Bulletin of Entomological Research (2006) 96, 295–304
DOI: 10.1079/BER2006426
Evaluation of temperature gradient
gel electrophoresis for the analysis of
prey DNA within the guts of
invertebrate predators
G.L. Harper1#, S.K. Sheppard1, J.D. Harwood1z, D.S. Read1,
D.M. Glen2·, M.W. Bruford1 and W.O.C. Symondson1 *
1
Cardiff School of Biosciences, Cardiff University, PO Box 915, Cardiff,
CF10 3TL, UK: 2IACR-Long Ashton Research Station, Department of
Agricultural Sciences, University of Bristol, Long Ashton, Bristol,
BS41 9AF, UK
Abstract
The utility of temperature gradient gel electrophoresis (TGGE) as a means of
analysing the gut contents of predators was evaluated. Generalist predators
consume multiple prey species and a species-specific primer approach may not
always be a practical means of analysing predator responses to prey diversity in
complex and biodiverse ecosystems. General invertebrate primers were used to
amplify the gut contents of predators, generating banding patterns that identified
component prey remains. There was no evidence of dominance of the polymerase
chain reaction (PCR) by predator DNA. When applied to field samples of the
carabid predator Pterostichus melanarius (Illiger) nine banding patterns were
detected, including one for aphids. To further distinguish between species, groupspecific primers were designed to separate species of earthworm and aphid. TGGE
of the earthworm PCR products generated banding patterns that varied with
haplotype in some species. Aphid and earthworm DNA could be detected in the
guts of carabids for up to 24 h using TGGE. In P. melanarius, with low numbers of
prey per insect gut (mean < 3), interpretation of banding patterns proved to be
tractable. Potential problems of interpretation of TGGE gels caused by multiple
prey bands, cryptic bands, haplotype variation, taxonomic uncertainties (especially
with regard to earthworms), secondary predation, scavenging and presence of
parasites and parasitoids in the prey or the predators, are discussed. The results
suggest that PCR, using combinations of general invertebrate and group-specific
primers followed by TGGE, provides a potentially useful approach to the analysis
of multiple uncharacterized prey in predators.
*Author for correspondence
Fax: 029 20 874305
E-mail: [email protected]
#Present address: School of Applied Sciences, University of Glamorgan, Pontypridd, Mid-Glamorgan, CF37 1DL, UK
zPresent address: Department of Entomology, University of Kentucky, S-225 Agricultural Science Center – North,
Lexington, KY 40546-0091, USA
·Present address: Styloma Research and Consulting, Phoebe, The Lippiatt, Cheddar, BS27 3QP, UK
S.K. Sheppard and G.L. Harper contributed equally to the experimental work in this paper and should be considered
joint first authors
296
G.L. Harper et al.
Keywords: aphid, carabid beetle, earthworm, gut content analysis, predator–prey
interactions, Pterostichus melanarius
Introduction
Interactions between generalist predators and their many
different prey are key components of ecological studies that
seek to explain the processes driving animal population
dynamics. DNA-based techniques are increasingly becoming
the methods of choice for research into predator–prey
relationships (reviewed in Symondson, 2002; Sheppard &
Harwood, 2005). Techniques and PCR primers have been
developed to detect predation by invertebrate predators
upon a range of other invertebrates including aphids
(Homoptera: Aphididae) (Chen et al., 2000; Cuthbertson
et al., 2003), psyllids (Homoptera: Psyllidae) (Agustı́ et al.,
2003b), Collembola (Agustı́ et al., 2003a), mosquitoes
(Diptera: Culicidae) (Zaidi et al., 1999), moths (Lepidoptera:
Noctuidae, Crambidae and Geometridae) (Agustı́ et al., 1999;
Hoogendoorn & Heimpel, 2001; Sheppard et al., 2004),
whiteflies (Homoptera: Aleyrodidae) (Agustı́ et al., 2000),
earthworms (Annelida) and molluscs (Harper et al., 2005). To
date, few studies have gone on to use these techniques to
analyse predation on natural populations of prey in the field
(Agustı́ et al., 2003a; Dodd et al., 2003; Kaspar et al., 2004;
Harper et al., 2005).
Although the study of predation on specific target
prey is usually the goal, this is not always the case and
important ecological questions can be asked using different
approaches. Here, the primary goal was to evaluate TGGE as
an approach that might be used to study predator responses
to prey diversity. It is known that, for generalist predators, a
diverse diet can enhance predator fitness (Uetz et al., 1992;
Toft & Wise, 1999; Oelbermann & Scheu, 2002) and that some
arthropod predators actively seek to balance their diets
(Greenstone, 1979; Mayntz et al., 2005). Therefore it might be
predicted that, if prey is sufficiently abundant, predators
in the field would specifically aggregate to diversity and
when they get there consume a diverse range of prey or,
in an alternative scenario, be more selective, optimizing
nutrient intake. Before undertaking such a study, we needed
to develop methods that were primarily designed to examine
diversity in the diet per se, rather than relying necessarily
upon identification of individual components within the diet
(although that too is possible). Recently, gut content analysis
has been made more rapid by multiplexing a broad range of
prey DNAs simultaneously using fluorescent labelled
primers and visualizing the prey-specific size fragments on
an automated sequencer (Harper et al., 2005). This approach
permits rapid analysis of field samples to study the diversity
of a specific range of target prey in the guts of predators.
However, it cannot be used to detect uncharacterized or
unexpected prey for which no specific primers have been
designed. This may be addressed by using general or groupspecific primers (Sutherland, 2000; Jarman et al., 2004), then
cloning and sequencing. The process is slow and expensive,
and relies completely upon the availability of matching
sequences on databases such as GenBank.
An alternative to these established methods is to use
molecular profiling technologies that have been used
extensively for the study of diversity amongst microbial
communities. Profiling has been primarily by TGGE or
denaturing gradient gel electrophoresis (DGGE) (e.g.
Muyzer et al., 1993; Felske et al., 1998; Nakatsu et al., 2000;
McCaig et al., 1999, 2001; Sheppard et al., 2005a). The aim
of such studies was to examine bacterial diversity within
habitats where diversity (rather than the identities of the
component species) was the primary criterion of interest.
The bands generated within these DNA profiles are
described as operational taxonomic units (OTUs) (e.g.
Torsvik et al., 1996) and can be counted to provide a gross
measure of diversity within a microbial community. Each
band does not necessarily represent a distinct species. It is
recognized that the general primers used in such work
sometimes amplify more than one band from some species
while cryptic bands (one superimposed on another) may
make separation incomplete. Nevertheless, overall, the
number of bands amplified will be greater in samples from
a more diverse than those from a less diverse community, as
shown in the microbial studies of soils (e.g. McCaig et al.,
1999, 2001; Sheppard et al., 2005a).
A TGGE-based approach allows sequences of the same
length to be separated where they vary by just a single base
pair substitution (Riesner et al., 1991). This may be important
in predation studies when unique primer sites cannot be
found to separate closely related species. Control DNA can
be run on the gels to allow species to be identified when
required. More importantly, it allows most species within
broad groups of prey to be detected and separated, if not
identified to species; it may tell us, for example, that three
different aphids or two different earthworm species have
been consumed even when the precise species are not
known. Thus TGGE can provide a method of analysing
predator responses to different levels of prey diversity in the
field, where ‘diversity’, rather than individual identity, is the
primary factor of interest. However, it can also be used to
reveal predation on unexpected prey, or for prospecting.
If TGGE, using general earthworm primers, revealed that
three species were being regularly consumed, then efforts
could be made to identify those prey by running earthworm
DNA from identified species from the field site through
TGGE gels, to characterize the mobility of their speciesdiagnostic band or bands.
Despite these advantages, TGGE has several potential
drawbacks. As this technique has never been used to study
predation before, our aim was to answer a series of critical
questions. If general primers are used, would predator DNA
dominate the PCR reaction and physically obscure prey
sequences on TGGE gels? Would cryptic and multiple bands
make interpretation of TGGE fingerprints impossible?
Would intraspecific haplotype diversity obscure specieslevel discrimination by TGGE? The following experiments
were designed to answer these questions.
Materials and methods
DNA extraction
DNA was extracted from invertebrate species (generally
n = 10), including earthworms (Annelida: Lumbricidae)
(Allolobophora chlorotica (Savigny), Aporrectodea caliginosa
(Savigny), Aporrectodea longa (Ude), Lumbricus rubellus
(Hoffmeister), Lumbricus terrestris Linnaeus, Lumbricus
Detecting prey diversity within predators
castaneus (Savigny), Octolasion cyaneum (Savigny); aphids
(Hemiptera: Aphididae) (Myzus persicae (Sulzer), Aphis fabae
Scopoli, Brevicoryne brassicae Linnaeus, Metopolophium dirhodum (Walker), Rhopalosiphum padi Linnaeus, Sitobion avenae
(Fabricius)), blowfly larvae Calliphora vomitoria (Linnaeus)
(Diptera: Calliphoridae), unfed carabids, Pterostichus melanarius (Illiger) (Coleoptera: Carabidae), and beetles from the
feeding experiments. The same basic procedure was used to
extract DNA from all specimens, including field-collected
beetles. Prior to DNA extraction, the earthworms were
starved for 48 h to allow soil to pass through their gut. Beetle
foreguts were removed and homogenized rather than
extracting DNA from the whole organism. Extraction was
carried out using the commercially available QIAamp1
DNA Mini Kit (QIAGEN GMBH, Hilden, Germany),
following the manufacturer’s instructions. The DNA was
resuspended in 75–200 ml of the elution buffer supplied, and
stored at x20 C.
Earthworm and aphid specific primers
General invertebrate primers were employed to
sequence-characterize part of the 12S rRNA gene of the
earthworms, SR-J-14233 and Sr-N-14588 (Simon et al., 1994).
PCR was conducted in a 25 ml reaction, containing 50–100 ng
of template DNA; 1 U Taq polymerase (Gibco BRL, Paisley,
UK); 0.5 mMoles of each primer; 20 mM (NH4)SO4; 75 mM
Tris-HCl, pH 8.8; 0.01% (v/v) Tween 20; 2 mM MgCl2; 0.2 mM
dNTPs (Abgene, Epsom, UK). The PCR reactions were
carried out on a GeneAmp 9700 PCR system (Applied
Biosystems, California, USA). The following conditions were
used: 1r94 C, 4 min; 30r(94 C, 45 s; 45 C, 45 s; 72 C, 1 min
15 s); 1r72 C for 10 min. The PCR amplified DNA was
then cleaned up using a Turbo GeneClean kit (Q-BIOgene,
Cambridge, UK), following the manufacturer’s instructions.
Sequencing reactions were carried out using Big-Dye
Terminators (PE-Applied Biosystems, California, USA) on a
GeneAmp 9700 PCR system (Applied Biosystems, California, USA) via the manufacturer’s instructions, then run out
on a Perkin Elmer ABI3100 automated sequencer. The region
was sequenced in both forward and reverse orientations.
Group-specific primers for earthworms (12SNF AAACTTAAAGATTTTGGCGGTGTCT and 12SNR GCTGCACTTTGACCTGACGTAT) were designed from the 12SrRNA gene
and synthesized (Sigma-Genosys, Gillingham, UK). These
primers proved to amplify earthworm sequences of 200 bp.
The general aphid primers Aph149F (AATCAAAATAAATGTTGATA) and Aph344R (GGAACAGGWACAGGATGAAC) designed by Harper et al. (2005) were used to
separate aphids.
PCR amplification with group-specific primers
In order to characterize a system for species separation
within groups, PCR–TGGE was carried out on freshly
extracted DNA from the earthworm and aphid species
listed above. Each 25 ml reaction contained 50–100 ng of
template DNA, 1 U Taq polymerase (Gibco BRL, Paisley,
UK), 0.2 mMoles of each primer, 20 mM (NH4)SO4, 75 mM
Tris-HCl, pH 8.8, 0.01% (v/v) Tween 20, 2 mM MgCl2 and
0.2 mM dNTPs (Abgene, Epsom, UK). The reaction was
carried out under the following conditions: 1r94 C, 4 min;
35r(94 C, 45 s; Ta, 45 s; 72 C, 1 min 15 s; 1r72 C for 10 min
(Ta: 57 C for earthworms (12SNF/12SNR), 55 C for aphids
297
(149F/344R)). Each primer pair was also tested for cross
reactivity against numerous other species of invertebrate
including Annelida, Arachnida, Collembola, Coleoptera,
Homoptera and Mollusca.
PCR amplification of the COI gene with ‘universal’
invertebrate primers
PCR amplification of relatively small fragments of the
COI gene was performed using universal PCR primers
targeting conserved regions of the insect mitochondrial
genome. Four primer sets were initially tested, but
the combination C1-J-1859/C1-N-2191 (Simon et al., 1994)
proved to be effective, and amplified a 332 bp fragment.
For the field-caught P. melanarius, species-specific PCRs
had been performed previously to identify the presence of
aphid in the predators’ guts as part of the study by Harper
et al. (2005). To prove the utility of the TGGE system fieldcaught predators were needed that had been shown, using a
proven methodology, to have eaten aphids. Sixteen beetles
were selected for PCR with general invertebrate primers and
subsequent TGGE analysis.
PCR Amplification was performed with 25 ml reaction
volumes consisting of 4 ml of extracted DNA, 2.5 ml 10rPCR
buffer (provided by the manufacturer), 10 mM dNTPs
(Boehringer Mannheim, Indianapolis, Indiana, USA), 5 U
Taq polymerase (Promega, Madison, Wisconsin, USA),
2.5 mM MgCl2 and 50 mM of each primer. Optimization of
PCR conditions was carried out in pilot experiments. All
reactions were carried out in a GeneAmp 9700 PCR system
(Applied Biosystems, California, USA). The final parameters
of the thermal cycle were: 95 C for 3 min, then 40 cycles of
94 C for 45 s, 47 C for 1 min and 72 C for 1 min 15 s, with a
final extension period at 72 C for 10 min. 5 ml of PCR
products were separated by electrophoresis in 1.5% agarose
and stained with 0.5 mg mlx1 ethidium bromide. Of the PCR
amplification products derived using primers C1-J-1859/C1N-2191, 20 ml or less were used in TGGE. These products
were selected as they were short enough (332 bp) to provide
an ecologically relevant detection period, whilst potentially
containing enough sequence variation for species separation.
A GC-clamp was used with the general invertebrate COI
primers and this improved the sensitivity/resolution of the
TGGE analyses (Myers et al., 1985; Sheffield et al., 1989).
However, they did not improve discrimination when used
with the aphid and earthworm group-specific primers and
therefore results for the latter were obtained without GC
clamps.
TGGE analysis
The DCode Universal Mutation System (Bio-Rad Inc.,
California, USA) was used as described in the manufactures
instruction manual. PCR samples were added to an equal
volume of loading buffer (0.5 g lx1 bromophenol blue,
0.5 g lx1 xylene cyanol, 70% glycerol in dH2O) (Bio-Rad,
California, USA) and loaded onto a 10% polyacrylamide gel.
TGGE gels (16r10r0.1 cm) contained 5 ml polyacrylamide/
BIS (37.5 : 1) in 1.25rTAE buffer (1rTAE: 40 mM Tris
acetate, 1 mM EDTA, pH 8.0), 6 M urea, 0.1% TEMED, and
0.1 g lx1 of ammonium persulphate. Outside lanes were run
as dye-only blanks (19 ml). Parallel gel electrophoresis was
performed in TGGE tanks (Bio-Rad, California, USA) with
7 litres of 1.25rTAE buffer. Changes were made to the
298
G.L. Harper et al.
temperature gradient and ramp rate following visualization
of perpendicular pilot gels and approximation of electrophoretic mobilities of the partially melted DNA fragments
from beetles and aphids. Numerous gels were produced
before optimal separation of minimum temperature melting
domains was obtained.
For the universal invertebrate primers, electrophoresis
proceeded at a constant voltage of 50–65 V for approximately
10 h on a temperature gradient increasing from 47 C to 54 C
at a ramp rate of 1 C hx1. The maximum temperature was
reached after 7 h but continued running of the gel enhanced
separation of partially melted DNA. For the aphid and
earthworm-specific amplicons, electrophoresis proceeded at
a constant voltage of 120 V. For the aphid DNA, gels were
run for 5.5 h, on a temperature gradient increasing from 45 C
to 50.5 C, on a ramp rate of 1 C hx1. For the earthworm
DNA, the gels were run for 4 h, over a temperature gradient
of 56 C to 60 C, with a ramp rate of 1 C hx1. Visualization
of gels was achieved by UV transillumination following
electrophoresis and gel staining for 20 min with SYBR gold
nucleic acid stain (1 : 10 000 dilution, Molecular Probes,
Leiden, The Netherlands). GC-clamps enhanced the separation of fragment amplified with the general primers.
Predators and field site
Carabid predators, P. melanarius, for both laboratory trials
and analysis of predation in the field, were collected from
crops of winter wheat at Long Ashton Research Station,
Bristol UK. Details of the trapping system are described in
Harper et al. (2005). A subsample of carabids was used in the
current experiments to demonstrate the ability of TGGE to
detect aphid DNA fragments known to be present, following
multiplex PCR (Harper et al., 2005).
Feeding experiments
Three feeding experiments were conducted to test the
ability of TGGE to detect semi-digested prey in predators.
Beetles used in the feeding experiments were maintained,
individually, in a controlled environment room (16+1 C)
with a 16 : 8 h L : D cycle, in transparent sealed perforated
plastic pots (9r6 cm) on 3 cm of sphagnum moss peat. They
were fed up to two blowfly larvae, C. vomitoria, per week for
one month to ensure a relatively constant nutrition status,
then starved for two weeks and transferred to separate
triple-vented Petri dishes (9r1.5 cm) containing damp filter
paper. Twenty P. melanarius (10 of each sex) were killed
(by freezing at x80 C) at the outset of the feeding trials to
provide unfed controls.
Carabid–mixed prey feeding experiment – detection
of multiple prey
The primary aim of this experiment, using universal
invertebrate primers (C1-J-1859/C1-N-2191, Simon et al.,
1994), was to ensure that the PCR reaction was not
dominated, or swamped, by DNA from the beetle predator.
Ten P. melanarius, five of each sex, were fed, ad libitum, a 1 : 1
homogenate (equal biomass of the two prey species
mascerated together in an Eppendorf tube) of the aphid
S. avenae and the blowfly C. vomitoria larvae for 2 h. Beetles
were killed by freezing at x80 C immediately after the
feeding period. DNA extraction, PCR and TGGE were
carried out as detailed above.
Carabid–aphid feeding experiment – post-feeding detection
period within predators
This experiment provided baseline data on the period of
time after feeding that aphid prey DNA could be detected
from within predator guts. Beetles were allowed to feed for
2 h ad libitum on aphids, S. avenae, which had just been killed
by freezing. Fourteen (seven male and seven female) beetles
were frozen immediately at x80 C after feeding. The
remaining beetles were moved to clean Petri dishes lined
with moist filter paper and 14 individuals (seven of each sex)
were killed, by freezing, at various times after feeding. Time
periods selected were based on previous studies of the
detectability of small multiple-copy prey DNA fragments in
invertebrate predator guts (Chen et al., 2000; Agustı́ et al.,
2003a). Beetles were killed at 1, 3, 6, 12, 24, 48 and 72 h after
feeding and stored at x80 C. All samples were tested by
PCR–TGGE using the same general primers (C1-J-1859/C1N-2191, Simon et al., 1994) as in the carabid–mixed prey
feeding experiment above.
Carabid–earthworm feeding experiment – post-feeding detection
period within predators
Beetles were fed on the earthworm Allolobophora chlorotica. The samples were taken from the first 24 h of
the feeding experiment described in detail in Harper et al.
(2005). Briefly, starved beetles were fed ad libitum for 2 h
on earthworms and only beetles observed feeding were
included in the experiment. Batches of beetles were killed by
freezing 0, 2, 4, 8, 16 and 24 h after feeding on the worms
and analysed using the group-specific earthworm primers
12SNF/12SNR.
Results
Carabid–mixed prey feeding experiment – detection of
multiple prey
Analysis was carried out on total DNA from aphids,
S. avenae, blowfly larvae, C. vomitoria, starved P. melanarius
and P. melanarius fed with a mixture of aphid and maggot
(n = 10) using the universal primer pair C1-J-1859/C1-N2191. TGGE analysis of three replicates of each category of
invertebrate are shown in fig. 1, and demonstrate that the
melting domains on the amplified fragments were sufficiently different, in this simple system, to provide clear
separation on acrylamide gels. Most importantly, the profiles
were not obscured by a preponderance of DNA amplified
from the predator nor was PCR dominance a factor. Distinct
and reproducible banding patterns are associated with
S. avenae, C. vomitoria and P. melanarius, although in each
case by a double band.
Carabid–aphid feeding experiment x post-feeding
detection period within predators
COI PCR products amplified from the beetles, again
using the general invertebrate primers C1-J-1859/C1-N-2191,
were separated by TGGE. The detection of the S. avenae
and P. melanarius COI banding pattern in the same lane
was indicative of a positive result and was recorded as
Detecting prey diversity within predators
1
2
3
4
5
6
7
8
9
10
299
11
12
Fig. 1. SYBR gold stained polyacrylamide gel of TGGE banding patterns of DNA extracted from Sitobion avenae (lanes 1–3), Calliphora
vomitoria (lanes 4–6), Pterostichus melanarius (starved) (lanes 7–9) and P. melanarius killed immediately after eating a homogenate of
S. avenae and C. vomitoria (lanes 10–12). TGGE analysis was on COI mitochondrial gene fragments following PCR amplification with
universal primers (C1-J-1859/C1-N-2191).
Carabid–earthworm feeding experiment – post-feeding
detection period within predators
12S PCR products amplified from the beetles, using the
group-specific earthworm primers 12SNF/12SNR, were
separated by TGGE. Figure 3 shows that up to 8 h after
ingestion the earthworm DNA was detectable by TGGE
in 100% of predators and in some individuals up to 24 h.
The single A. chlorotica bands amplified from the guts
of P. melanarius migrated in each case to one of two very
different points on the TGGE gel, probably because
A. chlorotica in fact comprises two cryptic earthworm
species/genotypes (see Discussion). Problems with earthworm taxonomy are discussed in Harper et al. (2005).
Aphid consumption by field-caught carabids
Of the numerous field-collected beetles (from the study
by Harper et al., 2005), 16 were analysed by TGGE following
2. 75
Lo g e n um ber of beet l es
p o s i ti ve fo r aphi d D N A
confirming beetle consumption of aphid. The number of
beetles testing positive for aphid from a total of 14 at each
time period is shown in fig. 2. Linear regression of the
number testing positive (loge transformed) against time (h)
showed a simple exponential decline in detectability, (loge
beetle number = 2.663x0.054 hours R2 99.2%, P < 0.001).
Aphid DNA was still detectable in 50% of the beetles 12 h
after feeding and 29% after 24 h.
2. 50
2. 25
2. 00
1. 75
1. 50
0
5
10
15
T i m e(h)
20
25
Fig. 2. Loge numbers of Pterostichus melanarius predators that
tested positive for aphid (Sitobion avenae) DNA between zero
and 24 h digestion (samples digested for 48 h and 72 h were all
negative). Detection was by PCR amplification of a 332 bp
fragment of COI mtDNA using the general invertebrate primers
C1-J-1859/C1-N-2191 (Simon et al., 1994).
PCR using the general invertebrate primers C1-J-1859/C1-N2191. These included 14 that had been shown to be positive
for aphid and two that were aphid-negative as controls. The
PCR products from these beetle gut samples were run on a
TGGE gel with a negative control, an unfed beetle, and a
positive control, aphid DNA (fig. 4). Pterostichus melanarius
300
G.L. Harper et al.
2h
1
4h
2
3
4
B 1
8h
2 3
4
−ve 1
16 h
2
3
4 −ve
24 h
−ve 1 2 3 4
1 2 3 4
Fig. 3. SYBR gold stained polyacrylamide gel of TGGE banding patterns of PCR-amplified earthworm DNA extracted from the guts of
Pterostichus melanarius beetles killed at 2, 4, 8, 16 and 24 h post-feeding. Also shown are negative controls (xve) (unfed beetle) and a PCR
negative (B).
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
}
H
C
E
A
B
F
D
*
*
*
*
*
*
*
*
*
*
*
*
G
Fig. 4. SYBR gold stained polyacrylamide gel of TGGE banding patterns of DNA extracted from Pterostichus melanarius collected from
the field. TGGE analysis was on COI mitochondrial gene fragments following PCR amplification with universal primers (C1-J-1859/C1N-2191). DNA bands from an unfed beetle and an aphid are in lanes 1 and 2 respectively. Lanes 3–16 are beetles previously shown by
multiplex-PCR to contain aphid, lanes 17–18 are beetles previously shown to be aphid negative. Lanes in which aphid positives were
recorded by TGGE are indicated ( *) on the figure. Non-aphid and non-beetle bands with the same melting domain (migration position)
are put into groups (A–H). Band H is a cluster of bands.
and aphids, S. avenae, gave distinct banding profiles
analogous to those recorded previously (fig. 1). Aphid
positives were recorded in 12 of the 14 specimens known
to have eaten aphid. No differences were detected between
the aphid banding patterns using these general invertebrate
primers. Known negatives gave negative results. The beetles
represented in lanes 6, 7, 9, 10, 11, 12, 13, 14, 15 and
16 contained additional non-aphid bands representing
Detecting prey diversity within predators
1
a
2
b
Aca
b
3
4
a
5
a
6
7
Alo
8
9
Lc
c
d
d
e
e
10 11 12 13 14
15
Ach
16 17 18 19
301
Lt
20
Oc
21 22 23 24
25
Lr
.
26 27 28
Fig. 5. SYBR gold stained polyacrylamide gel showing TGGE banding patterns for an earthworm-specific PCR amplicon, showing
mitochondrial 12S rRNA variation among seven different earthworm species (Aporrectodea caliginosa (Aca), Aporrectodea longa (Alo),
Lumbricus castanaus (Lc), Allolobophora chlorotica (Ach), Lumbricus terrestris (Lt), Octolasion cyaneum (Oc), Lumbricus rubellus (Lr)).
Amplicons for the two species showing intraspecific variation (Aca and Lc) were sequenced, and haplotypes are represented as a to e in
superscript.
non-aphid prey (or prey groups) consumed. Examination by
eye allows these bands to be put into eight groups (A–H)
representing genetically distinct prey categories. Band H
is a cluster of bands that appear together in several beetles
and probably represent a single prey category. Beetle 10
(fig. 4) appears to contain the remains of six different nonaphid prey.
Refining the approach to separate species within
a group (earthworms)
Whilst the PCR–TGGE approach using general invertebrate primers could expose predation on disparate groups
of invertebrates, it could not always distinguish between
individual species within these groups. However, using
earthworm-specific primers, and appropriate temperature
gradient conditions, species-specific bands can be resolved
(fig. 5) (similar results for aphid not shown). Among seven
earthworm species tested, using an c. 200 bp fragment of
the 12S rRNA gene, species-specific differences were found
(fig. 5). The addition of known and characterized positive
controls in gel lanes adjacent to field samples would enable
most (but not all) species/haplotypes to be correctly scored
if necessary. Figure 5 shows that there was haplotypic
variation among the A. caliginosa (two haplotypic states,
a and b) and the L. castaneus (three haplotypic states, c, d
and e). When the 12S rRNA amplicons from these individuals were sequenced, it became clear that only slight
sequence variations were necessary for marked changes
in melting profiles. This is exemplified in A. caliginosa, where
a single base pair substitution has caused distinct scorable
changes in the melting profiles of the two haplotypes.
GenBank accession numbers for the A. caliginosa and
L. castaneus haplotypes are DQ157851–DQ157855.
Discussion
There are many positive aspects to this analysis. The
results demonstrated, for the first time, that is it possible to
use TGGE for the detection of multiple prey species in the
guts of generalist predators using general or group-specific
primers. Previous work by Deagle et al. (2005), who
attempted to examine the dietary components in the gut of
a giant squid using DGGE, was plagued by PCR artefacts
not encountered here. By running known controls through
the gels for prey species of interest (in this case aphids as
a group, fig. 4), characteristic banding patterns can be
established and predation on the species involved can be
recorded. What is more it is also possible, as we have shown,
to record and track predation on uncharacterized prey
(fig. 4), something that is impossible using prey-specific
primers. Such work can provide the first indication that a
previously unrecognized prey was a significant component
of the predator’s diet. The unknown prey identified by the
bands in fig. 4 could be tracked down, either by reamplification from the TGGE bands, followed by sequencing
and a BLAST search, or by prospecting for, and extracting
DNA from, further prey species at the field site. Though very
limited in number, and included simply to illustrate the
potential of the system, the present results provide the first
field data using this technology.
A primary application, for which this system was
designed, was to study predator responses to prey diversity
in different habitats. In such an approach the parameter of
interest is the diversity of prey in the guts of the predators as
a separate measure from the component identities of species
in the diet. It is not suggested that the components of the diet
of generalist predators be ignored. However, the potential to
use TGGE profiles to measure dietary diversity in a range
of habitats is considerable and can be used to address
fundamental and applied ecological questions relating to the
effects of biodiversity on foodwebs. Bands can be clustered
into OTUs and populations compared simply in terms of
diversity. We suggest that such an approach is appropriate
for the study of diversity in the guts of predators, where
diversity may be of species but also of haplotype (fig. 5).
Most P. melanarius caught in the field had zero or one
identifiable prey species/groups in its gut when analysed
using multiplex PCR (Harper et al., 2005). These were a
subset of the prey species actually consumed, because the
multiplex PCR only detected species for which primers were
included. The beetles in fig. 4 are a biased sample, in that
302
G.L. Harper et al.
14 were known to contain aphid in advance. Even so, the
number of prey in each beetle was low, in most cases <3,
making separation of bands or groups of bands as distinct
OTUs relatively simple. Previous work using enzyme
electrophoresis attempted to create electrophoretic keys for
all of the possible prey that might be consumed by a
predator. This approach worked well where the predators
were relatively stenophagous (reviewed in Symondson,
2002) but was much less accurate where highly polyphagous
species contained the remains of several different prey
simultaneously that generated complex superimposed
banding patterns (Walrant & Loreau, 1995). The fact that
P. melanarius gut samples contained few prey does not
prevent such confusion completely, but the dangers of such
errors are reduced. For example, on a poorer gel than the one
illustrated (fig. 4) band E could become obscured within the
cluster representing the predator, especially in individuals
where band E is faint as a result of a prolonged digestion
period in the beetle gut. We also know that OTUs may not
always represent species. The aphids consumed by the
beetles in fig. 4 were not all the same species, but the general
primer/denaturing gradient combination used did not
separate them. It is not possible to tell from the gel whether
DNA from one or several species of aphid were in the guts of
the same individual predator at the same time without
using, in addition, aphid group-specific primers.
There are several possible causes of the multiple
bands generated by TGGE. Other publications show similar
multiple banding profiles from single PCR amplicons (e.g.
Foucher & Wilson, 2002), yet little attempt is ever made to
explain their significance. Extra bands may sometimes be
caused by nuclear copies and these have indeed been found
previously in Sitobion spp. aphids (Sunnucks & Hales, 1996)
(although in our own work resequencing failed to find any).
In practical terms the number of bands produced does not
matter as long as they are consistent and the different species
can be clearly distinguished.
Although TGGE has considerable potential for the
detection of haplotype variation amongst a prey population,
this may pose a problem for conventional species identification and can be seen amongst a subset of the earthworms
(A. caliginosa and L. castaneus) (fig. 5). The specimens used in
this study came from a number of geographically distant
field sites and some genetic variation might therefore be
expected. Although the 12S rRNA gene is more conserved
than, for example, the COI gene, in both earthworm species
the variation occurred within the less conserved loop
domains (Hickson et al., 1996). All TGGE-detectable haplotypes of these species at a given field site would need to be
characterized in advance. Very little molecular genetic work
has been applied to earthworms and species definitions are
probably outdated. For example, in Octolasion tyrtaeumi, the
presence of two size classes has recently been shown to
closely correlate with variation at the COII gene, indicating
the existence of two genetically distinct lineages (Heethoff
et al., 2004). It is thought that there could be several species
aggregates and cryptic species even amongst the wellstudied British earthworms (Simms & Gerard, 1985).
Molecular detection of predation on different haplotypes
offers a new and potentially rich area for future research.
Although it has been suggested for the study of haplotype
selection by owls feeding on small mammals (Taberlet &
Fumagalli, 1996) it has never, to our knowledge, been
attempted for predation on invertebrates.
Before this work was conducted we did not know
whether the use of general invertebrate primers would be
possible, given that it might be expected that the DNA from
the gut tissue of the beetle would be in better condition than
that of the semi-digested gut contents. This could have led to
complete dominance of the PCR reaction by the beetle DNA.
However, it is clear from figs 1 and 4 that this did not
happen and that prey bands are at least as strong as those for
the predator, often stronger. Thus it may be possible to use
general invertebrate primers, available to all, to study
predator responses to prey diversity in a range of different
invertebrate food webs.
The earthworm amplicons remained clearly detectable
even after digestion within the gut of P. melanarius for 16 h
(fig. 3). Even after 24 h, some samples could still be
amplified. Similar results were obtained where the carabids
had consumed aphids, with a third of samples still positive
for aphid DNA after 24 h (fig. 1). These data compare
favourably with those obtained by Chen et al. (2000) who,
using species-specific primers to amplify a smaller 198 bp
fragment, could detect aphid DNA on agarose gels in 50% of
predators 4–9 h after ingestion (depending upon predator
species). Comparable detection periods in other predator–
prey systems have been both shorter and longer (reviewed in
Symondson, 2002). Clearly, TGGE is capable of detecting
prey remains for significant periods after prey ingestion and
such detection periods need to be quantified where precise
rates of predation on particular target prey species are being
measured (Symondson, 2002; Sheppard & Harwood, 2005).
Interestingly, DNA from the earthworm A. chlorotica was
detected as one of two alternate and very different bands on
the TGGE gel (fig. 3). Haplotypic variation is possible, but
there is strong evidence that A. chlorotica comprises two
cryptic species (Satchell, 1967) and the observed separation
into two distinct genotypes would tend to support this
conjecture.
It is important to remember that predation is not the only
way that DNA may have entered the gut of a generalist
predator (Sunderland, 1996). Secondary predation (Harwood
et al., 2001; Sheppard et al., 2005b) and scavenging (Calder
et al., 2005; Foltan et al., 2005; Juen & Traugott, 2005) are also
possible sources of error. There is little doubt that the bands
and groups of bands (A–H) in fig. 4 mainly represent prey
species, but some could conceivably represent parasitoids or
invertebrate parasites on or in those prey items or even the
predators themselves.
To conclude, TGGE using group-specific primers
has potential as a relatively simple method for analysing
the gut contents of predators without the need for prior
design of species-specific primers. This approach may be
particularly useful for examining predator responses to
prey diversity and selection in the field of particular
prey haplotypes. However, caution needs to be used in
the interpretation of TGGE gels and their limitations
recognized.
Acknowledgements
These experiments were funded by the Biotechnology
and Biological Sciences Research Council (grant no. 72/
D14569) as part of a larger project looking at the effects
of alternative prey and prey diversity on the ability of
predators to control crop pests.
Detecting prey diversity within predators
References
Agustı́, N., de Vicente, C. & Gabarra, R. (1999) Development of
sequence characterized amplified region (SCAR) markers
of Helicoverpa armigera: a new polymerase chain reactionbased technique for predator gut analysis. Molecular Ecology
8, 1467–1474.
Agustı́, N., de Vicente, C. & Gabarra, R. (2000) Developing
SCAR markers to study predation on Trialeurodes vaporariorum. Insect Molecular Biology 9, 263–268.
Agustı́, N., Shayler, S.P., Harwood, J.D., Vaughan, I.P.,
Sunderland, K.D. & Symondson, W.O.C. (2003a) Collembola as alternative prey sustaining spiders in arable
ecosystems: prey detection within predators using molecular markers. Molecular Ecology 12, 3467–3475.
Agustı́, N., Unruh, T.R. & Welter, S.C. (2003b) Detecting
Cacopsylla pyricola (Hemiptera: Psyllidae) in predator guts
using COI mitochondrial markers. Bulletin of Entomological
Research 93, 179–185.
Calder, C.R., Harwood, J.D. & Symondson, W.O.C. (2005)
Detection of scavenged material in the guts of predators
using monoclonal antibodies: a significant source of error
in measurement of predation? Bulletin of Entomological
Research 95, 57–62.
Chen, Y., Giles, K.L., Payton, M.E. & Greenstone, M.H. (2000)
Identifying key cereal aphid predators by molecular gut
analysis. Molecular Ecology 9, 1887–1898.
Cuthbertson, A.G.S., Fleming, C.C. & Murchie, A.K. (2003)
Detection of Rhopalosiphum insertum (apple–grass aphid)
predation by the predatory mite Anystis baccarum using
molecular gut analysis. Agricultural and Forest Entomology 5,
219–225.
Deagle, B.E., Jarman, S.N., Pemberton, D. & Gales, N.J. (2005)
Genetic screening for prey in the gut contents from
a giant squid (Architeuthis sp.). Journal of Heredity 96,
417–423.
Dodd, C.S., Bruford, M.W., Symondson, W.O.C. & Glen, D.M.
(2003) Detection of slug DNA within carabid predators
using prey-specific PCR primers. pp. 13–20 in Dussart,
G.B.J. (Ed.) Slug and snail pests: agricultural, veterinary and
environmental perspectives. Alton, UK, British Crop Protection Council.
Felske, A., Akkermans, A.D.L. & de Vos, W.D. (1998)
Quantification of 16S rRNAs in complex bacterial communities by multiple competitive reverse transcription – PCR
in temperature gradient gel electrophoresis fingerprints.
Applied and Environmental Microbiology 64, 4581–4587.
Foltan, P., Sheppard, S.K., Konvicka, M. & Symondson,
W.O.C. (2005) The significance of facultative scavenging
in generalist predator nutrition: detecting decayed prey
in the guts of predators using PCR. Molecular Ecology 14,
4147–4158.
Foucher, A. & Wilson, M. (2002) Development of a polymerase
chain reaction-based denaturing gradient gel electrophoresis technique to study nematode species biodiversity using
the 18s rDNA gene. Molecular Ecology Notes 2, 45–48.
Greenstone, M.H. (1979) Spider behaviour optimises dietary
essential amino acid composition. Nature 282, 501–503.
Harper, G.L., King, R.A., Dodd, C.S., Harwood, J.D., Glen,
D.M., Bruford, M.W. & Symondson, W.O.C. (2005) Rapid
screening of invertebrate predators for multiple prey DNA
targets. Molecular Ecology 14, 819–827.
Harwood, J.D., Phillips, S.W., Sunderland, K.D. & Symondson, W.O.C. (2001) Secondary predation: quantification of
303
food chain errors in an aphid–spider–carabid system using
monoclonal antibodies. Molecular Ecology 10, 2049–2057.
Heethoff, M., Etzold, K. & Scheu, S. (2004) Mitochondrial
COII sequences indicate that the parthenogenic earthworm Octolasion tyrtaeum (Savigny 1926) constitutes of two
lineages differing in body size and genotype. Pedobiologia
48, 9–13.
Hickson, R.E., Simon, C., Copper, A., Spicer, G.S., Sullivan, J.
& Penny, D. (1996) Conserved sequence motifs, alignment,
and secondary structure for the third domain of animal 12S
rRNA. Molecular Biology and Evolution 13, 150–169.
Hoogendoorn, M. & Heimpel, G.E. (2001) PCR-based gut
content analysis of insect predators: using ribosomal ITS-1
fragments from prey to estimate predation frequency.
Molecular Ecology 10, 2059–2067.
Jarman, S.N., Deagle, B.E. & Gales, N.J. (2004) Group-specific
PCR for DNA-based analysis of species diversity and
identity in dietary samples. Molecular Ecology 13, 1313–1322.
Juen, A. & Traugott, M. (2005) Detecting predation and
scavenging by DNA gut content analysis: a case study
using a soil insect predator–prey system. Oecologia 142,
344–352.
Kaspar, M.L., Reeson, A.F., Cooper, S.J.B., Perry, K.D. &
Austin, A.D. (2004) Assessment of prey overlap between a
native (Polistes humilis) and an introduced (Vespula germanica) social wasp using morphology and phylogenetic
analysis of 16S rDNA. Molecular Ecology 13, 2037–2048.
Mayntz, D., Raubenheimer, D., Salomon, M., Toft, S. &
Simpson, S.J. (2005) Nutrient-specific foraging in invertebrate predators. Nature 307, 111–113.
McCaig, A.E., Glover, L.A. & Prosser, J.I. (1999) Molecular
analysis of bacterial community structure and diversity in
unimproved and improved upland grass pastures. Applied
and Environmental Microbiology 65, 1721–1730.
McCaig, A.E., Glover, L.A. & Prosser, J.I. (2001) Numerical
analysis of grassland bacterial community structure under
different land management regimes by using 16S ribosomal
DNA sequence data and denaturing gradient gel electrophoresis banding patterns. Applied and Environmental
Microbiology 67, 4554–4559.
Muyzer, G., De Waal, E.C. & Uitterlinden, A.G. (1993) Profiling
of complex microbial populations by denaturing gradient
gel electrophoresis analysis of polymerase chain reactionamplified genes coding for 16S rRNA. Applied and Environmental Microbiology 59, 695–700.
Myers, R.M., Fischer, S.G., Lerman, L.S. & Mantiatis, T. (1985)
Nearly all single base substitutions in DNA fragments
joined to a gc-clamp can be detected by denaturing gradient
gel electrophoresis. Nucleic Acids Research 13, 3131–3145.
Nakatsu, C.H., Torsvik, V. & Ovreas, L. (2000) Soil community
analysis using DGGE of 16S rDNA polymerase chain
reaction products. Soil Science Society of America Journal 64,
1382–1388.
Oelbermann, K. & Scheu, S. (2002) Effects of prey type and
mixed diets on survival, growth and development of a
generalist predator, Pardosa lugubris. Basic and Applied
Ecology 3, 285–291.
Riesner, D., Henco, K. & Steger, G. (1991) Temperature
gradient gel electrophoresis: a method for the analysis of
conformational transitions and mutations in nucleic acids
and proteins. Advances in Electrophoresis 4, 169–250.
Satchell, J.E. (1967) Colour dimorphism in Allolobophora
chlorotica Sav. (Lumbricidae). Journal of Animal Ecology 36,
623–630.
304
G.L. Harper et al.
Sheffield, V.C., Cox, D.R., Lerman, L.S. & Myers, R.M. (1989)
Attachment of a 40-base pair G+C-rich sequence (GC
clamp) to genomic DNA fragments by the polymerase
chain reaction results in improved detection of single base
pair changes. Proceedings of the National Academy of Sciences,
USA 86, 232–236.
Sheppard, S.K. & Harwood, J.D. (2005) Advances in molecular
ecology: tracking trophic links through predator–prey food
webs. Functional Ecology 19, 751–762.
Sheppard, S.K., Henneman, M.L., Memmott, J. & Symondson,
W.O.C. (2004) Infiltration of alien predators into invertebrate food-webs in Hawaii: a molecular approach. Molecular Ecology 13, 2077–2088.
Sheppard, S.K., McCarthy, A.J., Loughnane, J.P., Gray, N.D.,
Head, I.M. & Lloyd, D. (2005a) The impact of sludge
amendment on methanogen community structure in an
upland soil. Applied Soil Ecology 28, 147–162.
Sheppard, S.K., Bell, J., Sunderland, K.D., Fenlon, J., Skervin,
D. & Symondson, W.O.C (2005b) Detection of secondary
predation by PCR analysis of the gut contents of invertebrate generalist predators. Molecular Ecology 14, 4461–4468.
Simms, R.W. & Gerard, B.M. (1985) Earthworms. London, E.J.
Brill/W. Backhuys.
Simon, C., Frati, F., Beckenbach, A., Crespi, B., Liu, H. & Flook,
P. (1994) Evolution, weighting, and phylogenetic utility
of mitochondrial gene sequences and a compilation of
conserved polymerase chain reaction primers. Annals of the
Entomological Society of America 87, 651–701.
Sunderland, K.D. (1996) Progress in quantifying predation
using antibody techniques. pp. 219–455 in Symondson,
W.O.C. & Liddell, J.E. (Eds) The ecology of agricultural pests –
biochemical approaches. London, Chapman & Hall.
Sunnucks, P. & Hales, D.F. (1996) Numerous transposed
sequences of mitochondrial cytochrome oxidase I-II in
aphis of the genus Sitobion (Hemiptera: Aphididae).
Molecular Biology and Evolution 13, 510–524.
Sutherland, R.M. (2000) Molecular analysis of avian diets. PhD
Thesis, University of Oxford.
Symondson, W.O.C. (2002) Molecular identification of prey in
predator diets. Molecular Ecology 11, 627–641.
Taberlet, P. & Fumagalli, L. (1996) Owl pellets as a source
for genetic studies of small mammals. Molecular Ecology 5,
301–305.
Toft, S. & Wise, D.H. (1999) Growth, development and survival
of a generalist predator fed single- and mixed-species diets
of different quality. Oecologia 119, 191–197.
Torsvik, V., Sorheim, R. & Goksoyr, J. (1996) Total bacterial
diversity in soil and sediment communities: a review.
Journal of Industrial Microbiology 17, 170–178.
Uetz, G.W., Bischoff, J. & Raver, J. (1992) Survivorship of
wolf spiders (Lycosidae) reared on different diets. Journal of
Arachnology 20, 207–211.
Walrant, A. & Loreau, M. (1995) Comparison of iso-enzyme
electrophoresis and gut content examination for determining the natural diets of the groundbeetle species Abax
ater (Coleoptera: Carabidae). Entomologia Generalis 19,
253–259.
Zaidi, R.H., Jaal, Z., Hawkes, N.J., Hemingway, J. &
Symondson, W.O.C. (1999) Can the detection of prey
DNA amongst the gut contents of invertebrate predators
provide a new technique for quantifying predation in the
field? Molecular Ecology 8, 2081–2088.
(Accepted 1 February 2006)
Ó CAB International, 2006