Phycologia Volume 55 (3), 265–273

Published 28 March 2016

The nature of the diatom Leptocylindrus mediterraneus (Bacillariophyceae),
host of the enigmatic symbiosis with the stramenopile Solenicola setigera
FERNANDO GO´ MEZ1*, TAIZ L.L. SIMA˜O2, LAURA R.P. UTZ2

AND

RUBENS M. LOPES1

1

Laboratory of Plankton Systems, Oceanographic Institute, University of Sa˜o Paulo, Pra¸ca do Oceanogra´fico 191,
Cidade Universita´ria, Butanta˜, Sa˜o Paulo 05508–900, Brazil
2
Faculdade de Biociˆencias, Pontif´ıcia Universidade Cat´olica do Rio Grande do Sul, Av. Ipiranga 6681,
Porto Alegre 90619–900, Brazil
ABSTRACT: The consortium between the colonial stramenopile Solenicola setigera and the centric chain-forming diatom
Leptocylindrus mediterraneus is cosmopolitan throughout the world ocean yet rarely abundant. However, the nature of
the association remains enigmatic. A mutualistic symbiosis requires a live diatom host, but the frustule of L.
mediterraneus is apparently empty, lacking protoplasm and plastids. The parasitism requires free-living host cells to be
infected, but there is no evidence of populations of the free-living diatom. During experiments attempting to culture the
heterotrophic S. setigera, we successfully obtained a strain of the host diatom. In the small-subunit (SSU) rDNA
phylogeny, L. mediterraneus was closely related to Dactyliosolen blavyanus and to a new sequence of Dactyliosolen sp.
while distantly related to the genus Leptocylindrus. We proposed the reinstatement of L. mediterraneus in the genus
Dactyliosolen as D. mediterraneus. Even under optimal growth conditions, the frustule was nearly empty with reduced
protoplasm concentrated in the middle. When compared with congeneric species, D. mediterraneus showed a doublelayered structure that acts as substrate for Solenicola. The free-living diatom lacked the convex walls that D.
mediterraneus showed as host of Solenicola. The diatom in consortium with Solenicola maintained the photosynthetic
machinery to eventually proliferate as a free-living organism. The ecological and morphological observations suggested
that the diatom was successfully adapted to the mutualistic symbiosis with Solenicola. We discarded a parasitic
relationship in this exceptional example of symbiosis.
KEY WORDS: Bacillariophyta, Dactyliosolen, DNA bar coding, Epibiosis, Epibiont, Molecular phylogenetics, Phoresy
commensalism, Plankton, Parasitism, Rhizomonas, SSU rDNA, Symbiont, Symbioses

INTRODUCTION
The surface water of warm oceans, usually depleted in
nutrients, is an unfavourable environment for the proliferation of diatoms. Regardless, several diatoms can proliferate
in mutualistic symbioses with other plankton organisms. The
diatoms Rhizosolenia, Guinardia, Hemiaulus and Chaetoceros
are hosts of diazotrophic filamentous cyanobacteria (Villareal 1992; G´omez et al. 2005; Hilton et al. 2013). Diazotrophic
coccoid cyanobacteria are found in tropical diatoms such as
Climacodium (Carpenter & Janson 2000). Other diatoms
have modified morphologies to host ciliates in a phoresy
commensalism (transport of one organism by another
without physiological or biochemical dependence between
them), such as Chaetoceros tetrastichon Cleve and C. dadayi
Pavillard and the tintinnid Eutintinnus inquilinus
(O.F.M¨uller) Schrank (G´omez 2007a) or Chaetoceros coarctatus Lauder and the peritrich ciliate Vorticella oceanica
Zacharias (Nagasawa & Warren 1996).
The nature of the consortium of the centric, chain-forming
diatom Leptocylindrus mediterraneus (H.P´eragallo) Hasle
and the colonial heterotrophic flagellate Solenicola setigera
Pavillard [¼ Rhizomonas setigera (Pavillard) D.J.Patterson,
K.Nygaard, G.Steinberg & C.M.Turley] remains enigmatic.
The mutualistic symbiotic consortia of diatoms with
* Corresponding author (fernando.gomez@fitoplancton.com).
DOI: 10.2216/15-131.1
Ó 2016 International Phycological Society

cyanobacteria or ciliates allow those to proliferate in the
surface oligotrophic waters of the warm ocean. In contrast,
the consortium of Solenicola–Leptocylindrus is widespread
from polar to equatorial zones, from coastal to oceanic
waters, and often reaches high abundances (Hasle 1976;
Taylor 1982; Fryxell 1989; Buck & Bentham 1998; G´omez
2007b; Padmakumar et al. 2012). The most accepted view is
that Solenicola is a highly adapted epizoic or parasitic
organism; although, others have speculated that Solenicola is
a stage of the diatom life cycle (Margalef 1969; Scott &
Thomas 2005). With the aid of molecular biology, S. setigera
has been reported as the first morphologically identified
member of a ubiquitous and diverse clade of environmental
sequences known as MAST3 (Marine Stramenopiles III)
(G´omez et al. 2011). A second member of this clade was
further identified, confirming the nature of this group as
represented by unicellular heterotrophic organisms with a
flagellum used to capture picoplankton and small nanoplanktonic prey (Cavalier-Smith & Scoble 2013).
The host diatom was first described as Lauderia mediterranea H. P´eragallo 1888 from the Bay of Villefranche sur
Mer, France, later proposed as Dactyliosolen mediterraneus
(H.P´eragallo) H.P´eragallo 1892 and more recently as L.
mediterraneus (H.P´eragallo) Hasle 1975 (P´eragallo 1888,
1892; Hasle 1975). The diatom is a widespread species in the
world ocean (Hasle 1975, 1976; Fryxell 1989), and several
studies have investigated its morphology using scanning
electron microscopy (Hasle 1976; Taylor 1982; Buck &
Bentham 1998; G´omez et al. 2011). The frustule possesses an
265

266

Phycologia, Vol. 55 (3)

unusual double-layered structure (Hasle 1975). Transmission
electron microscopy revealed that the frustule was nearly
empty and that the protoplasm with mitochondria occupied
a very small part of the cell (Buck & Bentham 1998).
However, it is uncertain whether the mitochondria belonged
to Solenicola or the diatom. Based on light microscopy, the
central section of the frustule, usually covered by Solenicola,
is limited by two thin convex internal separations in the
valvar plane that were named convex walls (G´omez 2007b).
The convex walls enclosed the protoplasm inside the frustule
when the diatom was colonised by Solenicola (Buck &
Bentham 1998). These features, together with a doublelayered structure, are unusual in diatoms, leading Hasle &
Syvertsen (1997) to report that the taxonomic position of L.
mediterraneus was questionable. Nanjappa et al. (2013)
revised the genus Leptocylindrus from isolates of the Gulf
of Naples. Leptocylindrus mediterraneus was found in field
samples; although cultures were not established because only
empty frustules colonised by Solenicola were observed
(Nanjappa et al. 2013). Observations by epifluorescence
microscopy of the Solenicola–Leptocylindrus consortium do
not reveal the red fluorescence of the plastids as usual in
diatoms (Buck & Bentham 1998; G´omez et al. 2011). Scott &
Thomas (2005) reported free-living cells of L. mediterraneus.
However, their images, based on light and scanning electron
microscopy, correspond to another diatom. Padmakumar et
al. (2012) cited the presence of live cells of L. mediterraneus
but without reporting any evidence, such as a picture of the
live cells, their protoplasm or plastids.
To the best of our knowledge, there is no information on
the organelles (i.e. chloroplasts) or protoplasm of L.
mediterraneus as a free-living organism or as host of
Solenicola. Cultures have not been established, and there is
no molecular information on this widespread diatom. This is
an essential step to elucidate the nature of the relationship
between Solenicola and Leptocylindrus. The parasitism
requires the occurrence of a free-living host to be colonised
or infected by Solenicola. However, there is no evidence of
living individuals of L. mediterraneus. A mutualistic symbiosis requires a benefit for the diatom, but apparently L.
mediterraneus is not alive when colonised by Solenicola.
In this study, we successfully established a culture of the
host, the diatom L. mediterraneus, from an isolate obtained
from a frustule fully covered by Solenicola. For the first time,
the morphology of live specimens of L. mediterraneus and
the molecular phylogeny is reported. This provides key
information to understand the enigmatic nature of the
Solenicola–Leptocylindrus consortium. We additionally report a new sequence of an unidentified species of the genus
Dactyliosolen.

MATERIAL AND METHODS
Consortia of Solenicola–Leptocylindrus were gathered from
samples collected daily from surface waters with a phytoplankton net (20-lm mesh size) off Ubatuba, Sa˜o Paulo
State, Brazil (23831 0 27.80 00 S, 45804 0 59.48 00 W), from January
to December 2014. Aliquots of the net samples were left to
settle in a composite settling chamber, examined with an

inverted microscope (Nikon TS100, Nikon, Tokyo, Japan)
and photographed with a digital camera (Cyber-shot DSCW300, Sony, Tokyo, Japan) mounted on the microscope
eyepiece. After being photographed, the consortium of
Solenicola–Leptocylindrus was micropipetted individually
with a fine capillary into a clean chamber and washed
several times into a series of drops of 0.2-lm-filtered
seawater to remove other organisms. The consortium was
finally placed in a 12-well tissue culture plate with 0.2-lmfiltered and sterilised seawater supplemented with f/2
medium with silicates (Guillard & Ryther 1962) and
incubated at 238C, with 100 lmol photons m2 s1 from
cool-white tubes; the photoperiod was 12:12 light:dark. The
free-living diatoms were reisolated and placed into a six-well
tissue culture plate and further in 50-ml polystyrene tissue
culture flasks and cultured in the same conditions. In
addition, one cell of Dactyliosolen sp. was found in
September 2014. The cell was isolated and cultured as
described above.
Additional light microscopic observations of the culture of
L. mediterraneus were carried out at 3100 magnification
using an Olympus BX51 epifluorescence microscope
equipped with an Olympus DP72 camera (Olympus, Tokyo,
Japan). Cells fixed with glutaraldehyde (5% final concentration) were stained with DAPI (4’,6’diamino-2-phenylindole,
Sigma, St. Louis, Missouri USA). Live cells were observed at
363 magnification with an inverted confocal microscope
Leica TCS SP8 AOBS microscope (Leica, Wetzlar, Germany).
To preserve the cells of L. mediterraneus and Dactyliosolen
sp., glutaraldehyde (50% solution) was added to the culture
to make a final concentration of 5%. Cells were filtered onto
a 0.8-lm pore size Whatman Nuclepore membrane filter
(Fisher, Pittsburgh, Pennsylvania USA), washed with
distilled water, fixed with osmium tetroxide, dehydrated
with a graded series of ethanol (30%, 50%, 70%, 90%, 95%,
99%, 100%) and critical-point dried with CO2. Filters were
mounted on stubs, sputter coated with gold and viewed using
a Phillips XL30 (FEI, Hillsboro, Oregon USA) scanning
electron microscope.
For L. mediterraneus and Dactyliosolen sp., exponentially
growing cultures were harvested by centrifugation, and the
pellet was placed in a 1.5-ml Eppendorf tube filled with
absolute ethanol. The sample was kept at room temperature
and in darkness until the molecular analysis could be
performed. Prior to polymerase chain reaction (PCR), the
sample tube was centrifuged, and ethanol was evaporated by
placing the tube overnight in a desiccator at room
temperature; 100 ll of DNA lysis buffer (0.1 M EDTA pH
8.0, 1% SDS, 200 lg ml1 proteinase K) were used to
resuspend the cell pellets, and the resuspension was
transferred into a 1.5-ml tube. Genomic DNA was extracted
from 250 ll using a standard phenol/chloroform protocol
adapted from Sambrook et al. (1989). At the end of the
extraction process, DNA was eluted in 30 ll of purified
water. Fragments of the small-subunit (SSU) rDNA were
amplified using the universal primers EukA and EukB
(Medlin et al. 1988). PCR amplification was performed in a
25-ll reaction volume containing 0.5 units of Platinum Taq
DNA polymerase (Thermo Fisher, Waltham, Massachusetts
USA), 1.253 PCR buffer (supplied with the polymerase),

G´omez et al.: Leptocylindrus mediterraneus, host of Solenicola setigera
MgCl2 at 2.5 mM, dNTPs at 0.25 mM and the forward and
reverse primers at 0.2 lM. The PCR conditions were initial
denaturation (948C/4 min); 45 cycles of denaturation (948C/
45 s), annealing (508C/30 s), and extension (728C/1 min);
final extension (728C/7 min). PCR products were visualised
on a 1% agarose gel stained with GelRed (Biotium,
Hayward, California USA), purified enzymatically with
FastAP Thermosensitive Alkaline Phosphatase and Exonuclease I (Thermo Fisher) and sequenced bidirectionally with
an ABI3730xl sequencer (Macrogen Europe, Amsterdam,
Netherlands) using the same primers as used for PCR.
Sequence reads were aligned and assembled using the
software ChromasPro 1.75 (Technelysium, Brisbane, Australia). The newly generated sequences were deposited
in DDBJ/EMBL/GenBank under accession numbers
KU577433–KU577435.
Matrices comprising SSU rDNA sequences, respectively,
were assembled from most similar sequences identified using
BLAST (http://blast.ncbi.nlm.nih.gov/Blast.cgi). Furthermore, available sequences of Leptocylindrus spp. and other
centric diatoms were included (Ashworth et al. 2013;
Nanjappa et al. 2013; Medlin 2016), and a sequence from
the stramenopile Bolidomonas pacifica was used as outgroup. The computer program MEGA v6.05 (Tamura et al.
2013) was employed to align contigs using the CLUSTALW
algorithm, to select the best-fit model of nucleotide
substitution and to construct phylogenetic trees using the
maximum likelihood (ML), neighbour-joining (NJ) and
maximum parsimony methods. The model selection step,
using a ML approach and the Akaike information criterion,
indicated that the T92þGþI model was most appropriate.
We therefore assumed this model for the ML tree search. In
addition, a Bayesian inference of phylogeny was performed
with MrBayes v3.2 (Huelsenbeck & Ronquist 2001), also
assuming the T92þGþI model of nucleotide substitution,
along with the following options: Yule tree prior, strict
molecular clock, 10 million generations of Markov chain
Monte Carlo, state sampling every 200,000 generations and
the first 25% of the states discarded as burn-in.

RESULTS
Morphology of L. mediterraneus
Our observations of the Solenicola–Leptocylindrus consortia
off the coast of Ubatuba, Brazil, were rare and in all the cases
corresponded to frustules fully covered by Solenicola. The
diatom as a free-living organism was never detected. Our
observations and previous studies based on light and
epifluorescence did not reveal the presence of plastids in the
Solenicola–Leptocylindrus consortium. We assumed that the
diatom was dead. Our aim was to examine Solenicola under
culture conditions, and we fed this heterotrophic protist with
natural pico- and nanoplankton assemblages.
Filtered local seawater was enriched with f/2 medium
without silicates, and some drops of seawater sample were
added in order to grow natural pico- and small nanoplankton
populations. These cultures were also used to attempt to
culture heterotrophic planktonic organisms (flagellates, ciliates and dinoflagellates). Each consortium of Solenicola–

267

Leptocylindrus was individually placed in this medium with
natural pico-nanoplankton cells and kept in an incubator at
238C under light or sometimes covered with an aluminium foil
in order to reduce the growth of the potential prey.
Microscopic observations during the manipulation revealed
that Solenicola detached from the diatom frustule. After 24 h,
the culture plate was examined at low magnification (310 or
320) for the presence of the motile consortium of Solenicola–
Leptocylindrus. Solenicola was not observed, and an apparently empty diatom frustule remained at the bottom of the
culture plate. The culture was then discarded without any
further observations because the objective had not been
accomplished.
A consortium of Solenicola–Leptocylindrus observed on 25
February 2014 (Figs 1, 2) was treated with a different
protocol. This consortium was considered to be a photosynthetic organism and placed in 0.2-lm-filtered and sterilised
seawater supplemented with f/2 medium and incubated at
238C with a photoperiod of 12:12 light:dark. We expected a
single or broken frustule of L. mediterraneus free of Solenicola;
however, a group of colonies of a diatom that coincided with
the morphology of L. mediterraneus was observed. At higher
magnification, the frustules were nearly empty with small
amounts of plastid-bearing protoplasm. The cells were
isolated into new clonal cultures under the same culture
conditions. The growth of this diatom was slower when
compared to other diatoms under similar conditions. A
second consortium of Solenicola–Leptocylindrus was isolated
from plankton samples in April 2014 and isolated for culture
with no success. No further consortia were found in the
plankton samples.
Leptocylindrus mediterraneus showed different forms in
batch culture. During the first week, after the inoculation of a
few cells, L. mediterraneus showed tightly formed, straight
chains that reached up to eight cells. Despite the optimal
growth conditions, the inner frustule remained nearly empty,
with reduced protoplasm in the middle of the frustule
coinciding with the double-layered structure (Figs 3–6). The
protoplasm was displaced to one of the valve faces after each
division (Figs 5, 19). Then the double-layered structure
appeared in that valve face, and after the extension of the
frustule on that side, the double-layered structure and the
protoplasm returned to appear in the middle of the frustule
(Figs 16–18). The protoplasm was connected to each valve
face with two thin filaments (Fig. 6). In some cases, the
proximal parts of the filaments were pigmented (Fig. 4). A
small nucleus was visible by DAPI staining (Figs 7–10). In
some cases, two nuclei were observed, likely corresponding to
a cell prior to division (Fig. 10). Each cell showed 5–10
ellipsoid plastids (Figs 11, 12). There was no evidence of the
convex walls inside the frustule that has been reported when L.
mediterraneus was colonised by Solenicola. After 2 wk, singleor two-celled colonies of L. mediterraneus were dominant, and
curved frustules began to appear, sometimes forming sigmoidal colonies. The protoplasm possessed a single nonpigmented
filament towards the valve face, and the cell showed a large
vacuole (Figs 16–19). In the senescent phase, some of the
curved cells showed more intense pigmentation that covered
almost the entire frustule (Fig. 13). Round protoplasmic
masses, tentatively identified as auxospores, were observed
(Figs 13–15).

268

Phycologia, Vol. 55 (3)

Figs 1–21. Light micrographs of Leptocylindrus mediterraneus (Figs 1–19) and Dactyliosolen sp. (Figs 20, 21) from Ubatuba, Brazil. Scale bars
¼ 10 lm.
Figs 1, 2. Consortium of Solenicola–Leptocylindrus isolated for the strain culture.
Figs 3, 4. Culture of L. mediterraneus in exponential growth phase. The arrow points the pigmented proximal part of the filaments.
Figs 5, 6. Cells in stationary phase. The arrow points two filaments from the protoplasm to the valve faces.
Figs 7–10. Cells stained with DAPI. The blue area corresponds to the nucleus.
Fig. 11. Epifluorescence image of the plastids.
Fig. 12. Plastids release from the protoplasm.
Figs 13–15. Cells in senescent phase. Note the tentative auxospores.

G´omez et al.: Leptocylindrus mediterraneus, host of Solenicola setigera
The culture of L. mediterraneus was examined by scanning
electron microscopy (Figs 22–27). The ends of the girdle bands
were wedge shaped and formed a straight line (Figs 22–24).
The width of the girdle bands was different between cells of the
same culture (Fig. 22) and in the same cell (Fig. 23). As a
general trend, there were four or five half bands in 10 lm.
However, this number can be variable in the same cell (Figs 23,
24). The cells showed the double-layered structure in the
middle of the frustule surface (Figs 22, 23) or displaced
towards one side, depending of the growth stage (Fig. 24).
This double-layered structure was a unique feature of this
species. In senescent cultures, the frustules were more
elongated, more or less curved, and the double-layered
structure covered a larger proportion of the frustule (Fig.
25). The valve face showed a short labiate process displaced
from the centre, eccentric but not marginal (Figs 26, 27).
During plankton observations in September 2014, a cell of
an unidentified species of the genus Dactyliosolen was
observed (Fig. 20). As in the culture of L. mediterraneus, this
diatom showed the protoplast concentrated to the middle of
the cell (Fig. 20). Dactyliosolen sp. was isolated and cultured in
order to increase the representation of the genus Dactyliosolen
in the SSU rDNA molecular phylogeny. As in the case of L.
mediterraneus, even under optimal growth conditions, the
protoplast was reduced and concentrated to the middle (Fig.
21). The cells were straight, 20–50 lm in diameter and 60–120
lm in the pervalvar axis. The valves faces were flat (Figs 20,
21); whereas, in L. mediterraneus the valve faces were convex
(Figs 17–19). Using scanning electron microscopy, the ends of
the girdle half bands were wedge shaped and formed a straight
line (Figs 28, 29). The cells showed four to five girdle bands in
10 lm (Figs 28, 29). The valve face showed a short marginal
labiate process (Fig. 30). The double-layered structure was not
observed. The morphology of this species is close to D.
blavyanus (H. P´eragallo) Hasle.
Molecular phylogeny
We obtained the SSU rDNA sequences (1685 base pairs) of L.
mediterraneus and Dactyliosolen sp. A BLAST search was
conducted on the new sequences to find related sequences in
GenBank database. The new sequence of L. mediterraneus
(KU577433) was closely related to Dactyliosolen blavyanus
(KC309491, 99% sequence identity) and the new sequence of
Dactyliosolen sp. (KU577434–KU577435, 99% sequence
identity). The new sequence of L. mediterraneus was distantly
related to the type of the genus Leptocylindrus, L. danicus
Cleve (KC814808, 84%). In the molecular phylogeny, L.
mediterraneus and Dactyliosolen spp. formed a strongly
supported clade (Fig. 31). Two sequences of environmental
clones (SGYO1439, SGYO1107) from the Gulf Stream
surface waters formed a sister clade with L. mediterraneus/
Dactyliosolen spp. These sequences formed a sister clade with
Chaetoceros peruvianus Brightwell/C. rostratus Lauder and C.
muelleri Lemmermann/C. calcitrans (Paulsen) Takano. All

269

these species formed a highly supported clade together with
the clades of Acanthoceras sp., Urosolenia eriensis (H.L.Smith)
Round & R.M. Crawford, Hemiaulus spp., Eucampia spp. and
Cerataulina spp. The sequences of Leptocylindrus spp.
branched with other radial centric diatoms and were distantly
related to the clade of L. mediterraneus/Dactyliosolen spp.
(Fig. 31). Assuming that D. blavyanus and Dactyliosolen sp.
are truly representatives of the genus Dactyliosolen, we
proposed the reinstatement of L. mediterraneus under the
genus Dactyliosolen as D. mediterraneus (H.P´eragallo)
H.P´eragallo.

DISCUSSION
Identity of D. mediterraneus
The molecular phylogeny revealed that L. mediterraneus
should no longer belong to the genus Leptocylindrus. Until
now, the genus Dactyliosolen is represented in the GenBank
database only by the sequence of D. blavyanus (Ashworth et
al. 2013; Medlin 2016). The molecular phylogeny revealed that
L. mediterraneus should no longer belong to the genus
Leptocylindrus. The sequence of the type species, D.
antarcticus Castracane, is not available. However, the
resemblance of D. antarcticus, D. blavyanus and D. mediterraneus supports their placement within the same genus.
The species D. antarcticus and D. tenuijunctus (Manguin)
Hasle showed the girdle band ends in an oblique line; whereas,
D. mediterraneus and D. blavyanus showed the band ends in a
(nearly) straight line (Hasle & Syvertsen 1997). Von Stosch
(1986) illustrated the life cycle of D. blavyanus. He reported
that ‘D. blavyanus is one of the two centric diatoms known to
me . . . in which only the region of the thecal junction, where
the girdles overlap, seems to be ‘‘inhabited’’ by the protoplast,
though sometimes ‘‘pseudopodia’’, with or without plastids,
extend to neighbouring regions of the cincture’. Hasle &
Syvertsen (1997) reported that ‘Dactyliosolen blavyanus differs
from the other species of the genus by being the only one
known to form (endogenous) resting spores by having the
protoplasm concentrated to the middle of the cell (von Stosch
1986)’.
The protoplasm and the filaments, named ‘pseudopodia’ by
Stosch (1986), are a common feature for D. mediterraneus, D.
blavyanus and Dactyliosolen sp. (Figs 4, 21). Hasle (1976)
reported that valves of L. mediterraneus mounted in a medium
of a high refractive index are recognised by their coarse, or
double-layered, structure. Ashworth et al. (2013) provided the
sequence and light and scanning electron microscopic pictures
of D. blavyanus (available at http://www.protistcentral.org/
Taxa/get/taxa_id/585789). Their pictures show that D. blavyanus lacks the double-layered structure that characterises D.
mediterraneus and possesses a marginal labiate process. In
contrast, the labiate process of D. mediterraneus is eccentrically located but not marginal (Figs 26, 27). The valve face of

Figs 16–19. Cells observed with confocal microscopy. Red areas correspond to chlorophyll fluorescence. The arrowhead points the large
vacuole. The protoplasm possesses a single filament towards each valve face. Note the slightly convex valve faces.
Fig. 20. Cell of Dactyliosolen sp. isolated for the strain culture.
Fig. 21. Cultured cells of Dactyliosolen sp. Note the flat valve faces.

270

Phycologia, Vol. 55 (3)

Figs 22–30. Scanning electron microscopy of Leptocylindrus mediterraneus (Figs 22–27) and Dactyliosolen sp. (Figs 28–30) from Ubatuba,
Brazil. Scale bars ¼ 10 lm, except Figs 26, 27 and inset of Fig. 30 ¼ 1 lm.
Figs 22, 23. The arrowheads point the double-layered structure in the middle of the frustule.
Fig. 24. Double-layered structure close to the valve face. Note the different width of the girdle bands between different cells (Fig. 22) and in
the same cell (Figs 23, 24).
Fig. 25. The double-layered structure covers most of the cell.
Figs 26, 27. The arrowhead points the labiate process in the valve face of L. mediterraneus.
Fig. 28. Two-celled chain of Dactyliosolen sp.
Fig. 29. Detail of the girdle bands.
Fig. 30. The arrowhead points the labiate process in the valve face. The inset shows the labiate process.

G´omez et al.: Leptocylindrus mediterraneus, host of Solenicola setigera

271

Fig. 31. Bayesian phylogenetic tree of SSU rDNA sequences, based on 1387 aligned positions. The species newly sequenced in this study are
bold. Numbers at the end of each taxon name are GenBank accession numbers. Posterior probabilities from Bayesian analysis are noted to
the left of internodes. Bolidomonas pacifica was used as out-group. Scale bar ¼ 0.01 substitutions per site.

D. blavyanus is flat as well in Dactyliosolen sp. (Figs 20, 21)
while convex with slightly rounded edges in D. mediterraneus
(Figs 17–19).
Under routine phytoplankton microscopic analysis, D.
mediterraneus devoid of Solenicola can be misidentified with
other species of Dactyliosolen. The number of girdle bands in
10 lm is similar in D. mediterraneus, D. blavyanus (four to five

bands in 10 lm), Dactyliosolen sp. (four to five), lower in D.
antarcticus (two to three) and higher in D. tenuijunctus (five to
nine) (Table 1). Dactyliosolen blavyanus is more commonly
found in warm waters (Herna´ndez-Becerril et al. 2010;
Ashworth et al. 2013). Dactilyosolen antarcticus and D.
tenuijunctus are common in Antarctic waters, and D.
mediterraneus shows a cosmopolitan distribution (Hasle

272

Phycologia, Vol. 55 (3)

Table 1. Morphometric data of Dactyliosolen spp.
Taxon

Bands in 10 lm

Reference

D. antarcticus
D. tenuijunctus
D. tenuijunctus
D. blavyanus
D. blavyanus
D. mediterraneus
D. mediterraneus
D. mediterraneus1
D. mediterraneus
D. mediterraneus
Dactyliosolen sp.

2–3
5–9
6–8
4–5
5
1–5
4–5
7–8 or 12–14
4–5
4–5
4–5

Hasle & Syvertsen (1997)
Hasle & Syvertsen (1997)
Cefarelli et al. (2011)
Hasle & Syvertsen (1997)
Ashworth et al. (2013)
Hasle & Syvertsen (1997)
Buck & Bentham (1998)
Scott & Thomas (2005)
G´omez et al. (2011)
This study
This study

1

Misidentification.

1976). In Antarctic waters, some authors have pooled L.
mediterraneus and Dactyliosolen spp. (Waters et al. 2000).
Cefarelli et al. (2011) illustrated D. tenuijunctus and D.
mediterraneus from Antarctic waters, the latter taxon with
the typical empty frustules covered by Solenicola. Scott &
Thomas (2005), also from Antarctic waters, illustrated D.
antarcticus, D. tenuijunctus and another taxon identified as L.
mediterraneus. In contrast to other authors, Scott & Thomas
(2005) found L. mediterraneus devoid of Solenicola and even
dividing cells with plastids. They described D. mediterraneus
with heavily silicified walls; whereas, all previous studies
reported this species as lightly silicified. For most authors, the
frustule of D. mediterraneus possesses four to five girdle bands
in 10 lm (Table 1). However, Scott & Thomas (2005)
misidentified D. mediterraneus with other congeneric species
because their specimens showed more than seven girdle bands,
the valve faces were flat and the double-layered structure that
characterises D. mediterraneus was not visible. Scott &
Thomas (2005, p. 66) reported that ‘chains of cells have often
been illustrated with the epiphytic flagellate Rhizomonas
setigera attached to the girdle band zone (e.g. Hasle &
Syvertsen 1997, fig. 15a). Observations suggest that the
flagellated (or perhaps even amoeboid) stage emerges from
the central area of the cell and covers frustules in large
numbers in the spring. Rings of four flagellates on separate
thecae could represent a later gametic stage. Further study is
required, but it appears that R. setigera is a stage in the life
cycle of L. mediterraneus rather than a distinct epiphyte’.
Nature of the symbiotic consortium
Solenicola setigera is an independent organism phylogenetically unrelated to the diatoms (G´omez et al. 2011; CavalierSmith & Scoble 2013). In this study, we contributed new data
on the diatom to elucidate the nature of this enigmatic
symbiotic consortium. The heterotrophic Solenicola needs a
substrate to develop its colony. Many centric diatoms have the
same size and shape of D. mediterraneus. However, Solenicola
develops only on D. mediterraneus. Solenicola often forms
highly dense colonies with a mucilaginous layer on the diatom
frustule (Taylor 1982; Buck & Bentham 1998; G´omez et al.
2011). For any other diatom species, the colonization by
Solenicola will be a barrier for gases and nutrient exchange
through the frustule. Under these conditions, any other
diatom will be unable to carry out photosynthesis and will
finally die. If the diatom dies, the diatom colony decomposes,

the silica of the frustule will dissolve and Solenicola will need
to look for another host. Dactyliosolen mediterraneus possesses reduced protoplasm and apparently lacks chlorophyll when
colonised by Solenicola. However, D. mediterraneus is alive
and able to maintain the frustule that is the substrate for the
colony of Solenicola. There are no free-living cells of D.
mediterraneus in the surrounding waters. Consequently,
Solenicola does not need to look for another host if D.
mediterraneus is alive.
The diatom D. mediterraneus did not show chlorophyll
when colonised by Solenicola (Buck & Bentham 1998; G´omez
et al. 2011). However, this study provides evidences that D.
mediterraneus remains alive in the frustules fully colonised by
Solenicola. Dactyliosolen mediterraneus possesses a doublelayered structure that covers the reduced protoplasm. The cells
of the colony of Solenicola are preferentially placed on the
double-layered structure. Dactyliosolen mediterraneus is not
found as a free-living organism, except in our culture under
laboratory conditions. When compared to other diatoms, the
slow growth of D. mediterraneus could be attributed to its low
quantity of protoplasm (and low number of plastids). The
protoplasm and number of plastids do not extend inside the
frustule even under optimal growth conditions. Consequently,
D. mediterraneus with reduced protoplasm is able to provide a
wide substrate for Solenicola. The morphology of D.
mediterraneus is different when it acts as host of Solenicola
or as a free-living organism. In consortium with Solenicola, D.
mediterraneus does not produce chlorophyll, and it possesses
convex bands inside the frustule enclosing the protoplasm
(Buck & Bentham 1998, fig. 3; G´omez 2007b). That structure
is unknown in other diatoms. It is uncertain whether D.
mediterraneus maintains its metabolism from resources
provided by Solenicola. Solenicola setigera needs D. mediterraneus and vice versa because they are not found living
independently in natural conditions. The relationship is
successful because they are widely distributed throughout
the world’s oceans. Epibiosis was defined as a symbiosis in
which one organism lives on the outer surface of another
(Lincoln et al. 1982). A simple epibiosis does not explain the
success of Solenicola–Dactyliosolen. The nature of the
association is a phoresy commensalism because the flagellum
of Solenicola renders motility to the consortium. Phoresis is
defined as the provision of transport, shelter or other support
by one organism (the carrier or host) to another organism
without metabolic exploitation or exchange (Kinne 1980). We
have no evidence that D. mediterraneus uses its photosynthetic
capacity when behaving as host of Solenicola. It is uncertain
whether D. mediterraneus may benefit from exudates provided
by Solenicola. With the current evidence, we discard a
parasitic relationship, and we suggest that the consortium
Solenicola–Dactyliosolen could be an example of a unique and
widespread mutualistic symbiosis.

ACKNOWLEDGEMENTS
This research is supported by the Brazilian Conselho
Nacional de Desenvolvimento Cient´ıfico e Tecnolo´ gico
(grant numbers BJT 370646/2013–14 to F.G. and 402759/
2012–5 and 311936/2013–0 to R.M.L.).

G´omez et al.: Leptocylindrus mediterraneus, host of Solenicola setigera
REFERENCES
ASHWORTH M.P., NAKOV T. & THERIOT E.C. 2013. Revisiting Ross
and Sims (1971): toward a molecular phylogeny of the
Biddulphiaceae and Eupodiscaceae (Bacillariophyceae). Journal
of Phycology 49: 1207–1222.
BUCK K.R. & BENTHAM W.N. 1998. A novel symbiosis between a
cyanobacterium, Synechococcus sp., an aplastidic protist, Solenicola setigera, and a diatom, Leptocylindrus mediterraneus, in the
open ocean. Marine Biology 132: 349–355.
CARPENTER E.J. & JANSON S. 2000. Intracellular cyanobacterial
symbionts in the marine diatom Climacodium frauenfeldianum
(Bacillariophyceae). Journal of Phycology 36: 540–544.
CAVALIER-SMITH T. & SCOBLE J.M. 2013. Phylogeny of Heterokonta:
Incisomonas marina, a uniciliate gliding opalozoan related to
Solenicola (Nanomonadea), and evidence that Actinophryida
evolved from raphidophytes. European Journal of Protistology 49:
328–353.
CEFARELLI A.O., VERNET M. & FERRARIO M.E. 2011. Phytoplankton
composition and abundance in relation to free-floating Antarctic
icebergs. Deep-Sea Research II 58: 1436–1450.
FRYXELL G.A. 1989. Marine phytoplankton at the Weddell Sea ice
edge: seasonal changes at the specific level. Polar Biology 10: 1–
18.
GO´ MEZ F. 2007a. On the consortium of the tintinnid Eutintinnus and
the diatom Chaetoceros in the Pacific Ocean. Marine Biology 151:
1899–1906.
GO´ MEZ F. 2007b. The consortium of the protozoan Solenicola
setigera and the diatom Leptocylindrus mediterraneus in the
Pacific Ocean. Acta Protozoologica 46: 15–24.
GO´ MEZ F., FURUYA K. & TAKEDA S. 2005. Distribution of the
cyanobacterium Richelia intracellularis as an epiphyte of the
diatom Chaetoceros compressus in the western Pacific Ocean.
Journal of Plankton Research 27: 323–330.
GO´ MEZ F., MOREIRA D., BENZERARA K. & LO´ PEZ-GARC´IA P. 2011.
Solenicola setigera is the first characterized member of the
abundant and cosmopolitan uncultured marine stramenopile
group MAST–3. Environmental Microbiology 13: 293–302.
GUILLARD R.R. & RYTHER J.H. 1962. Studies on marine planktonic
diatoms. I. Cyclotella nana Hustedt and Detonula confervacaea
(Cleve) Gran. Canadian Journal of Microbiology 8: 229–239.
HASLE G.R. 1975. Some living marine species of the diatom family
Rhizosoleniaceae. Beihefte Nova Hedwigia 53: 99–140.
HASLE G.R. 1976. The biogeography of some marine planktonic
diatoms. Deep-Sea Research 23: 319–338.
HASLE G.R. & SYVERTSEN E.E. 1997. Marine diatoms. In:
Identifying marine phytoplankton (Ed. by C.R. Tomas), pp. 5–
385. Academic Press, San Diego, California.
HERNA´NDEZ-BECERRIL D.U., HERRERA-HERNA´NDEZ P., PE´ REZ-MENDOZA A. & CEBALLOS-CORONA J.G.A. 2010. Marine planktonic
diatoms of the order Rhizosoleniales (Bacillariophyta) from the
tropical Mexican Pacific. Vie et Milieu 60: 95–107.
HILTON J.A., FOSTER R.A., TRIPP H.J., CARTER B.J., ZEHR J.P. &
VILLAREAL T.A. 2013. Genomic deletions disrupt nitrogen
metabolism pathways of a cyanobacterial diatom symbiont.
Nature Communications 4: 1767.
HUELSENBECK J.P. & RONQUIST F. 2001. MrBayes: Bayesian
inference of phylogenetic trees. Bioinformatics 17: 754–755.

273

KINNE O. 1980. Diseases of marine animals: general aspects. In:
Diseases of marine animals (Ed. by O. Kinne), pp. 13–73. Wiley &
Sons, New York.
LINCOLN R.J., BOXSHALL G.A. & CLARK P.F. 1982. A dictionary of
ecology, evolution and systematics. Cambridge University Press,
London. 298 pp.
MARGALEF R. 1969. Composici´on espec´ıfica del fitoplancton de la
costa catalano-levantina (Mediterra´neo occidental) en 1962–1967.
Investigaciones Pesqueras 33: 345–380.
MEDLIN L.K. 2016. Evolution of the diatoms: major steps in their
evolution and a review of the supporting molecular and
morphological evidence. Phycologia 55: 79–103.
MEDLIN L., ELWOOD H.J., STICKEL S. & SOGIN M.L. 1988. The
characterization of enzymatically amplified eukaryotic 16S-like
rRNA-coding regions. Gene 71: 491–499.
NAGASAWA S. & WARREN A. 1996. Redescription of Vorticella
oceanica Zacharias, 1906 (Ciliophora: Peritrichia) with notes on
its host, the marine planktonic diatom Chaetoceros coarctatum
Lauder, 1864. Hydrobiologia 337: 27–36.
NANJAPPA D., KOOISTRA W.H.C.F. & ZINGONE A. 2013. A
reappraisal of the genus Leptocylindrus (Bacillariophyta) with
the addition of three species and the erection of Tenuicylindrus
gen. nov. Journal of Phycology 49: 917–936.
PADMAKUMAR K.B., CICILY L., SHAJI A., MANEESH T.P. &
SANJEEVAN V.N. 2012. Symbiosis between the stramenopile protist
Solenicola setigera and the diatom Leptocylindrus mediterraneus
in the North Eastern Arabian Sea. Symbiosis 56: 97–101.
PE´ RAGALLO H. 1888. Diatom´ees du Midi de la France. Diatom´ees de
la Baie de Villefranche (Alpes Maritimes). Bulletin de la Soci´et´e
d’Histoire Naturelle de Toulouse 22: 5–100.
PE´ RAGALLO H. 1892. Monographie du genre Rhizosolenia et de
quelques genres voisins. Diatomiste 1: 79–82, 99–117.
SAMBROOK J., FRITSCH E.F. & MANIATIS T. 1989. Molecular cloning:
a laboratory manual. Cold Spring Harbor Laboratory Press, New
York. 1626 pp.
SCOTT F.J. & THOMAS D.P. 2005. Diatoms. In: Antarctic marine
protists (Ed. by F.J. Scott & H.J. Marchant), pp. 13–201.
Australian Biological Resources Study, Canberra, and Australian
Antarctic Division, Hobart.
TAMURA K., STECHER G., PETERSON D., FILIPSKI A. & KUMAR S.
2013. MEGA6: molecular evolutionary genetics analysis version
6.0. Molecular Biology and Evolution 30: 2725–2729.
TAYLOR F.J.R. 1982. Symbioses in marine microplankton. Annales
de l’Institut Oc´eanographique, Paris 58: 61–90.
VILLAREAL T.A. 1992. Marine nitrogen-fixing diatom-cyanobacterial symbioses. In: Marine pelagic cyanobacteria, Trichodesmium
and other diazotrophs (Ed. by E.J. Carpenter, D.G. Capone & J.
Reuter), pp. 163–175. Kluwer Academic Press, Dordrecht, The
Netherlands.
VON STOSCH H.A. 1986. Some marine diatoms from the Australian
region, especially from Port Phillip Bay and tropical northeastern Australia. Brunonia 8: 293–348.
WATERS R.L., VAN DEN ENDEN R. & MARCHANT H.J. 2000. Summer
microbial ecology off East Antarctica (80–150 8E): protistan
community structure and bacterial abundance. Deep-Sea Research II 47: 2401–2435.

Received 17 November 2015; accepted 28 January 2016

Sign up to vote on this title
UsefulNot useful