Access to this full-text is provided by Frontiers.
Content available from Frontiers in Marine Science
This content is subject to copyright.
Gaining a (tube) foothold –
trends and status following two
decades of the long-spined
echinoid Diadema setosum
(Leske, 1778) invasion to the
Mediterranean Sea
Rotem Zirler
1,2
, Lynn Angele Leck
1,2
, Tamar Feldstein Farkash
1,2
,
Martina Holzknecht
3
, Andreas Kroh
3
, Vasilis Gerovasileiou
4,5
,
Mehmet Fatih Huseyinoglu
6,7
, Carlos Jimenez
8,9
,
Vasilis Resaikos
8
, Mehmet Baki Yokes¸
10
and Omri Bronstein
1,2
*
1
School of Zoology, Faculty of Life Sciences, Tel Aviv University, Tel Aviv, Israel,
2
The Steinhardt
Museum of Natural History, Tel Aviv University, Tel Aviv, Israel,
3
Natural History Museum Vienna
(NHMW), Vienna, Austria,
4
Department of Environment, Faculty of Environment, Ionian
University, Zakinthos, Greece,
5
Hellenic Centre for Marine Research (HCMR), Institute of Marine
Biology, Biotechnology and Aquaculture (IMBBC), Thalassocosmos, Heraklion, Crete, Greece,
6
Faculty of Maritime Studies, University of Kyrenia, Kyrenia, Cyprus,
7
Biosphere Research Center,
Istanbul, Türkiye,
8
Enalia Physis Environmental Research Centre, Nicosia, Cyprus,
9
Energy,
Environment and Water Research Center (EEWRC), The Cyprus Institute, Nicosia, Cyprus,
10
AMBRD
Laboratories, Istanbul, Türkiye
The Eastern Mediterranean Sea is an exceptional habitat. Its relative isolation and
distinct characteristics create a unique ecosystem recognized as a marine
biodiversity hot spot, where one-fifth of the species are endemic. Yet, native
Mediterranean biodiversity is under increasing threat, mainly due to massive alien
species invasions of Indo-Pacific origin. To date, more than 800 non-indigenous
species have been reported in the Eastern Mediterranean Sea, justifying its
reputation as one of the most severely affected habitats in the world in terms
of marine biological invasions. Here we summarized the Mediterranean invasion
dynamics of the long-spined echinoid Diadema setosum (Leske, 1778), one of
the most ubiquitous Indo-Pacific sea urchin species. We show an alarming
exponential population growth of D. setosum throughout the Eastern
Mediterranean since 2018, following more than a decade of ‘invasion lag’since
its first detection in 2006. Molecular analyses illustrate the presence of a single
genetic D. setosum clade in the Mediterranean Sea –corresponding to the
Arabian Peninsula clade of this species, reinforcing the notion of a Red Sea origin.
Our data support the current working hypothesis that the initial introduction of D.
setosum occurred in the Northern Levantine Basin from which it gradually
expanded in both north-west and south-east trajectories –in contrast to a
stepping-stone hypothesis of gradual advancement from the opening of the
Suez Canal. Demographic data of D. setosum along the Israeli Mediterranean
coastline reveals a well-established population of broad size distributions, from
juveniles to adult individuals of remarkably large size. Additionally, we provide
evidence of the reproductive capacity of D. setosum in its new environment. Due
Frontiers in Marine Science frontiersin.org01
OPEN ACCESS
EDITED BY
Pedro Morais,
Florida International University,
United States
REVIEWED BY
Jose
´Carlos Herna
´ndez,
University of La Laguna, Spain
Simone Farina,
Anton Dohrn Zoological Station Naples,
Italy
*CORRESPONDENCE
Omri Bronstein
bronstein@tauex.tau.ac.il
RECEIVED 27 January 2023
ACCEPTED 15 May 2023
PUBLISHED 31 May 2023
CITATION
Zirler R, Leck LA, Farkash TF, Holzknecht M,
Kroh A, Gerovasileiou V, Huseyinoglu MF,
Jimenez C, Resaikos V, Yokes¸MB and
Bronstein O (2023) Gaining a (tube)
foothold –trends and status following two
decades of the long-spined echinoid
Diadema setosum (Leske, 1778) invasion to
the Mediterranean Sea.
Front. Mar. Sci. 10:1152584.
doi: 10.3389/fmars.2023.1152584
COPYRIGHT
©2023Zirler,Leck,Farkash,Holzknecht,
Kroh, Gerovasileiou, Huseyinoglu, Jimenez,
Resaikos, Yokes¸andBronstein.Thisisan
open-access article distributed under the
terms of the Creative Commons Attribution
License (CC BY). The use, distribution or
reproduction in other forums is permitted,
provided the original author(s) and the
copyright owner(s) are credited and that
the original publication in this journal is
cited, in accordance with accepted
academic practice. No use, distribution or
reproduction is permitted which does not
comply with these terms.
TYPE Original Research
PUBLISHED 31 May 2023
DOI 10.3389/fmars.2023.1152584
to the magnitude of Diadema’s ecological footprint, it poses a severe threat to
the entire Eastern Mediterranean Sea, including the Levantine Basin and South
Aegean Sea, calling for rapid and coordinated action at both national and
regional scales.
KEYWORDS
Diadema, Levantine Basin, Mediterranean, non-indigenes species, population
outbreak, alien species, invasion dynamics
1 Introduction
The stability of ecosystems depends on their delicate food webs
and interactions among the fauna and flora of which they are
comprised (Pimm, 1984;McCann, 2000;Hooper et al., 2012;Loreau
and de Mazancourt, 2013). The frailty of such interactions is
revealed through dramatic changes in the local communities
following deviation from equilibrium (Downing et al., 2012;
Hooper et al., 2012). One of the many impacts of enhanced
anthropogenic activity is the intentional and coincidental transfer
of non-indigenous species (NIS) outside of their native distribution
range, which under certain circumstances may escalate to species
invasion (Galil, 2007). Consequently, invaded regions and their
native fauna may be severely affected (Vredenburg, 2004).
A typical invasion largely follows a sequential chain of events
(Figure 1) starting with (I) the dispersal of NIS beyond their native
range. This facilitates an initial introduction (II) which is normally
followed by a (III) lag-phase that may vary in length, during which the
new population is maintained at low densities in a restricted area, and
often remains undetected. Lag-phase duration may last for decades
before NIS become detectible (Crooks, 2005;Azzurro et al., 2016;
Zenetos et al., 2019). During establishment (IV), NIS population-
growth accelerates, their range expands, and they increasingly interact
with the new environment. During this phase, some NIS may turn
invasive as they reach exceptionally high abundances and trigger
undesired ecological and economic impacts (Mack et al., 2000). Under
certain circumstances, NIS populations may grow exponentially and
achieve exceptionally high abundances –further expanding their
range and changing the ecological equilibriums through competitive
exclusion of native species and exhaustion of natural resources. This
final stage is largely recognized as population outbreak (V) (Byers
et al., 2002;Blackburn et al., 2015).
The Eastern Mediterranean Sea is known as one of the most
invaded marine regions in the world (Katsanevakis et al., 2014b;
Tsiamis et al., 2020). Though the native biodiversity of the
Mediterranean Sea is characterized by a remarkable rate of
endemism (accounting for 20% of its species) and was therefore
referred as marine biodiversity hot spot (Coll et al., 2010;Gianni
et al., 2013;Katsanevakis et al., 2014a), the unique fauna of the
Mediterranean Sea is constantly threatened by a massive influx of
alien species of Red Sea (RS) origin (Galil, 2007;Coll et al., 2010;
Galil et al., 2017). The maritime connection between these two
adjacent seas was formed following the opening of the Suez Canal in
1869 (Coll et al., 2010;Soukissian et al., 2017). This manmade route
allows marine species to travel from the RS to the Mediterranean
(Galil and Goren, 2014)–a process termed Lessepsian Migration
(Por, 1971). Originally, the route through the canal was hindered by
a series of physical and biological barriers (such as the hyper-saline
barrier formed by the Bitter Lakes) that limited, at least temporarily,
the intensity of species migration between the RS and
Mediterranean (Aron and Smith, 1971;El-Serehy et al., 2014;El-
Serehy et al., 2018). However, most of these natural barriers eroded
over time, allowing direct (i.e., gamete propagation and active
locomotion) and indirect (i.e., human mediated transport)
transmission (Carlton, 1987). Consequently, the capacity for
successful invasions increased dramatically with time as reflected
by the mounting number of reports of new NIS of RS origin in the
Eastern Mediterranean (Galil and Goren, 2014;Galil et al., 2015)
currently including approximately 800 species (Rotter et al., 2020).
Recently, an ongoing migration is drawing increasing attention
as new evidence show alarming rates of population growth of the
long-spined sea urchin, Diadema setosum (Leske, 1778) in the
FIGURE 1
Schematic invasion curve specifying Diadema setosum invasion
timeline in the Mediterranean Sea. Following transport and
introduction to the new region, founder non-indigenous species
(NIS) populations remain restricted in both population size and
range. The invasion lag-phase occurs at the onset of NIS
establishment and is characterized by restricted spatial distributions
and low abundances, which are often undetected. As establishment
progresses, population size dramatically increases and includes
individuals at varying life stages. Ultimately, population outbreaks
may occur as NIS achieve exponential population-growth and range
expansion. Vertical dow-facing arrows indicate the year of first D.
setosum record for the respective countries.
Zirler et al. 10.3389/fmars.2023.1152584
Frontiers in Marine Science frontiersin.org02
Mediterranean Sea. Diadema setosum is one of the most
conspicuous Indo-Pacific shallow water echinoids (Bronstein and
Loya, 2014;Muthiga and McClanahan, 2020). The two recognized
genetic clades of the species are widely distributed throughout the
Indo-Pacific, ranging from the RS and Persian Gulf (clade b),
through the east coast of Africa in the Indian Ocean, to the west
Pacific, off the coast of Japan (clade a) (Lessios et al., 2001;Bronstein
et al., 2016;Muthiga and McClanahan, 2020). While diadematoids
are largely recognized as omnivores, they mainly feed by grazing on
algae scraped from hard substrates (Bronstein and Loya, 2014;
Muthiga and McClanahan, 2020). Consequently, Diadema spp., like
other members of this family, are potent ‘environmental engineers’,
capable of altering the structure and composition of entire benthic
communities (Hernandez et al., 2008;Ling et al., 2009;Bronstein
and Loya, 2014;Goh and Lim, 2015). As algal growth regulators,
Diadema grazing restricts algae proliferation, thereby supporting
the settlement and development of slower growing benthic
organisms –such as corals (Lawrence, 1975;Sammarco, 1982;
Bronstein and Loya, 2014;do Hung Dang et al., 2020).
Conversely, exceptionally high grazing intensities driven by large
Diadema populations, may drive degradation of benthic
communities (such as algae, macrophytes, and hard benthic
infrastructure; Carreiro-Silva and Mcclanahan, 2001), potentially
leading to collapse of entire habitats (Lawrence, 1975;Mokady et al.,
1996;Hernandez et al., 2008;Qiu et al., 2014).
The capacity of any given species to proliferate and reach high
abundances, largely depend on their ability to reproduce. For
invasive species in particular, one of the key challenges in
concurring new environments, is the ability to reach sexual
maturity and reproduce successfully, despite of potentially varying
environmental conditions with respect to their native range. While
reproduction was extensively studied across the genus Diadema (see
Muthiga and McClanahan, 2020 and references therein), and
several studies targeted RS populations of D. setosum (Pearse,
1970;Bronstein et al., 2016), none has so far targeted populations
from the Eastern Levantine Basin.
The first record of D. setosum in the Mediterranean Sea in 2006,
off the KasPeninsula, Turkey, (Yokes and Galil, 2006) was a
milestone in the successful establishment of large aggregations
currently occupying the entire Levantine Basin (Figure 2).
Here we follow the invasion dynamics of D. setosum in the
Mediterranean Sea, demonstrating its range expansion and recent
accelerated population growth reflecting clear signs of population
outbreak. We combine a thorough literature review with extensive
survey data and sampling complemented by citizen-science reports,
to provide a comprehensive report on the progress of this invasion
and depict its true scale. We use molecular data to identify the
genetic makeup of the invaders and determine their origin. We
show an alarming recent exponential increase in D. setosum
abundance in the Eastern Mediterranean and provide evidence of
effective reproductive capacity (i.e., the ability of an individual to
reproduce) in their new environment. Our results shed new light on
both temporal and spatial dynamics of biological invasions in the
Mediterranean Sea and facilitate better understanding of the life
history, invasion dynamics, genetic makeup, and reproductive
biology of the ubiquitous D. setosum.
2 Materials and methods
2.1 Field observations and
sample collection
Underwater surveys were conducted between 2016 and 2022
along the Greek, Turkish, Cypriot, and Israeli Mediterranean
coastline, and published data were obtained from the literature
and complimented by citizen reports (Table S1). Data from a total
of 670 reports comprising: 313 reports from scientific literature, 237
underwater survey dives by trained biologists, and 120 citizen
science reports, covering a depth range of 0-50 m and spanning
over 2,000 km of the Eastern Mediterranean Sea, were compiled.
Each identified specimen was noted, indicating the date of
observation, depth, substrate type, and precise location.
Data were used to construct two datasets: (1) ‘number of
individuals’–the total number of Diadema setosum individuals
encountered on a 45 min dive, and (2) ‘number of observations’–
each independent dive where Diadema setosum are observed was
counted as one, regardless of the number of individuals
encountered. At all dives, counts were conducted by a single
observer, regardless of the number of participants, and dives were
regarded as independent reports if conducted at different sites or on
different dates.
Considering the potential biases originating from citizen
science, strict criteria for evaluating these data were established.
Reports missing the exact number of individuals encountered, their
depths, and/or precise localities (coordinates) were omitted from
downstream analyses. In reports where the number of individuals
was ambiguous, values were determined following personal
communication with the reporter or else omitted. When depth
was given as range, mean values were calculated, and when
metadata were missing altogether, the observations were omitted
from the analysis. In total, 48 reports were omitted based on these
criteria. To avoid instances of misidentification of D. setosum by
FIGURE 2
Map of the Eastern Mediterranean Sea. Red dots (transparent red
symbols) represent sites of Diadema setosum observations compiled
over nearly two decades –from the detection of the first individual
in 2006 to the end of 2022. The map was created using QGIS v.
3.28.2 (QGIS Development Team, 2023).
Zirler et al. 10.3389/fmars.2023.1152584
Frontiers in Marine Science frontiersin.org03
untrained reporters, such data were included only if accompanied
by photographic evidence, allowing unambiguous identification
prior to inclusion in our dataset.
When possible, specimens were collected for morphological
examinations as well as molecular and histological analyses. In
total, 24 D. setosum specimens were collected along the Israeli
Mediterranean (specimens are deposited at the Steinhardt Museum
of Natural History collection, Tel Aviv University, Israel) (Table
S2). Twelve additional tissue samples of adult specimens were
collected in Crete, Greece, and five tissue samples were collected
from the native Red Sea population (Eilat, Israel), and processed as
described below.
2.2 Size distribution
Test diameters (measured at the ambitus) and heights of the 24
D. setosum specimens collected from the Israeli Mediterranean were
measured using a Mitutoyo 500-196-30 digital caliper to the nearest
0.01 mm (Table S2). A second distribution representing the native
Israeli Red Sea population was obtained from Bronstein’s
unpublished data collected between April 2007 and August 2008
in Eilat, Israel, including the sizes of 547 randomly selected D.
setosum specimens (Table S3). Both Israeli distributions were then
compared with the size distribution of the Greek population, which
was modified from Vafidis et al. (2021). The latter providing test
diameters of 160 randomly selected individuals, from the vicinity of
Dodecanese Islands, between December 2019 and July 2020.
2.3 Molecular analysis
Total genomic DNA was extracted from spine-muscles or
gonads using the DNeasy Blood and Tissue Kit (QIAGEN,
Hilden, Germany) following the manufacturer’s instructions. PCR
amplifications of two mitochondrial and one nuclear markers were
performed. Fragments of the mitochondrial (1) cytochrome c
oxidase subunit 1 gene (COI) and (2) Lysine-tRNA, ATPase-6
and ATPase-8 region (LYS), as well as the first exon region of the
nuclear (3) bindin gene (BIN) were amplified using the Hy-Taq
Ready Mix (2x) (Hylabs, Rehovot, Israel). Reaction conditions for
the COI fragment using the primers CO1f and CO1a (Lessios et al.,
2001) were: 3 min at 95°C followed by 35 cycles of 30 sec at 95°C, 30
sec at 58°C and 1 min at 72°C, ending with a final extension step of
10 min at 72°C (Table 1). Reaction conditions for the LYS fragment
using primers LYSa and ATP6b (Lessios et al., 2001) were 3 min at
94°C followed by 35 cycles of 30 sec at 94°C, 30 sec at 55°C and
1 min at 72°C, ending with a final extension step of 10 min at 72°C
(Table 1). Reaction conditions for the bindin fragment using the
primers DA5A and DAIR (Geyer et al., 2020) were 3 min at 94°C
followed by 40 cycles of 30 sec at 94°C, 30 sec at 51°C and 1 min at
72°C, ending with a final extension step of 10 min at 72°C (Table 1).
PCR products were visualized on a 1% agarose gel, purified using
ExoSAP-IT (Affymetrix) and sequenced in both directions at the
TAU sequencing facility. All sequences generated in the present
study were deposited in GenBank under accession numbers
MT430942-MT430943, MT434142-MT434143, MW387536,
MW394192, ON197106-ON197133, ON210773-ON210801 and
ON211043-ON211059 (Table S4).
Forward and reverse sequences of each locus were assembled,
inspected, and edited using SeqTrace (Stucky, 2012). Consensus
sequences were edited using AliView v.1.18 (Larsson, 2014) and
aligned using MAFFT v.7 alignment server (http://mafft.cbrc.jp/
alignment/server/), employing the E-INS-i algorithm. Ambiguous
positions were removed using TrimAl v.1.4 (Capella-Gutierrez
et al., 2009) and GUIDANCE2 (Sela et al., 2015), followed by a
final manual inspection.
Three datasets were created to facilitate further analyses: (1)
COI –comprising 108 COI sequences, 640 bp long, including all
publicly available sequences from the family Diadematidae (31
sequences generated in the current study); (2) LYS –comprising
172 Lysin-ATP6 Diadema spp. sequences, 580 bp long, representing
all extant species in the genus Diadema (20 sequences generated in
the current study); (3) BIN –comprising 157 sequences of the
nuclear bindin gene from all extant species of Diadema, 500 bp long
(29 sequences generated in the current study; Table S4).
Phylogenetic analyses were conducted using both Maximum
Likelihood (ML) and Bayesian Inference (BI) approaches following
Bronstein et al. (2017) and Bronstein and Kroh (2018). Briefly, a
heuristic search under the Bayesian Information Criterion (BIC)
(Schwarz, 1978), as implemented in PartitionFinder2 (Lanfear et al.,
2017) was employed to determine the optimal partitioning schemes
and models of molecular evolution. ML analyses were performed in
parallel with IQtree (Trifinopoulos et al., 2016) and raxmlGUI 2.0
(Edler et al., 2021). IQtree analyses uses ModelFinder
(Kalyaanamoorthy et al., 2017) to select the best fit models for
each partition, identifying the TN+F+G4 as the best-fit model
across all partitions for both COI and LYS datasets, while K2P
+G4 was better suited for the BIN datasets. Branch support was
evaluated using the ultrafast bootstrap on IQtree (UFBoot, 1000
TABLE 1 List of primers used for the molecular analysis.
Target gene Primer name Primer Sequence Tm (°C) Reference
Cytochrome c oxidase subunit 1 (COI) COIf
COIa
CCTGCAGGAGGAGGAGAYCC
TCATATTCGCAGACCCATCAG
66
59
Lessios et al., 2001
Lysine-tRNA, ATPase-6 and ATPase-8 region (LYS) LYSa
ATP6b
AAGCTTTAAACTCTTAATTTAAAAG
GCCAGGTAGAACCCGAGAAT
54
60
Lessios et al., 2001
Bindin (BIN) DA5A
DAIR
GATTTCTTTATGGGACATCGCAA
TCCGCACTGATGGTATCGTC
59
60
Geyer et al., 2020
Zirler et al. 10.3389/fmars.2023.1152584
Frontiers in Marine Science frontiersin.org04
replicates) as well as standard bootstrap (BS, 1000 replicates).
Additional ML analysis were performed with raxmlGUI 2.0
applying the settings ‘ML + thorough bootstrap’, 100 runs, 1000
replicates, using TrN+G4 (for COI and LYS datasets) and K80+G4
(for BIN dataset) as the best-fit models for all partitions as inferred
from PartitionFinder2. Bayesian analysis was carried out using
MrBayes v. 3.2.2 (Ronquist et al., 2012). We ran two independent
runs of three ‘heated’and one ‘cold’chain for 10 million
generations, sampling parameters and trees every 100
th
generations. The runs were inspected with Tracer 1.7.1 (Rambaut
et al., 2018) to assess convergence. In a conservative approach, the
first 25% of trees were discarded as burn-in, and a 50% majority-
rule consensus tree was calculated from the remaining trees.
Posterior Probabilities (PP) were obtained from the 50%
majority-rule consensus of the trees sampled during the
stationary phase.
2.4 Reproductive biology
To evaluate D. setosum’sreproductivecapacityinthe
Mediterranean, 12 individuals were sampled from Plakias (35°
09’16.9992”N 24°26’30.0012”E), in southern Crete (Greece),
during September 2020, from depths of 10-13 m. Export of
material was carried out under a Material Transfer Agreement
between HCMR and NHMW dated September 9
th
2020. Seven
additional samples were collected on different occasions from the
Israeli population between December 2019 and September 2021. To
facilitate a comparison between the Mediterranean population and
the native Red Sea population, five additional D. setosum specimens
were sampled from Eilat, Israel (29°30’6.966”N 34°55’3.3924”E)
during September 2020, from depth of 5-10 m. Given the rapid
decline of Red Sea D. setosum abundance over the past decade
(Eviatar and Bronstein, in prep), strict regulations on destructive
sampling have been imposed throughout the region, limiting
sample availability. Data on localities, sampling dates, moon-
phase, sex, and size of individuals used for the reproductive state
evaluation are provided in Table S5.
A single gonad from each specimen was extracted, fixed in
Bouin’s solution, and prepared for histological analysis following
the procedures described in Bronstein and Loya (2015) and
Bronstein et al. (2016).Briefly, following fixation, tissue was
embedded in paraffin, sectioned to 7 µm using a Shandon M1R
microtome, and stained using a standard Hematoxylin and Eosin
protocol. Stained slides were examined under a Nikon Eclipse Ni-U
light microscope to determine their reproductive stage. We followed
the four-stage system of Bronstein et al. (2016) to describe D.
setosum’s reproductive cycle: Stage I (spent): Gonads are largely
devoid of contents showing ova-free lumen in females and
spermatozoan-free lumen in males. A thin layer of nutritive
phagocytes (NPs) is present along the ascinal walls in both sexes
and may form a pale meshwork across the ascinus. Strongly
basophilic previtellogenetic oocytes or primary spermatocytes,
staining dark purple with Hematoxylin and eosin, are present
along the ascinal wall. Stage II (recovering): NPs proliferate from
the gonad ascinal wall, gradually filling the lumen of both ovaries
and testis. Limited groups of primary spermatocytes and clusters of
previtellogenetic oocytes start appearing in the testicular and
ovarian germinal epithelia, respectively, and may occasionally
project centrally. Stage III (growing): Both early and late
vitellogenetic oocytes may be present along the ovarian wall. All
stages of germ cells are evident in the male germinal epithelium and
continuously increase in number as new spermatogonia develop
basally while spermatocytes migrate to the testicular lumen, where
they accumulate as mature spermatozoa, forming visible columns of
darkly stained cells. NPs deplete and progressively occupy less space
in both males and females. Stage IV (mature): By the end of this
stage the NP layer in both ovaries and testes is largely exhausted.
Ovaries are packed with mature ova, while oocytes at different
maturation stages may still be evident in the germinal epithelium.
The testicular lumen is densely packed with spermatozoa.
Occasionally some ova and spermatozoa may be evident in
the coelom.
2.5 Statistical analyses
Statistical analyses were performed using R (RStudio Team,
2020). As data representing the relationship between years and the
number of observations or the number of individuals were counts,
and the variance of the count data was greater than the mean, we
used a generalized linear model (GLM) with a Poisson distribution
and a log-link function (to account for overdispersion) using the
stats package (R Core Team, 2022). ‘Year’was determined as the
independent variable, and the number of observations/individuals
was determined as the response variable. An a+2 offset was applied
to all count data. Pairwise comparisons were conducted between
each possible combination of years, for both the number of
observations and number of individuals, using the Tukey method
(implemented in the R package multcomp;Hothorn et al., 2008). p-
values were corrected for multiple comparisons using the
Bonferroni correction. Pairwise Kolmogorov-Smirnov tests were
performed to check for differences in size frequency distributions
between RS, Israeli Mediterranean, and Greek (Crete)
Mediterranean populations using the R package dgof (Arnold and
Emerson, 2011). Due to multiple testing for the size frequency
comparisons, p-value was corrected using the Bonferroni
correction. To test for the differences in male-female ratios, Chi-
square tests were performed.
3 Results
3.1 Demography
Reports on the presence of D. setosum in the Mediterranean Sea
increase with time since the original record in 2006 (Figure 3). By
the end of 2017, the number of new observations remained
consistently low, comprising no more than 13 annual
observations, and an average of 0-2 individuals per report, to a
total of 52 observations since the first report in 2006 (Table S1).
Between 2018 and 2020 a significant increase in number of
Zirler et al. 10.3389/fmars.2023.1152584
Frontiers in Marine Science frontiersin.org05
observations was noted (Generalized linear model, df=1, p> 0.001;
Table S6), summing up to a total of 268 reports (Table S1). Between
2020 through the end of 2022, a significant exponential increase in
the number of observations occurred (Generalized linear model,
df=1, p<0.001; Table S6 ;Figure 3). In 2018, when the significant
increase was first recorded, the sum of documented reports reached
74 (44 of which from the vicinity of Kas), and by 2021, a total of 205
observations were made throughout the Levantine Basin (Figure 3;
Table S1). A non-significant decrease in number of observations
was noted in 2022, totaling 145 annual observations (46 of which in
Kas). Prior to 2018, Cyprus held the highest number of annual
observations, however, since 2018, the majority of observations
shifted to Greece (Figure 3). Overall, Greece holds the highest
proportion of observations, accounting for 32% of the total
observations (Figure 3).
When considering the annual number of individuals (Figure 3),
a similar trend emerges. Between 2006 to 2014, most observations
were of single individual, totaling 36 reported individuals during
this period. From 2015 to 2017, there was a slight (yet non-
significant) increase in the number of reported individuals, with
the total number almost doubling within only three years (Figure 3).
A significant shift in trend occurred in 2018, during which a total of
757 individuals were observed (Generalized linear model, df=1,
p<0.001; Table S6;Figure 3), 428 of which from the vicinity of Kas.
In 2019, the number of individuals further increased (Generalized
linear model, df=1, p<0.001; Table S6), resulting in a threefold
increase in comparison to 2018 (n=2,563; Figure 3). A further
significant increase in the number of individuals occurred in 2020
(Generalized linear model, df=1, p<0.001; Table S6), totaling 3,552
documented individuals, reflecting an exponential growth in D.
setosum Mediterranean populations (Figure 3). Despite an increase
in the number of reports between 2020 and 2021 (n=122 and
n=205, respectively), 2021 was marked by a decrease in the number
of observed individuals (n=1847; Figure 3). Still, by the end of 2022,
the number of D. setosum individuals reached a record of 18,512
individuals (Generalized linear model, df=1, p<0.001; Table S6).
Overall, since the onset of invasion in 2006, the majority of
individuals were spotted around the coasts of Turkey (n=19,034)
and Greece (n=4,853).
Due to potential biases rising from unequal sampling efforts, a
dataset restricted to reports from Kas, Turkey, was constructed,
applying the same analyses for number of observations and number
of individuals, as conducted for the total count data. This dataset
was further restricted to data obtained through routine scientific
surveys. These reports contribute the majority of individuals
(63.5%) reported in this study. Furthermore, these reports were
collected under a controlled sampling design of routine annual
surveys (reporting presence and absences) and facilitated the widest
temporal comparison (with reports dating back to the first
observation in 2006). Results from these analyses were identical
to the results obtained for the pooled data (Generalized linear
model, df=1, p<0.001) for both number of observations and number
of individuals; Table S7 and S8, validating reported trends.
3.2 Population structure
3.2.1 Size distribution
Body sizes of Diadema setosum varied significantly within and
between Mediterranean and RS localities (Figure 4), with all
populations compared containing both juvenile and adult
individuals. Size-frequency distribution (SFD) of the Israeli
population showed specimens with test diameters ranging from a
minimum of 11.8 mm to a maximal test diameter of 97.03 mm (Table
S2). The majority of specimens (54%) were small-sized (test diameters
of 0–40 mm), with medium-sized (40–70 mm) and large-sized (70–
100 mm) individuals comprising 25% and 21% of the population,
respectively. Similar SFDs were measured in Greece (modified from
Vafidis et al., 2021; based on test diameter measurements of n=160
individuals sampled off the Dodecanese Islands) (Kolmogorov-
Smirnov; p=0.0964). Similar to the Israeli Mediterranean
population, very small (10–20 mm) and very large (80–90 mm)
individuals were the least frequent (1% and 4%, respectively). In
contrast, SFD of the Israeli population seems to be more biased
towards the smaller size groups (10–20 and 20–30 mm) than the
Greek population. Nevertheless, both Greek and Israeli populations
showed a high proportion of juveniles compared to adults. SFD of D.
setosum from its native RS range differed significantly from both
Israeli (Kolmogorov-Smirnov; p=0.0211) and Greek (Kolmogorov-
Smirnov; p<0.0001) Mediterranean populations. Specimens in the RS
FIGURE 4
Size-frequency distributions of Diadema setosum at three localities:
a native population at the Gulf of Aqaba, Red Sea (Is-RS; n=547),
measured between 2007 and 2008; Israeli Mediterranean
population (Is-Med; n=24), measured between 2019 and 2021; and
Greek population (Gre-Med; n=160), measured between 2019 and
2020 (modified from Vafidis et al., 2021).
FIGURE 3
Annual Diadema setosum number of observations and number of
individuals in the Mediterranean Sea between 2006 and 2022. Color
indicates the country from which the report originates. Column
pairs (per year) represent number of observations (left) and number
of individuals (right), and correspond to the blue and red scales,
respectively.
Zirler et al. 10.3389/fmars.2023.1152584
Frontiers in Marine Science frontiersin.org06
haddistinctlysmallerbodysize,rangingfrom1to65mm.Uniquely,
the RS was the only locality where individuals smaller than 10 mm
(7.5%) could be observed. The RS population was strongly skewed
towards smaller individuals with most individuals ranging between 10
to 30 mm in diameter. The occurrence of individuals larger than
50 mm in the RS was negligible (1.3%).
3.3 Reproductive biology
Similar male-female ratios were measured in both Crete (Cre-
Med; 7 males, 5 females; Chi-square test, X
2
=0.333,df=1, p-
value=0.56) and Israeli Mediterranean coastline (Is-Med; four males,
threefemales;Chi-squaretest,X
2
= 0.1428, df=1, p-value=0.7054). The
reproductive state of the Cretan specimens (collected in mid-
September; Figures 5A–F) appeared to be highly synchronous –
showing growing gametes (stage 3), with some individuals having
nearly mature oocytes (Figures 5E,F). This sexual synchronization
was evenly distributed between males and females (Table S54). All but
one of the Israeli Mediterranean specimens (collected during May, and
from September to December) were at the recovery stage (stage 2;
Figures 5C,D). The only exception was specimen DS21, a male at the
growing stage (stage 3), collected in August 2021.
3.4 Molecular analysis
Sequences of the mitochondrial COI gene of all Diadema
species as well as other members of the family Diadematidae were
obtained from public databases and included in the current analyses
(Figure 6). Complementary to the COI dataset, the mitochondrial
LYS gene dataset holds a higher number of Diadema spp. sequences
available for comparison with the Mediterranean sequences,
although it lacks the representation of D. clarki and other
diadematoids (Figure S1). The bindin gene dataset facilitates the
phylogenetic reconstruction of all extant Diadema species based on
a nuclear gene, thus complementing the two abovementioned
mitochondrial genes (Figure S2). As phylogenetic analyses
generated congruent topologies, only ML trees are presented –
showing both bootstrap support values and posterior probabilities
for the respective nodes.
All analyses, across methodology and loci, ascribe the Israeli
specimens to clade b of D. setosum (Figures 6, S1 and S2). In
agreement with previously published results (Lessios et al., 2001;
Bronstein et al., 2017;Geyer et al., 2020), D. setosum clade b was
resolved as sister to D. setosum clade a in all datasets. However,
while both COI and LYS support the early divergence of the former
two clades from the other Diadema species, the BIN dataset suggests
that the divergence of D. clarki and D. palmeri predated that split. In
agreement with the results of Lessios et al. (2001) and Bronstein
et al. (2017), both COI and LYS analyses support similar topologies
with extant Diadema species retrieved as monophyletic and highly
supported. Within the BIN dataset, only three of the eight
recognized species of Diadema were resolved as monophyletic
clades (D. palmeri,D. clarki and both clades of D. setosum), while
the other species remained unresolved (Figure S2).
4 Discussion
4.1 Molecular diagnosis
Two former studies on Mediterranean D. setosum included
genetic data in their reports (Bronstein et al., 2017;Bronstein and
FIGURE 5
Histological cross-sections of Diadema setosum gonads, representing the four reproductive stages used for classification (following Bronstein et al.,
2016). Ovaries (A, C, E, G), and testes (B, D, F, H) are presented. Stage 1 (spent; A,B): the post-spawned gonads are devoid of content, showing
mostly free lumen in both sexes. Stage 2 (recovering; C,D): nutritive phagocytes (NPs) are proliferating, filling the acinal lumen. Primary oocytes (C)
and spermatocytes (D) might be visible on the germinal epithelia. Stage 3 (growing; E,F) the maturating oocytes (E) migrate toward the lumen and
increase in size as vitellogenesis (the process during which the ova store nutrients to support the future larvae) advances. Similarly, in males,
spermatocytes migrate towards the lumen as they mature (F), while new spermatogonia keep forming along the germinal epithelium. As gametes
proliferate and grow, NPs are continuously depleted and become less abundant. Stage 4 (mature; G,H): the gonads of both sexes are packed with
mature oocytes (G) and spermatozoa (H), while NPs are largely exhausted. Photographs (B,G,H) were modified from Bronstein et al. (2016) to
illustrate the complete reproductive cycle of D. setosum. Scale bars represent 100 mm.
Zirler et al. 10.3389/fmars.2023.1152584
Frontiers in Marine Science frontiersin.org07
Kroh, 2018). Still, these studies only utilized two samples –the
original specimen collected near Kas(Turkey) in 2006, and the
first confirmed specimen from Israel (2017). The additional data
provided in the current study (24 new specimens), as well as the
inclusion of a nuclear marker, confirm the identity of
Mediterranean Diadema spp. as D. setosum clade b, the clade
native to the RS and Persian Gulf. Owing to the geographical
proximity of the RS and Mediterranean and given the favorable
conditions for clade b in the Eastern Mediterranean (Bronstein
et al., 2017), migration of the latter seemed inevitable. Current
data supports the notion of a single genetic clade of Diadema
(clade b) in the Mediterranean Sea. The Indo-Pacificcladea
seems to be absent in the Mediterranean at this time –in
agreement with the habitat suitability model of Bronstein et al.
(2017). Whether the present Mediterranean population represent
the descendants of a single invasion event prior to 2006, or the
consequent of multiple, periodic invasions, remains to be
answered. Molecular population genetics may help resolve this
question once sufficient samples are obtained from across the
Eastern Mediterranean.
FIGURE 6
Phylogenetic relationships of the family Diadematidae based on the COI dataset. Maximum likelihood topology is displayed (major clades collapsed),
representing all known species of Diadema as well as all publicly available sequences of additional members of the family (Centrostephanus
longispinus,Astropyga pulvinate,Astropyga radiata,Echinothrix calamaris, and Echinothrix diadema). All Mediterranean Diadema sequences as well
as RS sequences generated in the current study were clustered with D. setosum clade b by both analyses (ML and BI). Bootstrap support values
(<65%) and posterior probabilities (<0.65) are shown above nodes, before and after the slash, respectively. Specimens corresponding to GenBank
accession numbers LC037355/56/57 (marked with asterisk) were referred to in the literature as Diadema sp. and deposited under D. setosum,
however, based on the evidence provided in Chow et al. (2016), these specimens clearly belong to D. clarki. Details on the sequences used to
generate the tree are given in Table S4. Scale bar reflects number of changes per site.
Zirler et al. 10.3389/fmars.2023.1152584
Frontiers in Marine Science frontiersin.org08
4.2 Invasion dynamics
Given the conspicuous appearance and non-cryptic nature of D.
setosum, early detection of this species in the Mediterranean was
possible. Moreover, the hard substrate preference, and depth range
preference of Diadema, reaching the shallow depths of the lower
intertidal, makes it readily detectible at highly monitored sites along
the Mediterranean coastline. Consequently, the first report of this
species in 2006 (Yokes and Galil, 2006), likely occurred shortly
following the species arrival to the Mediterranean, offering a rare
opportunity to study the dynamics of marine invasions, from arrival
to population outbreak, in a natural setup.
Since the first detection of D. setosum off Kaspeninsula, Turkey
(Yokes and Galil, 2006), evidence for the successful establishment of
the species have been accumulating. What was originally
anecdotical observations of isolated individuals or small
aggregations (Gökoglu et al., 2007;Crocetta et al., 2015;Gökoglu
et al., 2016;Gerovasileiou et al., 2017;Bronstein and Kroh, 2018;
Galanos and Kritikos, 2019;Katsanevakis et al., 2014a;Nader and
Indary, 2011;Turan et al., 2011;Tsiamis et al., 2015;Mytilineou
et al., 2016;Yapıcı, 2018), has grown to aggregations of dozens
(Bilecenoglu et al., 2019;Katsanevakis et al., 2020;Ammar, 2021)
and hundreds (Gül and Aydin, 2021;Vafidis et al., 2021)
of individuals.
The exponential increase in observations and number of
individuals since 2018, reflect a rapid spread and range expansion
of D. setosum throughout the Eastern Mediterranean and
Southeastern Aegean Sea (Figure 3). During this time, large
aggregations of individuals became common, mostly around
Turkey, Greece, and Cyprus, marking the shift from
establishment to outbreak (Figures 1,3). The pattern of
prolonged periods of low densities upon NIS arrival followed by
an exponential increase in abundance, is typical of many biological
invasions (Figure 1) (e.g., Aikio et al., 2010;Blackburn et al., 2015;
Azzurro et al., 2016). During the initial ‘invasion lag’the new NIS
population is small and represented by only few individuals from
the founder population. Routine, intensive marine surveys
conducted annually since 2015 by Israel Nature and Parks
Authority (INPA) along the Israeli Mediterranean coastline, did
not report the presence of D. setosum between 2015 and 2017
(Lazarus et al., 2020). Therefore, the exponential increase in number
of observations in the following years, under identical sampling
effort, can only be attributed to the rapid proliferation of D. setosum
populations. Nevertheless, coordinated, regional scale monitoring is
needed to depict the fine scale dynamics of this invasion.
The primary requirement of all living organisms is the ability to
feed and experience environmental conditions within the species
tolerance range. In this respect, the Eastern Mediterranean,
characterized by high water temperatures and solar radiation (in
comparison to the western basin) causing high rates of evaporation
(Bethoux, 1979;Coll et al., 2010;Schroeder et al., 2016;Soukissian
et al., 2017), has recently been shown to be highly suitable for clade
bofD. setosum (Bronstein et al., 2017). In turn, these higher
temperatures and increased evaporation, drive elevated salinity,
forming a longitudinal salinity gradient being highest in the
southeast and gradually dissipating to the north and west (Coll
et al., 2010;Borghini et al., 2014;Soukissian et al., 2017).
Subsequently, the lower salinity near Gibraltar (36.2 psu;
Soukissian et al., 2017) resembles the salinity of the northeastern
Atlantic, whereas the higher salinity of the southeastern Levant
(38.6 psu; Soukissian et al., 2017), resembles the salinity of the
northern RS (40.7 psu; Biton and Gildor, 2011). Moreover, the
completion of the Aswan dam in 1965, restricted the seasonal
flooding of the Nile delta, playing a major role in the overall
reduction of Levantine productivity. As the Nile discharge drives
large amounts of mud and silt, the dam formation severely limited
nutrient drift into the Eastern Mediterranean, decreasing regional
productivity by ten folds (Azov, 1991). Consequently, a gradient in
habitat suitability for D. setosum now occurs in this region, being
most suitable in the Southeastern Levantine Basin, and decreasing
to the north and west (Bronstein et al., 2017). As elevated
temperatures have been shown to favor both settlement and
post-settlement survival of Diadema spp. (Hernandez et al.,
2010), the significantly higher abundances of D. setosum under
the less favorable conditions of the northern Levant (in
comparison to the Southeastern Levant), support the assumption
of gradual establishment starting in the north and gradually
expanding southwards.
The absence of native shallow water (<40m) diadematoids in
the Mediterranean suggests little direct niche competition with
potentially similar species. The only native Mediterranean
diadematoid, Centrostephanus longispinus, is more common at
depths of 40 m to over 200 m (Pawson and Miller, 1983;
Koukouras et al., 2007;Katsanevakis et al., 2017), with the highest
densities recorded at depths of 60–130 m (Furioand Templado,
2012). Still, bathymetric isolation between the native C. longispinus
and invasive D. setosum necessitate further investigation, as adult D.
setosum specimens were recorded down to depths of 49 m
(Katsanevakis et al., 2020) and even 55 m on coralligenous
patches off Cyprus (Katsanevakis et al., 2020). At the shallow end,
a decade long decline of native shallow-water echinoids in the
Eastern Mediterranean, namely Paracentrotus lividus and Arbacia
lixula, reached the point of near local extinction at some localities
(Yeruham et al., 2015;Rilov, 2016). Under these circumstances, the
addition of a strong echinoid competitor to the largely vacant niche
is likely to aid the expansion of D. setosum, and further restrict the
recovery of native species.
Following successful establishment, the founders of NIS must be
capable of reproducing in the new environment in order to
maintain sustainable populations over time. Our data based on
histological analyses of Mediterranean D. setosum gonads, show
evidence for reproductive capacity in populations comprising
sexually adult males and females at both Israeli and Greek
populations (Figures 5G,H;Table S5), similar to the evidence
recently provided from the Dodecanese Island complex (Vafidis
et al., 2021). Furthermore, the presence of a wide size distribution
among Mediterranean D. setosum, despite being skewed towards
juveniles, clearly show individuals of different ages, providing
further support for the capacity of D. setosum to successfully
reproduce in its new habitat, as reoccurring recruitment from
sources outside the Mediterranean at these scales is highly
unlikely. As the success of reproduction through broadcast
Zirler et al. 10.3389/fmars.2023.1152584
Frontiers in Marine Science frontiersin.org09
spawning is highly dependent on abundance of spawning
individuals (Figures 5G,H;Table S5), growing populations
gradually increase the chances of successful fertilizations, leading
to the formation of larger larval stocks. Consequently, a higher
number of juveniles will potentially be recruited, accelerating the
process with time, as these juveniles grow to sexual maturity.
Current evidence suggest population outbreak of D. setosum in
the Eastern Mediterranean, being most evident around Turkey and
Greece, and likely spreading south along the Eastern coast of the
Levant towards Lebanon and Israel (Figure 3). Yet, the true scale
and rate of progression from the assumed point of origin in Turkey,
is hindered by uneven sampling efforts, reflected, for example, by a
single published observation from Lebanon in 2009 (Nader and
Indary, 2011). Biased sampling effort is also evident in the drop in
reported numbers of individuals between 2020 and 2021 (Figure 3).
To the south, D. setosum seems to have recently (2021) expanded to
Egypt and Libya (Nour et al., 2022). Interestingly, despite their
proximity to the Mediterranean opening of the Suez Canal, Libya
and Egypt are the latest countries to report the presence of D.
setosum in the Mediterranean. West of the Suez Canal, NIS have to
disperse upstream in the prevailing currents, inevitably slowing
down their westbound progression (Koukouras et al., 2010). These
reports provide further support to the prevailing notion of an initial
introduction in Turkey (likely facilitated by man-mediated
transport) followed by gradual counter current (Pascual et al.,
2017) spread to the south (Bronstein et al., 2017). Considering D.
setosum’s delayed southern expansion, further establishment,
population growth and potentially outbreak in southeastern
Mediterranean countries seems highly probable. To the west, D.
setosum may be reaching the edge of its biological range (following
the habitat suitability model of Bronstein et al., 2017), potentially
providing an explanation to the lack of observation of this species
west of Zakynthos, Greece, Ionian Sea (Dimitriadis et al., 2023),
despite the close geographic proximity and favoring surface
currents from the assumed point of introduction in Kas.
Maritime traffic plays a major role in transporting NIS to new
territories by ballast water transport or as fouling communities on
ship hulls (Williams et al., 1988;Lavoie et al., 1999;Costello et al.,
2022). Numerous organisms and larvae migrate within ship ballast
water, consequently being introduced to new, sometimes distant
habitats. The Suez Canal, one of the world’s busiest marine trade
routes (Figure 7), provides easy access for alien Red Sea fauna
entering the Mediterranean Sea. D. setosum initial detection in the
northeastern Levant, 600 km away from Port Said (Egypt), at the
Mediterranean side of the Suez Canal, implies that introduction was
mediated by ballast water transport, as previously suggested by
Yokes and Galil (2006) and Bronstein et al. (2017). Of the three
main westbound shipping routes off Port Said, the northernmost
one travels to the Aegean Sea, passing tightly south of Rhodes, a
short distance from the site of first detection off Kas(Figure 7).
According to this scenario, from this initial point of introduction,
the species gradually extended its range, first to the west –being
aided by regional currents, and later, at a slower pace, to the south,
working its way against the prevailing counter-clockwise surface
currents in the Levant Basin (Millot and Taupier-Letage, 2005). The
elaborated network of regional shipping routes in the Eastern
Mediterranean likely contributed to D. setosum’s local expansion.
4.3 Insights on invasion dynamics from NIS
population structure
4.3.1 Size
As demonstrated by the Israeli Mediterranean population, size
distribution is multimodal and spans from juveniles to markedly
large individuals (Figure 4;Table S2), characteristic of a population
with ongoing recruitment and supply of settling larvae (Pecorino
et al., 2012). While size-age correlations are highly debated in
echinoderm research (Bluhm et al., 1998),itmayverywell
support a qualitative estimation of age. Hence, specimens at a size
range of 10–20 mm (e.g., DS4, DS8, and DS18; Table S2) can be
estimated as being several months old (Lewis, 1966), while
individuals with test diameters of 80–100 mm (such as DS3, DS5,
and DS6; Table S2) as being at least several years old. Estimates of
the Israeli SFD are based on a limited dataset and clearly more data
is needed to capture the true size structure of this population.
Nonetheless, the presence of individuals across the species size
range, suggests that D. setosum are capable of maintaining stable,
sustainable populations in the Mediterranean, demonstrating
ongoing recruitment and survival of offspring –most likely
originating from sexually reproducing local populations.
Ebert (1982) determined the maximum diameter of D. setosum
from Eilat (clade b) to be 83.57 mm –1.6 times larger than the Indo-
pacific clade a (Ebert, 1982;Muthiga and McClanahan, 2020). As
such, some of the unusually large Mediterranean specimens
(Figure 4,Table S2), reaching an unprecedented large size of
97.03 mm (DS3) may very well be a decade (or more) old. The
slow initial growth rate of D. setosum relative to other Diadema
species such as D. savignyi (Muthiga and McClanahan, 2020),
makes the presence of such large specimens even more surprising.
It reflects the high suitability of the new habitat for D. setosum
(Bronstein et al., 2017), which encourages body size enhancement.
Furthermore, the presence of such large, potentially decade old
FIGURE 7
Map of main shipping routes in the Eastern Mediterranean Sea
during 2021. Data obtained from https://www.marinetraffic.com/.
Color gradient scale represents –routes/2.45km
2
/year.
Zirler et al. 10.3389/fmars.2023.1152584
Frontiers in Marine Science frontiersin.org10
individuals, suggest that the true date of D. setosum’s arrival to
Israel, predates (potentially by several years) current confirmed
reports (Bronstein and Kroh, 2018). Similarly, the first specimen
collected in Kasin 2006 was 58 mm, and therefore already several
years old. Though the introduction of adults cannot be ruled out at
this stage, ballast water transport of larvae is still the most likely
means of transport. Under such circumstances, the presence of such
large specimens indicates a minimum two-decade long presence of
D. setosum in the Mediterranean Sea.
Further insights might be gained by comparing size frequency
distributions between the native population in the RS, the more
established Mediterranean population in the Aegean Sea and the
proliferating southern Levant population in Israel. Indeed, the
Greek population displayed size distributions similar to the Israeli
Mediterranean population, showing elevated frequencies of young
individuals in the size range of 10 to 30 mm (Figure 4). The
abundance of juveniles (compared to adults) at both Greek and
Israeli populations, suggest high larval supply for these populations
followed by high mortality rates post recruitment. While the drivers
of these selective adult mortalities are currently unknown, elevated
predation of the less cryptic adults may provide some explanations.
In contrast to the Mediterranean, the native RS population is
comprised of significantly smaller individuals (Figure 4). Moreover,
new recruits at a size range of 0–10 mm were only identified in the
RS, where the preferred recruitment sites, and early-life growth
niches of D. setosum, are within crevices and under peddles, rocks
and coral debris at shallow depths down to 1 m (Bronstein pers.
obs.). While the early recruits’size-class is inevitably present also in
the Mediterranean, its location is currently unknown. The bell-
shaped distribution of RS D. setosum suggests strong selective
pressure against both very small and very large individuals. Small
individuals are more susceptible for predation (McClanahan and
Kurtis, 1991;Clemente et al., 2007), whereas large individuals are
likely to accumulate fatal illnesses that eventually lead to mortality
(Jones, 1985;Buchwald et al., 2015). Thus, the enhanced body size
in Mediterranean populations may indicate the current lack of such
selective forces in the Levant. The lower abundance of potential
Diadema spp. predators (see below), species-specific illnesses,
currently present in the new habitat, is probably beneficial for D.
setosum populations, supporting enhanced body size and longevity,
which in turn increases the species reproductive potential and
accelerate the species proliferation.
4.3.2 Depth range
In its native RS range, D. setosum largely occupy shallow
subtidal depths down to 20 m, although most populations inhabit
much shallower depth of 0 to 10 m (Muthiga and McClanahan,
2020). The bulk of Levantine D. setosum observations were reported
from depths of 0–10 m (Table S1), in agreement with previously
determined depth range for native D. setosum (Muthiga and
McClanahan, 2020). Yet, about 11.5% of the Mediterranean
reports seem to exceed this range, occurring at depths of 20–30
m, occasionally in clusters of dozens, and 3.5% of reports come
from depths between 35 and 55 m. Several ecological factors may
drive this habitat preference pattern. The presence (or absence) of
predators may provide a convincing explanation to this deeper-
water shift. For instance, the fish species Diplodus sargus and
Balistes capriscus, which are known as Diadema spp. predators at
depths of 5–20 m in the Canary Islands, are also present the
Mediterranean Sea (Clemente et al., 2010;Clemente et al., 2011;
Kacem and Neifar, 2014;Exadactylos et al., 2019;Muthiga and
McClanahan, 2020). The invasive fish Lagocephalus sceleratus and
Torquigener hypselogeneion were also reported to feed on Diadema
spp. in the Southeastern Aegean Sea (Ulman et al., 2021; Jimenez
pers. Obs.; Huseyinoglu, pers. Obs.), and while the depth range of
this species is now documented down to 220 m, it is more abundant
at shallow depths of up to 10 m. The presence of such shallow water
predators in the Mediterranean may drive D. setosum out of its
preferred depth range to waters deeper than 20 m to avoid
predation. The capacity of Diadema spp. to occupy greater depths
in the Mediterranean may also explain, at least in part, some of the
lag phase of the invasion. Dives deeper than 20 m are scarce along
the Israeli Mediterranean coastline. In addition, D. setosum has
been also recorded in cryptic environments such as the shadowy
entrance of marine caves in Greece, Cyprus and Turkey (Digenis
et al., 2022;Gerovasileiou et al., 2022; Ragkousis et al., in press).
Given the chance that the first individuals spent their initial
establishment phase at limited-accessed locations and depths,
they might have simply been overlooked. Upon establishment,
they proliferated, gradually occupying shallower depths, and
became detectible. While the factor driving this ecological release
is currently unknown, the hypothesis of detection-avoidance-by-
depth gains further support by the most recent report of D. setosum
in Libya, reporting the presence of two individuals at a depth of
25 m (Nour et al., 2022).
4.4 Reproductive biology
The reproductive cycle of native Diadema setosum in the RS was
intensively studied by Bronstein et al. (2016), illustrating high
proportions of sexually mature individuals between July and
October (Figure 8). As such, sampling in the current study was
conducted during September –at the peak of the expected annual
reproductive cycle (Pearse, 1970;Bronstein et al., 2016). In Crete,
92% of the gonads sampled were at the advanced growing stage. The
almost uniform synchroneity of the Mediterranean population (for
both sexes), are congruent with the findings of Bronstein et al.
(2016) and support the ability of D. setosum to reproduce in its new
Mediterranean habitat (Figure 8). Like the Crete-Mediterranean
population, the RS population seemed synchronized, however, the
reproductive cycle at these two locations, varied markedly during
fall of 2021. In contrast to the sampling of the Greek population,
samples collected along the Israeli Mediterranean were compiled
over several months. These specimens demonstrate the presence of
both sexes, at the recovering stage (Bronstein et al., 2016), asserting
the reproductive capacity of the local population. Further
confirmation for D. setosum’s successful reproductive capacity
along the Israeli Mediterranean comes from specimen DS21,
demonstrating active gametogenesis during the growing stage
(following Bronstein et al., 2016). DS21 was sampled in early
Zirler et al. 10.3389/fmars.2023.1152584
Frontiers in Marine Science frontiersin.org11
August, suggesting that the peak of its reproductive cycle leading to
spawning likely occur during September to October –in congruence
with the RS cycle (Bronstein et al., 2016).
4.5 Ecology
Most recently, reports of D. setosum mortality started to
accumulate from near Kas(Turkey), spreading to adjacent coasts off
Turkey and Greece. The intensity of these die-offs suggest that the
largest, most established Mediterranean population of the species is
undergoing mass mortality since July 2022 (Zirler et al. in press).
Based on the observed pathology, the characteristic tissue and spine
loss suggest a water-born pathogen as the cause of mortalities
(currently under investigation), similar to the driver of mortalities
in other Diadema species, such as D. antillarum in the Caribbean
(Bak et al., 1984;Hughes et al., 1985;Lessios, 1988)andD.
africanum in the Canary Islands (Clemente et al., 2014;
Hernandez et al., 2020;Sangil and Hernandez, 2021). Though the
spatial scale of these events is, to the best of our knowledge, still
confined to the northern Levant, it seems to be rapidly spreading,
imposing an immediate threat to both local fauna as well as native
RS D. setosum populations, and call for close monitoring of
these events.
5 Conclusions
Our data clearly shows that D. setosum is now well established
in the Mediterranean Sea. The exponential growth phase currently
reached in the Northern Levant marks the shift of D. setosum
proliferation to population outbreak. While the full scope and
outcomes of this successful invasion are still unclear, the potential
risk for the already disrupted environment of the Eastern
Mediterranean, increases dramatically. Habitat degradation,
depletion of resources, competitive exclusion of native species,
hybridization, and the emerging signs of pathogenic infections,
are some of the major concerns associated with the current
invasion. To preserve the unique ecosystem of the Eastern
Mediterranean, coordinated, regional-scale action must be
implemented. Regional collaboration will be effective not only for
the current D. setosum invasion, but also aid in mitigation of
upcoming invasions by NIS that are already making their way to
the Mediterranean Sea.
Data availability statement
The datasets presented in this study can be found in online
repositories. The names of the repository/repositories and accession
number(s) can be found in the article/Supplementary Material.
Author contributions
Study conception and design: RZ and OB. Data collection: RZ,
LL, MHo, VG, MHu, CJ, VR and MY. Analysis and interpretation of
results: RZ, OB, LL, TF and AK. Draft manuscript preparation: RZ
and OB. All authors reviewed the results and approved the final
version of the manuscript.
Funding
This research was supported by the Israel Science Foundation
(ISF; grant number 2407/20) and Yad Hanadiv Foundation (grant
number 10699) to OB. Field research in Turkey was partially
supported by WWF-Turkey.
Acknowledgments
This work was facilitated by the Mediterranean Diadema
Response Network (MDRN) –an international collaboration of
scientists and stakeholders from Cyprus, Greece, Turkey, Israel, and
the United Arab Emirates, aimed at providing real-time, regional
scale monitoring of the invasive D. setosum. Institutional support
was provided by the Steinhardt Museum of Natural History, Tel
Aviv University, Israel. We thank Dr. Nir Stern of the Israel
BCA
FIGURE 8
Reproductive cycles of Diadema setosum. Israeli RS (Is-RS, A), Crete, Greece (Cre-Med, B) and Israeli Mediterranean (Is-Med, C) specimens collected
between 2019 to 2021. Colors indicate reproductive stages 1–4 (corresponding to stages: Spent (Blue), Recovering (Green), Growing (Purple) and
Mature (Red), respectively). Novel data generated during the current study was compared with the 2010 reproductive cycle of RS D. setosum
described in Bronstein et al. (2016) (represented as the opaque background illustrations of each radial plot).
Zirler et al. 10.3389/fmars.2023.1152584
Frontiers in Marine Science frontiersin.org12
Oceanographic and Limnological Research for facilitating this
international collaboration. We extend our gratitude to Dr. Sigal
Shefer, Dr. Shevy Rothman, Dr. Boaz Maizel, Dr. Liron Goren, and
Alex Geyzner for their contributions to data collection. Field work
was supported by the Dive2gether and Mare Mundi Marine Field
Station in Plakias, southern Crete, and Enalia Physis Environmental
Research Centre and Deep Dive and ScubaCyprus Dive Centers (CJ,
MFH, VR).
Conflict of interest
The authors declare that the research was conducted in the
absence of any commercial or financial relationships that could be
construed as a potential conflict of interest.
Publisher’s note
All claims expressed in this article are solely those of the
authors and do not necessarily represent those of their affiliated
organizations, or those of the publisher, the editors and the
reviewers. Any product that may be evaluated in this article, or
claim that may be made by its manufacturer, is not guaranteed or
endorsed by the publisher.
Supplementary material
The Supplementary Material for this article can be found online
at: https://www.frontiersin.org/articles/10.3389/fmars.2023.
1152584/full#supplementary-material
References
Aikio, S., Duncan, R. P., and Hulme, P. E. (2010). Lag-phases in alien plant invasions:
separating the facts from the artefacts. Oikos 119, 370–378. doi: 10.1111/j.1600-
0706.2009.17963.x
Ammar, I. A. (2021). First record and massive spread of tropical long-spined Sea
urchin Diadema antillarum in the Mediterranean Sea (Syrian coast). J. Earth Environ.
Sci. Res. 3 (5), 1–4. doi: 10.47363/JEESR/2021(3)158. * Misidentified as Diadema
antillarum.
Arnold, T. A., and Emerson, J. W. (2011). Nonparametric goodness-of-fit tests for
discrete null distributions. RJ.3, 34–39. doi: 10.32614/RJ-2011-016
Aron,W.I.,andSmith,S.H.(1971).Shipcanalsandaquaticecosystems:
equilibrium has not been achieved since the Erie, welland, and Suez canals were
built. Science 174, 13–20. doi: 10.1126/science.174.4004.13
Azov, Y. (1991). Eastern Mediterranean–A marine desert? Mar. pollut. Bull. 23, 225–
232. doi: 10.1016/0025-326X(91)90679-M
Azzurro, E., Maynou, F., Belmaker, J., Golani, D., and Crooks, J. A. (2016). Lag times
in lessepsian fish invasion. Biol. Invasions 18, 2761–2772. doi: 10.1007/s10530-016-
1184-4
Bak, R. P. M., Carpay, M. J. E., and de Ruyter van Steveninck, E. D. (1984). Densities
of the sea urchin Diadema antillarum before and after mass mortalities on the coral
reefs on curacao. Mar. Ecol. Prog. series. Oldendorf 17, 105–108. doi: 10.3354/
meps017105
Bethoux, J. P. (1979). Budgets of the Mediterranean sea. their dependance on the
local climate and on the characteristics of the Atlantic waters. Oceanol Acta 2, 157–163.
Bilecenoglu, M., Baki Yokes, M., and Draman, M. (2019). The invasive sea urchin
Diadema setosum provides shelter for coastal fish–first observations from the
Mediterranean Sea. Zool Middle East 65, 183–185. doi: 10.1080/09397140.
2.019.1571746
Biton, E., and Gildor, H. (2011). Stepwise seasonal restratification and the evolution
of salinity minimum in the gulf of aqaba (Gulf of eilat). J. Geophys Res. 116, 1–7.
doi: 10.1029/2010JC00686010.1029/2011JC006944
Blackburn, T. M., Lockwood, J. L., and Cassey, P. (2015). The influence of numbers
on invasion success. Mol. Ecol. 24, 1942–1953. doi: 10.1111/mec.13075
Bluhm, B. A., Piepenburg, D., and von Juterzenka, K. (1998). Distribution, standing
stock, growth, mortality and production of Strongylocentrotus pallidus (Echinodermata:
echinoidea) in the northern barents Sea. Polar Biol. 20, 325–334. doi: 10.1007/
s003000050310
Borghini, M., Bryden, H., Schroeder, K., Sparnocchia, S., and Vetrano, A. (2014). The
Mediterranean is becoming saltier. Ocean Sci. 10, 693–700. doi: 10.5194/os-10-693-
2014
Bronstein, O., Georgopoulou, E., and Kroh, A. (2017). On the distribution of the
invasive long-spined echinoid Diadema setosum anditsexpansioninthe
Mediterranean Sea. Mar. Ecol. Prog. Ser. 583, 163–178. doi: 10.3354/meps12348
Bronstein, O., and Kroh, A. (2018). Needle in a haystack-genetic evidence confirms
the expansion of the alien echinoid Diadema setosum (Echinoidea: diadematidae) to the
Mediterranean coast of Israel. Zootaxa 4497, 593–599. doi: 10.11646/zootaxa.4497.4.9
Bronstein, O., Kroh, A., and Loya, Y. (2016). Reproduction of the long-spined sea
urchin Diadema setosum in the gulf of aqaba - implications for the use of gonad-
indexes. Sci. Rep. 6, 1–11. doi: 10.1038/srep29569
Bronstein, O., and Loya, Y. (2014). Echinoid community structure and rates of
herbivory and bioerosion on exposed and sheltered reefs. J. Exp. Mar. Biol. Ecol. 456, 8–
17. doi: 10.1016/j.jembe.2014.03.003
Bronstein, O., and Loya, Y. (2015). Photoperiod, temperature, and food availability
as drivers of the annual reproductive cycle of the sea urchin Echinometra sp. from the
gulf of aqaba (Red Sea). Coral Reefs 34, 275–289. doi: 10.1007/s00338-014-1209-3
Buchwald, R. T., Feehan, C. J., Scheibling, R. E., and Simpson, A. G. B. (2015). Low
temperature tolerance of a sea urchin pathogen: implications for benthic community
dynamics in a warming ocean. J. Exp. Mar. Biol. Ecol. 469, 1–9. doi: 10.1016/
j.jembe.2015.04.006
Byers, J. E., Reichard, S., Randall, J. M., Parker, I. M., Smith, C. S., Lonsdale, W. M.,
et al. (2002). Society for conservation biology directing research to reduce the impacts
of nonindigenous species. Cons Biol. 16, 630–640. doi: 10.1046/j.1523-1739.
2002.01057.x
Capella-Gutierrez, S., Silla-Martınez, J. M., and Gabaldon, T.. (2009). trimAl: a tool
for automated alignment trimming in large-scale phylogenetic analyses. J. Bioinform.
25, 1972–1973. doi: 10.1093/bioinformatics/btp348
Carlton, J. T. (1987). Patterns of transoceanic marine biological invasions in the
pacific ocean. Bull. Mar. Sci. 41, 452–465.
Carreiro-Silva, M., and Mcclanahan, T. R. (2001). Echinoid bioerosion and
herbivory on Kenyan coral reefs: the role of protection from fishing. J. Exp. Mar.
Biol. Ecol. 262, 133–153. doi: 10.1016/s0022-0981(01)00288-x
Chow, S., Konishi, K., Mekuchi, M., Tamaki, Y., Nohara, K., Takagi, M., et al. (2016).
DNA Barcoding and morphological analyses revealed validity of Diadema clarki ikeda
1939 (Echinodermata, echinoidea, diadematidae). Zookeys 2016, 1–16. doi: 10.3897/
zookeys.585.8161
Clemente, S., Hernandez, J. C., and Brito, A. (2011). Context-dependent effects of
marine protected areas on predatory interactions. Mar. Ecol. Prog. Ser. 437, 119–133.
doi: 10.3354/meps09243
Clemente, S., Hernandez, J. C., Rodrıguez, A., and Brito, A. (2010). Identifying
keystone predators and the importance of preserving functional diversity in sublittoral
rocky-bottom areas. Mar. Ecol. Prog. Ser. 413, 55–67. doi: 10.3354/meps08700
Clemente, S., Hernandez, J. C., Toledo, K., and Brito, A. (2007). Predation upon
Diadema aff. antillarum in barren grounds in the canary islands. Sci. Mar. 71, 745–754.
doi: 10.3989/scimar
Clemente, S., Lorenzo-Morales, J., Mendoza, J. C., Lopez, C., Sangil, C., Alves, F.,
et al. (2014). Sea Urchin Diadema africanum mass mortality in the subtropical eastern
Atlantic: role of waterborne bacteria in a warming