Ibp

AIa

b

c

a

AAA

KIPAI

0d

Applied Animal Behaviour Science 138 (2012) 28– 35

Contents lists available at SciVerse ScienceDirect

Applied Animal Behaviour Science

journa l h o me pag e: www.elsev ier .com/ locate /applan im

nfluence of immunisation against GnRF on agonistic and mountingehaviour, serum testosterone concentration and body weight in maleigs compared with boars and barrows

nna-Katharina Albrechta,∗, Elisabeth grosse Beilageb, Ellen Kanitzc, Birger Puppec,mke Traulsena, Joachim Krietera

Institute of Animal Breeding and Husbandry, Kiel, Olshausenstr. 40, 24098 Kiel, GermanyField Station for Epidemiology, Buescheler Str. 9, 49465 Bakum, GermanyLeibniz Institute for Farm Animal Biology, Wilhelm-Stahl-Allee 2, 18196 Dummerstorf, Germany

r t i c l e i n f o

rticle history:ccepted 12 February 2012vailable online 24 March 2012

eywords:mprovaciglet castrationgonistic behaviour

mmunisation

a b s t r a c t

The aim of the present study was to assess the effects of vaccination against boar taint ontestosterone levels, body weight and behaviour in male fattening pigs when compared withboars and barrows. The study was carried out in two replicates over time. Pigs were allocatedto fattening pens according to treatment group (immunised males (IM), castrated males(CM), entire males (EM)) with 10 and 13 pigs per pen in batches one and two, respectively.IM pigs were vaccinated twice with ImprovacTM according to manufacturer’s instructions(first injection was given on the day of entry to the fattening unit, second injection wasperformed when pigs had an average weight of 76.6 kg and 80.4 kg in the two replicatesover time, respectively). During fattening, pigs were videotaped twice (test periods oneand two) for a consecutive period of 96 h during each test period and blood samples weretaken. Live weight measurements were conducted on a weekly basis. Before second vac-cination was given (test period one), IM pigs had serum testosterone levels comparableto those of entire boars. Two weeks after second ImprovacTM-injection (test period two)testosterone levels of IM pigs decreased significantly (p < 0.05). Agonistic as well as mount-ing behaviour of IM pigs was comparable to the behaviour of EM pigs during test period one.Two weeks after second ImprovacTM-injection fight time as well as fight duration decreasedsignificantly (p < 0.05) in the ImprovacTM-treated group to levels comparable to those of CMpigs, whereas both parameters remained on high levels in EM pigs. The number of fights,however, appeared not to be affected by vaccination with ImprovacTM, as IM and EM pigsrevealed higher numbers than CM pigs during both test periods. IM and EM pigs showedcomparable numbers of mountings as well as mounting time and mounting duration dur-ing test period one. Second ImprovacTM-injection was followed by a significant decrease inall three parameters in IM pigs (p < 0.05). When compared with the other treatment groupsCM pigs performed least agonistic and mounting behaviour during both test periods. IMand EM pigs performed significantly more manipulatory behaviour than CM pigs before

second ImprovacTM-injection is given. During test period two EM showed more manipula-tory behaviour than CM pigs with IM pigs in between. From our results we conclude thatimmunisation against 49 GnRH has the potential to reduce agonistic as well as mountingbehaviour in male fatteniwelfare when compared w

∗ Corresponding author.E-mail address: aalbrecht@tierzucht.uni-kiel.de (A.-K. Albrecht).

168-1591/$ – see front matter © 2012 Elsevier B.V. All rights reserved.oi:10.1016/j.applanim.2012.02.019

ng pigs to very low levels which may result in improved animalith fattening of entire boars.

© 2012 Elsevier B.V. All rights reserved.

al Beha

A.-K. Albrecht et al. / Applied Anim

1. Introduction

In most European countries it is common practice tosurgically castrate male piglets in order to prevent boartaint, the unpleasant sex odour developed by some malepigs after reaching puberty. Castration is usually done with-out anaesthesia and analgesia. Scientific as well as publicconcern, mostly on the grounds of animal welfare, haveincreased as this procedure is painful (Prunier et al., 2005,2006; Horn et al., 1999) and reduces growth performancein male pigs (Dunshea et al., 1993; Campbell et al., 1985).As a consequence many of alternative methods have beenproposed over the last years as reviewed by Prunier et al.(2006). One possible solution to inhibit testicular func-tion is down-regulation of the hypothalamic-pituitary-gonadal axis by vaccinating the pig against endogenousGonadotropin Releasing Factor (GnRF) using ImprovacTM

(Pfizer Animal Health). The effectiveness of ImprovacTM

in preventing boar taint as well as the possibility of rear-ing vaccinated pigs with superior growth performance andleaner carcasses compared to barrows has been shownin many studies (Skrlep et al., 2010; Fuchs et al., 2009;Dunshea et al., 2001). Regarding animal welfare, however,the behavioural consequence of castration using vacci-nation should be considered, as ImprovacTM-treated pigscan be regarded as entire males until the second vacci-nation is administered. Entire boars have higher levels ofthe male sex hormone testosterone, which in turn affectsbehaviour (Signoret, 1976). Other studies have shown thatboars are more likely to show aggressive (Cronin et al.,2003) and sexual behaviour, such as mounting (Fredriksenet al., 2004). This may result in higher levels of stressand injuries compromising animal welfare and decreas-ing growth performance. To date few studies have focusedon social and aggressive behaviour of ImprovacTM-treatedpigs (Baumgartner, 2010; Rydhmer et al., 2010).

The objective of this study was to evaluate the effectsof using ImprovacTM on agonistic and mounting behaviourin male pigs compared with boars and barrows rearedin groups. Additionally the intensity of manipulatorybehaviour within the groups was recorded and differencesbetween the treatment groups were analysed. Furthermorethe effects of ImprovacTM on serum testosterone concen-trations as well as body weight were examined.

2. Material and methods

The study period comprised two test periods dur-ing which body weight measurements, analysis of serumtestosterone levels and video observation were carriedout. The first test period started 4 days before secondImprovacTM-injection, the second test period was carriedout starting 13 days after the second vaccination. The studyperiod ended after the second test period and pigs wereslaughtered within 4 weeks after the end of study at acommercial abattoir.

2.1. Animals and handling

The study was carried out in two consecutivebatches consisting of 33 and 39 male pigs ((Large

viour Science 138 (2012) 28– 35 29

White × Landrace) × Pietrain). All animals were farrowedat the Institute’s own research farm Hohenschulen. Onstudy day 0, when equal to or less than 7 days old, pigswere randomly assigned to one of three study groups:ImprovacTM-vaccinated males (IM), surgically castratedmales (CM) and entire males (EM). Surgical castrationwas performed immediately after allocation to treatmentgroups and according to German regulations. Pigs in groupsIM and EM were left entire. After a suckling period of 28days, pigs were moved to the nursery unit and allocatedto two pens per treatment group blocked on weight (lightand heavy). All pens were provided with partially slattedfloor, one feeder providing space for seven to eight pigs,and eleven to twelve pigs per pen (eight to nine studypigs plus females). After an average nursery period of 6weeks surplus pigs were removed from the study and theremaining study pigs were moved to the fattening unitand treatment groups were housed in separate pens of3 m × 9 m with 11 and 13 pigs per pen in batches one andtwo, respectively. At the start of the fattening period sur-plus pigs were removed from the study and the two pensper treatment from the nursery unit were put togetherin one pen so as to form one group per treatment withan “inner-pen-recognition”/“degree of familiarity” of 50%among the pigs at the start of fattening. Every pen was pro-vided with partially slatted floor, one feeder which pigs hadaccess to from 07:00 to 00:00 and two drinking nipples. Allpigs were ear-taged with individual transponders so as toensure individual recognition at the feeder. Throughout thewhole fattening period feed (pellets) was offered restrict-edly, though on high levels. All pens included a playingdevice fixed to one edge of the pen wall.

2.2. ImprovacTM-treatment

Treatment with ImprovacTM was performed twice perbatch and administered subcutaneously behind the ear.The first injection was given on the day of entry to thefattening unit when pigs were 10–11 weeks old. The datefor the second vaccination was calculated from the desiredbody weight at slaughter of approximately 110 kg and per-formed when pigs had an average weight of 76.6 and 80.4in batches one and two, respectively.

2.3. Measurements and sampling

2.3.1. TestosteroneBlood samples were taken from all animals by punc-

turing the V. jugularis externa directly before the secondimmunisation was administered (test period one) andtested for serum testosterone levels. A second sam-ple was taken 13 days later (test period two), whentestosterone levels were expected to be on very lowlevels in ImprovacTM-treated pigs.Sample taking was car-ried out from 08:00 to 08:45 on every test day inorder to minimise the impact of diurnal variation inserum testosterone levels in the pig (Ellendorf et al.,

1975; Claus and Gimenez, 1977). Serum testosteroneconcentrations were analysed in duplicates using acommercially available 125I-RIA kit (DSL-4000, BeckmanCoulter GmbH, Sinsheim, Germany) according to the

3 al Behaviour Science 138 (2012) 28– 35

muaa1tc

2

sFss

2

i9tFiimpwmF

dovotacTwvcad

wiBPdow

2

sI(tcUf

Table 1Definition of behavioural traits.

Behavioural trait Unit Definition

Agonistic interaction Fight or displacement withphysical contact initiatedby one pig and featuringaggressive behaviouralelements, followed by anyform of submissivebehaviour performed bythe opponent (Langbeinand Puppe, 2004)

Fight Number of (n) Number of all agonisticinteractions per pig inwhich the pig was involvedduring one observationperiod (96 h)

Fighting time seconds (s) Total time a pig wasinvolved in agonisticinteractions during oneobservation period (96 h)

Fight duration seconds (s) Average fight duration perpig during one observationperiod (96 h)

Mounting behaviour Placing both front hoofs onthe back of a pen mate

Mounting Number of (n) Number of all mountingevents per pig in which thepig was involved duringone observation period(96 h)

Mounting time seconds (s) Total time a pig wasinvolved in mountingevents during oneobservation period (96 h)

Mounting duration seconds (s) Average mountingduration per pig during oneobservation period (96 h)

Manipulatorybehaviour

Rubbing belly of a penmate, nibbling, suckling orchewing on ear, tail orother part of the body of apen mate

Manipulation Number of (n) Number of allmanipulatory behavioursper treatment groupduring one observationperiod (96 h)

Manipulation time seconds (s) Total time spend on

0 A.-K. Albrecht et al. / Applied Anim

anufacturer’s guidelines. Cross-reactivities of antibodysed to 5�-dihydrotestosterone, 11-oxotestosterone andndrostenedione were 5.8%, 4.2% and 2.3%, respectively,nd to any further competing serum steroids lower than%. The assay was validated for use with porcine serum. Theest sensitivity was 0.01 ng/ml, and intra- and interassayoefficients of variation were 1.2% and 7.3%.

.3.2. Body weightLive weight measurements of each pig were taken on

tudy day 0 and on the day of allocation to nursery pens.rom the start of the fattening period live weight mea-urements were taken on a weekly basis until the end oftudy.

.3.3. Behavioural traitsAll behavioural traits were examined using videotap-

ng during the two test periods for a consecutive period of6 h each. Data recording included number, duration, ini-iator, recipient and outcome of the agonistic interaction.or mounting behaviour only duration, initiator and recip-ent were recorded. An agonistic interaction was recordedf the fight took longer than 3 s, mounting or attempted

ounting was recorded if both front hoofs of the mountingigs were lifted off the ground for longer than 1 s. If thereas an intervening period of more than 8 s, a new fight/ounting was considered to have started (Puppe, 1998).

or a more detailed ethogram, see Table 1.For manipulatory behaviour the number as well as the

uration were recorded and analysed on per-pen basis asbservation of individuals was too time consuming duringideo analysis. All video recording was performed usingne wide-angled camera located above each pen oppositeo the feeding station. Day and night time video taping wasssisted by artificial lighting located on the roof above theentre aisle. The derived data were stored in a database.he time points for the start of the observation periodsere calculated from the date of second ImprovacTM-

accination. The first video recording (test period one) wasarried out during the 4 days prior to second vaccinationnd the second recording (test period two) was started 13ays after second ImprovacTM-injection.

For the time of video recording all pigs in the penere spray painted with individual numbers on the back

n order to make recognition of individual pigs possible.ehavioural traits were recorded by two trained observers.rior to visual assessment, the observers were trained inefinition and identification of the behavioural traits. Inter-bserver-correlation for a representative video sequenceas >0.9 at the end of training.

.4. Data analysis

All statistical analysis was performed using the SAStatistical software package, version 9.1 (SAS Institutenc., 2005). The distribution of the behavioural traitsfight, fighting time, fight duration, mounting, mounting

ime, mounting duration) as well as testosterone con-entrations and body weights was analysed using theNIVARIATE-procedure (skewness and kurtosis ranged

rom 1.43 to 4.75 and −1.45 to 13.72, respectively).

manipulatory behaviourduring one observationperiod (96 h)

Due to log-transformation, the data were approximatedto normal distribution (skewness and kurtosis rangedfrom −0.7 to 0.49 and −1.45 to 1.11, respectively). Asserum testosterone concentration below the detectionlevel occurred frequently among the barrows and theImprovac-treated pigs at second sample taking (valueswere recorded as 0.0 ng/ml), the data set was transformedusing y = log(observation + 1). Further analyses were basedon transformed data.

An analysis of fixed effects was performed using theMIXED procedure (maximum likelihood estimation) (SAS,2005). Fixed effects of batch (batch one and two), treat-

ment group (IM, CM, EM), test period (test period one andtwo) and the interaction of treatment group and test periodwere included in the model as well as random effects foranimal within batch and treatment group.

A.-K. Albrecht et al. / Applied Animal Behaviour Science 138 (2012) 28– 35 31

Table 2p-Values of fixed effects for behavioural parameters.

Batch Treatment Test period Treatment × test period

Fights per pig (n) 0.6794 0.0238 0.0012 0.8791Fight time per pig (s) 0.0015 <0.0001 <0.0001 <0.0001Fighting duration (s) 0.0101 0.0488 <0.0001 <0.0001Mounts per pig (n) 0.8866 <0.0001

Mounting time per pig 0.916 <0.0001

Mounting duration 0.9021 0.0004

0

1

2

3

4

5

6

7

8

9

21test period

seru

m t

esto

ste

ron

e l

evels

(ng

/ml)

IM CM EM

a

a

bb

a

b

Fig. 1. LS mean serum testosterone levels of immunised males (IM), cas-

trated males (CM) and entire males (EM) for test periods one and two.Data are presented as retransformed LS mean values. a,b – different lettersindicate significant differences within testing (p < 0.05).

Manipulatory behaviour was analysed on per-pen basisusing a nonparametric test. The Wilcoxon rank-sum test(proc NPAR1WAY, SAS, 2005) was applied to identify signif-icant differences between treatment groups in the numberof and total time spent in manipulatory behaviours. Thetotal number of fights and total fighting time for the 4days within one test period were calculated and differencesbetween first and second test period were analysed.

All treatment differences were assessed at the two-sided 5% level of significance.

3. Results

3.1. Serum testosterone levels

Only the interaction of test period and treatmentgroup had a significant effect on serum testosterone levels(p < 0.05). ImprovacTM-treated males (IM) and entire males

(EM) showed significantly higher levels than castratedmales (CM) in test period one, prior to second ImprovacTM

injection (p < 0.05) (Fig. 1). Two weeks after the immunisa-tion (test period two), IM showed a significant decrease in

Table 3LS means and standard errors for behavioural traits as observed in immunised maand two.

Test period one

IM CM EM

Fights per pig (n) 18.0 (1.29) 7.1 (1.29) 15.3 (Fight time per pig (s) 495.9a (1.29) 51.6b (1.29) 380.0Fighting duration (s) 25.3a (1.33) 7.9b (1.33) 23.3a

Mounts per pig (n) 23.8a (1.23) 6.8b (1.27) 41.6a

Mounting time per pig 249.1a (1.40) 17.0b (1.43) 316.6Mounting duration 11.0a,c (1.28) 2.6b (1.29) 7.5c (1

Different superscript letters indicate significant differences within row (p < 0.05).

<0.0001 0.0381<0.0001 <0.0001

0.08 <0.0001

testosterone levels (p < 0.05) whereas testosterone in EMremained on high levels. Most of the castrated males hadtestosterone levels below the detection mark in both testperiods.

3.2. Body weight

Treatment group and test period had a significant influ-ence on body weight (p < 0.05) whereas the interaction ofthe two effects proved not to be significant (p > 0.05).

3.3. Behavioural parameters

Influences of fixed effects as included in the model arepresented in Table 2.

ImprovacTM-treated males (IM) and entire males (EM)had a significantly longer total fighting time per pig thancastrated males (CM) in test period one (Table 3). For allthree treatment groups a decrease in total fighting timeper pig from test period one to test period two couldbe observed, although differences only proved to be sig-nificant in ImprovacTM-treated males (IM) (p < 0.05). Thisresulted in a comparable total fighting time per pig inImprovacTM-treated males (IM) and castrated males (CM)in test period two, whereas the total fighting time perpig remained significantly higher in entire males (EM)(p < 0.05). In test period one ImprovacTM-treated males(IM) and entire males (EM) revealed significantly higherfight duration than castrated males (CM) (Table 3). Animalsfrom the ImprovacTM-treated group (IM) in test period twodecreased their fight duration to levels significantly lowerthan in test period one (p < 0.05) whereas castrated males(CM) and entire males (EM) remained on levels comparableto test period one (p > 0.05).

In test period one, castrated males showed significantlylower fight durations than the other treatments (p < 0.05)(Table 3). Although all three treatment groups revealedlower fight durations in test period two when compared to

les (IM), castrated males (CM) and entire males (EM) for test periods one

Test period two

IM CM EM

1.29) 12.1 (1.29) 5.4 (1.30) 10.9 (1.31)a,c (1.28) 79.4b (1.28) 38.0b (1.29) 202.5c (1.30)(1.33) 6.2b (1.33) 7.7b (1.35) 17.3a (1.36)(1.22) 6.1b (1.31) 3.4b (1.22) 32.0a (1.21)a (1.37) 16.0b (1.49) 9.5b (1.38) 159.1a (1.38).27) 2.7b (1.32) 2.9b,c (1.28) 14.0a (1.28)

Data are presented as retransformed LS mean values.

32 A.-K. Albrecht et al. / Applied Animal Behaviour Science 138 (2012) 28– 35

70

75

80

85

90

95

21

test period

bo

dy w

eig

ht

(kg

)

IM CM EM

F(

ti

wv(oiwtt((t(wmWmtst

ra

boiosctbilt(atd(

4

l

0

200

400

600

800

1000

1200 (3)

(4)

1 2

test period

tota

l n

um

ber

of

man

ipu

lati

ve

acti

on

s/ tr

eatm

en

t

IM CM EM

aa

a

c,d

b,d

c

0

5000

10000

15000

20000

25000

30000

35000

1 2

test periodto

tal

tim

e s

pe

nt

on

ma

nip

ula

tiv

e

ac

tio

n/

tre

atm

en

t

IM CM EM

a

a

b,d

a

b

c,d

Figs. 3 and 4. Total number of and total time spent on manipulative

ig. 2. LS mean body weight of immunised males (IM), castrated malesCM) and entire males (EM) for test periods one and two.

est period one, the difference only proved to be significantn ImprovacTM-treated males (IM) (p < 0.05).

A comparison of the number of mountings per pig asell as total mounting time revealed significantly higher

alues for Improvac-treated males (IM) and entire malesEM) than surgically castrated males (CM) in test periodne (p < 0.05) (Table 3). Both parameters decreased signif-cantly in IM pigs from first to second test period (p < 0.05)

hich resulted in comparable number of mountings andotal mounting time for Improvac-treated (IM) and cas-rated males (CM) in test period two whereas entire malesEM) performed significantly more mounting behaviournumber of mountings and total mounting time) thanhe other treatment groups during the second test periodp < 0.05). During test period one the mounting durationas longer for Improvac-treated males (IM) than castratedales (CM) with entire males (EM) in between (Table 3).hereas entire males (EM) showed a significant increase inounting duration in test period two (p < 0.05), Improvac-

reated males revealed a significant decrease (p < 0.05) andhowed mounting durations comparable to those of cas-rated males (CM) (Fig. 2).

All three parameters related to mounting behaviourevealed a significant decrease in ImprovacTM-treated pigsfter the second immunisation is given (p < 0.05).

Figs. 3 and 4 illustrate the total number of manipulatoryehavioural actions per pen as well as the total time spentn manipulating pen mates for each treatment group dur-ng the two test periods. At first videotaping (test periodne) Improvac-treated males (IM) and entire males (EM)howed significantly more manipulatory behaviour thanastrated males (CM). In ImprovacTM-treated pigs (IM)he total number as well as the total time spent on theseehavioural traits decreased significantly (p < 0.05) after

mmunisation two and both parameters were significantlyower than in boars at second videotaping (test periodwo) but still on higher levels than in surgical castratesCM). Manipulatory behaviour in boars (EM) even showed

tendency for increasing numbers whereas the totalime spent on manipulating pen mates revealed a smallecrease from first (test period one) to second videotapingtest period two).

. Discussion

Although the circadian rhythm of serum testosteroneevels in the pig has been the subject of a number of

actions, as observed in the three treatment groups for test periods one andtwo. Data are presented as retransformed LS mean values. a,b – differentletters indicate significant differences within testing (p < 0.05).

studies (Ellendorf et al., 1975; Claus and Gimenez, 1977;Edquist et al., 1980), no general agreement on the timeof peaks and low points of testosterone concentrationin the course of the day exists. Blood samples in thepresent study were taken at 08:00 on both days to over-come variations in daily testosterone concentrations andto minimise the effect of daytime. Entire males presenteda wide range of individual testosterone concentrations(2.78–12.47 ng/ml and 1.11–15.08 ng/ml on test periodsone and two, respectively). The serum testosterone con-centrations of ImprovacTM-treated males were comparableto those of entire males in test period one, but showed asignificant decrease after second immunisation was given.Our results are in line with findings presented by otherauthors. Zamaratskaia et al. (2007, 2008) and Dunshea et al.(2001) found that serum testosterone concentrations ofImprovacTM-treated pigs decreased to low levels within 2weeks of second immunisation is administered and thatthe change in the hormonal profile alters the behaviour inthe pig.

Entire as well as ImprovacTM-treated male pigs per-formed more agonistic behaviour than surgical castratesat the time point of second vaccination (test period one).These findings are in line with results presented by otherauthors (Rydhmer et al., 2010; Cronin et al., 2003). Atsecond test period less aggressive behaviour was observedin all three treatment groups. These results confirm that

the level of aggression decreases with time if groupsremain stable and unchanged (Rydhmer et al., 2006; Fraseret al., 1995). Fraser et al. (1995) argued that aggressionis linked to dominance hierarchies and once a hierarchy

al Beha

A.-K. Albrecht et al. / Applied Anim

is established, the number as well as the time spenton agonistic interactions decreases and that long-termaggression consists of brief threats and bites that onlyoccasionally escalate into more severe and longer fights(Fraser, 1984). The decrease in total fighting time as wellas fight duration was, however, much more pronounced inImprovacTM-treated males than in castrated or intact maleswhich suggests that the second ImprovacTM-injection hasan influence on fighting behaviour in pigs as far as fighttime and duration are concerned. The total number ofagonistic interactions on the other hand, seemed not tobe affected by immunisation, since ImprovacTM-treatedmales at second test period still fought at frequenciescomparable to those of boars. Fraser (1984) argues thatlong-term aggression often involves competition forspecific resources. In addition other authors suggest thatvaccinated pigs increase their feed intake disproportion-ally after second ImprovacTM-injection when comparedwith boars and barrows. Therefore, the limited access tofeed (restrictive feeding, only one feeder per pen) mighthave offered a constant source for competition in thepresent study. This might account for the markedly highernumber of fights in the ImprovacTM-treated group whencompared with surgical castrates during the second testperiod, although mean fight duration is at comparablelevels in both treatment groups.

In contrast to findings from other studies (Salmon andEdwards, 2006; Rydhmer et al., 2006; Giersing et al., 2000)no coherent relationships between aggressive behaviourand testosterone levels or body weight were found in thepresent study (unpublished results). One possible expla-nation might be the relatively small number of pigs andthe wide range in testosterone levels. Most authors, how-ever, agree that aggressive behaviour is not only influencedby gonadal hormones but that other factors are proba-bly even more important (Ruiz-de-la-Torre and Manteca,1999; Dixson and Herbert, 1976). Stability of the group aswell as group size, have influence on the level of aggressivebehaviour in the group (Rydhmer et al., 2006; Nielsen et al.,1995). In addition, Rushen (1985) found that animals assesstheir own fighting ability and that pigs learn from previ-ous social encounters and alter their behaviour dependingon the outcome (Rushen and Pajor, 1987). Moreover, it isargued that especially fight time and duration are not onlyaffected by testosterone levels but highly depend on weightasymmetries in the group. Andersen et al. (2000) foundlonger fight durations in groups with a smaller weightrange than in groups where pigs showed higher differencesin body weight. These findings may explain why correla-tions in the present study were incoherent or non-existent.Additionally, aggressive behaviour was not tested over aspecific resource. Since other authors assume that pigsform different hierarchies for different resources and thataggression is resource-related (Lindberg, 2001), behaviourof pigs in the present study might have been influenced byfactors not considered in the calculation.

On the one hand, mounting is part of pigs’ normal sex-

ual behaviour repertoire (Hemsworth and Tillbrook, 2007;Levis et al., 1997) but on the other also represents a meansto dominate low-ranking animals in order to attain accessto resources (Fredriksen et al., 2008). In the present study

viour Science 138 (2012) 28– 35 33

mounting behaviour was seen frequently in entire malepigs during both test periods. Whereas the number ofmountings as well as total mounting time and mountingduration decreased significantly in the ImprovacTM-treatedgroup after second immunisation, entire males maintainedhigh frequencies of mounting until the end of study. Ford(1990) suggested that sexual behaviour, such as mount-ing, is stimulated by testicular steroid hormones. Resultsof the studies conducted by Rydhmer et al. (2006) andFredriksen et al. (2004) confirm that entire males showmore sexual behaviour at the end of the fattening periodwhen compared to gilts or barrows. The low levels of allparameters related to mounting in surgical castrates andImprovac-treated pigs are in line with findings presentedby Cronin et al. (2003) and Rydhmer et al. (2010) andsuggest that mounting behaviour is strongly affected bycastration regardless of time point or method. Mountingbehaviour has detrimental consequences for animal wel-fare. Rydhmer et al. (2006) found that most skin lesionsand scratches on boars in single-sex pens were causedby mounting rather than aggressive behaviour whereasno relationship was found between number of scratchesand received aggression. Other authors also found signif-icantly more skin lesion in boars than in gilts or surgicalcastrates (Velarde et al., 2007; Rydhmer et al., 2006). Addi-tionally, mounting is suspected to cause leg problems(Rydhmer et al., 2006) and to impose high stress levelsand fear on frequently mounted pigs (Fabrega et al., 2010)which in turn results in reduced animal welfare (Björklundand Boyle, 2006). The positive effects on animal welfareobtained in the pigs’ early life by abandoning surgical cas-tration may therefore be impaired by increased levels ofaggression and mounting which results in stress and injury.Björklund and Boyle (2006) found that boars show evenhigher levels of aggression and mounting behaviour whenreared in single-sex groups than boars reared in mixed-sexgroups with females. Split marketing may even aggravatethe problem as the removal of animals causes disrup-tion in the dominance hierarchy and imposes high stresslevels and an increased risk of injury on the remainingpigs (Björklund and Boyle, 2006; Fredriksen and Hexeberg,2009).

Another factor influencing animal welfare is the manip-ulation of pen mates. There is limited research examiningthe effects of non-violent social interaction on pig welfare.From our results we conclude that castration influencesmanipulatory behavioural patterns. However, the tim-ing of castration or the castration technique seems toimpair the positive effects of castration on intensity ofmanipulatory behaviour. Entire male pigs as well as vacci-nated pigs before second ImprovacTM-injection performedsignificantly more manipulatory behaviour than surgicalcastrates. After second injection ImprovacTM-treated pigsreduced manipulatory behaviour to intermediate levelswhen compared with the other treatment groups. Themanipulation of pen mates at high frequencies may causestress for the receiver if pigs are hindered from resting or

feeding or in extreme cases, may lead to more injuries, andtherefore should be taken into consideration when assess-ing animal welfare consequences of surgical castration orits alternatives.

3 al Beha

5

nriawpbratenb(ammopebdw

R

A

B

B

C

C

C

D

D

D

E

E

F

4 A.-K. Albrecht et al. / Applied Anim

. Conclusion

From our results we conclude that the second immu-isation against GnRF, which effectively castrates the pig,educes aggressive and sexual behaviour in male fatten-ng pigs to levels comparable to those of surgical castratesnd may therefore be associated with improved animalelfare when compared with rearing of entire boars. Theositive effects of using ImprovacTM regarding agonisticehaviour may even be improved by adjusting feedingegimes to the voluntarily higher feed intake of vaccinatesfter second ImprovacTM-injection. As far as manipula-ory behaviour is concerned, further investigations on theffects of vaccination, especially timing of second vacci-ation, are needed, since pigs in the present study, haveeen re-vaccinated relatively early in the fattening periodapproximately 5 weeks before slaughter). Re-vaccinationt a later point in time may compromise the effects onanipulatory behaviour even further. However, in order toaximise positive effects related to growth performance

ther authors suggest to perform vaccination as late asossible (Lealiifano et al., 2009). This might have influ-nce on the behaviour and further investigation on theehavioural effects of late administration of the secondose of ImprovacTM is needed in order to ensure that pigelfare is not impaired by vaccination against GnRF.

eferences

ndersen, I.L., Andenaes, H., Boe, K.E., Jensen, P., Bakken, M., 2000. Theeffects of weight asymmetry and resource distribution on aggres-sion in groups of unacquainted pigs. Appl. Anim. Behav. Sci. 68 (2),107–120.

aumgartner, J., 2010. The behaviour of male fattening pigs followingeither surgical castration or immunisation with a GnRF vaccine. Appl.Anim. Behav. Sci. 124, 28–34.

jörklund, L., Boyle, L.A., 2006. Effects of finishing boars in mixed andsingle sex groups and marketing on pig welfare. Acta Vet. Scand. 48(Suppl. 1), P2.

ampbell, R.G., Taverner, M.R., et al., 1985. Effects of sex and energy-intake between 48-kg and 90-kg live weight on protein depositionin growing-pigs. Anim. Prod. 40 (June), 497–503.

laus, R., Gimenez, T., 1977. Diurnal rhythm of 5-alpha-androst-16-en-3-one and testosterone in peripheral plasma of boars. Acta Endocrinol.(Copenh.) 84, 200–206.

ronin, G.M., Dunshea, F.R., Butler, K.L., McCauley, L., Barnet, J.L.,Hemsworth, P.H., 2003. The effects of immuno- and surgical-castration on the behaviour and consequently growth of group-housed male finisher pigs. Appl. Anim. Behav. Sci. 81, 111–126.

ixson, A.F., Herbert, J., 1976. Testosterone, aggressive behaviour anddominance rank in captive adult male talapoin monkeys. Physiol.Behav. 18, 539–543.

unshea, F.R., King, R.H., Campbell, R.G., Sainz, R.D., Kim, Y.S., 1993.Interrelationships between sex and ractopamine on protein and lipiddeposition in rapidly growing pigs. J. Anim. Sci. 71, 2919–2930.

unshea, F.R., Colantoni, C., Howard, K., McCauley, I., Jackson, P., Long, K.A.,Lopaticki, S., Nugent, E.A., Simons, J.A., Walker, J., Hennessy, D.P., 2001.Vaccination of boars with a GnRH vaccine (Improvac) eliminates boartaint and increases growth performance. J. Anim. Sci. 79, 2524–2535.

dquist, L.-E., Einarsson, S., Larsson, K., Lundström, K., 1980. Diurnal vari-ations in peripheral plasma levels of testosterone, androstenone andcortisol in boars. Acta Vet. Scand. 21, 451–453.

llendorf, F., Parvizi, N., Pomerantz, D.K., Hartjen, A., König, A., Smidt, D.,Elsaesser, F., 1975. Plasma luteinizing hormone and testosterone inthe adult male pig: 24 hour fluctuations and the effect of copulation.

J. Endocrinol. 67, 403–410.

abrega, E., Velarde, A., Cros, J., Gispert, M., Suarez, P., Tibau, J., Soer, J., 2010.Effect of vaccination against gonodotroping-releasing hormone, usingImprovac, on growth performance, body composition, behaviour andacute phase proteins. Livest. Sci. 132, 53–59.

viour Science 138 (2012) 28– 35

Ford, J.J., 1990. Differentiation of sexual behaviour in pigs. J. Reprod. Fertil.40 (Suppl.), 311–321.

Fraser, D., 1984. The role of behaviour in swine production: a review ofresearch. Appl. Anim. Ethol. 11, 317–339.

Fraser, D., Kramer, D.L., Pajor, E.A., Weary, D.M., 1995. Conflict and cooper-ation: sociobiological principles and the behaviour of pigs. Appl. Anim.Behav. Sci. 44, 139–157.

Fredriksen, B., Hexeberg, C., 2009. The effect of removing animals forslaughter on the behaviour of the remaining male and female pigsin the pen. Res. Vet. Sci. 86, 368–370.

Fredriksen, B., Nafstad, O., Lium, B.M., Marka, C.H., Heier, B.T., Andresen,O., et al., 2004. Entire male pig production in ‘born-to-finishing-pens’.Project No. 6624. Norwegian Meat Research Centre.

Fredriksen, B., Lium, B.M., Marka, C.H., Mosveen, B., Nafstad, O., 2008.Entire male pigs in farrow-to-finishing pens – effects on animal wel-fare. Appl. Anim. Behav. Sci. 110, 258–268.

Fuchs, T., Nathues, H., Koehrmann, A., Andrews, S., Brock, F., Sudhaus, N.,Klein, G., grosse Beilage, E., 2009. A comparison of the carcase char-acteristics of pigs immunized with a ‘gonadotrophin-releasing factor(GnRF)’ vaccine against boar taint with physically castrated pigs. MeatSci. 83, 702–705.

Giersing, M., Lundström, K., Andersson, A., 2000. Social effects and boartaint: significance for production of slaughter boars (sus scrofa). J.Anim. Sci. 78, 296–305.

Hemsworth, P.H., Tillbrook, A.J., 2007. Sexual behaviour of male pigs.Horm. Behav. 52, 39–44.

Horn, T., Marx, G., von Borell, E., 1999. Behaviour of piglets duringcastration with and without a local anaesthesia. Deut. Tierärztl.Wochenschr. 106, 271–274.

Lealiifano, A.K., Pluske, J.R., Nicholls, R.R., Dunshea, F.R., Mullan, B.P., 2009.Altering the timing of an immunocastration vaccine to optimise pigperformance. In: van Barneveld, R.J. (Ed.), Manipulating Pig ProductionXII. Australasian Pig Science Association, Werribee, p. 184.

Levis, D.G., Ford, J.J., Christenson, R.K., 1997. An evaluation of three meth-ods for assessing sexual behaviour in boars. J. Anim. Sci. 75 (2),348–355.

Lindberg, C.A., 2001. Group life. In: Keeling, L.J., Gonyou, H.W. (Eds.), SocialBehaviour in Farm Animals. CABI Publ., Wallingford, pp. 37–58.

Nielsen, B.L., Lawrence, A.B., Whittemore, C.T., 1995. Effect of group sizeon feeding behaviour, social behaviour, and performance of growingpigs using single-space feeders. Livest. Prod. Sci. 44, 73–85.

Prunier, A., Mounier, A.M., Hay, M., 2005. Effects of castration, tooth resec-tion, or tail docking on plasma metabolites and stress hormones inyoung pigs. J. Anim. Sci. 83, 216–222.

Prunier, A., Bonneau, M., von Borell, E.H., Cinotti, S., Gunn, M., Fredriksen,B., Giersing, M., Morton, D.B., Tuyttens, F.A.M., Velarde, A., 2006. Areview of the welfare consequences of surgical castration in pigletsand the evaluation of non-surgical methods. Anim. Welfare 15 (3),277–289.

Puppe, 1998. Effects of familiarity and relatedness on agonistic pair rela-tionships in newly mixed domestic pigs. Appl. Anim. Behav. Sci. 58(3–4), 233–239.

Ruiz-de-la-Torre, J.L., Manteca, X., 1999. Effects of testosterone on aggres-sive behaviour after social mixing in male lambs. Physiol. Behav. 68,109–113.

Rushen, J., 1985. Explaining peck order in domestic chickens. Bird Behav.6, 1–9.

Rushen, J., Pajor, E., 1987. Offence and defence in fights between youngpigs (sus scrofa). Aggr. Behav. 13, 329–346.

Rydhmer, L., Zamaratskaia, G., Andersson, H.K., Algers, B., Guillemet, R.,Lundström, K., 2006. Aggressive and sexual behaviour of growing andfinishing pigs reared in groups, without castration. Acta Agric. Scand.Sect. A: Anim. Sci. 56, 109–119.

Rydhmer, L., Lundström, K., Andersson, K., 2010. Immunocastrationreduces aggressive and sexual behaviour in male pigs. Animal 4 (6),965–972.

Salmon, E.L., Edwards, S.A., 2006. Effects of gender contact on thebehaviour and performance of entire boars and gilts from 60 to 130 kg.In: Proceedings British Society of Animal Science, p. 72.

SAS, 2005. Version 9.1. SAS Institute, Cary, NC, USA.Signoret, J.P., 1976. Influence of anabolic agents on behaviour. Environ.

Quality Saf. Suppl. 5, 143–150.Skrlep, M., Segula, B., et al., 2010. Effect of immunocastration (Improvac

(R)) in fattening pigs. I: Growth performance, reproductive organs and

malodorous compounds. Slov. Vet. Res. 47 (2), 57–64.

Velarde, A., Gispert, M., Oliver, M.A., Soler, J., Tibau, J., Fabrega, E., 2007.The effect of immunocastration on the behaviour of pigs. In: Proceed-ings of the 41st International Congress of the International Society forApplied Ethology, 8–12 August 2007, Merida, Mexico, p. 117.

al Beha

A.-K. Albrecht et al. / Applied Anim

Zamaratskaia, G., Rydhmer, L., Andersson, H.K., Chen, G., Lowagie, S.,Andersson, K., Lundström, K., 2007. Long-term effect of vaccinationagainst gonadotropin-releasing hormone, using ImprovacTM, on hor-monal profile and behaviour of male pigs. Anim. Reprod. Sci. 108,37–48.

viour Science 138 (2012) 28– 35 35

Zamaratskaia, G., Andersson, H.K., Chen, G., Andersson, K., Madej, A.,Lundström, K., 2008. Effect of a gonadotropin-releasing HormoneVaccine (ImprovacTM) on steroid hormones, boar taint compoundsand performance in entire male pigs. Reprod. Domest. Anim. 43,351–359.