Page 2 - Chrysoula_al_2010

1680 C. Gubili et al. Historical dispersal of white sharks

Sicilian Channel and nearby waters. In the eastern basin, sterile conditions, with tools and surfaces cleaned frequently.
most observations are from the north, particularly the Blank samples were used to verify lack of contamination.
Adriatic, while the warmer, more saline south-eastern The six new sequences were aligned using CLUSTALX
region returns infrequent records. Instances of large [23] against 49 previously published sequences [18,19]
mature individuals, neonates and pregnant females in including animals sampled in the south-eastern Pacific
these waters [11,16,17] imply the existence of pupping (AU and NZ; AY026196–AY026209, AY026211), NEP
and nursery grounds. However, there are no accounts of (GU002321–GU002302) and western Indian Ocean (SA;
the origin and genetic profile of Mediterranean white AY026210, AY026212–AY026224). Additionally, two
sharks owing to rarity of samples. sequences from the porbeagle (Lamna nasus) were used as
The global phylogeography of C. carcharias is yet to be an outgroup. Measures of haplotype (h) and nucleotide (p)
studied comprehensively. Nuclear gene flow throughout diversity were calculated with ARLEQUIN v. 3.1.1 [24].
the Indian Ocean, and highly distinct mitochondrial A median-joining network was constructed to depict haplo-
haplotypes from populations either side of both the type relationships using NETWORK [25]. Based on
Indian and Pacific Oceans, suggest female philopatry likelihood-ratio tests (MODELTEST 3.7 [26]), the most
and long-term isolation [18]. However, the evolutionary appropriate model of evolution was the HKY þ G model,
history of remaining disjunct populations is poorly known. with a gamma distribution parameter of 0.5332. Phyloge-
Concordant with the hypothesis of natal philopatry, tagging netic relationships were examined using MRBAYES v. 3.1.2
studies reveal that although this species makes rapid and [27]. Markov chain Monte Carlo (MCMC) simulations
long transoceanic movements in both the Indian Ocean (a were run for 2 000 000 generations, and the first
round trip of 11 100 km [3]) and northeast Pacific (traver- 10 per cent less-optimal tree generations were discarded
sing 2000–5000 km [19]), individuals adhere to a highly as ‘burn-in’. PHYML v. 3.0 [28] was also used to calculate
predictable cycle, persistently returning to natal coastal the maximum-likelihood tree. An alternative tree topology
locations following migration [3,19,20], a behaviour which (NWA/SA as sister clade to the Mediterranean) was compared
places populations at risk from local threats. with the ML inferred tree using the Shimodaira–Hasegawa
Evidence of dramatic population declines in large (SH) test, as executed in PAUP* v. 4.0b10 [29]with RELL
predatory sharks in the Mediterranean [21] and the approximations and 1000 bootstrap replicates to produce a
potential consequences of trophic cascade [1] make it null distribution of differences in log-likelihoods. The SH
imperative to determine the utilization and connectivity test compares the difference in log-likelihoods of competing
of natural populations in vulnerable areas such as nur- tree topologies. Finally, evolutionary rates for the control
sery/pupping grounds, especially in species exhibiting region were calculated in two steps: initially, the nucleotide
natal philopatry. Here, we report the first genetic analysis substitutions per site (Da) were calculated between sequences
of Mediterranean white sharks, using the mtDNA control from the Pacific (AU, NZ, NEP) versus either the Atlantic
region. To elucidate the origins and relationships of rare (NWA) or eastern Indian ocean (SA) using DNASP v. 5.10
Mediterranean white sharks to other stocks their haplo- [30], and calibrated with either the rise of the Panama
types were compared with those from the north-eastern Isthmus (3.5 Ma [31]) or Sunda-Sahul shelves (5 Ma [32];
Pacific (NEP; California) [19], south-western Pacific the most protracted and, for an epipelagic species, the most
(Australia, AU; and New Zealand, NZ), western Indian significant lineage-splitting period of sea level drop early in
Ocean (South Africa, SA) [18] and the north-west white shark evolution).
Atlantic (NWA).

3. RESULTS
A 1083 bp sequence of the mtDNA control region was
2. MATERIAL AND METHODS
obtained for five of the new samples (three Mediterranean
(a) Specimens
and two NWA; accession nos HQ540294–HQ540298).
Mediterranean samples consisted of ethanol-preserved heart
Overall, 95 polymorphic sites revealed 45 haplotypes
tissue from two neonate sharks caught off the coast of
(figure 1) from 54 sequences examined, showing that
Altinoluk in the Bay of Edremit in the north-eastern
C. carcharias exhibits the highest haplotype (0.9888 +
Aegean Sea (eastern Mediterranean basin, Turkey) on 1
0.0075) and nucleotide (0.0223+ 0.0110) diversities of
and 4 July 2008 (from separate parents [17]), a fin clip
any shark studied to date [33–36]. Summary statistics
from a juvenile caught in Aras Dizra, Tunisia (also eastern
for each population are given in electronic supplementary
Mediterranean basin) on 20 April 2006 [22], together with
material, table S1. Yet remarkably, three of the four
dried tissue from a shark caught in a tuna net off
Mediterranean (MED) samples shared the same
Favignana, Sicily (western Mediterranean basin) in the late
haplotype (GW34), whereas each NWA sample was
1980s. Additionally, two NWA fin clip samples (immature
unique (GW43, GW44). The median-joining network
males, caught 1994) were collected off the west Florida
indicates that the MED haplotypes show little genetic
coast, USA.
differentiation from Indo-Pacific sharks, with only five
mutational steps separating them from either NEP
(b) Molecular methods and analysis (GW18 and GW17) or NZ (GW31; figure 1), and
Genomic DNA was extracted by phenol–chloroform pro- six steps from AU (GW30 and GW32) haplotypes.
cedures, and the mtDNA control region amplified using Partial sequences (297 bp; accession nos HQ540299–
primers and conditions from Pardini et al.[18]. Amplified HQ540300) from the Sicilian skin sample also gave a
products were purified using QIAquick (QIAGEN) columns unique haplotype, but was separated by only three
and commercially sequenced. Work on the historical sample hypothetical mutational steps from an NEP haplotype
was undertaken in a shark-DNA-free laboratory, under (electronic supplementary material, figure S1).
Proc. R. Soc. B (2011)