Download Unexpected prevalence of parasite 18S rDNA sequences in winter

Journal of
Plankton Research
plankt.oxfordjournals.org
Unexpected prevalence of parasite 18S
rDNA sequences in winter among
Antarctic marine protists
ALISON C. CLEARY1,2†* AND EDWARD G. DURBIN1
1
2
GRADUATE SCHOOL OF OCEANOGRAPHY, UNIVERSITY OF RHODE ISLAND, NARRAGANSETT, RI, USA AND UNIVERSITY CENTER IN SVALBARD, LONGYEARBYEN,
SVALBARD, NORWAY
†
PRESENT ADDRESS: MARINE BIOLOGICAL ASSOCIATION, PLYMOUTH, UK
*CORRESPONDING AUTHOR: [email protected]
Received August 7, 2015; accepted January 12, 2016
Corresponding editor: John Dolan
Parasites are not typically considered to be important components of polar marine ecosystems. It was therefore surprising when 18S rDNA surveys of protists in the West Antarctic Peninsula in winter revealed high abundances of
parasite sequences. Parasite sequences made up, on average, over half (52%) of sequence reads in samples from deep
water in winter. Winter surface water and sediment samples contained relatively fewer, but still strikingly high, parasite
sequence reads (13 and 9%, respectively), while surface water samples in summer contained fewer parasite sequences
(1.8%). A total of 1028 distinct parasite Operational Taxonomic Units were observed in winter, with the largest abundances and diversities within Syndiniales groups I and II, including Amoebophrya. Less abundant parasite sequence groups
included Apicomplexa, Blastodinium, Chytriodinium, Cryptocaryon, Paradinium, Perkinsidae, Pirsonia and Ichthyophonae. Parasite
sequence distributions suggested interactions with known hosts, such as diatom parasites which were mainly in the sediments, where resting spores of Chaetoceros spp. diatoms were abundant. Syndiniales sequences were correlated with radiolarian sequences, suggesting parasite–host interactions. The abundant proportions of parasite sequences indicate a
potentially important role for parasites in the Antarctic marine ecosystem, with implications for plankton population dynamics, the role of the microbial loop, carbon flows and ecosystem responses to ongoing anthropogenic climate change.
KEYWORDS: 18s rDNA; parasites; protists; plankton; West Antarctic Peninsula; Syndiniales
I N T RO D U C T I O N
The West Antarctic Peninsula (WAP) has traditionally been
thought of as a comparatively simple ecosystem, with the
food web dominated by relatively few types of organisms
and interactions; large phytoplankton being consumed
by krill, which were in turn consumed by penguins, seals
available online at www.plankt.oxfordjournals.org
# The Author 2016. Published by Oxford University Press. All rights reserved. For permissions, please email: [email protected]
Featured Article
J. Plankton Res. (2016) 38(3): 401–417. First published online Febraury 24, 2016 doi:10.1093plankt/fbw005
JOURNAL OF PLANKTON RESEARCH
j
VOLUME 
and whales (Huntley et al., 1991). More recently, the roles
of smaller phytoplankton, microzooplankton, bacteria
and the ensuing microbial loop have been increasingly recognized (Garrison, 1991; Sailley et al., 2013).
Alternative carbon pathways to the diatom–krill–penguin
model are thought to be particularly important outside of
the productive, and well-studied, spring bloom period
(Ducklow et al., 2007). The microbial loop is now known to
play an important role in carbon cycling and food web dynamics in the region, with bacterioplankton converting
dissolved organic matter (DOM) into biomass, which is in
turn consumed by small flagellates and other heterotrophic protists, creating an alternative, and more complex,
pathway for carbon to travel before reaching a size available to krill and other macrofaunal predators (Ducklow
et al., 2007; Sailley et al., 2013). Even as the role of many
of these smaller organisms, such as bacterioplankton and
heterotrophic flagellates, in the WAP marine ecosystem
has been increasingly recognized, one category of small
marine organisms which continues to escape attention are
the parasitic protists.
The abundance of protistan parasites is thought to
be under-accounted for in marine planktonic and sediment systems generally (Sousa, 1991; Skovgaard, 2014).
Protistan parasites can be difficult to observe, as they are
small and spend much of their lives hidden within their
hosts. These parasites may be invisible when analyzing
preserved plankton samples (Skovgaard and Daugbjerg,
2008). Estimates of protistan parasite abundance or
biomass in natural ecosystems are scarce. Analysis of
metazoan parasites in estuaries showed they accounted
for 2 – 3% of the total biomass (Kuris et al., 2008), while
in modeling studies, when included at all, total parasite
abundance has been much lower, at 0.01– 0.02% of total
biomass (Arias-Gonzáles and Morand, 2006).
Parasite diversity is also poorly known, particularly for
protistan parasites, and particularly in the marine environment. For many parasitic protist groups, there are only
rough estimates of the total species diversity and abundance. Even for described parasite organisms, the range
of possible hosts and the complete life cycle are often
unknown. It has been suggested that parasitic lifestyles
should encourage species diversity, such that there may
be an equal or greater diversity of parasites than of traditional predators in natural ecosystems (Toft, 1986).
Many protist groups which have been identified as obligate parasites include only a handful of described species
(Chambouvet et al., 2014). Yet, 18S ribosomal DNA
(rDNA) molecular surveys suggest these groups may be
highly diverse and contain hundreds or more species
(Chambouvet et al., 2014; Skovgaard, 2014).
Protistan parasites in the marine environment include a
diverse range of organisms from several major eukaryotic
j
NUMBER 
j
PAGES –
j

lineages. The Alveolata are particularly rich in marine
parasites. Marine Alveolate groups include the Syndiniales
and Amoebophrya which infect a variety of marine plankton,
including radiolarians, dinoflagellates and copepods, and
Euduboscquella which parasitizes radiolarians. Other alveolate parasites in the marine environment include Perkinsidae
which parasitize shellfish and dinoflagellates, and parasitic ciliates which infest crustaceans, fish and gelatinous
plankton, as well as other parasite groups (Skovgaard,
2014). Apicomplexan parasites, including gregarines,
parasitize marine crustaceans, polycheates, mollusks and
others, while Rhizaria parasites include Paradinium, which
infects copepods (Skovgaard, 2014). Additional parasites
from the amoebozoa, fungi, stramenopiles and euglenozoa
have also been observed in the marine environment
(Skovgaard, 2014). In the Antarctic, clone library studies
have shown large numbers of potential parasite sequences,
with the “vast majority of sequences” from deep water
plankton samples belonging to Marine Alveolate (MALV)
groups I and II (López-Garcı́a et al., 2001).
Although these protistan parasites are poorly known,
they may play important roles in ecosystem dynamics.
Protistan parasites can reduce the abundances of their
hosts, affecting their population dynamics, and inducing
mortality rivaling in magnitude that from zooplankton
grazing (Sousa, 1991; Coats, 1999; Coats and Park,
2002). Such reductions in the populations of hosts can
also have potential impacts on the predators of these hosts.
By killing, or reducing the fitness of, hosts in a speciesspecific and density-dependent manner, selective protistan parasites may also play a role in maintaining
ecosystem diversity (Hudson et al., 2006). Protistan parasite activity can lead to increased release of DOM and
particulate organic matter (POM) from their hosts as they
put them under increased physiological stress, or cause
them to lyse (Skovgaard, 2014). Such organic matter may
be taken up by bacteria or small eukaryotes, fueling the
microbial loop. Thus, understanding the roles of these
parasites is important to our overall understanding of
carbon flows and trophic interactions.
Newly generated next-generation sequencing (NGS)
data from the WAP marine ecosystem offered an opportunity to investigate the diversity, spatial and temporal
distributions, and relative abundance of protistan parasites. Millions of 18S rDNA barcode sequences from the
microeukaryote (0.2– 5000 mm) communities of the water
column and sediments in winter and the water column in
summer were used to obtain a first glimpse into the types
of parasites present in the region, how these parasites are
distributed in this ecosystem and in relation to potential
hosts, how parasite distributions change seasonally and
how much of the protist DNA in this ecosystem might be
attributed to parasites.
402
A. C. CLEARY AND E. G. DURBIN
j
ABUNDANT PARASITE 18S IN ANTARCTIC WINTER
METHOD
Field collections
Samples were collected in winter between 18 May and 3
June 2013 on the Research Vessel/Ice Breaker Nathaniel
B. Palmer cruise NBP1304. In summer, samples were collected between 10 and 21 December 2014 on the same
vessel on cruise NBP1410. Water samples were collected
in 12 L Niskin bottles on a CTD rosette. Surface water
was collected from the surface mixed layer at 20 m depth
in winter and at 5 – 10 m depth in summer (Table I).
Deep water was collected at 10 m above the seafloor within fjordic-bays and straits, and at 600 m in Palmer Deep
basin (bottom depth 1345 m) (Table I). In winter, for
each water sample, triplicate 2 L subsamples of whole
seawater were filtered by a peristaltic pump onto 0.2 mm
membrane filters, thus collecting all organisms or pieces
of organisms between 0.2 mm and 5 mm diameter (the
diameter of the pump tubing). In summer, for each water
sample, duplicate 100 or 200 mL subsamples of whole
seawater were filtered by gentle vacuum onto 0.2 mm
membrane filters. Much smaller volumes were used in
summer due to higher phytoplankton biomass. In winter,
surface and deep water samples were collected from
Flandres Bay, Gerlache Strait and Palmer Deep; surface
water only was collected in Andvord Bay (Fig. 1, Table I).
In summer, only surface water samples were collected,
and these were analyzed from Andvord, and Flandres’
Bays, and from Palmer Deep and the Gerlache Strait.
Filters were placed in individual cryovials and immediately frozen at 2808C. Temperature and salinity were
recorded simultaneously with sample collection with an
SBE 911plus CTD (SeaBird). Chlorophyll a was measured
fluorometrically from extracted pigments in triplicate on a
TD-700 fluorometer (Turner Designs) (Jespersen and
Christoffersen, 1987).
Sediment samples were collected in winter with a
megacorer (Ocean Scientific Instruments Limited). In
order to expose the sediment surface, overlaying water was
gently removed by a peristaltic pump and cores were
extruded to just below the level of the sediment surface.
The surface-most layer of sediment (3 mm) was sampled using sterile disposable scrapers, placed in cryovials
and immediately frozen at 2808C. Sediment samples collected all organisms less than 5 mm diameter.
Sediment was sampled from two locations in Wilhelmina
Bay, two locations in Andvord Bay, one location in
Flandres Bay and one location in Palmer Deep (Fig. 1,
Table I). Individual samples were collected from three
separate cores from one of the corings in Andvord Bay,
and triplicate samples were collected from a single core in
each of the other corings. All samples were shipped from
Chile to Rhode Island on dry ice and stored at 2808C
until analysis. Bottom depth was recorded by shipboard
Chirp 3260 echo sounder (Knudsen).
Laboratory processing
Total DNA was extracted from filters with the DNeasy
blood and tissue kit (Qiagen). Volumes of the initial lysis
buffers were all doubled to ensure the filter was submerged and all material was lysed. Total DNA was extracted
Table I: Sampling locations and metadata
Samples
Type
Date
Time
(local)
Depth
Latitude
Longitude
Location
Map
icon
Temp
(8C)
Salinity
(psu)
1, 2, 3
4, 5, 6
7, 8, 9, 79
10, 11, 12
13, 14, 15
16, 17, 18
19, 20, 21
22, 23, 24
25, 26, 27
28, 29, 30
31, 32, 33
34, 35, 36
37, 38, 39
66, 67
68, 69
70, 71
72, 73
74, 75
Water
Water
Water
Water
Water
Water
Water
Sediment
Sediment
Sediment
Sediment
Sediment
Sediment
Water
Water
Water
Water
Water
24 May
29 May
29 May
1 June
1 June
3 June
3 June
18 May
22 May
26 May
27 May
30 May
1 June
10 December
17 December
18 December
19 December
21 December
23:05
11:15
11:15
10:40
10:40
00:50
00:50
18:25
18:00
20:20
16:00
16:30
13:20
09:00
09:40
09:00
00:25
14:20
20
250
20
600
20
300
20
520
628
545
356
725
1345
5
5
10
5
10
264.8469
265.0468
265.0468
264.9668
264.9668
264.7949
264.7949
264.6857
264.5350
264.8123
264.8105
265.0030
264.9667
264.8245
264.9364
264.8954
265.0567
264.8149
262.6105
263.3014
263.3014
264.3549
264.3548
263.1214
263.1214
262.2350
262.2350
262.7333
262.7175
263.3110
264.3548
262.6412
264.3579
263.719
263.2035
262.6721
Andvord Bay
Flandres Bay
Flandres Bay
Palmer Deep
Palmer Deep
Gerlache Strait
Gerlache Strait
Wilhemina Bay
Wilhemina Bay
Andvord Bay
Andvord Bay
Flandres Bay
Palmer Deep
Andvord Bay
Palmer Deep
Gerlache Strait
Flandres Bay
Andvord Bay
1
2
2
3
3
4
4
5
6
7
7
8
3
9
10
11
12
13
20.89
0.72
20.77
1.43
20.86
20.24
20.58
33.66
34.47
33.64
34.66
33.65
34.51
33.84
0.26
0.02
0.18
0.01
0.26
0.02
0.24
0.19
21.33
21.03
0.20
0.51
33.86
33.33
33.39
33.55
34.07
6.8
0.42
1.42
13.14
Chlorophyll
Sample numbers correspond to independent water filters or sediment samples, except in the case of 79 which is a technical replicate for 9. Dates in May
and June are in 2013, dates in December are in 2014, time is in Chilean local 24 h time, depth is in meters, map symbol corresponds to Fig. 1, chlorophyll
a is in mg m23. Salinity and temperature data are not available for sediment samples.
403
JOURNAL OF PLANKTON RESEARCH
j
VOLUME 
j
NUMBER 
j
PAGES –
j

Fig. 1. Map of sampling locations. Numbers in the figure correspond to map icon numbers in Table I.
from 0.25 g of each sediment sample using the PowerSoil
kit (MoBio) with the bead-vortex lysis option as per the
manufacturer’s instructions. Only samples of the same
type (water/sediment) were extracted on the same day.
All extractions were conducted in a sterilized laminar
flow hood with project-dedicated pipettes, tips and chemicals, to minimize possible contamination.
Partial 18S rDNA was amplified over the v7 region
using universal eukaryotic primers (960f and 1200r of
Gast et al., 2004), with the following slight modifications.
One variable position (Y) was substituted for a T in the
Read1 primer because the primers as originally published have been observed not to efficiently amplify
Mesodinium spp. and related ciliates due to a 1 bp mismatch (A.C.C., unpublished data). This modification has
been applied in previous analyses amplifying a wide
range of marine protists (Cleary et al., 2015). Adaptors for
Illumina sequencing were added to the primers, along
with a variable number (0 – 3) of ambiguous bases to
offset the amplicons and increase the variability at each
read position for improved base calling, with the read 1
adaptor on the reverse primer and the read 2 adaptor on
the forward primer (Read1 primer: TCGTCGGCAG
CGTCAGATGTGTATAAGAGACAGfNNNgGGCTY
AATTTGACTCAACRCG; Read2 primer: GTCTCGT
GGGCTCGGAGATGTGTATAAGAGACAGfNNNg
GGGCATCACAGACCTG). In summer, the variable
number of NNNs was not included in the primers as
advances in illumina chemistry and base calling had
rendered these obsolete.
For winter samples, each reaction contained a final
concentration of 1 Pfu Ultra II clear buffer (Agilent),
1 bovine serum albumin (New England Biolabs),
0.25 mM equimolar mixture of all four deoxynucleotide
triphosphates (dNTPs) (Promega), 0.1 mM each primer
(forward and reverse), 1 Pfu Ultra II polymerase
(Agilent) and 20% by volume DNA template at extracted
concentrations. Summer samples were amplified with
1 GoTaq green master mix, 0.1 mM each primer (forward and reverse) and 20% by volume DNA template at
extracted concentrations. Differences in the fidelities of
Taq and Pfu polymerase mean comparisons of OTU
richness should be viewed with caution; however, these
polymerase differences are unlikely to affect the proportions of sequence reads recovered from broad taxonomic
groups. Thermocycling for all samples consisted of 958C
for 30 s, followed by cycles of 948C for 30 s, 588C for
45 s and 728C for 30 s, with a final extension of 728C for
5 min. Samples were amplified for the minimum number
of cycles necessary to obtain sufficient amplicon DNA for
404
A. C. CLEARY AND E. G. DURBIN
j
ABUNDANT PARASITE 18S IN ANTARCTIC WINTER
sequencing, in order to reduce amplification biases and
over-representation of abundant targets. Winter water
samples were all amplified for 35 cycles, as were sediment
samples 28, 29, 30, 35 and 37, while the remaining sediment samples and the summer water samples were amplified for 30 cycles. Amplicon presence and size (230–280
bp, excluding primers and adaptors) was confirmed with
gel electrophoresis and UV visualization. Amplifications of
no-template blanks included in each PCR showed no
signs of contaminating DNA in gel images.
Amplicon purification, tag addition and sequencing
were done at the URI Genomics and Sequencing Center.
Amplicons were AmPure (Agencourt) cleaned, re-amplified
in six cycles of PCR to add unique sample identification tags and quantified on a BioAnalyzer (Agilent). All
winter amplicons were pooled into one half Illumina
Miseq run, and sequenced for 2 250 cycles with V2
chemistry, allowing for almost complete overlap of the
amplicon. Summer amplicons were included in a separate sequencing run (along with 75 other samples) with
the same chemistry and parameters as the winter samples.
Each filter and each 0.25 g sediment sample were sequenced separately, and results are shown for all filters individually, with biological replicates shown side by side.
Data analysis
Data were demultiplexed in the Illumina default pipeline,
and analyses commenced from these demultiplexed reads.
Paired ends of reads were joined if the entire overlap
region was 100% identical in both read directions; if the
overlap region was not identical, both reads were discarded. Amplicons were then assigned sample-specific
names for each of the demultiplexed samples, and all
samples within a season pooled for further analysis. At
this stage, additional quality control on the sequences was
applied, following default parameters in split_libraries.py.
Primers, and any sequence data beyond the end of the
amplicons were trimmed by removing the reverse primer
and any following sequence, then reverse complementing
the sequences and repeating the procedure, for a final file
of trimmed, 50 -30 oriented sequences. Any sequences in
which both the exact primer sequences were not found
were again discarded. This fairly stringent approach to
quality control likely eliminated most sequencing errors,
as such errors are unlikely to occur identically in the two
directions of sequencing. Amplicons were then clustered
into 97% sequence identity Operational Taxonomic
Units (OTUs) with Uclust (Edgar, 2010). In analysis of
environmental 18S sequences, 97% is a common clustering threshold (Sogin et al., 2006). Any OTU with only a
single sequence in it was discarded, as such sequences
may be erroneous, and even if real, add significantly to
computation time without typically changing the overall
results of NGS data analysis. OTUs with chimeric sequences were detected with the blast-fragments approach
with a two fragment search parameter (Altschul et al.,
1990), and removed from the data. All of the above analyses were conducted in Qiime v.1.8 with default parameters unless otherwise specified above (Caporaso et al.,
2010).
Taxonomic identity was assigned to each OTU
through automated comparison in Qiime with the Silva
database v. 111 (Wang et al., 2007; Quast et al., 2013) and
confirmed for parasite OTUs by comparison with the
PR2 database (Guillou et al., 2012). By both database
comparisons, 94% of parasite OTUs sequence reads
were classified into the same groups and only results from
Silva are presented here. Taxonomic level of identification varied between OTUs, due to variation in the completeness of reference sequence sets, and the genetic
variability within the amplicon. Selected BLAST searching of the full GenBank database was used to confirm
taxonomic assignments (Altschul et al., 1990; Morgulis
et al., 2008). Because multicellular organisms are unlikely
to be quantitatively sampled in the small volumes of water
and sediment analyzed here, and thus their appearance
in samples was noisy and stochastic, these OTUs, which
accounted for 9% of the initial winter data, were also
removed from the final data set. Removing chimeras
and metazoans reduced the overall winter data set by
3085 OTUs and 826 742 sequence reads. All OTUs
were next manually classified as parasitic, non-parasitic
or unknown based on literature. Literature searches were
run on taxa name and “parasite”, beginning with kingdom level groups, and proceeding to lower taxonomic
levels. OTUs were only considered parasitic if all available literature on any organism in the group agreed with
the designation of parasitism, or if literature explicitly
stated that all members of a group were parasitic. If no
reports of parasitism within a group could be found, or if
even at the lowest taxonomic level sequences were
resolved to, known members of the group had both parasitic and other lifestyles, organisms were considered free
living. This approach leads to a conservative estimate of
parasitism as it excludes parasites which are closely
related phylogenetically to non-parasites.
To compare the communities of parasites and free
living organisms, a dissimilarity matrix of all winter samples was constructed using the Bray– Curtis metric (Bray
and Curtis, 1957) based on the sequence read counts for
each of the parasite OTUs, normalized by the overall
( parasite and free-living) total sequence read for each
sample. Likewise, a dissimilarity matrix was constructed
for all free-living OTUs. Parasite and free-living communities were compared in a side-by-side cluster analysis
405
JOURNAL OF PLANKTON RESEARCH
j
VOLUME 
based on Bray– Curtis distances. Principal coordinates for
the parasite assemblages in each sample (metric multidimensional scaling) were calculated and visualized in
MatLab. ANOSIM was used to determine the significance
of groupings of parasite OTU assemblages by source type
(surface water, deep water, sediment) as implemented by
the Fathom toolbox in MatLab (Jones, 2014).
Interactions between parasite OTUs and potential
host OTUs were explored using an OTU-wise Bray–
Curtis metric for all winter samples. Read counts were
normalized by the total for each OTU across all samples
(as opposed to the total for each sample across all OTUs
used above), in order to reduce the effects of variations in
copy number or organism size This approach creates in
effect a co-occurrence network of OTUs, with no a priori
assumptions of host– parasite interactions, and is similar
to that taken by Lima-Mendez et al. (Lima-Mendez et al.,
2015). Parasite – host interaction analysis was limited to
the 500 most abundant OTUs overall to avoid artifacts
due to stochasticity at low abundance. Parasite – host analysis also excluded organisms for which taxonomy was
not resolved below kingdom level, or with very phylogenetically close parasites and hosts (such as within the
dinoflagellates) to avoid erroneous correlations due to potentially imperfect OTU picking. Selected parasite – host
pairs identified in this exploratory analysis were further
investigated with linear correlations of OTU read counts
normalized by sample.
R E S U LT S
A total of 1028 parasite-associated OTUs which encompassed 309 192 sequence reads were recovered from the
Antarctic coastal environment in winter, out of a total
sample of 8539 OTUs and 1 990 675 cleaned sequence
reads. In the much smaller summer surface water data
set, only 184 parasite OTUs were identified, encompassing 11 723 sequences, out of a total sample of 3249
OTUs and 632 351 reads. In winter, 393 of these OTUs
were found in surface water samples, 592 were found in
deep water samples and 564 were found in sediment
samples. Winter sequence reads from parasite-associated
OTUs made up between 5.6% of the total sequence
reads in Sample 22 (Wilhelmina Bay sediment) and
73.0% of the total sequence reads in Sample 11 (Palmer
Deep deep water), with an overall winter mean of 21.4%
and median of 13.8% of the sequence reads in a sample.
In summer, parasites ranged in relative abundance from
0.9% (Flandres Bay surface waters) to 2.5% (Palmer
Deep surface water), with an overall summer mean and
median of 1.8%. Only two parasite OTUs were found
in every winter sample and both were classified as
j
NUMBER 
j
PAGES –
j

Syndiniales I (MALV I). A total of 108 OTUs were
present in all sample types in winter (surface water, deep
water, sediment), 83 of which were in MALV I and II.
Deep water winter samples showed the highest percentages of parasite-associated OTU reads, with 52.3% of
the deep water reads falling into these groups. About
13.1% of winter surface water sequences were classified
as parasite-associated OTUs. Sediments showed the
lowest relative abundance of parasite reads in winter,
with 9.1% of total reads classified into parasite-associated
OTUs (Table II).
Parasite OTUs included a diverse range of organisms.
Most of the parasite OTUs belonged to the alveolates.
The most abundant group of sequences belonged to the
Syndiniales, making up 11.3% of the total sequences,
and 62.2% of the parasite sequences. Within the syndinians, organisms classified solely as belonging with
Syndiniales group II made the greatest contributions to
both number of reads and number of OTUs. None of
these parasites could be classified to lower taxonomic
levels, suggesting a large biodiversity which remains underexplored. Also within this group, Amoebophyra spp.
(MALV II) were present mainly in water samples, but also
in sediments. In the MALV I, Euduboscquella spp. were
found mainly in deep water, and Haplozoon spp. were
found to be most abundant in surface waters (Table II,
Fig. 2). In MALV IV, we observed Hematodinium spp.
(MALV IV), found mainly in surface water samples.
Dinoflagellate parasites observed included: Chytriodinium
spp., found mainly in water samples, and Blastodinium spp.,
found in all sample types (Table II, Fig. 2). The related
Perkinsidae were observed mainly in the sediment.
Apicomplexa OTUs were predominantly gregarines, with
water samples containing mainly a Lankesteria spp. eugregarine, while sediment samples, where Apicomplexa were
most prominent, were more mixed in their apicomplexan
assemblage with notably more Lecudina spp. eugregarines,
and more Cryptosporida spp. Apicomplexa also included
three OTUs which could not be classified to lower taxonomic levels, again suggesting potentially currently
unknown diversity within this group. Other alveolate parasites found included Cryptocaryon spp. ciliates which were
observed in all sample types, and ellobiopsids which were
found mainly in water samples (Table II, Figs 2 and
Fig. 3).
Some parasites were also observed from Rhizaria.
Rhizarian parasites included Paradinium spp. which was
observed across all sample types, and phytomyxea which
was found mainly in sediments. Two groups of parasites
belonging to the stramenopiles were observed; Pirsonia
spp. was found mainly in deep water and sediments, and
Solenicola spp. was found mainly in surface waters. The
only parasite group found which did not fall within the
406
A. C. CLEARY AND E. G. DURBIN
j
ABUNDANT PARASITE 18S IN ANTARCTIC WINTER
Table II: Parasite groups encountered in 18S rDNA sequences from the WAP
Winter
Group
#OTUs
% Surface
% Deep
% Sediment
Summer
% Surface
Amoebophrya (MALV II)
Apicomplexa
Blastodinium
Chytriodinium
Cryptocaryon
Euduboscquella (MALV I)
Ellobiopsidae
Haplozoon
Hematodinium (MALV IV)
Ichthyophonae
Paradinium
Perkinsidae
Phytomyxea
Pirsonia
Solenicola
Syndiniales, unclassified
Syndiniales group I (MALV I)
Syndiniales group II (MALV II)
Totals
124
216
4
3
17
5
4
9
9
8
19
88
19
7
1
54
137
304
1028
2.766
0.058
0.047
0.001
0.002
0.003
0.002
1.185
0.006
0.004
0.029
0.003
0.001
0.001
0.001
0.683
4.835
3.499
13.125
3.687
0.130
0.008
0.001
0.015
0.377
0.340
0.390
0.210
0.020
0.036
0.062
0.002
0.032
0.000
1.401
9.562
36.063
52.337
0.013
3.004
0.003
0.000
0.032
0.024
0.000
0.000
0.002
0.027
0.055
0.589
0.132
0.037
0.000
0.463
1.645
0.277
9.144
0.376
0.003
0.000
0.000
0.000
0.000
0.000
0.017
0.000
0.003
0.000
0.000
0.000
0.001
0.000
0.051
1.055
0.349
1.828
Example known hosts
Ref
Dinoflagellates, ciliates,
Arthropods, polychaetes, Chaetognaths
Copepods
Copepod eggs
Fish
Cilliates
Crustaceans
Marine worms, mollusks, larvacea
Crustaceans
Fish, bivalves
Copepods
Mollusks, dinoflagellates
Diatoms
Diatoms
Diatoms
Copepods, radiolarians, amphipods,
larvacea,
1, 15, 16, 18
1, 2, 19
1, 3, 13,17
1
5
1, 16,17
1, 3
6, 14, 17
3
1, 7, 17
1, 8
9, 2
10
1, 4, 11
1
1, 12, 17
# of OTUs is the total number of observed distinct OTUs for each group. % indicates the percent of the total sequence reads in each sample type which
were attributable to each group. References: (1) Skovgaard (2014); (2) Moreira and López-Garcı́a (2003); (3) Konovalova (2008); (4) Tillmann et al. (1999);
(5) Wright and Colorni (2002); (6) Leander et al. (2002); (7) Glockling et al. (2013); (8) Skovgaard et al. (2008); (9) Chambouvet et al. (2014); (10) Neuhauser
et al. (2011); (11) Kühn et al. (2004); (12) Guillou et al. (2008); (13) de Vargas et al. (2015); (14) Sousa (1991); (15) Lima-Mendez et al. (2015); (16)
Chambouvet (2009); (17) Coats (1999); (18) Cachon (1964); (19) Takahashi et al. (2011).
Fig. 2. Distribution of sequence reads for parasite and free-living organism OTUs across all winter samples. Upper rectangles indicate sample
source (surface water, deep water, sediment); lower rectangles indicate sample location (Andvord Bay, Flandres Bay, Gerlache Strait, Palmer Deep),
sample numbers along the x-axis correspond to Table I and are arranged within sample type by increasing bottom depth. The high percentage of
parasites in sequencing reads from deep water samples is striking.
Stramenopiles –Alveolata – Rhizaria (SAR) complex was
the holozoan protist Ichthyopohonae spp. which was
observed mainly in deep water and sediment samples
(Table II, Figs 2 and 3).
Parasites showed different assemblages in winter between surface waters, deep waters and sediments. These
differences were statistically significant with ANOSIM
P ¼ 0.001, for all possible pairwise comparisons. Surface
407
JOURNAL OF PLANKTON RESEARCH
j
VOLUME 
j
NUMBER 
j
PAGES –
j

Fig. 3. Parasite sequence reads across all winter samples. Formatting follows Fig. 2. The high contribution from Syndiniales and related organisms
(Amoebophrya and Hematodinium) is striking, particularly in surface and deep water samples.
waters contained large fractions of Syndiniales and subgroups, as well as noticeable contributions from haplozoons (Fig. 3). Deep water parasite assemblages were
dominated by Syndiniales and subgroups, particularly
Group II, with relatively low contributions from other
groups. Sediment parasite assemblages showed much
lower abundances of syndinians than was observed in the
water column, with the assemblage composed largely of
Apicomplexa, particularly eugregarines, with contributions from Perkinsidae as well as small contributions from
Syndiniales and subgroups (Fig. 3). The notable exception to this trend is Palmer Deep, where samples showed
relatively more Syndiniales, particularly Group I. Palmer
Deep also showed differences within the Apicomplexa,
with all three surface water samples containing eugregarine sequences of an OTU, not prominent in other samples, which showed taxonomic affinities to eugregarines
isolated from flatfish. These differences in assemblages
are evident in a principal coordinates analysis (Fig. 4).
Samples cluster most strongly by sample type, but within
sample type also show clustering by location. Parasite
assemblages are closely tied to non-parasite assemblages.
Parasite OTU assemblages exhibited very similar clustering patterns to non-parasite OTU assemblages (Fig. 5).
Summer surface water parasite prevalences were over
a factor of 6 lower than those observed for similar samples
in winter. The observed assemblage of parasite types,
however, was similar (Fig. 6, Table II). Both summer and
winter surface water sample parasite assemblages were
dominated by syndinians. In winter, there were roughly
equal contributions from Group I, Group II and
Amoebophrya spp., while in summer, a lower relative abundance of Amoebophrya was observed, and Syndiniales Group
I made the largest fraction of the parasite sequences, particularly in the samples within fjordic-bays. Haplozoon spp.
were at lower relative abundances in the parasite assemblage in summer when compared with winter.
Considering relationships between specific parasites
and potential hosts, the strongest correlations observed
were between syndinians and radiolarians. Twelve syndinian OTUs correlated with r 2 .0.9 with individual radiolarian OTUs (Fig. 7). Considering all syndinians (Groups
I and II) and all radiolarians as groups, the positive linear
correlation between the percent abundances of these
groups over all samples showed a slope of 9.4 and an r 2 of
0.94 (Fig. 7).
DISCUSSION
Parasite OTU spatial distributions indicate active ecological interactions between specific parasite groups and
their hosts in the WAP. Parasite OTU assemblages clustered in a nearly identical pattern to free-living OTU
assemblages, potentially indicating close biological interactions between parasite and host species or groups
(Fig. 5). The Syndiniales (Groups I and II), the most
abundant parasite sequence groups observed, were correlated with distributions of radiolarians, with both of these
groups having their highest relative abundance in deep
408
A. C. CLEARY AND E. G. DURBIN
j
ABUNDANT PARASITE 18S IN ANTARCTIC WINTER
Fig. 4. Principal coordinates analysis (metric multidimensional scaling) of the winter parasite communities sampled. Shape indicates the location
of each sample, with shade indicating the sample type. Sample type is clearly the dominant structuring factor among these assemblages.
water samples. Summer research at a station within a few
miles of our sampling region has also shown tentative
correlations between poorly known alveolates and radiolarians, adding strength to the argument that this is a
biologically meaningful interaction (Luria et al., 2014),
and suggesting it may be a year-round relationship.
Sediment parasite assemblages were dominated by apicomplexans, mainly gregarines, and particularly OTUs
affiliated with the eugregarines. Gregarines may be parasitizing groups which were found mainly in the sediment,
such as nematodes and Platyhelminthes. It has been suggested that annelids are important hosts for gregarines
(Leander, 2007), but we observed annelid sequences
across all sample types, while gregarines were restricted
to the sediment, suggesting annelids may not be the most
important host for these groups in the WAP. Phytomyxea
and Pirsonia, both groups known to parasitize diatoms
(Tillmann et al., 1999), also had their highest abundances
in the sediment. Although initially counter-intuitive,
these diatom parasites in the sediment actually correlated
well with the distribution of diatom sequences, as diatom
biomass in the water column was very low in winter, but
sequences for what are assumed to be resting spores of
certain Chaetoceros groups were abundant in sediment
samples (Cleary, 2015). Parasites have been previously
observed within resting cysts of dinoflagellates, with the
ability to survive periods of unfavorable conditions, and
then become infective again when the host cell leaves
dormancy and returns to vegetative growth (Chambouvet
et al., 2011). The diatom parasite sequences observed in
sediments may suggest a similar life history strategy for
these organisms potentially parasitizing Chaetoceros hosts.
Parasite abundances in surface waters showed striking
differences between summer and winter, with ,2% of
summer sequences classified as parasite, while in winter,
this proportion was over 13%. All groups of parasites
showed this decline, with many groups which were present in winter not detected in summer, although this may
be partially due to a much smaller sample size. Summer
abundances of syndinians observed here (total 1.73%)
show very similar average relative abundance to early
results for summer in the adjacent Scotia Sea region (2%
MALV II) from DGGE (Diez et al., 2004). NGS data
provide only information on relative abundance, not
total abundance. Therefore, these sequence data alone
cannot determine whether parasites had lower overall
abundances in summer, or were just a lower proportion
of the total planktonic biomass. Previous analyses on
Syndiniales in an Arctic fjordic-bay with qPCR have also
seen a decrease from winter to spring, which was attributed to different water masses (Thomson, 2014). However,
there is no clear separation of our summer versus winter
samples in temperature– salinity (T– S) space, suggesting different water masses may not be the most likely explanation for the lower summer parasite abundances
observed here.
409
JOURNAL OF PLANKTON RESEARCH
j
VOLUME 
j
NUMBER 
j
PAGES –
j

Fig. 5. Cluster analysis of all winter samples for both parasite and non-parasite OTUs. Samples show strong clustering by sample type (surface
water, deep water, sediment), as well as clustering by location. Parasite and free-living communities show very similar clustering, suggesting
biological interactions.
The low summer relative abundance of parasites is
somewhat surprising, since it is during the summer months
when planktonic biomass peaks and biological activity is
at its greatest. The abundances of many potential hosts,
such as diatoms, photosynthetic dinoflagellates and copepods, were much higher in summer, particularly around
the seasonal spring bloom, than they were in winter, as
can be partly seen in the chlorophyll a values (Table II).
Protistan parasites have rapid generation times, on the
order of 24–48 h (Coats, 1999), and so should be able to
keep pace with the seasonal increases in available hosts.
Three potential explanations for the higher relative
abundance of parasite sequences in winter in the WAP are
worth considering. First, parasites may be targeting some
of the select species or groups which are more abundant in the fjordic-bays in winter. The one host group
identified in our parasite – host exploratory analysis, the
radiolarian, also showed a lower relative abundance in
surface waters in summer (0.010%) than in winter
(0.467%). However, Radiolaria in winter were more abundant in the deeper water column, and this environment
was not analyzed in summer, so this comparison should
not be over-interpreted. One prominent pelagic species
which was much more abundant in winter than in summer was Euphausia superba, the Antarctic krill (Cleary et al.,
in review). Apicomplexan parasites are known from krill
hosts (Takahashi et al., 2011), and their lower summer
abundance within fjordic-bays may be related to the
lower krill abundance in summer. Syndinians are known
to infect a wide range of crustaceans (Skovgaard, 2014),
and although they have not been described from krill, it
is possible that they too infect these keystone zooplankters. However, the spatial distribution of syndinians in
winter does not support the idea that their abundances
410
A. C. CLEARY AND E. G. DURBIN
j
ABUNDANT PARASITE 18S IN ANTARCTIC WINTER
Fig. 6. Summer distributions of parasite sequences. (A) All summer sequences, in which free-living organisms dominate the assemblage. (B)
Parasite sequences alone, showing the prevalence of Syndiniales, particularly Syndiniales I, similar to the observed distribution of parasites in
winter.
are driven by abundances of E. superba. Syndinians were
of equal (surface) or greater (deep) relative abundance in
the offshore Gerlache and Palmer Deep samples when
compared with the within-bay samples, whereas krill
were much more abundant within bays than in the more
open waters sampled (Cleary et al., in review).
Secondly, parasites may be more successful at infecting
physiologically stressed hosts. Protistan parasite infection
rates have previously been observed to be higher in photosynthetic hosts which are under nutrient stress (Coats,
1999). Phytoplankton hosts in winter are likely to be experiencing physiological stress due to very low light levels,
and zooplankton may also be experiencing stress due to
low availability of phytoplankton prey in this dark season.
The observed ratio of diatom – parasites to diatoms was
highest in winter sediments, and lowest in summer surface
waters. This distribution potentially corresponds to the
physiological stress of diatom cells, as cells which have
sunk down to the sediment in winter are likely stressed,
while diatoms in the sunlight surface waters of summer
may be much less so.
A third potential explanation of the lower observed
relative abundances of parasites in summer is differences
in the time required for parasites to locate a new host.
This explanation may be particularly relevant for parasites which infect motile metazoan hosts, such as the
dominant syndinians, which infect copepods and potentially other crustacean zooplankton. Water sampling protocols used here were designed to catch free-living
protists with relatively low motility and escape responses,
and are unlikely to have quantitatively sampled larger,
motile metazoans such as copepods. Copepods have welldeveloped escape responses and the ability to detect
sheer flows, and may therefore avoid the Niskin bottles,
or if captured, may avoid the outflow to the Niskin
spigot. That we did not quantitatively sample mesozooplankton is further suggested by the low (9%) abundance
of metazoan sequences in our data set. Given that we
may not be truly capturing the mesozooplankton, the
parasites which are inside these mesozooplankton hosts
may also not have been detected. Instead, the sequences
observed may be derived largely from spores or infective
stages of parasites. In summer, copepods, particularly
Metridia spp., were highly abundant in the sampled
fjordic-bays (E.G.D., data not shown), whereas in winter,
copepods were very sparse, with only a few individuals
per m3 (Cleary et al., in review). The distance between
evenly distributed potential hosts is proportional to (1/
hosts-per-volume)(1/3), so as the density of hosts increases
in the summer, the distance, and assuming a constant
swimming/drifting speed, time, for a parasite to find a
new host declines rapidly. With less time required to find
411
JOURNAL OF PLANKTON RESEARCH
j
VOLUME 
j
NUMBER 
j
PAGES –
j

Fig. 7. Correlations between Radiolaria and Syndiniales. All plots show radiolarian hosts on the x-axis, with syndinian parasites on the y-axis, with
both axis indicating sequence read abundance as a percent of total sequence reads per sample. (A) Various Syndiniales OTUs correlated with
Radiolaria OTU 13310; (B) various Syndinales OTUs correlated with Radiolaria OTU 15634; (C) all Syndiniales OTUs when compared with all
Radiolaria OTUs.
a new suitable host, each parasite would spend a lower
fraction of its life freely floating in the water column, and
by extension, at any given point in time, fewer of the
parasites would be in the water column, and available to
our filtering and sequencing analysis. Further research
with more quantitative and targeted approaches, such as
qPCR or FISH of parasites, or transcriptomic analyses of
hosts will be needed to assess the merits of these possible
explanations.
The high proportions of parasite sequences reads
observed in these winter samples from the Antarctic beg
the question: Are these observed high abundances of
parasite OTU sequences indicative of seasonally high
parasitic activity in Antarctic marine ecosystems? There
412
A. C. CLEARY AND E. G. DURBIN
j
ABUNDANT PARASITE 18S IN ANTARCTIC WINTER
are several potential explanations for the observed high
abundances: (i) the sequence abundances are an artifact
of the rDNA sequencing approach; (ii) groups classified
as parasitic in fact contain organisms with other lifestyles;
(iii) parasites have long-lived spores or resting stages; and
(iv) parasite abundance in the Antarctic in winter is high.
Each of these explanations likely plays a part in explaining the overall results, and the merits and likely impact of
each are discussed below.
As mentioned above, 18S NGS does not provide data
on the overall abundance of organisms, but it has traditionally been used to infer relative abundances of different organisms and groups of organisms (Not et al., 2009).
There is potential for biases in preservation efficiency,
18S copy numbers per cell, DNA extraction efficiency,
PCR primer binding, polymerase extension, amplicon
purification, sequencing and quality control. Despite
these potential sources of bias, studies with bacterial
mock communities have shown close resemblances of sequence read proportions with true cell abundance proportions (Jumpstart Consortium, 2012), and limited data
available for eukaryotes suggest that with appropriate
primers and data analysis, 18S sequence proportions are
similar to community biomass proportions (Egge et al.,
2013). Steps were taken to minimize the effect of all of
these potential sources of bias in this analysis. PCR
primers used have been optimized to efficiently amplify
all major groups of eukaryotes (Cleary et al., 2015).
Differences in the number of copies of the 18S gene per
cell or per unit bio-volume have been observed in some
groups (Zhu et al., 2005). However, some of the groups
notorious for high copy numbers were observed at low
abundances in these Antarctic samples. For example, the
free-living dinoflagellates have been suggested to be a
group prone to high 18S copy numbers, yet they only
made up 1.5% of the total sequences observed; if copy
number variations were the cause of the observed high
abundance of parasite sequences, one would expect these
closely related free-living organisms to also be abundant
in the data set. This suggests that while 18S copy variations are certainly present, they are unlikely to be the full
explanation for the observed high proportions of parasite
sequences. The relative abundance of parasite sequences
observed in summer falls within the range of parasite
biomass as a proportion of total biomass which has been
observed in other marine systems (Kuris et al., 2008).
That the summer, productive season shows values consistent with other methodologies suggests that the high
winter observations are unlikely to be fully explained as
an artifact of the 18S rDNA NGS sequencing approach.
Many of the groups of parasites found in this study are
very poorly known. Little is known about their diversity,
abundances, distributions, host preferences, morphology
or ecology (Skovgaard, 2014). This is particularly true of
the largest group of parasite sequences observed, the
Syndiniales (MALV Groups I and II), for which there are
many uncertainties (Bråte et al., 2012). Here, organisms
were classified as parasites if literature described the
group to which they belong as obligately parasitic, or if
all described organisms within the narrowest group to
which the OTU could be identified were obligate parasites. However, it may be possible that some of these
groups contain as-yet undiscovered species with other
lifestyles, such as free-living or mutualistic symbiosis.
This has been previously suggested for some radiolarianassociated alveolates (Bråte et al., 2012). Misclassification
of such hidden free-living organisms as parasites might
explain some of the high winter abundance of parasite sequences we observed; however, the inverse is also
true, that parasites within groups categorized as freeliving or mixed have been excluded as parasites, and
might be artificially depressing the abundance of parasite
sequences.
Many parasitic organisms have some form of spore or
infective stage, allowing them to spend time outside of a
host in the process of finding a new host (Hudson et al.,
2006). It is possible that some fraction of the parasite
sequences observed came not from active parasites but
from some type of spore. However, many of these spores
are very short-lived, suggesting that they would be unlikely to form a large reservoir of DNA sequences in the
environment. For example, syndinians and related organisms are thought to have very short-lived spores, and to
require new hosts within a matter of days (Coats and
Park, 2002). Amoebophyra spp. (MALV I) spores show exponential declines in abundance, with most spores disintegrated within 3 – 13 days, and even among surviving
spores, ability to infect declined rapidly over time since
production (Coats and Park, 2002).
It may be possible that some of the parasite organisms
observed in this study have longer-lived spores, potentially as an adaptation to the extreme seasonality in the
abundance and biomass of potential hosts in Antarctic
marine ecosystems. This has been suggested for other
dinoflagellate parasites, as a mechanism to survive
periods of low host abundance, and such cyst or cyst-like
cells have been observed for three parasitic dinoflagellate
groups (Paulsenella, Disodinium, Euduboscquella) (Coats,
1999). The production of long-lived resting cysts or
spores may partially explain the observed seasonal
pattern, with parasites, potentially in the form of overwintering spores, abundant in winter and much less
abundant during the productive summer season. In
austral winter in the Antarctic Peninsula, phytoplankton
biomass is very low, with measured values during the
time of sampling consistently ,0.5 mg chlorophyll a L21,
413
JOURNAL OF PLANKTON RESEARCH
j
VOLUME 
whereas in summer in these same regions, chlorophyll a
concentrations can exceed 13 mg L21 (Table II). It might
thus be advantageous for protistan parasites which rely
on planktonic hosts, such as diatoms or copepods, to have
the capacity to survive as a spore over the winter period
with low abundances of hosts in this ecosystem.
Correlations between parasite and host communities,
and between individual parasite and host groups, suggest
these parasites are likely to be active components of the
ecosystem, rather than simply a collection of resting
spores. The distribution of parasite sequences also may
argue against these sequences deriving from resting
spores; resting spores might be expected to sink out of the
water column, and yet the sediment actually showed the
lowest percentage of sequences belonging to parasite
groups and a very distinct assemblage from that in the
water column (Fig. 3). Previous analysis of Perkinsidae
parasite sequences have been shown to represent ribosomally active cells in marine sediments, suggesting their
DNA abundance is indicative of an ecologically important role (Chambouvet et al., 2014).
It appears likely that protist parasites are more abundant, and more ecologically important than has been
realized in the WAP coastal waters in winter, even given
all of the above potential secondary explanations and
caveats. Models of Antarctic marine food webs have not
typically included parasites (Melbourne-Thomas et al.,
2013), yet they are potentially important. Parasitism can
divert carbon and energy out of the classic phytoplankton – krill –whale food chain, and into the microbial loop
as POM and DOM released from ailing and dying hosts.
The microbial food web is thought to be increasing in
importance in this northern WAP region, as a result of
ongoing anthropogenic change (Sailley et al., 2013). Previous
sequencing in austral summer near our sampling site also
observed a sequence assemblage dominated by alveolates
related to dinoflagellates, which pigment analysis suggested were not photosynthetic (Luria et al., 2014). These
alveolates may be related to the syndinian parasite
groups which we report here, suggesting that potentially
the importance of the role of parasites may be temporally
variable over the year (Luria et al., 2014).
Parasites can also have effects on the population dynamics of hosts and the diversity of the ecosystem more
broadly. Parasitic infections among the plankton can
be widespread, with observations of over 80% of host
Gymnodinium dinoflagellates infected with Amoebophrya, and
29% of host Paracalanus copepods infected with Syndinium
(Coats, 1999). Parasites can cause mortality among microplanktonic hosts rivaling that caused by grazing zooplankton (Coats, 1999). Syndinium (MALV IV) infestation
is estimated to cause copepod mortality comparable with
predation mortality with rates as high as 42%
j
NUMBER 
j
PAGES –
j

mortality per day reported (Konovalova, 2008), Perkinsidae
or Haplosporidium infections can wipe out 90% of oyster
beds (Sousa, 1991) and Amoebophrya spp. (MALV II) have
even been suggested as a biocontrol on harmful algal
blooms as they can remove over 50% of their hosts daily
(Coats and Park, 2002; Skovgaard, 2014). Many parasites
are highly host-specific, parasitizing only a single species
(Tillmann et al., 1999; Hudson et al., 2006; Skovgaard,
2014). Such high-specificity parasites may play a role in
maintaining diversity and structuring seasonal succession
within Antarctic marine ecosystems; by causing mortality
or reducing fitness of an abundant species, parasites may
create opportunities for other species within the ecosystem.
Parasitism has traditionally been considered to be more
important in warmer ecosystems (Rhode, 1984), and thus
as temperatures continue to increase, parasitism may
play a role in the increasing importance of the microbial
loop. The prevalence of protozoan marine parasites within a single geographic area over interannual variations
has also been linked to environmental temperatures, with
increasing infections by Perkinsus spp. in the Gulf of
Mexico under warmer La Niña waters (Harvell et al.,
2002). The role of parasites in the WAP may therefore be
increasing as this region continues to experience ongoing
anthropogenic climate change and some of the most
rapid warming observed globally (Ducklow et al., 2007).
Thus, incorporating parasites and their roles in regulating plankton populations into ecosystem models may
allow for better understanding and predictions of the
trends in plankton species dynamics in the WAP.
Parasitism may be more important than commonly
considered in marine ecosystems more broadly. When
considering metazoan parasites on larger organisms such
as fish, the Antarctic has been found to have lower parasite loads than other parts of the world ocean (Rhode,
1984). Here, we present data on protistan parasites for a
limited area from the WAP. DNA sequencing technologies have been rapidly improving, and with the public
availability of reference databases for eukaryotes (Quast
et al., 2013), it has only very recently become feasible to
conduct and analyze broad-scale surveys of eukaryote
communities. Data from 18S NGS and clone libraries in
other regions of the world ocean suggest the unexpectedly
high prevalence of parasite sequences observed here may
be a wider phenomenon. The largest survey available to
date, throughout the temperate and tropical euphotic
zone, also found high abundances of parasite-associated
sequences, with these sequence reads accounting for 53%
of the heterotrophic protists in their smallest size fraction,
with lower per cent in the larger size fractions (de Vargas
et al., 2015). That abundance is still lower than the 52% of
all protists (of all sizes combined) observed in deep waters
414
A. C. CLEARY AND E. G. DURBIN
j
ABUNDANT PARASITE 18S IN ANTARCTIC WINTER
here, but indicates these parasites may be abundant globally. Sequence assemblages from the English Channel
included 18% parasite sequences (Genitsaris et al., 2015).
Meta-analyses of clone libraries suggest Syndiniales make
up over half of the dinoflagellate sequences observed in
marine samples (Guillou et al., 2008). Clone library
sequences of Antarctic deep water samples north of our
sampling region in austral summer consisted of 65–76%
unclassified Alveolata, many of which are related to the
Syndiniales parasite groups (López-Garcı́a et al., 2001).
Clone libraries in the high arctic also contained high
abundances of Syndiniales, with various groups making
up 44% of the clones (Sørensen et al., 2012). Radiolarians
sampled in the Arctic were associated with alveolates, as
observed here in the Antarctic (Fig. 7) (Bråte et al., 2012).
Similar syndinian-like alveolates have also been observed
near hydrothermal vent systems in both the Atlantic and
Pacific along with more well-known parasites such as
Perkinsus spp. and Pirsonia spp. (Edgcomb et al., 2002;
López-Garcı́a et al., 2003; Moreira and López-Garcı́a,
2003; Takishita et al., 2005). Understanding the relationships and biogeography of these global distributed yet
poorly known parasite groups, such as the Syndiniales,
may prove an interesting area for future research.
The much higher abundances of parasites observed in
winter, when compared with the better studied summer
season, underscore the importance of year-round observations, ideally with fine temporal resolution, in understanding plankton communities. As new observations over
diverse areas of the world ocean and over the seasonal
cycle become available, it will be interesting to see how the
importance of these parasite-associated sequence groups
varies globally and temporally, and begin to understand
the magnitude of their ecological roles more widely.
DATA A RC H I V I N G
DNA sequences of parasite OTUs are available from
GenBank under accession numbers KR864908 –
KR865949.
AC K N OW L E D G E M E N T S
Many thanks to Kerry Whitaker for assistance in winter
seawater sampling, and to Rebecca Robinson for assistance in sediment sampling. Thanks to Maria Casas, Iain
McCoy, Michelle Dennis, Jessica Perreault and David
Gleeson for assistance with summer water sampling.
Thanks to John Kirkpatrick for advice on molecular and
analytical techniques. Thanks also to the science party,
command, and crew of RVIB Nathaniel B Palmer cruises
1304 and 1410. This manuscript was improved by comments from four anonymous reviewers, who we thank.
FUNDING
This research is based in part upon work conducted
using the Rhode Island Genomics and Sequencing
Center, URI, which is supported in part by the National
Science Foundation under EPSCoR Grants Nos.
0554548 and EPS-1004057, and based in part using
resources and services at the Center for Computation
and Visualization, Brown University. This research was
supported by National Science Foundation Office of
Polar Programs grant #ANT-1142107 to E.G.D.
REFERENCES
Altschul, S. F., Gih, W., Miller, W., Myers, E. W. and Lipman, D. J.
(1990) Basic local alignment search tool. J. Mol. Biol., 215, 403–410.
Arias-Gonzáles, J. and Morand, S. (2006) Trophic functioning with
parasites: a new insight for ecosystem analysis. Mar. Ecol. Prog. Ser.,
320, 43–53.
Bråte, J., Krabberød, A. K., Dolven, J. K., Ose, R. F., Kristensen, T.,
Bjørklund, K. R. and Shalchian-Tabrizi, K. (2012) Radiolaria associated with large diversity of marine alveolates. Protist, 163, 767 –777.
Bray, J. R. and Curtis, J. T. (1957) An ordination of upland forest communities of southern Wisconsin. Ecol. Monographs, 27, 325– 349.
Cachon, J. (1964) Contribution a l’étude des Péridiniens parasites.
Cytologie. Cycles évolutifs. Ann. Sci. Nat. Zool., 12, 1– 158.
Caporaso, J. G., Kuczynski, J., Stombaugh, J., Bittinger, K., Bushman,
F. D., Costello, E. K., Fierer, N., Gonzalez-Pena, A. et al. (2010)
QIIME allows analysis of high-throughput community sequencing
data. Nat. Methods, 7, 335– 336.
Chambouvet, A. (2009) Les amoebophryidae (Syndiniales) parasitoı̈des
de dinoflagellés: cycle de vie, dynamique et spécificité in situ. PhD
thesis, University Pierre and Marie Curie, Paris, France.
Chambouvet, A., Alvez-de-Souza, C., Cueffa, V., Mariea, D., Karpova,
S. and Guillou, L. (2011) Interplay between the parasite Amoebophrya
sp. (Alveolata) and the cyst formation of the red tide dinoflagellate
Scripsiella trochoidea. Protist, 162, 637– 649.
Chambouvet, A., Berney, C., Romac, S., Audic, S., Maguire, F., De
Vargas, C. and Richards, T. A. (2014) Diverse molecular signatures
for ribosomally “active” Perkinsea in marine sediments. BMC
Microbiol., 14, 110.
Cleary, A. C. (2015) Distributions and interactions in three groups of
polar marine plankton. PhD dissertation, University of Rhode Island,
Graduate School of Oceanography.
Cleary, A. C., Durbin, E. G., Casas, M. C. and Zhou, M. (in review)
Winter distribution and size structure of Antarctic krill, Euphausia
superba, populations inshore in the West Antarctic Peninsula. Mar.
Ecol. Prog. Ser.
Cleary, A. C., Durbin, E. G., Rynearson, T. A. and Bailey, J. (2015)
Feeding by Pseudocalanus copepods in the Bering Sea: trophic linkages
and a potential mechanism of niche partitioning. Deep-Sea Res, II,
doi:10.1016/j.dsr2.2015.04.001.
415
JOURNAL OF PLANKTON RESEARCH
j
VOLUME 
Coats, D. W. (1999) Parasitic life styles of marine dinoflagellates.
J. Eukaryot. Microbiol., 46, 402– 409.
Coats, D. W. and Park, M. G. (2002) Parasitism of photosynthetic dinoflagellates by three strains of Amoebophrya (dinophyta): parasite survival, infectivity, generation time, and host specificity. J. Phycol., 38,
520–528.
De Vargas, C., Audic, S., Henry, N., Decelle, J., Mahé, F., Logares, R.,
Lara, E., Berney, C. et al. (2015) Eukaryotic plankton diversity in the
sunlit ocean. Science, 348, 1261605.
Diez, B., Massana, R., Estrada, M. and Pedrós-Alió, C. (2004)
Distribution of eukaryotic picoplankton assemblages across hydrographic fronts in the Southern Ocean, studied by denaturing gradient
gel electrophoresis. Limnol. Oceanogr., 49, 1022–1034.
Ducklow, H. W., Baker, K., Martinson, D. G., Quetin, L. B., Ross, R.
M., Smith, R. C., Stammerjohn, S. E., Vernet, M. et al. (2007)
Marine pelagic ecosystems: the west Antarctic Peninsula. Phil. Trans.
Roy. Soc. B: Biol. Sci., 362, 67–94.
j
NUMBER 
j
PAGES –
j

Jespersen, A. M. and Christoffersen, K. (1987) Measurements of
chlorophyll-a from phytoplankton using ethanol as extractions
solvent. Arch. Hydrobiol., 9, 445–454.
Jones, D. L. (2014) Fathom Toolbox for Matlab: Software for Multivariate
Ecological and Oceanographic Data Analysis. College of Marine Science,
University of South Florida, St Petersburg, FL. http://www.marine
.usf.edu/user/djones/.
Jumpstart Consortium Human Microbiome Project Data Generation
Working Group Evaluation of 16S rDNA-Based Community
Profiling for Human Microbiome Research. (2012) PLoS One, 7,
39– 315.
Konovalova, G. V. (2008) Parasitic dinoflagellates and ellobiopsids
(Ellobiopsidae) of the coastal waters of the Sea of Japan. Russ. J.
Protistol., 34, 28– 37.
Edgar, R. C. (2010) Search and clustering orders of magnitude faster
than BLAST. Bioinformatics, 26, 2460– 2461.
Kühn, S., Medlin, L. and Eller, G. (2004) Phylogenetic position of
the parasitoid nanoflagellate Pirsonia inferred from nuclearencoded small subunit ribosomal DNA and a description of
Pseudopirsonia n. gen. and Pseudopirsonia mucosa (Drebes) comb. nov. Protist,
155, 143– 156.
Edgcomb, V. P., Kysela, D. T., Teske, A., de Vera Gomez, A. and Sogin,
M. L. (2002) Benthic eukaryotic diversity in the Guaymas Basin
hydrothermal vent environment. Proc. Natl Acad. Sci. USA, 99,
7658– 7662.
Kuris, A. M., Hechinger, R. F., Shaw, J. C., Whitney, K. L.,
Aguirre-Macedo, L., Boch, C. A., Dobson, A. P., Dunham, A. J. et al.
(2008) Ecosystem energetic implication of parasite and free-living
biomass in three estuaries. Nature, 454, 515 –518.
Egge, E., Bittner, L., Andersen, T., Audic, S., de Vargas, C. and
Edvardsen, B. (2013) 454 Pyrosequencing to describe microbial eukaryotic community composition, diversity and relative abundance: a
test for marine haptophytes. PLoS One, doi:10.1371/journal.pone.0074371.
Leander, B. S. (2007) Marine gregarines: evolutionary prelude to the
apicomplexan
radiation?
Trends
Parasitol.,
doi:10.1016/
j.pt.2007.11.005.
Garrison, D. L. (1991) An overview of the abundance and role of protozooplankton in Antarctic waters. J. Mar. Syst., 2, 317– 331.
Gast, R. J., Dennett, M. R. and Caron, D. A. (2004) Characterization of
protistan assemblages in the Ross Sea, Antarctica, by denaturing gradient gel electrophoresis. Appl. Environ. Microbiol., 70, 2028–2037.
Genitsaris, S., Monchy, S., Viscogliosi, E., Sime-Ngando, T., Ferreira, S.
and Christaki, U. (2015) Seasonal variations of marine protist community structure based on taxon-specific traits using the eastern
English Channel as a model coastal ecosystem. FEMS Microbiol. Ecol.,
91 (doi: 10.1093/femsec/fiv034).
Glockling, S. L., Marshall, W. L. and Gleason, F. H. (2013) Phylogenetic
interpretations and ecological potentials of the Mesomycetozoea
(Ichthyosporea). Fungal Ecol., 6, 237–247.
Guillou, L., Bachar, D., Audic, S., Bass, D., Berney, C., Bittner, L., Boutte,
C., Burgaud, G. et al. (2012) The Protist Ribosomal Reference database
(PR2): a catalog of unicellular eukaryote small sub-unit rRNA
sequences with curated taxonomy. Nucleic Acids Res., 41, 597–604.
Guillou, L., Viprey, M., Chambouvet, A., Welsh, R. M., Kirkham, A.
R., Massana, R., Scanlan, D. J. and Worden, A. Z. (2008)
Widespread occurrence and genetic diversity of marine parasitoids
belonging to Syndiniales (Alveolata). Environ. Microbiol., 10,
3349– 3365.
Harvell, C. D., Mitchell, C. E., Ward, J. R., Altizer, S., Dobson, A. P.,
Ostfeld, R. S. and Samuel, M. D. (2002) Climate warming and disease
risks for terrestrial and marine biota. Science, 296, 2158–2162.
Leander, B. S., Saldarriaga, J. F. and Keeling, P. J. (2002) Surface
morphology of the marine parasite Haplozoon axiothellae Siebert
(Dinoflagellata). Eur. J. Protistol., 38, 287–297.
Lima-Mendez, G., Faust, K., Henry, N., Decelle, J., Colin, S., Carcillo, F.,
Chaffron, S., Ignacio-Espinosa, J. et al. (2015) Determinants of community structure in the global plankton interactome. Science, 348, 1262073.
López-Garcı́a, P., Phillippe, H., Gail, F. and Moreira, D. (2003)
Autochthonous eukaryotic diversity in hydrothermal sediment and
experimental microcolonizers at the Mid-Atlantic Ridge. Proc. Natl
Acad. Sci. USA, 100, 697–702.
López-Garcı́a, P., Rodrı́guez-Valera, F., Pedr?s-Alló, C. and Moreira, D.
(2001) Unexpected diversity of small eukaryotes in deep-sea Antarctic
plankton. Nature, 409, 603– 607.
Luria, C. M., Ducklow, H. W. and Amaral-Zettler, L. A. (2014) Marine
bacterial, archaeal, and eukaryotic diversity and community structure
on the continental shelf of the western Antarctic Peninsula. Aquat.
Microb. Ecol., 73, 107–121.
Melbourne-Thomas, J., Constable, A., Wotherspoon, S. and Raymond,
B. (2013) Testing paradigms of ecosystem change under climate
warming in Antarctica. PLoS One, doi:10.1371/journal.pone.0055093.
Moreira, D. and López-Garcı́a, P. (2003) Are hydrothermal vents oases
for parasitic protists? Trends Parasitol., 19, 556–558.
Morgulis, A., Coulouris, G., Raytselis, Y., Madden, T. L., Agarwala, R.
and Schäffer, A. A. (2008) Database indexing for production
MegaBLAST searches. Bioinformatics, 15, 1757– 1764.
Hudson, P. J., Dobson, A. P. and Lafferty, K. D. (2006) Is a healthy ecosystem one that is rich in parasites? Trends Ecol. Evol., 21, 381–385.
Neuhauser, S., Kirchmair, M. and Gleason, F. H. (2011) Ecological
roles of the parasitic phytomyxids ( plasmodiophorids) in marine ecosystems—a review. Mar. Freshwater Res., 62, 365–371.
Huntley, M. E., Lopez, M. D. G. and Karl, D. M. (1991) Top predators
in the Southern Ocean: a major leak in the biological carbon pump.
Science, 253, 64–66.
Not, F., del Campo, J., Balagué, V., de Vargas, C. and Massana, R.
(2009) New insights into the diversity of marine picoeukaryotes. PLoS
One, 4, e7143.
416
A. C. CLEARY AND E. G. DURBIN
j
ABUNDANT PARASITE 18S IN ANTARCTIC WINTER
Quast, C., Pruesse, E., Yilmaz, P., Gerken, J., Schweer, T., Yarza, P.,
Peplies, J. and Glöckner, F. O. (2013) The SILVA ribosomal RNA
gene database project: improved data processing and web-based
tools. Nucleic Acids. Res., 41, 590– 596.
Takahashi, K. T., Kawaguchi, S., Kobayashi, M., Toda, T., Tanimura,
A., Fukuchi, M. and Odate, T. (2011) Eugregarine infection within
the digestive tract of larval Antarctic krill, Euphausia superba. Polar Biol.,
34, 1167–1174.
Rhode, K. (1984) Zoogeography of marine parasites. Helgol. Meeresunters,
37, 35–52.
Takishita, K., Miyake, H., Kawato, M. and Maruyama, T. (2005)
Genetic diversity of microbial eukaryotes in anoxic sediment around
fumaroles on a submarine caldera floor based on the small-subunit
rDNA phylogeny. Extremophiles, 9, 185– 196.
Sailley, S. F., Ducklow, H. W., Moeller, H. V., Fraser, W. R., Schofield,
O. M., Steinberg, D. K., Garzio, L. M. and Doney, S. C. (2013)
Carbon fluxes and pelagic ecosystem dynamics near two western
Antarctic Peninsula Adélie penguin colonies: an inverse model approach. Mar. Ecol. Prog. Ser., 492, 253– 272.
Skovgaard, A. (2014) Dirty tricks in the plankton: diversity and role of
marine parasitic protists. Acta Protozool., 53, 51–62.
Skovgaard, A. and Daugbjerg, N. (2008) Identity and systematic position of Paradinium poucheti and other Paradinium-like parasites of
marine copepods based on morphology and nuclear-encoded SSU
rDNA. Protist, 159, 401– 413.
Sogin, M. L., Morrison, H. G., Huber, J. A., Welch, M. D., Huse, S. M.,
Neal, P. R., Arrieta, J. M. and Herndl, G. J. (2006) Microbial diversity
in the deep sea and the underexplored “rare biosphere”. Proc. Natl
Acad. Sci. USA, 103, 12115–12120.
Sørensen, N., Daugbjerg, N. and Gabrielsen, T. M. (2012) Molecular
diversity and temporal variation of picoeukaryotes in two Arctic
fjords, Svalbard. Polar Biol., 35, 519–533.
Sousa, W. P. (1991) Can models of soft-sediment community structure
be complete without parasites? Am. Zool., 31, 821–830.
Thomson, S. (2014) Seasonal abundance of parasitic Marine Alveolate
group II (MALV II) in an Arctic fjord, Svalbard. MS thesis,
University of Tromsø, Department of Arctic and Marine Biology.
Tillmann, U., Hesse, K. J. and Tillmann, A. (1999) Large-scale parasitic
infection of diatoms in the Northfrisian Wadden Sea. J. Sea Res., 42,
255 –261.
Toft, C. A. (1986) Communities of species with parasitic life-styles. In
Diamond, J. and Case, T. J. (eds), Community Ecology. Harper & Row,
New York, pp. 445 –463.
Wang, Q., Garrity, G. M., Tiedje, J. M. and Cole, J. R. (2007) Naive
Bayesian classifier for rapid assignment of rRNA sequences into the
new bacterial taxonomy. Appl. Environ. Microb., 73, 5261–5267.
Wright, A. D. G. and Colorni, A. (2002) Taxonomic re-assignment of
Crytocaryon irritans, a marine fish parasite. Eur. J. Protistol., 37,
375 –378.
Zhu, F., Massana, R., Not, F., Marie, D. and Vaulot, D. (2005) Mapping
of picoeukaryotes in marine ecosystems with quantitative PCR of the
18S rRNA gene. FEMS Microbiol., 52, 79–92.
417