Parasitology International 60 (2011) 480–487
Contents lists available at SciVerse ScienceDirect
Parasitology International
j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / p a r i n t
Morphometric and molecular characterization of Gyrodactylus teuchis Lautraite,
Blanc, Thiery, Daniel & Vigneulle, 1999 (Monogenea: Gyrodactylidae) from an
Austrian brown trout population
Christoph Hahn a,⁎, Tor A. Bakke a, Lutz Bachmann a, Steven Weiss b, Phil D. Harris a
a
b
Natural History Museum, University of Oslo, PO Box 1172 Blindern, 0318 Oslo, Norway
Institut für Zoologie, Karl-Franzens Universität Graz, Universitätsplatz 2, 8010 Graz, Austria
a r t i c l e
i n f o
Article history:
Received 8 May 2011
Received in revised form 8 August 2011
Accepted 11 August 2011
Available online 22 August 2011
Keywords:
Gyrodactylus teuchis
Gyrodactylus salaris
Monogenea
Atlantic salmon
Morphometry
Cryptic species
a b s t r a c t
Gyrodactylus teuchis is a widespread parasite of wild and farmed salmonids throughout Europe. It has been
frequently confused with the notifiable pathogen G. salaris, to which it bears a striking morphological
similarity. The species is frequently referred to as ‘cryptic’, and diagnoses are primarily based on molecular
evidence. We provide the first comprehensive re-description of G. teuchis from a natural wild brown trout
population in the Danube watershed, based on the state of the art morphometrics in addition to standard
molecular markers. We demonstrate that despite the lack of uni-variate diagnostic character measurements,
G. teuchis can be reliably distinguished from G. salaris using multivariate morphological approaches such as
Principal Component Analysis or Canonical Variate Analysis, suggesting that automated diagnostic
approaches for G. salaris can be modified to take account of potential G. teuchis in samples. This is the first
record of G. teuchis from a host population unlikely to have been modified by human stocking efforts. The
morphological variability observed in the samples collected from one site on 1 day reflects the overall level of
variation reported for European G. teuchis. We also report new sequence variants of the internal transcribed
spacer (ITS-1) of the nuclear ribosomal gene cluster with evidence for intra-individual heterogeneity of ITS-1
within this population of G. teuchis.
© 2011 Elsevier Ireland Ltd. All rights reserved.
1. Introduction
Gyrodactylus teuchis Lautraite, Blanc, Thiery, Daniel & Vigneulle,
1999 is the least studied Eurasian gyrodactylid infecting trout and
salmon, and as such is a source of confusion in the diagnosis of the
salmon pathogen Gyrodactylus salaris Malmberg, 1957. There is a
striking overlap in host range for the two species [1], and, according to
published accounts [2,3], the two species are said to show only subtle
differences in hook morphology. Initially G. teuchis was misidentified
based upon inadequate characterization of the V4 ribosomal fragment, leading to the erroneous reporting of G. salaris from Portugal
and France [4]. The initial description of G. teuchis corrected this error
[2], but within 2 years a re-description was necessary because the first
description failed to specify the type locality, deposit type material
and extensively describe the species morphologically [3]. Even so,
there is still insufficient morphological data to unambiguously
distinguish G. teuchis from G. salaris, and G. teuchis is considered to
be a cryptic species, which can only be adequately discriminated from
G. salaris using molecular approaches [2,3,5]. In Shinn et al. [6],
⁎ Corresponding author. Tel.: + 47 22851823; fax: + 47 22851837.
E-mail address: christoph.hahn@nhm.uio.no (C. Hahn).
1383-5769/$ – see front matter © 2011 Elsevier Ireland Ltd. All rights reserved.
doi:10.1016/j.parint.2011.08.016
appraising G. salaris diagnostic facilities, a fundamental conclusion
was the vulnerability of molecular diagnosis to systematic laboratory
failure. Morphological diagnosis represents a potentially important
control for the molecular identification and an important failsafe for
diagnosis of G. salaris, but presently is confounded by the lack of a
precise morphological description of G. teuchis. This hinders the
training of automated recognition systems [7,8] for identifying G.
salaris. Since an increasing body of work [2,3,5,6,9–15] suggests that G.
teuchis may be one of the most common gyrodactylids of European
salmonids, we have undertaken the current study to provide
comprehensive measurements of the anchors, bars and marginal
hooks of an Austrian population of this species, and to provide a
detailed comparison of morphology with that of its congeners, G.
salaris and G. thymalli.
2. Materials and methods
2.1. Parasites
Brown trout (Salmo trutta L.) were sampled from the river Kleiner
Kamp in Lower Austria (48°31, 15'N, 15°05, 53'E, water temperature:
5–6 °C) using a back-pack electric-fishing unit on a single day in
November 2008. Kleiner Kamp is a crystalline stream draining into the
C. Hahn et al. / Parasitology International 60 (2011) 480–487
river Kamp, a major Austrian tributary of the Danube. Selected 0+
trout (TL 6–10 cm, n = 26) were killed by a blow to the head and
stored immediately in 96% ethanol, before screening for Gyrodactylus
using a stereomicroscope at the Institute for Zoology, University of
Graz. Parasites were removed from the host fish and stored in 96%
ethanol prior to further processing at the Natural History Museum
Oslo. Epidemiological characteristics obtained for the parasite
population follow the definitions in Bush et al. [16]. For morphological
comparison, the two available paratypes of G. teuchis (NHM
2000.1.18.2-3) from French rainbow trout (Oncorhynchus mykiss
Walbaum) were borrowed from the NHM London. Comparative data
of G. salaris from the Norwegian river Drammenselva (n = 12), a
paralectotype (n = 1) and G. thymalli (n = 15) from the type locality
River Hnilec in Slovakia were kindly provided by Kjetil Olstad and are
more fully described in Olstad et al. [17].
2.2. Molecular analysis
Individual parasites were placed on polylysine coated microscopy
slides (Menzel-Gläser). A clean insect pin was used to cut the
opisthaptor from the body which was transferred in 5 μl 96% ethanol
to a sterile Eppendorf-tube. Total genomic DNA was extracted using
the E.Z.N.A.® Tissue DNA Kit (Omega Bio-Tek), with inclusion of an
overnight lysis step, and elution of DNA in a final volume of only
100 μl. The primer pair ITS-1A and ITS-2 [9] was used to amplify a ca
1300 bp fragment spanning the 3′-end of the 18S rDNA gene, ITS-1,
5.8S gene, and ITS-2 to the 5′ end of the 28S rDNA gene. Amplification
reactions contained 10 μl of 2× AmpliTaq Gold® Fast PCR Master Mix
(Applied Biosystems), 0.5 pM of each primer and an empirically
determined suitable amount of genomic Gyrodactylus template DNA.
The amplification protocol consisted of an initial denaturation step at
94 °C for 3 min, followed by 35 cycles of 94 °C for 30 sec, 55 °C for
20 sec, and 72 °C for 1 min, and a final extension at 72 °C for 3 min.
PCR-products were sequenced directly using either the initial PCRprimers or the internal primers ITS-3 and ITS-4.5 [9], and BigDye 3.1
chemistry (Applied Biosystems). Sequences were edited and aligned
using Sequencher 4.1.4 software (Gene Codes Corporation). Species
identification of G. teuchis was confirmed against GenBank entries
using the BLAST search algorithm [18]. In some individuals an
ambiguous site at ITS-1 position 638 could not be resolved. This
position formed part of the restriction site for the endonuclease BclI
(5′-TGATCA-3′), which was subsequently used to explore this
polymorphism. ITS-1 was amplified using conditions described
above using the primer pair ITS-1A and ITS-3 [9]. The products
(3 μl) were digested for 2 hours at 37 °C in a final volume of 10 μl.
Restriction patterns were then analyzed using an automated
electrophoresis system (Experion, Bio-Rad Laboratories).
2.3. Morphological and statistical analysis
The opisthaptors of individual parasites were processed as
previously described [19] with slight modifications. Enzyme digestion
was carried out directly on polylysine coated slides (Menzel-Gläser),
substantially improving hook adherence. Individual opisthaptors
were covered with 2 μl of digestion solution (Buffer TL from the
E.Z.N.A.® Tissue DNA Kit, Omega Bio-Tek supplemented with
10 mg/μl proteinase K) and incubated at room temperature under
continuous optical control. The digestion was stopped by removing
the digestion solution and washing twice in distilled H2O. Specimens
were mounted in 0.1% Sodium Dodecyl Sulphate and a Leica DM
6000B stereomicroscope equipped with a Leica DC 500 camera used to
obtain digital images of the anchors, marginal hooks and bars.
Morphological characters were measured using the Leica Application
Suite (version 2.6.0 R1). All analyses were performed by CH. Table 1
presents 32 morphological measurements used in the present study.
Apart from hamulus anchor shaft length (HASL) and ventral bar
481
Table 1
Morphological characters obtained in the present study. 1 Shinn et al. [21], 2 Olstad et al.
[17], 3 Mo [20], 4 introduced in present study. Characters marked with an asterisk were
used for PCA.
Character
number
Character
abbreviation
Character description
Hamuli
1
2
3
4
5
6
7
8
9
10
11
12
13
14
HAD1,2*
HPSW1,2*
HPL1,2*
HSL1,2*
HAA1
HIA1,2*
HRL1,2*
HTL1,2*
HDSW1,2*
HICL1,2*
HPCA1
HASL3
HDAA4
HTDA4
Hamulus
Hamulus
Hamulus
Hamulus
Hamulus
Hamulus
Hamulus
Hamulus
Hamulus
Hamulus
Hamulus
Hamulus
Hamulus
Hamulus
Ventral bar
15
16
17
18
19
20
21
22
23
24
VBTW1*
VBTL1,2*
VBPML1,2*
VBML1,2*
VBMBL1,2*
VBCL2*
VBMMW2*
VBLL2*
VBPL1,2*
VBPDD3
Ventral
Ventral
Ventral
Ventral
Ventral
Ventral
Ventral
Ventral
Ventral
Ventral
Marginal hooks
25
26
27
28
29
30
31
32
MHTL1,2*
MHSHL1,2*
MHSL1,2*
MHIH1,2*
MHSPW1,2*
MHSTL1,2*
MHSDW1,2*
MHAD1,2*
Marginal
Marginal
Marginal
Marginal
Marginal
Marginal
Marginal
Marginal
aperture distance
proximal shaft width
point length
shaft length
aperture angle
inner aperture angle
root length
total length
distal shaft width
inner curve length
point curve angle
anchor shaft length
distal aperture angle
tip-dorsal bar attachment angle
bar total width
bar total length
bar process-to-mid length
bar median length
bar membrane length
bar centre length
bar membrane maximal width
bar lateral length
bar process length
bar process distal distance
hook total length
hook shaft length
hook sickle length
hook instep/arch height
hook sickle proximal width
hook sickle toe length
hook sickle distal width
hook aperture distance
process distal distance (VBPDD) (corresponding to las and mdpvb in
[20]) the original character annotations have been retained as
introduced in previous works [17,21]. Hamulus distal aperture angle
(HDAA) and hamulus tip-dorsal bar attachment angle (HTDA) were
used for the first time in the present study, and are illustrated in Fig. 1.
Each character was measured three times per individual structure.
Statistical analyses were conducted using R version 2.12.0 (http://
www.r-project.org/). Morphological measurements were tested for
normality using the Shapiro–Wilk test and the slightly less conservative
one-sample Kolmogorov–Smirnov test. Ties in the data set were dealt
with by adding a randomly generated 10 −7 increment to each
measurement. The program PAST v. 2.07 [22] was used to perform
Principal Component Analysis (PCA) and Canonical Variate Analysis
(CVA) on a dataset consisting of the 26 character measurements
available for all three species G. teuchis, G. salaris and G. thymalli. The
newly introduced measurements HDAA and HTDA are not included in
the analyses, as they were not available for the reference material.
Further studies are therefore necessary to assess their usefulness for
species discrimination. Only individuals for which all opisthaptoral
measurements were available were included in the analyses.
3. Results
3.1. Epidemiology
A total of 26 S. trutta from the river Kleiner Kamp were processed in
this study. The prevalence of Gyrodactylus was 100% (mean intensity 14,
482
C. Hahn et al. / Parasitology International 60 (2011) 480–487
accession numbers JN628863-JN628864, and are compared with ITS
sequences from G. teuchis across Europe in Table 2.
Analyzing the variable ITS-1 position 638 using the endonuclease
BclI, confirmed the results of direct sequencing. Two distinct restriction
patterns were observed (see Fig. 2). For specimens displaying an
unambiguous A in the sequencing trace, the 740 bp PCR product was
always digested completely, resulting in two fragments of 85 bp and
655 bp. Specimens displaying the A/G ambiguity in the sequencing trace
always showed three fragments of 740 bp, 655 bp and 85 bp length after
digestion, inferring the presence of two distinct copies of ITS-1,
differentiated by a transition in position 638 and therefore selectively
targeted by the endonuclease BclI. Analyses of restriction chromatograms of the 20 individuals displaying this pattern revealed that the “Acopy” and the “G-copy” composed 40.9 ± 8.9% and 59.1 ± 8.9% of the
ITS-1 copies, respectively.
3.3. Morphological and statistical analyses
Fig. 1. Light microscope image of one of the anchors of G. teuchis illustrating the two
new morphological angle measurements introduced in this study. A—HDAA (Hamulus
distal aperture angle), B—HTDA (Hamulus tip-dorsal bar attachment angle). Scale
bar = 10 μm.
range 5–100 worms per fish), and the parasites occurred on both the
body and the fins. A total of 46 parasites randomly picked from the 26
trout were unambiguously identified as G. teuchis based on direct
sequencing of a nuclear ribosomal DNA fragment (ITS1 or ITS2), and no
other species were detected on the sampled fish. Only measurements
from specimens genetically confirmed as G. teuchis by direct sequencing
were included in the morphometric dataset.
3.2. Molecular analyses
Both strands of the 1371 bp nuclear ribosomal DNA fragment were
fully sequenced for 31 individuals. ITS-2 was invariant, whereas some
minor sequence variation was detected in ITS-1. A frequently
observed sequence ambiguity (in 23 individuals, 74.2%) was due to
a variable 6–7 A homopolymer at positions 45–51. The remaining eight
individuals (25.8%) displayed an unambiguous 6 A homopolymer at this
position. An A/G ambiguity was found at position 638 in ITS-1 for 20
(64.5%) G. teuchis specimens while there was an unambiguous A for 11
(35.5%) individuals. These ITS-1 sequences differed from GenBank
accession AJ249350[3], which includes the 6A tract while having the
unambiguous A at position 638; these new sequences are deposited under
Table A1 (see appendix) summarizes all morphological measurements for G. teuchis from river Kleiner Kamp, Austria. Previously
published morphological data for G. teuchis [2,3,10–12] are also
included, with the measurements that could be obtained from the
available paratypes. The two hook measurements (HTL: ~68 μm and
MHTL ~ 36 μm) reported by Paladini et al. [13] for G. teuchis from
farmed Italian rainbow trout were not included, but they fall within
the range observed by other authors. The overall reported morphological variability of European G. teuchis is similar to that seen in this
Austrian population. The only specimen which differs substantially in
a number of measurements (i.e. HAD, HIA, HAA, HDAA) is the
paratype (NHM 2000.1.18.2-3), reflecting obvious distortion due to
excessive pressure applied to the cover slip during preparation of the
slides. The hooks and bars of G. teuchis from river Kleiner Kamp are
illustrated in Fig. 3.
When analyzing the 26 morphometric characters utilized in the PCA
(see Table 1) for G. teuchis the Shapiro–Wilk test rejected normality for 3
(11.5%) of the characters (HPSW, p = 0.02366; HPL, p = 0.000839;
VBTW, p = 0.00665), while the less conservative one-sample Kolmogorov–Smirnov test did not indicate any significant deviations from a
normal distribution. However, Shapiro–Wilk test rejected normality for
4 (15.4%) morphometric characters in the G. salaris (MHSPW,
p = 0.02179; MHSDW, p = 0.003473; MHAD, p = 0.000455; VBML,
p = 0.008325) and G. thymalli (MHSTL, p = 0.04005; MHSDW,
p = 0.02362; HAD, p = 0.002947, HIA, p = 0.001078) datasets [17],
respectively. Therefore parametric statistics could only be applied to 16
measurements distributed normal across all three species. Accordingly,
Student's t-test (for the 16 characters which were normally distribution
in all three species) and the non-parametric Mann–Whitney U test (for
the 10 characters for which normality was rejected in one or more
species) were used to assess pairwise differences in each character
between the species G. teuchis, G. salaris and G. thymalli. Analyzing
the differences between the species pairs G. teuchis/G. salaris, G. teuchis/
G. thymalli and G. salaris/G. thymalli revealed that out of the 26
morphological characters a number of 15 (57.7%), 21 (80.8%) and 22
Table 2
Comparison of ITS rDNA nucleotide sequences for G. teuchis, in relation to geographical origin and host species of the variant. ST, Salmo trutta; OM, Oncorhynchus mykiss; f, farmed; w,
wild.
Gyrodactylus teuchis strain
ITS-1 position
ITS-2 position
Country
Host
Genbank accession
Reference
N
98
188
276
638
58
100
173
180
255
Austria
Austria
France, Denmark
Italy
Poland
Poland
ST (w)
ST (w)
Various
OM (f)
OM/ST (f)
OM/ST (f)
JN628863
JN628864
AJ249350
–
EF464679
EF464680
present study
present study
3
12
9
9
11
20
4/3
3
A
–
T
T
A
–
W
W
G
–
A
A
A
R
A
A
A
–
–
–
–
R
T
–
–
–
–
Y
A
–
–
–
–
R
G
–
–
–
–
R
C
–
–
–
–
Y
C. Hahn et al. / Parasitology International 60 (2011) 480–487
483
A
B
C
Fig. 2. Experion gel chip image and chromatograms illustrating the digestion patterns of PCR-products spanning ITS-1 digested with the restriction enzyme BclI. Simulated gel image:
lanes L and 10—Ladder, lane 1—undigested PCR product (740 bp, corresponds to chromatogram A); lanes 2–5—ITS-1 position 638 A, showing fully digested fragments of 655 and
75 bp respectively (corresponds to chromatogram B); lanes 6–9—ITS-1 position 638 A/G, showing digested 655 and 75 bp fragments, and the undigested G allele fragment (740 bp)
(chromatogram C).
(84.6%), respectively, differed significantly (see Table 3 for G. teuchis vs.
G. salaris). G. teuchis was relatively easy to distinguish from G. thymalli
based on point-to-point measurements, as it was generally smaller than
the latter species. The morphological characters HAD, HSL, HTL, VBCL,
VBMMW, MHTL and MHSHL are diagnostic in this case, as no overlap in
range was observed between the measured populations of G. teuchis and
G. thymalli for any of these characters. G. teuchis was more similar in size
to G. salaris, and although significant differences were observed
between these species for 15 characters of hamuli, ventral bar and
marginal hooks (see Table 3), the ranges for all 26 compared characters
overlapped between species. As seen no individual character was per se
diagnostic; it is the combination of characters that allows for species
discrimination (see Section 3.4 below). The characters MHSDW
(marginal hook sickle distal width) and MHAD (marginal hook aperture
distance) showed the smallest range overlap between G. salaris and G.
teuchis. Both measurements were significantly larger in G. teuchis
reflecting the qualitative visual observation of differences in the sickle
blade between the species as mentioned, but not quantified in previous
accounts [2,3]. Fig. 4 illustrates differences in marginal hook sickle shape
between G. teuchis and G. salaris utilizing selected morphological
characters. The marginal hook sickles of all three species are compared
in Fig. 5. Overall no clear size tendency can be observed as G. teuchis
appears to be significantly larger in the characters MHSPW, MHSDW,
MHAD, HPL, HDSW, VBPML, VBTW, VBMMW and VBPL, while
significantly smaller in MHTL, MHSHL, HAD, HIA, HICL and HRL in
comparison to G. salaris.
3.4. Principle component analysis (PCA)
Despite the overall similarity in size between G. salaris and G.
teuchis, PCA based on the variance–covariance matrix of 26 morphological characters readily discriminated G. teuchis from the morphologically similar G. salaris and G. thymalli (Fig. 6A). The eigenvalues of
the PC1 and PC2 were 114.8 and 16.3, accounting for 73.8% and 10.5%
of the observed variance, respectively. The loadings of PC1 were
mostly positive, confirming the well-known relationship of this first
component with overall size; PC2–PC6, accounting for 21.1% of the
total variance, are interpreted as reflecting shape, as the loadings were
both positive and negative.
Given that PCA was able to distinguish the three species clearly,
canonical variate analysis (CVA) was attempted after a priori
assignation of specimens to species groups. Assignation of G. teuchis
was based on molecular identification based on the rDNA sequence;
assignation of G. thymalli and G. salaris followed Olstad et al. [17].
Wilks' lambda, a test statistic conducted in the course of multivariate
analysis of variance (MANOVA), revealed significant differences
between the multidimensional means (calculated as a combination
of the 26 morphological characters) of the three species (Wilks'
lambda p = 1.6 −15). A CVA scatter-plot illustrating maximum separation between species is presented in Fig. 6B.
3.5. Formal redescription of Gyrodactylus teuchis
Family Gyrodactylidae Cobbold, 1864
Genus Gyrodactylus v. Nordmann, 1832
Gyrodactylus teuchis Latraite, Blanc, Thiery, Daniel & Vigneulle,
1999
Type-host: Rainbow trout, Oncorhynchus mykiss Walbaum
(Salmonidae)
Other hosts: Atlantic salmon, Salmo salar L. (Salmonidae); Brown
trout, Salmo trutta L. (Salmonidae), Brook trout, Salvelinus fontinalis
Mitchill (Salmonidae)
Site on host: Ectoparasite on fins and body skin.
Type material: Holotype (NHM 2000.1.18.1) and Paratypes (NHM
2000.1.18.2-3) from farmed rainbow trout in France (type locality not
specified: Brittany and the Western Pyrenees), deposited in the
Natural History Museum London [3].
Other localities/distribution: The species is widely distributed in
Europe, but appears absent in Fennoscandia and Iceland. The parasite
was originally reported from farmed and wild brown and rainbow
trout, and from wild Atlantic salmon in Brittany and the western
Pyrenees, France [2]. Reports from farmed rainbow trout in Lanarkshire, Scotland [3], and wild brown trout in mainland Jutland
(Denmark) and the island of Bornholm (politically part of Denmark)
484
C. Hahn et al. / Parasitology International 60 (2011) 480–487
Fig. 3. Comparison of light microscope images of the opisthaptoral hard parts obtained after digestion of randomly selected specimens of genetically identified G. teuchis from wild
brown trout Salmo trutta. A. Hamulus. B. Ventral bar. C. Marginal hook. Scale bar = 10 μm.
were added by Buchmann et al. [14,15] and Cunningham et al. [3].
Matejusová et al. [9] recorded this species from brown trout and
brook trout in the Czech Republic. G. teuchis was also recorded from
farmed rainbow trout and brown trout in Poland [10] and from
farmed rainbow trout in Italy [13] and Germany [12]. Records from
wild Atlantic salmon in Danish rivers were reported by von Gersdorff
Jørgensen et al. [11]. The present study is based on material from a
wild brown trout population in river Kleiner Kamp, Austria (48°31,
15′N, 15°05, 53′E) and represents the first record for a Danube
tributary (Kamp).
Table 3
Assessment of differences in morphometric characters between G. teuchis and G. salaris.
Level of significance: *0.005 b p b 0.05, **p b 0.005.
G. teuchis vs. G. salaris
Mann–WhitneyU
1
2
3
6
15
18
29
30
31
32
Student's t
HAD**
HPSW
HPL**
HIA**
VBTW**
VBML
MHSPW**
MHSTL
MHSDW**
MHAD**
4
7
8
9
10
16
17
19
20
21
22
23
25
26
27
28
HSL
HRL**
HTL
HDSW**
HICL**
VBTL
VBPML**
VBMBL
VBCL
VBMMW**
VBLL
VBPL*
MHTL*
MHSHL*
MHSL
MHIH
Fig. 4. Differences in marginal hook sickle shape between G. teuchis (bottom right,
closed squares) and G. salaris (top left, open squares) utilizing selected morphological
characters. MHSTL—marginal hook sickle toe length, MHSPW—marginal hook sickle
proximal width, MHSDW—marginal hook sickle distal width.
C. Hahn et al. / Parasitology International 60 (2011) 480–487
Fig. 5. Comparative light microscope images of marginal hook sickles for G. salaris
(A), G. teuchis (B) and G. thymalli (C). Scale bar = 10 μm.
Additional deposited material: Further specimens from Kleiner Kamp,
Austria, prepared for hook analysis (2 specimens, NHM C5281-C5282),
five whole specimens in Canada balsam (NHM C5277-C5280) and a
collection of brown trout fin clips with attached parasites preserved in
96% ethanol (NHM C5283) have been deposited in the NHM Oslo.
Additional hook preparations of G. teuchis have been deposited in The
Natural History Museum, London (2 specimens, NHMUK 2011.9.9.1-2)
and at the Natural History Museum, Vienna (2 specimens, NHMW Ev
varia Mikro 5581-5582).
Microscopical diagnosis: Fig. 3 illustrates the attachment hooks of G.
teuchis. Table A1 (see appendix) contains detailed opisthaptoral
measurements obtained from the Kleiner Kamp population, including
previously published data for G. teuchis [2,3,10–12]. Body measurements, based on the whole parasite mounts deposited in Oslo, are as
following: body spindle-shaped, 327–421 μm long and 79–110 μm at
the widest part of mid-body; opisthaptor, 69–95 μm long and 54–66 μm
wide. The penis was spherical, 11–13 μm in diameter and armed with
one large and 6 small spines. Hamuli typical for the G. wageneri species
group, 69.6 (65.5–74.1) μm in total length with roots 22.7 (18.3–25.3)
μm and points 36.8 (32.4–38.5) μm. Marginal hooks 37.3 (34.2–39.4),
shafts 30.1 (27.6–32.2), and sickles 7.8 (7.2–8.7) lengths; sickle shape
typical of the G. wageneri group. For total European hook variation see
Table A1 in appendix.
Fig. 6. PCA plot (A) and CVA plot (B) of the species G. teuchis, G. salaris and G. thymalli using
26 morphometric variables (see Table 1). Crosses—G. teuchis, open squares—G. salaris,
black squares—G. thymalli. Ellipses in (A) represent 95% confidence intervals about the
mean.
485
Molecular diagnosis: Nucleotide sequences of the rDNA gene
cluster stretching the 3′-end of the 18S subunit, ITS1, 5.8S gene,
ITS2 and the 5′-end of the 28S ribosomal subunit have been deposited
in GenBank under accession number AJ249349 (V4) and AJ249350
(ITS) [3]. Rokicka et al. [10] published an alternative ITS2 sequence
from a specimen from Polish farmed rainbow trout (GenBank
accession no. EF464680). Both ITS and V4 appear to be good species
markers according to Cunningham et al. [5]. The new ITS-1 and ITS-2
sequences for the Kleiner Kamp population obtained during this study
were deposited under accession no. JN628863 and JN628864, respectively. Published ITS sequences for G. teuchis are summarized in Table 2.
4. Discussion
The present study provides the first record of a Gyrodactylus teuchis
population from a wild brown trout population collected in the Danube
watershed. We present a comprehensive re-description of G. teuchis and
provide for the first time high resolution images of the anchors, marginal
hooks and bars of this species, as well as a detailed comparison with G.
salaris and G. thymalli. We show that it is possible to reliably distinguish
G. teuchis from G. salaris based on morphological data alone. G. teuchis is
clearly an abundant ectoparasite in Europe, frequently recorded from
both farmed and wild salmonids, including farmed and wild Salmo
trutta, Oncorhynchus mykiss and wild Salmo salar in Brittany and the
western Pyrenees in France [2], from farmed O. mykiss in Lanarkshire,
Scotland [3], and wild S. trutta in Sønderjyllands Amt (Jutland) and
Gyldenså (Bornholm), Denmark [3,14,15], from wild S. salar in Denmark
[11], S. trutta and Salvelinus fontinalis in the Czech Republic [9], and from
farmed Oncorhynchus mykiss in Germany [12], Poland [10] and Italy [13],
but it is generally considered harmless and non-pathogenic. As it infects
cultivated salmonids G. teuchis has been extensively distributed through
the aquaculture industry. It is interesting to note that the river Kleiner
Kamp population represents the first natural locality for G. teuchis in the
Danube basin. Kleiner Kamp has never been glaciated [23], and only
indigenous brown trout are thought to have been stocked. Most
importantly there is no evidence of stocking with rainbow trout. As
there is no evidence of anthropogenic introduction ([24], and
unpublished), the local trout population is therefore most likely to be
of natural origin. The Kleiner Kamp is so far the only Austrian river in
which the trout population seems to be infested solely with G. teuchis.
Although G. teuchis appears widespread in Austria, all other sampled
trout populations have shown a mixed pattern of infection with both G.
teuchis and the more abundant G. truttae (unpublished).
The ectoparasitic G. teuchis appears to be a non-pathogenic
generalist of salmonid fishes, able to infect at least four species of
three salmonid genera. In addition to the well-known susceptibility of
Atlantic salmon, brown trout, rainbow and brook trout, it has also
been found occasionally on European grayling (Thymallus thymallus
L.) in the Austrian Danube basin (personal observation). The
paradigm of gyrodactylids as relatively host specific ectoparasites
has been reassessed in recent years. According to Bakke et al. [25, rev.
in 26] only some 30% of the described Gyrodactylus species exclusively
infect a single host species (see also www.GyroDb.net). For G. teuchis,
it can only be speculated which species is the plesiomorphic host, and
it would be an interesting question as to whether this parasite species
originally was a generalist and therefore able to exploit the spread of
salmonids in aquaculture, or whether aquaculture has led to the
diversification of this species onto a range of salmonids. The relative
performance of G. teuchis on its various distinct host species and
strains remains also to be tested experimentally.
The morphological variation found in the G. teuchis population
from 26 brown trout collected on the same day in the river Kleiner
Kamp is of the same order of magnitude as that found for the entire
European G. teuchis metapopulation (see Table A1 in appendix). This
is a slightly surprising observation, as e.g. Lautraite et al. [2] provided
measurements from parasites pooled from several different localities
486
C. Hahn et al. / Parasitology International 60 (2011) 480–487
with varying water temperatures (6–15 °C). Temperature is well
known to affect size and shape of the hooks (rev. in [26], see also [27]).
The observation is reminiscent of that of Harris [28], who observed as
great a variation in inbred lines of G. gasterostei as in natural
populations, and concluded that much of the observed variation in
gyrodactylid populations must be environmental in origin, rather than
genetic.
Despite the restricted sample size of G. teuchis, it is noteworthy that
we detected novel ITS variants that are so far unique to the Danube
basin. ITS-1 differs by at least three nucleotides between Eastern
populations and the original western isolates sequenced by Cunningham et al. [3]. Matejusová et al. [9] likewise mentioned a different ITS-1
pattern, probably identical to the one collected here, but did not provide
detailed sequence information on either the position or the exact
number of substitutions. Small scale intra-specific ITS variation for such
a widely distributed gyrodactylid is not necessarily unusual [26].
However we have also been able to confirm intra-individual polymorphism in ITS-1 of G. teuchis. Rokicka et al. [10] reported that a large
proportion of the sequence traces of ITS-1 amplicons were unreadable in
their so called G. cf. teuchis samples. At first glance we also observed
unreadable ITS-1 sequences in the present study. However, careful
examination of the chromatograms revealed a one base indel connected
to a 6 or 7A homopolymer as the reason for the sequencing difficulties,
which may be genomic or may be a PCR artefact. No homogeneous PCR
product with a 7A homopolymer was detected.
Somewhat more interesting is the finding of two ITS-1 variants that
differ by an A/G transition at position 638. These were consistent among
individuals, and confirmed using a restriction site present only in one
sequence variant (see Fig. 2). In roughly 65% of the studied parasites the
“A” and the “G” variants occur with frequencies of 40.9% and 59.1%,
respectively. This may be taken as an indication for the presence of at
least two distinct rDNA loci in the genome of G. teuchis. If so, it is
however surprising that the two ITS variants were not discovered in all
individuals. One may also assume that the “A” and the “G” variants relate
to two different alleles with frequencies of 67.7% for the “A” and 32.3%
for the “G” allele. If true, the 20G. teuchis individuals with the “A” and the
“G” variants were then heterozygotes, we would also expect about 6
individuals to be homozygous for the “G” variant. As these were never
observed, one may still suggest two alleles, but one would be an “A”
allele and the other one an “A/G” allele with an only partly homogenized
rDNA cluster. However, we interpret the ITS-1 polymorphism as an
indication of a not fully homogenized rDNA locus with variable numbers
of “A” and “G” repeats. According to the concept of concerted evolution
[29] newly introduced nucleotide substitutions will either be lost or
show various intermediate levels of homogenization in tandemly
repeated sequences. Following the classification of Strachan et al. [30]
for the transitional stages of mutations at individual nucleotide
positions in repetitive DNA the observed A/G polymorphism falls in
class 3, i.e. both variants are only partly homogenized. However, in the
absence of further information, this interpretation can only be seen as a
hypothesis. Although rarely reported, intra-individual variation within
ITS sequences does occur; for example Hoste et al. [31] reported an
intra-individual polymorphism from a trichostrongylid nematode and
Králová-Hromadová et al. [32] detected large scale intra-individual ITS
variability in a triploid monozoic cestode.
In spite of the striking overlap in host preference, and reported
morphological similarity, G. salaris and G. teuchis are only rather
distantly related when the internal transcribed spacer regions (ITS) of
the ribosomal gene cluster are used to infer phylogenetic relationships
[33]. While G. teuchis was considered to represent a cryptic Gyrodactylus
species, difficult to distinguish morphologically from G. salaris, it was in
no sense seen as a sibling species of the latter. However, the present
study makes it clear that G. teuchis is quite different from G. salaris also in
terms of morphology, and cannot be considered a truly cryptic taxon. G.
teuchis can be readily distinguished from G. thymalli (an exception may
be the Trysil population analyzed by Olstad et al. [17]) because it is
smaller than G. thymalli in general. It is much more similar in size to G.
salaris, and although significant differences in a range of measurements
(e.g. hamulus aperture distance HAD, see Table 3) were observed, none
were reliably diagnostic. Nevertheless, appropriate morphometric
analyses such as PCA or CVA differentiated G. teuchis from G. salaris as
effectively as from G. thymalli (see Fig. 6). Due to morphological
similarities to the economically important G. salaris, molecular methods
have so far been the markers of choice for distinguishing G. teuchis.
Relatively speaking, the lack of economic importance of G. teuchis might
be the main reason why the morphological and genetic variability of
populations of this widely distributed species have never been explored
in detail, although frequently confusion with G. salaris through
inadequate diagnostics places a considerable hidden cost on the
salmonid aquaculture industry. The ease with which PCA and CVA
could distinguish G. teuchis from G. salaris lends hope that automated
identification systems [7,8] could learn to discriminate G. teuchis in
routine diagnostic procedures. This would remove a major source of
confusion which led to the poor resolution of G. salaris by taxonomic
experts in the identification trial of Shinn et al. [6].
5. Conclusions
We present a comprehensive re-description of G. teuchis, and, for the
first time, demonstrate that this so-called cryptic species can be reliably
discriminated using morphology from its most similar congenors. The
ability of multivariate methods to discriminate this species as
demonstrated suggests that the occurrence of false positives in G. salaris
diagnosis may be reduced using morphometric methods alone.
Supplementary materials related to this article can be found online
at doi:10.1016/j.parint.2011.08.016.
Acknowledgements
Data on G. salaris from the Norwegian river Drammenselva, and
G. thymalli from the type locality River Hnilec in Slovakia, were kindly
provided by Kjetil Olstad. We are grateful to Georg Holzer for giving
the opportunity to sample Kleiner Kamp trout. Thanks to Eileen Harris
for providing reference material and Øyvind Hammer for helpful
advice with multivariate statistics.
References
[1] Harris PD, Shinn AP, Cable J, Bakke TA. Nominal species of the genus Gyrodactylus
v. Nordmann 1832 (Monogenea: Gyrodactylidae), with a list of principal host
species. Syst Parasitol 2004;59:1–27.
[2] Lautraite A, Blanc G, Thiery R, Daniel P, Vigneulle M. Gyrodactylids parasitizing
salmonids in Brittany and Western Pyrénées water basins: epidemiological
features of infection and species composition. Bull Franç Pêche Pisc 1999;355:
305–25.
[3] Cunningham CO, Mo TA, Collins CM, Buchmann K, Thiery R, Blanc G, et al.
Redescription of Gyrodactylus teuchis Lautraite, Blanc, Thiery, Daniel & Vigneulle,
1999 (Monogenea: Gyrodactylidae); a species identified by ribosomal RNA
sequence. Syst Parasitol 2001;48:141–50.
[4] Johnston C, Mackenzie K, Cunningham CO, Eiras JC, Bruno DW. Occurrence of
Gyrodactylus salaris Malmberg, 1957, in Portugal. Bull Eur Ass Fish Pathol 1996;16:
89–91.
[5] Cunningham CO. Chapter 9. Gyrodactylus salaris Malmberg, 1957 (PLATYHELMINTHES: MONOGENEA). In: Cunningham CO, editor. Reviews: Methods and
Technologies in Fish Biology and Fisheries. Molecular Diagnosis of Salmonid
Diseases. Kluwer Academic Publishers; 2002. p. 235–65.
[6] Shinn AP, Collins C, Garcia-Vasquez A, Snow M, Matejusová I, Paladini G, et al.
Multi-centre testing and validation of current protocols for the identification of
Gyrodactylus salaris (Monogenea). Int J Parasitol 2010;40:1455–67.
[7] Kay JW, Shinn AP, Sommerville C. Towards an automated system for the
identification of notifiable pathogens: using Gyrodactylus salaris as an example.
Parasitol Today 1999;15:201–6.
[8] Shinn AP, Kay JW, Sommerville C. The use of statistical classifiers for the
discrimination of species of Gyrodactylus (Monogenea) parasitizing salmonids.
Parasitology 2000;120:261–9.
[9] Matejusová I, Gelnar M, McBeath AJA, Collins CM, Cunningham CO. Molecular
markers for gyrodactylids (Gyrodactylidae: Monogenea) from five fish families
(Teleostei). Int J Parasitol 2001;31:738–45.
C. Hahn et al. / Parasitology International 60 (2011) 480–487
[10] Rokicka M, Lumme J, Ziętara MS. Identification of Gyrodactylus ectoparasites in
Polish salmonid farms by PCR-RFLP of the nuclear ITS segment of ribosomal DNA
(Monogenea, Gyrodactylidae). Acta Parasitol 2007;52:185–95.
[11] von Gersdorff Jørgensen L, Heinecke RD, Kania P, Buchmann K. Occurrence of
gyrodactylids on wild Atlantic salmon, Salmo salar L., in Danish rivers. J Fish Dis
2008;31:127–34.
[12] Dzika E, Maciejewska IW, Hoffman RW, Oidtmann B. The Gyrodactylidae fauna of
rainbow trout Oncorhynchus mykiss Walbaum 1792 in the Rogg breeding pound in
Bavaria, Germany. Parasitol Res 2009;104:671–6.
[13] Paladini G, Gustinelli A, Fioravanti ML, Hansen H, Shinn AP. The first report of
Gyrodactylus salaris Malmberg, 1957 (Platyhelminthes, Monogenea) on Italian
cultured stocks of rainbow trout (Oncorhynchus mykiss Walbaum). Vet Parasitol
2009;165:290–7.
[14] Buchmann K, Lindenstrøm T, Bresciani J. En for videnskaben helt ny snylter ved
navn Gyrodactylus teuchis på de bornholmske ørreder i Gyldensåen. Bornholms
Natur, Tidsskriftet Fjældstauijn 1999;23:39–43 [In Danish].
[15] Buchmann K, Lindenstrøm T, Nielsen ME, Bresciani J. Diagnostik og forekomst af
ektoparasitinfektioner (Gyrodactylus spp.) hos danske laksefisk. Dansk Veterinærtidsskrift 2000;83:15–9.
[16] Bush AO, Lafferty KD, Lotz JM, Shostak AW. Parasitology meets ecology on its own
terms: Margolis et al. revisited. J Parasitol 1997;83:575–83.
[17] Olstad K, Shinn AP, Bachmann L, Bakke TA. Host-based identification is not
supported by morphometrics in natural populations of Gyrodactylus salaris and G.
thymalli (Platyhelminthes, Monogenea). Parasitology 2007;134:2041–52.
[18] Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ. Basic local alignment search
tool. J Mol Biol 1990;215:403–10.
[19] Harris PD, Cable J, Tinsley RC, Lazarus CM. Combined ribosomal DNA and
morphological analysis of individual gyrodactylid monogeneans. J Parasitol
1999;85:188–91.
[20] Mo TA. Seasonal variations of opisthaptoral hard parts of Gyrodactylus salaris
Malmberg, 1957 (Monogenea: Gyrodactylidae) on parr of Atlantic salmon Salmo
salar L. in the river Batnfjordselva, Norway. Syst Parasitol 1991;19:231–40.
[21] Shinn AP, Hansen H, Olstad K, Bachmann L, Bakke TA. The use of morphometric
characters to discriminate specimens of laboratory-reared and wild populations of
Gyrodactylus salaris and G. thymalli (Monogenea). Folia Parasitol 2004;51:239–52.
487
[22] Hammer Ø, Harper DAT, Ryan PD. PAST: Paleontological Statistics Software
Package for Education and Data Analysis. Palaeontol Electronica 2001;4:9.
[23] van Husen D. Die Ostalpen in den Eiszeiten. Aus der geologischen Geschichte
Österreichs. Wien: Geologische Bundesanstalt; 1987. p. 24.
[24] Weiss S, Schlötterer C, Waidbacher H, Jungwirth M. Haplotype (mtDNA) diversity
of brown trout Salmo trutta in tributaries of the Austrian Danube: massive
introgression of Atlantic basin fish—by man or nature? Mol Ecol 2001;10:1241–6.
[25] Bakke TA, Harris PD, Jansen PA, Hansen LP. Host specificity and dispersal strategy
in gyrodactylid monogeneans, with particular reference to Gyrodactylus salaris
Malmberg (Platyhelminthes, Monogenea). Dis Aquat Organ 1992;13:63–74.
[26] Bakke TA, Cable J, Harris PD. The biology of gyrodactylid monogeneans: the
“Russian-doll-killers”. Adv Parasitol 2007;64:161–376.
[27] Olstad K, Bachmann L, Bakke TA. Phenotypic plasticity of taxonomic and
diagnostic structures in gyrodactylosis-causing flatworms (Monogenea, Platyhelminthes). Parasitology 2009;136:1305–15.
[28] Harris PD. Ecological and genetic evidence for clonal reproduction in Gyrodactylus
gasterostei Glaser, 1974. Int J Parasitol 1998;28:1595–607.
[29] Dover GA, Brown S, Coen E, Dallas J, Strachan T, Trick M. The dynamics of genome
evolution and species differentiation. In: Dover GA, Flavell RB, editors. Genome
Evolution. Academic press; 1982. p. 343–72.
[30] Strachan T, Webb D, Dover GA. Transition stages of molecular drive in multiplecopy DNA families in Drosophila. EMBO J 1985;4:1701–8.
[31] Hoste H, Chilton NB, Gasser RB, Beveridge I. Differences in the second internal
transcribed spacer (Ribosomal DNA) between five species of Trichostrongylus
(Nematoda: Trichostrongylidae). Int J Parasitol 1995;25:75–80.
[32] Králová-Hromadová I, Štefka J, Špakulová M, Orosová M, Bombarová M, Hanzelová V,
et al. Intra-individual internal transcribed spacer 1 (ITS1) and ITS2 ribosomal
sequence variation linked with multiple rDNA loci: a case of triploid Atractolytocestus
huronensis, the monozoic cestode of common carp. Int J Parasitol 2010;40:175–81.
[33] Ziętara MS, Lumme J. Speciation by host switch and adaptive radiation in a fish
parasite genus Gyrodactylus (Monogenea, Gyrodactylidae). Evolution 2002;56:
2445–58.