Bird-Parasite Interactions - DiVA portal165972/FULLTEXT01.pdf · tests of the Hamilton-Zuk hypothesis, have also received criticism due to the somewhat opportunistic approaches ecologists

Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Science and Technology 557

_____________________________ _____________________________

Bird-Parasite Interactions

Using Sindbis Virus as a Model System

BY

KARIN LINDSTRÖM

ACTA UNIVERSITATIS UPSALIENSISUPPSALA 2000

Dissertation for the Degree of Doctor of Philosophy in Population Biology presented atUppsala University in 2000

Abstract

Lindström, K.M., 2000. Bird-parasite interactions. Using Sindbis virus as a model system.Acta Universitatis Upsaliensis. Comprehensive Summaries of Uppsala Dissertations fromthe Faculty of Science and Technology 557. 40 pp. Uppsala. ISBN 91-554-4773-2.

This thesis focuses on the evolutionary interactions between birds and a parasite, themosquito-borne Sindbis virus (Togaviridae, Alphavirus). In conclusion, the results showthat the Sindbis virus is widespread among birds, and that the fitness of infected hosts maybe reduced by the virus. Furthermore, virus clearance ability was revealed by maleplumage traits, and viraemia was related to hormonal- and social status.

The distribution of Sindbis virus infections among passerine birds was examined infive areas in Sweden. Almost all species tested were infected, and three species of thrusheswere identified as the main hosts. In a series of experimental infections, greenfinches(Carduelis chloris) kept in aviaries were used as hosts. First, the behavioural consequencesof an infection were investigated. During the infection, birds tended to reduce theirspontaneous locomotion activity, and when escaping from a simulated predator attack,infected birds had reduced take-off speed. Furthermore, when comparing virus clearancerate between male greenfinches, I found that males with large yellow tail ornaments hadfaster virus clearance rates as compared to those with smaller ornaments. Thus, male virusclearance ability was honestly revealed by the size of an ornament. Moreover, males withexperimentally elevated testosterone levels experienced a delayed, but not increasedviraemia as compared to controls. When the relationship between male social rank andviraemia was examined, I found no evidence that high-ranked males suffered reduced rankduring the infection. Nevertheless, viraemia patterns of males were related to their socialrank, so that low-ranked birds had a delayed viraemia as compared to high-ranked birds.

Key words: Carduelis chloris, Sindbis virus, host-parasite interaction, ornament, socialdominance, testosterone, sexual selection.

Karin M. Lindström, Dep. of Population Biology, Evolutionary Biology Centre, UppsalaUniversity, Norbyvägen 18D, SE-752 36 Uppsala, Sweden ( [email protected])

© Karin M. Lindström 2000

ISSN 1104-232XISBN 91-554-4773-2

Printed in Sweden by University Printers, Uppsala 2000

You can’t always get what you want, honey

you can’t always get what you want.

You can’t always get what you want,

but if you try, sometime you just might find

you get what you need!

Jagger/Richards

To the memory of my mother

This thesis is based on the following papers, which will be referred to in the text bytheir Roman numerals I-V.

I. Lundström, J.O., Lindström, K.M., Olsen, B., Dufva, R. & Krakower, D.S.Prevalence of Sindbis virus antibodies among Swedish passerines indicatethat thrushes are the main amplification hosts. Submitted manuscript.

II. Lindström, K.M., van der veen, I., Legault, B-A. & Lundström J.O.Behavioural alterations of Greenfinches (Carduelis chloris) during a virusinfection: are avian virus infections costly? Submitted manuscript.

III. Lindström, K.M. & Lundström, J.O. 2000. Male greenfinches (Carduelischloris) with brighter ornaments have higher virus infection clearance rate.Behav. Ecol. Sociobiol. 48:44-51.

IV. Lindström, K.M., Krakower, D. Lundström J.O. & Silverin, B. The effects ofTestosterone on a viral infection in greenfinches (Carduelis chloris): anexperimental test of the immunocompetence handicap hypothesis. Submittedmanuscript.

V. Lindström, K.M. Social rank and virus resistance in greenfinches. Submittedmanuscript.

Paper III is reproduced with permission from the publisher.

The order of the authors reflects their involvement in the papers. I havepersonally written and performed all analyses of papers II, III, IV and V. Co-authors in papers II, III and IV collected data, discussed ideas, analyses andcommented on the text. In paper I, I collected part of the field data, performedmost of the laboratory tests, discussed the analyses and commented on the text.

CONTENTS

INTRODUCTION ………………………………………………………..….……... 7

Birds and parasites in nature …………………………………………………. 7Parasite-mediated sexual selection …………………………………………... 8The immunocompetence-handicap …………………………………………... 9Life-history trade-offs ……………………………………………………….. 10

METHODS ………………………………………………………………………... 11Natural history of the greenfinch …………………………………………….. 11The Sindbis virus …………………………………………….………………. 13Greenfinch measurements …………………………….……………………… 15Virus infection and assays ……………………………………………………... 16Statistics and data analyses …………………………………………………... 17

RESULTS AND DISCUSSION ………………………………………………………. 18Distribution of Sindbis virus infections in the field …………………………..18Costs of the virus infection ………………………………………………..…. 19Sexual selection …………………………………………………………...…. 21Testosterone and virus clearance ……………………………………………… 23Social status and virus clearance……………………………………………….. 24

CONCLUSIONS AND PROSPECTS………………………………………...………... 26

ACKNOWLEDGEMENTS …………………………………………………….……. 27

REFERENCES …………………………………………………………………….. 28

SWEDISH SUMMARY – SVENSK SAMMANFATTNING …………………………….. 37

Bird-parasite interactions 7

INTRODUCTION

Most animals share their environments with a rich fauna of parasites. The need to survive

and reproduce successfully in these environments has therefore forced birds and other

animals to evolve a wide array of defences against parasites (Clayton and Moore 1997).

Host-parasite interactions have recently become a central concept in studies of

evolutionary ecology due to their impact on sexual selection (Hamilton and Zuk 1982,

Andersson 1994). Also, by integrating life-history theory and sexual selection, an

interesting new area of research has developed. With this approach it may become possible

to understand which factors may limit an individuals investment in immunity (Sheldon and

Verhulst 1996, Norris and Evans 2000). Although parasites increasingly have become the

focus of much scientific interest, many aspects of the interactions between hosts and

parasites are still poorly understood. The aim of this thesis has been to investigate one

host-parasite relationship in more detail, by testing some of the ideas put forth to

understand the complexity of the evolutionary interactions between hosts and parasites.

Birds and parasites in nature

A parasite can be defined as an organism that lives in, or on another living organism,

obtaining from it part or all of its organic nutrients, commonly exhibiting some degree of

structural modification, and causing some degree of damage to its host (Price 1980).

Viruses are small and simple organisms with an obligate parasitic lifestyle, that utilise host

cells for their own reproduction. Parasites in general, and viruses in particular, often have

short life-cycles. This gives them a capacity to change in genetic composition between

generations, and hosts have to be able to respond to these changes. On an evolutionary

time-scale, the interactions between hosts and parasites may therefore be viewed as a

never-ending arms race (Morse 1994, Domingo and Holland 1994, Ewald 1994). The

composition of the parasitic fauna may also change over time. Birds in nature are subject

to infections by many different parasites, some of which appear and disappear in cycles

(Clayton and Moore 1997). It is often assumed that parasitic infections are costly for the

8 Lindström

hosts, but in practice, these costs can be difficult to measure. Parasitic infections may

affect host fitness by causing disease symptoms, which may lead to death of the host. Mild

infections may cause morbidity that reduces the host’s ability to avoid predators or even

other parasitic infections. Parasites may also have negative impact on host reproduction or

competitive ability (Møller et al. 1990, Hudson and Dobson 1997). Paper I describes the

distribution of Sindbis virus infections among avian hosts in nature, and paper II

investigates if the Sindbis virus can impair infected birds ability to escape from predators.

Parasite-mediated sexual selection

In order to explain the evolution of extravagant male ornaments Darwin (1871) proposed

theory of sexual selection. Darwin reasoned that males with large ornaments could be

favoured by sexual selection if females preferred to mate with such males. He, however,

did not provide an explanation as to how females could benefit from such preferences. In

order to resolve the benefits of female choice, a series of handicap-models of sexual

selection were developed (Zahavi 1975, Grafen 1990). These models proposed that male

ornaments were costly to maintain, and therefore only high-quality males should be able to

afford these costs. Following this argument, if male quality and ornament size are

positively related, and heritable, females would benefit from their choice through increased

survival of their offspring. Hamilton and Zuk (1982) developed the handicap-models

further, and suggested that it was the ability to resist the currently prevailing fauna of

parasites that was the quality revealed by male ornaments. The hypothesis predicts that

within species there should be a negative correlation between ornament expression and

parasite load. It was mainly this hypothesis proposed by Hamilton and Zuk (1982), that

placed parasites in a central position in studies of sexual selection.

To date, a large number of studies on birds and other animals have demonstrated

that within species, male ornaments can be used as indicators of parasite load, or

immunocompetence, i. e. the ability of an individual’s immune system to resist and control

infections (Møller 1990, Andersson 1994, Hamilton and Poulin 1997). Although the

Hamilton-Zuk hypothesis has gained much empirical support, their idea has been

controversial. It has been argued that the hypothesis is impossible to falsify, as several


factors may explain why the predicted association was not found (Read 1990). It can for

example be argued that the cruzial parasite taxa were not included in the analyses. Also,

for most studies, the underlying assumption that parasite resistance ability is inherited by

the offspring has not been tested, and alternative hypothesis have often not been excluded

(Read 1990). Although females may use male ornament to select parasite free males, they

may also do so to obtain non-inherited direct benefits. These direct benefits could for

example include an increased feeding effort by a male in good condition (Hoelzer 1989).

Alternatively, females may choose highly ornamented males to gain indirect benefits

through the production of highly attractive sons (Fisher 1930, Jones et al. 1998). Previous

tests of the Hamilton-Zuk hypothesis, have also received criticism due to the somewhat

opportunistic approaches ecologists have adopted to evaluate immunocompetence (Siva-

Jothy 1995). Antibody response to novel or familiar antigens, level of serum globulin and

the size of immune organs have all been used to estimate variation in immunocompetence.

For several reasons, these measurements may not accurately reflect an individual’s

immunocompetence (Sheldon and Verhulst 1996, Penn and Potts 1998). So far, few studies

have attempted to test if expression of male ornament can be related to their ability to

control an experimental infection (Møller 1990, Hamilton and Poulin 1997). Also,

although viruses are common parasite for birds (Nuttall 1997), no study have related male

ornament expression to his ability to control a virus infection. This relationship is

examined in paper III.

The immunocompetence-handicap

The relationship between ornament expression and parasite resistance is physiologically

poorly understood. Several ideas have been put forward to explain how such a relationship

could evolve and be maintained. Carotenoid pigments have been proposed to function as

the physiological link between immunity and ornament expression (Olsen and Owens

1998). According to another influential model, the immunocompetence-hypothesis, the

relationship between male quality and ornament size could be guaranteed if the expression

of male ornaments required the presence of an immunosuppressive hormone, such as

testosterone (Folstad and Karter 1992). Although it was initially believed that testosterone

10 Lindström

was costly in terms of immunosuppression (Grossman 1985), several recent studies on

birds have now led behavioural ecologists to question this assumption (Hillgarth and

Wingfield 1997, Hasselqvist et al. 1999, Evans et al. 2000). It has also been argued that

testosterone does not regulate the expression of male plumage ornaments (Owens and

Short 1995). Although testosterone may have no direct influence of ornament expression, a

positive relationship between ornament expression and testosterone levels could be

mediated by social dominance (Poiani et al. 2000). It is also possible that interactions

between males per se can induce costs associated with male ornament expression

(Johnstone and Norris 1993). Following the same line of reasoning, several authors have

proposed that stress hormones like corticosterone could be the key relating ornament

expression to immune functions (Møller 1995, Hillgarth and Wingfield 1997, Evans et al.

2000). In paper IV, I investigated if increased levels of testosterone would suppress

immunity to the Sindbis virus infection.

Life-history trade-offs

Although investments in immunity are vital for survival, other investments are also

important. When resources are limited, individuals are unable to optimise all traits

simultaneously, so that each investment must be traded off against others. Within life-

history theory, several such trade-offs have been identified (Stearns 1992). The limiting

resources could be both energetic and structural components. The fact that mounting of

immune response can be costly has been demonstated for mammals (Lochmiller and

Deerenberg 2000), but not for birds (Svensson et al. 1998). To understand why individuals

may invest differently in immunity, it has been suggested that immunity should be studied

from a life-history perspective (Sheldon and Verhulst 1996). So far, two important trade-

offs between immunity and other life-history traits have been identified. In birds,

investments in sexual ornaments and reproduction have been shown to limit the resources

available for parasite defence (Gustafsson et al. 1994, Nordling et al. 1998, Verhulst et al.

1999). The relationship between social status and immunity could represent another

potential trade-off for the individual (Barnard et al. 1994, Zuk and Johnsen 2000). If

maintenance of a high social status is costly, this may have negative effects on immunity.


The relationship between parasite resistance ability and male social status is investigated in

paper V.

METHODS

The field study (I) was based on data collected between 1990-1998 from five study areas

in Sweden. All other data presented in the thesis were collected between 1996 and 2000.

For all experimental infections (II-V), I used greenfinches (Carduelis chloris) captured in

the Uppsala area (59°50’N; 17°50’E). The birds were caught in the autumn or winter at

feeding sites located in the Uppsala Botanical garden, at the Fyris River near the Uppsala

sewage treatment plant, or at Bäcklösa with mist-nets and ground-traps. After capture, all

birds were transferred to an aviary and kept in cages. Within a few days after capture, all

birds were ringed, sexed and aged into two age classes i.e.; yearlings and older (Svensson

1992). During the experiments birds were kept isolated (II-IV) or in pairs (V) in cages (0.5

x 0.5 x 0.5 m). Birds were supplied with food (sunflower-seeds supplemented with hemp,

millet, and vitamins) and water every second day and all cages contained bowls and sitting

rods. The aviary was maintained on a natural daylength cycle.

Natural history of the greenfinch

Greenfinches are medium sized (ca 30g) passerines that occur in most parts of Europe. It is

a sexually dimorphic species: males have distinct yellow colour patches on the breast,

primary wing feathers, and side of the tail (Merilä et al. 1999). The yellow pigments in the

plumage consist of carotenoids (Stradi et al. 1995). Females and juvenile birds have less

pronounced colour patches, and are therefore more cryptic. When nesting, the female

builds the nest and incubates. The male feeds the female during the incubation, and after

the chicks have hatched, both parents feed the chicks. Greenfinches lay an average of 4.83

eggs/clutch and can produce several clutches per year (Eley 1991). Due to a heavy

predation of eggs and nestlings, only half of the eggs produce offspring that survive to the

fledgling stage. Fledgling period is 14 days, and age of first breeding is one year (Cramp

12 Lindström

and Perrins 1994). Greenfinches are seedeaters that feed gregariously on the ground, or in

bushes. In greenfinch flocks, access to food is influenced by dominance status: thus a few

dominant birds have priority access to the food, and subordinate birds may have to wait for

their turn or may eventually leave the flock if they are not successful (Cramp and Perrins

1994). Thus, an individuals social status may influence survival. Greenfinches

communicate their social status through plumage signals and through social interactions

(Fig. 1). Plumage brightness in greenfinches has been shown to function as a badge of

social status, and brighter males have higher status (Rohwer 1975, Maynard Smith and

Harper 1988, Eley 1991, Cramp and Perrins 1994).

Fig. 1 Illustration of a male greenfinch, showing the position of the yellow tail patches on wings and tail.During social interactions, greenfinches may display threats by a head forward gaping posture with wingsand tails slightly lifted.

The mating system of greenfinches has previously been studied in an English

population during a three year field study (Eley 1991, also in Cramp and Perrins 1994).

This study found that while the majority of greenfinch males were monogamous, polygamy

(24%) was frequent. Among polygynous males, most were bigamous (74%) but nestings

with up to five females per male were recorded. Apparent non-breeders in the population

(28-34%) were mostly males (Eley 1991). Male plumage brightness was related to mating

success, and polygamous males had brighter than average plumages. Among polygynous

males, brighter males paired with more females, and males that were involved in extra-pair

copulations were also brighter than average (Eley 1991). These data imply that females

prefer to mate with the most colourful males, although this study was unable to distinguish

between female choice and male-male competition (Eley 1991). Among male greenfinches


infected with blood parasites (Haemoproteus sp.) in Sweden, Finland and Spain, plumage

brigthness was negatively related to blood parasite load (Merilä et al. 1999). For these

greenfinches, male plumage brightness was also found to be positively related to testis size

in the breeding season (Merilä and Sheldon 1999). Together these studys imply that

plumage brightness in male greenfinches can be an indicator of phenotypic quality.

The Sindbis virus

The Sindbis virus is a RNA virus that belongs to the Alpha viruses, and the virus family

Togaviridae (Lundström 1999). The virus consist of a12kb RNA strand, that has been

completely sequenced (Strauss and Strauss 1994). It is an arthropod-borne virus

(arbovirus), transmitted to birds by blood-sucking mosquitoes. It has a wide host-range and

infects almost all species of Swedish passerines (I). The virus is common in bird

populations during the summers when vector mosquitoes are numerous.

During a Sindbis virus infection, the virus enters the host cells through cell surface

receptors (Wang et al. 1992). It replicates by budding through the plasma membrane and

can be detected in the bloodstream of infected bird hosts 2-7 days after initial infection

(Strauss and Strauss 1994, Fig. 2). The infection triggers an immune response, and the

most important component for clearance is thought to be the production of neutralizing

antibodies against protein E2 (Griffin et al. 1997). The complement system is important for

clearance of the virus, and depletion of complement results in prolonged viraemia (Hirsch

et al. 1980). Sialic acid can influence the activation of the complement system, and it has

been demonstrated that the sialic acid content of hosts cell tissues can affect viraemia.

Thus, natural immunity is higher in hosts with less sialic acid in their tissues (Hirsch et al.

1983). For chicks with undeveloped immunity, the infection can be lethal (Lundström and

Turell, unpublished study). The virus can also infect humans. In humans the infection can

lead to rash and a persistent joint ache (Espmark and Niklasson 1984, Niklasson et al.

1988). After the infection has been cleared, antibodies can be maintained several years in

humans (Niklasson et al. 1998). For birds, antibodies may persist at least for several

months, and once an antibody response has been produced, birds appear to be immune to

subsequent infections (Lundström, Krakower, Lindström, unpublished study).

14 Lindström

In this thesis, I have used the clearance rate of the Sindbis virus infection as an

estimate of immunocompetence. By measuring the clearance rate of a natural infection, I

have been able to avoid the problem of determining which component of the immune

system is the best estimate of immunocompetence (Siva-Jothy 1995, Westneat and

Birkhead 1998, Norris and Evans 2000). Nevertheless, the assumption that has to be made

is that the ability to control the Sindbis infection is positively related to the ability to

control other infections. Clearance of the Sindbis virus infection is a complex process that

requires the activation of both general and specific components of the immune system.

Thus, at least some components of the immune response are general. Also, studies on

poultry have demonstrated that immunity to one type of infection can be positively related

to immunity to other infections (e.g. Cheng and Lamont 1987).

Fig. 2 Example of viraemia and antibody response from three greenfinches (823, 830 and 824) after aSindbis virus infection. The left part of the figure shows virus titres (day 1-7) from a plaque tests expressedas the number of plaque forming units (PFU)/ml of blood in each day. The right part of the figure showsantibody titres (day 7-90) from a PRNT-test expressed as the reciprocal of the blood dilution giving an 80%reduction in plaques numbers. During the infection, infectious virus particles can be detected in the blood ofinfected host up to one week after infection. Around this time, specific antibodies against the virus firstappear.

DAYS AFTER INFECTION

VIR

US

TITR

E (P

FU/M

L)

ANTI

BOD

Y TI

TRE

0

100

200

300

400

0

10 1

10 2

10 3

10 4

1 2 3 4 5 6 7 14 21 28 90

ANTIBODY RESPONSEVIRAEMIA

823 830 824


Greenfinch measurements

For all greenfinches included in the experiments, body mass was measured with a Pesola

spring balance, to the closest 0.5 g, and tarsus length was measured with digital callipers to

the closest 0.1 mm. For males, the brightness of the plumage was scored by measuring the

length of the yellow tail, and wing patch with callipers to the closest mm (Fig.1). I also

counted the number of yellow feathers on the tail, primaries and alula on the right side of

the bird. The coverage of yellow on the breast was estimated on a scale between 1-5. The

repeatability of these measurements were moderate to high. Two principal components of

plumage brightness (PC1 and PC2) were extracted, and these two components were used

as plumages brightness scores (III). The first component (PC1) mainly described the

brightness of the wings, and this brightness was closely related to the birds age. The

second component (PC2) mainly described the size of the tail patch, and was unrelated to

age. Both components were adjusted for age-dependant variation before the analyses.

Since the brightness of primarys and breast were uninformative of virus clearance rate

(III), the length of the tail patch was used as a predictor of viraemia in later experiments

(IV, V). Two studies (II and V) include behavioural observations of greenfinches. In the

first of these studies (II), spontaneous locomotion activity was measured by counting the

number of hopping or flying movements each bird made during 25 minutes each morning.

Furthermore, the body mass and speed and angle of take-off flights after a simulated

predator attack was measured, as described by van der Veen and Lindström (2000). In the

last study (V), dominance hierarchies within pairs of males was measured by observing

their interactions around a feeding site from a hide. Males with priority access to food and

water were classified as dominants following the methods described by Senar (1990). In

one experiment (IV), blood testosterone concentrations were measured. A detailed

description of the sampling procedure and analyses is presented in paper IV.

16 Lindström

Virus infections and assays

In all experimental infections (II-V), greenfinches were injected with infectious Sindbis

virus diluted in 0.1 ml saline solution (0.09 % NaCl) subcutaneously into the breast. In two

studies (II & III), I prepared a solution with 103 plaque-forming units (PFU)/100 µl and

infected birds immediately after preparation. The amount of virus injected was based on

the estimated amount of virus delivered by infected mosquitoes during blood feeding

(Chamberlain and Sudia 1961). When a large number of birds were infected (III-V), or

when birds were infected on different days (II), a virus solution was prepared with the

concentration of 104 PFU/100 µl. This solution was frozen in separate vials until the

infection. Freezing and thawing reduced the amount of virus about tenfold, so that the

inoculated dose became equal to the one obtained during a natural infection when thawed.

The Edsbyn 82/5 strain of the Sindbis virus was used in all infections (II-V). This strain

represents the north European genotype of Sindbis virus (Norder et al. 1996), and was

taken through three passages in Vero cells before being used in the experiments.

To measure daily virus concentrations (III-V), birds were sampled for blood at 24h

intervals during seven days after inoculation. To measure the amount of specific

neutralising antibodies, blood samples were taken once a week until four weeks after the

infection, and then at two and three month post infection (III-IV). All blood samples

(100µl) were taken from the jugular vein using a fine syringe (27G). After sampling, blood

was either diluted in 1:10 in Hanks’ balanced salt solution (I, III, IV and V) (Life

technologies), or stored as undiluted plasma (III). Samples were kept on ice until storage,

and thereafter kept in -70°C until assayed. Blood samples from day one to seven were

assayed for virus content (III-V), with a plaque-test (Lundström et al. 1990). In the plaque-

test, serial 10-fold dilution’s of the blood samples were seeded in duplicates on a

monolayer of Vero cells, cultured on 24-well plates. Cells and blood-samples were then

overlaid with a nutrient solution in agar. After one day, live cells were visualised by neutral

red staining, and infectious virus particles were counted as plaques of dead and unstained

cells. The lowest detectable level of virus in the assay was 102 PFU/ml blood. The amounts

of Sindbis-specific antibodies in the samples were measured by a plaque-reduction

neutralisation test (PRNT) on heat inactivated (30 min in 56° C) samples (I, III, IV). In


this test, blood samples were serially diluted four-fold starting at 1:20, using the same

buffer solution, cell cultures and plates as used in the plaque assay. In the PRNT, each well

was seeded with 30 - 80 PFU of Sindbis virus and the neutralising antibody titres of the

samples were expressed as the reciprocal of the dilution giving 80% reduction in plaque

numbers (Early et al. 1967, Francy et al. 1989).

Statistics and data analyses

I used parametric tests where data met the required assumptions of normality (Sokal and

Rohlf 1995). When data was non-normal, either transformed data, or non-parametric

statistical tests were used (Siegel and Castellan 1989). To test for differences in virus

clearance ability between individuals, total viraemia, defined as the sum of the log10

transformed daily virus titres, were used (III, IV and V). I also used peak viraemia,

defined as the highest recorded virus titre, and virus clearance rate, defined as the number

of days with detectable virus titres (Fig. 2., III). To analyse time differences in viraemia

patterns between individuals, I compared early and late viraemia (IV and V), as the sum of

the log10 transformed daily virus titres for the first days (early viraemia); corresponding to

the proliferation phase, and for the last days (late viraemia); corresponding to the clearance

phase. The borders between early and late were arbitrarily chosen. To test for differences

in antibody production (III and IV), I compared antibody titres at 21 days post infection, at

the antibody peak: defined as the highest detected antibody titre of all samplings, and the

antibody production rate: defined as the antibody peak divided by the time it took to reach

this peak (Fig. 2).

Since the Sindbis virus is a naturally occurring virus, some wild caught birds may

have developed immunity after previous exposure. Thus, to be able to compare the ability

to clear a primary infection, I wanted to exclude all birds that had encountered the virus in

nature. These birds could sometimes be identified by detectable amounts of specific

antibodies before the infection (III). However, even after experimental infections

antibodies can not always be detected at all samplings. Thereby, this method could not be

used to reveal all birds with previous exposures. In a pilot experiment in which birds where

challenged with a second Sindbis virus infection, I found that all birds that had been

18 Lindström

previously exposed had become immune to the virus, and produced no detectable viraemia.

Also, as their secondary antibody responses were both faster and stronger compared to a

primary response (Lindström, Lundström, Krakower, unpublished study). In experiments

with viraemia and antibody data available (III, IV and V), I identified and removed birds

that had previously been exposed to the virus.

RESULTS AND DISCUSSION

Distribution of Sindbis virus infections in the field

In the first study of the thesis (I), the distribution of Sindbis specific antibodies among

free-ranging Swedish passerines in five areas was investigated. Antibodies occurred in

almost all passerine species (13/15, or 87%) sampled in large numbers (≥20 individuals).

The average prevalence across species was 7.7%. This and other studies indicate that the

Sindbis virus has a very wide host range (McIntosh et al. 1969, Lundström et al. 1993,

Lundström 1999). Although it seemed likely that all species of passerine birds could be

utilised by the virus as hosts, antibodies were unequally distributed between species.

Higher than average antibody prevalences were found in three thrush species, the fieldfare

(Turdus pilaris) (43.3%), the redwing (Turdus iliacus) (37.0%), and the song thrush

(Turdus philomelos) (22.2%). The distribution of antibodies in relation to bird body mass

was analysed for Sindbis virus and a related North American alphavirus (eastern equine

encephephalomyelitis virus). Both these viruses are transmitted by mosquito-vectors. It

could be concluded that medium sized birds were the most common hosts of both these

viruses. A factor that potentially could create differences in infection prevalence between

species is that species may have differential exposure to vectors, or differ in anti-mosquito

behaviour (Anderson and Brust 1995). It has previously been demonstrated that animals of

large size exhibit less behavioural defences against mosquitoes (Edman and Scott 1987,

Scott and Edman 1991). Within species, there was no differences in infection prevalence

between sexes. Early in the summers (June and July), adult birds were more often infected

than hatchling year birds, but later in the summer (August), the infection prevalence was

equal between birds of different age classes.


Costs of the virus infection

No attempts have been made to describe symptoms of disease associated with Sindbis

virus infections in passerines. Thus, the study presented in paper II represents a first

attempt to examine whether the Sindbis virus infection was indeed costly for infected

birds. To estimate the potential costs of infection, the behaviour of Sindbis virus infected

greenfinches was measured. First the spontaneous activity of greenfinches before and

during an infection was compared. There was a tendency for birds to reduce their activity

during an infection. To investigate if infected birds are more vunerable to predation, I

measured the body mass, and the speed and angle of take-off flights from a simulated

predator attack on both morning and evening. When comparing differences in take-off

speed from before to after infection, between a virus-infected group and a control-treated

group, it was found that take-off speed in the virus treated group was reduced in the

evenings, but not in the mornings (Fig. 3). There was also a significant increase in body

mass in the virus treated group compared to controls (Fig. 3). When healthy, greenfinches

can maintain the speed of take-off flights, although their body mass increases throughout

the day (van der Veen and Lindström 2000). However, during the infection the increased

body mass may lead to reductions of flight speed. An impaired take-off ability could

reduce the fitness of infected hosts. In conclusion, the behavioural alterations observed

imply that the infection was energetically costly, and that birds responded to the increased

energetical costs by altering their behaviour during the infection. The behavioural

alterations observed could also result from direct symptoms of disease, such as cell or

tissue damage (Holmes and Zohar 1990).

20 Lindström

MORNING EVENING

CH

ANG

E IN

SPE

ED (m

/s)

-0.6

-0.3

0

0.3

0.6*

MORNING EVENING

CH

ANG

E IN

BO

DY

MAS

S (g

)

-0.8

-0.4

0.0

0.4

0.8

1.2

1.6 **

Fig. 3 Bars in the upper graph illustrate the mean (± s.e) change in take-off speed of greenfinches in themorning and the evening, from before until after a saline treatment (unfilled bars), or an infection with theSindbis virus (filled bars). Positive values on the x-axes result from a mean within-individual increase. Thelower graph illustrate mean (± s.e) change in body mass. Sample sizes are n = 10 for saline group in allcases, and n = 9 for virus group in all cases. Significant differences in post-hoc paired t tests are illustratedwith bars, * p < 0.05.


Sexual selection

According to the intra-specific prediction of the Hamilton-Zuk hypothesis, a negative

relationship between parasite load and ornament size should be expected (Hamilton and

Zuk 1982). In paper III, this prediction was tested using greenfinches infected with the

Sindbis virus. During an infection, greenfinches may vary in viraemia, virus clearance rate

and antibody production rate, although infected with equal virus doses (Fig. 2).

Greenfinches also vary in plumage brightness, and for males, plumage brightness may be a

sexually selected trait (Eley 1991, Merilä and Sheldon 1999, Merilä et al. 1999).

Two principal components (PC1 and PC2) were used to describe plumage

brightness of males. For the second principal component, there was a positive relation

between virus clearance rate and plumage brightness. Thus, males with large tail patches

had a faster clearance of the infection (Fig. 5). Males with large tail patches also had lower

virus titres during the infection-period, this was revealed by the negative relation between

total viraemia and tail patch size (Fig. 5). This is the first study to demonstrate that the

clearance rate after a virus infection can be predicted by male ornament size. I found no

clear relationship between male plumage characters and antibody titres produced after the

infection. Also, there were no clear relationships between viraemia and antibody

production rate (Table 1). Thus, at least for this virus infection, the amount or speed of

antibody production after infection was a poor predictor of parasite load or parasite

clearance rate.

AB day 21 AB peak AB production rater p r p r p

Virus clearance 0.00 0.99 -0.13 0.53 0.17 0.37Total viraemia 0.10 0.60 0.24 0.21 -0.31 0.10

Table 1 Pearson correlation coefficients (r) of virus clearance and total viraemia against antibody titres 21days after the infection (AB day 21), antibody peak levels (AB peak) and the antibody peak level/peak week(AB production rate). The sample size was 28 in all cases.

22 Lindström

Fig. 5 Virus clearance rate (r = 0.58, p = 0.001, n = 28) and total viraemia (r = -0.46, p = 0.01, n = 28) as afunction of plumage brightness. Plumage brightness is measured as PC2 scores, adjusted for age dependentvariation. Males with large yellow tail patches had higher virus clearance rate and lower total viraemiacompared to males with smaller tail patches. There were no clear relationships between any measurement ofantibody production and plumage brightness.

PLUMAGE BRIGHTNESS (PC2)

VIR

US

CLE

ARAN

CE

RAT

E

-2,0

-1,5

-1,0

-0,5

0,0

0,5

1,0

1,5

2,0

-3 -2 -1 0 1 2 3

PLUMAGE BRIGHTNESS (PC2)

TOTA

L VI

RAE

MIA

-1,5

-1,0

-0,5

0,0

0,5

1,0

1,5

2,0

-3 -2 -1 0 1 2 3


Testosterone and virus clearance

In paper IV, I tested if the immunocompetence-handicap model (Folstad and Karter 1992)

could be applied to the greenfinch-Sindbis virus system. According to this model,

increased testosterone levels are expected to suppress immunity (Folstad and Karter 1992).

If the model holds, an increased viraemia or reduced antibody production of testosterone-

implanted males would be expected. Also, if the quality signal discovered in paper III is

honest, the cost of increased testosterone should be lower for males of better quality

(Grafen 1990). I therefore predicted that males with large tail patches should have lower

costs associated with the elevated testosterone levels. Thus, a birds response to the

treatment should be related to his tail patch size. I tested these predictions by introducing

either a testosterone- or control-implant into males, that were matched in pairs according to

their age and tail patch size. All birds were then infected with Sindbis virus after which

viraemia and antibody production rate were compared between treatments. Following the

implantations, testosterone concentrations differed between treatments. Still, I found no

differences in either antibody production rate or total viraemia between treatments.

Furthermore, there was no evidence that males with small tail patches reacted differently to

the treatment compared to males with large tail patches. Thus, I could not find support for

the assumption of the immunocompetence-handicap hypothesis. However, there were

differences between treatments in the timing of the infection (Fig. 6). Early in the infection

(Day 1-3), testosterone treated males had lower viraemia compared to control treated

males. Late in the infection (day 4-7) testosterone treated males had highest viraemia. Thus

overall the implant caused a delay, but not an increase, in viraemia.

In other recently published studies on birds investigating the immunosuppressive

effect of testosterone, results have been equivocal (e.g. Hasselqvist et al. 1999, Evans et al.

2000, Peters et al. 2000). Apparently, testosterone can interact with immunity, but these

interactions are not straightforward (Marsh 1992). The experiment with greenfinches

shows that increased testosterone concentrations can result in both increased and decreased

virus titres during an infection. A factor that further complicates the interpretation of these

results is that increased levels of testosterone can cause a release of stress hormones (e.g.

corticosteroids). In two recent studies on house sparrows (Passer domesticus), the effects

24 Lindström

of testosterone implants on immunity were entirely explained in terms of an accompanying

release of corticosteroids (Evans et al. 2000, Poiani et al. 2000). Thus, it is possible that the

effect on viraemia resulted from a release of stress hormones rather than differences in

levels of testosterone (Apanius 1998). It has also recently been argued that testosterone

redistribute immune cells, rather than inhibits immune functions (Braude et al. 1999).

Consideration of these poinst suggests that the immunocompetence-handicap model may

need to be revised.

DAY POST INFECTION

VIR

US

TITR

E (L

og10

PFU

/ml)

0

1

2

3

4

5

1 2 3 4 5 6 7

T-IMPLANTCONTROL

Fig. 6 Mean (± s.e) of daily virus titres of testosterone and control implanted male greenfinches. Plottedvalues are log10 transformed Sindbis virus titres (PFU/ml blood) for day 1-7 after infection. The sample sizewas 15 in each group. Early in the infection (sum of titres day 1-3), testosterone implanted males had lowervirus titres (t-test, t28 = 2.48, p = 0.020), and late in the infection (sum of titres day 4-7), testosteroneimplanted males had higher virus titres (t-test, t28 = -2.42, p = 0.022).

Social status and virus clearance

The final experiment (V) focuses on the relationship between the Sindbis virus infection

and social dominance. I tested if infected males would lose rank against an uninfected male

and whether the virus clearance ability of males was related to their social status. First, I

carried out daily observations of males kept together in pairs, in which one male in the pair

was infected. During the infection, the number of altered ranks between subsequent days

were no higher than what could be expected by chance (Fig. 7). Also, when an infected and


an uninfected male were first introduced, there was no association between infection status

and the outcome of the dominance test. Instead, the outcome of the dominance test could

be predicted from the outcome of a male’s previous encounters in all cases. Thus, none of

the results indicated that social status decreased during the Sindbis virus infection.

When viraemia patterns between subdominant and dominant males were compared,

it was found that dominant males had higher early viraemia (Fig. 8). There was a

significant positive correlation between within-group rank and early viraemia, and birds

with high rank had high early viraemia. Total viraemia did not differ between dominants

and subdominants, nor was it related to within group rank. Late viraemia was higher

among subdominant birds, and at this time, there was a significant negative relationship

between social rank and late viraemia.

A bird’s social rank may be the trait that is most strongly connected to fitness.

Dominant individuals may gain access to both food and partners, especially when these are

in short supply and there is competition over these resources. Although I had previously

found indications that other behavioural traits, such as take-off flight and locomotion

activity were down-regulated during an infection-period (II), the results of this experiment

(V) indicate that social status was maintained during the infection. The close relationship

between social status and viraemia was an interesting and somewhat unexpected finding.

The differences found in viraemia patterns between males of different rank could be due to

a physiological trade-off. If males with high rank initially had invested less in immunity,

they would become more easily infected, and thereby obtain initial high virus titres.

However, rank may also be positively related to quality, so that high-ranked males respond

to their heavy infection with an efficient and rapid virus clearance. Mechanistically, these

differences could be due to differences in hormone levels. A potential hormone candidate

that could create such a difference is testosterone. A high social status in birds and

mammals is commonly associated with elevated testosterone levels, although this

relationship does not always hold true (Wingfield et al. 1990, Evans et al. 2000). In this

study, the viraemia pattern of subordinate males were similar to that of testosterone

implanted males, and a more likely hormone candidate is therefore corticosteroids.

Subordinate males in this study and testosterone implanted males in the previous study

(IV) could both have experienced increased levels of stress hormones. Stress hormones,

26 Lindström

like corticosteroids are potent regulators of immunity, and increased stress can function

both as an enhancer and a repressor of immune responses or parasite load (Apanius 1998).

Fig. 7 The proportion of pairs with stable ranks between subsequent test days. In each pair, one male wasinfected with Sindbis virus at day 0. On the following four days, 16 pairs of males were ranked daily. Thesolid line indicate the proportion of ranks (89%, n = 31) that remained stable during the control week, whennone of the males were infected. The area above the lower (hatched) line indicates the variation that wasexpected due to chance. Thus although one male was infected, ranks remained stable.

Fig. 8 Mean (± s.e) of daily virus titres for dominant and subordinate males after an experimental infection.The sample size is 8 in each group. Early in the infection (sum of titres day 1-4), subordinate males hadlower virus titres, and late in the infection (sum of titres day 5-7), subordinate birds had higher virus titres.

DAYS POST INFECTION

% S

TABL

E R

ANKS

0

20

40

60

80

100

DAY 1 DAY 2 DAY 3 DAY 4

Days post infection

Viru

s tit

re (l

og P

FU/m

l)

0

0.5

1.0

1.5

2.0

2.5

3.0

1 2 3 4 5 6 7

SUBORDINATESDOMINANTS


CONCLUSIONS AND PROSPECTS

The distribution of infections in nature may vary for many reasons. In the first paper of this

thesis (I), data showed that passerine species sharing the same environment may differ in

infection prevalence. Since exposure to infections may differ, this may create differences

in infection prevalence between individuals. Although many infections are not directly

lethal, the fitness of infected individuals may still be reduced if for example escape-

behaviours are impaired during the infection (II).

Although infected with equal doses of the virus, greenfinches varied in their ability

to control the infection. The reasons for this variation is not well understood. For this virus

infection, there was no clear relationship between antibody-production and viraemia. Since

much of the genetic variation in the immune system of vertebrates is located in the major

histability complex (MHC) (Grahn 2000), an interesting prospect for the future would be

to investigate if there is a connection between variation in viraemia and MHC genotype.

The results in paper III show that the size of the tail patch in males reveals

information about his ability to control the Sindbis virus infection. If virus clearance ability

is heritable, female greenfinches that prefer males with large tail patches could potentially

gain indirect genetic benefits (Kirkpatrick and Ryan 1991). It remains to be investigated if

the ability to control the Sindbis virus is heritable. In a study on bluethroats (Luscinia

svecica), it was recently demonstrated that females enhance the immunocompetence of

their offspring through extra-pair copulations (Johnsen et al. 2000). That virus resistance

can be a heritable trait, have been demonstated in studies of poultry (e.g. Bumstead et al.

1993).

It has become increasingly clear that male ornament expression can reveal aspects

of quality, but it remains unclear how these factors are regulated. I found no support for the

prediction that increased testosterone levels in males were costly in terms of immuno-

supression (IV). However, I did find that viraemia patterns could be modulated by

testosterone, and similar differences in viraemia patterns were also related to an individuals

social status (V). Since an individuals behaviour may be regulated by hormone levels, a

model that could fully explain how quality signals are maintained would probably have to

consider both the effects of testosterone and social stress on immunity (Poiani et al. 2000).

28 Lindström

Many male traits can function both as cues in female choice and in male-male interactions

simultaneously, and for the expression of such traits the costliness of female directed

signals could be influenced by social interactions with other males (Berglund and Bisazza

1996, Qvarnström and Forsgren 1998, Poiani et al. 2000). Also, since plumage colour of

greenfinches are carotenoid based, the interactions between carotenoids and parasite load

is an interesting area that deserves further investigation (Olson and Owens 1998). The

integration of these models will most certainly provide an interesting challenge for future

research.


ACKNOWLEDGEMENTS

To become a doctor on greenfinches is somehow not exactly what I had expected……

Although I feel a bit surprised, I am very glad I did this, and this four years long journey

into the scientific world has been both stimulating and rewarding. Of course, it would not

have been possible for me to complete the project without the help and inspiration from

many other people. First, I would therefore like to thank everyone at ‘Zootis’ for making it

into the fun, friendly and interesting place that it is to work in. THANK YOU ALL!

Although, that includes most of the people mentioned below, some persons deserves to be

thanked at least twice:

The first one of these is of course my wonderful supervisor-roommate-coauthor Jan

Lundström. Janne has taken the time to help with all sorts of practical things, we have had

many interesting discussions over the years and he has read and commented on all my

manuscript without loosing his good mood. I am most grateful for all time and support you

have provided, for inviting me to your nice family, and for being a friend! Thank you!

I would also like to thank Jan Ekman for getting me started by hiring me, and Jacob

Höglund for helping me to reach the finish line. Ingrid Ahnesjö is a great person that has

helped me to maintain calm and organised. Douglas Krakower made a huge effort feeding

and bleeding birds, and even sang encouraging songs while doing so. He also turned out to

be an excellent English proofreader. I really enjoyed working with you Special-D! Ineke

helped me a lot with both texts and statistics, and kept reminding me to take coffee breaks.

You are a rock (or klippa in Swedish)! Anna K helped to improve many of my texts and

she also kept my knowledge of nudibranchs and flower maintenance updated. These years

would have been much less fun without you! Tina S kept my supervisor busy, so that I

could occasionally rule the office. She also sometimes changed the channels on my TV.

Boris-Antoine did his master project on greenfinches and patiently counted their jumps

while learning Chinese. Arne and Pekka sometimes provided me with greenfinches from

the garden. Reija analysed a lot of greenfinch blood-smears, Theresa J and Stein-Are

helped me improve the text in the summary. Henk gave comments on my text and has been

a good friend. Juha gave great comments on any manuscript that I gave him, Måns A

taught me that answers can sometimes be found in Poultry Science, Christer H showed me

30 Lindström

how to hold a bird and Lotta helped me (without knowing it) by writing a very good thesis.

Also, a lot of other people like Maria S, Ane, Simon, Katrin, Sami, Fredrik W, Lisa, Johan

W, Robert, Marcus, Anders, Jon and Olle has kept the party spirit alive at the department.

A great deal of practical help was provided by Nisse. Over the years, Nisse taught me one

or two things about carpentry and once was nice enough to give me a sandwich at

Bäcklösa. Marianne and Ingela also helped with many practical things and kept the

department organised. Karin Söderhäll did a great job keeping the Vero cells alive, and

Jonas Blomberg and his colleagues at clinical virology kindly let me use their lab to grow

Vero cells.

Other people have contributed to this work in a more indirect but equally important

way. Pysselklubben and the Wednesday Clay-Day with Barbro kept my weekdays fun. On

other days, all my wonderful friends have kept my life busy and happy. Kursargänget have

been great party company at all times. The weekends with Västeråsbrudarna gave much

important inspiration and always filled my batteries with new energy. Göran H is a special

friend that deserves special mentioning. Lene fed me with tasty meals and much good

advice. Sofia has been a great adventure and holiday companion. Thank you all for being

so great friends!

Finally, I would like to thank my dear family: My sister Lena taught me that

sometimes you need to take things one step at the time, she also helped by providing me

with a working computer. Lena, Leif, my father Sam and Arja have provided much

encouragement and personal support. Thank you all for being there! My last and biggest

thanks goes to my wonderful mother Margit for believing so much in me, and for

supporting all weird things that I have done in my life. You will always be there for me in

my heart!

Financial support was provided by the Olle Engqvist Byggmästare Foundation, the

Crawfoord Foundation, the Zoological Foundation and the Royal Swedish Academy of

Science.


REFERENCES

Anderson RA & Brust RA, 1995. Blood feeding success of Aedes aegypti and Culex

nigripalus (Diptera: Culicidae) in relation to defensive behaviour of Japanese Quail

(Coturix japonica) in the laboratory. J Vect Ecol 21:94-104.

Andersson M, 1994. Sexual selection. Princeton: Princeton University Press.

Apanius V, 1998. Stress and immune defence. Adv Stud Beh 27:133-153.

Barnard CJ, Behnke JM & Sewell J, 1994. Social behaviour and susceptibility to infection

in house mice (Mus musculus): effects of group size, aggressive behaviour and

status-related hormonal responses prior to infection on resistance to Babesia microti.

Parasitol 108:487-496.

Berglund A, Bisazza A & Pilastro A, 1996. Armaments and ornaments: an evolutionary

explanation of traits of dual utility. Biol J Linn Soc 58:385-399.

Braude S, Tang-Martinez Z & Taylor GT, 1999. Stress, testosterone and the

immunoredistribution hypothesis. Behav Ecol 10:345-350.

Bumstead N, Reece RL, Cook JKA, 1993. Genetic differences in susceptibility of chicken

lines to infection with infectious bursal disease virus. Poultry Sci 72:403-410.

Chamberlain RW & Sudia WD, 1961. Mechanisms of transmission of viruses by

mosquitoes. Annu Rev Entomol 6:371-390.

Cheng S & Lamont SJ, 1987. Genetic analyses of immunocompetence measures in a white

leghorn chicken line. Poultry Sci 67:989-995.

Clayton DH & Moore J, 1997. Host-parasite evolution: general principles and avian

models. Oxford: Oxford University Press.

Cramp S & Perrins CM, 1994. The birds of the western palearctic. Vol. VIII.

Oxford:Oxford University Press.

Darwin C, 1871. The descent of man and selection in relation to sex. London: Murray.

Domingo E & Holland JJ, 1994. Mutation rates and rapid evolution of RNA viruses. In:

The evolutionary biology of viruses (Morse SS, ed) pp. 161-184. New York: Raven

Press.

Earley E, Peralta PH & Johnson KM, 1967. A plaque neutralization method for

arboviruses. Proc Soc Exp Biol Med 125:741-747.

32 Lindström

Edman JD & Scott TW, 1987. Host defensive behaviour and the feeding success of

mosquitoes. Insect Sci Applic 8:617-622.

Eley C, 1991. Status signalling in the western greenfinch, Carduelis chloris. PhD Thesis,

University of Sussex.

Espmark Å & Niklasson B, 1984. Ockelbo disease in Sweden: epidemiological, clinical,

and virological data from the 1982 outbreak. Am J Trop Med Hyg 33:1203-1211.

Evans MR, Goldsmith AR & Norris SRA, 2000. The effect of testosterone on

antibodyproduction and plumage coloration in male house sparrows (Passer

domesticus). Behav Ecol Sociobiol 47:156-163.

Ewald PW, 1994. The evolution of infectious disease. Oxford: Oxford University Press.

Fisher RA, 1930. The genetical theory of natural selection. Oxford: Claredon Press.

Folstad I & Karter AJ, 1992. Parasites, bright males, and the immunocompetence

handicap. Am Nat 139:603-622.

Francy DB, Jaenson TGT, Lundström JO, Schildt E-B, Espmark Å, Henriksson B &

Niklasson B, 1989. Ecologic studies of mosquitoes and birds as hosts of Ockelbo

virus in Sweden and isolation of Inkoo and Batai viruses from mosquitoes. Am J

Trop Med Hyg 41:355-363.

Grafen A, 1990. Biological signals as handicaps. J Theor Biol 144:517-546.

Grahn M, 2000. MHC genotype and ornamentation. In: Animal signals, signalling and

signal design in animal communication. (Espmark Y & Rosenqvist G eds)

pp. 421-436. Trondheim: Tapir academic press.

Griffin D, Levine B, Tyor W, Ubol S & Desprès P, 1997. The role of antibody in recovery

from alphavirus encephalites. Immunol Rev 159:155-161.

Grossman CJ, 1985. Interactions between the gonadal steroids and the immune system.

Science 227:257-261.

Gustafsson L, Nordling D, Andersson MS, Sheldon BS & Qvarnström A, 1994. Infectious

disease, reproductive effort and the cost of reproduction in birds. Phil Trans R Soc

Lond B 346:323-331.

Hamilton WD & Zuk M, 1982. Heritable true fitness and bright birds: a role for parasites?

Science 218:384-387.

Hamilton WJ & Poulin R, 1997. The Hamilton and Zuk hypothesis revisited: a meta-

analytical approach. Behaviour 134: 299-320.


Hasselquist D, Marsh JA, Sherman PW & Wingfield JC, 1999. Is avian humoral

immunocompetence suppressed by testosterone? Behav Ecol Sociobiol 45:167-175.

Hillgarth N & Wingfield J, 1997. Parasite-mediated sexual selection: endocrine aspects. In:

Host-parasite evolution: general principles and avian models (Clayton D & Moore J,

eds) pp. 78-104. Oxford: Oxford University Press.

Hirsch RL, Griffin DE & Winkelstein JA, 1983. Natural immunity to Sindbis virus is

influenced by host sialic acid content. Proc Natl Acad Sci USA 127:1740-1743.

Hirsch RL, Winkelstein JA & Griffin DE, 1980. The role of complement in viral

infections. III. Activation of the classical and alternative pathway by Sindbis virus. J

Immunol 124:2507-2510.

Hoelzer GA, 1989. The good parent process of sexual selection. Anim Behav 38:1067-

1078.

Holmes JC & Zohar S, 1990. Pathology and host behavior. In: Parasitism and host

behavior. (Barnard CJ & Behnke JM eds) pp. 34-63. London: Taylor & Francis.

Hudson PJ & Dobson AP, 1997. Host-parasite processes and demographic consequences.

In: Host-parasite evolution, general principles and avian models. (Clayton DH &

Moore J eds). pp. 128-154. Oxford: Oxford University Press.

Johnsen A, Andersen V, Sunding C & Lifjeld JT, 2000. Female bluethroats enhance

offspring immunocompetence through extra-pair copulations. Nature 406:296-299.

Johnstone RA & Norris R, 1993. Badges of status and the cost of aggression. Behav Ecol

Sociobiol 32:127-134.

Jones TM, Quinnell RJ & Balmford A, 1998. Fisherian sand flies: benefits of female

choice in a lekking sandfly. Proc R Soc Lond B 265:1651-1657.

Kirkpatrick M & Ryan MJ, 1991. The evolution of mating preferences and the paradox of

the lek. Nature 350:33-38.

Lochmiller RL & Deerenberg C, 2000. Trade-offs in evolutionary immunology: just what

is the cost of immunity? Oikos 88:87-98.

Lundström JO, 1999. Mosquito-borne viruses in Western Europe: A review. J Vect Ecol

24:1-39.

Lundström JO, Niklasson B & Francy BD, 1990. Swedish Culex torrentium and Cx.

pipiens (Diptera: Culicidae) as experimental vectors of Ockelbo virus. J Med

Entomol 27:561-564.

34 Lindström

Lundström JO, Turell MJ & Niklasson B, 1993. Viremia in three orders of birds

(anseriformes, galliformes and passeriformes) inoculated with Ockelbo virus. J Wildl

Dis 29:189-195.

Marsh J, 1992. Neuroendocrine-immune interactions in the avian species- a review.

Poultry Sci Rev 4:129-167.

Maynard Smith J & Harper DGC, 1988. The evolution of aggression: can selection

generate variability? Phil Trans R Soc Lond B 319:557-570.

McIntosh BM, Madsen W, Dickinson DB, 1969. Ecological studies on Sindbis virus and

West nile viruses in South Africa. VI. The antibody response of wild birds. S Afr J

Med Sci 34:83-91.

Merilä J & Sheldon B, 1999. Testis size variation in the greenfinch Carduelis chloris:

relevance for some recent models of sexual selection. Behav Ecol Sociobiol 45:115-

123.

Merilä J, Sheldon BC & Lindström K, 1999. Plumage brightness in relation to hematozoan

infections in the Greenfinch, Carduelis chloris- bright males are a good bet.

Ecoscience 6:1-7.

Morse SS, 1994. The evolutionary biology of viruses. New York: Raven Press.

Møller AP, 1990. Parasites and sexual selection: current status of the Hamilton and Zuk

hypothesis. J Evol Biol 3:319-328.

Møller AP, 1995. Hormones, handicaps and bright birds. Trends Ecol Evol 10:121

Møller AP, Allander K & Dufva R, 1990. Fitness effects of parasites on passerine birds: a

review. In: Population Biology of passerine birds: An integrated approach. (Blondel

J, Gosler A, Lebreton JD & McCleery RH eds) pp 269-280. Berlin: Springer Verlag.

Niklasson B, Espmark Å & Lundström J, 1988. Occurrence of arthralgia and specific IgM

antibodies three to four years after Ockelbo disease. J Inf Dis 157:832-835.

Norder H, Lundström JO, Kozuch O & Magnius LO, 1996. Genetic relatedness of Sindbis

virus strains from Europe, Middle East and Africa. Virology 222:440-445.

Nordling D, Andersson M, Zohari S & Gustafsson L, 1998. Reproductive effort reduces

specific immune response and parasite resistance. Proc R Soc Lond B 265:1291-

1298.

Norris K & Evans MR, 2000. Ecological immunology: life history trade-offs and immune

defence in birds. Beh Ecol 11:19-26.


Nuttall PA, 1997. Viruses, bacteria, and fungi of birds. In: Host-parasite evolution, general

principles and avian models. (Clayton DH & Moore J eds) pp. 271-302. Oxford:

Oxford University Press.

Olson VA & Owens IPF, 1998. Costly sexual signals: are carotenoids rare, risky or

required? Trends Ecol Evol 13:510-514.

Owens IPF & Short RV, 1995. Hormonal basis of sexual dimorphism in birds: implications

for new theories of sexual selection. Trends Ecol Evol 10:44-47.

Penn D & Potts WK, 1998. Chemical signals and parasite-mediated sexual selection.

Trends Ecol Evol 13:391-396.

Peters A, 2000. Testosterone treatment is immunosuppressive in superb fairy wrens, yet

free-living males with high testosterone are more immunocompetent. Proc R Soc

Lond B

Poiani A, Goldsmith AR & Evans MR, 2000. Ectoparasites of house sparrows (Passer

domesticus): an experimental test of the immunocompetence handicap hypothesis.

Behav Ecol Sociobiol 47:230-242.

Price PW, 1980. The evolutionary biology of parasites. Princeton: Princeton University

Press.

Qvarnström A & Forsgren E, 1998. Should females prefer dominant males? Trends Ecol

Evol 13:498-501.

Read AF, 1990. Parasites and the evolution of host sexual behaviour. In: Parasitism and

Host behaviour. (Barnard CJ & Benkhe JM eds) pp. 117-157 London: Taylor &

Francis.

Rohwer S, 1975. The social significance of avian plumage variability. Evolution 29:593-

610.

Scott TW & Edman JD, 1991. Effects of avian host age and arbovirus infection on

mosquito attraction and blood-feeding success. In: Bird-parasite interactions (Loye

JE & Zuk M eds) pp. 179-204 Oxford: Oxford University Press.

Senar JC, 1990. Agonistic communication in social species: what is communicated?

Behaviour 112:270-283.

Sheldon BC & Verhulst S, 1996. Ecological immunity: costly parasite defences and trade-

offs in evolutionary ecology. Trends Ecol Evol 11:317-321.

36 Lindström

Siegel S, Castellan NJ, 1989. Nonparametric statistics for the behavioural science, 2nd ed.

New York: McGraw-Hill

Siva-Jothy MT, 1995. “Immunocompetence”: conspicuous by its absence. Trends Ecol

Evol 10:205-206.

Sokal, RR. & Rohlf FJ, 1995. Biometry 3rd edn. San Francisco: Freeman.

Stearns SC, 1992. The evolution of life histories. Oxford: Oxford University Press.

Stradi R, Celentano G, Possi E, Rovati G & Pastore M, 1995. Carotenoids in bird plumage:

I. The carotenoid pattern in a series of Palearctic Carduelinae. Comp Biochem

Physiol B Biochem Mol Biol 110:131-143.

Strauss JH & Strauss EG, 1994. The alphaviruses: gene expression, replication and

evolution. Microbiol Rev 58:491-562.

Svensson E, Råberg L, Koch C & Hasselqvist D, 1998. Energetic stress,

immunosupression and the cost of an antibody response. Func Ecol 12:912-919.

Svensson L, 1992. Identification guide to European Passerines. Stockholm:Lars Svensson.

van der Veen I & Lindström K, 2000. Escape flights of yellowhammers and greenfinches:

more than just physics. Anim Behav 59:593-601.

Verhulst S, Dieleman SJ & Parmentier HK, 1999. A trade-off between immunocompetence

and sexual ornamentation in domestic fowl. Proc Natl Acad Sci USA 96:4478-4481.

Wang K-S, Kuhn RJ, Strauss EG & Ou S, 1992. High-affinity laminin receptor is a

receptor for Sindbis virus in mammalian cells. Virology. 66:4992-5001.

Westneat DF & Birkhead TR, 1998. Alternative hypothesis linking the immune system and

mate choice for good genes. Proc R Soc Lond B 265:1065-1073.

Wingfield JC, Hegner RE, Dufty AM & Ball GF, 1990. The “challenge hypothesis”:

Theoretical implications for patterns of testosterone secretion, mating systems, and

breeding strategies. Am Nat 136:829-846.

Zahavi A, 1975. Mate selection- a selection for a handicap. J Theor Biol 53:205-214.

Zuk M & Johnsen TS, 2000. Social environment and immunity in the red jungle fowl.

Behav Ecol 11:146-153.


SWEDISH SUMMARY - SVENSK SAMMANFATTNING

Samspelet mellan fåglar och parasiter - med Sindbisvirus som modellsystem.

Fåglar och andra djur riskerar under sitt liv att drabbas av en mängd olika infektioner.

Individer som effektivt kan kontrollera dessa infektioner kan både få en bättre överlevnad

och en större avkomma. Därför är immunitet mot parasitangrepp en viktig egenskap som

kan utvecklas över tiden genom det naturliga urvalet. På senare tid har man även upptäckt

att motståndskraft mot parasiter kan utvecklas genom sexuellt urval. Om honor föredrar att

para sig med hanar med god motståndskraft mot parasiter får dessa hanar en större

reproduktiv framgång, och immuniteten mot parasiter kan föras vidare till nästa generation.

I den här avhandlingen har jag använt en virusinfektion hos fåglar som modellsystem för

att studera hur värddjurs försvar mot parasiter kan utvecklas genom evolution.

I alla undersökningar har jag använt mig av Sindbisviruset. Sindbis är ett vanligt

förekommande virus hos fåglar, och det tycks kunna infektera de flesta arter tättingar i

Sverige. Fåglar smittas av viruset via fågelbitande myggor som fungerar som bärare. Mer

sällan infekteras också människor av viruset. Sjukdomen som viruset orsakar hos

människor kallas “Ockelbosjukan”, och kan leda till långvariga ledbesvär.

Den första artikeln i avhandlingen kartlägger utbredningen av Sindbisinfektioner

hos vilda fåglar i Sverige (Artikel I). För att ta reda på vilka fåglar som varit infekterade

fångades fåglar in med slöjnät på sommaren vid fem olika platser under flera år.

Blodprover från ca 1000 fåglar av ett 40-tal arter analyserades för förekomst av

antikroppar mot Sindbisviruset. Förekomst av antikroppar visar att en fågel varit infekterad

av viruset. Hos nästan alla fågelarter där jag undersökte mer än 20 fåglar fanns antikroppar

mot viruset hos någon individ. Detta visar att Sindbisviruset verkar kunna utnyttja de flesta

arter av tättingar som värddjur. Andelen infekterade individer skilde sig dock mellan arter,

och tre trastarter; björktrast (Turdus pilaris), rödvingetrast (Turdus iliacus) och taltrast

(Turdus philomelos), var oftare infekterade än övriga tättingar. Hos dessa trastar var

mellan 22% och 44% av alla testade individer bärare av antikroppar, medan genomsnittet

för samtliga arter var 8%. Tidigt under somrarna var äldre fåglar oftare infekterade än

årsungar, men denna skillnad försvann under sensommaren. Hanar och honor hade

38 Lindström

antikroppar i lika hög grad. Eftersom så stor andel trastar var infekterade kan förekomsten

av trastar i ett område vara en betydelsefull faktor som reglerar Sindbisvirusets utbredning.

För övriga artiklar i avhandlingen användes grönfinkar (Carduelis chloris) som

värddjur för viruset (Fig. 1). Jag valde att arbeta med grönfinkar eftersom de infekteras av

viruset i naturen, och är lätta att hålla i bur. I den andra artikeln undersökte jag om

fåglarnas beteende påverkades av en infektion (Artikel II). De beteenden som mättes hos

fåglar var deras aktivitet och deras förmåga att fly undan en simulerad rovfågelsattack.

Under infektionen fanns en tendens till att grönfinkarna var mindre aktiva. Infekterade

fåglar hade också en högre kroppsvikt. När infekterade fåglar flydde undan en simulerad

rovfågelsattack så lyfte de med en lägre hastighet, kanske för att de var tyngre än vanligt.

Om fåglar som infekteras med viruset har svårare att fly från rovfåglar, kan detta försämra

deras chanser att överleva. Detta kan i sin tur innebära att fåglar med god förmåga att

motstå infektionen gynnas i det naturliga urvalet.

Fig. 1 Hos grönfinken (Carduelis chloris) är hanar färggrannare än honor, och hanens färger utgör en signalom hanens kvalitet. Denna signal kan uppfattas av både honor och andra hanar. Cirka 10% av allagrönfinkar i Sverige blir på sommaren infekterade med Sindbisviruset, som överförs via infekterade myggor.


Det är sedan tidigare känt att hanar använder sig av ornament (utsmyckningar) för

att locka till sig honor. Många forskare tror idag att hanliga ornament i själva verket ger

honor information om hanens förmåga att motstå parasitinfektioner, och att hanar med

stora ornament har en god förmåga att motstå infektioner. I den tredje artikeln undersökte

jag därför hur omfattande viremi (viruskoncentration i blodet) grönfinkar med olika stora

ornament fick efter en experimentell infektion (Artikel III). I försöket fann jag att

grönfinkshanar med stora gula ornament på stjärtfjädrarna hade lägre viremi jämfört med

hanar med mindre ornament. Hanar med stora ornament var alltså mer motståndskraftiga

mot infektionen. Andra studier på grönfinkar har visat att hanar med stora ornament också

är mer populära hos honor. Hanens förmåga att motstå infektioner kan alltså vara en

egenskap som är betydelsefull i det sexuella urvalet.

Artikel III visade att det fanns ett samband mellan storleken på en hanes ornament

och hans förmåga att motstå en infektion. Man kan fråga sig hur ett sånt samband kan

upprätthållas? Några forskare har föreslagit att det hanliga könshormonet testosteron kan

fungera som en viktig länk mellan immunförsvar och uttrycket av hanliga ornament. Om

det krävs höga halter av testosteron för att utveckla stora ornament och testosteron

samtidigt har negativa effekter på immunförsvaret, har bara hanar med starkt

immunförsvar råd att utveckla stora ornament. En grundläggande förutsättning för att

denna modell ska gälla är att testosteron verkligen har negativa effekter på hanens

immunförsvar. Jag undersökte om detta stämde genom att höja testosteronhalten hos

grönfinkshanar före en experimentell virusinfektionen (Artikel IV). Det visade sig att

hanar med förhöjda halter testosteron fick infektionen senare än obehandlade hanar, men

infektionen höll i sig längre. Hanar som behandlats med testosteron fick däremot inte en

mer omfattande infektion. Trots att testosteron inverkade på infektionsförloppet, kunde

man alltså inte dra slutsatsen att ökade halter testosteron försämrade förmågan att motstå

en infektion.

Grönfinkshanarnas fjäderdräkt används inte bara som en kvalitetssignal till honor;

den kan även ge information om hanens rang, eller sociala status, i flocken. Tidigare

undersökningar av grönfinkar visar att grönfinkshanar med färgrika fjäderdräkter är av hög

rang. I den femte artikeln undersökte jag om hanarnas rang kunde påverkas av en

virusinfektion, och om det finns något samband mellan en hanes rang och hans förmåga att

motstå en virusinfektion (Artikel V). Genom att studera två grönfinkshanars beteende när

de skulle äta vid en gemensam matskål kunde jag fastställa deras rang. Vid matskålen vek

40 Lindström

nämligen hanar med låg rang undan för dominanta hanar. Under infektionen förlorade inte

de infekterade grönfinkshanarna i rang gentemot oinfekterade hanar. Däremot fanns det ett

samband mellan en hanes rang och de viruskoncentrationer de fick efter en infektion.

Tidigt under infektionen (1-3 dagar) hade hanar med hög rang höga viruskoncentrationer i

blodet. Sent under infektionen (4-7 dagar) hade istället de lågt rankade hanarna höga

viruskoncentrationer. Av detta försök kan man dra slutsatsen att en hanes sociala status

verkade påverka hans förmåga att kontrollera en infektion. Hanar med låg rang var de som

var bäst på att motstå infektionen i början. Detta kan bero på att deras immunförsvar redan

var mobiliserat för att vara redo att bekämpa eventuella skador. Högt rankade hanar blev

snabbast virusfria. Detta kan bero på att högrankade hanar var i bättre kondition, och när

immunförvaret väl mobiliserats så hade de inga problem med att hantera infektionen.

Många faktorer kan avgöra om en individ kommer att drabbas av en infektion. För

Sindbisviruset visade det sig att vissa arter av trastar var mer drabbade än övriga fåglar. En

möjlig orsak till detta mönster kan vara att myggor söker sig till trastar för att hitta sin

föda. Det sätt som en infektion sprids har ofta stor inverkan på dess spridningsmönster.

Parasiter kan påverka sitt värddjurs överlevnad på flera sätt. Även infektioner som inte är

dödliga kan få fundamentala konsekvenser för värddjuren om det visar sig att infekterade

djur har svårare att undkomma rovdjur. Hos grönfinkar visade det sig att storleken på

hanens ornament kunde användas för att förutsäga hans förmåga att motstå en virus

infektion. Detta innebär att honor kan använda ornamentens storlek för att välja ut hanar

med god motståndskraft till partners. Förhöjda halter testosteron hade inte någon entydigt

negativ effekt på infektionsförloppet. Dessutom var en hanes rang relaterat till hans

förmåga att kontrollera virusinfektionen. Det är möjligt att sociala interaktioner är viktiga

för att upprätthålla sambandet mellan uttrycket av hanliga ornament och resistensförmåga.

Att kunna försvara sig mot parasitangrepp är en egenskap som är nödvändig för att

kunna överleva i naturen. Genom att studera hur försvaret mot parasiter formas av

evolutionen kan vi förstå vilka processer som måste få verka för att värddjur ska kunna

upprätthålla ett naturligt skydd mot parasitangrepp.

Documents

Bird-Parasite Interactions - DiVA portal165972/FULLTEXT01.pdf · tests of the Hamilton-Zuk hypothesis, have also received criticism due to the somewhat opportunistic approaches ecologists