Syst. Biol. 62(3):411–423, 2013
© The Author(s) 2013. Published by Oxford University Press, on behalf of the Society of Systematic Biologists. All rights reserved.
For Permissions, please email: [email protected]
DOI:10.1093/sysbio/syt008
Advance Access publication February 15, 2013
Is Permanent Parasitism Reversible?—Critical Evidence from Early Evolution of House
Dust Mites
PAVEL B. KLIMOV∗
AND
BARRY OCONNOR
Department of Ecology and Evolutionary Biology, University of Michigan, Ann Arbor, MI 48109-1079, USA
to be sent to: Department of Ecology and Evolutionary Biology, University of Michigan, Ann Arbor, MI 48109-1079, USA;
E-mail: [email protected].
∗ Correspondence
Received 9 June 2012; reviews returned 30 August 2012; accepted 4 February 2013
Associate Editor: Elizabeth Jockusch
Abstract.—Long-term specialization may limit the ability of a species to respond to new environmental conditions and
lead to a higher likelihood of extinction. For permanent parasites and other symbionts, the most intriguing question is
whether these organisms can return to a free-living lifestyle and, thus, escape an evolutionary “dead end.” This question
is directly related to Dollo’s law, which stipulates that a complex trait (such as being free living vs. parasitic) cannot reevolve again in the same form. Here, we present conclusive evidence that house dust mites, a group of medically important
free-living organisms, evolved from permanent parasites of warm-blooded vertebrates. A robust, multigene topology (315
taxa, 8942 nt), ancestral character state reconstruction, and a test for irreversible evolution (Dollo’s law) demonstrate that
house dust mites have abandoned a parasitic lifestyle, secondarily becoming free living, and then speciated in several
habitats. Hence, as exemplified by this model system, highly specialized permanent parasites may drastically de-specialize
to the extent of becoming free living and, thus escape from dead-end evolution. Our phylogenetic and historical ecological
framework explains the limited cross-reactivity between allergens from the house dust mites and “storage” mites and the
ability of the dust mites to inhibit host immune responses. It also provides insights into how ancestral features related to
parasitism (frequent ancestral shifts to unrelated hosts, tolerance to lower humidity, and pre-existing enzymes targeting
skin and keratinous materials) played a major role in reversal to the free-living state. We propose that parasitic ancestors
of pyroglyphids shifted to nests of vertebrates. Later the nest-inhabiting pyroglyphids expanded into human dwellings to
become a major source of allergens. [Ancestral ecology; Dollo’s law; evolutionary “dead end”; house dust mites; permanent
parasitism; Pyroglyphidae.]
“Once a parasite, always a parasite.” This seeming
truism appears in recent online discussions of
everything from noxious computer software, to
welfare cheats, to politicians. Curiously, in evolutionary
biology, there is also a strongly rooted supposition that
highly specialized traits or ecologies, such as permanent
(or full-time) parasitism, result in irreversible or
unidirectional evolution (Futuyma and Moreno 1988;
Agnarsson et al. 2006; Cruickshank and Paterson 2006;
Goldberg and Igic 2008). Although the term “parasite”
has many definitions, all involve some degree of
dependence between organisms and lasting exploitation
of one organism by another, with permanent parasites
spending their entire lives on or in the body of a host
(Price 1980). The natural outcome of this situation is
that parasites may quickly evolve highly sophisticated
mechanisms for host exploitation and lose their ability
to function away from the host body. Parasites often
experience seemingly irreplaceable degradation or loss
of many genes as their functionality is no longer required
in a rich, predictable environment where hosts provide
both the living space and nutrition (Sakharkar et al.
2004; Dieterich and Sommer 2009; Mendonca et al. 2011).
Many perceive such specialization as evolutionarily
irreversible and, consequently, believe that parasites
can never return to their ancestrally free-living lifestyle
again (Toft et al. 1991; Combes and Simberloff 2005).
High degrees of specialization may evolve by
short-term selective advantage. However, long-term
specialization may limit the ability to respond to new
environmental conditions leading to a higher likelihood
of extinction and the proverbial evolutionary “dead
end” (Takebayashi and Morrell 2001; Termonia et al.
2001; Agnarsson et al. 2006). The question of whether
or not highly specialized lineages can enter a different
ecological niche, and thus, avoid the fate of evolutionary
“dead ends” is a subject of ongoing debates (reviewed
in Colles et al. 2009). Several studies suggest that
various organisms indeed can offset the constraints
imposed by specialization (Holmes 1977; Lanyon 1992;
Armbruster and Baldwin 1998; D’Haese 2000; Janz
et al. 2001; Termonia et al. 2001; Morse and Farrell
2005; Nosil and Mooers 2005; Stireman 2005; Colles
et al. 2009; Prendini et al. 2010). Particularly, supposed
irreversibility of parasitism has been placed under severe
scrutiny, resulting in a number of empirical studies that
have seemingly brought this parasitological dogma into
disrepute (Siddall et al. 1993; Radovsky et al. 1997; Smith
1998; Apakupakul et al. 1999; Light and Siddall 1999;
Dorris et al. 2002; Salewski 2003; Borda and Siddall 2004;
Cruickshank and Paterson 2006). However, advances
in analytical approaches suggest that simple ancestral
character state inference, as done in these studies, may
be confounded if irreversible evolution (Dollo’s law) is
assumed (Stireman 2005; Goldberg and Igic 2008).
Standard Markov (mk) models of trait evolution might
wrongly reject the hypothesis of irreversible evolution if
the net diversification rate (speciation minus extinction)
is greater for lineages having the ancestral state of
a character subject to Dollo’s law (Goldberg and Igic
411
[10:33 28/3/2013 Sysbio-syt008.tex]
Page: 411
411–423
412
SYSTEMATIC BIOLOGY
2008). In cases of parasitism, it means that if free-living
lineages (ancestral state) diversify faster than parasitic
ones, then on the phylogeny, a small portion of them
may appear surrounded by numerous parasitic lineages.
In this situation, “standard” methods of ancestral
character state reconstruction will wrongly infer reversal
of an ancestral state. Therefore, methods accounting for
speciation, extinction, and character state transition rates
should be employed to test for irreversible evolution
(Goldberg and Igic 2008). In this analytical framework,
one influential study rejecting Dollo’s law under the mk
model was demonstrated to be inconclusive (Goldberg
and Igic 2008), but a few violations of Dollo’s law have
since been reported (Kohlsdorf et al. 2010; Lynch and
Wagner 2010; Wiens 2011). The question of whether a
free-living state can re-evolve in permanent parasites is
still in doubt, and indisputable examples are currently
lacking.
Here, we analyze the long-standing problem of the
ancestral ecology of house dust mites (family
Pyroglyphidae), a medically important, yet understudied group of microscopic arthropods. Pyroglyphid
house dust mites are the most common cause of allergic
symptoms in humans, affecting 65 million to 1.2 billion
people worldwide (Cunnington and Gregory 1968;
Basagana et al. 2004; Holt and Thomas 2005; Colloff
2009; Hammad et al. 2009; Lloyd 2009). Dust mites
feed on organic debris (such as shed skin) and flourish
in nests of vertebrates, including human dwellings.
The mites’ severe allergenic properties are linked
to their powerful digestive and molting enzymes,
salivary secretions, and other water-soluble molecules.
For example, large quantities of digestive enzymes
accumulate in mites’ fecal material—minute particles
that can easily become airborne. When inhaled or in
contact with skin, the residual mite enzymes may break
up tight junctions between the epithelial cells and
trigger allergic reactions (Tovey et al. 1981; Arlian 1991;
Holt and Thomas 2005; Hammad et al. 2009; Lloyd
2009).
Pyroglyphid mites belong to a large acarine lineage,
the Astigmata, comprising more than 6100 described
species. This lineage contains an ensemble of several
mutually paraphyletic, mostly free-living lineages (“freeliving Astigmata”) and a large monophyletic parasitic
lineage—Psoroptidia.
Free-living Astigmata (ca. 2300 species) are usually
saprophagous, feeding on decaying organic material,
fungi, and bacteria found in patchy or ephemeral
habitats—fungal sporocarps, sap flows, dung, carrion,
seaweeds, insect and vertebrate nests, and stored food
(hence the colloquial name “storage mites”). Storage
mites often develop large populations in dust and stored
biological products (e.g., grain, straw), and can cause
allergies in humans (Sidenius et al. 2001; Ong and Chew
2010). Specialized deutonymphs often disperse among
habitat patches through phoretic associations with
arthropods and vertebrates that also use these resources.
Phoretic deutonymphs usually do not feed during
dispersal, but there are several unrelated free-living
[10:33 28/3/2013 Sysbio-syt008.tex]
VOL. 62
lineages with parasitic deutonymphs: Hypoderatidae,
Glycyphagidae, Chortoglyphidae, and Echimyopodidae
(Fig. 1, color-coded yellow). These deutonymphs occur
inside hair follicles or under the skin, where they acquire
nutrition from hosts (the exact mechanism is still not
understood—these deutonymphs do not have a mouth
or normal gut).
Psoroptidia (ca. 3800 species) are permanent or fulltime parasites. This group is represented by highly
specialized parasites that spend their entire lives on or
in a suitable vertebrate host, unable to complete their
life cycle without it, and lack a free-living dispersal
stage. Host specificity is relatively high, with 80% of mite
species specific to host species or genera. The typical
host range includes birds and mammals; occasionally
they are hyperparasitic on insect parasites. Psoroptidian
mites include several ecological groups (respiratory
endoparasites, quill, skin-surface and skin-burrowing
mites, Fig. 2c–i) that are conventional parasites causing
negative fitness effects on their hosts. However, some
mites feeding on uropygial gland secretions (serving
to waterproof and protect bird plumage) may inflict
minimal or negligible harm to the host fitness. They
not only rarely cause pathology (reviewed by Proctor
2003) but also can opportunistically feed on blood (PBK
personal observation). Sometimes called commensals,
such feather vane-dwelling mites (Fig. 2a,b) fit within the
classical definition of parasitism because they consume
a functionally important component of the body and
are unable to survive away from their hosts. However,
if these mites are not accepted as being true parasites,
then it would be more appropriate to define our research
question as “Is full-time symbiosis evolutionarily
reversible?”
Relationships of the pyroglyphid dust mites and the
nonpsoroptidian and psoroptidian Astigmata are poorly
understood. Sixty-two different published hypotheses
argue whether pyroglyphids originated from (1) a freeliving ancestor (e.g., storage mites) that also gave rise to
parasitic lineages or (2) a parasitic ancestor that became
secondarily free-living. Although these hypotheses have
been discussed on numerous occasions in relation
to morphological, molecular, or immunological data
(Sidenius et al. 2001; Loo et al. 2003; Suarez-Martinez
et al. 2005; Lee et al. 2006; Cui et al. 2008; Klimov and
OConnor 2008; Dabert et al. 2010; Ong and Chew 2010;
Yang et al. 2010; Bochkov and Mironov 2011), there is
no consensus, and none of them have been evaluated
statistically. The latter hypothesis represents the most
intriguing question—can permanent parasites return to
a free-living state over evolutionary time and, thus,
escape the fate of being evolutionary “dead ends” in
violation of Dollo’s law? The most recent treatment
dubbed this hypothesis as “impossible” (Bochkov and
Mironov 2011).
We sequenced 5 nuclear genes (8942 nt, after removing
introns and unalignable regions) of 315 terminal taxa,
including all outgroups ever proposed for house dust
mites. Using a robust phylogeny, we (i) evaluated
the 62 hypotheses of alternative placement of house
Page: 412
411–423
2013
413
KLIMOV AND OCONNOR—REVERSIBILITY OF PERMANENT PARASITISM
a)
c)
b)
FIGURE 1.
Phylogram of pyroglyphid house dust mites and outgroups inferred by ML analysis a) (315 taxa, 6164 sites, nuclear rDNA
and protein-coding genes translated into amino acids). Four numbered key nodes leading to the dust mites are supplemented with bootstrap
support values, posterior probabilities (BA), and pie charts showing results of ancestral character state reconstruction for ecology (parasitic and
parasitic as deutonymphs vs. free-living) for these nodes (here values for Bayesian 1-rate Markov model are shown, see Table 2 for probabilities
calculated using other methods). Node 1 = Hypoderatidae+Heterocoptidae+Psoroptidia; 2 = Heterocoptidae+Psoroptidia; 3 = Psoroptidia;
4 = EP complex. b) ML tree of HSP70 (315 taxa, 569 amino acids, part of tree shown). c) BEAST species tree analysis (46 taxa, 4146 nt, proteincoding genes only, part of tree shown). The inferences are uncertain if in the dust mites a free-living state evolved twice a) or once as suggested
by the HSP70 topology b) or species tree topology inferred from protein-coding genes c). Both scenarios have low support. Scale bars represent
expected changes per site. Bootstrap support or posterior probability of 70–100% is shown by proportionally increased line weight (see actual
values and taxon names in Supplementary Figs. S3–S5).
dust mites using bootstrap replicates generated from
site-wise log-likelihoods calculated for each hypothesis
in Consel 1.20 (Shimodaira 2002); (ii) conducted an
ancestral character state reconstruction to determine the
ancestral ecology of house dust mites in BayesTraits
(Pagel et al. 2004) using, separately, maximum likelihood
(ML) parametric bootstrap trees and Bayesian stationary
trees to account for phylogenetic uncertainty; and
(iii) tested for irreversible evolution of parasitism
[10:33 28/3/2013 Sysbio-syt008.tex]
in a recently developed framework (Goldberg and
Igic 2008) comparing unconstrained models with
those constrained for irreversible evolution, while
accounting for speciation, extinction, and character
state transition rates in DiversiTree (FitzJohn 2010).
These tests provide decisive evidence for the historical
ecology of house dust mites and answer the question
of whether permanent parasitism is evolutionarily
reversible.
Page: 413
411–423
414
VOL. 62
SYSTEMATIC BIOLOGY
a)
b)
c)
d)
e)
f)
g)
h)
i)
FIGURE 2. Specialization to parasitic life style in Psoroptidia. a) Analges sp.: anterior legs with apophyses, spines, and modified setae (arrows)
aiding in maneuvering among barbs of undertail covert feathers (museum voucher BMOC 05-1023-001), b) Amerodectes sp.: cylindrical body
shape matching the mite restricted environment—spaces between barbs of primary feathers of the wing, the ambulacra (arrows) are enlarged
to hold on during the flight (BMOC 84-1002-001), c) Yunkeracarus otomys: extreme reduction of body setae, intranasal parasite of rodents (BMOC
91-1350-114), d) Labidocarpellus eonycteris: legs and gnathosoma are modified (arrows) to attach to the host hair, associated with bats (MAH
85-0131-001), e) Listrophorus squamiferus: body elongated, legs and gnathosoma are modified (arrows) to attach to the host hair, associated with
rodents (BMOC 93-1010-001), f) Myocoptes japonensis: legs III (arrow) are modified to attach to the host hair, skin parasite of rodents (BMOC 931010-001), g) Knemidokoptes jamaicensis: legs are extremely short, body spherical, skin-burrowing on birds’ legs (BMOC 99-0420-011), h) Sarcoptes
scabiei: same as above, ambulacra extremely elongated (arrows), skin-burrowing on humans and domestic animals (BMOC 82-0521-030) and i)
Heteropsorus sp.: ambulacra hypertrophied to attach to the upper skin of birds’ wings (BMOC 07-1015-093).
[10:33 28/3/2013 Sysbio-syt008.tex]
Page: 414
411–423
2013
KLIMOV AND OCONNOR—REVERSIBILITY OF PERMANENT PARASITISM
MATERIALS AND METHODS
Taxon Sampling, DNA Isolation, and Sequencing
Five genes, 18S, 28S rDNA, EF1-, SRP54, and HSP70
(8942 nt aligned, including 4775 nt of rDNA, 1389
amino acids of protein coding genes, no missing
data due to sequencing failures), were sequenced
for 315 taxa (Supplementary Table S1, available from
http://datadryad.org, doi:10.5061/dryad.rd1bc). The
total number of nucleotides in our alignment is
comparable (55–119%) to recent multigene phylogenomic studies (Dunn et al. 2008; Kocot et al. 2011;
Struck et al. 2011). As such, our sampling strategy
represents the broadly sampled, multigene approach,
which has been proven to be very successful in several
studies (Parfrey et al. 2010; Rothfels et al. 2012). A total
of 1535 new sequences were deposited tin GenBank
under accession numbers JQ000032–JQ001566. Another
40 sequences were utilized from our previous studies
(Klimov and OConnor 2008; Knowles and Klimov
2011; see Supplementary Table S1). Taxa were selected
to account for all previously proposed molecular
and morphological hypotheses on dust mite origins
(reviewed Klimov and OConnor 2008). The distant
outgroups in the acarine order Acariformes were
“Endeostigmata” (Lee et al. 2006; Domes et al. 2007)
and Oribatida (Norton 1998; Dabert et al. 2010; Pepato
et al. 2010). Potential close outgroups were representative
sets of nonpsoroptidian (including storage mites) and
psoroptidian (associated with birds and mammals)
astigmatid mites. Our data set represents nearly 4.8%
of known diversity of Astigmata, but it is well
balanced, and does not contain obvious taxonomic
biases. For example, the percentages of species in the
3 focal groups (free-living Astigmata, nonpyroglyphid
Psoroptidia, and pyroglyphids) are comparable with
the observed values (26.3 vs. 37.3, 70.0 vs. 61.8, and 3.7
vs. 0.9, respectively). It is almost impossible to obtain
a complete taxonomic sampling of astigmatid mites
given the large size of this group (6150 known species,
representing, perhaps, only 5–10% of the real diversity).
The ingroup pyroglyphid taxon sampling was nearly
exhaustive at the family–subfamily level and included
all common species causing allergies. DNA extraction,
rDNA secondary structure alignment, oligonucleotide
primers, amplification, and sequencing were previously
described (Knowles and Klimov 2011).
Phylogenetic Analyses
Alignment of rDNA was based on secondary
structures of Apis mellifera (Gillespie et al. 2006) and
Saccharomyces cerevisiae available from the Comparative
RNA Web site (Cannone et al. 2002). Stem regions
were further evaluated in the program mfold v. 3.1,
which folds rRNA based on free energy minimization
(Mathews et al. 1999; Zuker 2003) using the default
settings. Compensatory mutations in stem regions
were detected using a large comparative data set
[10:33 28/3/2013 Sysbio-syt008.tex]
415
(543 sequences of mites, unpublished). Unalignable,
hypervariable regions of rDNA (no common secondary
structure can be found) and introns were excluded
to avoid erroneous homology assignment. Exons of
protein-coding genes, with a few 3- or 6-nt indels, were
aligned in MacClade and then checked by eye. Orthology
of the genes was unambiguous based on the absence of
heterogeneous amplicons. All sequences were checked
against the nucleotide database using BLAST (Altschul
et al. 1997) to identify potential contaminants. Sequences
of protein-coding genes were translated into protein
prior to analyses.
Models of nucleotide or amino acid evolution were
selected based on corrected Akaike information criterion
(AICc) in the programs jModelTest 2.0 (Darriba et al.
2011a) and Prottest 3 (Darriba et al. 2011b), respectively.
We explored several partition strategies (by rDNA and
amino acids, individual genes, rDNA stem and loop
regions) and selected the following partition strategy
based on the lowest AICc value: rDNA stem, rDNA loop,
EF1-, SRP54, and HSP70 (Supplementary Table S2).
Phylogenetic relationships were inferred in a ML and
Bayesian framework in Garli 2.0.1019 (Zwickl 2006) and
MrBayes 3.1.2 (Ronquist and Huelsenbeck 2003; Altekar
et al. 2004) using a 52-node Mac OS X computer cluster.
At least 5 (Garli) or 6 (MrBayes) independent runs were
performed. Concatenated and separate analyses for each
partition were run in Garli (Supplementary Fig. S3)
to estimate potential biases introduced by discordant
gene genealogies on the total evidence tree (Degnan
and Rosenberg 2009). We used stem and loop regions
of rDNA as separate partitions because they have very
different levels of saturation (Klimov and OConnor 2008)
and, therefore, may provide insights on phylogenetic
signal present in the data set.
For Bayesian analyses (BA), convergence of model
parameters and topology were assessed by the standard
MrBayes convergence diagnostics (i.e., the average
standard deviation of split frequencies values below 0.01
and potential scale reduction factor values approaching
1.00) and the program Are We There Yet? (AWTY)
(Nylander et al. 2008). Adequacy of the posterior sample
size was evaluated through autocorrelation statistics as
implemented in Tracer v. 1.5 (Rambaut and Drummond
2009)—all effective sample size values substantially
exceeded 200. We ran MrBayes analyses for 10 million
generations discarding the first 100 000 trees as burnin as determined in Tracer. To investigate whether our
partitioned BA may be trapped in regions of parameter
space characterized by unrealistically long trees (Brown
et al. 2010; Marshall 2010), we compared the average post
burn-in tree lengths reported by Tracer and the ML tree
length estimate. The latter was shorter (9.136 vs. 10.357)
and outside of the 95% Bayesian credibility interval
(9.967–10.719), suggesting that our Bayesian inference
may somehow overestimate branch lengths. However,
these estimates may not be completely unrealistic in
contrast to studies reporting them as several orders of
magnitude longer than corresponding ML estimates,
while recovering the same topology (Brown et al. 2010).
Page: 415
411–423
416
SYSTEMATIC BIOLOGY
Given a potential problem with Bayesian branch length
estimates, we restrict our study to tests relying on
ML topologies. In the only test utilizing Bayesian trees
(BayesTraits), results were remarkably similar to those
obtained for ML trees.
ML trees were annotated with nonparametric
bootstrap support values (104–108 replicates) using the
program SumTrees (Sukumaran and Holder 2010) and
visualized in FigTree 1.3.1 (Drummond and Rambaut
2007). Matrices and trees from this study are available
from TreeBASE (http://www.treebase.org) accession
number 12087.
Because monophyly of Dermatophagoidinae+
Pyroglyphinae (major free-living lineages of the dust
mites) was not recovered, and because the status of this
group has potential implications for the evolutionary
loss of parasitism, we investigated these relationships
more closely by estimating a species tree using the
coalescent-based species tree method in BEAST v1.7.3
(Drummond and Rambaut 2007; Heled and Drummond
2010). For this analysis, a subset of taxa (n = 46), including
the epidermoptid–psoroptid (EP) complex (Fig. 1a: node
4) and 8 outgroups of the family Analgidae, was used.
A series of analyses utilizing different partitions and
their combinations (see above, partitioning strategy
by codon position was also used for protein-coding
genes) was conducted using at least 2 independent
Markov chain Monte Carlo (MCMC) analyses run for
1.2–3.8 ×108 generations with parameters sampled every
1000 steps. Independent runs were combined using
the program LogCombiner v. 1.4.6 (Drummond
and Rambaut 2007) and burn-in samples were
discarded. Convergence and adequacy of the posterior
sample size were determined as above for MrBayes
analyses.
Hypothesis Testing
Evaluation of 62 previously proposed hypotheses of
alternative placement of house dust mites was done in
Consel 1.20 using the AU statistic as the primary test
(Shimodaira 2002). Consel calculates P-values for the AU
and other types of statistics using bootstrap replicates
generated from site-wise log-likelihoods with the RELL
resampling method—a fast and accurate approximation
technique (Shimodaira 2001). We generated a matrix
of site-wise log-likelihoods in Garli using constraints
corresponding to each of the 62 hypotheses on
phylogenetic placement of pyroglyphids. We also report
the proportion of trees sampled from the posterior in our
preferred BA.
Ancestral character state reconstruction was done in
ML and Bayesian frameworks in BayesTraits (Pagel et al.
2004). To account for uncertainty in our phylogenetic
inferences, bootstrap trees (ML, 108 nonparametric
bootstrap replicates) and post burn-in trees sampled
from the posterior probability distribution (BA) were
used. For the latter, 180 000 post burn-in trees were
thinned to obtain 1000 trees in Burntrees v. 0.1.9
[10:33 28/3/2013 Sysbio-syt008.tex]
VOL. 62
(Nylander 2011). Polytomies were randomly resolved
in the APE module (Paradis et al. 2004) in R (R
Development Core Team 2010). Because the variable
describing parasitism in our system has 3 levels (freeliving, parasitic as deutonymph, permanently parasitic,
see below), reconstructions were performed under a
6-rate model (transition rates between all character
states are different). The single-rate model (all transition
rates are equal) was also evaluated to see if it
better fits the data. BA were run under reversible
jump MCMC, where models are visited in proportion
to their posterior probabilities. An exponential prior
was used seeded from a uniform on the interval
0–30.
We tested the hypothesis of irreversibility of
parasitism (ecological interpretation of Dollo’s law—no
morphological characters subject to Dollo’s law could
be found) using a recently developed framework
(Goldberg and Igic 2008) in DiversiTree containing
implementations of binary state speciation and
extinction (BiSSE) (Maddison et al. 2007), multiple
state speciation and extinction (MuSSE) (FitzJohn 2010),
and Markov models of character state evolution (mk2
and mk3) (Pagel 1994). We used 3 possible schemes of
character state coding: (i) 3-state coding (free-living,
parasitic as deutonymphs, and parasitic in all stages)
and 2 approaches of binary coding (ii–iii) where the
“parasitic as deutonymphs” state was assigned to
either “free-living” or to “parasitic,” respectively.
Models corresponding to the 3 coding schemes were
constrained to conform to the assumptions of irreversible
evolution (root is fixed to the free-living state; back
transition rates are set to zero) and then compared
to unconstrained models using the AIC. For the
unconstrained models, stationary probabilities based on
the assumption of equilibrium in the state frequencies
at the root were used. The importance of estimating
diversification rates can be demonstrated by the
following example. If diversification of free-living mites
(state A) is greater than that of parasitic mites (state B),
then the B-to-A transition rate would be overestimated
by simple mk models leading to an incorrect rejection
of irreversible evolution. If, however, diversification
is independent from state A, then mk models can
be used to assess unidirectional evolution. For these
reasons, we give results of different analyses estimating
state transitions rates only (mk) or both transition and
diversification rates (BiSSE and MuSSE). Irreversible
evolution is rejected when all tests prefer unconstrained
models over models constrained for irreversible
evolution with AIC differences of more than 10
(Burnham and Anderson 2004).
Estimating diversification rates, as in the above tests,
may be affected by incomplete or biased taxonomic
sampling (Cusimano and Renner 2010). Our data set is
incomplete, but the taxon sampling is nearly unbiased
(see the section “Taxon Sampling, DNA Isolation, and
Sequencing”). Consequently, results obtained by the
BiSSE and MuSSE analyses should be treated with
caution.
Page: 416
411–423
2013
Estimating Host Specificity
Parasites associated with a single host, or closely
related hosts, are more likely to be specialists, while
those parasitizing multiple, distantly related hosts are
expected to be generalists. Generalists would not have
adaptations to a particular host taxon but, having a
general morphology and means of active dispersal to
different hosts, rather employ the strategy of being jacksof-all-trades (Krasnov et al. 2004). In contrast to highly
specialized parasites, they have a greater evolutionary
potential and are more likely to be involved in a possible
transition from a parasitic to free-living state.
To estimate host specificity in Psoroptidia, we
prepared a host–parasite database consolidating data
provided by H. Proctor (bird mites) (Proctor 2012) and
A. Bochkov (mammal mites, unpublished). All records
involving unidentified mites or hosts were excluded.
The MS Excel function vlookup was used to verify mite
names using the Index to Organism Names compiled
from Zoological Record (Thomson Reuters 2012); host
names were similarly verified and standardized using
the online databases Mammal Species of the World
(Wilson and Reeder 2005) and the Clements Checklist
of Birds of the World (Clements et al. 2011). After
these procedures, our database contained a total of
9483 unique host records for 3901 named species of
psoroptidian mites (excluding Pyroglyphidae). These
data were then used to extract host range values
(expressed as the number of host taxa per mite species)
at 4 taxonomical levels (species-, genus-, family-, and
order-specific) using Pivot Table summary functions in
Excel. Averaged values were also calculated for each
family of mites. Parasites were arbitrarily considered as
“generalists” if they were associated with multiple host
orders, while species- to family-specific parasites were
considered as “specialists.”
RESULTS
Phylogenetic Analyses
Our analyses recovered pyroglyphids deeply nested
within a large monophyletic lineage otherwise
comprising parasites of mammals and birds—an
unranked clade termed Psoroptidia (BS 100, PP 1.00;
Fig. 1a: node 3). This group was first proposed by
Murray (1877), and named so by Yunker (1955), but its
monophyly has been contested multiple times. OConnor
(1982) offered convincing morphological evidence for
a monophyletic Psoroptidia and also proposed its
rank-free nature. Monophyly was confirmed by DNA
sequence analyses (Klimov and OConnor 2008).
The family Heterocoptidae, which is parasitic on
insects, is consistently recovered as the sister group
of Psoroptidia (BS 100, PP 1.00; Fig. 1a: node 2).
The internal lineage including the dust mites, the
EP complex (Klimov and OConnor 2008) (BS 99, PP
1.00; Fig. 1a: node 4) was inferred as an assemblage
of bird nasal endoparasites (family Turbinoptidae:
[10:33 28/3/2013 Sysbio-syt008.tex]
417
KLIMOV AND OCONNOR—REVERSIBILITY OF PERMANENT PARASITISM
TABLE 1. Support for 4 key nodes leading to the dust mites under
different analytical techniques, total evidence, and independent data
partition analyses
Partition
All genes (BA)
All genes (ML)
rDNA
Protein-coding genes
18S
28S
EF1-
SRP54
HSP70
Node
1
2
3
4
0.99
46
31a
44
nr
1b
0b
8a
7a
1.00
100
100
89
94
99
6c
8c
38
1.00
100
98
73
96
96
6c
8c
39
1.00
99
96
69
44
57
0d
52
0
Notes: Node 1, Hypoderatidae+Heterocoptidae+Psoroptidia; Node
2, Heterocoptidae+Psoroptidia; Node 3, Psoroptidia; Node 4, EP
complex (Fig. 1). Support is indicated as Bayesian posterior
probabilities (BA) and nonparametric bootstrap support values
(ML analysis; for brevity, ML is not indicated in rows 3–7); nr =
not recovered.
a Also includes Suidasiidae. b Also includes several families of
free-living mites. c With Heterocoptidae as ingroup. d Recovered
monophyletic, except for Gymnoglyphus.
Congocoptes, Schoutedenocoptes), bird skin mites (family
Epidermoptidae: Microlichus, Myialges, Promyialges,
the latter 2 genera are also partially hyperparasitic
on insects), feather mites (family Psoroptoididae:
Picalgoides, Mesalgoides, Hyomesalges, Temnalges), and
mammal skin mites (family Psoroptidae: Otodectes,
Psoroptes and family Lobalgidae: Echimytricalges).
Three key nodes (Fig. 1a: nodes 2–4) render the dust
mites deeply nested within parasitic lineages and are
particularly strongly supported by different types of
total evidence, as well as independent data partition,
and alternative close and distant outgroup analyses
(Table 1; Supplementary Figs. S3 and S4). For the
total evidence analyses support was high (BS 99–100),
decreasing for analyses including only rDNA (BS
96–100) or protein-coding genes (BS 69–89) (Table 1).
Single-gene analyses also recovered these clades, but
often with low or no support (Table 1; Supplementary
Fig. S3).
In agreement with other studies (Fain 1962; Cui et al.
2009; Arlian and Morgan 2011), we note paraphyly
of the large pyroglyphid genus Dermatophagoides. The
total evidence tree (Fig. 1a) and individual analyses of
4 genes (Supplementary Fig. S3) show that the freeliving pyroglyphids are not monophyletic. However, one
gene, HSP70, does render them monophyletic (Fig. 1b),
suggesting that reversal to a free-living state might
have occurred only once. To further investigate the
discordance across gene genealogies, we estimated a
species tree focusing on the free-living pyroglyphid
clades using the coalescent-based species tree method.
Free-living pyroglyphids were monophyletic (PP 0.49)
on a species tree inferred from a subset of the 3 proteincoding genes entered into analysis as DNA (Fig. 1c).
Two gene trees (EF1- and HSP70) supported these
relationships (PP 0.54–0.67), while another gene (SRP54)
Page: 417
411–423
418
also included Onychalges (bird parasites) here as an
ingroup (PP 1.00) (Supplementary Fig. S5). Translating
these loci into protein, adding rDNA, or using rDNA
only resulted in nonmonophyly of the free-living dust
mites (Supplementary Fig. S5). Given these results, a
single origin of the free-living lifestyle in pyroglyphids
cannot be ruled out. However, to accurately estimate the
probability of this event, as well as internal relationships
of Pyroglyphidae, a larger number of protein-coding loci
should be used.
Alternative Relationships of the Dust Mites
We statistically tested 62 existing hypotheses of dust
mite phylogenetic affinities based on morphological,
molecular, and immunological evidence and rejected
57 of them. Although several hypotheses may be of
interest to mite taxonomists only, we note a subset
of 16 hypotheses that assume close relationships of
pyroglyphids to the storage mites (Supplementary
Table S6). These relationships were proposed chiefly
based on the large, superficially similar female ovipore
and associated structures in the free-living family
Chortoglyphidae and the dust mites (Berlese 1897;
Cunliffe 1958). With the replacement of chortoglyphids
by other nonpsoroptodian Astigmata, this idea reemerged in several recent molecular studies (Loo
et al. 2003; Suarez-Martinez et al. 2005; Dabert et al.
2010; Yang et al. 2010). The controversy also extends
to immunology, with different studies demonstrating
similar cross-reactivity of the dust mites and either
storage mites or parasitic psoroptidian mites (reviewed
Sidenius et al. 2001; Ong and Chew 2010). Given that
the mites mostly have species-specific allergens (Arlian
et al. 2009), one would expect higher positive covariation
between sensitization involving phylogenetically related
groups. All hypotheses involving close relationships of
the storage and dust mites were statistically rejected by
our Consel analyses.
The 5 hypotheses that cannot be rejected by our
analyses suggest a parasitic sister group of the dust
mites. These hypotheses indicate (i) the possibility
of close relationships of the free-living pyroglyphids
(Dermatophagoidinae+Pyroglyphinae), and, thus, a
single origin of the free-living style (Supplementary
Table S6, row 22); (ii) mammal parasites (family
Psoroptidae) being their sister group (Supplementary
Table S6, row 6); (iii) the removal of Ptyssalgidae
(hummingbird quill inhabitants) from the house dust
mite complex with placement within the pteronyssid
feather mites (Supplementary Table S6, row 39); and
(iv–v) the higher-level relationships of Psoroptidia
with the 4 critical nodes as shown on Figure 1a
(Supplementary Table S6, rows 1, 52). Hypotheses i
and ii (with the inclusion of Turbinoptidae) were only
recovered in the analysis utilizing species tree estimation
under the coalescent model (Fig. 1c), whereas other
supported hypotheses were consistently recovered on
[10:33 28/3/2013 Sysbio-syt008.tex]
VOL. 62
SYSTEMATIC BIOLOGY
our concatenated phylogeny (Fig. 1a) and all other types
of analyses.
Irreversible Evolution
Our analyses indicate that unconstrained models
perform substantially better than models constrained
for unidirectional evolution (the root is fixed to the
free-living state, the transition rate of parasitism-to-freeliving state is set to zero) in all pairwise comparisons
representing different coding strategies for the variable
“parasitism.” The full-rate models (Table 2, rows 3, 4,
9, 10, 15, 16) were preferred over the single-rate models
(rows 1, 2, 7, 8, 13, 14), justifying the use of the former
over the transition-only models (mk). In these best,
unconstrained, full-rate models, speciation rates were
inferred to be much higher, and the corresponding
extinction rates were estimated to be much lower for
the parasitic state than the free-living state (Table 2,
rows 3, 9, 15). In contrast, under the assumptions of
irreversible evolution, the increase in diversification
rates should be associated with the free-living state.
No differences in results were observed in 3 separate
analyses utilizing different coding strategies for mites
with parasitic deutonymphs but otherwise free living
[i.e., assigned to a separate category (Table 2, rows 1–6),
or classified as either free living (rows 7–12) or parasitic
(rows 13–18)].
Because unconstrained, full-rate models accounting
for diversification and character state transition were
preferred over models constrained for irreversible
evolution (AIC 33.4, 17.8, 22.1 for the 3 coding
strategies; Table 2), and the increase of diversification
rates is associated with the parasitic state rather than the
free-living state, irreversible evolution is rejected in this
system.
Ancestral Ecology
Ancestral character state reconstruction inferred that
the respective common ancestors of 2 major lineages
leading to the dust mites, Psoroptidia+Heterocoptidae
and Psoroptidia (Fig. 1a: nodes 2, 3) were parasitic in
all stages of their life cycle (probabilities 0.983–1.000;
Supplementary Table S7). Probabilities for ancestral
parasitism in the EP complex (Fig. 1a: node 4) were
also high (0.684–0.992) and statistically significant
(Supplementary Table S7). We note high congruence
between 2 independent tests relying on trees inferred
by ML and BA. Absolute differences in probabilities
yielded by these tests were 0.000–0.154, 0.018±0.033
(range, average ± SD). ML trait models tend to be slightly
more conservative than Bayesian models in estimating
probabilities for ancestral parasitism (Supplementary
Table S7). Constraining the 6-parameter ML trait models
to 1 parameter made very little difference to the
likelihood (AIC 0.1–0.2, favoring the 6-rate models over
1-rate models). In contrast, Bayesian trait framework
Page: 418
411–423
2013
TABLE 2.
#
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
419
KLIMOV AND OCONNOR—REVERSIBILITY OF PERMANENT PARASITISM
Comparison of models constrained for irreversible evolution (Dollo’s law) of parasitism vs. unconstrained models
States Model
3
2a
2b
Root
MuSSE Stationary
Fixed
Stationary
Fixed
mk3 Stationary
Fixed
BiSSE Stationary
Fixed
Stationary
Fixed
mk2 Stationary
Fixed
BiSSE Stationary
Fixed
Stationary
Fixed
mk2 Stationary
Fixed
1
2
3
1
2
3
q12
q13
q21
7.060
1.322
0.000 0.014 3.312
7.060
1.322
0.170 0.513 —
4.878 7.749 8.629 3.365 0.000 0.000 0.000 0.000 6.107
4.832 7.393 8.808 1.041 7.208 0.000 0.340 0.433 —
—
—
—
—
—
— 0.143 0.151 0.119
—
—
—
—
—
— 0.171 0.512 —
7.060
—
1.322
— 0.404 — 0.078
7.060
—
1.322
— 0.688 —
—
4.770 8.546 —
1.812 0.000 — 0.185 — 0.239
4.730 8.635 —
1.067 0.000 — 0.590 —
—
—
—
—
—
—
— 0.404 — 0.078
—
—
—
—
—
— 0.688 —
—
7.060
—
1.322
— 0.176 — 0.094
7.060
—
1.321
— 0.487 —
—
4.971 8.659 —
1.802 0.000 — 0.127 — 0.125
5.002 8.811 —
1.513 0.000 — 0.397 —
—
—
—
—
—
—
— 0.176 — 0.094
—
—
—
—
—
— 0.487 —
—
q23
q31
q32
Ln L
df
0.536 0.122 0.000 218.095 8
0.000 —
— 202.889 5
0.516 0.152 0.000 242.490 12
0.000 —
— 222.772 9
0.096 0.101 0.000 −56.006 6
0.000 —
— −67.568 3
—
—
— 220.632 4
—
—
— 212.007 3
—
—
— 239.803 6
—
—
— 229.900 5
—
—
— −49.825 2
—
—
— −58.449 1
—
—
— 236.047 4
—
—
— 224.707 3
—
—
— 256.141 6
—
—
— 244.089 5
—
—
— −34.410 2
—
—
— −45.749 1
AIC
AIC
-420.190
-395.777
-460.979
-427.545
124.011
141.136
-433.264
-418.014
-467.606
-449.800
103.649
118.899
-464.093
-443.415
-500.282
-478.177
72.820
93.498
40.789
65.202
0
33.434
0
17.125
34.342
49.591
0
17.806
0
15.250
36.190
56.868
0
22.105
0
20.678
Notes: Analyses are based on selection between models explicitly accounting for speciation (), extinction (), character state transition rates
(q), and character state at root options (root) versus those favoring irreversible evolution (i.e., back character state transitions are set to zero and
the root is fixed to the free-living state, root = “fixed,” q21, q31, q32 = “—”). When MuSSE or BiSSE single-rate models (1, 2, 7, 8, 13, 14) are
preferred over full-rate models (3, 4, 9, 10, 15, 16) then mk models (5, 6, 11, 12, 17, 18) should be used instead. Here, however, the full-rate models
are preferred. Among them, unconstrained models present a better fit to the data (as evidenced by their lowest AICs), thus, rejecting irreversible
evolution in this system. States = 3-state coding: free-living, parasitic as deutonymphs, parasitic at all stages; 2a,b = binary coding 1 (categories
1 and 2 joined) and 2 (categories 2 and 3 joined), respectively; Model = BiSSE, MuSSE, mk2-3 (Markov 2 and 3 state models of character state
evolution) as implemented in DiversiTree; root = character states at root are either stationary or fixed to state 1 (complex state); = speciation
rates; = extinction rates; q = character state transition rates; ln L = ML; df = number of free parameters (degree of freedom); AIC = models
selected by AIC have AICs of zero.
using Bayesian trees shows clear superiority of the 1-rate
model (AIC 10.7) (Supplementary Table S7).
Specialization, Host Range, and Evolutionary Plasticity
Eighty percent of nonpyroglyphid prosroptidian mites
(3109 of 3901 species) are specialists, associated with a
single host or closely related hosts belonging to the same
genus (Supplementary Fig S8a,b). The large percentage
of species with limited host ranges is not surprising
because most psoroptidians do not have a specialized
dispersal stage, so their main method of infecting new
hosts is vertical transfer (from parent to offspring),
or, more rarely, horizontal transfer during mating or
occasional close contact between potential hosts.
Only 3% (126) of nonpyroglyphid proroptidians are
associated with hosts belonging to multiple orders
(Supplementary Fig S8a,b). Their ability to live on
a range of phylogenetically and morphologically
different hosts and maintain gene flow between
these seemingly isolated populations suggests that
these mites are generalists with potentially higher
evolutionary plasticity than single-host mites. On a
macroevolutionary scale, such mites are more likely to
avoid “dead ends” eminent for single-host parasites with
the extinction of their hosts. To successfully colonize
multiple hosts, mites should develop (i) “general”
morphologies suitable for parasitizing disparate hosts
and (ii) means for active dispersal across hosts, including
[10:33 28/3/2013 Sysbio-syt008.tex]
morphological and behavioral adaptations for a shortterm “free-living” (or hyperparasitic) period, during
which the mites are seeking a new host. Indeed, if
psoroptidian families are ordered by their average
host range (Supplementary Table S9), the taxa with
the broadest host ranges (Psoroptidae, Sarcoptidae,
Psoroptoididae, Dermoglyphidae, Knemidokoptidae,
Epidermoptidae; host order count per mite species is
from 1.122 to 1.767) tend to be skin parasites. This can be
explained by the fact that skin, in contrast to feathers or
fur, can be similar across many unrelated host lineages,
so there is no need for host-specific morphological
specializations (such as various elaborate modifications
of legs and body to attach to a specific place on the
host, as in fur mites, Fig. 2d–f). This is especially true for
certain psoroptids parasitizing the mammalian ear—this
habitat is relatively well protected from the outside and
does not require any specialized structures for fastening
to the host. Fur mites and plumaceous feather mites
live essentially in a three-dimensional environment, and
have very sophisticated morphological adaptations for
this. Some feather mites even develop asymmetrical
body shapes to match the asymmetry of individual
feathers (Gaud and Atyeo 1996). Mammal mites living
in upper, dead layers of the epidermis (without a
primary contact with the host immune system) are more
likely to have broader host range as compared with
mites living on the hairs themselves [P = 0.0273–0.0002
for comparisons of species- and genus-specific against
order(s)-specific, respectively, our multinomial logistic
regression analysis, unpublished].
Page: 419
411–423
420
SYSTEMATIC BIOLOGY
Several lines of evidence indicate that the EP complex,
a group that gave rise to the dust mites (Fig. 1a:
node 4), is characterized by a broader host range than
most other psoroptidian mites. First, the proportion of
mites associated with multiple host orders is relatively
higher in the EP complex than in other mites (23.0%
vs. 2.3–5.0%) (Supplementary Fig. S8). Second, of
the 6 families with the broadest average host range
(see above, Supplementary Table S9), 3 (Psoroptidae,
Psoroptoididae, and Epidermoptidae) are part of the
EP complex. Third, in the EP complex, mammal
and bird-associated lineages (Psoroptidae, Lobalgidae
vs. Psoroptoididae, Epidermoptidae, Turbinoptidae)
appear to be closely related and intermixed (Fig. 1a:
node 4). This indicates relatively frequent host shifts
across avian and mammalian hosts—something that is
not characteristic of other Psoroptidia. Finally, the only
family of nonpyroglyphid psoroptidian mites adapted
to horizontal transfer belongs to the EP complex
(some epidermoptid females are hyperparasitic on
hippoboscid flies). These lines of evidence suggest
that ancestors of the pyroglyphids were probably
unspecialized, multihost mites, parasitizing skin or
related environments.
DISCUSSION
Results from our phylogenetic analyses, topologybased tests for alternative placement of house dust mites,
ancestral character state reconstructions, and a test for
irreversibility of parasitism all suggest that the common
ancestor of pyroglyphid house dust mites underwent
reversal from a permanently parasitic lifestyle to become
secondarily free living. By inference, this violates the
ecological interpretation of Dollo’s law defining the
parasitic state as irreversible because many of the
adaptations necessary for life away from the host
(“complex state”) are assumed to be lost in parasites
(Cruickshank and Paterson 2006). Thus, as exemplified
by house dust mites, specialized organisms can adapt
to new ecological niches via de-specialization, escaping
evolutionary “dead ends.”
Recognizing the possibility of irreversible evolution in
this host–parasite system is very important. A pattern of
free-living clades deeply nested within parasitic lineages
may be generated if free-living lineages speciate faster
and go extinct more slowly than parasitic lineages where
parasitism is irreversible (Goldberg and Igic 2008).
Interestingly, a recently proposed historical ecological
scenario for pyroglyphids (Bochkov and Mironov 2011)
strikingly resembles this pattern. Under this hypothesis,
parasitism has evolved as many as 8 times in Psoroptidia.
Pyroglyphids were treated as “living fossils”—the
only surviving remnants of the ancestral stem group
that gave rise to all of the spectacular diversity
of parasitic psoroptidian mites. The authors gave 2
arguments against reversible parasitism: (i) there is no
co-divergence of pyroglyphids with their avian hosts
and (ii) living on a host versus its nest provides a strong
[10:33 28/3/2013 Sysbio-syt008.tex]
VOL. 62
selective advantage. However, strict co-divergence is not
expected in this system or elsewhere in Psoroptidia
(e.g., their and our cladograms show some mammalassociated mites as the sister group to pyroglyphids).
Similarly, the selective advantage argument is not
applicable here because mites living on a host versus its
nest occupy different ecological niches, so they cannot
compete with each other. In addition, because these
authors did not use the family Heterocoptidae, a critical
parasitic sister group of Psoroptidia, their assumption
about the free-living ancestor of psoroptidian mites
also should be re-evaluated. Our analyses, accounting
for both diversification and character state transition
rates, strongly reject the irreversible parasitism scenario
in this system. Potentially, these estimates can be
confounded by incomplete taxon sampling (Cusimano
and Renner 2010), but our tests that do not rely on
evaluations of diversification rates arrive at the same
conclusion.
How might the ecological shift associated with the
reversal to the free-living state have occurred? Close
relatives of pyroglyphids were inferred to be mammal
skin mites and avian endoparasites [Psoroptidae and
Turbinoptidae (Congocoptes)], but more distant relatives
might include feather-inhabiting mites (e.g., Analgidae)
(Fig. 1a). Although ancestral hosts cannot be determined
with certainty due to frequent host shifts, there is
little doubt that early free-living dust mites were nest
inhabitants. Indeed, nests of birds, mammals, or both
are the principal habitat of all modern free-living
pyroglyphid species (a few exceptions probably involve
accidental records). We propose that a combination
of several characteristics of their parasitic ancestors
played an important role in abandoning permanent
parasitism: tolerance of low humidity, development
of powerful digestive enzymes allowing feeding on
skin and keratinous materials, and low host specificity
with frequent shifts to unrelated hosts. As compared
to nonpyroglyphid free-living mites, efficient water
balance mechanisms have appeared within parasitic
lineages of Psoroptidia (Gaede and Knülle 1987) in
response to living in specific regions of the host body
(e.g., skin surface, flight feathers) and feeding on a
dry, fat-rich diet (keratinaceous materials and lipids
of the skin, sebaceous or uropygial gland secretions).
The specific diet of the mites and the need to
evade the host immune response probably promoted
development of powerful, specialized enzymes, such
as cysteine protease (group 1 allergen) (Kato et al.
2005). These features, occurring in almost all parasitic
mites, were potentially important precursors enabling
mite populations to thrive in host nests despite low
humidity and scarce, low quality food resources, largely
consisting of shed skin and feathers. Furthermore, as
evidenced by frequent ancestral shifts to unrelated host
lineages, selection in early pyroglyphids appears to
have favored an unspecialized morphology (such as the
absence of elaborate leg and gnathosomal modifications
specialized for attachment to specific places on the
hosts, Fig. 2), which probably enabled the switch from
Page: 420
411–423
2013
KLIMOV AND OCONNOR—REVERSIBILITY OF PERMANENT PARASITISM
permanent parasitism to a completely or partially freeliving condition. This transition might have occurred
through an intermediate state, with mites living both
in the nest and on the body of the host, utilizing
the host chiefly for dispersal. This intermediate state
can still be seen in several Dermatophagoides species,
such as Dermatophagoides evansi (personal observation)
and Dermatophagoides pteronyssinus (Fain et al. 1990).
With the advent of human civilization, nest-inhabiting
pyroglyphids could have shifted to human dwellings
from the nests of synanthropic birds or rodents,
where, thanks to potent digestive enzymes and other
immunogenic molecules, they became a major source of
allergies.
From a broader perspective, understanding
phylogenetic relationships of house dust mites and
other astigmatan mites may provide insights into
allergenic properties of their immunogenic proteins
and the evolution of paralogs and orthologs of genes
encoding allergens. Our phylogenic and historical
ecological inferences provide a highly predictive
framework for research in immunological aspects
of pyroglyphids. Given that the mites mostly have
species-specific allergens (Arlian et al. 2009), the great
phylogenetic distance between the dust and storage
mites might explain the limited cross-reactivity between
allergens from these mites demonstrated by some, but
not all studies (reviewed Sidenius et al. 2001; Arlian et al.
2009). Similarly, cross-antigenicity between allergens
of house dust mites and other parasitic psoroptidians
(Arlian et al. 1991) is not surprising given their close
phylogenetic affinities. Finally, the ancestrally parasitic
lifestyle of the dust mites is a very plausible explanation
for the ability of present day dust mites to modulate
host immunity and downgrade host immunological
response in laboratory settings (Arlian and Morgan
2011). One would expect this feature to be present only
in true parasites, having a direct contact with the host
immune system.
SUPPLEMENTARY MATERIAL
Data files and/or other supplementary information
related to this paper have been deposited at Dryad
(http://www.datadryad.org/) under doi: 10.5061/
dryad.rd1bc.
FUNDING
This work was supported by National Science
Foundation (NSF) [DEB-0613769, 9521744, 0118766 to
B.M.OC.], and also benefitted in part from specimens
collected by the Field Museum’s Emerging Pathogens
Project, funded by the Davee Foundation and the Dr.
Ralph and Marian Falk Medical Research Trust. The
molecular work of this study was conducted in the
Genomic Diversity Laboratory of the University of
Michigan Museum of Zoology. We greatly appreciate
comments and careful proofreading of a previous
[10:33 28/3/2013 Sysbio-syt008.tex]
421
version of the manuscript by E. Jockusch, J. Wiens, and
R. Cruickshank.
ACKNOWLEDGMENTS
Specimens for this study were collected by a network
of 64 biologists in 19 countries. We thank T. M. Pérez
Ortíz, G. Montiel Parra, J. B. Morales-Malacara, L. Luna
Wong, W. Stanley, and Z. Sayakova for organizing our
fieldtrips and obtaining collecting permits in Mexico,
Peru, Tanzania, and Kazakhstan; S. V. Mironov for
sharing his unpublished technique of sampling from live
birds without euthanizing them; L. Arlian for discussion
on allergy issues; J. Hubert for discussion on house dust
mite enzymes; J. Brown and H. Lanier for discussion on
various issues of molecular phylogenetics; M. Colloff,
S. V. Mironov, and A. V. Bochkov for discussion; G.
Bauchan and R. Ochoa for taking low temperature
scanning electron microscope photographs (LT-SEM) of
the American house dust mite Dermatophagoides farinae;
H. Abraham, J. Dikema, E. Foot Perkowski, and K. Mar
for assistance with molecular laboratory work.
REFERENCES
Agnarsson I., Aviles L., Coddington J.A., Maddison W.P. 2006. Sociality
in theridiid spiders: repeated origins of an evolutionary dead end.
Evolution 60:2342–2351.
Altekar G., Dwarkadas S., Huelsenbeck J.P., Ronquist F. 2004. Parallel
Metropolis coupled Markov chain Monte Carlo for Bayesian
phylogenetic inference. Bioinformatics 20:407–415.
Altschul S.F., Madden T.L., Schaffer A.A., Zhang J., Zhang Z., Miller W.,
Lipman D.J. 1997. Gapped BLAST and PSI-BLAST: a new generation
of protein database search programs. Nucleic Acids Res. 25:
3389–3402.
Apakupakul K., Siddall M.E., Burreson E.M. 1999. Higher level
relationships of leeches (Annelida: Clitellata: Euhirudinea) based
on morphology and gene sequences. Mol. Phylogenet. Evol. 12:
350–359.
Arlian L., Morgan M., Vyszenski-Moher D., Sharra D. 2009. Crossreactivity between storage and dust mites and between mites and
shrimp. Exp. Appl. Acarol. 47:159–172.
Arlian L.G. 1991. House-dust-mite allergens: a review. Exp. Appl.
Acarol. 10:167–186.
Arlian L.G., Morgan M.S. 2011. Immunomodulation of skin cytokine
secretion by house dust mite extracts. Int. Arch. Allergy Immunol.
156:171–178.
Arlian L.G., Vyszenski-Moher D.L., Ahmed S.G., Estes S.A. 1991. Crossantigenicity between the scabies mite, Sarcoptes scabiei, and the
house dust mite, Dermatophagoides pteronyssinus. J. Invest. Dermatol.
96:349–354.
Armbruster W.S., Baldwin B.G. 1998. Switch from specialized to
generalized pollination. Nature 394:632.
Basagana X., Sunyer J., Kogevinas M., Zock J.P., Duran-Tauleria
E., Jarvis D., Burney P., Anto J.M. 2004. Socioeconomic
status and asthma prevalence in young adults: the European
Community Respiratory Health Survey. Am. J. Epidemiol. 160:
178–188.
Berlese A. 1882–1897. Acari Myriopoda et Scorpiones hucusque in Italia
reperta. Ordo Cryptostigmata (Sarcoptidae). Portici (Italy): Typis
Vesuvianae.
Bochkov
A.V.,
Mironov
S.V.
2011.
Phylogeny
and
systematics
of
mammal-associated
psoroptidian
mites
(Acariformes : Astigmata : Psoroptidia) derived from external
morphology. Invertebr. Syst. 25:22–59.
Page: 421
411–423
422
SYSTEMATIC BIOLOGY
Borda E., Siddall M.E. 2004. Arhynchobdellida (Annelida: Oligochaeta:
Hirudinida): phylogenetic relationships and evolution. Mol.
Phylogenet. Evol. 30:213–225.
Brown J.M., Hedtke S.M., Lemmon A.R., Lemmon E.M. 2010. When
trees grow too long: investigating the causes of highly inaccurate
Bayesian branch-length estimates. Syst. Biol. 59:145–161.
Burnham K.P., Anderson D.R. 2004. Multimodel inference—
understanding AIC and BIC in model selection. Sociol. Method Res.
33:261–304.
Cannone J., Subramanian S., Schnare M., Collett J., D’Souza L., Du
Y., Feng B., Lin N., Madabusi L., Muller K., Pande N., Shang Z.,
Yu N., Gutell R. 2002. The Comparative RNA Web (CRW) Site: an
online database of comparative sequence and structure information
for ribosomal, intron, and other RNAs. BMC Bioinformatics 3:2.
Clements J.F., Schulenberg T.S., Iliff M.J., Sullivan B.L., Wood
C.L., Roberson D. 2011. The Clements checklist of birds
of the world: Version 6.6 [Internet]. Available from: URL
http://www.birds.cornell.edu/clementschecklist/downloadableclements-checklist.
Colles A., Liow L.H., Prinzing A. 2009. Are specialists at risk
under environmental change? Neoecological, paleoecological and
phylogenetic approaches. Ecol. Lett. 12:849–863.
Colloff M.J. 2009. Dust mites. Dordrecht (The Netherlands): Springer.
Combes C., Simberloff D. 2005. The art of being a parasite. Chicago
(IL): University of Chicago Press.
Cruickshank R.H., Paterson A.M. 2006. The great escape: do parasites
break Dollo’s law? Trends Parasitol. 22:509–515.
Cui Y., Zhou P., Peng J., Peng M., Zhou Y., Lin Y., Liu L. 2008. Cloning,
sequence analysis, and expression of cDNA coding for the major
house dust mite allergen, Der f 1, in Escherichia coli. Braz. J. Med.
Biol. Res. 41:380–388.
Cui Y.B., Gao C.X., Zhou Y., Peng Z., Peng M., Lin Y., Peng J. 2009.
Phylogenetic analysis of house dust mites. Cent. Eur. J. Med. 5:69–74.
Cunliffe F. 1958. Pyroglyphus morlani, a new genus and species of
mite forming a new family, Pyroglyphidae, in the Acaridiae. Proc.
Entomol. Soc. Wash. 60:85–86.
Cunnington A.M., Gregory P.H. 1968. Mites in bedroom air. Nature
217:1271–1272.
Cusimano N., Renner S.S. 2010. Slowdowns in diversification rates from
real phylogenies may not be real. Syst. Biol. 59:458–464.
Dabert M., Witaliñski W., Kazmierski A., Olszanowski Z., Dabert J.
2010. Molecular phylogeny of acariform mites (Acari, Arachnida):
strong conflict between phylogenetic signal and long-branch
attraction artifacts. Mol. Phylogenet. Evol. 56:222–241.
Darriba D., Taboada G.L., Doallo R., Posada D. 2011a. HPC selection of
models of DNA substitution. Proceedings of the 9th International
Conference on Computational Methods in Systems Biology (CMSB
’11). Paris, September 2011, p. 65–72.
Darriba D., Taboada G.L., Doallo R., Posada D. 2011b. ProtTest 3:
fast selection of best-fit models of protein evolution. Bioinformatics
27:1164–1165.
Degnan J.H., Rosenberg N.A. 2009. Gene tree discordance,
phylogenetic inference and the multispecies coalescent. Trends
Ecol. Evol. 24:332–340.
D’Haese C. 2000. Is psammophily an evolutionary dead end?
A phylogenetic test in the genus Willemia (Collembola :
Hypogastruridae). Cladistics 16:255–273.
Dieterich C., Sommer R.J. 2009. How to become a parasite—lessons
from the genomes of nematodes. Trends Genet. 25:203–209.
Domes K., Althammer M., Norton R.A., Scheu S., Maraun M. 2007. The
phylogenetic relationship between Astigmata and Oribatida (Acari)
as indicated by molecular markers. Exp. Appl. Acarol. 42:159–171.
Dorris M., Viney M.E., Blaxter M.L. 2002. Molecular phylogenetic
analysis of the genus Strongyloides and related nematodes. Int. J.
Parasitol. 32:1507–1517.
Drummond A.J., Rambaut A. 2007. BEAST: Bayesian evolutionary
analysis by sampling trees. BMC Evol. Biol. 7:214.
Dunn C.W., Hejnol A., Matus D.Q., Pang K., Browne W.E., Smith
S.A., Seaver E., Rouse G.W., Obst M., Edgecombe G.D., Sorensen
M.V., Haddock S.H.D., Schmidt-Rhaesa A., Okusu A., Kristensen
R.M., Wheeler W.C., Martindale M.Q., Giribet G. 2008. Broad
phylogenomic sampling improves resolution of the animal tree of
life. Nature 452:745–U745.
[10:33 28/3/2013 Sysbio-syt008.tex]
VOL. 62
Fain A. 1962. Un acarien remarquable combinant les caracteres de
plusieurs families: Evansacarus lari n.g., n. sp. Bull. Ann. Soc. R.
Belge Entomol. 98:125–145.
Fain A., Guerin B., Hart B.J. 1990. Mites and allergic disease. Varennes
an Argonne, France: Allerbio.
FitzJohn R.G. 2010. Quantitative traits and diversification. Syst. Biol.
59:619–633.
Futuyma D.J., Moreno G. 1988. The evolution of ecological
specialization. Annu. Rev. Ecol. Syst. 19:207–233.
Gaede K., Knülle W. 1987. Water vapour uptake from the atmosphere
and critical equilibrium humidity of a feather mite. Exp. Appl.
Acarol. 3:45–52.
Gaud J., Atyeo W.T. 1996. Feather mites of the World (Acarina,
Astigmata): the supraspecific taxa. Part 1. Text. Koninklijk
Museum voor Midden Afrika Tervuren Belgie Annalen Zoologische
Wetenschappen 277:1–193.
Gillespie J.J., Johnston J.S., Cannone J.J., Gutell R.R. 2006.
Characteristics of the nuclear (18S, 5.8S, 28S and 5S) and
mitochondrial (12S and 16S) rRNA genes of Apis mellifera (Insecta:
Hymenoptera): structure, organization, and retrotransposable
elements. Insect Mol. Biol. 15:657–686.
Goldberg E.E., Igic B. 2008. On phylogenetic tests of irreversible
evolution. Evolution 62:2727–2741.
Hammad H., Chieppa M., Perros F., Willart M.A., Germain R.N.,
Lambrecht B.N. 2009. House dust mite allergen induces asthma via
Toll-like receptor 4 triggering of airway structural cells. Nat. Med.
15:410–416.
Heled J., Drummond A.J. 2010. Bayesian inference of species trees from
multilocus data. Mol. Biol. Evol. 27:570–580.
Holmes E.B. 1977. Is specialization a dead end. Am. Nat. 111:
1021–1026.
Holt P.G., Thomas W.R. 2005. Sensitization to airborne environmental
allergens: unresolved issues. Nat. Immunol. 6:957–960.
Janz N., Nyblom K., Nylin S. 2001. Evolutionary dynamics of hostplant specialization: a case study of the tribe Nymphalini. Evolution
55:783–796.
Kato T., Takai T., Mitsuishi K., Okumura K., Ogawa H. 2005. Cystatin
A inhibits IL-8 production by keratinocytes stimulated with Der p 1
and Der f 1: biochemical skin barrier against mite cysteine proteases.
J. Allergy Clin. Immunol. 116:169–176.
Klimov P.B., OConnor B.M. 2008. Origin and higher-level relationships
of psoroptidian mites (Acari: Astigmata: Psoroptidia): evidence
from three nuclear genes. Mol. Phylogenet. Evol. 47:1135–1156.
Knowles L., Klimov P.B. 2011. Estimating phylogenetic relationships
despite discordant gene trees across loci: the species tree of a
diverse species group of feather mites (Acari: Proctophyllodidae).
Parasitology 138:1750–1759.
Kocot K.M., Cannon J.T., Todt C., Citarella M.R., Kohn A.B., Meyer
A., Santos S.R., Schander C., Moroz L.L., Lieb B., Halanych K.M.
2011. Phylogenomics reveals deep molluscan relationships. Nature
477:452–456.
Kohlsdorf T., Lynch V.J., Rodrigues M.T., Brandley M.C., Wagner G.P.
2010. Data and data Interpretation in the study of limb evolution: a
reply to Galis et al. On the reevolution of digits in the lizard genus
Bachia. Evolution 64:2477–2485.
Krasnov B.R., Poulin R., Shenbrot G.I., Mouillot D., Khokhlova I.S. 2004.
Ectoparasitic "jacks-of-all-trades": relationship between abundance
and host specificity in fleas (Siphonaptera) parasitic on small
mammals. Am. Nat. 164:506–516.
Lanyon S.M. 1992. Interspecific brood parasitism in blackbirds
(Icterinae)—a phylogenetic perspective. Science 255:77–79.
Lee K.H., Yu H.S., Park S.K., Lee S.J., Lee K.A., Kim S.M., Ock M.S.,
Jeong H.J. 2006. Phylogeny of mite taxa (Acari:Sarcoptiformes)
based on small subunit ribosomal RNA sequences. J. Life Sci.
16:71–75.
Light J.E., Siddall M.E. 1999. Phylogeny of the leech family
Glossiphoniidae based on mitochondrial gene sequences and
morphological data. J. Parasitol. 85:815–823.
Lloyd C.M. 2009. Dust mites’ dirty dealings in the lung. Nat. Med.
15:366–367.
Loo A.H.B., Tan S.P.L., Angus A.C., Kuay K.T., Reginald K., Gao Y.F.,
Chew F.T. 2003. Genetic relationship between allergy-causing dust
mites: phylogenetic inference from random amplified polymorphic
Page: 422
411–423
2013
KLIMOV AND OCONNOR—REVERSIBILITY OF PERMANENT PARASITISM
DNA (RAPD) markers, housekeeping gene (18S rDNA), and group
2 allergens. J. Allergy Clin. Immunol. 111:S162.
Lynch V.J., Wagner G.P. 2010. Did egg-laying boas break Dollo’s law?
Phylogenetic evidence for reversal to oviparity in sand boas (Eryx:
Boidae). Evolution 64:207–216.
Maddison W.P., Midford P.E., Otto S.P. 2007. Estimating a binary
character’s effect on speciation and extinction. Syst. Biol. 56:
701–710.
Marshall D.C. 2010. Cryptic failure of partitioned Bayesian
phylogenetic analyses: lost in the land of long trees. Syst. Biol.
59:108–117.
Mathews D.H., Sabina J., Zuker M., Turner D.H. 1999. Expanded
sequence dependence of thermodynamic parameters improves
prediction of RNA secondary structure. J. Mol. Biol. 288:
911–940.
Mendonca A.G., Alves R.J., Pereira-Leal J.B. 2011. Loss of genetic
redundancy in reductive genome evolution. Plos Comput. Biol.
7:1–10.
Morse G.E., Farrell B.D. 2005. Ecological and evolutionary
diversification of the seed beetle genus Stator (Coleoptera :
Chrysomelidae : Bruchinae). Evolution 59:1315–1333.
Murray A. 1877. Economic Entomology, Aptera: prepared at the request
of the Lords of the Committee of Council on education, and
published for them. London: Chapman and Hall.
Norton R.A. 1998. Morphological evidence for the evolutionary
origin of Astigmata (Acari: Acariformes). Exp. Appl. Acarol. 22:
559–594.
Nosil P., Mooers A.O. 2005. Testing hypotheses about ecological
specialization using phylogenetic trees. Evolution 59:2256–2263.
Nylander J.A., Wilgenbusch J.C., Warren D.L., Swofford D.L. 2008.
AWTY (are we there yet?): a system for graphical exploration
of MCMC convergence in Bayesian phylogenetics. Bioinformatics
24:581–583.
Nylander J.A.A. 2011. Burntrees v. 0.1.9. Perl script for manipulating
MrBayes tree and parameter files. Program distributed by the
author. Uppsala (Sweden): Evolutionary Biology Centre, Uppsala
University.
OConnor B.M. 1982. Astigmata. In: Parker S.P., editor. Synopsis and
classification of living organisms. New York: McGraw-Hill Book
Company, p. 146–169.
Ong S.T., Chew F.T. 2010. Reconstructing the repertoire of mite
allergens by recombinant DNA technology. In: Pawankar R.,
Holgate S.T., Rosenwasser L.J., editors. Allergy frontiers: future
perspectives. Japan: Springer. p. 49–64.
Pagel M. 1994. Detecting correlated evolution on phylogenies: a general
method for the comparative analysis of discrete characters. Proc. R.
Soc. Lond. B Biol. Sci. 255:37–45.
Pagel M., Meade A., Barker D. 2004. Bayesian estimation of ancestral
character states on phylogenies. Syst. Biol. 53:673–684.
Paradis E., Claude J., Strimmer K. 2004. APE: analyses of phylogenetics
and evolution in R language. Bioinformatics 20:289–290.
Parfrey L.W., Grant J., Tekle Y.I., Lasek-Nesselquist E., Morrison
H.G., Sogin M.L., Patterson D.J., Katz L.A. 2010. Broadly sampled
multigene analyses yield a well-resolved eukaryotic tree of life. Syst.
Biol. 59:518–533.
Pepato A.R., da Rocha C.E., Dunlop J.A. 2010. Phylogenetic position of
the acariform mites: sensitivity to homology assessment under total
evidence. BMC Evol. Biol. 10:235.
Prendini L., Francke O.F., Vignoli V. 2010. Troglomorphism,
trichobothriotaxy and typhlochactid phylogeny (Scorpiones,
Chactoidea): more evidence that troglobitism is not an evolutionary
dead-end. Cladistics 26:117–142.
Price P.W. 1980. Evolutionary biology of parasites. Princeton (NJ):
Princeton University Press.
Proctor H.C. 2003. Feather mites (Acari: Astigmata): ecology, behavior
and evolution. Annu. Rev. Entomol. 48:185–209.
Proctor H.C. 2012. List of feather mites and their hosts [Internet].
Available from: URL http://www.biology.ualberta.ca/faculty/
heather_proctor/?Page=5626.
R Development Core Team. 2010. R: a language and environment
for statistical computing, reference index version 2.11.1 [Internet].
[10:33 28/3/2013 Sysbio-syt008.tex]
423
Vienna (Austria): R Foundation for Statistical Computing. ISBN
3-900051-07-0. Available from: URL http://www.R-project.org.
Radovsky F.J., Krantz G.W., Whitaker J.O. 1997. A remarkable example
of predation in the parasitic mite family Macronyssidae. Int. J.
Acarol. 23:3–6.
Rambaut A., Drummond A.J. 2009. Tracer v1.5 [Internet]. Available
from: URL http://beast.bio.ed.ac.uk/Tracer.
Ronquist F., Huelsenbeck J.P. 2003. MRBAYES 3: Bayesian phylogenetic
inference under mixed models. Bioinformatics 19:1572–1574.
Rothfels C.J., Larsson A., Kuo L.Y., Korall P., Chiou W.L., Pryer K.M.
2012. Overcoming deep roots, fast rates, and short internodes to
resolve the ancient rapid radiation of eupolypod II ferns. Syst. Biol.
61:490–509.
Sakharkar K.R., Dhar P.K., Chow V.T.K. 2004. Genome reduction
in prokaryotic obligatory intracellular parasites of humans: a
comparative analysis. Int. J. Syst. Evol. Microbiol. 54:1937–1941.
Salewski V. 2003. Satellite species in lampreys: a worldwide trend for
ecological speciation in sympatry? J. Fish Biol. 63:267–279.
Shimodaira H. 2001. Multiple comparisons of log-likelihoods and
combining nonnested models with applications to phylogenetic tree
selection. Commun. Stat. Theory Methods 30:1751–1772.
Shimodaira H. 2002. An approximately unbiased test of phylogenetic
tree selection. Syst. Biol. 51:492–508.
Siddall M.E., Brooks D.R., Desser S.S. 1993. Phylogeny and the
reversibility of parasitism. Evolution 47:308–313.
Sidenius K.E., Hallas T.E., Poulsen L.K., Mosbech H. 2001. Allergen
cross-reactivity between house-dust mites and other invertebrates.
Allergy 56:723–733.
Smith B.P. 1998. Loss of larval parasitism in parasitengonine mites. Exp.
Appl. Acarol. 22:187–199.
Stireman J.O. 2005. The evolution of generalization? Parasitoid flies
and the perils of inferring host range evolution from phylogenies.
J. Evol. Biol. 18:325–336.
Struck T.H., Paul C., Hill N., Hartmann S., Hosel C., Kube M.,
Lieb B., Meyer A., Tiedemann R., Purschke G., Bleidorn C. 2011.
Phylogenomic analyses unravel annelid evolution. Nature 471:
95–98.
Suarez-Martinez E.B., Montealegre F., Sierra-Montes J.M., Herrera R.J.
2005. Molecular identification of pathogenic house dust mites using
12S rRNA sequences. Electrophoresis 26:2927–2934.
Sukumaran J., Holder M.T. 2010. DendroPy: a Python library for
phylogenetic computing. Bioinformatics 26:1569–1571.
Takebayashi N., Morrell P.L. 2001. Is self-fertilization an evolutionary
dead end? Revisiting an old hypothesis with genetic theories and a
macroevolutionary approach. Am. J. Bot. 88:1143–1150.
Termonia A., Hsiao T.H., Pasteels J.M., Milinkovitch M.C. 2001. Feeding
specialization and host-derived chemical defense in chrysomeline
leaf beetles did not lead to an evolutionary dead end. Proc. Natl.
Acad. Sci. U. S. A. 98:3909–3914.
Thomson Reuters. 2012. Index to Organism Names (ION) [Internet].
Available from: URL http://www.organismnames.com/.
Toft C.A., Aeschlimann A., Bolis L. 1991. Parasite-host associations:
coexistence or conflict? New York: Oxford University Press.
Tovey E.R., Chapman M.D., Platts-Mills T.A. 1981. Mite faeces are a
major source of house dust allergens. Nature 289:592–593.
Wiens J.J. 2011. Re-evolution of lost mandibular teeth in frogs after more
than 200 million years, and re-evaluating Dollo’s law. Evolution
65:1283–1296.
Wilson D.E., Reeder D.M. 2005. Mammal species of the world. A
taxonomic and geographic reference. 3rd ed. Baltimore (MD): The
Johns Hopkins University Press.
Yang B., Cai J., Cheng X. 2010. Identification of astigmatid mites using
ITS2 and COI regions. Parasitol. Res. 108:497–503.
Yunker C.E. 1955. A proposed classification of the Acaridiae (Acarina:
Sarcoptiformes). Proc. Helminthol. Soc. Wash. 22:98–105.
Zuker M. 2003. Mfold web server for nucleic acid folding and
hybridization prediction. Nucleic Acids Res. 31:3406–3415.
Zwickl D.J. 2006. Genetic algorithm approaches for the phylogenetic
analysis of large biological sequence datasets under the maximum
likelihood criterion [Ph.D. dissertation]. [Austin (TX)]: The
University of Texas at Austin.
Page: 423
411–423