Stable-isotope signatures (δ¹³C and δ¹⁵N) of different tissues of Pinna nobilis Linnaeus, 1758 (Bivalvia): isotopic variations among tissues and between seasons

Abstract

The fan mussel Pinna nobilis is a rapidly growing suspension-feeding bivalve endemic to the Mediterranean Sea and usually inhabits seagrass meadows. This endangered bivalve assimilates carbon and nitrogen isotopic ratios in tissues from its food sources. The δ¹³C and δ¹⁵N signatures were used to demonstrate isotopic variation among three tissues (digestive gland, gills and muscle) and (for digestive gland and gills) between summer and winter. Isotopic values were in the order: gills ≈ muscle > digestive gland. The mean values of the isotopic signal for δ¹³C were –20.41 ± 0.12‰, –19.78 ± 0.07‰ and –19.30 ± 0.5‰ and for δ¹⁵N were 2.84 ± 0.11‰, 4.05 ± 0.14‰ and 3.51 ± 0.15‰ for the tissues digestive gland, gills and muscle, respectively. Seasonal variations (summer–winter) were found in the isotopic signals (δ¹³C and δ¹⁵N) for digestive gland and gills of fan mussel. As a result, this study suggests muscle as the appropriate tissue for future trophic studies and indicates that isotopic signals should be compared in the same season. These δ¹³C and δ¹⁵N values are among the first available for P. nobilis and contribute to the library of δ values for marine invertebrates.

INTRODUCTION

The fan mussel Pinna nobilis is a long-lived species (Galinou-Mitsoudi, Vlahavas & Papoutsi, 2006) endemic to the Mediterranean Sea and is one of the largest bivalves in the world (García-March, 2003). It is a common inhabitant of meadows of the seagrass Posidonia oceanica, where it lives partially buried in sand, anchored among the seagrass shoots and hidden by the leaves (Vicente, 1990). Populations of P. nobilis have been greatly reduced during the past few decades by human impact (Katsanevakis, 2007; Rabaoui et al., 2007) and, indirectly, through the regression of the P. oceanica meadows (Marbà et al., 1996). Consequently, the fan mussel has been listed as an endangered species (EEC, 1992; Centoducati et al., 2007). Since knowledge about the biology and ecology of P. nobilis is fragmentary, further investigation is needed for its conservation (García-March, Perez-Rojas & García-Carrascosa, 2007).

Organisms assimilate carbon and nitrogen stable isotopes from their food sources (Pinnegar & Polunin, 1999), and the analysis of isotopic composition has become an effective method for the study of trophic food webs (Cabana & Rasmussen, 1996; Pinnegar & Polunin, 2000; Fisher, Brown & Willis, 2001). The ratios of stable isotopes ¹³C to ¹²C and ¹⁵N to ¹⁴N is expressed in δ notation with respect to deviations from standard reference material. Values of δ¹³C have mostly been used to identify primary food sources, whereas δ¹⁵N values indicate the trophic level (Post, 2002). Some studies have interpreted the ecological meaning of δ¹³C and δ¹⁵N signatures without taking into account the possibility of variation between tissues and with time (Peterson, 1999). Physiological studies in terrestrial animals have suggested that isotopic enrichment depends on tissue turnover rates and, consequently, it will vary according to tissue type (Tieszen et al., 1983; Gannes, del Rio & Koch, 1998). Moreover, isotopic composition of biological communities can vary as a result of spatial and temporal ecological factors (Cabana & Rasmussen, 1996; Post, 2002; Rubenstein & Hobson, 2004).

Many previous studies have explored isotopic variation in marine organisms (Goering, Alexander & Haubenstock, 1990; Riera et al., 1996; Kang et al., 1999; Rolff, 2000) including macrophytes (Stephenson, Tan & Mann, 1984), P. oceanica (Vizzini et al., 2003), fishes (Gaston & Suthers, 2004), tuna and dolphins (Das et al., 2000) and the sea turtle Chelonia mydas (Seminoff et al., 2006), and some have obtained different isotopic composition according to tissue type (Deudero et al., 2009). Significant differences in isotopic signals for different tissues have been demonstrated in Pecten maximus (Lorrain et al., 2002).

Seasonal variations of food availability from the primary producers (Madurell, Fanelli & Cartes, 2008) affect growth in fishes (Sweeting et al., 2007), mussels (Peharda et al., 2007) and oysters (Sara & Mazzola, 1997; Bayne & Svensson, 2006). Seasonal variations in isotopic signatures may thus reflect variations in food availability (Madurell, Fanelli & Cartes, 2008), for example in P. maximus (Lorrain et al., 2002), Crassostrea gigas and Crepidula fornicata (Decottignies et al., 2007).

Previous work on stable isotopes in P. nobilis has mainly been done to estimate growth and age, oxygen isotopes (¹⁸O/¹⁶O) (Richardson et al., 1999) and the ratios of Mg:Ca and Sr:Ca (Richardson et al., 2004). Reconstruction of sea-surface temperatures and ontogenetic records of metabolic CO₂ incorporation have been studied with oxygen (skeletal δ¹⁸O) and carbon (skeletal δ¹³C) isotopic profiles (Kennedy et al., 2001b). Surprisingly, only one previous investigation has analysed δ¹³C and δ¹⁵N isotopic signatures in P. nobilis; Kennedy et al. (2001a) demonstrated that the fan mussel and its symbiotic shrimp Pontonia pinnophylax assimilated similar food sources and belonged to the same trophic level. To our knowledge, isotopic fractionation at the level tissue has not previously been examined in this bivalve. Consequently, the aim of this study was to evaluate isotopic variation in the fan mussel P. nobilis by dual isotopic analysis (δ¹³C and δ¹⁵N) among three tissues (digestive gland, gills and muscle). A second objective was to assess the seasonal variation in δ¹³C and δ¹⁵N in this bivalve.

MATERIAL AND METHODS

Sampling stations

Pinna nobilis was sampled at two stations on Mallorca Island and three stations around Ibiza Island (Balearic Islands, Spain), in each case from Posidonia oceanica seagrass meadows. These five stations were selected for their similar characteristics and to minimize the impact on vulnerable P. nobilis populations.

Palma and Dragonera were the two stations around Mallorca. Palma (5–10 m depth; SE of Mallorca Island; 39°28′29″N, 2°43′05″E) lies in a marine protected area (MPA) within a zone where neither collecting nor boat anchorage are allowed. Dragonera (7–10 m depth; SW of Mallorca Island; 39°34′48″N, 2°20′54″E) is a small islet that was declared a protected area in 1995 regarding its biodiversity and its natural and pristine characteristics.

Espardell, Talamanca and Esponja were the three sampling stations around Ibiza. Espardell (7–10 m depth; SE of Ibiza; 38°48′10″N, 1°28′42″E) is included in an MPA. Esponja (20–25 m depth; SE of Ibiza; 38° 52′34″N, 1°25′37″E) is a small islet where the bottom shelves steeply. Talamanca (7–10 m depth; SE of Ibiza; 38°54′50″N, 1°28′13″E) was the last sampling station.

Sample collection and biometric characterization of samples

Pinna nobilis is a protected species and listed as endangered. According to European Council Directive 92/43/EEC on the conservation of natural habitats and wild fauna and flora, P. nobilis is under strict protection (Annex IV) and all forms of deliberate capture or killing of this bivalve are prohibited (EEC 1992; Centoducati et al., 2007). Consequently, the individuals of P. nobilis were collected under licence from the Government of the Balearic Islands by experienced scuba divers and this collection was conducted to minimize the impact on the populations. Therefore, the total number of fan mussels sampled was 37, and numbers from each station ranged from 3 individuals at Palma to 12 at Dragonera.

Shell size varied in each population. On each specimen, the parameters maximum shell width (W_m), maximum shell length (L_m) and maximum length of the posterior adductor muscle scars (L_ad) were measured. Age was determined by counting the number of adductor muscle scar rings (R) on right valve of the shell. According to Richardson et al. (1999), the first year's muscle scar ring is absent or inconspicuous, and therefore the estimated age was the number of rings plus 1.

The mussels were quickly transferred into cooler boxes and returned to the laboratory in order to extract the tissues.

Seasonal isotopic variability

Temporal sampling was only conducted at the stations on Ibiza. A total of 22 fan mussels were analysed. Thirteen individuals were collected in summer 2007 and nine in winter 2008. Digestive gland and gills were processed to determine seasonal variation in their carbon and nitrogen isotopic signatures (δ¹³C and δ¹⁵N).

Isotopic analysis and processing

Each fan mussel was dissected in order to separate the digestive gland, muscle and gills. All tissues samples were dried at 60°C for 24 h and then ground to a fine powder using pestle and mortar. For each tissue, a sample of powder (2 ± 0.1 mg) was placed in a cadmium tin cup and combusted for ¹⁵N and ¹³C isotope analysis by continuous-flow isotope-ratio mass spectrometry (CF-IRMS) using a Thermo Delta X Plus mass spectrometer. Analyses were conducted at the Stable Isotopes Laboratory of the Science Technical Service of the University of the Balearic Islands. Samples of an internal reference material were analysed after every eight samples to calibrate the system and to compensate for drift over time. The reference material used was Bovine Liver Standard (1577b; US Department of Commerce, National Institute of Standards and Technology, Gaithersburg, MD 20899, USA). The analytical precision was based on the standard deviation of replicates of the internal standard (BSL) and was 0.03‰ for δ¹³C and 0.08 for δ¹⁵N. Stable isotope abundances were measured by comparing the ratios of the isotopes (¹³C:¹²C and ¹⁵N:¹⁴N) in the sample with those of the standard. Carbon and nitrogen isotopic ratios were expressed in δ notation in terms of parts per thousand (‰) deviation from the standard according to the equation:

where X is ¹³C or ¹⁵N and R the corresponding ¹³C/¹²C and ¹⁵N/¹⁴N ratio.

Statistical analysis

Two-way crossed analysis of similarity (ANOSIM) was performed in order to analyse the variation of the isotopic signatures (δ¹³C and δ¹⁵N) of P. nobilis by age.

A permutational multivariate ANOVA (PERMANOVA; Anderson, 2001) was used to test simultaneous responses of the fan mussel variation isotopic signatures (δ¹³C and δ¹⁵N) to the factor Tissue (with three levels: digestive gland, muscle and gills) and to the factor Locality (with five levels: Palma, Dragonera, Espardell, Talamanca and Esponja). For the factor Tissue, a posteriori pairwise comparisons of levels were performed. Additionally, a distance-based test for homogeneity of multivariate dispersions (PERMDISP) was done for the factor Tissue at each locality.

PERMANOVA was used to analyse season variations in isotopic signatures in relation the factors Tissue (two levels: digestive gland and gills), Locality (three levels: Espardell, Talamanca and Esponja) and Season (two levels: summer and winter).

All statistical analyses were performed using the statistical package PRIMER^® 6.1.10 and PERMANOVA with β20 software.

RESULTS

Morphometry and age of fan mussels

Maximum shell width (W_m), maximum shell length (L_m), maximum length of the posterior adductor muscle scars (L_ad) and number of muscle scar ring (R) were 20.37 ± 2.22, 30.14 ± 2.51, 17.17 ± 1.32 and 5.24 ± 0.55 (mean ± SE), respectively, for the 37 sampled specimens. No differences were found in the isotopic signatures taking into account the age of Pinna nobilis (ANOSIM, P > 0.05).

Isotopic composition of tissues

The δ¹³C and δ¹⁵N isotopic signatures of P. nobilis showed significant differences among the three tissues (Table 1; PERMANOVA 1). Pairwise tests indicated differences among δ¹³C isotopic signatures for the three analysed tissues. In relation to δ¹⁵N signals, differences only appeared between digestive gland and the others tissues (gills and muscle), while no differences in the δ¹⁵N signals between muscle and gills were found. The multivariate dispersions test (PERMDISP) showed small dispersions of the isotopic values for each tissue (Table 1).

Table 1.

Permutational ANOVA (PERMANOVA 1) analyses to analyse tissue variation in Pinna nobilis.

Source of variations	PERMANOVA 1
	δ13C				δ15N
	df	SS	MS	Pseudo-F	df	SS	MS	Pseudo-F
Tissue	2	32.33	16.16	42.46*	2	24.86	12.43	45.32*
Locality	4	1.47	0.37	0.97n.s.	4	45.13	11.28	41.13*
Tissue × Locality	7	5.22	0.74	1.96n.s.	7	2.98	0.42	1.55n.s.
Residual	76	28.93	0.38		76	20.85	0.27
Total	89	71.65			89	100.02
Pairwise within level of factor Tissue	Digestive gland ≠ muscle ≠ gills				Digestive gland ≠ muscle = gillsn.s.
	Pairwise comparisons
	δ13C				δ15N
Groups	t				t
Digestive gland, muscle	3.03n.s.				0.52n.s.
Digestive gland, gills	1.97n.s.				1.81n.s.
Muscle, gills	0.40n.s.				1.36n.s.
	PERMDISP means and standard errors
	δ13C				δ15N
Group	Mean		SE		Mean		SE
Digestive gland	0.65		0.08		0.59		0.08
Muscle	0.36		0.05		0.65		0.09
Gills	0.39		0.09		0.88		0.02

PERMANOVA based on the modified Euclidean distance dissimilarity of the δ¹³C and δ¹⁵N isotopic signatures (9999 permutations) for 37 mussels. The studied factors were five sampling stations and three tissues (digestive gland, muscle and gills). Pairwise test on the significant interaction term Tissue is shown. Pairwise comparison of PERMDISP test of the factor Tissue is based on the modified Euclidean distance dissimilarity of the δ¹³C and δ¹⁵N isotopic signatures.

*P < 0.001.

^n.s.Nonsignificant difference.

The δ¹³C and δ¹⁵N variation of the three tissues (digestive gland, muscle and gills) of P. nobilis is shown in Figure 1A. Comparison among tissues showed that values for the digestive gland were always lower than for muscle and gills. The mean values of the isotopic signals for δ¹³C were –20.41 ± 0.12‰, –19.78 ± 0.07‰ and –19.30 ± 0.5‰ for digestive gland, gills and muscle, respectively. The δ¹⁵N for digestive gland, gills and muscle were 2.84 ± 0.11‰, 4.05 ± 0.14‰ and 3.51 ± 0.15‰, respectively. Variation among the tissues for both δ¹³C and δ¹⁵N was in the order: gills ≈ muscle > digestive gland.

Figure 1.

Stable isotopes signatures (δ¹⁵N vs δ¹³C) of Pinna nobilis. A. Isotopic variation among tissues (digestive gland, n = 20; muscle, n = 18; gills, n = 19) of the five sampling stations. B. Dragonera. C. Palma. D. Espardell. E. Esponja. Symbols: circle, digestive gland; square, muscle; triangle, gills.

Table 1 (PERMANOVA 1) shows the comparison between the different stations. Differences were observed among stations for δ¹⁵N, but not for δ¹³C. For each sampling station, the isotopic signature for the digestive gland was lower than those for muscle and gills (Fig. 1B–E).

Seasonal isotopic variation

Strong differences between summer and winter were observed in δ¹³C and δ¹⁵N for both digestive gland and gills tissues (Table 2; PERMANOVA 2). Differences in δ¹³C and δ¹⁵N signatures have been demonstrated here between tissues and localities, but the interaction tissue–season was significant only for δ¹³C. The carbon isotopic signature decreased between summer and winter samples by 1.45‰ for the digestive gland and 0.94‰ for gills. The corresponding decreases for nitrogen were 0.99‰ for the digestive gland and 0.88‰ for gills (Fig. 2).

Figure 2.

Seasonal variation in the isotopic signatures (δ¹⁵N vs δ¹³C) for digestive gland and gills of Pinna nobilis. A. Digestive gland. B. Gills. Open symbols indicate winter and closed symbols the summer season.

Table 2.

Permutational ANOVA (PERMANOVA 2) to analyse seasonal isotopic variations.

Source of variations	δ13C				δ15N
	df	SS	MS	Pseudo-F	df	SS	MS	Pseudo-F
Tissue	1	10.33	10.33	68.69***	1	12.34	12.34	55.59***
Locality	2	1.84	0.92	6.10**	2	7.43	3.72	16.75**
Season	1	13.45	13.45	89.43***	1	9.64	9.64	43.44***
Tissue × Locality	2	0.55	0.03	0.18n.s.	2	0.25	0.12	0.56n.s.
Locality × Season	2	0.11	0.05	0.37n.s.	2	0.37	0.19	0.84n.s.
Tissue × Season	1	0.67	0.67	4.49*	1	0.01	0.01	0. 06n.s.
Locality × Tissue × Season	2	0.04	0.02	0.12n.s.	2	0.21	0.11	0.49n.s.
Residual	27	4.06	0.15		27	5.99	0.22
Total	38	31.05			38	35.91

PERMANOVA based on the modified Euclidean distance dissimilarity of the δ¹³C and δ¹⁵N isotopic signatures for 22 Pinna nobilis (9999 permutations). The studied factors were three sampling stations, two tissues (digestive gland and gills) and two seasons (winter and summer). Pairwise test on the significant interactions term Tissue is shown.

*P < 0.05.

**P < 0.01.

***P < 0.001.

^n.s.Nonsignificant difference.

DISCUSSION

The present study has demonstrated variation in δ¹³C and δ¹⁵N isotopic signatures among three tissues (digestive gland, muscle and gills) of the fan mussel Pinna nobilis. Both δ¹³C and δ¹⁵N showed a similar pattern of variation among tissues (gills ≈ muscle > digestive gland), with a small dispersion for each tissue. The interpretation of isotopic data may therefore require separate consideration of the values for each tissue (Lorrain et al., 2002). A stable isotope study of P. nobilis by Kennedy et al. (2001a) reported that the mean value of δ¹³C for muscle was –18.3‰, which is similar to the present value (–19.3‰), considering the 1‰ variation related to the diet of this species (Kennedy et al., 2001a).

Several studies have shown that isotopic enrichment processes are tissue-specific (Yokoyama et al., 2005; Logan et al., 2006). Differences between tissues could be explained by the suggestion that isotopic enrichment depends on tissue turnover rates (Tieszen et al., 1983; Frazer et al., 1997; MacAvoy, Macko & Garman, 2001). Tissues with low turnover rates, such as the muscle, integrate dietary isotopic signatures over a longer time period than tissues that have higher turnover rates, such as digestive gland (Watanabe, Seikai & Tominaga, 2005; Bodin, Le Loc'h & Hily, 2007). This may explain the higher values reported here for muscle and gills (tissues with low turnover rates) than for digestive gland (with high turnover rate). Another factor that could contribute to the higher δ¹³C values in muscle and gills and lower value for digestive gland is the negative linear relationship between δ¹³C and the lipid content of the tissues (Tieszen et al., 1983; Lorrain et al., 2002; Thompson et al., 2002). This is in agreement with the lower δ¹³C signatures for the tissue with high lipid content (digestive gland) than for the tissues with low lipid percentages (muscle and gills).

Isotopic variation among different tissues has been found in many other marine organisms (Stephenson, Tan & Mann, 1984; Goering, Alexander & Haubenstock, 1990; Riera et al., 1996; Kang et al., 1999; Das et al., 2000; Rolff, 2000; Vizzini et al., 2003; Gaston & Suthers, 2004). Bodin, Le Loc'h & Hily (2007) documented isotopic enrichment for δ¹³C and δ¹⁵N in the white muscle of the decapod crustacean Maja brachydactyla compared to gonads, which in turn showed higher values than the digestive gland. A similar pattern has been reported in Pecten maximus, where muscle tissue was found to have larger ¹³C and ¹⁵N values than those of the digestive gland (Lorrain et al., 2002). Deudero et al. (2009) found an isotopic variation pattern in the order: gills ≈ muscle > digestive gland using Mytilus galloprovincialis. Our finding of lower values in digestive gland than muscle and gills is in agreement with these previous studies.

The present results for P. nobilis showed seasonal (summer–winter) differences in the isotopic signatures of digestive gland and gills, and that this was most pronounced in the digestive gland. Larger values were found in summer than in winter. Decottignies et al. (2007) have also demonstrated significant differences among seasons in δ¹³C and δ¹⁵N in Crassostrea gigas and Crepidula fornicata. These variations could be related to the availability of food sources (Decottignies et al., 2007), since it has been shown that the food web determines isotopic signatures of the filter-feeding Mytilus edulis, C. gigas and C. fornicata (Riera, 2007). Likewise, seasonal variations have been demonstrated in P. maximus and the largest variations occurred in the digestive gland (Lorrain et al., 2002).

Knowledge of the isotopic values is important for understanding trophic pathways and food source partitioning within and between ecosystems. The feeding ecology of filter feeders is difficult to study, and isotopic studies have contributed to an understanding of their role in the food webs in marine ecosystems (Lorrain et al., 2002; Yokoyama et al., 2005). The present work provides δ¹³C and δ¹⁵N values for an endangered filter feeder, P. nobilis, and contributes to the library of δ values for marine invertebrates. From our results, we suggest that muscle is the most appropriate tissue for future trophic studies of this kind, taking into account its uniform isotopic signatures and its low tissue-turnover rates. Previous studies have indeed used muscle to determine the diet and trophic level of the P. nobilis (Kennedy et al., 2001a). In addition, we emphasize that isotopic comparisons should be conducted at the same time of year, because values may show seasonal variation as reported here.

ACKNOWLEDGEMENTS

The authors thank the Services Scientist-Technical (UIB) for collaboration in isotopic analysis, and especially the support offered by M. Ribas (IUNICS). Also, we are grateful for logistic support from S. Tejada, E. Álvarez, S. Sardu, A. Martín, Y. Bertrand, J. Jimenez and A. Box. This study was partly financed, and the author M.C.-R. received a fellowship from: the Acció Especial ‘Desarrollo de técnicas para evaluar juveniles del bivalvo endémico Pinna nobilis mediante dispositivos de asentamiento’ of the DG Recerca, Desenvolupament Tecnològic i Innovació, Conselleria d'Economia Hisenda i Innovació, CAIB.