JOURNAL OF PLANKTON RESEARCH
j
VOLUME
34
j
NUMBER
1
j
PAGES
72 – 82
j
2012
Copepod feeding and digestion rates
using prey DNA and qPCR
EDWARD G. DURBIN*, MARIA C. CASAS AND TATIANA A. RYNEARSON
UNIVERSITY OF RHODE ISLAND, GRADUATE SCHOOL OF OCEANOGRAPHY, SOUTH FERRY ROAD, NARRAGANSETT, RI
02882, USA
*CORRESPONDING AUTHOR: edurbin@gso.uri.edu
Received December 15, 2010; accepted in principle August 30, 2011; accepted for publication September 2, 2011
Corresponding editor: Roger Harris
Copepod feeding and digestion rates were measured using quantitative polymerase
chain reaction (qPCR) to amplify the prey Thalassiosira weissflogii and Heterocapsa triquetra from the guts of Acartia tonsa. Using species-specific primers, prey 18S rDNA
could be detected routinely and quantified in the guts and fecal pellets of A. tonsa.
Recovery of gut contents DNA using two fixation methods was compared. Prey
18S copy numbers were .10-fold higher in copepods fixed in 95% ethanol
(3260 + 822 copies copepod21) compared with anesthetized and frozen copepods
(210 + 19 copies copepod21). Experiments using 95% ethanol fixation showed
rapid prey DNA digestion rates during the initial 2 min after ingestion (0.7 min21)
after which they slowed 10-fold. Chlorophyll pigment disappearance rates were
slower (0.015 min21). Rates of gut filling measured by DNA and gut pigments
differed, reaching 95% of the asymptote, Imax, in 3 and 58 min, respectively, likely
reflecting differences in rates at which biomarkers were digested. Gut fullness
measured by DNA increased with prey concentration, reaching Imax at
9760 copies copepod21 and a critical concentration (Icrit) at 1530 cells mL21.
These results demonstrate that qPCR analysis of prey DNA in copepod guts can
be used to provide a quantitative index of feeding rates.
KEYWORDS: Acartia tonsa; Thalassiosira weissflogii; Heterocapsa triquetra
I N T RO D U C T I O N
The predatory role of mesozooplankton in marine food
webs can only be understood through the knowledge of
their in situ feeding behavior. Gaining this information
has been a major challenge in biological oceanography.
Approaches used include bottle incubation with natural
plankton, extrapolation of laboratory-derived functional
curves to field samples and stomach content analyses of
copepods collected in the field (reviewed in Bamstedt
et al., 2000). However, the first two methods do not
provide a direct in situ measurement of consumption
and are subject to various artifacts arising from lengthy
bottle incubations and the difficulty of ensuring that a
representative prey field is collected. The third approach
requires information on stomach contents in the field at
intervals over 24 h and a measure of how rapidly food
is being digested in the stomachs, i.e. food disappearance rates (Durbin and Campbell, 2007). This latter
approach has been applied to copepods using chlorophyll pigments as tracers of gut contents (Mackas and
Bohrer, 1976). Although this approach has been used
extensively, it is limited to herbivory. HPLC has been
used to discriminate among phytoplankton taxa using
pigments or other biomarkers (e.g. Kleppel and Pieper,
1984; Kleppel et al., 1988; Head and Harris, 1996; Juhl
et al., 1996), while stable isotopes, particularly 15N, may
be used to provide information on the trophic level at
which the predator is feeding. An approach to estimate
the in situ carnivory of individual prey species has been
to use immunological analysis of gut contents (Ohman,
1992). More recently, molecular techniques have been
doi:10.1093/plankt/fbr082, available online at www.plankt.oxfordjournals.org. Advance Access publication October 18, 2011
# The Author 2011. Published by Oxford University Press. All rights reserved. For permissions, please email: journals.permissions@oup.com
E. G. DURBIN ET AL.
j
COPEPOD FEEDING
diatom Thalassiosira weissflogii (CCMP1336) and the dinoflagellate Heterocapsa triquetra (CCMP448). These cultures
were grown in f/2 media (Guillard, 1975), with (T. weissflogii) or without Si (Rhodomonas sp., H. triquetra). Both
A. tonsa and phytoplankton prey were raised at 188C. All
experiments were carried out at 208C.
Phytoplankton cultures were harvested in a midexponential phase of growth for determination of cell
carbon, nitrogen, chlorophyll and 18S rDNA gene
sequence and copy number per cell. Cells of each
species were preserved in Lugol’s iodine for later enumeration. Samples (15 mL) for cell carbon and nitrogen
were filtered onto 13 mm GF/A pre-combusted glass
fiber filters, dried at 608C and analyzed with a NA1500
CHN Analyzer (Carlo Erba). Samples (3 mL) for chlorophyll a (Chl a) were filtered onto 25 mm diam GF/F
glass fiber filters. These were placed immediately in
13 mm diameter glass tubes with 6-mL 90% acetone,
incubated at – 208C for 24 h, read on a fluorometer
(Turner Designs Model 10) before and after acidification and Chl a and pheopigments calculated (Parsons
et al., 1984). Four to five replicate filters were prepared
for each different analysis.
developed using polymerase chain reaction (PCR) to
identify prey DNA in the guts of both terrestrial and
marine predators (e.g. Jarman et al., 2002; Symondson,
2002; Blankenship and Yayanos, 2005; Pons, 2006;
Read et al., 2006). This approach has been used both in
the laboratory and in the field together with speciesspecific primers to detect phytoplankton prey DNA in
the guts and fecal pellets of copepods (Nejstgaard et al.,
2003; Vestheim et al., 2005; Haley et al., 2010). Although
these studies demonstrated the feasibility of using endpoint PCR and species-specific primers to detect prey,
they were not quantitative.
To quantify prey-specific DNA in zooplankton guts,
quantitative PCR (qPCR) has been used (e.g. Passmore
et al., 2006; Troedsson et al., 2007; Durbin et al., 2008;
Nejstgaard et al., 2008; Barofsky et al., 2010). This
approach has the advantage of providing information
on the amount of prey DNA present within zooplankton guts. However, both Nejstgaard et al. (Nejstgaard
et al., 2008) and Durbin et al. (Durbin et al., 2008) noted
that the amount of DNA detected with qPCR was
lower than expected. One possible explanation is that
digestion of DNA is very rapid, permitting prey DNA
to be partially digested during handling (Simonelli et al.,
2009). These possibilities would result in an underestimation of stomach contents that in turn would affect
the estimates of feeding rates.
Below, we describe experiments in which we use
quantitative measures of prey DNA in the guts of copepods to determine the rate of gut filling, the effect of
food concentration on stomach contents and the shortterm digestion rate of prey DNA. Our results demonstrate that qPCR analysis of prey DNA in copepod guts
can be used to provide a quantitative index of zooplankton feeding rates.
Phytoplankton genomic DNA extraction
To obtain T. weissflogii genomic DNA for 18S rDNA
sequencing and determination of cell 18S copy
numbers, 2 mL of culture (107 000 cells) were filtered
onto 13 mm diam, 3 mm pore size filters (Nucleopore),
frozen at 2808C and processed using the DNeasy Plant
Kit (Qiagen). Initial attempts to obtain genomic DNA
from H. triquetra using this method gave low and variable copy numbers cell21. Three methods to obtain
genomic DNA from H. triquetra were subsequently compared. Four 2-mL aliquots of culture (22 500 cells) were
centrifuged at 4000g for 5 min, the supernatant aspirated and the pellet frozen at 2808C. Genomic DNA
was extracted using the DNeasy Blood & Tissue Kit
(Qiagen) after vortexing (3200 rpm and 1 min) with
lysis buffer (DNeasy buffer ATL with proteinase-K) and
three to five 425 – 600 mm acid washed glass beads
(Sigma). This method was compared with extraction of
DNA from pelleted cells using the DNeasy Plant Kit
(Qiagen) with and without glass beads. DNA from these
extractions was quantified using species-specific primers
and qPCR as described below. Genomic DNA from the
former method was used for 18S rDNA sequencing.
METHODS
Culturing methods
Acartia tonsa adult females were used as the predator in all
experiments. A culture was established by isolating adult
females collected from the pier at the University of
Rhode Island, Graduate School of Oceanography in
Narragansett Bay, RI (41829.50 N, 71825.10 W). Eggs from
these females were collected for initiation of the culture.
Copepods were maintained in 20-L polycarbonate containers and fed Rhodomonas sp. (CCMP768). When larger
numbers of copepods were required for experiments,
groups of eggs were collected over 1 or 2 days and a
cohort of was reared in 120-L containers. Prey used in
the feeding and digestion rate experiments were the
Copepod fixation and DNA extraction
We compared the efficacy of two copepod fixation
methods in recovering ingested prey DNA; 95% ethanol
73
JOURNAL OF PLANKTON RESEARCH
j
VOLUME
and freezing. Acartia tonsa were fed T. weissflogii at a concentration of 10 000 cell mL21 for 2 h. A portion of
these copepods was fixed immediately by collecting
them on a sieve and immediately placing this in a
beaker containing 95% ethanol. These were rinsed
several times with clean ethanol to remove any phytoplankton that may have been present and then transferred to 1.5-mL microcentrifuge tubes with ethanol for
later DNA extraction. Prior to DNA extraction, these
copepods were sorted in four groups of eight into clean
microcentrifuge tubes. The other group was anesthetized in MS-222 (0.5 g L21 SW; Durbin and Durbin,
1978), rinsed in filtered seawater and sorted quickly in
four groups of eight into microcentrifuge tubes and
frozen at 2808C. Copepods processed using each
method were ground in the microcentrifuge tubes with
a disposable sterile pellet pestle (Kimble Chase Kontes)
and genomic DNA purified using the DNeasy Blood &
Tissue Kit (Qiagen). The eluted volume was then concentrated by a factor of 10 (Thermo Savant DNA110
SpeedVac) to increase our ability to quantify low concentrations of prey DNA.
34
j
NUMBER
1
j
PAGES
72 – 82
j
2012
Table I: Primers used in this study
Primer
Sequence
Thalassiosira weissflogii qPCR F 50 -ACC GGT CTC ACA CCC TGT-30
Thalassiosira weissflogii qPCR R 50 -TGC CCG ACA ACT GAA TGC CAA
ACA-30
Heterocapsa triquetra qPCR F
50 -ATG GCC GTT CTT AGT TGG TGG
AGT-30
Heterocapsa triquetra qPCR R
50 -TAA GAA GTA GCC CAC GTA ACC
GAG-30
Methods used to quantify the 18S copy number in
samples were similar to those described in Durbin et al.
(Durbin et al., 2008). Standards for qPCRs were
prepared from end-point PCR amplicons for both
T. weissflogii and H. triquetra. The copy number mL21 in
both standards was calculated from the amplicon concentration (quantified using a NanoDrop ND-1000) and
length. Standard curves were prepared in a 10-fold serial
dilution (5 107, 5 106, 5 105, 5 104, 5 103,
5 102 and 50 copies per reaction). The reaction
volume for qPCR was 25 mL and contained 50 ng of
template, 1 SYBRGreen Master Mix (Stratagene),
250 nM of each primer (Univ18S-550F and R) and
0.03 mM ROX Reference Dye (Stratagene). For each run
of unknowns, we carried out a duplicate standard curve
together with a no template control. Amplifications were
performed on an Mx3005P (Stratagene) and consisted of
10 min at 958C, followed by 40 cycles of 30 s at 958C,
1 min at 578C and 30 s at 728C. Melting curve analysis
was used to verify amplicon purity. Copy numbers were
determined from the Ct value (the fractional PCR cycle
number where the fluorescence signal crosses a certain
threshold) for each prey species, template concentration
and the standard curve. Amplification efficiency
was calculated from: PCR efficiency ¼ 10(21/slope) 2 1.
In the experiments reported here, all but one (84.4%)
were .96%.
18S rDNA sequencing
For 18S rDNA sequencing of A. tonsa, genomic DNA
copepods were starved for 24 h and fixed in 95%
ethanol. Individual copepods were then placed in clean
microcentrifuge tubes and DNA extracted as described
above. An 1750-bp segment of the 18S rDNA gene
of each prey species and of A. tonsa was amplified in
reaction mixtures containing 1 Taq-PCR Master Mix
(Qiagen), 0.5 mM of primers 18SA and 18SB (Medlin
et al., 1988) and 3.0 mL (10 – 15 ng) template of genomic
DNA. Amplification was conducted in a Mastercycler
Epigradient S thermal cycler (Eppendorf ) with the
following thermal regime: 958C for 30 s (1); 948C
for 30 s, 608C for 60 s and 728C for 2 min (35); 728C
for 10 min (1). Amplicons were cleaned using the
QIAquick PCR Purification Kit (Qiagen). Sequences
were obtained using Big Dye terminators and primers
18SA and 18SB and analyzed on a 3130xl (ABI).
Prey DNA and chlorophyll disappearance
rates in copepod guts
Experiments were carried out to measure the short-term
disappearance rates of prey (T. weissflogii and H. triquetra)
DNA and Chl a pigments in copepod guts. For each
experiment, copepods were anesthetized with MS-222
(0.5 g L21 SW) and sorted in groups of 30 into 0.15 mm
sieves (40 mm i.d.) sitting in 100-mL glass beakers to
which 60 mL of filtered seawater had been added.
Beakers (13 to 17) were prepared providing 1 control
each (no food) for gut pigments and DNA and 12 and
16 experimentals (8 – 11 for DNA and 3 – 5 for gut pigments). After 24 h, food was added to each
Design of species-specific qPCR primers
and methods
Species-specific qPCR primers for T. weissflogii and
H. triquetra were designed with the aid of PrimerQuest
(IDT) and are listed in Table I. These primer sets were
tested with genomic DNA from each species by qPCR
to confirm specificity and were subsequently used to
quantify prey DNA copy numbers in all experiments
using procedures described below.
74
E. G. DURBIN ET AL.
j
COPEPOD FEEDING
added to the remaining beakers to provide a final concentration of 10 000 cells mL21. At 2.5, 5, 10 15, 20,
25 and 30 min, the groups of copepods were fixed with
95% ethanol, while other groups were anesthetized and
processed for gut pigment determination as described
above.
Another experiment was carried out to determine the
relationship between food concentration and gut fullness as measured in terms of copy numbers. The experimental design was similar to that described above.
After the 24 h food-deprivation period, T. weissflogii was
added to eight beakers in final concentrations of 0, 200,
400, 800, 1200, 2000, 5000 and 10 000 cells mL21.
The copepods were allowed to feed for 5 min and then
sieves immediately transferred to 95% ethanol. 18S
copy numbers were determined as described above.
experimental beaker to provide a final concentration of
either 8000 cells mL21 (T. weissflogii) or 3000 cells mL21
(H. triquetra). Prior to each experiment, the appropriate
volume of food was dispensed into 13 mm diam glass
tubes to speed the addition of food to beakers containing
copepods. This food was added rapidly and after a brief
feeding period (2 – 60 min depending on experiment),
one of the sieves containing copepods was transferred to
a beaker containing 95% ethanol, while the copepods in
another were rinsed with MS-222 into a Petri dish and
sorted in three groups of 8 25 mm diam GF/A glass
fiber filters for initial gut pigment determination. The
remaining sieves were immediately transferred to
100-mL beakers containing 60 mL of filtered seawater.
A sieve was transferred from the filtered seawater to 95%
ethanol at either 30 s (T. weissflogii) or 20 s (H. triquetra)
intervals for 2 min, and then longer intervals for the
next 20 – 60 min. At 5, 10, 20 and 30 min, an additional
sieve in filtered seawater was rinsed with MS-222 into a
Petri dish and sorted for gut pigments. Copepods
without food were also either fixed in 95% ethanol or
processed for background gut pigment measurements.
Chl a and pheogut pigments (as Chl a equivalents)
were determined using the methods described above
for phytoplankton chlorophyll and expressed as ng
pigment copepod21 (Durbin et al., 1995). Copepods in
95% ethanol in beakers were rinsed several times with
clean ethanol to remove any phytoplankton that may
have been present and then transferred to 1.5-mL microcentrifuge tubes with ethanol for later DNA extraction.
DNA was extracted (three groups of eight copepods;
DNeasy Blood & Tissue Kit, Qiagen), and the 18S gene
copy number determined through qPCR using the
methods described above. While it was not logistically
possible to replicate treatments within an experiment, by
dividing copepods into 3 groups of 8 for analysis
( pseudo replication), chances of loss of a whole sample
during processing were minimized. However, each shortterm disappearance rate experiment was repeated two
(H. triquetra) or three (T. weissflogii) times.
The negative exponential curves describing the rate of
gut filling and the effect of food concentration on gut
fullness were fitted using Proc NLIN (SAS 9.2). Similar
procedures were used to fit the exponential decline in
gut fullness as measured by gut pigments in all experiments and for DNA in the 60-min feeding experiment.
Data for the short-term feeding experiments were analyzed using a segmented model with two exponential
curves and a breakpoint (C) in SAS (V9.2). For the left
line Y1 ¼ A1 exp(2B1 X) for X C, while for the
right line Y2 ¼ A2 exp(2B2 X) for X C. For the
left and right regression to be continuous at point C,
Y1 ¼ Y2, allowing B2 to be calculated.
Gut filling rate and food concentration
experiments
R E S U LT S
Fecal pellet copy numbers
Fecal pellets produced by A. tonsa females after being
fed T. weissflogii at 10 000 cells mL21 for 30 min were
placed in groups of 16– 23 onto 13 mm 3 mm filters
(Nucleopore) and frozen in microcentrifuge tubes. 18S
copy numbers were determined as described above.
Statistical analysis
An experiment was carried out to determine the rate of
gut filling as measured by the 18S copy number and by
gut pigments. In this experiment, T. weissflogii was used
as prey. Groups of 30 copepods were placed in filtered
seawater in sieves as described above and left for 24 h.
One group was transferred to 95% ethanol to provide
an initial no food DNA gut content measure while
another was processed for background gut pigment
measurements. Thalassiosira weissflogii culture was then
Comparison of fixation and DNA extraction
methods
We compared the number of 18S copies obtained using
different fixation and extraction methods. Prey 18S
copy numbers were over 10-fold higher in copepods
that had been fixed in 95% ethanol (3260 + 822 18S
copies copepod21) compared with those that had
been anesthetized, sorted and frozen (210 + 19 18S
75
JOURNAL OF PLANKTON RESEARCH
j
VOLUME
copies copepod21). This indicated that prey DNA was
lost during sorting prior to, and/or after, freezing. In all
subsequent experiments, copepods were fixed in 95%
ethanol.
Extraction of H. triquetra DNA using the DNeasy
Blood & Tissue Kit (Qiagen) with disruption using
glass beads yielded 1690 + 405 18S copies cell21.
Considerably fewer 18S copy numbers cell21 were
recovered using the DNeasy Plant Kit (Qiagen) (430 +
282 18S copies cell21 with beads and 303 + 106 18S
copies cell21 without beads). All subsequent extractions
of H. triquetra DNA for 18S sequencing and copy
number determination were performed using the
DNeasy Blood & Tissue Kit (Qiagen) with disruption
using glass beads.
34
j
NUMBER
1
j
PAGES
72 – 82
j
2012
equivalent to 14.7 cells for a copy number cell21 of 760
(Fig. 1). The rate of increase was very rapid and
reached 95% of the maximum, Imax, in 3 min. In
contrast, gut pigments increased more slowly reaching
95% of the Imax (0.97 ng Chl a equiv. copepod21,
equivalent to 324 cells copepod21) in 58 min (Fig. 1).
Gut fullness also increased rapidly with concentration
of T. weissflogii reaching an asymptote (Imax) of
9580 copies copepod21, equivalent to 12.6 cells (Fig. 2).
The critical concentration (Icrit), 90% of this
asymptotic value, was reached at a concentration of
1530 cells mL21.
Prey DNA and prey chlorophyll
disappearance rates in the guts of copepods
Disappearance of prey DNA in A. tonsa that were fed
for short periods (2 – 6 min) showed a rapid decline for
2 min after removal from food (Figs 3 and 4). After
this initial rapid decline, prey DNA declined more
slowly and was still present at 30 min when each experiment was terminated. This pattern of decline was best
fit by a segmented regression providing a slope for each
phase of decline. The initial slopes (B1) varied between
0.30 and 1.17 min21 (Table III) with no significant
difference (t-test, P ¼ 0.73) between prey species.
Phytoplankton cell characteristics
Heterocapsa triquetra contained between 4- and 8-fold
more carbon, nitrogen and chlorophyll cell21 than
T. weissflogii (Table II). Differences in 18S copy
numbers cell21 were much less (1690 and
760 copies cell21, respectively). The ratios of 18S copy
numbers to cell carbon were 9.83 and 2.52 copies pg
carbon21 for T. weissflogii and H. triquetra. The 18S
sequence obtained for H. triquetra was identical to that
already in GenBank (accession numbers AB183670 and
AF022198) while those for T. weissflogii and A. tonsa were
submitted (accession numbers GU594638 and
FJ422281).
Gut filling rate and effects of food
concentration on gut contents
Food-deprived adult female A. tonsa were offered
T. weissflogii cells at 10 000 cell mL21 as prey for differing time periods. Gut contents (as prey 18S copy
numbers) increased with feeding time reaching an
asymptote (Imax) of 11 190 18S copies copepod21,
Table II: Average and standard deviations of
carbon, nitrogen, Chl a and 18S copies per
cell of T. weissflogii and H. triquetra fed
to A. tonsa in feeding and digestion rate
experiments
Thalassiosira
weissflogii
Heterocapsa
triquetra
Carbon
(pg cell21)
Nitrogen
(pg cell21)
Chl a
(pg cell21)
18S copies
(cell21)
77.3 + 5.4
17.5 + 0.85
2.99 + 0.35
760 + 9.8
671 + 37
134 + 6.8
11.4 + 0.54
1690 + 414
Fig. 1. Gut filling rates for adult female A. tonsa. The copepods were
fed T. weissflogii at 10 000 cell mL21 for different times before fixing.
(A) Gut contents as 18S copies copepod21 and (B) gut contents as gut
pigments (ng Chl a equiv. cop.21). The equations for a negative
exponential fit to the data are shown.
76
E. G. DURBIN ET AL.
j
COPEPOD FEEDING
Measurement of phytoplankton DNA
in fecal pellets of copepods
We were routinely able to quantitatively measure the
amount of phytoplankton prey DNA (as 18S gene copy
numbers) in fecal pellets of adult female A. tonsa. Fecal
pellets collected from copepods that had been fed
T. weissflogii at 8000 cells mL21 contained an average of
625 + 160 18S copies pellet21.
DISCUSSION
Digestion rates of prey DNA by copepods
Our study confirms previous suggestions that prey DNA
is digested very rapidly (Durbin et al., 2008; Nejstgaard
et al., 2008). In experiments where copepods were
deprived of food then fed for a brief period, we
observed an initial, very rapid, disappearance during
the first few minutes after ingestion. This was followed
by a slower phase in which DNA disappears at 1/10
of the initial rapid rate. Because of this clear break in
the data, we fit a segmented model rather than a single
exponential curve as has been more traditionally used
for evacuation rate data. We suggest that these two
phases reflect two quite different processes controlling
the decline in DNA in the guts of copepods. First, there
is a very rapid enzyme-mediated decline in DNA after
maceration of the food and exposure to digestive
enzymes in the gut. Second, there is a slower decline in
DNA resulting from possible lower rates of enzymatic
degradation after incorporation of food into fecal pellets
inside the gut. However, the measurable quantities of
DNA in the feces indicate that the DNA was not totally
broken down in the gut and that this slower phase also
represents fecal pellet release.
The very rapid initial decline in DNA is a result of
the very efficient system copepods have for macerating
food and exposing it to digestive enzymes as they ingest
it with mandibles that have a spined (toothed) distal
edge with abrasive tips of silica (Miller et al., 1980).
After maceration, food passes through the copepod alimentary canal where cellular contents are exposed to
non-specific nucleases, or DNases, that cleave the DNA
strands (Linn and Roberts, 1982; Linn et al., 1993).
Similar to most other enzyme-catalyzed reactions, these
nucleases act very rapidly, cleaving nucleic acids internally (endonucleases), chewing in from the ends (exonucleases) or attacking in both of these modes. Reaction
rate constants for reactions catalyzed by these DNases
were between 0.03 and 0.05 s21 (Light et al., 1974). By
comparison, the DNA disappearance rates that we
Fig. 2. (A) Gut fullness in terms of 18S copies of T. weissflogii. The
copepods were fed T. weissflogii at different concentrations for 5 min
before fixing. The critical concentration, Icrit, the cell concentration at
which the ingestion rate was 90% of the maximum ingestion rate,
Imax, was 1590 cell mL21. (B) Ingestion of adult female A. tonsa
measured in terms of T. weissflogii cells. The copepods were fed
T. weissflogii at different concentrations for 24 h and ingestion rates
determined from changes in cell concentrations in experimental and
control jars. The critical concentration, the cell concentration at
which the ingestion rate was 90% of the maximum ingestion rate,
Imax, was 943 cell mL21 (data from Thompson et al., 1994).
During the second phase the slopes (B2) varied between
0.016 and 0.18 min21, again with no significant difference (t-test, P ¼ 0.89) between prey species (Table III).
In two experiments, one each with T. weissflogii and
H. triquetra, the copepods were fed for 60 min.
Compared with the short-term feeding experiments,
prey DNA disappearance rates were more variable with
a lower initial slope and a longer period (5 min)
during which prey DNA remained relatively high
before declining to low levels (Fig. 5). A single exponential curve provided the best fit to the data.
Gut pigment disappearance rates were measured over
longer time intervals between samples because of the
time required to sort them after anesthesia. However,
there was no indication of a rapid decline between 0
and 5 min in contrast with the rapid decline in prey
DNA (Figs 3 – 5). While rates of decline in the shortterm feeding experiments were approximately twice
as fast when T. weissflogii was the prey compared with
H. triquetra (0.025 and 0.012 min21 respectively,
Table IV), these differences were not significant (t-test,
P ¼ 0.07).
77
JOURNAL OF PLANKTON RESEARCH
j
VOLUME
34
j
NUMBER
1
j
PAGES
72 – 82
j
2012
Fig. 3. 18S copy number and gut pigment disappearance rates for adult female A. tonsa fed T. weissflogii. The copepods were fed for 2 (A and
B), 4 (C and D) and 6 min (E and F) and placed in filtered SW for differing times (0– 30 min) before fixing. The decline in 18S copy numbers
was biphasic and segmented regressions were fit to the data. A single exponential curve was fit to the decline in gut pigments.
Fig. 4. 18S copy number and gut pigment disappearance rates for adult female A. tonsa fed H. triquetra in two different experiments (A–D). The
copepods were fed for 5 min in each experiment and placed in filtered SW for differing times (0– 30 min) before fixing. The decline in 18S copy
numbers was biphasic and segmented regressions were fit to the data. A single exponential curve was fit to the decline in gut pigments.
observed were a little slower (0.85 min21 or 0.014 s21
for T. weissflogii and 0.011 s21 for H. triquetra) but still
have the same order for an enzyme-catalyzed reaction.
The slower disappearance rates we observed may have
resulted from incomplete mixing of reactants.
The copepod alimentary canal has a mid-gut that is
divided into three regions; Region 1 extending from
the esophagus to approximately the posterior end of the
cephalosome, Region 2 behind this to the first or the
second metasomal segment and Region 3 from there to
78
E. G. DURBIN ET AL.
j
COPEPOD FEEDING
Table III: Exponential disappearance rates of 18S prey DNA for adult female A. tonsa fed
T. weissflogii and H. triquetra
Feeding time (min)
Thalassiosira weissflogii
2
4
6
Mean
60
Heterocapsa triquetra
2
2
Mean
60
A1
B1
4973 (4093 –5852)
11 765 (6068 –17 462)
12 800 (10 060 –15 540)
1.17 (0.67 – 1.67)
1.01 (20.24 –2.26)
0.38 (0.099 –0.14)
0.853
0.11
1.61 (0.55 –2.68)
1.53 (21.9 –4.95)
3.36 (0.073 –6.66)
2.16
1.055 (0.66 – 1.45)
0.300 (0.028 –0.57)
0.68
0.118
1.912 (0.079 –3.74)
6.59 (27.5 –20.7)
4.25
15 520
17 820 (14 683 –20 950)
11 600 (8253 –14 950)
4470
c
B2
0.039 (20.066 –0.145)
0.13 (20.6 – 0.86)
0.088 (20.8 – 0.25)
0.085
0.184 (20.46 –0.82)
0.0164 (20.25 –0.28)
0.10
For the experiments where the copepods were fed for a short-time (2 –6 min) segmented regressions were fit to the data. A1 is the y-intercept and
B1 the slope for the initial exponential decline, c the time at which the breakpoint occurs and B2 the slope of the second period of exponential
decline. Approximate 95% are given. For the experiments where the copepods were fed for 60 min a segmented regression did not provide a good
confidence intervals fit to the data and a single exponential curve was fit for the whole period of measurement (0 – 30 min).
Fig. 5. 18S copy number and gut pigment disappearance rates for adult female A. tonsa fed H. triquetra (A and B) and T weissflogii (C and D) for
60 min and then placed in filtered SW for differing times (0–30 min) before fixing. A single exponential curve is fit to the data.
the hindgut in the posterior part of the uresome
Table IV: Exponential disappearance rates of (Mauchline, 1998). DNA digestion takes place in
chlorophyll pigments in adult female
Regions 1 and 2. Extracellular digestion probably takes
Feeding time (min)
Food: T. weissflogii
2
4
6
Mean
60
Food: H. triquetra
5
5
Mean
60
Obs. time (min)
Exp. slope (min21)
R2
0–30
0–30
0–10
0.018
0.030
0.026
0.025 + 0.001
0.033
0.62
0.44
0.15
0–30
0–30
0–30
0–30
0.011
0.013
0.012 + 0.006
0.010
place in Region 1 through secretion of digestive
enzymes by F-cells lining the gut in this region (Arnaud
et al., 1980), while intracellular digestion appears to
occur in Region 2 through absorption into B-cells lining
the gut where material is digested (Nott et al., 1985).
These B-cells subsequently burst releasing contents into
the gut. There appears to be cyclical changes in the
presence of B-cells in the gut wall between feeding and
non-feeding periods (Nott et al., 1985), and we suggest
that this may affect feeding activity of individual
copepods.
0.28
0.98
0.54
0.99
Acartia tonsa fed T. weissflogii and H. triquetra.
79
JOURNAL OF PLANKTON RESEARCH
j
VOLUME
34
j
NUMBER
1
j
PAGES
72 – 82
j
2012
stage prior to feeding with unutilized B-cells. The initial
rapid digestion may reflect the synchronized utilization
of all of these cells. During longer feeding periods the
copepods may lose this synchrony and be in differing
stages of this cycle. This could reduce the availability of
digestive enzymes resulting in lower digestion rates.
We hypothesize that rapid digestion terminates when
digested material is incorporated into fecal pellets in
Region 3. The DNA disappearance rate then decreases
to about one-tenth of the initial rate, representing both
a low rate of continued DNA digestion and fecal pellet
release. In copepods, fecal pellets are lined with a peritrophic membrane that binds fecal material together
(Gauld, 1957; Brunet et al., 1994). During this process,
undigested DNA may become protected from enzyme
activity. However, even with potential enzyme activity
during the second phase, DNA remains in fecal pellets
after release by copepods (Nejstgaard et al., 2003;
Vestheim et al., 2005; this study). In our experiments,
fecal pellet release began between 5 and 10 min after
feeding terminated consistent with this explanation.
In contrast to the rapid decline in DNA, the disappearance rate of chlorophyll pigments in copepod guts
during the same time frame was slower by a factor of
3 –8 and was best described by a single exponential
curve. We suggest that these differences reflect the role
of two different processes controlling these rates; a
decrease in chlorophyll pigments following feeding
solely as a result of defecation (Durbin and Campbell,
2007), whereas most of the DNA is degraded enzymatically via DNases followed by the release of the remaining amount through defecation, Interestingly, there have
been suggestions that chlorophyll may also be very
rapidly degraded to colorless compounds ( pigment
destruction; Head and Harris, 1996). If so, there should
be an initial rapid disappearance similar to that which
we observed for DNA. Unfortunately, however, the
methods we used for gut pigment determination where
copepods were anesthetized, sorted into tubes and
acetone added, were not rapid enough to have detected
any such possible rapid disappearance. Notably, the
rates of decline in gut pigments we observed were consistently lower than the rate predicted (0.061 min21) for
the same temperature (208C) from an equation presented by Mauchline (Mauchline, 1998). This may have
been due to the relatively small amount of food ingested
during the short feeding time used in the experiments
(2 – 6 min). Gut evacuation of copepod rates appear to
be negatively affected by the amount of food in the guts
(Dagg and Walser, 1987; Pasternak, 1994; Durbin et al.,
1995; Irigoien, 1998).
For copepods fed for 60 min on both prey species,
there was a suggestion of a slower DNA digestion rate
and a slightly longer period during which prey DNA
remained elevated. This may be due to the utilization
of B-cells as they burst (Nott et al., 1985) with the cyclical changes in the gut wall between feeding and nonfeeding periods affecting the digestion capabilities of the
copepods. All copepods are presumed to be in the same
Quantification of prey DNA in guts
and fecal pellets of copepods
This study demonstrates that prey DNA can be detected
routinely and quantified in the guts and fecal pellets of
copepods using species-specific primers and qPCR. The
relationships we observed between gut fullness as
measured by DNA, and prey concentration and time,
were similar to what we would have expected using traditional bottle incubation techniques and cell counts.
The concentration of T. weissflogii at which gut contents
reached 90% of the asymptotic value (Ccrit) was
1530 cells mL21 when gut contents were measured as
DNA; similar to the value for Ccrit of 940 cells mL21
when adult female A. tonsa were fed the same prey
(T. weissflogii) at 208C and ingestion determined by cell
counts (Thompson et al., 1994; Fig. 2).
The rate of gut filling as measured by DNA was very
rapid, reaching 95% of the maximum gut content
(Smax) at 3 min. This contrasts with a time of 58 min for
prey chlorophyll pigments in the gut to reach 95% of
Smax. While this appears to be a large difference, it is
what we would expect based on a consideration of
digestion rate on the rate of gut filling. With copepods
previously deprived of food the rate of gut filling is controlled by the rate of disappearance of the food tracer
through either digestion or evacuation. The effect of
tracer disappearance rate on gut filling time is illustrated
in Fig. 6 where the rate of gut filling was calculated
from:
St ¼ S0 eRt þ
F
ð1 eRt Þ
R
where St is the gut content at time t, S0 the initial
amount of food, R the exponential stomach content disappearance rate and F the rate of food consumption
(Elliott and Persson, 1978). For this example, we used
two exponential disappearance rates, 0.5 min21 a rate
similar to our results for DNA digestion and
0.061 min21 for the disappearance rate of gut pigments
in copepods at 208C (Mauchline, 1998), and expressed
the gut contents as a percent of maximum gut fullness.
DNA in the gut reached 95% of the maximum gut
content (Smax) at 6 min, while phytoplankton pigment
in the gut reached 95% of Smax at 50 min. While food
tracer disappearance rates have a strong effect on the
80
E. G. DURBIN ET AL.
j
COPEPOD FEEDING
Fig. 6. (A) Changes in gut fullness with time expressed as a percentage of the maximum for two different exponential disappearance rates, R ¼
0.06 and R ¼ 0.5. (B) Rate of gut filling in terms of 18S copies copepod21 for different feeding rates (F ¼ 10, 50 and 100 18S copies s21) where
the exponential disappearance rate is R ¼ 0.5 min21.
rate of gut filling as shown above, different rates of
feeding have no effect on the rate of gut filling for the
same exponential disappearance rate R. For example,
using a value of R ¼ 0.5 min21, and feeding rates of
100, 50 and 10 DNA copies s21 (Fig. 6), the same proportional stage of gut filling was reached at the same
time for each feeding rate.
REFERENCES
Arnaud, J., Brunet, M. and Mazza, J. (1980) Structure et ultrastructure comparées de l’intestin chez plusiers espèces de Copépodes
Calanoides (Crustacea). Zoomorphologie, 95, 213– 233.
Bamstedt, U., Gifford, D. J., Irigoien, X., Atkinson, A. and Roman,
M. (2000) Feeding. In Harris, R. P., Wiebe, P. H., Lenz, J., Skjoldal,
H. R. and Huntley, M. (eds), ICES Zooplankton Methodology Manual.
Academic Press, San Diego, CA, pp. 455 –532.
Barofsky, A., Simonelli, P., Vidoudez, C., Troedsson, C., Nejstgaard,
J. C., Jakobsen, H. H. and Pohnert, G. (2010) Growth phase of the
diatom Skeletonema marinoi influences the metabolic profile of the
cells and the selective feeding of the copepod Calanus spp. J. Plankton
Res., 32, 263–272.
CONCLUSIONS
We have shown that the digestion rates of prey DNA in
copepods can be measured directly. The very rapid
rates we saw in experiments where copepods were
deprived of food and then fed for only a few minutes
suggests that DNA degradation is an enzyme-mediated
reaction. However, it should be noted that our understanding of factors such as temperature, short-term
feeding periodicity and prey type controlling these rates
is still very limited. These rapid DNA digestion rates
have implications for using prey DNA to estimate in situ
feeding by zooplankton. First, tow lengths should be
kept to a minimum and zooplankton fixed in 95%
EtOH as quickly as possible. Second, the prey DNA
present in the guts represents recent feeding, perhaps
only over the previous 20– 30 min at the temperatures
examined here.
Blankenship, L. E. and Yayanos, A. A. (2005) Universal primers and
PCR gut contents to study marine invertebrate diets. Mol. Ecol., 14,
891 –899.
Brunet, M., Arnaud, J. and Mazza, J. (1994) Gut structure and digestive cellular processes in marine Crustacea. Oceanogr. Mar. Biol. Ann.
Rev., 32, 335– 367.
Dagg, M. J. and Walser, W. E. (1987) Ingestion, gut passage time, and
egestion by the copepod Neocalanus plumchrus in the laboratory and
in the sub-Arctic ocean. Limnol. Oceanogr., 32, 178– 188.
Durbin, E. G. and Campbell, R. G. (2007) Reassessment of the gut
pigment method for estimating in situ copepod ingestion. Mar. Ecol.
Progr. Ser., 331, 305 –307.
Durbin, E. G., Campbell, R. C., Gilman, S. and Durbin, A. G. (1995)
In situ feeding of Calanus finmarchicus in the southern Gulf of Maine
during late spring. Cont. Shelf Res. 15, 539–570.
Durbin, E. G., Casas, M. C., Rynearson, T. A. and Smith, D. C.
(2008) Measurement of copepod predation on nauplii using qPCR
of the cytochrome oxidase gene. Mar. Biol., 153, 699– 707.
Durbin, E. G. and Durbin, A. G. (1978) Length-weight relationship of
Acartia clausi in Narragansett Bay, R.I. Limnol. Oceanogr., 23,
958 –969.
FUNDING
Elliott, J. M. and Persson, L. (1978) The estimation of daily rates of
food consumption for fish. J. Anim. Ecol. 47, 977– 991.
This research was supported by NSF (0748963 and
909415) awarded to E.G.D. and T.A.R. This research
was conducted using facilities supported by the
National Science Foundation under EPSCoR
(0554548), including the URI flowing seawater facility,
the URI Molecular Ecology Lab and the Rhode Island
Genomics and Sequencing Center.
Gauld, D. T. (1957) A peritrophic membrane in calanoid copepods.
Nature, 180, 510.
Guillard, R. R. L. (1975). Culture of phytoplankton for feeding
marine invertebrates. In Smith, W. L. and Chaney, M. H. (eds),
Culture of Marine Invertebrate Copepods. Plenum Press, New York, USA,
pp. 29–60.
81
JOURNAL OF PLANKTON RESEARCH
j
VOLUME
34
j
NUMBER
1
j
PAGES
72 – 82
j
2012
Nejstgaard, J. C., Frischer, M. E., Raule, C. L., Grubel, R., Kohlberg,
K. E. and Verity, P. G. (2003) Molecular detection of algal prey in
copepod guts and fecal pellets. Limnol. Oceanogr. Methods, 1, 29– 38.
Haley, S. T., Juhl, A. R., Keafer, B. A., Anderson, D. M. and Dyhrman,
S. T. (2010) Detecting copepod grazing on low-concentration populations of Alexandrium fundyense using PCR identification of ingested
prey. J. Plankton Res., doi: 10.1093/plankt/fbq151.
Nejstgaard, J. C., Frischer, M. E., Simonelli, P., Troedsson, C., Brakel,
M., Adiyaman, F., Sazhin, A. F. and Artigas, L. F. (2008)
Quantitative PCR to estimate copepod feeding. Mar. Biol., 153,
565 –577.
Head, E. J. H. and Harris, L. R. (1996) Chlorophyll destruction by
Calanus spp. grazing on phytoplankton: kinetics, effects of ingestion
rate and feeding history, and a mechanistic interpretation. Mar. Ecol.
Progr. Ser., 135, 223–235.
Nott, J. L., Corner, E. D. S., Mavin, L. J. and O’Hara, S. C. M.
(1985) Cyclical contributions of the digestive epithelium to faecal
pellet formation by the copepod Calanus helgolandicus. Mar. Biol., 89,
271 –279.
Irigoien, X. (1998) Gut clearance rate constant, temperature and
initial gut contents: a review. J. Plankton Res., 20, 997–1003.
Jarman, S. N., Gales, N. J., Tierney, M., Gill, P. C. and Elliott, N. G.
(2002) A DNA-based method for identification of krill species and
its application to analyzing the diet of marine vertebrate predators.
Mol. Ecol., 11, 2679–2690.
Ohman, M. D. (1992) Immunochemical recognition of oligotrich ciliates. Mar. Biol., 114, 653– 660.
Parsons, T. R., Maita, Y. and Lalli, C. M. (1984) A Manual of Chemical
and Biological Methods for Seawater Analysis. Pergamon Press, New York.
Juhl, A. R., Ohman, M. D. and Goericke, R. (1996) Astaxanthin in
Calanus pacificus: assessment of pigment-based measures of omnivory.
Limnol. Oceanogr., 41, 1198–1207.
Passmore, A. J., Jarman, S. N., Swadling, K. M., Kawaguchi, S.,
McMinn, A. and Nicol, S. (2006) DNA as a dietary biomarker in
Antarctic krill, Euphausia superba. Mar. Biotechnol., 8, 686–696.
Kleppel, G. S., Frazel, D., Pieper, R. E. and Holliday, D. V. (1988)
Natural diets of zooplankton off southern California. Mar. Ecol.
Progr. Ser., 49, 231– 241.
Pasternak, A. F. (1994) Gut fluorescence in herbivorous copepods: an
attempt to justify the method. Hydrobiologia, 292/293, 241– 248.
Kleppel, G. S. and Pieper, R. E. (1984) Phytoplankton pigments in
the gut contents of planktonic copepods from coastal waters off
southern California. Mar. Biol., 78, 193– 198.
Pons, J. (2006) DNA-based identification of prey from non-destructive,
total DNA extractions of predators using arthropod universal
primers. Mol. Ecol. Notes, 6, 623–626.
Light, A., Taniguchi, H. and Chen, R. F. (1974) A kinetic study of the
complementation of fragments of staphylococcal nuclease. J. Biol.
Chem., 249, 2285– 2293.
Read, D. S., Sheppard, S. K., Bruford, M. W., Glen, D. M. and
Symondson, W. O. C. (2006) Molecular detection of predation by
soil micro-arthropods on nematodes. Mol. Ecol., 15, 1963– 1972.
Linn, S. M., Lloyd, R. S. and Roberts, R. J. (1993) Nucleases, 2nd edn.
Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
Simonelli, P., Troedsson, C., Nejstgaard, J. C., Zech, K., Larsen, J. B.
and Frischer, M. E. (2009) Evaluation of DNA extraction and handling procedures for PCR-based copepod feeding studies. J. Plankton
Res., 31, 1465–1474.
Linn, S. M. and Roberts, R. J. (1982) Nucleases. Cold Spring Harbor
Laboratory Press, Cold Spring Harbor, NY.
Mackas, D. L. and Bohrer, R. (1976) Fluorescence analysis of zooplankton gut contents and an investigation of diel feeding patterns.
J. Exp. Mar. Biol. Ecol., 25, 77– 85.
Symondson, W. O. C. (2002) Molecular identification of prey in predator diets. Mol. Ecol., 11, 627– 641.
Thompson, A. M., Durbin, E. G. and Durbin, A. G. (1994) Temporal
changes in the maximum ingestion rate of Acartia tonsa in
Narragansett Bay. Mar. Ecol. Progr. Ser., 108, 91–105.
Mauchline, J. (1998) The biology of plankton copepods. In Advanced
Marine Biology, Vol. 33. Academic Press, San Diego.
Medlin, L., Elwood, H. J., Stickel, S. and Sogin, M. L. (1988) The
characterization of enzymatically amplified eukaryotic 16S-like
rRNA coding regions. Gene, 71, 491– 499.
Troedsson, C., Frischer, M. E., Nejstgaard, J. C. and Thompson,
E. M. (2007) Molecular quantification of differential ingestion
and particle trapping rates by the appendicularian Oikopleura dioica
as a function of prey size and shape. Limnol. Oceanogr., 52, 416– 427.
Miller, C. B., Nelson, D. M., Guillard, R. R. L. and Woodward, B. L.
(1980) Effects of media with low silicic acid concentrations on tooth
formation in Acartia tonsa Dana (Copepoda: Calanoida). Biol. Bull.,
159, 349 –363.
Vestheim, H., Edvardsen, B. and Kaardvedt, S. (2005) Assessing
feeding of a carnivorous copepod using species-specific PCR. Mar.
Biol., 147, 381 –385.
82