TMP E39

NMP5[v.2007/04/19 7:46;] Prn:21/05/2007; 12:39 Tipas:?? F:nmp506.tex; /Austina p.
1 (72-184)
Nematology Monographs & Perspectives, 2007, Vol. 5, 693-733
Chapter 6
Phylogeny and evolution

Byron J. A DAMS, Scott M. P EAT and Adler R. D ILLMAN
Microbiology & Molecular Biology Department, and Evolutionary
Ecology Laboratory, Brigham Young University, Provo,
UT 84602-5253, USA
Introduction
Nematodes originated during the Precambrian explosion over 500

million years ago (Wray et al., 1996; Ayala & Rzhetsky, 1998; Rod-
riguez-Trelles et al., 2002). The phylogenetic position of the Nematoda
relative to other metazoans has historically been one of contention.
Originally circumscribed within the Vermes Linnaeus, 1758 and later the
Aschelminthes Grobben, 1910 (Claus & Grobben, 1910), the Nematoda
are now believed to belong to a clade of moulting animals, the Ecdysozoa
(Aguinaldo et al., 1997), and share a most recent common ancestor with
arthropods, kinorhynchs, nematomorphs, onychophorans, priapulids and
tardigrades.
O RIGINS OF ENTOMOPATHOGENIC NEMATODES (EPN)

Entomopathogenic nematodes of the Steinernematidae and Het-
erorhabditidae are not monophyletic, but likely began independently to
explore biotic relationships with arthropods and Gram-negative enteric
bacteria (Enterobacteriaceae) by the mid-Palaeozoic (approximately 350
million years ago) (Poinar, 1993). Their origins were probably not syn-
chronous and the ages of their respective lineages appear to be signif-
icantly different. Evidence for the disparate origins and relative age of
these Families is illustrated in Figure 240. Assuming even a somewhat
sloppy molecular clock, the long branch lengths of Steinernema, both
within the genus and relative to its most recent common ancestor, imply
that it has been evolving independently of other nematode lineages for a
longer period of time than Heterorhabditis. This could explain, in part,
© Koninklijke Brill NV, Leiden, 2007 693

NMP5[v.2007/04/19 7:46;] Prn:21/05/2007; 12:39 Tipas:?? F:nmp506.tex; /Austina p. 2 (184-184)
B.J. Adams et al.
Fig. 240. Phylogenetic tree of 434 representative taxa from the Rhabditida,
Plectida and Monhysterida. The tree was generated from 18S rDNA sequences
by the neighbour joining BioNJ algorithm using log determinant transformed
distances in order to account for rate and nucleotide usage heterogeneity
among lineages. EPN branches are in bold.
694 Nematology Monographs & Perspectives

the differences in species richness between the two genera (i.e., Het-
erorhabditis has had less time to generate species). However, if EPN
in general have been coevolving with their insect hosts for 350 mil-
lion years, why do we not see numbers of entomopathogenic nematodes
equal to the species richness of their insect hosts? Even if the number
of species of insect hosts were whittled down to only those that have
a soil-borne larval stage, the host-parasite species deficit would still be
tremendous.
O RIGINS OF EPN SPECIES
Little is known of the modes of speciation that have produced the

extant biodiversity of EPN. Because most EPN sister species appear
to be (or have been) geographically isolated (Hominick et al., 1996;
Hominick, 2002), vicariant allopatric speciation, in most cases followed
by dispersal, is the most likely mode of speciation to explain their
distribution (Lynch, 1989). There are several hypothetical reasons as
to why we might predict that there should be more species of EPN
than have been described to date. First of all, there appears to be
a correlation between species size and the probability of extinction
(Jackson, 1974; Jablonski, 1987). Species with large range sizes, such
as bison and woolly mammoths, require more resources and space and,
therefore, have a lower probability of speciating and a higher probability
of extinction. By contrast, animals such as EPN that have a small body
size may be able to partition more niche space per unit area (Hutchinson
& MacArthur, 1959; Morse et al., 1985). Therefore, because nematodes
have small spatial requirements, they should have higher speciation rates
and greater niche filling or niche partitioning opportunities. Just such
partitioning is evident in studies of habitat preference (Sturhan, 1999)
and could be maintained phylogenetically (Spiridonov et al., 2004).
Fig. 240. (Continued). The upper bold lines depict the lineages of Steinernema
carpocapsae and S. glaseri relative to their most recent common ancestor. The
lower bold lines illustrate the branching point of Heterorhabditis marelatus
and H. bacteriophora relative to their most recent common ancestor. Note the
long branch length of the steinernematid lineage relative to the heterorhabditid
lineage, and the long branch lengths of the steinernematid species relative to
the heterorhabditid species. The multiple sequence alignment for this clade was
retrieved from NemaTOL (nematol.unh.edu).
Vol. 5, 2007 695

B.J. Adams et al.
However, nematodes could have high rates of dispersal, as evidenced

by the inter- and transcontinental geographic ranges of many species
(Adams et al., 2006); their effective range requirements could be vast,
effectively decreasing the probability of speciation. In the case of
EPN, their small body size, coupled with an association with mobile
insect hosts, could facilitate dispersal and gene flow across expansive
geographic ranges, resulting in retarded speciation rates or extinction
probabilities comparable to those of megafauna (but see Blouin et al.,
1999).
Additional explanations for the patterns of EPN biodiversity are legion
but the rate of molecular evolution can probably be ruled out, as it has
been shown that for multiple genes nematodes evolve at anomalously
high rates relative to other metazoans (e.g., Wray et al., 1996). Rather,
mechanisms of host specificity and virulence may exert greater influence
on the origin of species (or lack thereof) in these genera than other
factors (Figs 241, 242). For example, if the nematode depends on the
survival or extended mobility of its insect host to vector it to other
potential hosts (offspring, mates, etc.), then there will be strong selection
pressure to reduce virulence because if the host is killed too quickly
the nematode will have extinguished the resources or transmission
mechanisms required to continue its life cycle (net positive differential
reproduction). Reliance on the host as a vector should favour nematode-
insect host specificity because the nematode must closely track the
evolutionary trajectory of its host. According to this scenario, as the
insect speciates, the nematode species must also adapt to changes in
its host that involve the nematode’s ability to maintain a symbiotic
relationship. This would result in tight host-tracking and a longer-term
relationship between an individual nematode and its insect host that leads
to decreased opportunities for gene flow among other members of its
population that have a broader host range. Such a scenario promotes
evolutionary fidelity between the nematode and insect host such that if
the insect has a high rate of speciation, the nematode must also speciate
rapidly as it tracks its host over time. Thus, entomogenous nematodes
with low virulence are predicted to have simultaneously narrower host
ranges, lower cophylogenetic fidelity, lower rates of gene flow, and
comparable rates of speciation relative to those of their insect host.
Under this scenario, we would expect strong selection for the evolution
of avirulence. Production of toxins or reproduction would increase only
to the point that it hindered host mobility or longevity (Fig. 241).

Fig. 241. Selection between hosts, where inclusive fitness is a function of

host mobility/survival. When insect host mobility or longevity is required for
successful completion of the nematode life cycle, EPN/B virulence, as measured
by either toxin production or reproduction, increases at a maximal rate to
establish infection and reproduction, but tapers off as it begins to influence host
survival. Such a scenario favours the evolution of increasing avirulence, narrow
host ranges, increased nematode-insect host fidelity (cophylogeny), decreased
gene flow among nematode populations, and increased rates of speciation.
Fig. 242. Selection within hosts, where inclusive fitness is a function of the
rate of reproduction in the host. Competition for resources among different
pathogens in an insect host can lead to selection pressures that favour
increased virulence. Accordingly, virulence, as measured by toxin production
or reproduction, increases at a maximum rate culminating with the rapid death
of the host insect. Such a scenario favours the evolution of increasing virulence,
broad host ranges, decreased host fidelity (cophylogeny), increased gene flow,
and decreased rates of speciation.
Vol. 5, 2007 697

B.J. Adams et al.
Relative to other families of nematodes, or clades for that matter (Figs

240, 243), the Steinernematidae and Heterorhabditidae contain relatively
few species. This may be due, in part, to the discrepancy between the
number of EPN species that currently exist and the number of EPN
species that have been described (Liu et al., 2000). Taxonomic efforts
aside, it may be that the highly virulent nature of these nematode-
bacterium complexes works to retard speciation in these nematodes. For
example, the rapid rate at which the bacterium kills its insect host may
have evolved in response to the highly competitive environment inside
the host cadaver where bacterial colonies are under strong selection
pressure to be as virulent as possible. Considering that the insect
represents a rich carbon resource for soil microbes and the tremendous
diversity of soil microbes positioned to utilise it, individual bacterial
colonies that can get into the insect, utilise cadaver resources and
replicate the fastest have the greatest chance of being vectored into
the next insect by the emergent IJ. Under these conditions, natural
selection for high rates of within-host competition results in virulence
which continually increases to an optimum that is tempered only by the
rate at which the nematode can evolve tolerance to the bacterial toxins
(Fig. 3). The rapid death of the host obviates its ability to vector the
nematodes (and their endosymbiotic bacteria) back to insect host nests,
breeding sites, mates or offspring. The result is that as the infective
juveniles emerge from their host’s cadaver, they may be equally, if
not more, likely to encounter a potential insect host that is a different
species to the one from which they just emerged. Thus, the increase
in virulence that is generated by competition within the host reinforces
the requirement for a broad host range, and vice versa. Similarly, high
virulence and broad host ranges decrease the requirement for nematodes
to establish host-specific relationships. Subsequently, as hosts speciate,
there is no selective pressure on the nematodes to do the same. Thus,
nematodes that are highly virulent and have broad host ranges will tend
to be less speciose, so long as dispersal is high and barriers to gene
flow remain relatively inconsequential. Nematodes that have narrow host
ranges and lower virulence should in general be more species rich. The
prediction that an increase in host range is concomitant with increased
gene flow is likely countered by the rate of mortality of infected hosts
(low mobility) and the insularity of the soil environment (Blouin et al.,
1999). The effect (evolution of increasing virulence) is crudely reflected
across larger phylogenetic space in Figure 243. That virulence could

Fig. 243. Species richness of select entomogenous nematode clades. Het-

erorhabditis and Steinernema have relatively few species, broad geographic
ranges, little host fidelity and high virulence. Although it is predicted that an
increase in host range is concomitant with increased gene flow, this is likely
countered by the rate of mortality of infected hosts (low mobility) and the in-
sularity of the soil environment (Blouin et al., 1999). Note also the influence of
sampling effort (some groups have been investigated more intensively than oth-
ers) and species richness increases over evolutionary time (earlier branching
taxa are also more speciose). Tree modified from Blaxter et al. (1998).
Vol. 5, 2007 699

B.J. Adams et al.
Fig. 244. Plot of times to coalescence in units of genetic distance for each pair
of Heterorhabditis-Photorhabdus cospeciation events. The coalescence time is
the time two lineages last had a common ancestor. On an ultrametric tree this
corresponds to the distance between the ancestral node and any one of its
descendants (5A). Given ultrametric Heterorhabditis and Photorhabdus trees
(not shown), plots of the coalescence time of each pair of cospeciation nodes
(5B) reveal if the times to coalescence are correlated (they are: r = 0.82;
P < 0.05) and also whether the timing of the codivergence events are pre-
emptive (plotted line crosses a positive y-axis; Photorhabdus diverges earlier),
synchronous (line crosses at the origin; they diverge at the same time) or
delayed (line crosses a negative y-axis; Heterorhabditis diverges earlier (Page
& Hafner, 1996). The number pairs of the plots refer to the paired nodes of
the Heterorhabditis-Photorhabdus trees, respectively. The positive slope in the
figure suggests pre-emptive evolution of the bacterium followed by a tracking
response of the nematode host.
be maintained, and even increased over evolutionary time, counters the

widely held view that virulence is largely only manifest in the initial
stages of symbiosis, and that the pathogenicity of the parasite decreases

Fig. 245. Phylogenetic position of Steinernema relative to 20 other most closely

related nematodes for which comparable sequences are publicly available.
Numbers on branches indicate node support (bootstrap, 1000 BioNJ reps)
mapped on a tree produced with SSU sequences using the BioNJ neighbour-
joining algorithm (Gascuel, 1997). Note sister taxon relationship with sampled
Strongyloidoidea and Panagrolaimidae; see also Figure 243 and Nadler et al.
(2006).
over time. Work has shown that the evolution of virulence can be highly
plastic and driven by adaptive virulence-associated responses by both
host and parasite (Perlman & Jaenike, 2003), that often these responses
can involve increases in virulence (Herre, 1993) and are correlated with
a lack of concerted host-parasite coevolution (Perlman et al., 2003). In
Heterorhabditis a piece of tenuous evidence supports the notion that the
nematode is tracking changes that first occur in its bacterial symbiont.
When times to coevolving coalescent events (time since two lineages
shared a common ancestor)between species of Heterorhabditis and their
bacterial endosymbionts Photorhabdus are plotted in terms of genetic
distance, the correlation is significant (r = 0.82; P < 0.05) and
the slope crosses the y axis positively, suggesting that the nematode
is exhibiting a delayed response to evolutionary diversification of its
symbiont even though the gene used to reconstruct the phylogenetic
history of Heterorhabditis (ITS) is evolving at a faster rate than the
gene used to reconstruct Photorhabdus phylogeny (16S) (Fig. 245).
Vol. 5, 2007 701

B.J. Adams et al.
Such analyses provide insight into the degree and mechanisms by which
nematode and endosymbiont can influence each other’s evolutionary
trajectory. The small sample sizes and somewhat crude analyses caution
that it is still preliminary to suggest that much, if anything, at this point
can be inferred with confidence about correlations between the evolution
of virulence, coevolution, host-range, gene flow and biodiversity, but
EPN are well positioned to serve as model organisms for comparative
and experimental research programmes in this area.
Phylogeny
Numerous molecular approaches have been applied to the evolution-

ary relationships of EPN. Initial efforts included PCR RFLP analyses
of the internal transcribed spacer (ITS) regions of the ribosomal DNA
tandem repeat unit (Reid, 1994; Reid et al., 1997). Other approaches in-
cluded combinations of random amplified polymorphic DNA (RAPD)
and morphological data (Liu & Berry, 1996). These papers still contain
valid data, although their interpretations have been questioned due to an-
alytical drawbacks, primarily insufficient numbers of phylogenetically
informative characters, tenuous homology statements and tree build-
ing methods that are suboptimal for reconstructing evolutionary history
(Stock et al., 2001). While the use of morphological characters in phy-
logenetic analyses remains robust, the performance of RAPD markers
has fallen out of favour, primarily because of problems associated with
replication and sensitivity of the approach.
The ribosomal small subunit (SSU, or 18S gene) was the first to
be sequenced in an effort to resolve phylogenetic relationships within
and between Steinernema and Heterorhabditis (Liu et al., 1997). While
variation among species in both genera could be identified, the particular
5 segment of the gene that was sequenced and analysed is evolving
too slowly to produce the quantity of phylogenetically informative
characters (substitutions) needed appropriately to infer relationships
among species in their respective genera. Subsequently, the ribosomal
large subunit (LSU), ITS and mitochondrial NADH dehydrogenase
subunit 4 (ND4), and cytochrome oxidase I gene (Cox1) and small
mitochondrial ribosomal RNA gene (16S) have been used to reconstruct
the phylogenetic relationships among EPN (Adams et al., 1998; Liu
et al., 1999; Szalanski et al., 2000; Stock et al., 2001; Nguyen et al.,

2001, 2004; Saeb et al., unpubl.). As the SSU is too conserved to be

informative of relationships among most EPN species, the ITS, 16S,
COI and ND4 appear to be evolving more rapidly than is optimal for
phylogenetic resolution at this taxonomic level (Adams et al., 1998;
Liu et al., 1999; Szalanski et al., 2000; Nguyen et al., 2001) where
intraspecific and even intraindividual variability, although rare (see
Adams et al., 1998; Spiridonov et al., 2004), exists. Their fairly rapid
rate of evolution means that the ITS, 16S, COI and ND4 genes perform
well at revealing relationships among closely related species, and even
populations (Adams et al., 1998; Spiridonov et al., 2004) but they are
unable to resolve more basal nodes with confidence (Adams, 2001;
Nguyen et al., 2001; Spiridonov et al., 2004). Over its entire length, the
LSU is highly conserved, but within it are stretches of expansion regions
that have relatively higher substitution rates than the gene overall. The
LSU (Duncan et al., 1999; Kaplan et al., 2000) D2D3 segment is one
such expansion region that has been shown to evolve more slowly than
the above mentioned genes but more rapidly than the SSU, making it (in
most cases) more appropriate for resolving more deeply branching nodes
in phylogenetic trees, while at the same time being sensitive enough to
reveal differences among most species (De Ley et al., 1999; Duncan
et al., 1999; Kaplan et al., 2000; Nadler et al., 2000; Tenente et al.,
2004; He et al., 2005). This property also makes it, along with some
of the more variable regions of the SSU, an appealing tool for species
delimitation (Adams, 2001; Nadler, 2002; Sites & Marshall, 2004; Gozel
et al., 2006b) and for use as a molecular barcode (Powers, 2004; Blaxter
et al., 2005).
Whilst phylogenetic analyses of these genes appears to be straightfor-
ward, the ND4 and COI mitochondrial genes are protein coding genes
that undergo selection to maintain an open reading frame, whereas the
ribosomal genes are non-protein coding, and are thought to be under less
selective constraint to maintain their length (Hillis & Dixon, 1991) (but
note that there is some apparent selection pressure to maintain certain se-
quence and size fidelity for the proper maturation of the complete subunit
(Van Nues et al., 1994)). Therefore, while insertion-deletion events (in-
dels) are infrequent in the protein-coding genes, they are often the most
common form of substitution in the ITS regions of EPN (Adams et al.,
1998; Nguyen et al., 2001). Although robust models of nucleotide sub-
stitution have been developed for protein coding genes that incorporate
different substitution parameters as a function of codon position, similar
Vol. 5, 2007 703

B.J. Adams et al.
models for rDNA (i.e., that can account for indels and differential rates of
substitution in stems vs loops of the RNA secondary structure) await full
development. This is a problem for parsimony and model-based analy-
ses. At the moment, the stem-loop issue can be addressed by determining
the empirical transition/transversion ratios in stem and loop partitioned
datasets and then analysing them under mixed model assumptions (Ron-
quist & Huelsenbeck, 2003) but see also (He et al., 2005). Indels can be
accounted for by gap coding (Adams et al., 1998; Muller, 2006).
Optimal analyses of EPN phylogenetic relationships will incorporate
every type of character for which homologous states can be identified
across as many taxa as possible (Kluge, 1998, 2004; Giribet et al., 2001).
However, the analyses of Liu and Berry (1996), Stock et al. (2001)
and Spiridonov et al. (2004) remain the most deliberate attempts to
integrate morphological and molecular characters in a cohesive analysis.
As with all phylogenetic analyses, ongoing taxonomic work renders
every previous solution obsolete as soon as it is published. Below
are analyses that sample the taxa and comparable DNA sequence data
currently available in public databases.
P HYLOGENY OF STEINERNEMA SPECIES
Spiridonov and Belostotskaya (1983) suggested that morphological

features of Steinernema imply a most recent common ancestry with
strongyloidids and alloionematids. Poinar (1993) subsequently inferred
from morphological and bionomic characters that the Steinernematidae
might have evolved from a proto-Rhabditonema-like ancestor. Subse-
quent analyses of SSU rDNA depict the Steinernematidae sharing com-
mon ancestry, if not sister taxon to the Strongyloidoidea and Panagro-
laimidae (Blaxter et al., 1998, 2000) (Figs 243, 245).
Phylogenetic relationships among species of Steinernema have been
carried out using morphological characters and RAPDs (Liu & Berry,
1996), PCR-RFLP of the ITS region (Reid, 1994; Reid et al., 1997),
and DNA sequences of the SSU (Liu et al., 1997), ITS, COII and 16S
(Szalanski et al., 2000), ITS (Nguyen et al., 2001; Spiridonov et al.,
2004) and LSU (Stock et al., 2001) genes. In the present analysis, we
conservatively reconstruct phylogenetic relationships among nominal
taxa using LSU sequences available in public databases (Fig. 246).
Using the profile alignment mode of Clustal X (Thompson et al., 1997),
individual sequences were aligned to the most complete alignment of

Fig. 246. Phylogenetic relationships among species of Steinernema based on

LSU rDNA sequence data, summarising Bayesian, likelihood and parsimony
analyses. Values above each branch represent Bayesian posterior probabilities,
parsimony bootstrap indices (1000 replicates), where concordant with the
Bayesian analyses, appear below. Clade designations (I-V) follow Spiridonov
et al. (2004).
Stock et al. (2001) available in TreeBase (Sanderson et al., 1994). For

the parsimony analyses all characters were considered unordered and
unweighted. Of 1159 total aligned characters, 538 are constant and 438
are phylogenetically informative. One hundred eighty-three characters
Vol. 5, 2007 705

B.J. Adams et al.
varied but are likely phylogenetically uninformative autapomorphies.

Gaps were treated as missing data. Starting trees for the heuristic search
were obtained by stepwise addition with a random addition sequence,
with TBR branch-swapping. Parsimony bootstrapping was carried out
the same way, but with 1000 replicates.
Maximum likelihood searches were carried out based on the
GTR + G + I model of nucleotide evolution (Posada & Crandall,
1998) with six substitution types and substitution rate matrix and nu-
cleotide frequencies estimated from the empirical data (transformation
series). The model assumed that approximately 15% of the sites (homol-
ogous nucleotides) were invariable, with the variable sites approximat-
ing a gamma distribution (shape parameter 0.6213, four rate categories).
Heuristic searches were carried out the same way as described above for
the parsimony analyses.
Bayesian analyses were performed using MrBayes 3.1 (Huelsenbeck
& Ronquist, 2001) based on the same models used for the ML searches.
For each analysis we carried out runs for 7 000 000 generations with
four incrementally heated chains and then sampled at intervals of 1000
generations to include 7000 data points. Stationarity was estimated by
plotting (with Microsoft Excel 2004) the log likelihood scores against
generation time and assumed stationarity when the curves flattened
out. This phase was reached between 5000 and 6000. The first three
to six trees (burn-in) were discarded and 50% majority rule trees
were obtained from the remaining 6994-6995 data points with the
purpose of obtaining the posterior probability values. To avoid local
entrapment on a suboptimal peak in the tree space, we performed two
independent analyses and compared these for convergence to similar
log likelihood mean values (Huelsenbeck & Bollback, 2001; Leache
& Reeder, 2002). We also compared the posterior probabilities for
individual clades obtained from the separate analyses for congruence to
ensure convergence of the two analyses. All phylogenetic analyses were
performed on a RackSaver 64 node dual Opteron processor computing
cluster. The strength of support for nodes in the Bayesian tree is
estimated by posterior probabilities. For Bayesian analyses, posterior
probabilities greater than 0.95 are typically considered robust (Leache &
Reeder, 2002), 0.70 for maximum parsimony and maximum likelihood
bootstrap analyses (Hillis & Bull, 1993).
The phylogeny presented in Figure 246 presents few deviations from
previous hypotheses (Nguyen et al., 2001; Stock et al., 2001; Spiridonov

et al., 2004) but includes several taxa previously unrepresented in

comprehensive phylogenies of the genus, namely: S. guangdongense, S.
hermaphroditum, S. aciari, S. leizhouense, S. yirgalemense, S. websteri,
S. anatoliense, S. akhursti and S. beddingi. Steinernema pakistanense,
S. weiseri, S. tami and S. neocurtillae are sampled in the analyses of
Spiridonov et al. (2004) but are not included in the present analysis for
lack of LSU sequence data for these taxa. Steinernema pakistanense
most likely nests within clade IV, S. weiseri within clade III and
S. tami within clade II. The position of S. neocurtillae within the
Steinernematidae can be inferred to diverge somewhere between clades
I and II, but its placement is one of the most enigmatic (in terms of
support) of all the taxa in the genus (Nguyen et al., 2001; Spiridonov et
al., 2004). Another problematic taxon is S. rarum (see discussion below).
The addition of taxa to a dataset is typically seen as more important than
adding more characters, particularly where there may be problems of
long-branch attraction (LBA) (Graybeal, 1998; Poe, 1998). However, the
only potential LBA event in Steinernema involves the branch between
the species in clade II and the rest of the genus, and none of the additional
taxa appears to intersect this branch.
The overall structure of the Steinernema tree is not completely
resolved without some ambiguity; clades II, IV and V all contain taxa
with weakly supported relationships, and S. rarum appears not to belong
to any predetermined clade at all. Spiridonov et al. (2004) placed this
taxon as sister to the remaining members of clade V, Stock et al. (2001)
depict this taxon as sister to clades III and IV, and the present analyses
suggest it is more likely the sister taxon to clades III, IV and V.
One of the earliest morphological trends to emerge from mapping
character traits on phylogenetic trees was the length of the infective ju-
veniles where there appeared to be clades of ‘long’ and ‘short’ morphs
(Reid, 1994; Reid et al., 1997) but, as analyses have since become in-
clusive of increasing numbers of taxa, this character has been shown to
be fairly plastic and homoplasious (Stock et al., 2001). Although Stock
et al. (2001) suggest that the majority of morphological characters com-
monly used for classification and phylogenetic relationships are either
homoplastic or plesiomorphic, Spiridonov et al. (2004) suggested that
the structure of bacterial vesicles, colour of male copulatory structures,
position of the excretory pore and sperm morphology could be useful
synapomorphies for future phylogenetic analyses. Detailed information
on steinernematid morphological characters is provided elsewhere (see
Vol. 5, 2007 707

B.J. Adams et al.
Chapter 3), but in the context of the present phylogeny it would be inter-
esting to see if, for example, S. beddingi has yellow to colourless spicules
and whether S. guangdongense and S. hermaphroditum lack a true bac-
terial vesicle as predicted by Spiridonov et al. (2004). It is also possi-
ble that S. websteri and S. anatoliense actually have horn-like cephalic
structures (in contradiction to their original descriptions), although this
observation requires examination of exsheathed infective juveniles as it
is easily overlooked (Nguyen & Adams, 2003).
P HYLOGENY OF HETERORHABDITIS SPECIES
Poinar (1993) suggested that, based on morphological, physiological,

distributional and biological evidence, Heterorhabditis most likely
shared a most recent common ancestor with a Pellioditis-like ancestor.
This argument is further corroborated by SSU gene genealogy (Fig. 247).
Virtually all molecular systematic approaches to relationships among
species and populations of Heterorhabditis have been done using the
ITS rDNA gene (Adams et al., 1998, 2006; Phan et al., 2003; Nguyen et
al., 2004, 2006), although the ND4 gene has also been used effectively
(Blouin et al., 1999; Liu et al., 1999). In the present analysis, we
conservatively reconstruct phylogenetic relationships among nominal
taxa using ITS and ND4 sequences available in public databases.
Accordingly, individual ITS sequences were aligned using the profile
alignment mode of Clustal X to the most complete alignment of Adams
et al. (1998). Alignment of the ND4 sequences was straightforward as
there were no insertion or deletion events.
For the parsimony analyses of the ITS sequences, all characters were
considered unordered and unweighted. Of 1167 total aligned characters,
533 were constant and 389 were phylogenetically informative. Some
243 characters varied, but were probably phylogenetically uninformative
autapomorphies. Gaps were treated as missing data. Starting trees for
the heuristic search were obtained by stepwise addition with a random
addition sequence with TBR branch swapping. Parsimony bootstrapping
was carried out the same way, but with 1000 replicates. Likelihood
analyses were run using the GTR-G model selected by AIC (Posada
& Crandall, 1998; Posada & Buckley, 2004) (six substitution types,
empirical rate matrix and nucleotide frequencies with no invariable sites
and a gamma distribution of rate shape parameter of 0.9 across four

Fig. 247. Phylogenetic position of Heterorhabditis relative to 35 other most

closely related nematodes for which comparable sequence information is
publicly available. Numbers on branches indicate node support (bootstrap,
1000 reps) mapped on a tree produced with SSU sequences using the BioNJ
neighbour-joining algorithm (Gascuel, 1997). Note sister taxon relationship
to the Filaroides and Strongyloidea; see also Figure 243 and Hoglund et al.
(2003).
Vol. 5, 2007 709

B.J. Adams et al.
rate categories). Starting trees were obtained by stepwise addition with

random addition sequence and TBR branch-swapping.
For the parsimony analyses of the ND4 sequences, of 495 total
characters 173 were informative; 257 were constant, and 65 varied but
were parsimony-uninformative. Tree search strategy was the same as that
described above for the ITS analyses. Likelihood analyses of the ND4
data matrix assumed a GTR + I + G model of sequence evolution, as
estimated by ModelTest (Posada & Crandall, 1998). Accordingly, six
substitution types, their rate matrix and nucleotide usage frequencies
were estimated empirically from the multiple sequence alignment.
Approximately 4% of the inferred homologous sites were considered
to be invariant, with the distribution of rates at the variable sites
approximating a gamma distribution with a shape parameter of 0.36 and
four rate categories (mean rate for each category). Bayesian analyses
for the ITS and ND4 regions employed the same models of sequence
evolution as were utilised in the ML analyses and proceeded as described
above for the Steinernema LSU analyses.
There is general topological agreement between the two trees except
for the placement of H. zealandica and H. marelatus to sister taxa
H. downesi and H. megidis (Figs 248, 249). The ITS solution depicts
H. marelatus as sister to H. downesi + H. megidis, whereas the ND4
solution places H. zealandica as sister taxon to this clade. The node
associated with this discrepancy is tenuously supported by Bayesian
posterior probability and parsimony bootstrapping and is consistent with
previous analyses (Adams & Nguyen, 2002), which also document the
discordance between the present ND4 tree topology and that of Liu et al.
(1999).
Species delimitation
D ELIMITATION OF NEMATODE SPECIES

The body of literature surrounding species and species concepts is
amongst the largest in biology. Much progress has been made, both
in terms of determining what species are (ontology) and how best to
find them (epistemology) but, upon inspection, the majority of metazoan
species descriptions published over the last 10 years do not readily reflect
the progressive trend. This is not surprising. The language of the best
ideas that this literature has to offer is dense and burdensome to read,

Fig. 248. Phylogenetic relationships among species of Heterorhabditis based

on ITS sequences, summarising Bayesian, maximum likelihood and parsimony
analyses. Values above each branch represent Bayesian posterior probabilities.
Parsimony bootstrap indices (1000 replicates) appear below (where trees are
concordant).
Fig. 249. Phylogenetic relationships among species and some populations of

Heterorhabditis based on ND4 mtDNA sequence data, summarising Bayesian,
maximum likelihood and parsimony analyses. Values above each branch
represent Bayesian posterior probabilities, parsimony bootstrap indices (100
replicates), where concordant, appear below.
Vol. 5, 2007 711

B.J. Adams et al.
if not outright impenetrable – for example, a sure cure for insomnia is

Adams (1998). The result is that, despite the philosophical and analytical
tools that have been developed to recover and represent evolutionary
history accurately, few taxonomists are actually taking advantage of
them. Few species descriptions are explicit as to what notion of species
they are attempting to delimit, or the analytical tools they are using to
do so. The confusion surrounding definitions of species is not unlike the
definition of what is ‘obscene’ under United States law. In 1964, Justice
Potter Steward summarised the court’s position as indefinable, but that
“I know it when I see it . . .” (Anon., 1964). As applied to taxonomy,
this problem defers to the common quip that a species is “what a good
taxonomist says it is” (Crum, 1985). Not state of the art, but status quo
nonetheless. Thus it appears “[taxonomists] defend the status quo long
past the time when the quo has lost its status” (with apologies to Peter;
Peter & Hull, 1969).
Analytical methods for morphological and molecular data types
vary somewhat, but there is no conceptual difference for their use
in systematics and conflicts between morphological and molecular
approaches have been overemphasised. The divisions that do exist are
being bridged rapidly by research in developmental and molecular
genetic processes and systematic theory. Both types of data can be
used to identify, diagnose and delimit species. As with all scientific
endeavours, systematic hypotheses (including the validity of species
and their phylogenetic relationships) are supported or rejected based
on the preponderance of evidence. The most robust hypotheses are
those that withstand falsification by the greatest amount of corroborative
evidence – whether molecular, morphological or behavioural. The only
prerequisite for systematic data (characters) is that they have a heritable
genetic basis, and demonstrate the level of fixation and variation
appropriate to resolve the question at hand.
The identification and diagnosis of species is often confused with
species delimitation, or the process of testing the hypothesis that the
entity in question is a species. Identifying species involves the recog-
nition of differences and similarities among described taxa. Diagnosis is
merely the description of the taxon. Less intuitively, species delimitation
involves making a determination and prediction about the history and
fate of an evolutionary lineage. The theoretical difference between de-
limitation and diagnosis is that the former results in something that has
meaning in a historical, evolutionary context. A diagnosis cannot make

the same claim. In practice, species delimitation differs from diagnosis in

that it requires character argumentation, or determining whether homol-
ogous characters are plesiomorphic (primitive) or apomorphic (derived).
In contrast, diagnosis consists of simply determining whether homolo-
gous characters are ‘similar’ or ‘different’. It is crucial that systematists
make this distinction because the two activities are as disparate as saying,
“this entity looks different” and “this entity is a species” – see Kiontke
et al. (2002) and Sudhaus and Koch (2004) for clear examples of this
exercise.
The most salient theoretical definition of species is the evolutionary
species concept. According to this concept, a species is “an entity
composed of organisms that maintains its identity from other such
entities through time and over space, and which has its own independent
evolutionary fate and historical tendencies” (Wiley & Mayden, 2000).
The best way to find species is to identify independent evolutionary
lineages. There are numerous ways to achieve this, depending on the
biological properties of the nematodes, and the types of data that are
accessible. The particular mechanics of these methods are described as
those that emphasise the tokogeny-phylogeny interface (i.e., incorporate
inferences about gene flow) (Sites & Marshall, 2003). The gist of these
approaches is that if a lineage (species) is evolving independently, it will
accumulate fixed changes among its populations that do not appear in
any other lineage. Conversely, shared or variable character states are
indicative of non-exclusive reticulation and reflect genealogical patterns
consistent with outcrossing populations. A cartoon illustrating these
principles is presented in Figure 250.
D ELIMITATION OF EPN SPECIES
Although the first EPN species was described in 1923, over 80% of
them have been described since 1990 (Adams et al., 2006). The recent
increase in species descriptions is driven primarily by the potential for
these species to find applications in biological control, but they are
quickly becoming the focus of basic biological questions of coevolution
(ffrench-Constant et al., 2003; Ciche et al., 2006) and ecology (Lewis et
al., 2006), which also spurs the search for biodiversity.
Thus far, virtually all species of entomopathogenic nematodes have
been delimited in Linnaean (phenetic) fashion (Mayr, 1963) based on
diagnosable differences in morphology, morphometrics, and even mole-
Vol. 5, 2007 713

B.J. Adams et al.
Fig. 250. Examples of discovery operations required to delimit species.

Character states are for illustrative purposes only and are not meant to depict
those actually found in entomopathogenic nematodes.

cules. In some cases crossbreeding tests are done to examine potential

reproductive isolation to clarify whether the criteria of biological species
can be met (Mayr, 1964). However, despite the thoroughness of these de-
scriptive investigations, delimiting species solely on the basis of overall
similarity or potential to interbreed can misrepresent the actual number
of unique evolutionary species that exist in nature and their inferred his-
torical relationships. For example, delimitation in the same group based
on reproductive compatibility can result in a situation where two sister
populations can be reproductively isolated, but each is interfertile with
a more distantly related population (Adams et al., 1998). Delimitation
based on overall similarity or reproductive compatibility betrays histor-
ical relationships and probably fails to accurately reflect the number of
species that actually exist (Adams, 1998).
Adams (1998) argued that lineages that evolve unique character states
show evidence of lineage independence. Identifying unique characters
(autapomorphies) requires character argumentation (polarisation) via
outgroup comparison with all the other known members of the group.
The requirement that species possess autapomorphies protects against
misrepresenting the actual number of species and ensures that all
valid species are consistent with evolutionary history (Adams, 1998).
Spiridonov et al. (2004) take exception to this point. As evidence,
they point to instances where their molecule of choice (ITS) failed to
yield autapomorphies for either of two sister taxa (in this case S. karii
and S. siamkayai). This leads to their conclusion that “autapomorphies
for some well established species may disappear as the number of
species studied increase” and that therefore the utility of autapomorphies
for species delimitation is not a “sound procedure” (see Spiridonov
et al., 2004, pp. 560-561). We disagree that sampling additional
Fig. 250. (Continued). A: Demonstration of lineage exclusivity (independent

evolution of autapomorphic character states), evidence of two species; B:
Sampled populations are polymorphic, no evidence of lineage independence; C:
Diagnostic variation exists in the form of a unique combination of characters,
but taxon Ψ does not show evidence of independent evolution (is a privative
group). D-F: The reason a single multi-state character cannot be used to delimit
two species. D: Character states for a fixed, multi-state character appear to
be partitioned exclusively among taxa Ψ and γ , but there are two equally
logical paths (E, F) that are operationally indistinguishable from the scenario
presented in C.
Vol. 5, 2007 715

B.J. Adams et al.
loci from additional populations and analysing them with an eye

for autapomorphies implies that Adams’ approach is flawed or will
result in decreased systematic accuracy. Nor are we troubled by the
fact that additional data might suggest that what we think are two
‘well established’ species in reality represent two or more populations
of a single species. Where morphological data suggest otherwise,
the addition of molecular characters usually supports the original
species descriptions (Nadler et al., 2000) or sheds light on cryptic
speciation (Gozel et al., 2006a). Also, if the two species are indeed well
supported by other characters (autapomorphies?), then the absence of
autapomorphies in the ITS data is irrelevant because there is no a priori
reason to believe that the molecular characters are somehow superior to
the morphological characters used in the original or subsequent species
descriptions. As an alternative to phylogenetic methods for delimiting
species, Spiridonov et al. (2004) suggest that sequence divergence is
a better indication of lineage independence. Unfortunately, sequence
divergence cannot reveal lineage independence and, therefore, is a
suboptimal tool to find independent lineages (Farris, 1980, 1981, 1983).
The use of sequence divergence alone to delimit species is arbitrary; no
more informative of lineage divergence than say, body length divergence.
It is a poor indicator of species boundaries (Ferguson, 2002). We caution
those who wish to interpret them in this manner.
Molecular methods
The allure of molecular data for work in systematics is their clear

genetic basis, a compelling advantage given the significant amount
of environmental and host-induced morphological variation displayed
by entomopathogenic nematodes (Nguyen & Smart, 1996; Hominick
et al., 1997). Potentially sensitive, objective and powerful, molecular
characters can also be misleading when wielded carelessly. The use of
common genetic markers for EPN research, together with a discussion
of the relative merits of using DNA sequencing methods as opposed to
RAPD, PCR-RFLP and AFLP markers, has been addressed previously
(Adams & Nguyen, 2002; Adams et al., 2006). Below we provide
some basic protocols for acquiring DNA sequence data for use in the
molecular systematics of EPN.

DNA EXTRACTION
DNA extraction from individual nematodes

For most applications, it is important that DNA template used for
RFLP or DNA sequencing come from individual nematodes because
variation within and between individual animals must be accounted for.
This is especially important when working with variable markers that are
informative at the species boundary. Individuals of any sex or stage may
be used, but large females commonly yield the largest amounts of DNA.
DNAzol® DNA extraction (modified from Steve Nadler, pers. comm.)

Place a single nematode in a 0.2 ml PCR tube with no more than 5 µl
of water. Add 100 µl of the digestion solution to the tube and incubate
overnight at 56◦ C, vortexing the tube occasionally. Check tissues for
digestion using a dissection or other low-powered light microscope. If
undigested, add 1 µl proteinase K and incubate longer (eggs will not
usually digest). ‘Pump-mix’ digested tissues. Centrifuge the sample for
5 min at 10 000 g in a microfuge.
Use a pipette to remove ca 90 µl of the supernatant and place it in a
new 0.2 ml PCR tube (the important thing is to leave the last ca 10 µl
containing the cellular debris at the bottom of the tube to be discarded).
Heat kill the proteinase by incubating the tube at 95◦ C for 15 min in a
thermal cycler with heated lid (hot bonnet).
Add the contents of the PCR tube from step 4 to 250 µl of DNAzol®
in a new 1.7 ml Eppendorf tube. Add 4 µl of the Polyacryl carrier (stored
in refrigerator at 4◦ C; mix the carrier well before using).
Mix the contents of the tube by inverting it ca five times. Add 250
µl of 100% ethanol. Mix by inversion ten times. Let sample sit at room
temperature for 5 min.
Pellet DNA by centrifugation at 5000 g for 5 min at room temperature
(spin with cap ‘tabs’ facing out).
Wash the DNA twice with 800 µl of 75% ethanol. To wash, invert the
tube containing the sample and ethanol ca six times. Briefly centrifuge
the tube to pellet the DNA/carrier pellet, if necessary (if it has broken
loose). Carefully decant the ethanol (or remove by pipetting as required).
Remove remaining alcohol from the tube using a fine pipette tip/
pipettor. Do NOT spin-dry the DNA/carrier. Once excess ethanol is
removed, leave tubes open in the refrigerator (ensure that tubes are
slightly covered or tilted on their side to prevent contamination) or on
Vol. 5, 2007 717

B.J. Adams et al.
ice in a hood for 1-2 h to allow DNA pellet to dry. This should be done
in the refrigerator or on ice under a hood. Once all ethanol has evaporated
and the pellet has dried, resuspend the DNA pellet in 6-10 µl of TE. Try
using 2 µl of the resuspension for a 25 µl PCR reaction, though more
can be used if the DNA concentration is low.
• Preparation 100 mM Tris-HCl pH 7.6
Tris hydroxymethyl 15.8 g
Distilled water to 1000 ml
Dissolve 15.8 g of Tris in 800 ml of distilled water. Adjust pH with
NaOH. Add distilled water to bring the solution to 1000 ml. Sterilise by
autoclaving at 121◦ C for 15 min. Store at room temperature.
• Preparation 200 mM NaCl
Sodium chloride 11.7 g
Dissolve 11.7 g of NaCl in 800 ml of distilled water, and then add
distilled water to bring the solution to 1000 ml. Sterilise by autoclaving
at 121◦ C for 15 min. Store at room temperature.
• Preparation 0.5 M EDTA pH 8.0
EDTA 186.1 g
Add 186.1 g of EDTA in 500 ml of distilled water stir and heat at 30◦ C.
Add ca 100 ml of 1 M NaOH to bring the pH to 8.0. Add distilled water
to bring the solution to 1000 ml while adding 1 M NaOH as needed
for pH. Sterilise by autoclaving at 121◦ C for 15 min. Store at room
temperature.
• Preparation 1 M NaOH (for making EDTA)
Sodium hydroxide 40 g
Dissolve 40 g of sodium hydroxide in 800 ml of distilled water, and
then add distilled water to bring the solution to 1000 ml. Sterilise by
• Preparation 10% Sarkosyl
Sodium N-lauroyl sarcosine 100 g
Dissolve 100 g of sodium N-lauroyl sarcosine in 800 ml of distilled
water, then add distilled water to bring the solution to 1000 ml. Sterilise
by autoclaving at 121◦ C for 15 min. Store at room temperature.

• Preparation Proteinase K
Proteinase K 0.1 g
Dissolve 0.1 g of proteinase K in 9 ml of distilled water, then add
distilled water to bring the solution to 10 ml. Aliquot this into 0.2 PCR
tubes and store at −20◦ C.
• Preparation Digestion Solution
100 mM Tris HCl pH 7.6 200 µl
200 mM NaCl 200 µl
10% Sarkosyl 200 µl
0.5 M EDTA pH 8.0 400 µl
Proteinase K (10 mg ml−1 ) 20 µl
Ultrapure water 980 µl
(Final solution is 10 mM Tris-HCl pH 7.6, 20 mM NaCl, 100 mM
EDTA pH 8.0, 1% Sarkosyl, 0.1 mg ml−1 proteinase K.)
Worm Lysis Buffer DNA extraction (modified after Jones et al., 2006)
This method is rapid and yields the largest volumes of template for
PCR, but does not work equally well with all taxa (works poorly for
some Tylenchomorpha). Place a single nematode in a 0.2 ml PCR tube
with no more than 5 µl of water. Add 40 µl of the digestion solution to
the tube and incubate at 60◦ C for 1 h or until the individual is completely
digested. Vortex the tube occasionally. Manually rupturing the individual
organism can aid in rapid and complete digestion. This can be done by
using a pipette tip or pulled Pasteur pipette with a blunted end.
Check tissues for digestion using a dissection or light microscope. If
undigested, add 1 µl proteinase K and incubate longer (eggs will not
usually digest). ‘Pump-mix’ digested tissues.
Heat-kill the proteinase by incubating the tube at 95◦ C for 15 min in
a thermal cycler with heated lid (hot bonnet). Use 2-5 µl of lysate as the
DNA template in a 25 µl or 50 µl PCR reaction. Lysate should be stored
at −20◦ C or −80◦ C.
• Preparation 100 mM Tris pH 8.2
Tris (hydroxymethyl) aminomethane 12.1 g
Dissolve 12.1 g of Tris in 800 ml of distilled water. Adjust pH with
NaOH. Add distilled water to bring the solution to 1000 ml. Sterilise by
Vol. 5, 2007 719

B.J. Adams et al.
• Preparation Worm Lysis Buffer 10 ml

Proteinase K 10 mg
100 mM Tris pH 8.2 1 ml
KCl 37.3 mg
MgCl2 5.1 mg
Tween-20 45 µl
NP-40 (IGEPAL) 45 µl
Gelatin 5.0 mg
Ultrapure water to 10 ml
−1
(Final solution is 1 mg ml proteinase K, 10 mM Tris pH 8.2, 50 mM
KCl, 2.5 mM MgCl2 , 0.45% Tween-20, 0.45% NP-40, 0.05% gelatin.)
Dissolve 10 mg of proteinase K, 37.3 mg of KCl, 5.1 mg of MgCl2 ,
and 5.0 mg of gelatin in 8 ml of ultrapure water. Then add 1 ml of 100
mM Tris (pH 8.2), 45 µl of Tween-20, and 45 µl of NP-40. Add ultrapure
water to bring the solution to 10 ml. Store at 4◦ C in the refrigerator.
DNeasy® DNA extraction
Another extraction method that is effective in isolating nematode
DNA is the Qiagen DNeasy® tissue kit, which is our method of choice
for ethanol preserved specimens. Use the protocol for purification of
total DNA from animal tissues with the following modifications: In step
1, a single nematode is manually crushed using a pipette tip or pulled
Pasteur pipette with a blunted end. Follow step 2 but incubate at 60◦ C for
2-3 h while vortexing every 15 min. Follow the rest of the steps as written
in the DNeasy® tissue handbook. After the final step, dry the sample
down using a vacuum speed dryer for 2 h or until dry and resuspend the
DNA in 20 µl of ultrapure water or TE.
S EQUENCE ANALYSIS
Polymerase chain reaction (PCR)
PCR is a powerful molecular technique used to amplify target genes,
although the specifics vary depending on the primers and commercial
DNA polymerase that is used. Here is a generic PCR protocol that works
with many brands of Taq and several primer sets as well. All reagents
should be stored at −20◦ C or −80◦ C. Thaw reagents on ice:
Buffer (commercially available Taq comes with its own specific
buffer)
Thermus aquaticus (Taq) DNA Polymerase

DNA Primers (both forward and reverse)

Deoxynucleotide triphosphates (dNTPs)
Ultrapure water
DNA template (from extraction or other source)
Once the reagents have completely thawed, they are combined in a
1.5 ml Eppendorf tube in the following order and quantities, forming
the reaction mixture also known as the master mix. This step avoids
pipetting minute amounts into multiple individual tubes:
Reagent 1 Reaction 10 Reactions x Reactions

ThermoPol buffer 2.5 µl 25 µl 2.5 × x µl
Ultrapure water 17.3 µl 173 µl 17.3 × x µl
Forward Primer 1.25 µl 12.5 µl 1.25 × x µl
Reverse Primer 1.25 µl 12.5 µl 1.25 × x µl
dNTPs 0.5 µl 5 µl 0.5 × x µl
Taq 0.2 µl 2 µl 0.2 × x µl
Total Master Mix Volume 23 µl 230 µl 23 × x µl
To allow for pipetting error, it is recommended that one extra reaction

be added to the calculation for every ten reactions needed. The master
mix should be mixed by pump mixing, flicking the tube or gently
inverting.
Transfer 23 µl of reaction mixture into a PCR tube for each DNA
template.
Add 2 µl of DNA template to each PCR tube for a total reaction
volume of 25 µl. If more DNA template is used, subtract the extra
volume from the water added so that each reaction still has a total volume
of 25 µl.
Load tubes into thermal cycler. The exact cycling parameters used
depend on the gene targeted, or in other words the primers being used.
The following cycling parameters are generic and work with a variety of
primer sets, although optimisation may be required in order to obtain the
best results for your primer set.
Steps Step instructions
1. Initial denaturation Incubate at 95◦ C for 10:00 min
2. Denaturation Incubate at 94◦ C for 1:00 min
3. Annealing Incubate at 50◦ C for 1:00 min
Vol. 5, 2007 721

B.J. Adams et al.
4. Elongation Incubate at 72◦ C for 2:00 min

5. Cycling step Cycle to step 2 for 39 more times
6. Final elongation Incubate at 72◦ C for 10:00 min
7. Cycle termination Incubate at 4◦ C forever
Once the thermal cycler reaches step 7, the cycle can be stopped and
the tubes with PCR product removed. PCR product should be stored at
−20◦ C or −80◦ C and should always be kept on ice when in use.
• Preparation Primer Resuspension
DNA forward and reverse primer lyophilised
Ultrapure water variable
Primers are shipped lyophilised. To make a master stock, check the
number of nanomoles on the tube. Multiply this number by 10 then add
that number of µl of ultrapure water to the tube. For example, in a tube
with 72.5 nm of primers, add 617 µl of ultrapure water. This creates a
final stock of about 100 pmol µl−1 which should be stored at −20◦ C
or −80◦ C. Mix by inversion or by gently vortexing. Working stocks of
primer for PCR should be 20 pmol µl−1 by diluting at a ratio of 1 : 5.
This is done by adding 20 µl of your final stock primer to 80 µl of
ultrapure water in a 0.2 ml PCR tube.
• Preparation dNTP solution
100 mM dATP 62.5 µl
100 mM dCTP 62.5 µl
100 mM dGTP 62.5 µl
100 mM dTTP 62.5 µl
Ultrapure water 2.25 ml
Pipette 2.25 ml of ultrapure water into a 15 ml conical tube, then
pipette 62.5 µl of each dNTP solution into the tube. Mix by inversion
or by gently vortexing. Aliquot 500 µl of the mixture in five separate
0.5 ml tubes to minimise freeze thawing and to prevent contamination.
This solution should be stored at −20◦ C or −80◦ C and should always be
thawed on ice before use.
Agarose gel electrophoresis
PCR reactions can be verified and quantified using gel electrophoresis.
Make a 1% agarose gel by adding 0.5 g agarose to 50 ml of 1X TBE. This
is heated on a Bunsen burner or in a microwave until the solution is uni-
form with no visible agar grains. While still warm but no longer steaming
add 4 µl of ethidium bromide, which allows the DNA to be visualised

in the gel. Pour the still warm mixture into a gel mould. Allow the gel to
cool, setting in combs that provide the amount of wells as needed.
Once solid, the gel can be transferred to a gel box and submerged in
1X TBE. PCR product, thawed on ice, can be added to the individual
wells by pipetting 3 µl of DNA tracking dye onto a sheet of Parafilm®
for each sample to be run. Then 5 µl of PCR product is added to the bead
of tracking track, pump mixed, and then pipetted into a well. DNA ladder
should be added in the same manner but using 1 µl of ladder (amount of
ladder required will vary by manufacturer and concentration) and 1 µl of
tracking dye. The purpose of the dye is to allow the otherwise clear PCR
product to be visualised as it is pumped into each well. Before running
the gel, add an additional 4 µl of ethidium bromide into the TBE at the
cathode end of the gel box.
Be sure that the end of the gel with the wells is toward the anode and
the bottom of the gel is positioned toward the cathode. The gel should be
run using 90 volts for ca 30 min, allowing for the tracking dye to reach
the bottom of the gel.
DNA bands are then visualised using UV light. Based on the DNA
ladder used, both the fragment size and quantity of DNA in the PCR
product can be quantified using the position of the band in the gel and
the brightness of each band, respectively.
• Preparation 1X TBE Electrophoresis Buffer
Tris (hydroxymethyl) aminomethane 10.8 g
Boric acid 5.5 g
0.5 M EDTA 2 ml
Dissolve 10.8 g of Tris base and 5.5 g of boric acid in 800 ml of
distilled water, then add 2 ml of 0.5 M EDTA. Add distilled water to
bring the solution to 1000 ml. Store at room temperature.
• Preparation 1% Ethidium bromide
Ethidium bromide 0.1 g
Distilled water 100 ml
Dissolve 0.1 g of ethidium bromide in 100 ml of distilled water. Mix
by shaking or inversion. Store this in a light-protected bottle in the
refrigerator. This represents the stock solution. Make a working stain
by diluting 0.4 ml of the stock solution with 400 ml of distilled water.
This can be stored at room temperature. Exercise caution when making
ethidium bromide as it is mutagenic.
Vol. 5, 2007 723

B.J. Adams et al.
• Preparation DNA Tracking Dye (with Glycerol or Sugar)

Bromophenol Blue 25 mg
Glycerol 3 ml
Sucrose 4g
Dissolve 25 mg bromophenol blue in 7 ml of distilled water. Add 3 ml
of glycerol or 4 g of sucrose and mix by vortexing. Store at 4◦ C.
ExoSAP-IT® purification of PCR product

For sequencing and further analysis of the product resulting from
PCR, excess primers and dNTPs need to be removed. There are
several options to do this, but we find ExoSAP-IT® fast and relatively
inexpensive. Transfer 5 µl of PCR product into a new 0.2 ml PCR tube
and then add 2 µl of ExoSAP-IT® for a total volume of 7 µl. Place this
tube in the thermal cycler and have it run the following programme:
1. Enzyme activation Incubate at 37◦ C for 15 min
2. Enzyme denaturation Incubate at 80◦ C for 15 min
3. Cycle termination Incubate at 4◦ C forever
The resulting PCR product is free of excess nucleotides and primer
and is now ready for further analysis (sequencing, SNP analyses, etc.).
ExoSAP-IT® and cleaned PCR product should be stored at −20◦ C.
Big Dye cycle sequencing

This technique is used to determine the nucleotide sequence of genes
amplified using polymerase chain reaction. It is very similar to PCR
but utilises chain terminating, fluorescently labelled ddNTPs. Reagents
should be stored at −20◦ C and thawed on ice:
5X Buffer (commercially available Big Dye comes with its own
specific buffer) Big Dye
DNA Primers (only one direction per reaction; forward or reverse
only)
Ultrapure water
Purified PCR product
Once the reagents have completely melted, they are combined in a
1.5 ml Eppendorf tube in the following order and quantities, forming
the reaction mixture also known as the master mix. This step avoids

pipetting minute amounts into multiple individual tubes:
Reagent 1 Reaction 10 Reactions x Reactions

Ultrapure water 6.4 µl 64 µl 6.4 × x µl
5X Buffer 2.1 µl 21 µl 2.1 × x µl
Primer (forward or reverse) 1.0 µl 10 µl 1.0 × x µl
Big Dye 0.5 µl 5.0 µl 0.5 × x µl
Total Master Mix Volume 10 µl 100 µl 10 × x µl
To allow for pipetting error, it is recommended that one extra reaction

worth of reagents be added for every ten reactions needed. The master
mix should be mixed by pump mixing, flicking the tube or gently
inverting.
Transfer 10 µl of reaction mixture into a PCR tube for each DNA
template and primer direction. In order to obtain sequence for the
forward and reverse directions for each sample, two reactions must be
done, one with the forward primer and one with the reverse.
Add 2 µl of cleaned PCR product to each 0.2 ml PCR tube for a total
reaction volume of 12 µl. Load tubes into thermal cycler. For Big Dye
cycle sequencing reactions use the following sequencing parameters:
1. Initial denaturation Incubate at 95◦ C for 10 min
2. Denaturation Incubate at 94◦ C for 1 min
3. Annealing Incubate at 50◦ C for 1 min
4. Elongation Incubate at 72◦ C for 2 min
5. Cycling step Cycle to step 2 for 39 more times
6. Final elongation Incubate at 72◦ C for 10 min
7. Cycle Termination Incubate at 4◦ C forever
Once the thermal cycler reaches step 7, the cycle can be stopped and
the tubes with sequencing product removed. Sequencing product should
be stored at −20◦ C or −80◦ C and should always be kept on ice when in
use.
Cycle sequencing Sephadex™ clean-up

Sequencing reactions must be completely cleaned before submitted
for electrophoresis. Much like PCR product, the extra primers and
dNTPs must be removed from the reaction mix while preserving the
Vol. 5, 2007 725

B.J. Adams et al.
sequenced target gene. This can be done by column purification using

Sephadex™ G-50 Fine. The column is set up in a 96-well MultiScreen™
Millipore filter plate. This is done using a Sephadex™ loading plate
and filling it with Sephadex™ G-50 Fine. The column is hydrated using
300 µl of distilled water and allowed to set up at room temperature for
20 min. Excess water is pulled from the column using centrifugation.
The MultiScreen™ Millipore filter plate is attached to a 96-well
collection plate by a MultiScreen™ centrifuge alignment frame and
then centrifuged for 4 min at ca 500 g. The collection plate is emptied
and new sterile collection plate is attached. Before adding the 12 µl
sequencing reaction product to the column, 15 µl of distilled water is
added to the sequencing product, adding volume. The now diluted 27
µl sequencing reaction mixture is added to one well of the 96-well
MultiScreen™ Millipore filter plate where the Sephadex™ columns
have been set up. This is then pulled through the column by centrifuging
the plate at ca 500 g for 4 min. The collection plate is now ready for
electrophoresis.
Acknowledgements
We wish to acknowledge Bishwo Adhikari, Jim Baldwin, Ernie

Bernard, Dana Blackburn, Mark Blaxter, Dan Brooks, John Chaston,
T.J. Bliss, Tim Davie, Paul De Ley, Virginia Ferris, Ugur Gozel, Martin
Garcia, David Hull, Heather Koppenhöfer, Hugo Kons, John Lynch,
Steve Nadler, Tom Powers, Rebecca Scholl, Sergei Subbotin, Patricia
Stock, Don Strong, Walter Sudhaus and Sergei Spiridonov for helpful
discussions on various topics in this paper (whether they agree with us
or not) and/or their roles in developing and optimising the molecular
protocols. Errors in the present work should be attributed to us, not them.
References
A NON . (1964). Jacobellis v. Ohio. Supreme Court of Ohio.

A DAMS , B.J. (1998). Species concepts and the evolutionary paradigm in
modern nematology. Journal of Nematology 30, 1-21.
A DAMS , B.J. (2001). The species delimitation uncertainty principle. Journal
of Nematology 33, 153-160.

A DAMS , B.J. & N GUYEN , K.B. (2002). Taxonomy and systematics. In:
Gaugler, R. (Ed.). Entomopathogenic nematology. Wallingford, UK, CABI
Publishing, pp. 1-33.
A DAMS , B.J., B URNELL , A.M. & P OWERS , T.O. (1998). A phylogenetic
analysis of Heterorhabditis (Nemata: Rhabditidae) based on internal tran-
scribed spacer 1 DNA sequence data. Journal of Nematology 30, 22-39.
A DAMS , B.J., F ODOR , A., KOPPENHÖFER , H.S., S TACKEBRANDT, E.,
S TOCK , S.P. & K LEIN , M.G. (2006). Biodiversity and systematics of
nematode-bacterium entomopathogens. Biological Control 37, 32-49.
AGUINALDO , A.M.A., T URBEVILLE , J.M., L INFORD , L.S., R IVERA , M.C.,
G AREY, J.R., R AFF , R.A. & L AKE , J.A. (1997). Evidence for a clade of
nematodes, arthropods and other moulting animals. Nature 387, 489-493.
AYALA , F.J. & R ZHETSKY, A. (1998). Origin of the metazoan phyla:
Molecular clocks confirm paleontological estimates. Proceedings of the
National Academy of Sciences of the United States of America 95, 606-611.
B LAXTER , M., D ORRIS , M. & D E L EY, P. (2000). Patterns and processes in
the evolution of animal parasitic nematodes. Nematology 2, 43-55.
B LAXTER , M., M ANN , J., C HAPMAN , T., T HOMAS , F., W HITTON , C.,
F LOYD , R. & A BEBE , E. (2005). Defining operational taxonomic units
using DNA barcode data. Philosophical Transactions of the Royal Society
B – Biological Sciences 360, 1935-1943.
B LAXTER , M.L., D E L EY, P., G AREY, J.R., L IU , L.X., S CHELDEMAN ,
P., V IERSTRAETE , A., VANFLETEREN , J.R., M ACKEY, L.Y., D ORRIS ,
M., F RISSE , L.M., V IDA , J.T. & T HOMAS , W.K. (1998). A molecular
evolutionary framework for the phylum Nematoda. Nature (London) 392,
71-75.
B LOUIN , M.S., L IU , J. & B ERRY, R.E. (1999). Life cycle variation and the
genetic structure of nematode populations. Heredity 83, 253-259.
C ICHE , T.A., DARBY, C., E HLERS , R.-U., F ORST, S. & G OODRICH -
B LAIR , H. (2006). Dangerous liaisons: the symbiosis of entomopathogenic
nematodes and bacteria. Biological Control 38, 22-46.
C LAUS , C.F.W. & G ROBBEN , K. (1910). Lehrbuch der Zoologie. Mar-
burg/Lahn, Germany, Elwert, 1002 pp.
C RUM , H. (1985). Traditional make-do taxonomy. The Bryologist 88, 221-222.
D E L EY, P., F ELIX , M.A., F RISSE , L.M., NADLER , S.A., S TERNBERG , P.W.
& T HOMAS , W.K. (1999). Molecular and morphological characterisation of
two reproductively isolated species with mirror-image anatomy (Nematoda:
Cephalobidae). Nematology 1, 591-612.
D UNCAN , L.W., I NSERRA , R.N., T HOMAS , W.K., D UNN , D., M USTIKA ,
I., F RISSE , L.M., M ENDES , M.L., M ORRIS , K. & K APLAN , D.T. (1999).
Molecular and morphological analysis of isolates of Pratylenchus coffeae
and closely related species. Nematropica 29, 61-80.
Vol. 5, 2007 727

B.J. Adams et al.
FARRIS , J.S. (1980). Naturalness, information, invariance, and the conse-

quences of phenetic criteria. Systematic Zoology 29, 360-381.
FARRIS , J.S. (1981). Distance data in phylogenetic analysis. In: Funk, F.A. &
Brooks, D.R. (Eds). Advances in cladistics. Proceedings of the First Meeting
of the Willi Hennig Society. New York, USA, New York Botanical Garden,
pp. 3-23.
FARRIS , J.S. (1983). The logical basis of phylogenetic analysis. In: Platnick,
N.I. & Funk, V.A. (Eds). Advances in cladistics, volume 2. Proceedings of
the Second Meeting of the Willi Hennig Society. New York, USA, Columbia
University Press, pp. 1-47.
F ERGUSON , J.W.H. (2002). On the use of genetic divergence for identifying
species. Biological Journal of the Linnean Society 75, 509-516.
FFRENCH -C ONSTANT, R., WATERFIELD , N., DABORN , P., J OYCE , S.,
B ENNETT, H., AU , C., D OWLING , A., B OUNDY, S., R EYNOLDS , S. &
C LARKE , D. (2003). Photorhabdus: towards a functional genomic analysis
of a symbiont and pathogen. FEMS Microbiology Reviews 26, 433-456.
G ASCUEL , O. (1997). BIONJ: An improved version of the NJ algorithm based
on a simple model of sequence data. Molecular Biology and Evolution 14,
685-695.
G IRIBET, G., E DGECOMBE , G.D. & W HEELER , W.C. (2001). Arthropod
phylogeny based on eight molecular loci and morphology. Nature 413,
157-161.
G OZEL , U., A DAMS , B.J., N GUYEN , K.B., I NSERRA , R.N., G IBLIN -DAVIS ,
R.M. & D UNCAN , L.W. (2006a). A phylogeny of Belonolaimus populations
in Florida inferred from DNA sequences. Nematropica 36, 155-171.
G OZEL , U., L AMBERTI , F., D UNCAN , L., AGOSTINELLI , A., ROSSO , L.,
N GUYEN , K. & A DAMS , B.J. (2006b). Molecular and morphological
consilience in the characterisation and delimitation of five nematode species
from Florida belonging to the Xiphinema americanum-group. Nematology 8,
521-532.
G RAYBEAL , A. (1998). Is it better to add taxa or characters to a difficult
phylogenetic problem? Systematic Biology 47, 9-17.
H E , Y., S UBBOTIN , S.A., RUBTSOVA , T.V., L AMBERTI , F., B ROWN , D.J.F.
& M OENS , M. (2005). A molecular phylogenetic approach to Longidoridae
(Nematoda: Dorylaimida). Nematology 7, 111-124.
H ERRE , E.A. (1993). Population structure and the evolution of virulence in
nematode parasites of fig wasps. Science 259, 1442-1445.
H ILLIS , D.M. & B ULL , J.J. (1993). An empirical test of bootstrapping as
a method for assessing confidence in phylogenetic analysis. Systematic
Biology 42, 182-192.
H ILLIS , D.M. & D IXON , M.T. (1991). Ribosomal DNA: molecular evolution
and phylogenetic inference. The Quarterly Review of Biology 66, 411-453.

H OGLUND , J., M ORRISON , D.A., D IVINA , B.P., W ILHELMSSON , E. &

M ATTSSON , J.G. (2003). Phylogeny of Dictyocaulus (lungworms) from
eight species of ruminants based on analyses of ribosomal RNA data.
Parasitology 127, 179-187.
H OMINICK , W.M. (2002). Biogeography. In: Gaugler, R. (Ed.). Entomopatho-
genic nematology. Wallingford, UK, CABI Publishing, pp. 115-143.
H OMINICK , W.M., R EID , A.P., B OHAN , D.A. & B RISCOE , B.R. (1996).
Entomopathogenic nematodes – biodiversity, geographical distribution and
the convention on biological diversity. Biocontrol Science and Technology 6,
317-331.
H OMINICK , W.M., B RISCOE , B.R., D EL P INO , F.G., H ENG , J.A., H UNT,
D.J., KOZODOY, E., M RÁČEK , Z., N GUYEN , K.B., R EID , A.P., S PIRI -
DONOV, S., S TOCK , P., S TURHAN , D., WATURU , C. & Y OSHIDA , M.
(1997). Biosystematics of entomopathogenic nematodes: current status, pro-
tocols and definitions. Journal of Helminthology 71, 271-298.
H UELSENBECK , J.P. & B OLLBACK , J.P. (2001). Empirical and hierarchical
Bayesian estimation of ancestral states. Systematic Biology 50, 351-366.
H UELSENBECK , J.P. & RONQUIST, F. (2001). MRBAYES: Bayesian infer-
ence of phylogenetic trees. Bioinformatics 17, 754-755.
H UTCHINSON , G.E. & M AC A RTHUR , R.H. (1959). A theoretical ecological
model of size distributions among species of animals. The American
Naturalist 93, 117-125.
JABLONSKI , D. (1987). Heritability at the species level: Analysis of geographic
ranges of Cretaceous mollusks. Science 238, 360-363.
JACKSON , J.B.C. (1974). Biogeographic consequences of eurytopy and
stenotopy among marine bivalves and their evolutionary significance. The
American Naturalist 108, 541-560.
J ONES , K.L., T ODD , T.C. & H ERMAN , M.A. (2006). Development of
taxon-specific markers for high-throughput screening of microbial-feeding
nematodes. Molecular Ecology Notes 6, 712-714.
K APLAN , D.T., T HOMAS , W.K., F RISSE , L.M., S ARAH , J.-L., S TANTON ,
J.M., S PEIJER , P.R., M ARIN , D.H. & O PPERMAN , C.H. (2000). Phylo-
genetic analysis of geographically diverse Radopholus similis via rDNA se-
quence reveals a monomorphic motif. Journal of Nematology 32, 134-142.
K IONTKE , K., H IRONAKA , M. & S UDHAUS , W. (2002). Description of
Caenorhabditis japonica n. sp (Nematoda: Rhabditida) associated with the
burrower bug Parastrachia japonensis (Heteroptera: Cydnidae) in Japan.
Nematology 4, 933-941.
K LUGE , A.G. (1998). Total evidence or taxonomic congruence: Cladistics or
consensus classification. Cladistics – the International Journal of the Willi
Hennig Society 14, 151-158.
Vol. 5, 2007 729

B.J. Adams et al.
K LUGE , A.G. (2004). On total evidence: for the record. Cladistics – the
International Journal of the Willi Hennig Society 20, 205-207.
L EACHE , A.D. & R EEDER , T.W. (2002). Molecular systematics of the Eastern
Fence Lizard (Sceloporus undulatus): a comparison of parsimony, likelihood,
and Bayesian approaches. Systematic Biology 51, 44-68.
L EWIS , E.E., C AMPBELL , J., G RIFFIN , C., K AYA , H. & P ETERS , A. (2006).
Behavioral ecology of entomopathogenic nematodes. Biological Control 38,
66-79.
L IU , J. & B ERRY, R.E. (1996). Phylogenetic analysis of the genus Steinernema
by morphological characters and randomly amplified polymorphic DNA
fragments. Fundamental and Applied Nematology 19, 463-469.
L IU , J., B ERRY, R.E. & M OLDENKE , A.F. (1997). Phylogenic relationships
of entomopathogenic nematodes (Heterorhabditidae and Steinernematidae)
inferred from partial 18S rRNA gene sequences. Journal of Invertebrate
Pathology 69, 246-252.
L IU , J., B ERRY, R.E. & B LOUIN , M.S. (1999). Molecular differentiation and
phylogeny of entomopathogenic nematodes (Rhabditida: Heterorhabditidae)
based on ND4 gene sequences of mitochondrial DNA. Journal of Parasitol-
ogy 85, 709-715.
L IU , J., P OINAR J R , G.O. & B ERRY, R.E. (2000). Control of insect pests
with entomopathogenic nematodes: the impact of molecular biology and
phylogenetic reconstruction. Annual Review of Entomology 45, 287-306.
LYNCH , J.D. (1989). The gauge of speciation: On the frequencies and
modes of speciation. In: Otte, D. & Endler, J.A. (Eds). Speciation and its
consequences. Sunderland, MA, USA, Sinauer Associates, Inc., pp. 527-553.
M AYR , E. (1963). Animal species and evolution. Cambridge, MA, USA, The
Belknap Press of Harvard University Press, 797 pp.
M AYR , E. (1964). Systematics and the origin of species, from the viewpoint of
a zoologist. New York, USA, Dover Publications, 334 pp.
M ORSE , D.R., L AWTON , J.H., D ODSON , M.M. & W ILLIAMSON , M.H.
(1985). Fractal dimension of vegetation and the distribution of arthropod
body lengths. Nature 314, 731-733.
M ULLER , K. (2006). Incorporating information from length-mutational events
into phylogenetic analysis. Molecular Phylogenetics and Evolution 38, 667-
676.
NADLER , S.A. (2002). Species delimitation and nematode biodiversity: phylo-
genies rule. Nematology 4, 615-625.
NADLER , S.A., A DAMS , B.J., LYONS , E.T., D ELONG , R.L. & M ELIN ,
S.R. (2000). Molecular and morphometric evidence for separate species
of Uncinaria (Nematoda: Ancylostomatidae) in California sea lions and
northern fur seals: hypothesis testing supplants verification. Journal of
Parasitology 86, 1099-1106.

NADLER , S.A., D E L EY, P., M UNDO -O CAMPO , M., S MYTHE , A.B.,

S TOCK , S.P., B UMBARGER , D., A DAMS , B.J., D E L EY, I.T., H OLOVA -
CHOV, O. & BALDWIN , J.G. (2006). Phylogeny of Cephalobina (Nema-
toda): molecular evidence for recurrent evolution of probolae and incongru-
ence with traditional classifications. Molecular Phylogenetics and Evolution
40, 696-711.
N GUYEN , K.B. & A DAMS , B.J. (2003). SEM and systematic studies of
Steinernema abbasi Elawad et al., 1997, and S. riobrave Cabanillas et al.,
1994 (Rhabditida: Steinernematidae). Zootaxa 179, 1-10.
N GUYEN , K.B. & S MART, G.C. (1996). Identification of entomopathogenic
nematodes in the Steinernematidae and Heterorhabditidae (Nemata: Rhabdi-
tida). Journal of Nematology 28, 286-300.
N GUYEN , K.B., M ARUNIAK , J. & A DAMS , B.J. (2001). Diagnostic and
phylogenetic utility of the rDNA internal transcribed spacer sequences of
Steinernema. Journal of Nematology 33, 73-82.
N GUYEN , K.B., S HAPIRO -I LAN , D.I., S TUART, R.J., M C C OY, C.W.,
JAMES , R.R. & A DAMS , B.J. (2004). Heterorhabditis mexicana n. sp.
(Rhabditida: Heterorhabditidae) from Tamaulipas, Mexico, and morphologi-
cal studies of the bursa of Heterorhabditis spp. Nematology 6, 231-244.
N GUYEN , K.B., G OZEL , U., KOPPENHÖFER , H.S. & A DAMS , B.J. (2006).
Heterorhabditis floridensis n. sp. (Rhabditida: Heterorhabditidae) from
Florida. Zootaxa 1177, 1-19.
PAGE , R.D.M. & H AFNER , M.S. (1996). Molecular phylogenies and host-
parasite cospeciation: gophers and lice as a model system. In: Harvey,
P.H., Brown, L., Maynard Smith, J. & Nee, S. (Eds). New uses for new
phylogenies. Oxford, UK, Oxford University Press, pp. 255-270.
P ERLMAN , S.J. & JAENIKE , J. (2003). Evolution of multiple components of
virulence in Drosophila-nematode associations. Evolution 57, 1543-1551.
P ERLMAN , S.J., S PICER , G.S., S HOEMAKER , D.D. & JAENIKE , J. (2003).
Associations between mycophagous Drosophila and their Howardula ne-
matode parasites: a worldwide phylogenetic shuffle. Molecular Ecology 12,
237-249.
P ETER , L.J. & H ULL , R. (1969). The Peter principle. New York, USA,
William Morrow & Company, Inc., 179 pp.
P HAN , K.L., S UBBOTIN , S.A., N GUYEN , N.C. & M OENS , M. (2003). Het-
erorhabditis baujardi sp. n. (Rhabditida: Heterorhabditidae) from Vietnam
and morphometric data for H. indica populations. Nematology 5, 367-382.
P OE , S. (1998). Sensitivity of phylogeny estimation to taxonomic sampling.
Systematic Biology 47, 18-31.
P OINAR J R , G.O. (1993). Origins and phylogenetic relationships of the
entomophilic rhabditids, Heterorhabditis and Steinernema. Fundamental
and Applied Nematology 16, 333-338.
Vol. 5, 2007 731

B.J. Adams et al.
P OSADA , D. & B UCKLEY, T. (2004). Model selection and model averaging

in phylogenetics: Advantages of Akaike information criterion and Bayesian
approaches over likelihood ratio tests. Systematic Biology 53, 793-808.
P OSADA , D. & C RANDALL , K.A. (1998). Modeltest: testing the model of
DNA substitution. Bioinformatics 14, 817-818.
P OWERS , T.O. (2004). Nematode molecular diagnostics: from bands to
barcodes. Annual Review of Phytopathology 42, 367-383.
R EID , A.P. (1994). Molecular taxonomy of Steinernema. In: Burnell, A.M.,
Ehlers, R.-U. & Mason, J.P. (Eds). Genetics of entomopathogenic nematode-
bacterium complexes. Proceedings of a symposium and workshops held
at St Patrick’s College, Maynooth, Co. Kildare, Ireland. Luxembourg-
Belgium, European Commission Directorate-General XII, Science, Research
and Development Environment Research Programme, pp. 49-58.
R EID , A.P., H OMINICK , W.M. & B RISCOE , B.R. (1997). Molecular tax-
onomy and phylogeny of entomopathogenic nematode species (Rhabditida:
Steinernematidae) by RFLP analysis of the ITS region of the ribosomal DNA
repeat unit. Systematic Parasitology 37, 187-193.
RODRIGUEZ -T RELLES , F., TARRIO , R. & AYALA , F.J. (2002). A method-
ological bias toward overestimation of molecular evolutionary time scales.
Proceedings of the National Academy of Sciences of the United States of
America 99, 8112-8115.
RONQUIST, F. & H UELSENBECK , J.P. (2003). MrBayes 3: Bayesian phyloge-
netic inference under mixed models. Bioinformatics 19, 1572-1574.
S ANDERSON , M.J., D ONOGHUE , M.J., P IEL , W. & E RIKSSON , T. (1994).
TreeBASE: a prototype database of phylogenetic analyses and an interactive
tool for browsing the phylogeny of life. American Journal of Botany 81, 183.
S ITES , J.W. & M ARSHALL , J.C. (2003). Delimiting species: a Renaissance
issue in systematic biology. Trends in Ecology and Evolution 18, 462-470.
S ITES , J.W. & M ARSHALL , J.C. (2004). Operational criteria for delimiting
species. Annual Review of Ecology Evolution and Systematics 35, 199-227.
S PIRIDONOV, S.E. & B ELOSTOTSKAYA , F.N. (1983). [New data about
morphology and biology of the nematodes of the genus Neoaplectana and
their position in the system of rhabditids.] Parasitologyia 17, 119-125.
S PIRIDONOV, S.E., R EID , A.P., P ODRUCKA , K., S UBBOTIN , S.A. &
M OENS , M. (2004). Phylogenetic relationships within the genus Stein-
ernema (Nematoda: Rhabditida) as inferred from analyses of sequences of
the ITS-5.8S-ITS2 region of rDNA and morphological features. Nematology
6, 547-566.
S TOCK , S.P., C AMPBELL , J.F. & NADLER , S.A. (2001). Phylogeny of Stein-
ernema Travassos, 1927 (Cephalobina: Steinernematidae) inferred from ribo-
somal DNA sequences and morphological characters. Journal of Parasitol-
ogy 87, 877-889.

S TURHAN , D. (1999). Prevalence and habitat specificity of mentioned nema-

todes in Germany. In: Gwynn, R.L., Smith, P.H., Griffin, C.T., Ehlers, R.-U.,
Boemare, N. & Mason, J.-P. (Eds). Entomopathogenic nematodes: applica-
tion and persistence of entomopathogenic nematodes. Brussels-Luxembourg,
European Commission, COST819, pp. 123-132.
S UDHAUS , W. & KOCH , C. (2004). The new nematode species Poikilolaimus
ernstmayri sp. n. associated with termites, with a discussion on the phylogeny
of Poikilolaimus (Rhabditida). Russian Journal of Nematology 12, 143-156.
S ZALANSKI , A.L., TAYLOR , D.B. & M ULLIN , P.G. (2000). Assessing
nuclear and mitochondrial DNA sequence variation within Steinernema
(Rhabditida: Steinernematidae). Journal of Nematology 32, 229-233.
T ENENTE , G., D E L EY, P., D E L EY, I.T., K ARSSEN , G. & VANFLETEREN ,
J.R. (2004). Sequence analysis of the D2/D3 region of the large subunit
RDNA from different Meloidogyne isolates. Nematropica 34, 1-12.
T HOMPSON , J.D., G IBSON , T.J., P LEWNIAK , F., J EANMOUGIN , F. &
H IGGINS , D.G. (1997). The CLUSTAL_X windows interface: flexible
strategies for multiple sequence alignment aided by quality analysis tools.
Nucleic Acids Research 25, 4876-4882.
VAN N UES , R.W., R IENTJES , J.M.J., VAN D ER S ANDE , C.A.F.M., Z ERP,
S.F., S LUITER , C., V ENEMA , J., P LANTA , J. & R AUE , H.A. (1994). Sep-
arate structural elements within internal transcribed spacer 1 of Saccha-
romyces cerevisiae precursor ribosomal RNA direct the formation of 17S
and 26S rRNA. Nucleic Acids Research 22, 912-919.
W ILEY, E.O. & M AYDEN , R.L. (2000). The evolutionary species concept. In:
Wheeler, Q.D. & Meier, R. (Eds). Species concepts and phylogenetic theory:
a debate. New York, Columbia University Press, pp. 70-89.
W RAY, G.A., L EVINTIN , J.S. & S HAPIRO , L.H. (1996). Molecular evidence
for deep Precambrian divergences among metazoan phyla. Science 274, 568-
573.
Vol. 5, 2007 733

TMP E39

Uploaded by

Copyright:

Available Formats

TMP E39

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

TMP E39

Uploaded by

Copyright:

Available Formats

NMP5[v.2007/04/19 7:46;] Prn:21/05/2007; 12:39 Tipas:?? F:nmp506.tex; /Austina p.

Nematology Monographs & Perspectives, 2007, Vol. 5, 693-733

Phylogeny and evolution

Nematodes originated during the Precambrian explosion over 500

O RIGINS OF ENTOMOPATHOGENIC NEMATODES (EPN)

© Koninklijke Brill NV, Leiden, 2007 693

B.J. Adams et al.

694 Nematology Monographs & Perspectives

Phylogeny and evolution

O RIGINS OF EPN SPECIES

Little is known of the modes of speciation that have produced the

Vol. 5, 2007 695

B.J. Adams et al.

However, nematodes could have high rates of dispersal, as evidenced

696 Nematology Monographs & Perspectives

Phylogeny and evolution

Fig. 241. Selection between hosts, where inclusive fitness is a function of

Vol. 5, 2007 697

B.J. Adams et al.

Relative to other families of nematodes, or clades for that matter (Figs

698 Nematology Monographs & Perspectives

Phylogeny and evolution

Fig. 243. Species richness of select entomogenous nematode clades. Het-

Vol. 5, 2007 699

B.J. Adams et al.

be maintained, and even increased over evolutionary time, counters the

700 Nematology Monographs & Perspectives

Phylogeny and evolution

Fig. 245. Phylogenetic position of Steinernema relative to 20 other most closely

Vol. 5, 2007 701

B.J. Adams et al.

Numerous molecular approaches have been applied to the evolution-

702 Nematology Monographs & Perspectives

Phylogeny and evolution

2001, 2004; Saeb et al., unpubl.). As the SSU is too conserved to be

Vol. 5, 2007 703

B.J. Adams et al.

P HYLOGENY OF STEINERNEMA SPECIES

Spiridonov and Belostotskaya (1983) suggested that morphological

704 Nematology Monographs & Perspectives

Phylogeny and evolution

Fig. 246. Phylogenetic relationships among species of Steinernema based on

Stock et al. (2001) available in TreeBase (Sanderson et al., 1994). For

Vol. 5, 2007 705

B.J. Adams et al.

varied but are likely phylogenetically uninformative autapomorphies.

706 Nematology Monographs & Perspectives

Phylogeny and evolution

et al., 2004) but includes several taxa previously unrepresented in

Vol. 5, 2007 707

B.J. Adams et al.

P HYLOGENY OF HETERORHABDITIS SPECIES

Poinar (1993) suggested that, based on morphological, physiological,

708 Nematology Monographs & Perspectives

Phylogeny and evolution

Fig. 247. Phylogenetic position of Heterorhabditis relative to 35 other most

Vol. 5, 2007 709

B.J. Adams et al.

rate categories). Starting trees were obtained by stepwise addition with