Functional Ecology 2009, 23, 187–195
doi: 10.1111/j.1365-2435.2008.01482.x
Measured immunocompetence relates to the proportion
of dead parasites in a wild roach population
Blackwell Publishing Ltd
Anssi Vainikka1,2*, Jouni Taskinen2, Katja Löytynoja2, E. Ilmari Jokinen2 and Raine Kortet2,3
1
Institute of Coastal Research, Swedish Board of Fisheries, Box 109, SE-74222, Öregrund, Sweden; 2Department of
Biological and Environmental Science, University of Jyväskylä, P.O. Box 35, FIN-40014 University of Jyväskylä, Finland;
and 3University of Oulu, Department of Biology, P.O. Box 3000, FI-90014 Oulu, Finland
Summary
1. Although various methods are used to measure immunocompetence, their relationship with the
actual parasite clearance or parasite load is seldom demonstrated in natural systems.
2. We combined nine measures of immune function using principal component analysis (PCA),
and examined the relationship of the collective measures with (i) the proportion of parasites killed
by the host, (ii) the burden of several parasite species and (iii) a viral disease in a wild population of
the roach, Rutilus rutilus. We also studied if these variables were associated with the concentration
of steroids (testosterone and oestradiol).
3. Most significant correlations between the loads of ecto- and gill parasites and the measures of
immunity were positive, suggesting either a parasite-induced immunoactivation or temporal
covariation. When the temporal covariation was statistically removed, the functional aspects
(respiratory burst, chemotaxis) of phagocytotic cells and the total concentration of immunoglobulin
M correlated positively with the proportion of dead Rhipidocotyle campanula (Digenea) parasites.
However, no relationships remained between the parasite loads and the measures of immunity or
concentrations of the hormones.
4. The present results suggest that the measures of phagocytotic activity and IgM concentrations
relate to an observable competence of roach to eliminate parasites. However, new experimental
approaches are needed to reveal the causal mechanisms behind the sometimes ambiguous correlations.
Key-words: immune defence, innate immunity, parasite resistance, trade-off
Functional Ecology (2008) xx, 000–000
Introduction
Parasites and pathogens impose strong natural selection
pressures on their hosts (Clarke 1979; Price 1980; Goater &
Holmes 1997), and therefore, efficient immune function may
be favoured both by natural and sexual selection (Hamilton &
Zuk 1982; Andersson 1994). Handicapping sexual signalling
(Zahavi 1975, 1977) and unavoidable energetic trade-offs
between immune function and reproduction link parasite
load to the fitness of a host, mainly via endocrinological
mechanisms (Folstad & Karter 1992; Wedekind & Folstad
1994; Kurz et al. 2007). In immunological ecology, various
methods are used to estimate individuals’ immunocompetence
(e.g. Skarstein, Folstad & Liljedahl 2001; Kortet et al. 2003a;
Måsvær, Liljedahl & Folstad 2004). However, in addition to
the rather well-established connection between specific
antibodies and disease resistance in vertebrates (Casadevall
2004), surprisingly few data show how the commonly used
non-specific measures of immune function actually relate to
abundance of parasites and diseases, or to individuals’ ability
*Correspondence author. E-mail: [email protected]
to control the cost of parasitism (e.g. Pulsford et al. 1995;
Sakai et al. 1996; Hakoyama et al. 2001).
It is clear that more detailed knowledge on the relationships between the measured indices of immune function and
parasite load and parasite resistance in natural systems is
needed to explain the causal mechanisms of these interactions, and to reveal whether high measures of immunity are
protective against parasitic infections. Studies of this kind
could also help us to estimate the ecological consequences of
stress and pollutants that affect several immune functions
(Pulsford et al. 1995). Moreover, specific information on how
different aspects of immune function are related to parasite load in the wild is a necessary background for future
interpretation of optimal immune responses (Viney, Riley &
Buchanan 2005).
Within an individual’s life-history, the development, maintenance and activation of immune defence can be considered
as a significant sink of energy that is traded against growth
and reproduction (e.g. Tschirren & Richner 2006; Simková
et al. 2008). Since many life-history traits are physiologically
regulated by hormonal mechanisms, these trade-offs are
often detected as correlations between concentrations of
© 2008 The Authors. Journal compilation © 2008 British Ecological Society
188
A. Vainikka et al.
reproductive hormones and measures of immune function or
parasite loads (Hillgarth & Wingfield 1997; Braude, TangMartinez & Taylor 1999; Kortet & Vainikka 2008). However,
the sign of correlations between sex hormone concentrations,
measures of non-specific immunity and parasite loads is often
difficult to predict, since many relationships are bi-directional
and may be shaped by the efficiency of the adaptive immunity
determined by the set of available MHC-alleles (SchmidHempel & Ebert 2003; Kurtz et al. 2004; Ottová, Simková &
Morand 2007). Therefore, additional descriptive information
about patterns in parasite–immune defence interactions is
needed in wild animals.
Males of many cyprinids signal both parasite resistance
(Taskinen & Kortet 2002; Kortet et al. 2004) and parasite
load (Wedekind 1992; Kortet & Taskinen 2004; Ottová et al.
2005) to females by the expression of androgen-dependent
secondary sexual characters; breeding tubercles (Kortet et al.
2003b). Previously, we found using the same data as in the
current study, that individual measures of immunity of roach,
Rutilus rutilus, showed seasonal changes that to some extent
related to reproduction (Kortet et al. 2003a). However, the
concentrations of testosterone and oestradiol were not
negatively, but often positively, correlated with the individual
measures of immune function (Vainikka et al. 2004a).
Here, we present novel data on parasite abundances on
roach sampled for the study of Kortet et al. (2003a) and
further analysed for the correlations between the measures of
immunity and sex hormone concentrations in Vainikka et al.
(2004a). We reanalyse the old data by combining the nine
measures of immune function (relative counts of (1) lymphocytes,
(2) thrombocytes, (3) granulocytes and (4) macrophages,
(5) total IgM concentration, (6) relative mass of the spleen,
(7) total count of leukocytes in blood, (8) chemiluminescence
and (9) migration differential of head kidney phagocytes)
into principal components and comparing their expression
between sexes during the year. The main aim of this study is to
examine the relationships between the combined measures of
immunity and (1) load of several parasite species within and
among seasons, (2) proportion of dead Raphidascaris acus
and Rhipidocotyle campanula parasites, (3) occurrence of a
viral disease, and finally (4) testosterone and oestradiol
concentrations. We aim to describe the relationships between
the measures of immunity and the occurrence of parasites,
and dead parasites, thereby potentially revealing which
measures of immunity relate to the host’s ability to eliminate
parasites.
Materials and methods
Roach ( N = 149, length 158·8 ± 21·4 mm, mass 55·8 ± 27·2 g,
mean ± SD) were captured by angling in five occasions (approximately 15 males and 15 females per time) during the year 1999 from
Lake Jyväsjärvi, Jyväskylä, Finland. Fish were reared in accordance
with the Guidelines for the Use of Animals in Research (Anon.
1996) under license from the Central Finland Regional Environmental
Center (permission LS-2/99). Methods for fish sampling and physiological
investigations are given in more detail in Kortet et al. (2003a) and
Vainikka et al. (2004a).
ANALYSES OF IMMUNITY AND HORMONE LEVELS
An overview of the methods is as follows; the fish were anesthetized
immediately after capture using 0·01% MS-222 (Sigma Chemical
Co., St Louis, MO), transported to the laboratory in aerated lake
water and kept at the ambient temperature of the lake before
examination not later than 2 h after capture. Whole blood (50 µL)
was stained with Shaw dye for enumeration of red and white blood
cells (Shaw 1930), and plasma was separated by centrifugation and
frozen for the later determination of immunoglobulin M (IgM) and
steroid concentrations. IgM was analysed using an enzyme-linked
immunoassorbent assay (ELISA) for roach IgM (Aaltonen, Jokinen
& Valtonen 1994; Vainikka et al. 2004a). Spleen somatic index was
calculated as the mass of the spleen the total body mass. Sex was
identified from gonads. The chemiluminescence and migration of
head kidney phagocytes (Scott & Klesius 1981) were assayed as
described in Kortet et al. (2003a). The difference between spontaneous
and directed migration (chemotactic differential) was used in the
analyses. Plasma hormone concentrations were determined using
RIA-based kits (TESTO-CTK, DiaSorin, Italy, for testosterone and
ESTR-CTK-4, DiaSorin, Italy, for 17β-oestradiol).
PARASITES AND DISEASES
Mucous was collected from gills and skin of the fish using a scalpel,
pressed on a slide under cover glass, and examined directly for
Chiliophora protozoan parasites with a microscope using transmitted
light and 400× magnification. Parasite infestations from the gills, fins
and liver were examined microscopically for Sporozoa protozoans
and metazoan parasites by pressing the tissues between two large
glass plates and using transmitted light with 100× magnification. We
found Apiosoma, Trichodina, Scyphidia and Ichthyopthirius multifiliis
chiliophores from the mucous of the gills and skin. Their numbers
were estimated as a mean of three visual fields. I. multifiliis, known
as the white spot disease, occurred so rarely that it was not included
in further analyses. Instead, the sum of Apiosoma-, Trichodina- and
Scyphidia-type chiliophores, named as ‘ectoprotozoa on gills and
epidermis’ (Table 1) was used in the statistical analyses. Pooling of
these parasites is justified as they co-occur and have a similar way of
living; they usually forage on bacterial flora and organic particles
from the water but may start utilizing epidermal cells of fish and
cause damage to skin when occurring in high numbers. These protozoans are moderately common parasites of roach in the present
study area (Halmetoja, Valtonen & Taskinen 1992), and their
numbers are known to respond to changes in host susceptibility, for
example due to pollution (Lehtinen 1989; Khan et al. 1994). We
encountered three sporozoan-type protozoan taxa; the myxozoans
Myxobolus sp. from the gills, Myxobolus mülleri from the liver and
an unidentified sporozoan from the fins. M. mülleri in the liver was
very rare and was therefore excluded from further analyses. These
sporozoans, named as ‘endoprotozoa’ (Table 1), are abundant
among roach in the present study area (Brummer Korvenkontio,
Tellervo Valtonen & Pugachev 1991) and can be harmful to the fish
host (Shul’man 1990).
Among metazoan parasites, we found seven taxa belonging to the
phylum Platyhelmithes within the gills, one of them being Dactylogyrus
spp. (Monogenea);the sum of different Dactylogyrus species
occurring on the gills of roach in the present study area (Table 1)
(Koskivaara & Valtonen 1992). High abundance of Dactylogyrus
spp. has also been suggested to be associated with pollution-induced
impacts on the immune defence of roach (Valtonen, Holmes &
Koskivaara 1997). Another monogenean in the gills, Paradiplozoon
© 2008 The Authors. Journal compilation © 2008 British Ecological Society, Functional Ecology, 23, 187–195
Parasites and immune function in roach 189
Table 1. Prevalence, mean abundance and mean intensity for the three parasite species used in parametric analyses (three uppermost rows) and
several mostly ectoparasitic, and less abundant species not included in parametric tests (lowermost rows). CI stands for the confidence interval
Prevalence
Rhipidocotyle campanula
Rhipidocotyle fennica
Raphidascaris acus
Diplostomum spathaceum
Ectoprotozoa on gills and epidermis
Monogenea on gills
Ergasilus briani on gills
Myxobolus on gills
Tylodelphus clavata
Endosporozoa
Neoergasilus japonicus
Glochidia total
Abundance
Intensity
Mean
95% CI
Mean
95% CI
Mean
95% CI
94·6
88·4
84·6
84·6
29·1
48·3
41·6
30·2
23·5
7·4
53·7
18·8
89·7–97·7
82·1–93·1
77·7–90·0
77·7–90·0
21·9–37·1
40·1–56·6
33·6–50·0
23·0–38·2
16·9–31·1
3·7–12·8
45·4–61·9
12·9–26·0
9·4
28·5
14·4
5·2
1·2
4·7
2·3
1·2
1·4
0·3
3·4
0·5
8·1–10·8
21·8–35·2
10·2–18·5
2·7–7·8
0·4–2·0
3·3–6·2
1·5–3·2
0·7–1·7
0·4–2·4
0·1–0·5
2·3–4·5
0·3–0·7
10·0
32·3
17·0
6·2
4·1
9·8
5·6
4·0
6·1
3·9
6·3
2·6
8·6–11·3
24·9–39·6
12·2–21·8
3·2–9·1
1·7–6·6
7·3–12·3
3·8–7·4
2·7–5·4
2·1–10·1
1·5–6·3
4·5–8·1
1·7–3·6
homoion, was present too infrequently to be included in the analyses.
Bucephalid digenean Rhipidocotyle campanula was encountered on
the gills and R. fennica on the fins. Numbers of R. fennica were
counted on the tail of the fish, as it appears to be the main site of
infection for this species, whereas R. campanula infects almost
exclusively the gills (Taskinen, Valtonen & Gibson 1991; Gibson,
Valtonen & Taskinen 1992). These Rhipidocotyle species are the
most common and abundant parasites of roach in the study area
(Valtonen et al. 1997; Taskinen & Kortet 2002), and roach develops
acquired immunity against Rhipidocotyle parasites (Aaltonen,
Valtonen & Jokinen 1997). Bucephalid digeneans can be harmful to
a fish host (Hoffmann et al. 1990); R. campanula can interrupt the
blood circulation in the gills. Dead and live specimens of R. campanula
in the gills of roach were identified as described by Taskinen & Kortet
(2002). Among the observed digenean eye flukes, Diplostomum
spathaceum and Tylodelphus clavata, the former is known to impair
vision of fish by causing cataracts (e.g., Karvonen, Seppälä &
Valtonen 2004). For the nematode Raphidascaris acus, which damages
liver tissue (Valtonen, Haaparanta & Hoffmann 1994), we observed
both living (moving) and dead (encapsulated, motionless, degenerated)
individuals. Among ectoparasites, the crustacean parasites Ergasilus
sieboldi on gills and Neoergasilus japonicus on skin occurred frequently,
together with the unionid bivalve glochidia. These findings are in
accordance with a previous study on roach in the same area (Tuuha,
Valtonen & Taskinen 1992).
Three of the most prevalent species were used in the parametric
analyses between the measures of immune function, hormone
concentrations and parasite abundance (Table 1). Furthermore, nine
parasite taxa (or groups) (prevalence 7·4 – 84·6%), and the proportion
of dead R. acus individuals were used in non-parametric correlations
between the measures of immune function, hormone concentrations
and parasite abundance (Table 1). The proportion of dead R. campanula
parasites of all R. campanula was not defined if the total number of
parasites was less than three (thus N = 118). A similar criterion was
used for dead R. acus; thus N = 82. The proportion of dead specimens
was used as a proxy of the host’s ability to eliminate these parasites.
However, the ratio is susceptible to uneven exposure to these parasites,
and the delay after which dead specimens disappear is not known.
These facts set further constrains for the interpretation.
Occurrence of the viral disease, epidermal papillomatosis, was
examined visually and by hand (Korkea-Aho, Vainikka & Taskinen
2006a). Recent studies indicate that epidermal papillomatosis is
associated with pollution and stress in roach (Korkea-Aho et al.
2006b, 2008).
STATISTICAL ANALYSES
Prior to analyses spleen-somatic index, relative proportions of
differential leukocyte counts and the proportion of dead R. campanula
and R. acus parasites were arcsine square root transformed.
However, the proportion of dead R. acus was not successfully
normalised and was therefore used only for non-parametric analyses.
All other parameters and parasite counts were Ln(X + 1) transformed, but only the total load of R. campanula, R. fennica, and
R. acus met the assumption of normality after transformation, and
so only these were used in the parametric analyses. Principal component analysis was run on all nine immunological parameters, and
principal components (for this on immunological components, IC:
s) having eigenvalues over one (IC1, IC2, IC3 and IC4) were considered for further analysis. Correlations between hormones,
measures of immune function and successfully normalised parasite
intensities were first examined within each sex and period separately.
The individual correlation coefficients were then meta-analysed as
described in Hedges & Olkin (1985) to achieve 95% confidence intervals for through-year values. Statistical analyses were performed
using AV Bio-statistics 4·8 (available at http://web.telia.com/
~u25601709/avbs/) and SPSS 15·0·0 (SPSS Inc, USA).
Results
PRINCIPAL COMPONENT ANALYSIS
The four principal components (PC) having eigenvalues over
1 explained 70·8% of the total variance in immune parameters
and divided the individual immune parameters into four
functional categories (Table 2). The first PC (IC1) was
positively related to the total number of leukocytes and relative count of lymphocytes, and correspondingly negatively
related to the relative count of thrombocytes (Table 2). The
second PC (IC2) related most clearly to phagocytosis and
© 2008 The Authors. Journal compilation © 2008 British Ecological Society, Functional Ecology, 23, 187–195
190
A. Vainikka et al.
Table 2. Component score matrix of the three principal components
of immune function. Varimax rotation with Kaiser normalization was
used, and only eigenvalues over 1 are considered. The most important
parameters contributing to the principal components are in bold
Score coefficients
Variable
IC1
IC2
IC3
IC4
Lymphocytes
Thrombocytes
Granulocytes
Macrophages
Spleen-somatic index
Total leukocyte count
Chemiluminescence
IgM concentration
Chemotaxis
Cumulative R2
0·92
−0·95
−0·06
0·04
0·18
0·41
−0·14
0·25
−0·04
24·45
−0·04
0·02
0·06
−0·26
−0·07
−0·08
0·81
0·45
0·84
44·71
−0·28
−0·19
0·84
0·72
−0·11
−0·47
−0·10
−0·03
−0·02
58·96
0·06
−0·09
0·03
−0·15
0·88
−0·42
−0·09
−0·31
0·10
70·78
migration activity of head kidney phagocytes, but also to the
IgM concentration. The third PC (IC3) was indicative of the
presence of phagocytotic cells in the circulation, i.e. the
relative counts of granulocytes and macrophages. The fourth
PC was dominated by the spleen somatic index (Table 2).
SEASONAL CHANGES AND SEX-DIFFERENCES
Since the seasonal changes in immune function have been
reported in Kortet et al. (2003a), we do not report details about
the individual immune parameters here. All principal components of immunity (IC1-IC4) varied seasonally (ancova,
F4,112 ≥ 8·1, P < 0·001, Fig. 1a–d). All but IC2 also differed
between sexes (ancova, F1,112 ≥ 6·0, P < 0·02). Furthermore,
the interaction between sex and season was significant for IC1
and IC3 (ancova, F4,112 ≥ 3·9, P < 0·01).
Fig. 1. (a–d) Seasonal variation in the four principal components of immune function (See Table 2 for description for PC: s). (e–h) Seasonal
variation in the proportion of dead Rhipidocotyle campanula parasites and load of R. campanula, R. fennica, and Raphidascaris acus. Filled
circles represent males and open boxes represent females. Error bars represent standard error of mean.
© 2008 The Authors. Journal compilation © 2008 British Ecological Society, Functional Ecology, 23, 187–195
Parasites and immune function in roach 191
Table 3. Correlations between testosterone, oestradiol, the four principal components of immune function (Table 1), proportion of killed
Rhipidocotyle campanula and load of R. campanula, R. fennica and Raphidascaris acus. In the upper right diagonal there are the Pearson’s
correlation over all periods and both sexes, and in the lower left diagonal there are the meta-analytically combined 95% confidence intervals for
correlations calculated within each period and sex. N.S. is for not significant at 0·05 level
Testosterone
Estradiol
IC1
IC2
IC3
IC4
Dead R. campanula
R. campanula load
R. fennica load
R. acus load
Testosterone
Oestradiol
IC1
IC2
IC3
IC4
Dead
R. camp.
R. camp.
load
R. fennica
load
R. acus
load
1
na.
n.s.
0·00–0·40
n.s.
n.s.
n.s.
n.s.
n.s.
n.s.
0·391***
1
n.s.
n.s.
n.s.
n.s.
n.s.
n.s.
n.s.
n.s.
n.s.
−0·186*
1
na.
na.
na.
−0·54–0·11
n.s.
n.s.
n.s.
0·393***
0·201*
n.s.
1
na.
na.
0·01–0·45
n.s.
n.s.
n.s.
n.s.
−0·211*
n.s.
n.s.
1
na.
n.s.
n.s.
n.s.
n.s.
n.s.
−0·202*
n.s.
n.s.
n.s.
1
n.s.
n.s.
n.s.
n.s.
n.s. n.s.
n.s.
n.s.
n.s.
0·342***
n.s.
1
na.
na.
na.
n.s.
0·273**
n.s.
n.s.
n.s.
n.s.
n.s.
1
na.
na.
n.s.
0·268**
n.s.
n.s.
n.s.
n.s.
−0·275**
0·517***
1
na.
n.s.
n.s.
n.s.
−0·183*
n.s.
n.s.
0·182*
n.s.
1
*P < 0·05, **0·05 < P < 0·01, ***P < 0·001.
The proportion of dead R. campanula varied between
seasons (ancova, F4,108 = 4·5, P = 0·002) and by the interaction
term season × sex (F4,108 = 2·7, P = 0·032, Fig. 1e). The total
load of R. campanula did not vary seasonally (F4,140 = 1·3,
P = 0·280), but was higher in females than in males (F1,140 = 5·2,
P = 0·02, Fig. 1f ). The interaction term was not significant in
this case. Also the other Rhipidocotyle species, R. fennica was
more abundant in females than in males (F4,139 = 6·1,
P = 0·015), and did not vary between the seasons even in the
interaction with sex (Fig. 1g). Raphidascaris acus was in
general more abundant in females than in males (F1,140 = 7·3,
P = 0·008, Fig. 1h), but also varied between seasons
(F4,140 = 2·7, P = 0·034), and in the interaction between sex
and season (F4,140 = 2·9, P = 0·025, Fig. 1h). Loads of other
parasite species were not studied for seasonal changes or sex
differences since their abundances were rather low (Table 1),
and these aspects were not the focus of the study.
IMMUNE FUNCTION AND PARASITE LOADS
IC3 correlated positively with the proportion of dead Rhipidocotyle campanula parasites and negatively with the load of
Raphidascaris acus in the pooled samples (Table 3). Measures
of immune function were mainly positively correlated with
the load of several parasites when analysed using Spearman’s
rank correlation in pooled samples (Table 4).
When the samples and sexes were analysed separately, and
the individual correlation coefficients were combined
meta-analytically, IC1 was found to correlate negatively with
the proportion of dead R. campanula, whereas the IC2 correlated positively with it (Table 3). None of the parasite loads were
significantly related to the measures of immunity (Table 3).
STEROIDS AND IMMUNE FUNCTION
Correlations of testosterone and oestradiol with individual
immune measures are reported in Vainikka et al. (2004a).
Plasma testosterone concentration correlated positively with
the IC2 both when the samples and sexes were pooled, and in
meta-analysed correlation based on seasonal correlations
(Table 3). Concentration of oestradiol correlated negatively
with all immune components, except for the positive correlation
with IC2 in the pooled samples (Table 3). However, concentration of oestradiol was not correlated with any of the
measures of immunity when the season effect was removed by
meta-analysing the individual correlations (Table 3).
STEROIDS AND PARASITE LOADS
Plasma oestradiol concentration correlated positively with
the total load of R. campanula and R. fennici parasites in the
pooled samples (Table 3). In non-parametric analyses,
testosterone correlated negatively with the load of monogenean
parasites on gills when the samples and sexes were pooled
(Table 4). Oestradiol was negatively correlated with the loads
of ectoprotozoa and Monogenea (Table 4). However, the
concentrations of sex hormones were not related to the load
of parasites in the meta-analytical results (Table 3).
EPIDERMAL PAPILLOMATOSIS, IMMUNE FUNCTION
AND STEROIDS
Logistic regression with a backward model selection procedure starting from the model including sex (as factor),
period (as factor), fish length, testosterone, oestradiol, IC1, IC2,
IC3, IC4, and interactions testosterone × sex, oestradiol × sex
was used to examine factors contributing to the probability
of being diseased by the epidermal papillomatosis. The final
model included the parameters; sampling period, length,
oestradiol, testosterone and oestradiol × sex interaction but
none of the immune measures. The final model explained
65·3% of the total variance and predicted 89·5% of the fish to
the correct health/diseased class. Testosterone had a strong
positive effect (B = 0·94) and oestradiol a negative effect
(B = −0·66, in interaction with sex, B = −0·70) on the occurrence of papillomatosis.
© 2008 The Authors. Journal compilation © 2008 British Ecological Society, Functional Ecology, 23, 187–195
192
A. Vainikka et al.
Table 4. Spearman’s rank correlations of testosterone, oestradiol and the four principal components of immune function (Table 2) with the load
of several parasites and the proportion of dead Raphidascaris acus parasites of all R. acus individuals over both sexes and all periods. The
proportion was defined only for fish that had at least three parasite individuals
Species & location
Total ectoprotozoa
Gills and epidermis
Monogenea total
Gills
Ergasilus sieboldi
Gills
Myxobolus sp.
Gills
Glochidia total
Gills
Endoprotozoa
Fins
Neoergasilus japonicus
Fins
Diplostomum sp.
Eye
Tylodelphys clavata
Eye
Prop. of dead R. acus
Liver
R
P
N
R
P
N
R
P
N
R
P
N
R
P
N
R
P
N
R
P
N
R
P
N
R
P
N
R
P
N
Testosterone
Oestradiol
IC1
IC2
IC3
IC4
−0·14
0·096
144
−0·25
0·003
144
−0·11
0·176
144
0·16
0·054
144
0·12
0·161
144
0·13
0·128
144
0·13
0·133
144
−0·13
0·128
144
−0·01
0·862
144
0·07
0·564
77
−0·19
0·027
141
−0·24
0·005
141
−0·02
0·850
141
0·10
0·242
141
0·02
0·826
141
0·06
0·467
141
0·09
0·263
141
−0·16
0·063
141
0·11
0·207
141
0·04
0·734
78
−0·17
0·054
123
−0·02
0·833
123
0·02
0·838
123
−0·13
0·163
123
−0·10
0·278
123
−0·20
0·029
123
0·05
0·620
123
−0·03
0·784
123
0·20
0·030
123
0·02
0·855
63
0·02
0·864
123
0·01
0·886
123
−0·15
0·092
123
0·14
0·132
123
0·18
0·046
123
0·07
0·420
123
0·18
0·047
123
0·02
0·820
123
−0·01
0·884
123
−0·06
0·661
63
0·27
0·002
123
0·38
< 0·001
123
0·12
0·171
123
0·21
0·022
123
0·20
0·025
123
0·00
0·970
123
−0·02
0·787
123
−0·01
0·920
123
−0·04
0·693
123
−0·05
0·726
63
0·02
0·809
123
−0·01
0·953
123
−0·06
0·494
123
−0·10
0·258
123
0·19
0·038
123
0·16
0·074
123
−0·15
0·106
123
0·01
0·877
123
−0·23
0·011
123
−0·19
0·127
63
Discussion
We were able to divide several measures of immune function
of roach to four functional categories by principal component
analysis: (1) total leukocyte counts dominated by lymphocytes,
(2) function of phagocytes including chemotaxis and oxidative
burst, and IgM antibody concentration, (3) proportion of
phagocytotic cells in the blood, and (4) the relative size of
the spleen. Our results show that the functional qualities of
phagocytes (i.e. respiratory burst and chemotaxis) relate to
the proportion of dead R. campanula parasites, potentially
killed by host, in a wild fish species. Similarly, our results
suggest that the measurement of IgM antibody concentration
is positively associated to the ability of the fish to control
parasites. In contrast, the (IC1) relative proportion of
lymphocytes and count of leukocytes was negatively, and
relative count of thrombocytes positively associated with the
proportion of dead R. campanula. Previously, Vainikka et al.
(2005) used a similar approach for tench (Tinca tinca) and
also demonstrated divergent associations between the head
kidney and blood phagocytes. The fourth immune component did not relate to the count of parasites, or proportion
of dead parasites, which suggests that the relative size of the
spleen may be a poor indicator of immune function in roach,
and not related to the more direct measures of immunity
(c.f. Vainikka et al. 2005).
The observed positive correlations between immune
measures and parasite loads suggest that many measures
of immune function may be up-regulated in response to
increased parasite loads. However, the causal relationships
are difficult and sometimes impossible to confirm in correlative studies. As detected in another cyprinid, bream (Abramis
brama L.), allelic composition of major histocompatability
(MHC) genes can affect the occurrence of certain parasites
and even the size of the spleen (Ottová et al. 2007). Therefore,
experimental studies are urgently needed to link the MCH
to immune function and susceptibility to parasites both in
cyprinids and in general. In our study, seasonal covariation
probably reflected to the observed correlations in the pooled
data, but also affected the formation of the immune components. However, seasonal covariation between the parameters
of immunity likely indicates a common functional category
rather than a fully artificial collinear seasonal fluctuation.
Therefore, our results suggest that relatively simple measures
of immune function, such as those included in the IC2, can be
biologically meaningful; here roach with high immune
component two (IC2) values had efficiently eliminated
R. campanula parasites in their body.
© 2008 The Authors. Journal compilation © 2008 British Ecological Society, Functional Ecology, 23, 187–195
Parasites and immune function in roach 193
The observed relationship between the IC2 and the proportion
of dead R. campanula parasites is interesting, particularly in
light of the following. Firstly, R. campanula has been
suggested to be harmful to roach (Baturo 1977), and it is
prevalent and abundant in the current study population, as
well as in many other roach populations (Taskinen & Kortet
2002). Secondly, breeding tubercles of roach are known to
convey information about the proportion of dead R. campanula
(Taskinen & Kortet 2002); therefore sexual ornaments of
roach may signal good functional immunocompetence.
Roach are known to produce effective specific antibodies
against another Rhipidocotyle species; R. fennica that occurs
predominantly on fins (Aaltonen et al. 1997). Thus, it is likely
that the resistance against R. campanula is also mediated by
specific antibodies (see also Aaltonen et al. 1994). The
positive association between the concentration of antibodies
and functional measures of phagocytes was also expected
since phagocytosis is strongly enhanced by specific antibodies
(Verho et al. 2005). However, the proportion of dead
parasites observed on individuals captured in the wild is
susceptible to ecological factors, and may not represent a
direct measure of immunocompetence. Uneven exposure to
these parasites might bias also the proportions especially over
time, as older individuals might accumulate more dead individuals than young fish. The clearance time of dead parasite
specimens is not known. However, at population level there
are no reasons to expect biasing effects in addition to large
variation in the abundance of these parasites.
Granulocytes and macrophages form the most important
cellular first-line defence in fish (Secombes & Fletcher 1992).
Granulocytes are able to migrate towards chemotactic
stimulus and kill ingested pathogens by destructive enzymes
and oxygen radicals in the reaction known as respiratory
burst. Increased number of leukocytes, especially granulocytes, is a common consequence of infection (e.g. Ellis 1999;
Pulsford et al. 1995). Thus, the observed positive correlation
between parasite loads and the relative proportion of phagocytotic cells in blood in the pooled materials was expected.
Also, it was found that the functional characteristics of these
cells were positively related to the proportion of dead R. campanula
parasites, suggesting that the chemiluminescence method
assaying the respiratory burst may be a valuable tool to
estimate functional immunocompetence. Supporting the
importance of chemiluminescence in resisting parasites, the
scuticociliate parasite Uronema marinum was found to inhibit
the respiratory burst response of olive flounder Paralichthys
olivaceus as a way to increase virulence (Kwon et al. 2002). Moreover, interspecies comparison between coho salmon (Oncorhynchus
kisutch), rainbow trout (O. mykiss) and common carp (Cyprinus
carpio) suggests that the differences in the chemiluminescence
response between these species related to their resistance
against Renibacterium salmonarium (Sakai et al. 1996).
Spleen size is commonly used as an indicator of immunocompetence (e.g. Saino, Calza & Møller 1997; Simková et al.
2008), is often sexually dimorphic (Møller, Sorci & Erritzøe
1998; Kortet et al. 2003b), and correlates positively with the
abundance of parasites in a within species comparison of
cyprinid fishes (Simková et al. 2008). However in our analysis
on roach, the relative spleen size did not relate to the other
more direct measures of immune function and the IC4, most
strongly affected by the spleen size, was not related to the
parasite counts or proportions of dead parasites. Although
the size of the spleen may increase following an infection (e.g.
Lefebvre et al. 2004), our analysis suggests that spleen size
might not represent immunocompetence of roach and should
be interpreted with caution in immunological studies.
Several aspects of immune function show seasonal changes
as a response to challenging time periods such as winter,
and physiological demands such as reproduction (Kortet &
Vainikka 2008). Reproduction has been shown to impair
immune function and increase parasitism e.g. in Arctic charr,
Salvelinus alpinus (Skarstein et al. 2001). Breeding-related
immunosuppression is thought to result from a high breedingtime concentration of androgens (Aida 1988; Folstad &
Karter 1992; Hou, Suzuki & Aida 1999a,b; Muñoz et al.
2000), although changes in water temperature and in the
concentration of corticosteroids may also play a role in
explaining this annual variation (Zapata, Varas & Torroba
1992; Collazos, Ortega & Barriga 1994; Alcorn, Murray &
Pascho 2002). The seasonal variation in the individual measures
of immunity we used has been described previously by Kortet
et al. (2003a). However, the present combined measures of
immunity showed somewhat different patterns: changes in
IC1 suggested breeding-related suppression in male roach,
whereas both IC2 and IC3 showed high values during the
spawning period in spring. These changes may reflect the
direct effects of immunosuppressive hormones but also
indirect effects of increased parasite loads at the time of
breeding. Therefore, experimental studies are needed to
address the causal factors behind these seasonal changes.
Interestingly, size-adjusted parasite loads were often higher in
females than in males, suggesting that the relatively high
reproductive effort of females in terms of egg production may
contribute to the observed gender differences more than the
potential androgen-dependent immunosuppression.
The occurrence of epidermal papillomatosis has been
shown to be highly season-dependent, peaking during the
spawning period, and being more common in males than in
females (Kortet, Vainikka & Taskinen 2002). The occurrence
of the disease has been suggested to be related to the concentrations of both sex (Kortet et al. 2003b) and stress
hormones (Vainikka, Kortet & Taskinen 2004b). In this
study, we confirmed that high testosterone levels contribute to
the development of the disease. Interestingly, it seemed that
high oestradiol levels prevented outbreak; which might
explain the clear sex difference in the disease occurrence.
However, in this study we did not determine cortisol levels,
and the relationship between sex hormones and the epidermal
papillomatosis could have been caused by a correlated stress
response (Vainikka et al. 2004b).
Although immunosuppressive effects of testosterone
were thought to be well understood, at least in mammals
(Grossman 1985); the recent meta-analysis of studies testing the
immunocompetence-handicap hypothesis (Folstad & Karter
© 2008 The Authors. Journal compilation © 2008 British Ecological Society, Functional Ecology, 23, 187–195
194
A. Vainikka et al.
1992) by Roberts, Buchanan & Evans (2004) suggests that in many
species testosterone may increase susceptibility to ectoparasites, but direct effects on immune function may not exist. For
example, in red-winged blackbirds (Agelaiuis phoenicus) a
clear absence of a response on immune function by testosterone
has been reported (Hasselquist et al. 1999). In contrast, many
studies of lizards clearly support the immunosuppressivity of
testosterone (e.g. Saad, Khalek & El Ridi 1990; Klukowski &
Nelson 2001). At the population level, the trade-offs predicted
by life-history theory might be difficult to observe because
individuals of better overall quality may show superior
performance in several traits, even if trade-offs operate
between the traits at individual level (Van Noordwijk & de
Jong 1986). This might have contributed to our lack of detection
of negative correlations between testosterone or oestradiol
concentrations and the measures of immunity. However,
oestradiol concentration correlated positively with the load
of R. campanula and R. fennica parasites in the pooled samples.
Since the loads of these parasites did not vary seasonally, this
relationship may suggest that high reproductive investment
caused the increase in the numbers of these parasites.
In conclusion, we found correlations between several
individual measures of immune function, and that these
formed functional groups. The functional capacity of
phagocytotic cells, together with antibody concentration, was
most clearly related to the ability of the fish to eliminate
R. campanula, measured as the proportion of dead parasites.
Conversely, the number of phagocytotic cells in circulation was
related to a high parasite load when the seasonality in these
patterns was not controlled for. Our study demonstrates that it is
of prominent importance to assay immune functions at seasonal
scales to make reliable and ecologically relevant conclusions.
Some correlations arose because of parallel seasonal changes in
immunity and the parasites’ life cycle, but when seasonal
effects were accounted for, only a few associations remained.
Our results suggest that the quantitative and functional measures
of phagocytosis and antibodies can provide information
about parasite resistance, whereas numbers of white blood
cells may increase as a response to high parasitic exposure.
Acknowledgements
The study was conducted in the Biological Interactions graduate school of the
Finnish Ministry of Education (A.V., R.K.). During the manuscript finalisation
AV was funded by European Marie Curie Research Training Network
FishACE (Fisheries-induced Adaptive Changes in Exploited Stocks), funded
through the European Community’s Sixth Framework Programme (Contract
MRTN-CT-2004-005578) and RK by University of Helsinki and by University
of Oulu. We thank Raahe Steel for financial support of the study. We are
grateful to A. Bagge, P. Hoskonen, A. Tuomainen, A. Tarvainen, M. Saarinen,
H. Salo and T. Sinisalo for assistance in the field and laboratory. We thank The
Academy of Finland (project 204837), The Biological Society of Finland
Vanamo and Emil Aaltonen foundation for grants to RK. We also thank Maj
and Tor Nessling Foundation and Niilo Helander Foundation for grants to JT.
We thank Heidi Pardoe for commenting the manuscript.
References
Aaltonen, T.M., Jokinen, E.I. & Valtonen, E.T. (1994) Antibody synthesis in
roach (Rutilus rutilus): analysis of antibody secreting cells in lymphoid
organs with ELISPOT-assay. Fish & Shellfish Immunology, 4, 129–140.
Aaltonen, T.M., Valtonen, E.T. & Jokinen, E.I. (1997) Humoral response of
roach (Rutilus rutilus) to digenean Rhipidocotyle fennica infection.
Parasitology, 114, 285–291.
Aida, M. (1988) A review of plasma hormone changes during ovulation in
cyprinid fishes. Aquaculture, 74, 11–21.
Alcorn, S.W., Murray, A.L. & Pascho, R.J. (2002) Effects of rearing temperature
on immune functions in sockeye salmon (Oncorhynchus nerka). Fish & Shellfish
Immunology, 12, 303–334.
Andersson, M. (1994) Sexual Selection. Princeton University Press, Princeton.
Anonymous (1996) Guidelines for the Use of Animals in Research. Animal Behaviour, 51, 241–246.
Baturo, B. (1977) Bucephalus polymorphus Baer, 1827 and Rhipidocotyle illense
(Ziegler, 1883) (Trematoda, Bucephalidae): morphology and biology of
developmental stages. Acta Parasitology Polonica, 24, 203–220.
Braude, S., Tang-Martinez, Z. & Taylor, G.T. (1999) Stress, testosterone, and
the immunoredistribution hypothesis. Behavioral Ecology, 10, 345– 350.
Brummer Korvenkontio H., Tellervo Valtonen E. & Pugachev O.N. (1991)
Myxosporea parasites in roach, Rutilus rutilus (Linnaeus), from four lakes in
central Finland. Journal of Fish Parasitology, 38, 573–586.
Casadevall, A. (2004) The methodology for determining the efficacy of
antibody-mediated immunity. Journal of Immunological Methods, 291, 1–10.
Clarke, B.C. (1979) The evolution of genetic diversity. Proceedings of the Royal
Society of London B, 205, 453–474.
Collazos, M.E., Ortega, E. & Barriga, C. (1994) Effect of temperature on the
immune system of a cyprinid fish (Tinca tinca L.). Seasonal variation in the
titres. Fish & Shellfish Immunology, 4, 231–238.
Ellis, A.E. (1999) Immunity to bacteria in fish. Fish & Shellfish Immunology, 9,
291–308.
Folstad, I. & Karter, A.J. (1992) Parasites, bright males, and the immunocompetence handicap. The American Naturalist, 139, 603–622.
Gibson, D.I., Valtonen, E.T. & Taskinen, J. (1992) Studies on bucephalid
digeneans parasitising molluscs and fishes in Finland. II. The description of
Rhipidocotyle fennica n. sp. and its discrimination by principal component
analysis. Systematic Parasitology, 23, 67–79.
Goater, C.P. & Holmes, J.C. (1997) Parasite-mediated natural selection.
Host-parasite Evolution: General Principles and Avian Models (eds D.H.
Clayton & J. Moore), pp. 9–29. Oxford University Press, Oxford.
Grossman, C.J. (1985) Interactions between the gonadal steroids and the
immune system. Nature, 18, 257–261.
Hakoyama, H., Nishimura, T., Matsubara, N. & Iguchi, K. (2001) Difference in
parasite load and nonspecific immune reaction between sexual and gynogenetic
forms of Carassius auratus. Biological Journal of the Linnean Society, 72,
401–407.
Halmetoja, A., Valtonen, E.T. & Taskinen, J. (1992) Trichodinids (protozoa)
on fish from four central Finnish lakes of differing water quality. Aqua
Fennica, 22, 59–70.
Hamilton, W.D. & Zuk, M. (1982) Heritable true fitness and bright birds: a role
for parasites? Science, 218, 384–386.
Hasselquist, D, Marsh, J.A., Sherman, P.W. & Wingfield, J.C. (1999) Is avian
humoral immunocompetence suppressed by testosterone? Behavioral Ecology
and Sociobiology 45, 167–175.
Hedges, L.V. & Olkin, I. (1985) Statistical Methods for Meta-Analysis. Academic
Press, Orlando, DL.
Hillgarth, N. & Wingfield, J.C. (1997) Parasite-mediated sexual selection:
endocrine aspects. Host-parasite Evolution: General Principles and Avian
Models (eds D.H. Clayton & J. Moore), pp. 78–104. Oxford University
Press, Oxford.
Hoffmann, R.W., Körting, W., Fischer-Scherl, T. & Schäfer, W. (1990) An
outbrake of bucephalosis in fish of the Main river. Angewandte Parasitologie,
31, 95–99.
Hou, Y., Suzuki, Y. & Aida, K. (1999a) Effects of steroids on the antibody
producing activity of lymphocytes of rainbow trout. Fisheries Science, 65,
850–855.
Hou, Y.Y., Suzuki, Y. & Aida, K. (1999b) Effects of steroid hormones on immunoglobulin M (IgM) in rainbow trout, Oncorhynchus mykiss. Fish Physiology
and Biochemistry, 20, 155–162.
Karvonen, A., Seppälä, O. & Valtonen, E.T. (2004) Eye fluke-induced cataract
formation in fish: quantitative analysis using an opthalmological microscope.
Parasitology, 129, 473–478.
Khan, R.A., Barker, D.E., Williamsryan, K. & Hooper, R.G. (1994) Influence
of crude oil and paper and pulp mill effluent on mixed infections of Trichodina
cottidarium and T. saintjohnsi (Ciliophora) parasitising Myxocephalus
octodecemspinosus and M. scorpius. Canadian Journal of Zoology, 72, 247 – 251
Klukowski, M. & Nelson, C.E. (2001) Ectoparasite loads in free-ranging northern fence lizards, Sceloporus undulatus hyacinthinus: effects of testosterone
and sex. Behavioral Ecology and Sociobiology, 49, 289–295.
© 2008 The Authors. Journal compilation © 2008 British Ecological Society, Functional Ecology, 23, 187–195
Parasites and immune function in roach 195
Korkea-Aho, T.L., Partanen, J.M., Vainikka, A. & Taskinen, J. (2006b)
Association between environmental stress and epidermal papillomatosis of
roach (Rutilus rutilus L.). Diseases of Aquatic Organisms, 72, 1– 8.
Korkea-Aho, T.L., Partanen, J.M., Kukkonen, J.V.K. & Taskinen, J. (2008)
Hypoxia increases intensity of epidermal papillomatosis in roach Rutilus
rutilus. Diseases of Aquatic Organisms, 78, 235–241.
Korkea-Aho, T., Vainikka, A. & Taskinen, J. (2006a) Factors affecting the
intensity of epidermal papillomatosis in populations of roach, Rutilus
rutilus L., estimated as scale coverage. Journal of Fish Diseases, 29, 115–
122.
Kortet, R., Taskinen, J., Sinisalo, T. & Jokinen, I. (2003a) Breeding-related
seasonal changes in immunocompetence, health state and condition of the
cyprinid fish, Rutilus rutilus, L.. Biological Journal of the Linnean Society,
78, 117 –127.
Kortet, R. & Vainikka, A. (2008) Seasonality of innate immunity; evolutionary
aspects and latest updates. New Research on Innate Immunity (eds M.
Durand & C.V. Morel). Nova Science Publishers, New York. In press.
Kortet, R., Vainikka, A., Rantala, M.J., Jokinen, I. & Taskinen, J. (2003b)
Sexual ornamentation, androgens and papillomatosis in male roach (Rutilus
rutilus). Evolutionary Ecology Research, 5, 411–419.
Kortet, R., Vainikka, A., Rantala, M.J., Myntti, J. & Taskinen, J. (2004) In vitro
embryo survival and early viability of larvae in relation to male sexual
ornaments and parasite resistance in roach, Rutilus rutilus L. Journal of
Evolutionary Biology, 17, 1337–1344.
Kortet, R., Vainikka, A. & Taskinen, J. (2002) Epizootic cutaneous papillomatosis in roach Rutilus rutilus: sex and size dependence, seasonal occurrence
and between-population differences. Diseases of Aquatic Organisms, 52,
185 –190.
Koskivaara, M. & Valtonen, E.T. (1992) Dactylogyrus (Monogenea) communities on the gills of roach in three lakes in central Finland. Parasitology,
104, 263 – 272.
Kurtz, J., Kalbe, M., Aeschlimann, P.B., Häberli, M.A., Wegner, K.M.,
Reusch, T.B.H. & Milinski, M. (2004) Major histocompatibility complex
diversity influences parasite resistance and innate immunity in sticklebacks.
Proceedings of the Royal Society of London, Series B, 271, 197–204.
Kwon, S.R, Kim, C.S., Chung, J.K., Lee, H.H. & Kim, K.H. (2002) Inhibition
of chemiluminescence response of olive flounder Paralichthys olivaceus
phagocytes by the scuticocliliate Uronema marineum. Diseases of Aquatic
Organisms, 52, 119 –122.
Lefebvre, F., Mounaix, B., Poizat, G. & Grivelli, A.J. (2004) Impacts of the
swimbladder nematode Anguillicola crassus on Anguilla anguilla: variations
in liver and spleen masses. Journal of Fish Biology, 64, 435–447.
Lehtinen, K.-J. (1989) Survival, growth and disease of three-spined stickleback,
Gasterosteus aculeatus L., brood exposed to bleached kraft mill effluents
(BKME) in mesocosms. Annales Zoologici Fennici, 26, 133–144.
Måsvær M., Liljedahl, S. & Folstad, I. (2004) Are seconday sex traits, parasites
and immunity related to variation in primary sex traits in the Arctic charr?
Proceedings of the Royal Society, Biology Letters, 271, S40–S42.
Møller, A.P., Sorci, G. & Erritzøe, J. (1998) Sexual dimorphism in immune
defence. The American Naturalist, 152, 605–619.
Muñoz, F.J., Galván, A., Lerma, M. & De la Fuente, M. (2000) Seasonal
changes in peripheral blood leukocyte functions of the turtle Mauremys
caspica and their relationship with corticosterone, 17-β-estradiol and
testosterone serum levels. Veterinary Immunology and Immunopathology, 77,
27 – 42.
Ottová, E., Simková, A., Jurajda, P., Dávidová, M., Ondrakovác, M., Pecínková,
M. & Gelnar, M. (2005) Sexual ornamentation and parasite infection in
males of common bream (Abramis brama): a reflection of immunocompetence
status or simple cost of reproduction? Evolutionary Ecology Research, 7,
581– 593.
Ottová, E., Simková, A. & Morand, S. (2007) The role of major histocompatibility
complex diversity in vigour of fish males (Abramis brama L.) and parasite
selection. Biological Journal of the Linnean Society, 90, 525–538.
Price, P.W. (1980) Evolutionary Biology of Parasites. Princeton University
Press, Princeton, New Jersey.
Pulsford, A.L., Crampe, M., Langston, A. & Glynn, P.J. (1995) Modulatory
effects of disease, stress, copper, TBT and vitamin E on the immune system
of flatfish. Fish & Shellfish Immunology, 5, 631–643.
Roberts, M.L., Buchanan, K.L. & Evans, M.R. (2004) Testing the immunocompetence handicap hypothesis: a review of the evidence. Animal Behaviour,
68, 227 – 239.
Saad, A.H., Khalek, N.A. & El Ridi, R. (1990) Blood testosterone level: a seasondependent factor regulating immune reactivity in lizards. Immunobiology,
180, 184 –194.
Saino, N., Calza, S. & Møller, A.P. (1997) Immunocompetence of nestling barn
swallows in relation to brood size and parental effort. Journal of Animal
Ecology, 66, 827–836.
Sakai, M., Miyama, K., Atsuta, S. & Kobayashi, H. (1996) The chemiluminescent
responses of phagocytic cells of coho salmon, rainbow trout and carp
against Renibacterium salmonarium. Fish & Shellfish Immunology, 6, 71– 73.
Schmid-Hempel, P. & Ebert, D. (2003) On the evolutionary ecology of specific
immune defence. Trends in Ecology and Evolution, 18, 27–32.
Scott, A.L. & Klesius, P.H. (1981) Chemiluminescence: a novel analysis of
phagocytosis in fish. Development in Biological Standards, 49, 243– 254.
Secombes, C.J. & Fletcher, T.C. (1992) The role of phagocytes in the protective
mechanisms of fish. Annual Review of Fish Diseases, 2, 53–71.
Shaw, A.F.B. (1930) A direct method for counting the leukocytes, trombocytes
and erytrocytes of bird’s blood. Journal of Pathology and Bacteriology, 33,
833–835.
Shul’man, S.S. (1990) Myxosporidia of the USSR. A.A. Balkema, Rotterdam.
Simková, A., Lafond, T., Ondracková, M., Jurajda, P., Ottová, E. & Morand,
S. (2008) Parasitism, life history traits and immune defence in cyprinid fish
from Central Europe. BMC Evolutionary Biology, 8, 29.
Skarstein, F., Folstad, I. & Liljedahl, S. (2001) Whether to reproduce or not:
immune suppression and cost of parasites during reproduction in the Arctic
charr. Canadian Journal of Zoology, 79, 271–278.
Taskinen, J. & Kortet, R. (2002) Dead and alive parasites: sexual ornaments
signal parasite resistance in the male fish, Rutilus rutilus. Evolutionary
Ecology Research, 4, 919–929.
Taskinen, J., Valtonen, E.T. & Gibson, D.I. (1991) Studies on bucephalid
digeneans parasitising molluscs and fishes in Finland. I. Ecological data and
experimental studies. Systematic Parasitology, 19, 81–94.
Tschirren, B. & Richner, H. (2006) Parasites shape the optimal investment in
immunity. Proceedings of the Royal Society of London, Series B, 273, 1773 –
1777.
Tuuha, H., Valtonen, E.T., & Taskinen, J. (1992) Ergasilid copepods as parasites
of perch Perca fluviatilis and roach Rutilus rutilus in central Finland:
seasonality, maturity and environmental influence. Journal of Zoology, 228,
405–422.
Vainikka, A., Jokinen, E.I., Kortet, R. & Taskinen, J. (2004a) Gender- and
season-dependent relationships between testosterone, oestradiol and immune
functions in wild roach Rutilus rutilus L. Journal of Fish Biology, 64, 227 –
240.
Vainikka, A., Kortet, R., Paukku, S., Rantala, M.J. & Pirhonen, J. (2005) What
do male tench, Tinca tinca, advertise with morphological ornaments? Acta
Ethologica, 8, 70–78.
Vainikka, A., Kortet, R. & Taskinen, J. (2004b) Epizootic cutaneous papillomatosis, cortisol and male ornamentation during and after breeding in roach
Rutilus rutilus. Diseases of Aquatic Organisms, 60, 189–195.
Valtonen, E.T., Haaparanta, A. & Hoffmann, R.W. (1994) Occurrence and
histological response of Raphidascaris acus (Nematoda: Ascaridoidea) in
roach from four lakes differing in water quality. International Journal for
Parasitology, 24, 197–206.
Valtonen, E.T., Holmes, J.C. & Koskivaara, M. (1997) Eutrophication,
pollution, and fragmentation: effects on parasite communities in roach
(Rutilus rutilus) and perch (Perca fluviatilis) in four lakes in central Finland.
Canadian Journal of Fisheries and Aquatic Sciences, 54, 572–585.
Van Noordwijk, A.J. & de Jong, G. (1986) Acquisition and allocation of
resources: their influence on variation in life history tactics. The American
Naturalist, 128, 137–142.
Verho, S., Järvinen, S., Nikoskelainen, S. & Lilius, E-M. (2005) Biological effect
of vaccination can be assessed directly from diluted whole blood of rainbow
trout using homologous blood phagocytes as immunosensors. Fish &
Shellfish Immunology, 19, 175–183.
Viney, M.E., Riley, E.M. & Buchanan, K.L. (2005) Optimal immune
responses: immunocompetence revisited. Trends in Ecology and Evolution,
20, 665–669.
Wedekind, C. (1992) Detailed information about parasites revealed by sexual
ornamentation. Proceedings of the Royal Society of London B, 247, 169 –174.
Wedekind, C. & Folstad, I. (1994) Adaptive or nonadaptive immunosuppression by sex hormones? The American Naturalist, 143, 936–938.
Zahavi, A. (1975) Mate selection – a selection for a handicap. Journal of
Theoretical Biology, 53, 205–214.
Zahavi, A. (1977) The cost of honesty (further remarks on the handicap principle).
Journal of Theoretical Biology, 67, 603–605.
Zapata, A.G., Varas, A. & Torroba, M. (1992) Seasonal variations in the
immune system of lower vertebrates. Immunology Today, 13, 142–147.
Received 30 March 2008; accepted 13 August 2008
Handling Editor: Mike Siva-Jothy
© 2008 The Authors. Journal compilation © 2008 British Ecological Society, Functional Ecology, 23, 187–195