Abstract |
The giant red shrimp Aristaeomorpha foliacea (Risso, 1827) is a large-sized decapod
crustacean first described in the Ligurian Sea. Together with Aristeus antennatus (blue and
red shrimp), are the only species of the family Aristeidae found in the Mediterranean,
constituting the main target species for the demersal deep-sea fishery in the Mediterranean.
The species geographic distribution is wide. It has been reported to occur in the
Mediterranean, the Atlantic, the Indian Ocean, the western Pacific and South Africa. In the
Mediterranean Sea, the distribution of the giant red shrimp is patchy in nature, with the
highest abundances found in the central-eastern basins. Aristaeomorpha foliacea is a deepwater benthopelagic shrimp with a reported depth distribution of 120-1300 m, generally
inhabiting muddy bottoms. The species aggregates in submarine trenches and canyons along
the continental slope and peaks in abundance at 300-800 m depths. The economic
importance of the giant red shrimp enhanced the scientific interest to study its populations
and evaluate its stocks. Until now, however, research on the genetic diversity of the species
in the Mediterranean is limited. Only a small number of studies have been published aiming
at detecting the genetic structure of the species in the Mediterranean. All these studies share
in common the small number of: (a) samples collected, (b) areas sampled, and (c) molecular
markers used which were either microsatellites or mitochondrial/nuclear DNA sequences.
According to these studies, there seems to be a lack of genetic differentiation among
populations living in different locations in the Mediterranean.
The present work attempts to study the genetic structure of A. foliacea populations in the
Mediterranean. The research approach is based on an extensive number of samples, collected
in the context of the European project MED_UNITS. A Genotyping-by-Sequencing (GBS)
methodology constructing reduced-representation libraries was selected to generate a set of
restriction site associated DNA (RAD) fragments isolated from a given number of individuals
that are barcoded and sequenced together. Single Nucleotide Polymorphism (SNP) markers
newly isolated following the double-digest restriction site associated DNA (ddRAD)
sequencing were used and a high number of samples were genotyped. SNPs have nowadays
become the marker of choice for a variety of molecular ecology studies, since advances in
sequencing technologies coupled with increases in computational power have resulted in a
shift towards genome‐scale data analysis. The main advantages of this approach are the high
number of SNPs usually identified, their wide distribution in the genome, and the ability given
by High Performance Computing (HPC) systems to analyze data from a large number of
individuals. Also, this method does not require a fully sequenced reference genome as loci
can be reconstructed de novo from sequencing reads.
The sampling effort resulted in 1,692 samples (tissues preserved in absolute ethanol) covering
32 geographic sub-areas (GSA), from the Balearic Islands in the W. Mediterranean to the
Cypriot and Egyptian waters in the E. Mediterranean. DNA quality was in general above
average and almost 75% of the samples were used for the library preparations and
downstream analyses. Sequenced reads were analyzed using STACKS v.2.4 pipeline in order
to quality control the reads, identify the genomic loci sequenced, genotype each individual,
and conduct basic population genetics analysis.
Using all individuals genotyped, the species "catalogue" comprised approximately 2.4 million
ddRAD loci but the number of (Ustacks) loci varied significantly among specimens (from 17 to
244,066) and among populations on average (1,217 to 69,625). Thus, samples with low
number of stacks loci (we opted for less than 6,000) had to be filtered out and all
phylogeographic analyses were run with a dataset composed of 661 samples (for 30 localities)
and 734 higher quality SNPs present in at least 70% of the samples. Results point out an
evident lack of genetic differentiation and are generally in agreement with previous studies
conducted at smaller geographic scales and less extended sampling points in the
Mediterranean Sea. In particular, observed heterozygosities (Ho) were generally low (ranging
from 7.3 to 18.6%) and smaller than the expected (23.8 to 45.5%). Low Ho values were found
mostly in samples with low number of specimens which seem to also have a smaller number
of common polymorphic loci from the dataset of 734 SNPs.
Pairwise Fst values were low and not significant, ranging from -0.080 to 0.058 using an
uncorrected probability threshold of 0.05, and a Benjamini-Hochberg correction. The Fst
values were found not significant and remained low even when populations with a small
number of specimens were grouped together, in order to increase the sampling size and make
the analysis more powerful (-0.06263 to 0.02865). Such weak differentiation was evident both
by DAPC where all populations seemed to belong to a single cluster, and STRUCTURE analysis.
Using the STRUCTURE software, the best k value selected using the EVANNO method was 2,
and with most individuals assigned to both clusters (admixed). Moreover, AMOVA results for
several scenarios of grouping A. foliacea populations did not indicate any significant genetic
structure but rather a panmictic situation in the Mediterranean Sea. AMOVA showed a small
FCT = 0.00196 (0,2%) in a scenario involving 15 populations (after merging neighboring
populations of low sample size) divided in 3 groups (Western, Central, Eastern
Mediterranean). Slightly lower FCT= 0.00183 was calculated when populations were merged
to 20 and even less FCT = 0.00088 when they were used as they originally defined (30
populations divided in 3 groups). All FCT values were statistically significant (P<0.0001). Thus,
all scenarios indicated the existence of very weak but significant differentiation among
proposed groups and the greatest part of the identified genetic variation was attributed to
differences among individuals in the populations, and much less among groups.
Current results refer to samples from different areas throughout the Mediterranean Sea
where the species occurs and point out an evident lack of genetic differentiation; this has
already been reported at smaller geographic scales and less extended sampling points. The
present study, conducted with a high number of nuclear DNA markers and a big number of
samples/sampling areas, also revealed no distinct sub-populations and low genetic
heterogeneity. This can be attributed to the high dispersal abilities of the red shrimps during
both adult and larval stages. Last, we discuss the need to obtain a high-quality data set for
the analysis, with no or little missing data, and the potential problems that may arise because
of them.
|