Regional patterns of δ13C and δ15N for European common cuttlefish (Sepia officinalis) throughout the Northeast Atlantic Ocean and Mediterranean Sea

Stable isotope compositions of carbon and nitrogen (expressed as δ13C and δ15N) from the European common cuttlefish (Sepia officinalis) were measured in order to evaluate the utility of using these natural tracers throughout the Northeast Atlantic Ocean and Mediterranean Sea (NEAO-MS). Mantle tissue was obtained from S. officinalis collected from 11 sampling locations spanning a wide geographical coverage in the NEAO-MS. Significant differences of both δ13C and δ15N values were found among S. officinalis samples relative to sampling location. δ13C values did not show any discernable spatial trends; however, a distinct pattern of lower δ15N values in the Mediterranean Sea relative to the NEAO existed. Mean δ15N values of S. officinalis in the Mediterranean Sea averaged 2.5‰ lower than conspecifics collected in the NEAO and showed a decreasing eastward trend within the Mediterranean Sea with the lowest values in the most eastern sampling locations. Results suggest δ15N may serve as a useful natural tracer for studies on the population structure of S. officinalis as well as other marine organisms throughout the NEAO-MS.


Introduction
The European common cuttlefish, Sepia officinalis Linnaeus, 1758 is a coastal nektobenthic species ranging from the Shetland Islands through the Northeast Atlantic Ocean and Northwest Africa into the Mediterranean Sea [1]. This species constitutes one of the most economically valuable cephalopod resources in the Northeast Atlantic Ocean, supporting an important fishery resource [2,3]. Sepia officinalis has a relatively short lifespan of 1-2 years, early sexual maturity and an extended spawning season laying eggs on the seafloor with direct benthic, large hatchlings [4,5]. Given this species geographical distribution combined with limited dispersal, it has been a targeted model species to examine connectivity throughout the Northeast Atlantic Ocean and Mediterranean Sea (hereafter NEAO-MS) [6].
Natural biomarkers such as stable isotopes are commonly used to examine food web structure and ecosystem connectivity in marine environments [7,8]. Stable isotopes of carbon (δ 13 C) and nitrogen (δ 15 N) are particularly useful tracers due to their natural abundance being influenced by the environment and ease of measurement in body tissues without having to track individuals in a population. δ 13 C is traditionally used to trace carbon pathways because little fractionation occurs between predator and prey, and different primary producers (energy sources) often have unique δ 13 C values [9]. δ 13 C values of consumers are a product of the primary producers' composition and influenced by the dissolved inorganic carbon (DIC) pool, as well as local abiotic factors including sea surface temperature, and can differ across ocean basins [10] and region-specific freshwater to marine gradients [9]. δ 15 N becomes enriched with increasing trophic level and is used to infer trophic position [7], but can also differ at the base of the food web. Depending upon the types of nutrients available to stimulate growth, δ 15 N values can be used to track energy flow in high-nutrient (nitrate) and low-nutrient (N 2 fixation) ecosystems as well as new nitrogen (upwelled nitrate) versus regenerated nitrogen (ammonia, urea). Combining both δ 13 C and δ 15 N offers the potential to study the connectivity and population structure of species because longitudinal and latitudinal gradients exist throughout marine ecosystems [11,12], including the NEAO-MS [8,13].
The population structure of S. officinalis has been found throughout the NEAO-MS [14]. Rooker et al. [6] used stable isotopes of δ 13 C and δ 18 O in the cuttlebones and found regional differences throughout the NEAO-MS due to different environmental conditions among regions, in addition to potential movement and connectivity between the Strait of Gibraltar and western Mediterranean Sea. Another study combined genetics, morphometrics and trace elements in cuttlebones with results suggesting discrete population structure within the eastern Mediterranean Sea [15]. Drábková et al. [16] showed the degree of genetic diversification of S. officinalis throughout the Mediterranean Sea and suggested this pattern resulted from a low dispersion capability and a complex structure with different degrees of genetic exchange among these subpopulations. Further, Pasqual et al. [17] performed a metaanalysis of species connectivity (based on life-history traits) throughout the Mediterranean Sea supporting limited connectivity and a high probability of population structuring for species such as S. officinalis. Previous studies have found patterns of δ 13 C and δ 15 N in other marine species throughout the NEAO-MS suggesting the potential to use these tracers to evaluate population structure and connectivity [13,18].
The primary purpose of this study was to examine patterns in stable isotopes of δ 13 C and δ 15 N from mantle tissue of S. officinalis collected throughout the NEAO-MS in order to evaluate the utility of using these natural tracers as tools to examine population structure. Previous studies on S. officinalis have used royalsocietypublishing.org/journal/rsos R. Soc. Open Sci. 8: 210345 a variety of techniques including genetics [19], cuttlebone geochemistry [6,20] and electronic tagging [4], thus this study will expand upon the ecological toolbox available to researchers. . Samples were acquired from artisanal or commercial trawlers, and local market sampling. For market sampling, special care was taken to acquire small body sizes by local fishers in order to ensure samples were from the specific region of collection. No special permission to conduct the research was needed and no ethical assessment was required. Upon collection, S. officinalis samples were measured using straight mantle length (ML) to the nearest mm and white muscle tissue samples were excised from the mantle and immediately freeze dried or preserved in 70% ethanol and later freeze dried. Subsamples of cuttlefish were paired with both (i) immediately freeze dried and (ii) ethanol-preserved specimens later freeze dried from Bay of Biscay-South (n = 20) and Gulf of Cadiz (n = 19) sample locations in order to assess if sample preservation techniques affected δ 13 C and δ 15 N.

Material and methods
Freeze-dried muscle tissue for stable isotope analysis was homogenized using a mortar and pestle, weighed, wrapped in tin capsules and shipped to the University of California at Davis' Stable Isotope Facility for analysis. Analysis of muscle tissue δ 13 C and δ 15 N was carried out using an elemental analyser (PDZ Europa ANCA-GSL) coupled with an isotope ratio mass spectrometer (PDZ Europa 20-20). The long-term standard deviation of UC Davis' stable isotope facility is 0.2‰ for δ 13 C and 0.3‰ for δ 15 N. Lipids were not extracted from tissue; however, C : N ratios were low (less than or equal to 3.75) across the size spectrum of S. officinalis samples, indicating a low lipid content and little influence of lipids on tissue δ 13 C values [21]. Mean C : N ratio was 3.21 ± 0.13 s.d. and ranged from 2.87 to 3.75. A study by Ruiz-Cooley et al. [22] also showed a significant effect on stable isotope values as a result of lipid extractions on squid tissues and recommended researchers use caution with this approach for ecological studies on cephalopods. Isotopic ratios are reported relative to Vienna PeeDee belemnite for carbon and atmospheric N 2 for nitrogen.
Multivariate analysis of covariance (MANCOVA) was used to test for differences in muscle tissue δ 13 C and δ 15 N. Dependent variables included both δ 13 C and δ 15 N, cuttlefish size (ML) was used as royalsocietypublishing.org/journal/rsos R. Soc. Open Sci. 8: 210345 the covariate, and sampling location as the independent factor. Pillai's trace statistic was used to test for significance. Univariate tests for both δ 13 C and δ 15 N were also performed using analysis of covariance (ANCOVA) and a posteriori differences among means were analysed with Tukey's honestly significant difference (HSD) test. A t-test was used to test for differences in δ 13 C and δ 15 N of cuttlefish tissue based on preservation technique (freeze-dried versus ethanol-preserved specimens). Normality was evaluated using a Shapiro-Wilk test and the equal variance assumption was assessed by the Spearman rank correlation between the absolute value of the residuals and the observed value of the dependent variable. Quadratic discriminant function analysis (QDFA) was used to evaluate the best possible classification accuracy of δ 13 C and δ 15 N values of cuttlefish tissue based on (i) sampling location (sites A-K) and (ii) NEAO and MS regions. Statistical significance for all tests was determined at the alpha level of 0.05.  with lowest values in the most eastern sampling locations (Gulf of Corinth, Aegean Sea-West, Aegean Sea-East) (figure 3). While size ranges of cuttlefish slightly differed among sample locations, no main effect of size was found on δ 15 N (ANCOVA, p = 0.481).

Results
Results from QDFA showed moderate overall classification success of 60% among the 11 sampling locations. Classification success was highest at sampling locations in the Atlantic with Bay of Biscay-North at 90% and Bay of Biscay-South at 98%, while lowest classification success was from sampling locations in the Mediterranean Sea (0% success for both Adriatic Sea and Aegean Sea-East). Overall classification success was 93% when contrasting between the Atlantic (96%) and Mediterranean Sea (89%) highlighting the differences in δ 13 C and δ 15 N of cuttlefish tissue from the two regions.

Discussion
Life-history strategy of S. officinalis combined with previous tracer studies suggests limited dispersal and connectivity leading to population structuring in the NEAO-MS. A relatively short lifespan, early sexual maturity and a nektobenthic strategy lacking a pelagic larval stage supports population structuring of this species throughout its range [17,23]. Additionally, genetic studies of S. officinalis within the royalsocietypublishing.org/journal/rsos R. Soc. Open Sci. 8: 210345 NEAO-MS highlight the complex population structure and a high degree of genetic diversification at a relatively fine spatial scale [14,16,24]. Pérez-Losada et al. [19] suggested that the population structure of S. officinalis in the NEAO-MS was driven by isolation through distance and the results of stable isotopes in muscle tissue presented here show support for this based on the isoscapes throughout the range examined. These results parallel those from a similar study conducted by Rooker et al. [6] that found area-specific geochemical signatures in cuttlebone chemistry over the same geographical range. While this study did not have the spatial resolution to examine potential connectivity through the Strait of Gibraltar as Rooker et al. [6], results show clear gradients and potential for stable isotopes of δ 13 C and δ 15 N as natural tracers for this species. Spatial patterns of δ 15 N in S. officinalis muscle tissue showed a distinct gradient with higher values in samples collected from the NEAO and a gradual depletion into the Mediterranean Sea from west to east. These findings parallel those of McMahon et al. [8] where authors performed a meta-analysis of δ 15 N for zooplankton examining spatial patterns throughout the Atlantic Ocean and found δ 15 N values were on average 2‰ higher than values in the Mediterranean Sea. McMahon et al. [8] also found a similar gradient within the Mediterranean Sea of higher δ 15 N values in the west near the Strait of Gibraltar relative to the east. Similar δ 15 N patterns have been found for a wide variety of marine taxa ranging from krill [25], fishes [18,26], whales [13] and seabirds [27] with consistently higher values in the NEAO relative to the Mediterranean Sea. Patterns of δ 15 N are most likely due to the differential utilization of nitrogen sources. Higher latitudes of the Atlantic are fuelled by nitrate in contrast with the oligotrophic Mediterranean Sea where nitrogen fixation by diazotrophy occurs [28], particularly in the eastern Mediterranean Sea [29]. Estimates indicate that up to 90% of the nitrogen in the eastern Mediterranean Sea may be derived from biological N 2 fixation in contrast with only 20% in the west [28], probably owing to the distinct gradient in δ 15 N observed in the Mediterranean Sea from this study and others.
Isoscapes in δ 13 C of S. officinalis did not show any discernable spatial trends. δ 13 C values of marine organisms are largely a function of the local DIC, and a meta-analysis of zooplankton δ 13 C in the North Atlantic did not find any notable patterns in the NEAO-MS [8]. The complexity of hydrodynamical, biogeochemical and ecological features in the NEAO-MS [30] may explain the lack of a clear gradient in δ 13 C values of S. officinalis. Further, the heterogeneous patchiness of the phytoplankton community structure throughout pelagic waters of the Mediterranean Sea may owe to the δ 13 C results in this study [31]. Thus, a combination of the complex physical oceanography combined with the biological community of the ecosystem may contribute to the lack of any δ 13 C gradients observed. Two notable collection locations (Portugal-South and Aegean Sea-West) had higher δ 13 C values in S. officinalis muscle tissue than other sampling locations. Specific areas of sample collection at Aegean Sea-West are unknown due to market-based sample collections; however, Portugal-South collections occurred near shore relative to other locations in this study. Near-shore waters with increased nutrient loads, higher productivity and 13 C-enriched seagrass input probably contribute to higher δ 13 C values than offshore environments [32]. In addition, 13 C inputs from benthic algae and terrestrial sources are 13 C heavy probably owing to the higher δ 13 C values from this particular study location [32]. Future studies may want to consider increased sampling resolution as this study demonstrated local conditions probably affect the stable isotope values measured in consumer tissue. royalsocietypublishing.org/journal/rsos R. Soc. Open Sci. 8: 210345 Preservation techniques for S. officinalis muscle tissue did not affect stable isotopes of δ 13 C and δ 15 N, demonstrating both methods are adequate. This finding is particularly useful for researchers collecting tissue samples of this species without immediate access to equipment such as freezers or freeze driers. Previous studies using stable isotopes on tissues of consumers including cephalopods, fishes and zooplankton have reported varying results ranging from major effects of preservation techniques to little or no impact on results [22,33,34].
Findings from this study support the use of stable isotopes as natural tracers to examine the population structure of S. officinalis in the NEAO-MS and add to the growing ecological toolkit of tracers available. Specifically, δ 15 N may serve as a useful natural tracer to track movement and connectivity for marine species in the region. These findings can be extended to other marine organisms from zooplankton to apex predators to examine movement and highlight specific gradients throughout the NEAO-MS. Furthermore, S. officinalis are opportunistic feeders with diets comprising bony fishes, crustaceans, molluscs and polychaetes [5,35,36] and serve as prey to larger bony fishes, sharks and marine mammals [3] thereby serving as a model species for future trophic and population connectivity studies throughout the NEAO-MS.
Ethics. No ethical assessment was needed in order to carry out this research. All collections were from artisanal or commercial fishing operations, and they were caught for fish markets (not specifically for this study). All cuttlefish at markets were dead (no live specimens obtained).