Chronic stressors, tonic immobility and leucocytic responses in the domestic fowl

Physiology & Behawor, Vol 46, pp 439--442 © Pergamon Press plc, 1989 Pnnted in the U S A

0031-9384/89 $3 00 + 00

Chronic Stressors, Tonic Immobility and Leucocytic Responses in the Domestic Fowl R. B R Y A N J O N E S A F R C Institute o f Animal Physiology and Genetics Research, Edinburgh Research Station Roslin, Midlothian EH25 9PS, Scotland, U.K. R e c e i v e d 24 January 1989

JONES, R. B. Chromc stressors, tomc tmmoblhty and leucocyttc responses m the domesticfowl PHYSIOL BEHAV 46(3) 439--442, 1989.--Tonic immobility (TI) fear reactions and heterophll/lymphocyte (H/L) ratios were measured in individually caged Brown Leghorn pullets before and after their exposure to one of three procedures lastmg approximately 72 hr These were" ad lib feeding (controls), fasting (removal of food) or frustratton of feeding (fasting plus denial of access to vistble food). H/L ratios were elevated after fasting or frustration, illustrating the stressful nature of these procedures, but remmned unaltered in the controls. Pretreatment TI responses were slrmlar in all groups. Susceptibility and response duration were reduced upon retestlng presumably through habituation, but tlus effect was relatively homogeneous across groups. Any direct effect of fastmg/frustraUon upon fearfulness seemed unlikely because the mean TI responses of control and stressed birds were smular. However, whereas a rank order based on TI durations was repeatable upon retestmg in the controls, it was destabihsed m the stressed groups. Posmve lntramdividual correlations were found between pretreatment TI duratmns and subsequent H/L responses to fasting/frustration. This suggests that leucocytic responses to chronic stressors may be greater in fearful pullets than in less fearful birds Chronic stress

Domestic fowl

Fear

Heterophd/lymphocyte ratio

Tonic immobility

avoidance of the experimenter (4). Tonic immobility (TI) is a fear-potentiated state of reduced responsiveness ehclted by physical restraint (11,18). It is thought to provide a useful measure of general, underlying fearfulness because close associations have been found between TI reactions and responsiveness to a variety of fear-eliciting situations m Japanese quail, domestic chicks and adult laying hens (18, 26, 29, 34). Chronic infusion of physiological levels of corticosterone via osmotic minipumps prolonged TI reactions and increased heterophil/lymphocyte (H/L) ratios in adult White Leghorn hens (21). It was consequently suggested that long-term elevation of plasma cortlcosterone concentrations may not only impair growth, reproductive capacity and leucocytic responsiveness, but might also predispose chickens to react more fearfully to alarming stimuli (21). Chronic fasting or frustration of feeding are known to elevate circulating cortlcosterone concentrations (1, 9, 31). Their effects on the TI responses and H/L ratios of Brown Leghorn pullets were, therefore, measured in the present study. The relationships between initial fearfulness, posttreatment TI reactions and leucocytic responsiveness were also assessed.

THE Farm Animal Welfare Council in the U.K. has recommended that livestock should be permitted certain basic freedoms including freedom from fear or distress, freedom from hunger and the freedom to display most normal, socially acceptable patterns of behaviour (35). Indeed, prolonged exposure to frightening stimulation, food deprivation or frustration are all commonly regarded as stressors which are likely to reduce welfare and performance. For example, their undesirable consequences in chickens may include energy wastage, physical injury, impaired growth and reproductive performance and the appearance of behavioural abnormalities (3,17). However, the possible existence and nature of any interrelationships between these potentially stressful states is unclear. Underlying fearfulness or temperament varies between chickens and fearful birds normally show more pronounced reactions to frightening stimuli than do their less fearful counterparts (8, 19, 26). Emotional and physiological responses to aversive stimulation are often closely integrated (6, 15, 22, 24) and recent findings have suggested that underlying fearfulness and adrenocortical actavataon may also be positively associated. Firstly, for instance, broilers showed prolonged tonic immobility fear reactions after transportation (5), a procedure known to elevate plasma corticosterone concentrations (10). Secondly, hens which consistently laid eggs with abnormal shells, a stress-related phenomenon (16), were more fearful than normal layers in a variety of situations (25). Thirdly, the adrenocortical response to capture and restraint was highest in those hens which showed the greatest home-cage

METHOD

Animals Used Forty-four Brown Leghorn pullets, bred at the Institute of Animal Physiology and Genetics Research, were used. They had

439

440

JONES

TABLE 1 LEUCOCYTIC PROFILES BEFORE AND AFTER A 72-HR REGIME OF EITHER AD LIB FEEDING, FASTINGOR FRUSTRATION OF FEEDING (MEANS _+ SEM) Ad Lib Feeding (n = 15)

Heterophfls (no) Lymphocytes (no) Heterophd/lymphocyte ratJo

Fasting (n 14)

Frustranon of Feeding (n = 15)

=

Pretreatment

Posttreatment

Pretreatment

Posttreatrnent

Pretreatment

Posttreatment

4 6 __. 0 3* 91 6 ± 0 5* 0051 ___ 0.004*

4 5 + 0 4* 91 8 ± 0.6* 0050 ± 0 005*

4 4 --- 0 4* 91 9 ± 0.5* 0 048 ± 0 004*

20.8 ± 2 0~ 72 6 --- 2.2t 0 301 --- 0.041t

4 9 - 0 5* 91 9 ± 0 7* 0054 ± 0006"

20 4 ± 1 7+ 72 9 ±- 1 9t 0 295 ± 0 039t

(no) = number Values with different superscripts are different at p < 0 01 been reared socially on wire until their transfer at 12 weeks of age to individual cages (45 × 30 × 45 cm) in the bottom uer of a three-tier battery system. The birds were randomly allocated to 1 of 3 treatment groups upon transfer to the battery and group members were caged consecutively. A continuous series of cages were subdivided into 6 blocks (4 blocks of 7 and 2 of 8). Each treatment thus occupied 2 separate blocks and posit~on of the block in the row was assigned at random. Four cages at each end of the series remained vacant and 2 empty cages separated each block. The poultry house accommodated approximately 500 other hens Food and water were provided ad lib until testing began, the photoperiod ran from 05.00 to 19.00 hr and light intensities recorded at the fronts of the cages were approximately 18 lux

Procedures Blood sampling and heterophd/lymphocyte ratios. At 17 weeks of age (Day 1) each bird was removed from its cage and restrained while 2 drops of blood were qmckly withdrawn from the wing vein and smeared onto each of 2 glass slides All blood samples were taken between 16.00 and 17.00 hr. The smears were stained within 2 hr of preparation with May-Grunwald-Glemsa stain One hundred leucocytes, including heterophils, lymphocytes, monocytes, basophils and eosinophils, were counted on each slide and the H/L ratios were determined by dividing the number of heterophils by that of lymphocytes. The two counts were added together and the average H/L ratios were calculated for each bird. This procedure was repeated later (see below). Tonic immobility testing. After a 3-day recovery penod the birds' TI reactions were measured. Each bird was carried approximately 10 m to a separate room and tested individually. Tonic immobility was mduced by placing the bird on its back in a U-shaped wooden cradle covered with cloth and restraining it for 15 sec (23). The observer sat nearby within sight of the bird and recorded the following indices: 1) the number of inducuons (15-sec periods of restramt) necessary to attain TI lasting at least 10 sec and 2) the duration of TI, i.e., until the bird righted itself. If the bird remained in TI throughout the 20-min observation period, a maximum score of 1200 sec was given for duration Tests were carried out between 08.30 and 15.30 hr and no time of day effects were apparent. Treatment regimes and subsequent testing. A further 4 days were allowed to elapse before the pullets were exposed to one of three treatments beginning at 09.00 hr and lasting approximately 72 hr. Firstly, 15 birds were allowed to feed ad lib as normal, these acted as controls. Secondly, 14 birds were deprived of food, i.e., their food hoppers were emptied. Thirdly, the feeding behavior of 15 pullets was frustrated by placing tight-fitting clear perspex covers over their food hoppers. This treatment combined fasting with exposure to inaccessible food. Treatment duration varied slightly with the order of subsequent testing, but each group was similarly treated.

The TI reactions of each bird were measured again between 08.00 and 15.00 hr on Day 11, i.e., immediately after the treatment periods. Blood samples were then taken at the same time of day (16.00 to 17.00 hr) as the prewous ones and H/L ratios were calculated. Data handling and analysts. The Kruskal-Walhs one-way analysis of variance was used to compare TI responses and H/L ratios across groups both before and after treatment, while treatment effects on TI and H/L were assessed within groups using the Wilcoxon matched-pairs, signed-ranks test, two-tailed (33). The birds were ranked within treatment groups according to their TI durations, both before and after treatment. The degree of association between the pre- and posttreatment TI ranks were estimated within groups using the Spearman rank correlation coefficient. Since fasted and frustrated pullets showed similar behavioural and leucocytic profiles, both before and after treatment, the data from these two groups was pooled for further analysis. The btrds were then ranked from low to hlgh, (Le., from 1 to 29), on each of the following measures' 1) pretreatment TI duranons, 2) posttreatment duraUons of immobility and 3) H/L responses to treatment. The Spearman rank correlation coefficient was again used to assess intraindwldual relationships between TI and leucocyt~c responses. RESULTS All three groups showed slmtlar heterophil/lymphocyte ratios prior to treatment and this measure remained virtually unaltered m the control birds maintained on an ad lib diet (Table 1). Conversely, H/L ratios were sigmficantly increased following 72-hr periods of either fasting or frustration of feeding. Pretreatment TI responses were similar across groups (Table 2). The overall susceptibility to TI. as measured by the number of reductions, was significantly reduced upon retesting ( p < 0 . 0 2 ) and TI duration showed a similar, albeit nonsignificant ( p > 0 . 0 5 < 0 . 1 ) trend. However, this response attenuation was relatively homogeneous and there were no significant overall differences between group means. The hierarchy based on pretreatment TI durations was repeatable upon retesting in the controls but was totally destabilised in the fasted and frustrated groups (Table 3). Significant, positive intraindividual correlations (Table 4) were found between pretreatment TI durations and subsequent H/L responses to fasting and frustration (pooled data). Posttreatment immobility reactions and H/L elevations, however, were not significantly correlated. DISCUSSION Heterophil/lymphocyte ratios are known to rise progressively in the domestic fowl upon treatment with corticosterone or

STRESS, TONIC IMMOBILITY AND H/L RATIOS

441

TABLE 2 TONIC IMMOBILITY(TI) RESPONSESBEFOREAND AFTERTREATMENT(MEANS _ SEM) Treatments Ad Ltb Feeding (n = 15)

Pretreatment

Posttreatment

Inducuons (no) Duration of TI (s) Inductmns (no) Duration of TI (s)

Fastang (n = 14)

Frustration of Feeding (n = 15)

348 2 +-- 70.9

1 5 0.3 314.0 - 64.8

1.3 - 0.2 307.4 - 95.2

1.7 - 0.3 191 9 - 39.4

22 - 0.3 164 2 +-- 43.0

1.7 - 0.2 172.7 - 38.7

1 3 +--

03

(no) = number; (s) = seconds.

exposure to presumed stressors, such as fasting, water deprivation or social tension (13, 14, 21). The H/L ratios found here were generally low in relation to those of 0.39 (prefasting) and 0.73 (posffasting) reported elsewhere (14) and may have reflected strain and age differences. However, the 5- to 6-fold elevations which followed 72-hr periods of either fasting or frustration of feeding illustrated the stressful nature of these procedures. Of course, the effects of fasting and of frustration cannot be entirely separated with the approach used here. Not only were the frustrated birds also deprived of food, but the fasted pullets may also have experienced frustration because their assumed expectation of finding food in the hoppers was thwarted. The pretreatment similarities in TI responses suggested that general fearfulness did not vary significantly between groups. Although 3-4 trials are generally required for any appreciable attenuation of the TI reaction (18), both susceptibility and response duration were then reduced upon retesting in the present study, though the latter trend failed to reach significance. Response diminution was relatively homogeneous across groups and presumably reflected habituation. Both fasting and frustration could be expected to increase general arousal, but neither procedure seemed to have affected TI directly here because the mean responses of control and stressed birds were similar. Previous manipulations of arousal via food deprivation (12) or the administration of d-amphetamine (2) have also failed to affect chicks' TI reactions or attenuated their responses, respectively. These findings suggest that tonic immobility and, presumably, underlying fearfulness, are not affected in any direct, systematic way by nonspecific variations in generalized arousal, or at least those associated with fasting, frustration or amphetamine injection.

TABLE 3 SPEARMANRANK CORRELATIONCOEFFICIENTSBETWEENPRE- AND POSTTREATMENTTI RANKS

Groups

Spearman rs

Ad hb Fasted Frustrated N.S. = not significant.

.642 .284 .128

t 3.02 0.896 0.465

However, an interesting picture emerged upon the examination of intraindividual relationships rather than overall effects. The significant intraindividual correlation found across TI tests in the ad lib control group suggested that its members had maintained relatively stable positions in a fearfulness hierarchy. Similar repeatability of fear ranks has also been reported in adult hens (20). Conversely, the TI hierarchy was destabilised in both stressed groups but the reason is unknown. Examination of intraindividual relationships may yield more information than that of group means in studies such as this. Fasting is known to elevate circulating corticosterone concentrations (l, 9, 31) and TI was significantly prolonged in adult White Leghorn hens at 4 and 11 days after implantation of osmotic infusion minipumps releasing corticosterone solution (21). Longer durations of TI nught, therefore, have been expected to follow fasting/frustration in the present study. However, the mean immobility responses of control and stressed birds were similar. There is no obvious explanation for this apparently paradoxical finding, but the destabilisation of the TI hierarchy in stressed birds and the differences between studies in strain, age and duration of treatment may be important factors. A previous finding that implantation of minipumps releasing only vehicle increased H/L ratios without affecting circulating corticosterone levels also suggested that there may be some dissociation between adrenocortical and leucocytic responses, at least in certain circumstances (21). Indeed, exposure to stressors is known to weaken the correlation between plasma corticosterone concentrations and H/L ratios (14). The assessment of behavioural, adrenocortical and leucocytic responses to stressors applied at varying ages to a number of genetic stocks might clarify the situation. TABLE 4 SPEARMANRANKCORRELATIONCOEFFICIENTSBETWEENRANKED H/L RESPONSESTO FASTING/FRUSTRATIONAND PRE- AND POSTTREATMENTTI RANKS(N = 29) Spearman rs

t

p

Pretreatment TI vs. I-I/L responses

48

2.81

<0.01

Posttreatment TI vs. H/L responses

09

0.47

N.S.

p <0.02 N.S N.S

N.S. = not slgmficant; FIlL = heterophd/lymphocyte ratm

442

JONES

Pretreatment TI durations and subsequent H/L responses to fasting or frustration were positively correlated within individual birds. Leucocyt~c responses to chronic stress may, therefore, be more pronounced in birds with high rather than low underlying levels o f fearfulness, although causality is not necessarily implied. Because adrenocortical activation precedes lymphocytosis (32), the above suggestion is consistent with two recent reports. Firstly, chicks selected genetically for high open-field activity were not only less fearful than those o f the corresponding reactive line, but they also showed lower resting plasma corticosterone levels and a lower adrenocorttcal response to experimental coccidial infection (7). Secondly, a consistent, albe~t nonsignificant, trend towards greater adrenocortical responsiveness to a frustrauon o f feeding regtme was found among White Leghorn hens classed as high rather than low fear responders (1) It is conceivable that fearful chickens may experience more frequent and profound d~sturbance than their less fearful counterparts. Resultant differences in the

degree o f n e u r o e n d o c n n e activation may, m turn, perpetuate or exaggerate the behawoural dichotomy. Indeed, behavioural inhibition, indicative o f heightened fearfulness, a c c o m p a m e s chronic elevations o f circulating corticosterone, not only in chickens (21) but also in ducklings (28), rats (27,30) and pigs (6). The above findings suggest that simple behavioural measures, such as tonic immobility or open-field responses, may have pre&ctlve value concerning subsequent leucocytic and possibly adrenocortlcal responsiveness to stressful sumulat~on, at least m the domestic fowl. This potentmlly ~mportant notion merits further study. ACKNOWLEDGEMENTS This study was comnussloned and supported by the Ministry of Agriculture, Fisheries and Food I am grateful to Miss M. W. Thomson for her techmcal assistance

REFERENCES 1 Beuvmg, G , Jones, R. B , Blokhms, H J Adrenocortlcal and heterophil/lymphocyte responses to challenge m hens showing short or long tomc lmmobdlty reactmns Br Poult. Scl 30 175-184; 1989 2. Boren, J L.; Gallup, G G , Jr. Amphetamine attenuation of tonic immobility tn chickens. Physiol. Psychol. 4:429-432, 1976 3 Broom, D M. Welfare problems and how to recogmse them. Mlsset Int. Poult 3:40--41, 1987 4 Broom, D. M , Kmght, P. G., Stansfield, S. C Hen behavlour and hypothalanuc-pitultary-adrenal responses to handhng and transport Appl Atom. Behav Sci. 16.98; 1986 5. Cashman, P. J.; Nlcol, C J , Jones, R. B Effects of transportauon on the tomc mamobihty fear reactmns of broders Br Poult Scl 30 211-221; 1989 6 Dantzer, R.; Morm6de, P Stress m farm animals a need for reevaluation. J. Amm. Scl 57 6-18, 1983 7 Faure, J. M. To adapt the envtronment to the bird or the bird to the environment9 In. Moss, R., ed The laying hen and its enwronment Brussels. ECSC, EEC, EAEC, 1980"19-42 8 Faure, J. M. Analyse g6n6tlque du comportement en open field du jeune poussm (Gallus gallus domesticus) Ann. Genet Sel Atom. 14 187-212; 1982 9 Freeman, B M Stress and the domestic fowl physiological fact or fantasy? Wlds. Poult. Sci J 41'45-51, 1985. 10 Freeman, B. M ; Kettlewell, P J , Manning, A C C , Berry, P S The stress of transportation of broilers. Vet. Rec 114:286-287, 1984 11. Gallup, G. G., Jr. Tomc immobility as a measure of fear m domestic fowl. Atom. Behav. 27 316-317, 1979 12 Gallup, G. G., J r , Wdhamson, G T. Effect of food deprivation and a wsual cliff on tomc lmmobdlty. Psychon. Scl. 29 301-302, 1972 13 Gross, W. B , Chlckenng, W. Effects of fasting, water depnvauon and adrenal-blocking chermcals on resistance to Escherwhia colt challenge Poult. SCL 66 270-272; 1987 14. Gross, W B.; Siegel, H. S. Evaluation of the heterophd/lymphocyte raUo as a measure of stress in chickens. Avian Dis 27 972-979; 1983 15 Harvey, S.; Phflhps, J G., Rees, A , Hall, T R Stress and adrenal funcaon. J. Exp. Zool 232 633--645, 1984 16. Hughes, B. O.; Gilbert, A. B., Brown, M F Categonsauon and causes of abnormal eggshells, relaUonshlp with stress Br Poult Scl 27'325-337, 1986. 17 Jones, R B Fearfulness and adaptabihty m the domestac fowl. IRCS Med. ScL 13:797-800; 1985 18 Jones, R. B. The tomc tmmobdlty reaction of the domestic fowl a review. Wlds. Poult. Sc~. J 42:82-96, 1986 19 Jones, R B. The assessment of fear in adult laying hens correlational analysis of methods and measures. Br. Poult SCl 28.319-326, 1987.

20 Jones, R. B Repeatabdlty of fear ranks among adult laying hens Appl Anlm Behav Sct 19:297-304; 1988. 21 Jones, R. B., Beuvmg, G , Blokhms, H J Tomc lmmobdtty and heterophll/lymphocyte responses of the domestic fowl to corUcosterone infusion Physiol. Behav 42.249-253, 1988 22 Jones, R. B., Duncan, I J H., Hughes, B O. The assessment of fear m domestic hens exposed to a looming human snmulus Behav. Proc 6 121-133; 1981 23 Jones, R B , Faure, J. M. Sex and strata comparisons of tomc tmmobtlity ("righting ume") m the domestic fowl and the effects of various methods of mductlon. Behav Proc 6 47-55, 1981 24. Jones, R. B , Harvey, S. Behavloural and adrenocomcal responses of domestac chicks to systemaUc reductions m group s~ze and to sequential d~sturbance of compamons by the experimenter Behav Proc 14 291-303, 1987 25 Jones, R B , Hughes, B. O Fearfulness and abnormahtles m the chicken's egg shell--is there a hnk 9 Proc Int Symp. Appl. Ethol Farm Atom. Balatonfured, Hungary, 1986'171-175 26. Jones, R B., Mills, A D EstlmaUon of fear m two hnes of the domesUc chick correlations between various methods Behav Proc 8'243-253, 1983 27 Katz, R J . Roth, K A , Carroll, B J Acute and chromc stress effects on open field acUvlty m the rat Imphcanons for a model of depression Neuroscl Blobehav Rev 5:247-251; 1981 28 Landsberg, J. W , Weiss, J Stress and increase of the cortlcosterone level prevent lmpnntmg m duckhngs Behavlour 57:173-189, 1977 29 Mills, A. D., Faure, J M The estlmaUon of fear m domestic quaff. correlations between various methods and measures. Blol Behav 11:235-243, 1986 30 Rodriguez Echandm, E. L., Gonzalez, A S , Cabrera, R , Fracchm, L N A further analysis of behawoural and endocnne effects of unpre&ctable chronic stress Physlol Behav. 43.789-795, 1988 31 Scanes, C. G ; Merrill, G F , Ford, R , Mauser, P , Horowltz, C Effects of stress (hypoglycaemm, endotoxm, and ether) on the peripheral circulating concentrauon of cortlcosterone m the domesuc fowl (Gallus domesticus) Comp Blochem. Physlol 66C 183-186, 1980 32 Siegel, H S Immunological responses as indicators of stress Wlds Poult Sc~ J 41:36--44, 1985 33 Siegel, S. Non-parametric staUstlcs for the behavloural sciences New York McGraw-Hdl, 1956. 34 Suarez, S. D.; Gallup, G. G., Jr Predatory overtones of open-field testing m chickens Atom. Learn. Behav 9 153-163, 1981 35 Webster, A. J. F , Nlcol, C. J. The case for welfare In. Cages for the Future. Cambridge Poultry Conference, ADAS, 1988" 11-21