Content uploaded by Olivia Roth
Author content
All content in this area was uploaded by Olivia Roth on Mar 05, 2014
Content may be subject to copyright.
The University of Chicago
Male Pregnancy and Biparental Immune Priming
Author(s): Olivia Roth, Verena Klein, Anne Beemelmanns, Jörn P. Scharsack and Thorsten B.
H. Reusch,
Reviewed work(s):
Source:
The American Naturalist,
(-Not available-), p. 000
Published by: The University of Chicago Press for The American Society of Naturalists
Stable URL: http://www.jstor.org/stable/10.1086/668081 .
Accessed: 01/11/2012 06:29
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp
.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.
.
The University of Chicago Press, The American Society of Naturalists, The University of Chicago are
collaborating with JSTOR to digitize, preserve and extend access to The American Naturalist.
http://www.jstor.org
vol. 180, no. 6 the american naturalist december 2012
Male Pregnancy and Biparental Immune Priming
Olivia Roth,
1,
* Verena Klein,
1
Anne Beemelmanns,
1
Jo¨rn P. Scharsack,
2
and
Thorsten B. H. Reusch
1
1. Helmholtz Centre for Ocean Research Kiel (GEOMAR), Evolutionary Ecology of Marine Fishes, Du¨sternbrooker Weg 20, 24105 Kiel,
Germany; 2. Institute for Evolution and Biodiversity, Animal Evolutionary Ecology, University of Mu¨nster, Hu¨fferstrasse 1, 48149
Mu¨nster, Germany
Submitted May 8, 2012; Accepted July 26, 2012; Electronically published October 31, 2012
Online enhancement: appendix. Dryad data: http://dx.doi.org/10.5061/dryad.r6624.
abstract: In vertebrates, maternal transfer of immunity via the
eggs or placenta provides offspring with crucial information on pre-
vailing pathogens and parasites. Males contribute little to such trans-
generational immune priming, either because they do not share the
environment and parasite pressure of the offspring or because sperm
are too small for transfer of immunity. In the teleost group of Syng-
nathids (pipefish, seahorses, and sea dragons), males brood female
eggs in a placenta-like structure. Such sex-role-reversed species pro-
vide a unique opportunity to test for adaptive plasticity in immune
transfer. Here, males and females should both influence offspring
immunity. We experimentally tested paternal effects on offspring
immunity by examining immune cell proliferation and immunegene
expression. Maternal and paternal bacterial exposure induced off-
spring immune defense 5 weeks after hatching, and this effect per-
sisted in 4-month-old offspring. For several offspring immune traits,
double parental exposure (maternal and paternal) enhanced the re-
sponse, whereas for another group of immune traits, the transgen-
erational induction already took place if only one parent wasexposed.
Our study shows that sex role reversal in connection with male
pregnancy opens the door for biparental influences on offspring im-
munity and may represent an additional advantage for the evolution
of male pregnancy.
Keywords: parental effects, host-parasite interaction, immune defense,
placenta, pipefish, Vibrio.
Introduction
The non-DNA-based transfer of heritable information
from parents to offspring has recently received increasing
attention. Darwinian selection changes genotype frequen-
cies at the population level, whereas non–genetically in-
herited traits have the potential for adaptive plasticity from
one generation to another within a pedigree (Mousseau
and Fox 1998; Haig 2000). One of the strongest selection
pressures is parasites (Hamilton et al. 1990). To counter
* Corresponding author; e-mail: oroth@geomar.de.
Am. Nat. 2012. Vol. 180, pp. 000–000. 䉷2012 by The University of Chicago.
0003-0147/2012/18006-53836$15.00. All rights reserved.
DOI: 10.1086/668081
negative fitness effects by parasites, hosts evolved an ef-
ficient and often highly specific immune system (Altizer
et al. 2003; Boots and Bowers 2004). The resulting co-
evolutionary dynamics between hosts and parasites lead
to rapid changes in parasite genotypes and species. How-
ever, certain parasite groups often persist in the environ-
ment for more than just one host generation (Van Valen
1973; Hamilton 1980; Eizaguirre et al. 2012). Because an
offspring shares the same parasite environment as a parent
early in its life, the transfer of immunological experience
(transgenerational immune priming) on abundance and
composition of pathogens to offspring represents a fast
alternative way for hosts to effectively react to pathogens
(Boulinier and Staszweski 2008; Hasselquist and Nilsson
2009).
Such transgenerational immune priming is a phenotypic
plastic effect (Lemke et al. 2004; Poulin and Thomas 2008)
that is defined over parental effects that boost the immune
defense of the offspring. It occurs in vertebrates via the
deposition of maternal antibodies into eggs of fish and
birds (Bly et al. 1986; Fuda et al. 1992; Grindstaff et al.
2003) or via the transfer of antibodies over the placenta
and during breast feeding in mammals (Glezen 2003;
Lemke et al. 2004). Transfer of immunity is not limited
to acquired immunity (maternal antibodies) but also in-
cludes components of the innate immune defense (e.g.,
lysozyme enzymes; Hanif et al. 2004) and components of
the complement system (Ellingsen et al. 2005). The trans-
fer of as-yet-unknown components of the innate immune
defense was recently shown also for invertebrates (Sadd et
al. 2005; Moret 2006; Sadd and Schmid-Hempel 2007;
Roth et al. 2010; Zanchi et al. 2011), which points to the
universal presence of adaptive plasticity in transgenera-
tional transfer and induction of immunity across the an-
imal kingdom.
An activation of the immune defense is costly and
should be traded off against other life-history components
(Sheldon and Verhulst 1996). Hence, because investment
000 The American Naturalist
into immune defense can reduce life expectancy (Bon-
neaud et al. 2004), immunochallenged individuals may
increase their investment in the current reproductive out-
put as the chances of surviving to reproduce later in life
are reduced (formulated as the “terminal investment hy-
pothesis”; Clutton-Brock 1984). Higher investment on an
immunochallenge or infection in terms of egg-laying rate
(Adamo 1999), pheromone production (Sadd et al. 2006),
and brood care (Hanssen 2006) has been shown. The pos-
itive effects gained by transgenerational immune priming
on offspring immune defense can thus even be enhanced
by terminal investment. As yet, however, the majority of
experiments have revealed decreased investment in repro-
duction, such that cases of decreased egg-laying rates
(Shoemaker et al. 2006), parental effort (Ra˚berg et al.
2000), and production of nuptial gifts (Reaney and Knell
2010) were reported; probably costs of immune activation
were too high to permit allocation of more resources to
reproduction.
Thus far, the major mechanisms of transfer of immunity
in vertebrates have been suggested to be functionally re-
stricted to mothers, because sperm cells are probably too
small to transfer antibodies or components of the com-
plement system to the zygote (Wassarmann et al. 2001;
Arnqvist and Rowe 2005). This is in line with findings
that no relationships among antibody titers were detected
in vertebrates when fathers but not mothers were primed
(Gasparini et al. 2002; Reid et al. 2006). However, in some
sex-role-reversed animals, such as the bony fish group of
Syngnathids (pipefishes and seahorses), males have
evolved a unique placenta-like structure. This structure
fulfils certain requirements of the mammal placental sys-
tem, and we thus hypothesize that pipefish could poten-
tially have evolved a mechanism for the transfer of im-
munological information from fathers to offspring during
male pregnancy. In Syngnathids, females transfer eggs to
the male brood pouch where they are provided with nu-
trients and oxygen (Wilson et al. 2001; Dzyuba et al. 2006;
Harlin-Cognato et al. 2006; Sto¨ lting and Wilson 2007; Rip-
ley and Foran 2009). Because pipefish males invest a higher
proportion of resources in the offspring, males are limited
in their reproductive potential according to Bateman’s
principle (Bateman 1948; Jones et al. 2000, 2005). Hence,
their life history is selected for longevity, which is consis-
tent with a more efficient immune response compared with
that of females (Roth et al. 2011). The transfer of im-
munological knowledge from fathers to offspring would
thus also fulfill another function that is usually restricted
to mothers under conventional sex roles. In addition, off-
spring are born into the environment of the father and
are therefore exposed to the same parasite genotypes as
the father. We thus hypothesize that, in sex-role-reversed
animals with placenta-like structures, selection for paternal
transfer of immunity should be more important for im-
mune transfer than transfer via mothers from both a
mechanistic and an evolutionary point of view.
To test these ideas, we examined whether male pipefish
influence immune defense of the offspring in the broad-
nosed pipefish Syngnathus typhle in a fully crossed labo-
ratory experiment. To do so, the mother, the father, both,
or neither were exposed to injections with a mix of several
phylotypes of heat-killed Vibrio bacteria to simulate the
abundance of a community of potentially harmful path-
ogens. Thus, our study mainly addressed constitutive up-
regulation of basal immunocompetence in the offspring,
in particular the vertebrate’s costly innate immunity (Ra˚-
berg et al. 2002). The use of heat-killed bacteria allowed
us to exclusively concentrate on host immune defense
without confounding pathogenicity effects. The F
1
gen-
eration was collected and exposed to an immune challenge
to address the persistence of possible transgenerational
immune effects after 5 weeks and after 4 months. Direct
immune measurements (in 4-month-old individuals only)
and large-scale immune gene expression patterns (in 5-
week-old and 4-month-old individuals) were then as-
sessed. In addition, body length was measured after 4
months to address possible effects due to terminal in-
vestment into reproduction. Our main goal was to assess
whether both maternal and paternal immune experience
interact or whether strength of offspring immune defense
depends only on paternal or maternal effects.
Material and Methods
Experimental Design
Our study species is the European broad-nosed pipefish
Syngnathus typhle. Unmated pipefish were caught around
the island of Gotland, Sweden, in May 2011 before the
breeding season. Within 48 hours, animals were trans-
ported to our laboratory facility in Kiel, Germany, where
they were sex separated and acclimatized to local water
(∼18⬚C, 15 psu) over a 3-week period in 200-L tanks. We
used only pipefish that had entered their first breeding
season. All animals reached sexual maturity within the 3
weeks in Kiel; females started egg production, and males
developed a brood pouch. Two weeks before mating, pipe-
fish were injected with either 50 mLof10
8
cells mL
⫺1
heat-
killed Vibrio (a combination of 8 allopatric strains: D11K1,
I11E3, I2K1, D1K1, SH54, D1K3, D1E3, and D12K2) ac-
cording to Roth et al. (2012) in phosphate-buffered saline
(PBS) or PBS only as sham control (Roth et al. 2011).
After injection, animals were kept in pairs with sexes and
injection treatments separated in thirty-six 80-L aquaria
under continual water exchange. They were fed twice a
day with live and frozen mysids and Artemia salina.Two
Biparental Immune Priming 000
weeks after injection, mating pairs were formed in a fully
reciprocal design, with seven pairs per parental treatment,
resulting in a total of 28 mating pairs: 乆⫹么⫹,乆⫺么⫹,
乆⫹么⫺,乆⫺么⫺, where ⫹and ⫺indicate injection with
Vibrio and sham control, respectively. We obtained 18 suc-
cessful broods with offspring numbers, approximately
equally distributed among treatments ( breeding
Np4–5
pairs per parental treatment). Immediately after birth, off-
spring were separated from the parental aquaria and placed
by family into 22-L aquaria. Offspring were fed 4 times
d
⫺1
with A. salina naupliae and copepods. At the age of
5 weeks, between 28 and 40 offspring per family were
challenged with either the heat-killed Vibrio strain com-
bination homologous to the parental exposure, a different
(heterologous) single bacteria strain (I2K3), or PBS or
were left naive (7–10 animals per treatment per family).
For this, a needle was dipped into a solution of 10
9
cells
mL
⫺1
heat-killed Vibrio, and the fish was pricked intra-
peritoneally. After the challenge, pipefish were kept in 7-
L beakers (one beaker per offspring treatment per family).
Animals were killed by MS222 24 h later, immersed in
RNA for 24 h, and frozen at ⫺80⬚C. A subset of animals
from 16 families was grown until the age of 4 months (1–
4 animals per family; a total of 58 animals) and then either
exposed to 20 mL of heat-killed Vibrio intraperitoneally
(10
8
cells mL
⫺1
), homologous to the parental Vibrio com-
bination, or kept naive (1–2 animals per treatment per
family). For individual identification, fluorescent visible
implant tags were injected subcutaneously. Twenty-four
hours later, animals were killed with MS 222. Three ani-
mals died during those 24 h, which resulted in a total of
55 animals available for analysis. Head kidneys, where lym-
phocytes are produced, were dissected and used for direct
central immune measurements, whereas blood was col-
lected for direct peripheral immune measurements, be-
cause lymphocytes migrate into the blood to fight a bac-
terial infection. Because Syngnathids lack a spleen
(Matsunaga and Rahman 1998), additional immune or-
gans could not be included for immune measurements.
Gills were preserved in RNA for gene expression mea-
surements, whereas a piece of the tail was collected for
additional DNA extraction.
Blood cells and head kidney cells were prepared ac-
cording to Roth et al. (2011) and Landis et al. (2012). Cells
were kept on ice until cellular immune activity (proportion
of monocytes and lymphocytes and lymphocyte activity,
measured using cell cycle analysis [resting cells are from
G
0–1
phase and the dividing cells are from S⫹G
2-M phase
])
was assessed with a fluorescence-activated cell scanner us-
ing head kidney cells as proxy for central immune activity
and blood cells as proxy for peripheral immune activity.
Assessment of Immune Function via Quantitative Real-
Time Polymerase Chain Reaction (q-RT-PCR)
The amount of mitochondrial RNA of 14 immune-related
genes was quantified using q-RT-PCR relative to a house-
keeping gene. Primers for 10 immune genes were available
from an earlier study (Birrer et al. 2012). For another four
immune genes, novel primers were designed on the basis
of annotated genes in an expressed sequence tag library.
We designed primer pairs using the software primer 3
(v.04.0; Rozen and Skaletzky 1997) for putative homolo-
gous genes encoding allograft inflammation factor (allo),
complement component 9 (c9), B cell receptor-associated
protein 31 (bcell), and translocator protein (tspo). Ho-
mology was inferred from BLASTX hits of other teleost
species with e-values !10
E⫺20
. The primer efficiency was
assessed using dilution series (1 : 1 to 1 : 36 of diluted
cDNA) as templates. All genes attained slopes of log
2
cycle
number versus log fluorescent intensity between 0.9 and
1, and all correlation coefficients were . Melt-
2
0.9 !R!1
ing curves after the q-RT-PCR were inspected to exclude
nonspecific amplification products (Birrer et al. 2012).
Gene Expression
Immune gene expression was measured using two different
platforms (StepOnePlus [Applied Biosystems] and Flu-
idigm Biomark [HD Systems]). Intercalibration assured
that results are directly comparable. In the 4-month-old
group, gene expression was measured with q-RT-PCR
(StepOnePlus). Four to five individuals per parental and
offspring treatment were randomly chosen, which resulted
in 36 samples that were used for RNA extraction with the
Invitrap Universal HTS 96 Kit/C from Invitek. The RNA
yield was measured with a Nanodrop ND-1000 (peQLab).
Reverse transcription was performed with the Quanti-
TectReverseTranscription kit (Qiagen) and incorporated a
genomic DNA (gDNA) digestion. The q-RT-PCR reactions
were performed with 5 ng/mL total transcribed RNA fol-
lowing the method described by Birrer et al. (2012) in-
cluding non-reverse-transcribed negative controls and
nontemplate controls. Expression of 10 immune genes
(allo,bcell,grcsf,tlr,ig ant,il,tnf,tspo,c9, and cf )plus
housekeeping gene expression (ubiquitin [ubi]) was as-
sessed. Data were analyzed with the StepOnePlus Real-
Time PCR System, and cycle time (C
T
) values were cal-
culated for every reaction using a fixed fluorescence
threshold of 0.5.
In a second set up, gene expression was measured for
both 5-week-old and 4-month-old animals using a Flu-
idigm Biomark based on chips, and RNA was48 #48
extracted from a total of 384 animals (both juveniles and
mature animals). For juveniles, offspring from 15 families
000 The American Naturalist
were taken, with up to seven replicates per offspring treat-
ment (the exact number of animals used per family was
as follows: family 1, 25; family 2, 26; family 5, 22; family
7, 24; family 9, 26; family 10, 26; family 11, 26; family 13,
16; family 15, 25; family 16, 25; family 19, 7; family 20,
25; family 24, 17; family 25, 26; and family 28, 9). For
mature animals, all replicates were taken. For reverse tran-
scription, the QuantiTectReverseTranscription kit was
used, including a gDNA wipeout buffer, as described
above. Preamplification steps were performed for a mix-
ture of 13 immune gene primers (allo,bcell,c9,grcsf,hsp,
ig ant,il,kin,nramp,tlr,tnf,tspo, and cf ) plus the house-
keeping gene primer for all samples, and 2.5 mL TaqMan
PreAmp Master Mix (Applied Biosystems) and 1.25 mL
200 nM pooled primer mix were mixed with 1.25 mL
cDNA. The mixture was amplified for 14 cycles in a ther-
mocycler (activation, 10 min at 95⬚C; thereafter, 14 cycles
of 15 s at 95⬚C and 4 min at 60⬚C). Thereafter, samples
were diluted 1 : 2, and sample mix was prepared. Three
microliters 2#TaqMan Gene expression Master Mix (Ap-
plied Biosystems), 0.3 mL20#DNA Binding Dye Sample
Loading Reagent (Fluidigm), 0.3 mL20#EvaGreen Dye
(Biotium), and 0.9 mL1#low EDTA TE Buffer were
mixed with 1.5 mL of the sample. Assay mix was prepared
by mixing 3.6 mL 20 uM of forward and reverse primer
mix, 4 mLof2#Assay Loading Reagent (Fluidigm), and
1#low EDTA TE Buffer. During the preparation, the
chip was primed according to Fluidigm priming48 #48
protocols. The chip was thereafter loaded with 5 mLof
each sample and assay mix into the appropriate inlets, and
the run was started.
Every chip included standards, no template controls,
and controls of the non-reverse-transcribed RNA after the
initial gDNA wipeout step. Because of the space used for
those controls (eight samples), not all 382chips p382
experimental samples could be run, and some had to be
randomly excluded. The 48 available slots for primers were
used for the 14 primers, which resulted in 3–4 technical
replicates per sample and primer. The comparison of the
results from two different methods to quantify gene ex-
pression in 4-month-old pipefish suggested consistent
effects.
Analysis of q-RT-PCR Data. For all technical replicates,
the mean C
T
and the standard deviation (SD) were cal-
culated. Samples with an SD among the technical replicates
of 10.4 were removed from additional analyses to avoid
confounding the biological variation in the data due to
measurement errors. The housekeeping gene ubi served as
internal control, and the mean C
T
values of ubi were used
to quantify the relative gene expression of each immune
gene and to calculate ⫺DC
T
values. All statistical analyses
and plots are based on ⫺DDC
T
values. On exclusion of
several animals either for the complete gene expression
analysis or for only some specific genes, we ended up with
a total of 36 4-month-old individuals that were used for
the standard qPCR and 47 4-month-old individuals and
281 5-week-old juveniles that were used for the Fluidigm
gene expression analyses.
The figures that display gene expression data in 4-
month-old individuals represent the mean of the two
qPCR methods applied (figs. 1, 2).
Data Analysis and Statistics
To assess parental effects (differences in offspring im-
munity attributable to parental treatment) and offspring
effects (differences in offspring immunity attributable to
offspring treatment) on immune gene expression and di-
rect immune parameters, ANOVA were performed, in-
cluding the appropriate tests of variance homogeneity. For
the 5-week-old juveniles, nested analyses were performed
with parental and offspring treatment as fixed effect, and
the random factor “family” was nested within parental
treatment. Differences between the four parental treat-
ments were assessed with post hoc tests (Tukey HSD test).
Before all univariate tests, multivariate analysis of variance
(MANOVA) was performed to assess effects over all genes
and to protect the analysis against inflation of Type 1 error.
To assess effects per gene, single ANOVAs were calculated
using the same factors as for the MANOVA. Only effects
that were initially significant in the MANOVA were in-
terpreted. For 4-month-old offspring, only 1–2 individuals
per family per offspring treatment were available. Exper-
imental replication was thus on the family level and not
within families. Hence, family effects could not be cal-
culated. However, significant treatment effects would be
accounted conservative, because such effects can be found
only if they are stronger than family effects. Parental and
offspring treatment were taken as fixed effects for MAN-
OVA and ANOVA to assess effects on gene expression.
Gene expression data were statistically analyzed as DC
T
and only visualized in figures 1 and 2 as DDC
T
relative to
the control treatments.
Results
Direct Immune Parameters (Blood Cells and
Head Kidney Cells)
We found a marked effect of parental immune challenge
on peripheral immune activity of offspring blood immune
cells (table 1A; table A1, available online). This parental
effect was mainly driven by enhanced lymphocyte activity,
which was only induced when both parents were exposed
to heat-killed bacteria, which suggested an additive effect
5
5
5
24
-2
P
M
P&M complement c9
-2
granulocyte colony
stimulating factor
-2
toll-like receptor
-2
heat shock protein
lymphocyte antigen 75
interleukin 10
X Data
04
kinesin
4
translocator protein
-4 0 4
tumor necrosis factor
X Data
-5 0 5
Y Data
P
M
P&M
-5
Y Data
P
P&M
5
-1 -0.5 0.5 1-2 -1 012
0
∆∆C
T
five-week-old four-month-old
-5 5
5
P&M
5
coagulation factor
allograft inflammation
factor
no data
no data
complement c3
P
M
P&M
P
M
P&M
-2
P
M
P&M
-2
P
M
P&M
-2
-2
P
M
P&M
-2
-2
P
M
P&M
-2
-2
P
M
P&M
-2
-2
P
M
P&M
-2
-2
P
M
P&M
-2
-2
P
M
P&M
-2
-2
P
M
P&M
Figure 1: Gene expression profiles in juveniles (left) and 4-month-old (right) offspring on paternal (P), maternal (M), or biparental (P&M)
immune priming. Gene expression data are standardized to the housekeeping gene ubiquitin and are depicted as DDC
T
values relative to gene
expression in sham-exposed offspring. White bars indicate no significant difference, black bars indicate significant differences ( ), andP≤.05
gray bars indicate differences with a trend toward significance ( ). Detailed results of statistical analyses are found in tables A1–A5,.05 !P≤.10
available online.
24
-4 -2 2 4-4 -2 024
0
five-week-old four-month-old
no data
aV+
V-
V+
aV+
V-
V+
aV+
V-
V+
complement c9
granulocyte colony
stimulating factor
toll-like receptor
heat shock protein
lymphocyte antigen 75
interleukin 10
kinesin
translocator protein
tumor necrosis factor
coagulation factor
allograft inflammation
factor
complement c3
no data
aV+
V-
V+
aV+
V-
V+
aV+
V-
V+
aV+
V-
V+
aV+
V-
V+
aV+
V-
V+
aV+
V-
V+
aV+
V-
V+
aV+
V-
V+
ΔΔC
T
Figure 2: Gene expression profiles in 5-week-old (left) and 4-month-old (right) offspring on homologous Vibrio exposure (V⫹), phosphate-
buffered saline exposure (V⫺), or exposure to heterologous Vibrio (aV⫹). Gene expression data are depicted relative to unexposed individuals,
both standardized initially to cycle time (C
T
) values of the housekeeping gene ubiquitin (i.e., DDC
T
). All statistical analyses are based on DC
T
.
Biparental Immune Priming 000
Table 1: Evaluation of influences of parental and offspring effects on offspring immune responses
using multivariate analysis of variance (MANOVA)
MANOVA Value Fdf numerator df denominator P
A. Blood cells
Parental .472 3.193 12 117 .0006
Offspring .064 .709 4 44 .590
Parental #offspring .907 .366 12 117 .973
B. Head kidney cells
Parental .830 .719 12 117 .731
Offspring .352 3.870 4 44 .009
Parental #offspring .738 1.182 12 117 .304
C. Four-month-old gene expression
Parental .060 3.012 30 56 .0002
Offspring .796 1.512 10 280 .1575
Parental #offspring .307 .932 30 56 .574
D. Four-month-old gene expression
Parental .197 1.702 36 83 .024
Offspring 1.174 2.738 12 28 .014
Parental #offspring .298 1.176 36 83 .269
E. Five-week-old gene expression
Parental .597 4.313 33 740 !.0001
Offspring .726 2.578 33 740 !.0001
Parental #offspring .689 .977 99 1781 .547
Family (parental) .582 3.331 44 962 !.0001
Note: Models test effects on direct immune parameters in the periphery (blood cell activity; A) and in head kidney
cell activity (B), on immune gene expression in 4-month-old individuals using either standard quantitative real-time
polymerase chain reaction (q-RT-PCR; C) or q-RT-PCR on a microfluidic PCR system (Fluidigm; D), and on immune
gene expression in 5-week-old juveniles using Fluidigm (E).
of paternal and maternal priming (fig. 3). The composition
and abundance of head kidney cells as indicators of central
immune response were only affected by offspring treat-
ment (table 1B; table A2, available online). Offspring
effects are instructive to examine whether our treatment
with Vibrio presented a challenge to the immune system
as desired. When we examined single-response variables,
we found that head kidney monocyte counts, the
ratio, and the lymphocyte activitylymphocy te : monocyte
were affected by offspring treatment. Juvenile animals ex-
posed to Vibrio had fewer monocytes but more lympho-
cytes than did control animals. The lymphocyte activity
was reduced 24 h after the challenge if pipefish were ex-
posed to a Vibrio challenge (more lymphocytes in the rest-
ing stage [G
0–1
] and fewer lymphocytes in the active stage
[S⫹G
2-M
phase of the cell cycle]) compared with control
animals (table A2).
Growth
The parental treatment affected offspring size and growth
rate (ANOVA: parental effect , ; postFp3.331 Pp.027
3, 55
hoc test: maternal exposure and biparental exposure
greater than paternal exposure and unexposed). If either
both parents or only mothers were exposed to a Vibrio
challenge, offspring grew larger than the offspring of con-
trol-injected parents (fig. 3), who grew, on average, 2.5 cm
less than the former group. Thus, only maternal effects
seem to influence the offspring growth rate.
Immune Gene Expression
Gene expression measurements were a key variable, be-
cause these allowed for simultaneous assessment of im-
mune system parameters in small juveniles and larger sub-
adults. The gene expression of seven genes was affected
by parental Vibrio exposure in 5-week-old and in 4-
month-old offspring (table 1C–1E; tables A3–A5, available
online). In addition, significant effects were found for off-
spring exposure and for family effects (only tested in 5-
week-old offspring). There was no parental #offspring
interaction. In 5-week-old juveniles, parental effects im-
pacted mainly genes of the innate and rather unspecific
immune system, such as c9,tlr,kin, and tnf (for full gene
names, see table 2). In contrast, in 4-month-old pipefish,
genes with an interactive effect on the innate and adaptive
immune system were influenced, such as c3,c9,grcsf,and
il. Although gene expression in 5-week-old juveniles in-
creased 1.5–2-fold on parental Vibrio exposure, effects in
4-month-old offspring were more pronounced (2–4-fold).
000 The American Naturalist
Figure 3: Effects of parental treatment on mean length of 4-month-old individuals and on the mean proportion of active lymphocytes (lymphocyte
proliferation). Bars show means plus standard errors.
This may be attributable to the different tissues measured,
which needed to be entire animals in the 5-week-old off-
spring group, whereas only gills were used in the 4-month-
old offspring group. The only persistent upregulation in
both age groups was found for the complement system
(for c9, because c3 was only measured in the 4-month-old
group). This suggests that, on parental challenge, general
bacteria opsonization is persistently induced in offspring.
In juveniles, c9 was upregulated if mother, father, or both
were challenged, whereas in 4-month-old pipefish, c3 and
c9 expression was only induced if both parents were ex-
posed to Vibrio, which implies an interactive effect of ma-
ternal and paternal exposure on the complement system
activity. Upregulation of tlr in 5-week-old animals implies
an activated recognition of conserved pathogen-associated
patterns via the toll-like receptor pathway, which would
also include Vibrio epitopes. The opposite pattern was
found for kin, where immune gene expression was down-
regulated when both parents had been exposed to Vibrio.
Expression of tnf, which regulates the immune defense
activity, was upregulated if the father was exposed to Vib-
rio, compared with expression when both parents or only
the mother were exposed. A trend for induced expression
of hsp was shown on paternal exposure compared with no
Biparental Immune Priming 000
Table 2: Immune system components in pipefish offspring (Syngnathus typhle), their function, abbreviation, and putative classification into the adaptive or the innate
immune system
Name Immune effector Abbreviation Immune function Innate/adaptive Juvenile Mature
Blood lymphocyte activity Cellular b lympho act Periphery cellular Adaptive NM P&M
Blood lymphocytes Cellular b lympho Periphery cellular Adaptive NM NS
Head kidney lymphocyte activity Cellular hk lympho act Central cellular Adaptive NM NS
Head kidney lymphocytes Cellular hk lympho Central cellular Adaptive NM NS
Blood monocytes Cellular b mono Periphery cellular Innate NM NS
Head kidney monocytes Cellular hk mono Central cellular Innate NM NS
Allograft inflammation factor Immune gene allo Tissue rejection Innate NS P, M, P&M
Complement component c3 Immune gene c3 Complement system Innate and adaptive NM P&M
Complement component c9 Immune gene c9 Complement system Innate and adaptive P, M, P&M P&M
Granulocyte colony-stimulating factor Immune gene grcsf Granulocyte proliferation Innate NS P, M, P&M
Toll-like receptor Immune gene tlr Self-nonself difference
over PRR/PAMPs
Innate P, M, P&M NS
Heat shock protein 1 Immune gene hsp General stress Innate P, M, P&M NS
Lymphocyte antigen 75 Immune gene ig ant Tissue rejection Adaptive NS NS
Interleukin 10 Immune gene il Regulation of macrophage
activity
Innate and adaptive NS P, M, P&M
Kinesin Immune gene kin Intracellular transport Innate P&M NS
Coagulation factor II receptor-like 1 Immune gene cf Blood clotting Innate NM P
Translocator protein Immune gene tspo Inflammatory response Innate NS P&M
Tumor necrosis factor protein Immune gene tnf Regulation of immune
defense
Innate and adaptive P NS
B-cell receptor-associated protein Immune gene bcell Regulation of B cell
activity
Adaptive NS NS
Natural-resistance-associated macrophage protein Immune gene nramp Macrophage activation Innate NS NS
Length Life history length Development Innate and adaptive NM M
Note: Significance and direction of effect evaluated in juveniles and in mature individuals. M peffect mediated if only maternal parent was immunochallenged; NM pnot measured; NS pnot
significant; P peffect mediated if only paternal parent was immunochallenged; PAMP ppathogen-associated molecular patterns; PRR ppattern recognition receptors; P&M peffect mediated if both
parents were immunochallenged.
000 The American Naturalist
parental exposure. Furthermore, in 4-month-old pipefish,
grcsf expression was increased if parents were not exposed
to Vibrio, which suggests that granulocyte production is
less important after parental challenge, independent of
whether mother, father, or both were exposed to Vibrio.
Upregulation of il occurred if parents stayed unexposed
or if only the mother was exposed, which indicates that,
if paternal components of the immune system were lack-
ing, a general inflammation response was of more im-
portance. Several additional immune genes revealed effects
that were marginally significant and suggestive of parental
effects ( ). The gene cf, which is linked to bloodP!.10
clotting, showed a trend of higher expression in offspring
with paternal Vibrio exposure than in the offspring of
unexposed parents. The gene tspo, which is responsible for
the regulation of the steroid hormone synthesis and in-
duces controlled cell death (apoptosis), showed a trend of
upregulation if both parents were exposed to Vibrio.
Offspring effects on gene expression were abundant (ta-
ble 1; tables A3–A5; fig. 2) and were significant for allo
(regulation of tissue rejection), bcell (regulation of lym-
phocyte activity), c9 (bacteria opsonisation), tlr (self non-
self reconition), and tspo (inflammation responses) in 5-
week-old pipefish and for il (regulation of macrophage
activity), allo (regulation of tissue rejection), and c9 (com-
plement system) and trended to significance for bcell (lym-
phocyte activity regulation) and tspo (inflammation re-
sponse) in 4-month-old individuals. In 4-month-old
pipefish, gene expression was upregulated on Vibrio ex-
posure, whereas in 5-week-old offspring, allo,bcell,and
tlr expression was downregulated if offspring remained
naive, whereas c9 expression was induced if offspring were
either exposed to homologous or heterologous Vibrio.
Moreover, tspo was downregulated if offspring were ex-
posed to PBS. Family effects were strong overall and were
found for all genes except for c9,il, and tspo in juveniles.
Discussion
In a sex-role-reversed animal with male pregnancy, we
simulated the abundance of a potentially harmful cocktail
of bacteria epitopes in only the mother, only the father,
or both parents. We then experimentally examined the
potential parental effects on both innate and acquired off-
spring immunity. We further evaluated the persistence of
the parental effects from juveniles to 4-month-old fish and
measured offspring length after 4 months. Contrary to our
initial hypothesis, which predicted mostly male but not
female immune transfer, we identified complex additive
effects on offspring immunity from both fathers and moth-
ers in 5-week-old juveniles and in 4-month-old animals.
Overall, we found three categories of responses: (i) The
same extent of regulation when father, mother, or both
have been exposed to a challenge with heat-killed Vibrio.
In this case, chances are increased (effectively doubled)
that the crucial information on prevailing parasites in the
parental generation is collected and transferred to the off-
spring. In 5-week-old offspring, this was the case for tlr
and c9 and was a trend for hsp. All of these genes are
involved in the unspecific innate humoral inflammation
response. In 4-month-old individuals, we found this pat-
tern for grcsf and il and found a trend for allo. These three
immune genes regulate the activity of the cellular innate
immune system (macrophage and granulocyte prolifera-
tion). (ii) Effects on offspring immunity were detectable
only if both parents were exposed to heat-killed Vibrio,
and an enhanced effect was detectable only through an
interaction of maternal and paternal immune status. This
is difficult to explain in evolutionary terms, because the
required double exposure of both parents to a pathogen
decreases the likelihood that parental influence takes place.
The requirement that both parents be exposed for an up-
regulation in offspring may reflect a dose dependence of
offspring immune defense. Hence, exposure of parents
with higher dosages or applying live bacteria could have
resulted in more intense effects on both the paternal and
maternal sides. In 5-week-old offspring, we detected pre-
cisely this pattern for kin, which is responsible for intra-
cellular transport of molecules; in 4-month-old animals,
we detected this pattern for the direct assessment of blood
lymphocyte activity, c9, and c3 and detected a trend for
tspo. The complement system is responsible for bacteria
opsonization and the activity of the lymphocytes, whereas
tspo mediates an inflammatory response. (iii) Immune de-
fense is only affected after paternal exposure. We identified
this for the regulation of tnf in juveniles and detected a
trend for cf in 4-month-old individuals. Upregulation of
the gene tnf prevents an overreaction of the immune sys-
tem. On maternal challenge, more intense activity of the
immune system seems necessary to fight prevailing par-
asites and pathogens.
Interestingly, parental effects that were associated with
maternal exposure alone were not identified for a single
immune parameter. However, maternal pathogen chal-
lenge significantly influenced offspring growth, as reflected
by larger offspring size after 4 months. This effect poten-
tially arises because offspring of immunochallenged moth-
ers have an improved constitutive immune defense and
hence need fewer resources for immune defense upregu-
lation (Sheldon and Verhulst 1996). Alternatively or in
addition, immunochallenged mothers may invest more
into the current clutch because of an increased risk to not
survive to the next reproductive event as a result of the
high costs of immune defense (Clutton-Brock 1984). This
terminal investment seems to be a maternally associated
trait even in a sex-role-reversed species, or it may poten-
Biparental Immune Priming 000
tially just be of less importance in pipefish males than in
males of other species because of their more efficient im-
mune defense (Roth et al. 2011).
By exposing juveniles to the same (homologous) Vibrio
phylotypes as the parental generation or to different (het-
erologous) phylotypes, we investigated potential specificity
of the parental effects on immunity versus a rather general
upregulation of innate immune defense. However, no dif-
ferences in the degree of immune gene induction between
the homologous and heterologous combinations could be
detected. Our specific study design was intended to mimic
the natural situation in applying a bacterial cocktail as a
second challenge, which precludes any conclusion about
the specificity of paternal and maternal immune priming.
This is because we cannot exclude that the Vibrio phylotype
applied in the offspring generation I2K3 harbors epitopes
present in the strains of the parental cocktail.
In 5-week-old juveniles, only genes of the innate im-
mune system were induced, whereas in 4-month-old in-
dividuals, genes with interactive effects on adaptive and
innate immune defense were also affected (in particular,
c3,c9,grcsf, and il). This suggests that fish of several
months of age are already capable of a limited usage of
adaptive immune responses (Magnadottir et al. 2005). This
difference between 5-week-old and 4-month-old offspring
is further supported by similar upregulation in both Vib-
rio-challenged and PBS-challenged individuals for several
immune genes in 5-week-old offspring (allo,bcell,tlr,c9,
and tspo). Note that 4-month-old animals were exposed
to Vibrio alone or were left naive. However, earlier studies
confirmed that sham injection with PBS will not result in
immune gene upregulation in pipefish of that age (Birrer
et al. 2012).
The parental animals were wild-caught around the is-
land of Gotland. The Vibrio phylotypes used during this
study were isolated from different pipefish populations and
were thus allopatric and most likely novel to the immune
system of the Gotland pipefish tested here (Roth et al.
2012). Still, the generally high degree of cross-reactivity in
fish immune defense (Magnadottir et al. 2005) may con-
found the desired experimental parental exposure with
earlier exposure of parents to other Vibrio phylotypes in
the field. However, this would make our results rather
more conservative, because effects were even visible against
the background of other possible Vibrio phylotypes that
the parental pipefish were previously exposed to in the
wild.
To date, parental effects have been assumed to be of
short duration (Picchietti et al. 2001). Our data suggest
that beneficial influences on the immunocompetence of
offspring do not cease after a few weeks but persist in 4-
month-old individuals. However, the genes that are im-
portant during the ontogeny process change as resembled
in different patterns of up- or downregulation in juveniles
and 4-month-old pipefish. This suggests a specific pattern
of immune regulation, rather than a generally higher in-
vestment into clutches as a result of perceived parental
infection status. The most consistent parental effect was
observed for genes associated with the complement system,
in which effects present in juveniles persisted at the 4-
month-old stage, which was not the case for most other
genes. The complement system is responsible for bacteria
opsonization and clearance and provides a functional link
between the innate and the adaptive immune response by
activating antibody proliferation. This second function is
consistent with the induced lymphocyte activity in off-
spring on parental exposure, because lymphocytes are re-
sponsible for antibody production. Interestingly, in 4-
month-old pipefish, components of the complement
system and lymphocyte proliferation were only induced if
both father and mother were exposed to a challenge with
heat-killed Vibrio (fig. 1). This suggests an intimate in-
terplay between maternal and paternal resources.
Possible mechanisms of transgenerational immune
priming on the maternal side are probably comparable to
those in conventional-sex-role species (e.g., deposition of
antibodies into the eggs and transfer of components of the
complement system), but the mechanisms of paternal im-
mune priming remain unclear. In conventional vertebrate
sex role systems, transfer of immunity from fathers to
offspring has never been found. This makes it likely that
the evolution of paternal transgenerational immune prim-
ing was associated with the evolution of sex-role reversal.
The placenta-like structure in the pipefish Syngnathus ty-
phle thus makes this species a prime candidate for studying
the mechanistic basis of paternal immune priming.
Fish larvae hatch in an ocean environment thriving with
bacterial pathogens (Belkin and Colwell 2006). At the same
time, their immunological capacity is still severely limited
(Mulero et al. 2007), because the maturation of the ac-
quired immune defense as a response to interactions with
the present pathogens takes several months (Magnadottir
et al. 2005). As an alternative defense, we identified here
the constitutive upregulation of innate immunity across
generations under a scenario in which potentially harmful
bacteria are abundant. In a species with male pregnancy
and an enclosed brood pouch, offspring receive the in-
formation whether or not abundant bacteria are present
predominantly via males. Such transgenerational influ-
ences on offspring immunity are probably adaptive, be-
cause innate immunity is costly and should not be induced
in environments where pathogens are scarce. Male pipefish
are thus not only in control of which female’s offspring
will be born (Paczolt and Jones 2010) but also provide
crucial immunological information on pathogen abun-
dance to offspring. This may represent another fitness ben-
000 The American Naturalist
efit leading to the evolution of male pregnancy in seahorses
and pipefish.
Acknowledgments
We thank S. Birrer, R. Ho¨glund, S. Landis, L. Miersch, G.
Rosenqvist, P. Schubert, and J. Sundin for support in ex-
perimental and field work. We thank A. Minder and T.
Torossi and the Genetic Diversity Center at the Eidge-
no¨ ssische Technische Hochschule Zurich for access to the
Fluidigm and their help in running the assays and F. Brun-
ner and D. Refardt for support during gene expression
measurements. Comments and input by B. Sadd improved
the manuscript considerably. O.R. thanks S.B.B. and D.B.
for silent support during the writing of this article. Thanks
to two anonymous reviewers for their helpful comments.
This study was financed by a grant from the Volkswagen
Foundation “Evolution” program to O.R. The experiment
was performed according to current national legislation of
the Ministerium fu¨ r Landwirtschaft, Umweltund la¨ndliche
Ra¨ume des Landes Schleswig-Holstein (project: “Effects of
global change on the immunological interaction of pipefish
and their natural bacterial communities”). All data have
been deposited at Dryad under the doi:10.5061/
dryad.r6624.
Literature Cited
Adamo, S. A. 1999. Evidence for adaptive changes in egg laying in
crickets exposed to bacteria and parasites. Animal Behavior 57:
117–124.
Altizer, S., D. Harvell, and E. Friedle. 2003. Rapid evolutionary dy-
namics and disease threats to biodiversity. Trends in Ecology &
Evolution 18:589–596.
Arnqvist, G., and L. Rowe. 2005. Sexual conflict. PrincetonUniversity
Press, Princeton, NJ.
Bateman, A. J. 1948. Intra-sexual selection in Drosophila. Heredity
2:349–368.
Belkin, S., and R. R. Colwell. 2006. Oceans and health: pathogens in
the marine environment. Springer, Berlin.
Birrer, S. C., T. B. H. Reusch, and O. Roth. 2012. Salinity change
impairs pipefish immune defense. Fish and Shellfish Immunololgy
32:1238–1248. doi:10.1016/j.fsi.2012.08.028.
Bly, J. E., A. S. Grimm, and I. G. Morris. 1986. Transfer of passive
immunity from mother to young in a teleost fish: haemagglu-
tinating activity in the serum and eggs of plaice Pleuronectes pla-
tessa L. Comparative Biochemistry Physiology A 84:309–313.
Bonneaud, C., J. Mazuc, O. Chastel, H. Westerdahl, and G. Sorci.
2004. Terminal investment induced by immune challenge and fit-
ness traits associated with major histocompatibility complex in the
hourse sparrow. Evolution 58:2823–2830.
Boots, M., and R. G. Bowers. 2004. The role of ecological feedbacks
in the evolution of host defence: what does theory tell us? Phil-
osophical Transactions of the Royal Society B: Biological Sciences
364:27–36.
Boulinier, T., and V. Staszweski. 2008. Maternal transfer of antibodies:
raising immuno-ecology issues. Trends in Ecology & Evolution 23:
282–288.
Clutton-Brock, T. H. 1984. Reproductive effort and terminal in-
vestment in iteroparous animals. American Naturalist 123:212–
229.
Dzyuba, B., K. J. W. Van Look, A. Cliffe, H. J. Koldewey, and W. V.
Holt. 2006. Effect of parental age and associated size on fecundity,
growth and survival in the yellow seahorse Hippocampous kuda.
Journal of Experimental Biology 209:3055–3061.
Eizaguirre, C., T. L. Lenz, M. Kalbe, and M. Milinski. 2012. Rapid
and adaptive evolution of MHC genes under parasite selection in
experimental vertebrate populations. Nature Communications 3:
621. doi:10.1038/ncomms1632.
Ellingsen, T., C. Strand, E. Monsen, J. Bogwald, and R. A. Dalmo.
2005. The ontogeny of the comlement component C3 in the spot-
ted wolffish (Anarhichas minor Olafsen). Fish and Shellfish Im-
munology 18:351–358.
Fuda, H., A. Hara, F. Yamazaki, and K. Kobayashi. 1992. A peculiar
immunoglobulin M (IgM) identified in eggs of chum salmon (On-
corhynchus keta). Developmental and Comparative Immunology
1993:241–248.
Gasparini, J., K. D. McCoy, T. Tveraa, and T. Boulinier. 2002. Related
concentrations of specific immunoglobulins against the Lyme dis-
ease agent Borrelia burgdorferi sensu lato in eggs, young and adults
of the kittiwake (Rissa tridactyla). Ecology Letters 5:519–524.
Glezen, P. 2003. Effect of maternal antibodies on the infant immune
response. Vaccine 21:3389–3392.
Grindstaff, J. L., E. D. Brodie III, and E. D. Ketterson. 2003. Immune
function accross generations: integrating mechanism and evolu-
tionary process in maternal antibody transmission. Proceedings of
the Royal Society B: Biological Sciences 270:2309–2319.
Haig, D. 2000. The kinship theory of genomic imprinting. Annual
Review of Ecology and Systematics 31:9–32.
Hamilton, W. D. 1980. Sex versus non-sex versus parasite. Oikos 35:
282–290.
Hamilton, W. D., R. Axelrod, and R. Tanese. 1990. Sexual repro-
duction as an adaptation to resist parasites (a review). Proceedings
of the National Academy of Sciences of the USA 87:3566–3573.
Hanif, A., V. Bakopoulos, and G. J. Dimitriadis. 2004. Maternal trans-
fer of humoral specific and non-specific immune parameters to
sea bream (Sparus aurata) larvae. Fish Shellfish Immunol 17:411–
435.
Hanssen, S. A. 2006. Costs of an immune challenge and terminal
investment in a long-lived bird. Ecology 87:2440–2446.
Harlin-Cognato, A., E. A. Hoffman, and A. G. Jones. 2006. Gene
cooption without duplication during the evolution of a male-preg-
nancy gene in pipefish. Proceedings of the National Academy of
Sciences of the USA 103:19407–19412.
Hasselquist, D., and J.-A. Nilsson. 2009. Maternal transfer of anti-
bodies in vertebrates: trans-generational effects on offspring im-
munity. Philosophical Transactions of the Royal Society B: Bio-
logical Sciences 364:51–60.
Jones, A. G., G. Rosenqvist, A. Berglund, S. J. Arnold, and J. C. Avise.
2000. The Bateman gradient and the cause of sexual selection in
a sex-role-reversed pipefish. Proceedings of the Royal Society B:
Biological Sciences 267:677–680.
Jones, A. G., G. Rosenqvist, A. Berglund, and J. C. Avise. 2005. The
Biparental Immune Priming 000
measurement of sexual selection using Bateman’s principles: an
experimental test in the sex-role-reversed pipefish Syngnathus ty-
phle. Integrative Comparative Biology 45:874–884.
Landis, S. H., M. Kalbe, T. B. H. Reusch, and O. Roth. 2012. Con-
sistent pattern of local adaptation during an experimental heat
wave in a pipefish-trematode host-parasite system. PLoS ONE 7:
e30658.
Lemke, H., A. Coutinho, and H. Lange. 2004. Lamarckianinheritance
by somatically acquired maternal IgG phenotypes. Trends in Im-
munology 25:180–186.
Magnadottir, B., S. Lange, S. Gudmundsdottir, J. Bogwald, and R.
A. Dalmo. 2005. Ontogeny of humoral immune parameters in
fish. Fish and Shellfish Immunology 19:773–783.
Matsunaga, T., and A. Rahman. 1998. What brought the adaptive
immune system to vertebrates? the jaw hypothesis and the sea-
horse. Immunology Review 166:177–186.
Moret, Y. 2006. “Trans-generational immune priming”: specific en-
hancement of the antimicrobial immune response in the meal-
worm beetle, Tenebrio molitor. Proceedings of the Royal Society
B: Biological Sciences 273:1399–1405.
Mousseau, T. A., and C. W. Fox. 1998. The adaptive significance of
maternal effects. Trends in Ecology & Evolution 13:403–407.
Mulero, I., A. Garcia-Ayala, J. Meseguer, and V. Mulero. 2007. Ma-
ternal transfer of immunity and ontogeny of autologous immu-
nocompetence of fish: a minireview. Aquaculture 268:244–250.
Paczolt, K. A., and A. G. Jones. 2010. Post-copulatory sexual selection
and sexual conflict in the evolution of male pregnancy. Nature
464:401–404.
Picchietti, S., G. Scapigliati, M. Fanelli, F. Barbato, S. Canese, L.
Mastrolia, M. Mazzini, et al. 2001. Sex-related variations of serum
immunoglobulins during reproduction in gilthead sea bream and
evidence for a transfer from the female to the eggs. Journal of
Fish Biology 59:1503–1511.
Poulin, R., and F. Thomas. 2008. Epigenetic effects of infection on
the phenotype of host offspring: parasites reaching across host
generations. Oikos 117:331–335.
Ra˚berg, L., J.-A. Nilsson, P. Ilmonen, M. Stjernmann, and D. Has-
selquist. 2000. The cost of an immune response: vaccination re-
duces parental effort. Ecology Letters 3:382–386.
Ra˚berg, L., M. Vestberg, D. Hasselquist, R. Holmdahl, E. Svensson,
and J.-A. Nilsson. 2002. Basal metabolic rate and the evolution of
the adaptive immune system. Proceedings of the Royal Society B:
Biological Sciences 269:818–821.
Reaney, L. T., and R. J. Knell. 2010. Immune activation but not male
quality affects female current reproductive investment in a dung
beetle. Behavioral Ecology 21:1367–1372.
Reid, J. M., P. Arcese, L. F. Keller, and D. Hasselquist. 2006. Long-
term maternal effects on offspring immune response in song spar-
rows Melospiza melodia. Biology Letters 2:573–576.
Ripley, J. L., and C. M. Foran. 2009. Direct evidence for embryonic
uptake of paternally-derived nutrients in two pipefishes (Syng-
nathidae: Syngnathus spp.). Journal of Comparative Physiology B
179:325–333.
Roth, O., G. Joop, H. Eggert, J. Hilbert, J. Daniel, P. Schmid-Hempel,
and J. Kurtz. 2010. Paternally derived immune priming for off-
spring in the red flour beetle, Tribolium castaneum. Journal of
Animal Ecology 79:403–413.
Roth, O., I. Keller, S. H. Landis, W. Salzburger, and T. B. H. Reusch.
2012. Hosts are ahead in a marine host-parasite coevolutionary
arms race: innate immune system adaptation in pipefish Syng-
nathus typhle against Vibrio phylotypes. Evolution 66:2528–2539.
Roth, O., J. P. Scharsack, I. Keller, and T. B. H. Reusch. 2011. Bate-
man´ s principle and immunity in a sex-role reversedpipefish. Jour-
nal of Evolutionary Biology 24:1410–1420.
Rozen, S., and H. Skaletzky. 1997. PRIMER3. http://www.genome
.wi.mit.edu/genome/other/primer3.html.
Sadd, B. M., L. Holman, S. A. O. Armitage, F. Lock, R. Marland,
and M. T. Siva-Jothy. 2006. Modulation of sexual signalling by
immune challenged male mealworm beetles (Tenebrio molitor,L.):
evidence for terminal investment and dishonesty. Journal of Evo-
lutionary Biology 19:321–325.
Sadd, B. M., Y. Kleinlogel, R. Schmid-Hempel, and P. Schmid-
Hempel. 2005. Trans-generational immune priming in a social
insect. Biology Letters 1:386–388.
Sadd, B. M., and P. Schmid-Hempel. 2007. Facultative but persistent
transgenerational immunity via the mother’s eggs in bumblebees.
Current Biology 17:R1046–R1047.
Sheldon, B. C., and S. Verhulst. 1996. Ecological immunology: costly
parasite defences and trade-offs in evolutionary ecology. Trends
in Ecology & Evolution 11:317–321.
Shoemaker, K. L., N. M. Parsons, and S. A. Adamo. 2006. Egg-laying
behaviour following infection in the cricket Gryllus texensis. Ca-
nadian Journal of Zoology 84:412–418.
Sto¨lting, K. K., and A. B. Wilson. 2007. Male pregnancy in seahorses
and pipefish: beyond the mammalian model. Bioessays 29:884–
896.
Van Valen, L. 1973. A new evolutionary law. Evolutionary Theory 1:
1–30.
Wassarmann, P. M., L. Jovine, and E. S. Litscher. 2001. A profile of
fertilization in mammals. Nature Cell Biology 3:E59–E64.
Wilson, A., A. Vincent, I. Ahnjesjo¨, and A. Meyer. 2001. Male preg-
nancy in seahorses and pipefishes (Family Syngnathidae): rapid
diversification of paternal brood pouch morphology inferred from
a molecular phylogeny. Journal of Heredity 92:159–166.
Zanchi, C., J.-P. Troussard, G. Martinaud, J. Moreau, and Y. Moret.
2011. Differential expression and costs between maternally and
paternally derived immune priming for offspring in an insect.
Journal of Animal Ecology 80:1174–1183.
Associate Editor: Elizabeth Adkins-Regan
Editor: Troy Day
