: hereditary versus adaptive? An obvious difference in the MT system between Littorina littorea and Pomatias elegans is the diversification in the latter species towards two distinct Cd-responsive MT isoform genes with two distinct gene products (PeMT1 and PeMT2) that differ by their divergent domain number and tertiary structure (Fig. 3). This supports the assumption that either of them may have, in spite of their common tasks in Cd binding, a distinct ecophysiological and toxicological significance. The apparent advantage of the three-domain PeMT1 is its increased loading capacity upon Cd binding (9 Cd2+ equivalents per protein molecule) compared to the lower metal loading potential of PeMT2 (6 Cd2+ equivalents per protein molecule). In this respect, PeMT1 resembles very much the three-domain MT of its marine relative, Littorina littorea, sharing with the latter protein not only the overall 3D structure (Fig. 3B), but in addition a high similarity at the primary structure level (Fig. 2B) (Baumann et al., 2017; Benito et al., 2017). A high degree of amino acid sequence homology with LilMT and PeMT1 is also observed for PeMT2, irrespective of its differing 3D structure owing to the missing third domain (Fig. 3C). This strongly suggests that the underlying twodomain consensus sequence (consisting of one N-terminal and one Cterminal domain) contained in of all three MTs may have been inherited by Littorina littorea and Pomatias elegans from a common marine littorinoid ancestor (see Fig. 2D). Upon adaptation to their specific habitats, both species may independently have evolved a three-domain MT (PeMT1 in Pomatias elegans and LilMT in Littorina littorea), by addition of a second N-terminal domain, enhancing in this way the Cd-binding potential (Palacios et al., 2017) of their cognate MT proteins. The huge Cd accumulation capacity in the midgut gland of both species may be a consequence of this evolution (Fig. 1A) (Benito et al., 2017). At the same time, the genes of PeMT1 and PeMT2 have conserved their responsiveness towards Cd exposure (see Fig. 5), in a similar manner as observed for LilMT in Littorina littorea (Benito et al., 2017). Overall, this clearly demonstrates that after transition to terrestrial life, the inherited MT proteins of Pomatias elegans have maintained their original Cd specificity, binding stoichiometry and detoxifying function under completely changed environmental conditions. In spite of sharing their common responsiveness to Cd exposure, however, PeMT1 and PeMT2 significantly differ in their response intensity towards the metal stressor. PeMT2 is quicker and more sensitive in its response towards Cd stress, with transcriptional upregulation rates more than twice as high than those of PeMT1 upon metal exposure (Fig. 5). An obvious interpretation of this may be that the more responsive PeMT2 gene is activated as a quick and acute response mechanism upon rapidly changing conditions, when Cd2+ ions enter the midgut gland cells owing to external exposure. This suggests that this gene may be particularly important as a regulatory protective mechanism against Cd insults during activity phases of Pomatias elegans, in order to prevent Cd impairment induced by ingestion. In contrast, the slower reacting PeMT1 may be more significant in situations of a prolonged Cd impact. This may be the case, for instance, upon intracellular mobilization of Cd stores from impaired cellular structures – such as leaking lysosomes – due to re-organization of the midgut gland cells and tissue under metal stress or other stressful conditions, as demonstrated previously for Littorina littorea (Cajaraville et al., 1990; Marigómez et al.,

569

1993; Zaldibar et al., 2007; Benito et al., 2017). This is consistent with the fact that the expressed PeMT1 protein possesses, in compensation for the slower response ability of its encoding gene, a higher metal loading capacity due to one additional Cd binding domain. As such, the expression of PeMT1 may become increasingly important upon chronic Cd exposure, remaining available for long-term Cd buffering. Thus, in contrast to PeMT1, the buffering capacity of PeMT2 may be more important under conditions of low metabolic activity, perhaps during aestivation phases of Pomatias elegans, in order to overcome adverse environmental conditions. This is supported by the fact that the homologous LilMT gene of Littorina littorea can be induced not only by Cd exposure (Benito et al., 2017), but also due to non-metallic environmental stressors (English and Storey, 2003). Overall, the clear diversification into two differently responsive genes with two gene products possessing divergent Cd loading capacities may represent an adaptive feature of the MT system in Pomatias elegans that has emerged and proofed successful upon transition of littorinoid snails from the sea to a true terrestrial life style.

3.6. Ecotoxicological implications and biomarker relevance As a small soil-dwelling gastropod with a main distribution area in Western Mediterranean and Central European countries, Pomatias elegans prefers calcareous habitats with a thick layer of well aerated organic substrate (Kerney and Cameron, 1979; Pfenninger, 2002). In order to become active, the snails require a high degree of relative air humidity, normally above 90% (Platts et al., 2003). During activity, snails seem to feed intermittently, with intensive feeding phases of a few days alternating with days of starvation (Triebskorn, 1987). It was shown previously that cyclic phases of activity may contribute significantly to fluctuating expression patterns of MT genes (Pedrini-Martha et al., 2016). Thus, it is conceivable to assume that the quicker responsive PeMT2 gene may be particularly important as a regulatory Cdresponsive protective mechanism during alternating phases of activity. In the field, Pomatias elegans plays an important role in substrate decomposition, apparently in a cooperative manner with other soildwelling invertebrates (De Oliveira et al., 2010). Owing to its particular features and preferences, Pomatias elegans often exhibits a patchy distribution, being split up onto small and often shrinking habitats with little mutual gene flow among isolated snail populations (Jordaens et al., 2001). It remains questionable whether the protective function of the MT system of Pomatias elegans would compensate for these threats in metal-contaminated habitats. Against this background, Cd exposure in the field may pose a particular risk to the patchy distributed Pomatias elegans populations, especially in combination with additional, nonmetallic environmental stressors. On the other hand, differentially expressed MT genes within one species may – as in Pomatias elegans - represent a particularly appropriate biomarker tool to be applied for purposes of biomonitoring in terrestrial field sites, as already proposed for other terrestrial snail species (Dallinger et al., 2004a, 2004b). In Pomatias elegans, for example, PeMT2 could serve as a biomarker tool for rather short-term, acute impacts due to Cd exposure, whereas the slower and less intensively responding PeMT1 gene could perhaps be more relevant as a biomarker for chronic exposure and/or periods of snail aestivation or inactivity. A still open question is, whether PeMT1 may also respond to nonmetallic stressors, as shown for its cognate three-domain LilMT gene from Littorina littorea (English and Storey, 2003). The real value of the potential dual biomarker system of Pomatias elegans may lie, however, in a more advanced concept by applying the induction of PeMT1 and PeMT2 within a biomarker battery (Kammenga et al., 2000) for ecotoxicological evaluation and risk assessment of metal-contaminated soils in specific habitats colonized by this particular species. Supplementary data to this article can be found online at https://doi. org/10.1016/j.scitotenv.2018.07.426.

570

L. Schmielau et al. / Science of the Total Environment 648 (2019) 561–571

Acknowledgements This study was supported by a grant of the Austrian Science Foundation (FWF, DACH proj. No. I 1482-N28) to R.D. References Barker, G.M., 2001. Gastropods on land: phylogeny, diversity and adaptive morphology. In: Barker, G. (Ed.), The Biology of Terrestrial Molluscs. CAB International, Wallingford (UK), pp. 1–146. Baumann, C., Beil, A., Jurt, S., Niederwanger, M., Palacios, Ò., Capdevila, M., Atrian, S., Dallinger, R., Zerbe, O., 2017. Structural adaptation of a protein to increased metal stress: NMR structure of a marine snail metallothionein with an additional domain. Angew. Chem. Int. Ed. 2017 (56), 4617–4622. Bebianno, M.J., Langston, W.J., 1998. Cadmium and metallothionein turnover in different tissues of the gastropod Littorina littorea. Talanta 46 (2), 301–313. Bebianno, M.J., Langston, W.J., Simkiss, K., 1992. Metallothionein induction in Littorina littorea (Mollusca: Prosobranchia) on exposure to cadmium. J. Mar. Biol. Assoc. UK 72 (2), 329–342. Benito, D., Niederwanger, M., Izagirre, U., Dallinger, R., Soto, M., 2017. Successive onset of molecular, cellular and tissue-specific responses in midgut gland of Littorina littorea exposed to sub-lethal cadmium concentrations. Int. J. Mol. Sci. 18 (1815), 1–26. https://doi.org/10.3390/ijms18081815 2017. Berger, B., Dallinger, R., 1989. Accumulation of cadmium and copper by the terrestrial snail Arianta arbustorum L.: kinetics and budgets. Oecologia 79, 60–65. Biggar, K.K., Kornfeld, S.F., Maistrovski, Y., Storey, K.B., 2012. MicroRNA Regulation in extreme environments: differential expression of microRNAs in the intertidal snail Littorina littorea during extended periods of freezing and anoxia. Genom. Proteom. Bioinform. 10, 302–309. Björklund, Å.K., Ekman, D., Elofsson, A., 2006. Expansion of protein domain repeats. PLoS Comput. Biol. 2 (8), 2259–2269. Boshoff, M., Jordaens, K., Backeljau, T., Lettens, S., Tack, F., Vandecasteele, B., De Jonge, M., Bervoets, L., 2013. Organ- and species-specific accumulation of metals in two land snail species (Gastropoda, Pulmonata). Sci. Total Environ. 449, 470–481. Cajaraville, M.P., Marigómez, J.A., Angulo, E., 1990. Short-term toxic effects of 1naphthol on the digestive gland-gonad complex of the marine prosobranch Littorina littorea (L): a light microscopic study. Arch. Environ. Contam. Toxicol. 19, 17–24. Capdevila, M., Bofill, R., Palacios, Ò., Atrian, S., 2012. State-of-the-art of metallothioneins at the beginning of the 21st century. Coord. Chem. Rev. 256, 46–62. Churchill, T.A., Storey, K.B., 1996. Metabolic responses to freezing and anoxia by the periwinkle Littorina littorea. J. Therm. Biol. 21 (1), 57–63. Dallinger, R., 1993. Strategies of metal detoxification in terrestrial invertebrates. In: Dallinger, R., Rainbow, P.S. (Eds.), Ecotoxicology of Metals in Invertebrates. Lewis Publishers, CRC Press, Boca Raton, USA, pp. 245–289. Dallinger, R., Wieser, W., 1984. Patterns of accumulation, distribution and liberation of Zn, Cu, Cd and Pb in different organs of the land snail Helix pomatia L. Comp. Biochem. Physiol. 79C (1), 117–124. Dallinger, R., Wang, Y., Berger, B., Mackay, E.A., Kägi, J.H.R., 2001. Spectroscopic characterization of metallothionein from the terrestrial snail, Helix pomatia. Eur. J. Biochem. 268, 4126–4133. Dallinger, R., Chabicovsky, M., Berger, B., 2004a. Isoform-specific quantification of metallothionein in the terrestrial gastropod Helix pomatia. I. Molecular, biochemical, and methodical background. Environ. Toxicol. Chem. 23 (4), 890–901. Dallinger, R., Chabicovsky, M., Lagg, B., Schipflinger, R., Weirich, H.G., Berger, B., 2004b. Isoform-specific quantification of metallothionein in the terrestrial gastropod Helix pomatia. II. A differential biomarker approach under laboratory and field conditions. Environ. Toxicol. Chem. 23 (4), 902–910. Davis, J.J., Hill, H.A.O., Kurz, A., Jacob, C., Maret, W., Vallee, B.L., 1998. A scanning tunneling microscopy study of rabbit metallothionein. Phys. Chem. Comm. 1, 12–22. De Oliveira, T., Hättenschwiler, S., Handa, I.T., 2010. Snail and millipede complementarity in decomposing Mediterranean forest leaf litter mixtures. Funct. Ecol. 24 (4), 937–946. Dolderer, B., Echner, H., Beck, A., Hartmann, H.-J., Weser, U., Luchinat, C., Del Bianco, C., 2007. Coordination of three and four Cu(I) to the α- and β-domain of vertebrate Zn-metallothionein-1, respectively, induces significant structural changes. FEBS J. 274, 2349–2382. Egg, M., Paulitsch, M., Ennemoser, Y., Wüstenhagen, A., Schwerte, T., Sandbichler, A.M., Fiechtner, B., Köblitz, L., Prem, C., Pelster, B., 2014. Chronodisruption increases cardiovascular risk in zebrafish via reduced clearance of senescent erythrocytes. Chronobiol. Int. 31, 680–689. English, T.E., Storey, K.B., 2003. Freezing and anoxia stresses induce expression of metallothionein in the foot muscle and hepatopancreas of the marine gastropod Littorina littorea. J. Exp. Biol. 206, 2517–2524. Espart, H.A., 2015. Molecular evolution of unicellular eurkaryote metallothioneins: tandem repetition of coordinating domains. PhD thesis. University of Barcelona (281 pp.). Fischer, E.H., Davie, E.W., 1998. Recent excitement regarding metallothionein. Proc. Natl. Acad. Sci. U. S. A. 95, 3333–3334. Grabherr, M.G., Haas, B.J., Yassour, M., Levin, J.Z., Thompson, D.A., Amit, I., Adiconis, X., Fan, L., Raychowdhury, R., Zeng, Q., Chen, Z., Mauceli, E., Hacohen, N., Gnirke, A., Rhind, N., di Palma, F., Birren, B.W., Nusbaum, C., Lindblad-Toh, K., Friedman, N., Regev, A., 2011. Full-length transcriptome assembly from RNA-seq data without a reference genome. Nat. Biotechnol. 29 (7), 644–652.

Gundacker, C., 1999. Tissue-specific heavy metal (Cd, Pb, Cu, Zn) deposition in a natural population of the zebra mussel Dreissena polymorpha Pallas. Chemosphere 38, 3339–3356. Gutierrez, J.C., Amaro, F., Díaz, S., de Francisco, P., Cubas, L.L., Martín-Gonzàlez, A., 2011. Ciliate metallothioneins: unique microbial eukaryotic heavy-metal-binder molecules. J. Biol. Inorg. Chem. 16, 1025–1034. Hamed, S.S., Abdelmeguied, N.E., Essawy, A.E., Radwan, M.A., Hegazy, A.E., 2007. Histological and ultrastructural changes induced by two carbamate mollucicides on the digestive gland of Eobania vermiculata. J. Biol. Sci. 7 (6), 1017–1037. Haszprunar, G., 1992. The first molluscs – small animals. Bolletino di zoologia 59 (1), 1–16. Hispard, F., Schuler, D., De Vaufleury, A., Scheifler, R., Badot, P.-M., Dallinger, R., 2008. Metal distribution and metallothionein induction after cadmium exposure in the terrestrial snail Helix aspersa (Gastropoda, Pulmonata). Environ. Toxicol. Chem. 27 (7), 1533–1542. Höckner, M., Stefanon, K., Schuler, D., Fantur, R., De Vaufleury, A., Dallinger, R., 2009. Coping with cadmium exposure in various ways: the two helicid snails Helix pomatia and Cantareus aspersus share the metal transcription factor-2, but differ in promoter organization and transcription of their Cd-Metallothionein genes. J. Exp. Zool. 311A, 776–787. Höckner, M., Stefanon, K., de Vaufleury, A., Monteiro, F., Pérez-Rafael, S., Palacios, Ò., Capdevila, M., Strian, S., Dallinger, R., 2011. Physiological relevance and contribution to metal balance of specific and non-specific metallothionein isoforms in the garden snail, Cantareus aspersus. Biometals 24, 1079–1092. Jenny, M., Payton, S.L., Baltzegar, D.A., Lozier, J.D., 2016. Phylogenetic analysis of molluscan metallothioneins: evolutionary insight from Crassostrea virginica. J. Mol. Evol. 83 (3–4), 110–125. Jordaens, K., Platts, E., Backeljau, T., 2001. Genetic and morphological variation in the land winkle Pomatias elegans (Müller) (Caenogastropoda: Pomatiasidae). J. Molluscan Stud. 67 (2), 145–152. Kägi, J.H.R., 1993. Evolution, structure and chemical activity of class I metallothioneins: an overview. In: Suzuki, K.T., Imura, N., Kimura, M. (Eds.), Metallothionein III. Birkhäuser Verlag, Basel, Switzerland, pp. 29–55. Kammenga, J.E., Dallinger, R., Donker, M.H., Köhler, H.-R., Simonsen, V., Triebskorn, R., Weeks, J.M., 2000. Biomarkers in terrestrial invertebrates for ecotoxicological soil risk assessment. Rev. Environ. Contam. Toxicol. 164, 93–147. Kelley, L.A., Mezulis, S., Yates, C.M., Wass, M.N., Sternberg, M.J.E., 2015. The Phyre2 web portal for protein modeling, prediction and analysis. Nat. Protoc. 10 (6), 845–858. Kerney, M.P., Cameron, R.A., 1979. Field Guide to the Land Snails of Britain and Northwest Europe. Collins, London (288 pp.). Klekowski, R.Z., 1963. The influence of low salinity and desiccation on the survival, osmoregulation and water balance of Littorina littorea (L.) (Prosobranchia). Pol. Arch. Hydrobiol. 11, 241–250. Kolkman, J.A., Stemmer, W.P.C., 2001. Directed evolution of proteins by exon shuffling. Nat. Biotechnol. 19, 423–428. Larade, K., Storey, K.B., 2007. Arrest of transcription following anoxic exposure in a marine mollusc. Mol. Cell. Biochem. 303, 243–249. Little, C., 1989. Comparative physiology as a tool for investigating the evolutionary routes of animals on to land. Earth Environ. Sci. Transact. Roy. Soc. Edinburgh 80 (3–4), 201–208. Lobo-da-Cunha, A., 1999. Ultrastructural and cytochemical aspects of the basophilic cells in the hepatopancreas of Aplysia depilans (Mollusca, Opisthobranchia). Tissue Cell 31 (1), 8–16. Ma, L.Q., Rao, G.N., 1997. Chemical fractionation of cadmium, copper, nickel, and zinc in contaminated soils. J. Environ. Qual. 26, 259–264. Manzl, C., Krumschnabel, G., Schwarzbaum, P.J., Dallinger, R., 2004. Acute toxicity of cadmium and copper in hepatopancreas cells from the Roman snail (Helix pomatia). Comp. Biochem. Physiol. C 138, 45–52. Marcotte, E.M., Pellegrini, M., Yeates, T.O., Eisenberg, D., 1998. A census of protein repeats. J. Mol. Biol. 293, 151–160. Marigómez, I.A., Cajaraville, M.P., Angulo, E., 1990. Cellular cadmium distribution in the common winkle, Littorina littorea (L.) determined by X-ray microprobe analysis and histochemistry. Histochemistry 94, 191–199. Marigómez, I., Soto, M., Etxeberria, M., Angulo, E., 1993. Effects of size, sex, reproduction and trematode infection on the quantitative structure of digestive tubules in stressed winkles. In: Fischer, G. (Ed.), Zoologische Jahrbücher. Abteilung für Anatomie und Ontogenie der Tiere. 123, pp. 319–336 Jena (Germany). Marigómez, I., Soto, M., Cajaraville, M.P., Angulo, E., Giamberini, L., 2002. Cellular and subcellular distribution of metals in molluscs. Microsc. Res. Tech. 56, 358–392. Marshall, D.J., McQuaid, C.D., 1991. Metabolic rate depression in a marine pulmonate snail: pre-adaptation for a terrestrial existence? Oecologia 88 (2), 274–276. Miller, J., McLachlan, A.D., Klug, A., 1985. Repetitive zinc-binding domains in the protein transcription factor IIIA from Xenopus oocytes. EMBO J. 4 (6), 1609–1614. Nahmani, J., Rossi, J.P., 2003. Soil macroinvertebrates as indicators of pollution by heavy metals. Comptes Rendus - Biol. 326, 295–303. Narula, S.S., Brouwer, M., Hua, Y., Armitage, I.M., 1995. Three-dimensional solution structure of Callinectes sapidus metallothionein-1 determined by homonuclear magnetic resonance spectroscopy. Biochemistry 34, 620–631. Niederwanger, M., Bacher, K., Dvorak, M., Schnegg, R., Dallinger, R., 2016. Gene structure, protein sequence and function of a novel multi-domain metallothionein from the freshwater snail Biomphalaria glabrata. 30th ESCPB Congress: Unraveling Complexity: From Molecules to Ecosystems, p. 66 (Barcelona, 2016). Abstract book. Niederwanger, M., Dvorak, M., Schnegg, R., Pedrini-Martha, V., Bacher, K., Bidoli, M., Dallinger, R., 2017. Challenging the metallothionein (MT) gene of Biomphalaria glabrata: unexpected response patterns due to cadmium exposure and temperature stress. Int. J. Mol. Sci. 18 (1747), 1–15. https://doi.org/10.3390/ijms18081747.

L. Schmielau et al. / Science of the Total Environment 648 (2019) 561–571 Nott, J.A., Bebianno, M.J., Langston, W.J., Ryan, K.P., 1993. Cadmium in the gastropod Littorina littorea. J. Mar. Biol. Assoc. U. K. 73, 655–665. Palacios, Ò., Pagani, A., Pérez-Rafael, S., Egg, M., Höckner, M., Brandstätter, A., Capdevila, M., Atrian, S., Dallinger, R., 2011. Shaping mechanisms of metal specificity in a family of metazoan metallothioneins: evolutionary differentiation of mollusc metallothioneins. BMC Biol. 9 (4), 1–20 2011. Palacios, Ò., Jiménez-Marti, E., Niederwanger, M., Gil-Moreno, S., Zerbe, O., Atrian, S., Dallinger, R., Capdevila, M., 2017. Analysis of metal-binding features of the wild type and two domain-truncated mutant variants of Littorina littorea Metallothionein reveals its Cd-specific character. Int. J. Mol. Sci. 2017 (18), 1–16. Pedrini-Martha, M., Niederwanger, M., Kopp, R., Schnegg, R., Dallinger, R., 2016. Physiological, diurnal and stress-related variability of cadmium-metallothionein gene expression in land snails. PLoS ONE 11 (3), 1–19. https://doi.org/10.1371/journal. pone.0150442 e0150442. Pérez, L.M., Fittipaldi, M., Adrados, B., Morató, J., Codony, F., 2013. Error estimation in environmental DNA targets quantification due to PCR efficiencies differences between real samples and standards. Folia Microbiol. (Praha) 58, 657–662. https://doi.org/ 10.1007/s12223-013-0255-5. Pettersen, E.F., Goddard, T.D., Huang, C.C., Couch, G.S., Greenblatt, D.M., Meng, E.C., Ferrin, T.E., 2004. UCSF Chimera - a visualization system for exploratory research and analysis. J. Comput. Chem. 25 (13), 1605–1612. Pfenninger, M., 2002. Relationship between microspatial population genetic structure and habitat heterogeneity in Pomatias elegans (O.F. Müller 1774) (Caenogastropoda, Pomatiasidae). Biol. J. Linn. Soc. 76, 565–575. Platts, E., Bailey, S., McGrath, D., McGeough, G., 2003. A survey of the land winkle Pomatias elegans (Müller, 1774) in the Burren. Co. Clare. Proc. Roy. Irish Acad. 103 (3), 197–201. Porcel, D., Bueno, J.D., Almendros, A., 1996. Alterations in the digestive gland and shell of the snail Helix aspersa Müller (Gastropoda, Pulmonata) after prolonged starvation. Comp. Biochem. Physiol. 115A (1), 11–17. Rosenberg, G., 1996. Independent evolution of terrestriality in Atlantic truncatellid gastropods. Evolution 50 (2), 682–693. Sawasdee, B., Köhler, H.-R., Triebskorn, R., 2011. Histopathological effects of copper and lithium in the ramshorn snail, Marisa cornuarietis (Gastropoda, Prosobranchia). Chemosphere 85 (6), 1033–1039. Schmitt-Wrede, H.P., Koewius, H., Tschuschke, S., Greven, H., Wunderlich, F., 2004. Genomic organization of the cadmium-inducible tandem repeat 25-kDa metallothionein of the oligochaete worm Enchytraeus buchholzi. Biochim. Biophys. Acta 1680, 24–33.

571

Sokolova, I.M., Ringwood, A.H., Johnson, C., 2005. Tissue-specific accumulation of cadmium in subcellular compartments of eastern oysters Crassostrea virginica Gmelin (Bivalvia: Ostreidae). Aquat. Toxicol. 74, 218–228. Storey, K.B., Lant, B., Anozie, O.O., Storey, J.M., 2013. Metabolic mechanisms for anoxia tolerance and freezing survival in the intertidal gastropod, Littorina littorea. Comp. Biochem. Physiol. A Mol. Integr. Physiol. 165 (4), 448–459. Sturba, L., Liberatori, G., Vannuccini, M.L., Ancora, S., Corsi, I., 2018. Uptake and biological responses in land snail Cornu aspersum exposed to vaporized CdCl2. Ecotoxicol. Environ. Saf. 148, 377–383. Sumner, A.T., 1965. The cytology and histochemistry of the digestive gland cells of Helix. J. Microsc. Sci. 106, 173–192. Taïeb, N., Vicente, N., 1999. Histochemistry and ultrastructure of the crypt cells in the digestive gland of Aplysia punctata (Cuvier, 1803). J. Molluscan Stud. 65, 385–398. Taylor, P.M., Andrews, E.B., 1988. Osmoregulation in the intertidal gastropod Littorina littorea. J. Exp. Mar. Biol. Ecol. 122, 35–46. Triebskorn, R., 1987. Pomatias elegans: Zur Ernährungsbiologie einer prosobranchen Landschnecke. Zulassungsarbeit zur ersten Staatsprüfung für das Lehramt an Gymnasien. Diploma thesis. University of Heidelberg (117 pp.). Triebskorn, R., Köhler, H.-R., 1992. Plasticity of the endoplasmic reticulum in three cell types of slugs poisoned by molluscicides. Protoplasma 169, 120–129. Van Rijk, A., Bloemendal, H., 2003. Molecular mechanisms of exon shuffling: illegitimate recombination. Genetica 118, 245–249. Vega, M.M., Marigómez, J.A., Angulo, E., 1989. Quantitative alterations in the structure of the digestive cell of Littorina littorea on exposure to cadmium. Mar. Biol. 103, 547–553. Webb, B., Sali, A., 2014. Comparative protein structure modeling using Modeller. Curr. Protoc. Bioinformatics 5 (6), 1-5.6.32 John Wiley & Sons, Inc. Zaldibar, B., Cancio, I., Marigómez, I., 2007. Reversible alterations in epithelial cell turnover in digestive gland of winkles (Littorina littorea) exposed to cadmium and their implications for biomarker measurements. Aquat. Toxicol. 81, 183–196. Zangger, K., Öz, G., Otvos, J.D., Armitage, I.M., 1999. The three-dimensional solution structure of mouse [Cd7]-metallothionein-1 by homonuclear and heterornuclear NMR spectroscopy. Prot. Sci. 8, 2630–2638. Zhu, J., Lu, K., Zhang, C., Liang, J., Hu, Z., 2011. Biochemical and ultrastructural changes in the hepatopancreas of Bellamya aeruginosa (Gastropoda) fed with toxic cyanobacteria. Sci. World J. 11, 2091–2105.