- Email: [email protected]
Effects of two different housing systems on behavior, physiology and milk yield of Comisana ewes
Small Ruminant Research 41 (2001) 151±161
Effects of two different housing systems on behavior, physiology and milk yield of Comisana ewes D. Casamassimaa, A. Sevib,*, M. Palazzoa, R. Ramacciatoa, G.E. Colellaa, A. Bellittia a
Dipartimento di Scienze Animali, Vegetali e dell'Ambiente, via De Sanctis, 86100 Campobasso, Italy Istituto di Produzioni e Preparazioni Alimentari, FacoltaÁ di Agraria di Foggia, via Napoli 25, 71100 Foggia, Italy
b
Accepted 9 January 2001
Abstract Two groups of 20 early-lactating Comisana ewes were used in this study, and were allocated to either an indoor or outdoor daytime environment. The indoor environment was a 3 m 12 m straw bedded pen inside a pre-fabricated building. The outdoor environment was a 200 m2 paddock during daytime (09.00±19.00 h) with ewes being moved to the shed, as described for the indoor treatment, at night. Behavior of ewes was recorded at 14 days interval from 09.00 to 19.00 h. A phytohemagglutinin (PHA) skin test was performed at weeks 6, 11, 15 and 18 of the experiment to induce a non-speci®c delayed-type hypersensitivity in ewes. Jugular blood samples were taken at the beginning and at weeks 5, 10, 14 and 18 to determine endocrine and metabolic responses of animals to housing system and to changes in climatic conditions and stage of lactation. Ewe milk yield was recorded daily and individual milk samples were analyzed for milk composition, coagulating properties and somatic cell count (SCC) at 14 days interval. No differences were found between groups for endocrine and immune responses. Outdoor ewes had increased locomotor activities (P < 0:01) and decreased idling (P < 0:05) compared to indoor animals. The outdoor group had higher levels of blood creatinine (P < 0:01) and inorganic phosphorus (P < 0:05) as well as lower urea (P < 0:01) and glucose (P < 0:001) concentrations compared to the indoor group. Milk yield and composition were not changed by the housing system, though signi®cant time treatment interactions were found for milk constituents, with indoor-housed ewes having higher (P < 0:05) milk protein, fat and lactose concentrations during the middle of the trial. Outdoor ewes had lower SCC (P < 0:003) and pH (P < 0:001) in their milk than indoor ewes, whereas renneting parameters were not different across treatments. Results suggest that ewe welfare and productivity were not substantially affected by the housing system. The provision of feeding rations that meet the greater energy demand for maintenance is required to sustain productivity in outdoor reared sheep. # 2001 Elsevier Science B.V. All rights reserved. Keywords: Sheep; Housing system; Behavior; Welfare; Metabolism; Milk yield
1. Introduction Increased size of specialized ovine milk producing ¯ocks, especially in the northern countries of the *
Corresponding author. Tel.: 39-881-589222; fax: 39-881-740211. E-mail address: [email protected] (A. Sevi).
Mediterranean basin, has recently warranted the attention of scientists wanting to assess the interrelationships between microclimate, management practices and sheep welfare. In intensive production systems, animals do not face unpredictable and challenging environments and are generally protected from climate extremes and inadequate nutrition; undoubtedly this can result in improved welfare and
0921-4488/01/$ ± see front matter # 2001 Elsevier Science B.V. All rights reserved. PII: S 0 9 2 1 - 4 4 8 8 ( 0 1 ) 0 0 2 0 1 - 2
152
D. Casamassima et al. / Small Ruminant Research 41 (2001) 151±161
health of livestock (Berge, 1997; Brosh et al., 1998). However, published evidence proves that a number of factors, primarily space allocation (Sevi et al., 1999) and indoor climate and hygiene (Curtis, 1983; Hartung, 1994), are critical to ensure the health and welfare of housed animals. Furthermore, living in highly predictable and structured environments may also cause problems to farm animals (Wemelsfelder and Birke, 1997), that cannot perform many of the behaviors they would normally use to attain their functional goals (Petherick and Rushen, 1997). There is a widespread belief that behavioral restriction or deprivation will cause animals to suffer (Hughes and Duncan, 1988). It is known that sheep habituate to con®nement less easily than other farm animals. Indeed, sheep are gregarious and relatively defenseless animals and husbandry operations, such as simple handling, may result in the motivational state of fear, anxiety and frustration (Lynch et al., 1992). It has been found that housing sheep in enclosures induces an increase in heart rates in rams (Palestrini et al., 1998) as well as long-term physiological disturbance and increased cortisol secretion in ewes (McNatty and Young, 1973; Pearson and Mellor, 1976). Done-Currie et al. (1984) also found that movement of sheep from pasture to an indoor environment results in a higher incidence of stereotypic activities and that the performance of abnormal behaviors increases with length of con®nement. Thus, adopting less restrictive housing conditions might reduce the impact of con®nement on sheep well-being. When assessing the effects of the rearing system on lactating ewes, Celi et al. (1993) found higher body weight gains and similar yields of milk in indoorhoused animals, compared to grazing. Alliston and Lucas (1979) did not observe any signi®cant difference in ewe reproductive performance and lamb growth rate when comparing outdoor rearing to housing for all or part of the winter. Number of lambs born, wool production and feed intake are reported by Berge (1997) to be similar for sheep under different housing conditions. There is a paucity of data describing the effect of housing system on ewe milk yield and welfare. The present study was undertaken to assess the behavioral, endocrine and metabolic responses and production performance of lactating ewes when
housed indoors or maintained in an outdoor enclosure during the day and housed indoors during the night. 2. Materials and methods 2.1. Experimental design and animal management The experimental site was approximately 10 km northwest of Campobasso, Molise, Italy (latitude: 418340 N and longitude: 148390 E), with an elevation of about 650 m above sea level. The climate of this area is Mediterranean, with about 650 mm of annual rainfall, mainly distributed in late autumn and winter, and 15.7 and 8.48C mean maximum and minimum temperatures over the last 40 years. In this area, ¯ocks are maintained on pasture during the daytime and moved to a shed at night, except for the middle part of the winter season during which sheep are not on pasture, due to weather severity and reduced herbage availability. The experiment, which lasted 18 weeks, was conducted in the spring and summer (April± August) of 1999 and involved 40 Comisana ewes that had lambed in winter. At weaning (50 3 days after lambing), ewes were divided into two groups of 20 each, balanced for age, parity, time of lambing, number of lambs suckled, body weight (44:6 0:7 kg, means S:E.) and milk yield and composition. The indoor group was housed in a 3 m 12 m straw bedded pen inside a pre-fabricated building throughout the study period, whereas outdoor group was maintained in a 200 m2 paddock, designated with a wire-mesh fence, during daytime (09.00±19.00 h) and moved to a pen of the same area and adjacent to the pen that indoor ewes were housed in during nighttime. The experimental building was 4 m high and provided with transom windows placed at a height of 2.5 m from the ¯oor. The paddock was on ¯at ground and herbage was mowed to ground level at the commencement of the experiment. Pens and paddock were provided with two mangers and a hay crib (feeder space 0:4 m per animal). The feeding area in the paddock being covered by a 3 m 4 m wooden canopy. Experimental pens were not cleaned during the study period, but a layer of straw (about 0.3 kg/m2) was added daily to each pen. Ewes were fed on a diet composed of 67.5% alfalfa hay and 32.5% pelleted concentrate, which was
D. Casamassima et al. / Small Ruminant Research 41 (2001) 151±161
administered in three meals daily (09.00, 14.30 and 21.00 h) in equal amounts to the experimental groups. The chemical composition of dry matter was determined by AOAC (1990) methods and was 16.3% crude protein, 3.1% fat, by ether extract, 20.7% crude ®bre and 8.6% ash. Throughout the trial, the amount of food given to the ewes was modi®ed based on the mean productive levels, keeping the forage to concentrate ratio 0.625±0.375 constant. Average daily DM intake was 2.2 kg per ewe during the study period. Ambient temperature and relative humidity inside the pre-fabricated building and in the paddock were monitored throughout the trial using thermo-hygrographs (LSI, 20090 Settala Premenugo, Italy) placed at a height of 2 m from the ¯oor. Temperature±humidity index (THI) was calculated using the formula of Kelly and Bond (1971). 2.2. Behavioral observation Behavioral observations were recorded by two trained observers equipped with video cameras from 09.00 to 19.00 h at fortnightly intervals. Three different focal animals were chosen at random each time in both groups. All three of the focal animals were videotaped and the tapes were then reviewed continuously for each animal. Times spent in each of the following seven behavioral categories were recorded: eating, drinking, ruminating, lying, walking, idling and other activities (social and aggressive interactions, scratching oneself, bleating and fence or manger or bucket biting). 2.3. In vivo cell-mediated immunity The phytohemagglutinin (PHA) skin test was performed to induce a non-speci®c delayed-type hypersensitivity. At weeks 6, 11, 15 and 18 of the experiment, 1 mg of PHA (Sigma-Aldrich, 20151 Milan, Italy) dissolved in 1 ml of sterile saline solution was injected intradermally into the middle of two 2 cm wide circles stamped on shaved skin in the upper side of each shoulder. The skinfold thickness was determined before PHA injection and 24 h after with a caliper. For each animal an average increase in skinfold thickness (24 h thickness Ð pre-injection thickness) was compared using an average of the two measurements gathered from each shoulder.
153
2.4. Blood sampling and analyses Jugular blood samples were taken from ewes at the beginning of the experiment and at weeks 5, 10, 14 and 18 of the trial. Blood samplings were collected in 10 ml vacuum tubes at 07.00 h. Samples were centrifuged for 20 min at 3500 rpm and blood serum was stored at 208C except for a sub-sample which was immediately analyzed enzymatically for serum concentrations of glucose. The following metabolites and enzymes were determined within 3 days from blood sampling using standard procedures and a semi-automatic UVICON 941 spectrophotometer (Kontron, Milan, Italy): total protein, albumin, total bilirubin, NEFA, calcium and inorganic phosphorus, using colorimetrical methods, and total cholesterol, triglycerides, urea, creatinine, aspartate-amintransferase (AST/GOT), alanine-aminotransferase (ALT/GPT), lactate-dehydrogenase (LDH) and alkaline phosphatase (ALP), using enzymatical methods. The concentrations of cortisol, free-triiodothyronine (F-T3) and free-thyroxine (F-T4) were determined using an enzyme-linked ¯uorescent assay (ELFA) and commercial reagents (Biomerieux, F-69280 Marcy l'Etoile, France). 2.5. Sampling and analyses of milk Ewes were hand milked twice daily (08.00 and 20.00 h). Morning and evening milk was collected in metal buckets and weighed using an electronic scale. Milk samples, consisting of proportional volumes of morning and evening milk, were individually collected in plastic containers at 14 days interval, refrigerated (48C) and transported to our laboratory. Upon arrival at the laboratory (30± 60 min after collection), the following measurements were made on milk: pH, total protein, fat and lactose content, using an IR spectrophotometer (Milko Scan 133B, Foss Electric, DK-3400 Hillerùd, Denmark) according to the IDF (1990) standard; somatic cell count (SCC), using a Foss Electric Fossomatic 90 cell counter (IDF, 1995); and renneting parameters (clotting time, rate of clot formation and clot ®rmness after 30 min) using a Foss Electric Formagraph and the method of Zannoni and Annibaldi (1981). At the beginning of the experiment and on the day before each milk sampling, all the ewes were
154
D. Casamassima et al. / Small Ruminant Research 41 (2001) 151±161
examined carefully to assess the absence of signs of clinical mastitis, such as fever, pain or gland swelling. A small quantity of milk was checked visually for signs of mastitis. Body weights of the ewes were recorded at the beginning and at weeks 9 and 18 of the study period after the machine milking in the morning but before feeding. 2.6. Statistical analysis Data from two ewes, one from the indoor and the other from the outdoor group, were excluded from the data set due to suspected mastitis. Behavioral data were analyzed using the Wilcoxon rank-sum test in the NPAR1WAY procedure of the SAS (1990) statistical software, having housing system and time as categorical factors. Milk and blood variables were tested for normality using the Shapiro and Wilk (1965) test and milk SCC were log-transformed to normalize their frequency distributions before performing statistical analysis. A linear model that included the effects of housing system, time, housing system time and residual error was used (SAS, 1990). Pre-treatment values were used as covariates for blood parameters. Body weights and body weight changes of the ewes
were analyzed using a linear model that included housing system and residual error. 3. Results 3.1. Meteorological data Averages of indoor and outdoor temperatures during daytime (09.00±19.00 h) were substantially similar throughout the experiment (Table 1). Differences only exceeded 38C during weeks 5, 13 and 17. Through the trial, humidity ranged from 41 to 63% in the shed and from 33 to 72% in the paddock. During weeks 15, 17 and 18, averages of indoor and outdoor THI exceeded 72, which is considered the upper critical THI at least for lactating cows (Armstrong, 1994). Outdoor THI values were also over 72 during weeks 9, 10 and 13. 3.2. Behavior Feeding activity (Fig. 1) was not affected by treatment (26.9 versus 21.7% in indoor and outdoor group, respectively, P 0:079), whereas drinking was signi®cantly lower in housed than in outdoor reared
Table 1 Averages of indoor and outdoor temperatures (8C), humidities (%) and temperature±humidity indexes (THI) during daytime (09.00±19.00 h) Trial week
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Temperature
Relative humidity
THI
Indoor
Outdoor
Indoor
Outdoor
Indoor
Outdoor
14.0 11.7 14.6 20.7 17.8 20.8 18.4 21.9 24.7 24.1 20.6 21.1 25.0 24.0 26.9 22.6 30.4 31.5
12.8 10.9 14.2 23.0 22.0 21.7 17.0 23.4 25.4 25.6 21.7 22.6 28.1 22.9 25.1 21.6 26.9 28.5
55.0 56.9 63.1 59.6 61.7 51.0 56.4 51.4 41.0 44.0 56.3 41.1 47.6 51.0 67.7 61.7 51.6 47.5
61.0 59.0 67.0 53.4 72.0 64.1 72.2 52.7 53.7 68.4 44.3 46.1 62.1 42.1 55.9 54.7 33.6 32.9
57.4 54.2 58.2 66.7 62.8 66.3 63.4 67.8 70.5 70.1 66.4 66.1 71.5 70.6 76.4 69.6 79.1 79.8
55.7 53.1 57.6 69.4 69.5 68.5 61.9 69.9 72.7 74.6 67.1 68.3 77.4 68.4 72.2 67.7 72.2 73.9
D. Casamassima et al. / Small Ruminant Research 41 (2001) 151±161
155
Fig. 1. Behavioral traits of ewes when allocated to an indoor or outdoor environment during daytime. Values are least square means S:E.
animals (0.9 versus 2.1%). Time spent in ambulatory activities was about three-fold greater in the outdoor than in indoor ewes (13.4 versus 4.5%, P < 0:001). Conversely, the incidence of idling was higher in indoor compared to outdoor ewes (21.5 versus 13.3%, P < 0:05). Irrespective of housing system, eating and ruminating decreased (P < 0:001) over time, whereas lying, idling and drinking activities increased (P < 0:001). Aggressive interactions and
anomalous behaviors were sporadically observed in both indoor and outdoor ewes. 3.3. Immune response The skinfold thickness after PHA injection (Fig. 2) was similar across treatments throughout the experiment (5.31 and 5.11 mm, in indoor and outdoor groups, respectively, P 0:407).
Fig. 2. Immune response to PHA injection in ewes when allocated to an indoor or outdoor environment during daytime. Values are least square means S:E.
156
D. Casamassima et al. / Small Ruminant Research 41 (2001) 151±161
3.4. Endocrine response Neither cortisol, nor F-T3, or F-T4 blood levels (Fig. 3) were changed by housing system. However, cortisol concentrations tended to decrease in
outdoor compared to indoor ewes at week 14 of the study (8.36 versus 9.67 ng/ml, P 0:081). An effect of time of sampling was observed for F-T3 (P < 0:05) and F-T4 (P < 0:01), with ewes having the lowest blood levels of both thyroid hormones
Fig. 3. Plasma levels of cortisol, free-triiodothyronine and free-thyroxine in ewes when allocated to an indoor or outdoor environment during daytime. Values are least square means S:E.
D. Casamassima et al. / Small Ruminant Research 41 (2001) 151±161
157
Table 2 Changes in blood metabolites and enzymes in ewes when allocated to an indoor or outdoor environmenta Item
Trial week
Glucose (mmol/l)
Urea (mmol/l)
Group
S.E.
Indoor (n 19)
Outdoor (n 19)
5 10 14 18
3.37 a 2.80 3.68 a 4.07
3.06 b 2.71 2.18 b 3.69
5 10 14 18
6.36 7.69 9.54 a 5.38
5.68 6.26 6.86 b 6.14
Effects, P Treatment
Time
Treatment time
0.2 0.1 0.2 0.2
0.0001
0.0767
0.0001
0.6 0.6 0.6 0.4
0.0075
0.4986
0.0336
Creatinine (mmol/l)
5 10 14 18
51.3 b 30.5 11.5 13.0
81.3 a 35.2 13.6 14.6
4.1 2.5 2.1 1.4
0.0011
0.0001
0.0051
AST/GOT (U/l)
5 10 14 18
58.2 a 40.5 46.0 b 38.4
32.3 b 38.5 58.6 a 33.6
2.6 3.1 2.9 3.1
0.0180
0.2532
0.0001
5 10 14 18
13.8 8.6 b 8.4 7.9
11.6 16.9 a 10.0 8.5
1.2 2.6 0.7 0.7
0.0605
0.8205
0.0023
0.1 0.1 0.2 0.1
0.0464
0.9961
0.3685
ALT/GPT (U/l)
Inorganic phosphorus (mmol/l)
a
5 10 14 18
1.31 b 1.78 1.32 b 1.19
1.86 a 1.93 2.22 a 1.19
Values are least square means S.E. Means followed by different letters differ signi®cantly at P < 0:05.
Fig. 4. Milk yield (g per day) of ewes when allocated to an indoor or outdoor environment during daytime. Values are least square means S:E.
158
D. Casamassima et al. / Small Ruminant Research 41 (2001) 151±161
at week 18 of the experiment regardless of housing system. 3.5. Blood metabolites and enzymes Blood glucose levels were higher in indoor than in outdoor ewes at weeks 5 (P < 0:05) and 14 (P < 0:001) of the experiment (Table 2), whereas blood concentrations of glucose only tended to increase (P 0:06) in housed compared to outdoor reared animals at week 18. Indoor ewes also exhibited higher urea (P < 0:001) and lower AST/GOT levels
(P < 0:01) than outdoor animals at week 14. In contrast, the latter group had higher levels of blood creatinine (P < 0:001) at week 5, of ALT/GPT (P < 0:05) at week 10 and of inorganic phosphorus (P < 0:01) at weeks 5 and 14 of the experiment. No differences were found for any of the other parameters measured. 3.6. Milk yield Average milk yield and constituents were not affected by housing system. However, signi®cant
Fig. 5. Total protein and fat content (%) in the milk of ewes when allocated to an indoor or outdoor environment during daytime. Values are least square means S:E.
D. Casamassima et al. / Small Ruminant Research 41 (2001) 151±161
159
Table 3 Somatic cell count (SCC), pH and renneting parameters of milk in ewes when allocated to an indoor or outdoor environmenta Item
Group
SCC (cells/ml 103) pH Clotting time (min) Rate of clot formation (min) Clot firmness (mm) a
S.E.
Indoor (n 19)
Outdoor (n 19)
437 6.45 9.82 2.16 59.8
224 6.37 9.56 2.14 62.8
49 0.01 0.39 0.07 1.61
Effects, P Treatment
Time
Treatment time
0.0031 0.0001 0.6385 0.7985 0.3111
0.1947 0.0001 0.0001 0.0003 0.0001
0.0464 0.0001 0.0761 0.6511 0.2004
Values are least square means S.E.
time treatment interactions were found for all parameters investigated. Indeed, indoor ewes yielded greater volumes of milk (P < 0:05) at weeks 2 and 6 of the experiment than outdoor ewes (Fig. 4). In addition, housed animals had higher milk contents (P < 0:05) of protein at weeks 8 and 14, of fat (Fig. 5) at weeks 6 and 14 and of lactose at week 8 of the experiment compared to outdoor reared ewes. Milk from outdoor ewes had a lower pH (P < 0:001) and somatic cell counts (P 0:003) than indoor animals, with the difference being greater during the second half of the trial (Table 3). In contrast, no signi®cant differences were found in the coagulating properties of milk between groups. 3.7. Body weight change Body weight and body weight changes were similar across treatments. On average, weight gains were 44:6 17:2 and 45:0 21:1 g per day in the indoor and outdoor groups, respectively. 4. Discussion Outdoor ewes spent more time walking than indoor animals, due to a greater available space and possibly to a stronger kinetic drive to investigate the environment (Carson, 1985). Walking is part of exploratory behavior (Gluesing and Balph, 1980) and it is well known that sheep spend a long time exploring any new ®eld they are put in (Fraser and Broom, 1990). Indoor ewes spent more time idling compared to outdoor animals, probably as a consequence of an increased threshold of stimuli perception due to a narrower range of stimuli experienced (Wemelsfelder, 1993).
Inactive behaviors in group-housed animals have also been interpreted as a strategy to reduce physical interactions with conspeci®cs (Fraser and Broom, 1990; Barnett et al., 1992; Hanlon et al., 1994). The reduction in milk yield and the rise in ambient temperature with the advancement of the trial were probably responsible for decreased feeding and ruminating activities of ewes and for increased thermoregulatory behaviors, such as lying, idling and drinking (Shref¯er and Hohenboken, 1980). Friend et al. (1985) did not observe signi®cant changes of neutrophil and lymphocyte counts in the blood of Holstein calves when housed in pens, hutches and yards. Accordingly, no differences in the immune response were found in this study between indoor and outdoor reared animals. These results are consistent with the lack of differences in blood cortisol concentrations, because the levels of circulating corticosteroids are known to have a direct effect on ewe immune function (Cockram et al., 1994). The changes in climatic conditions through the study period may explain the behavior of both cortisol and thyroid hormones through the experiment. Indeed, animals respond to rising THI values with a reduction in the secretion of thermogenic hormones to lower their basal metabolism (Habeeb et al., 1992). Reduction in ewe metabolic rate with the decline of lactation also contributed to lowering of blood serum levels of thyroid hormones at week 18 of the experiment (Hart, 1983). Increased energy demand and muscular tissue metabolism, due to enhanced locomotion, resulted in higher blood levels of creatinine and lower blood glucose concentrations in outdoor compared to indoor ewes at week 5 of the experiment (Finco, 1989). The reduction of glucose levels in outdoor reared animals
160
D. Casamassima et al. / Small Ruminant Research 41 (2001) 151±161
was probably magni®ed by the exposure to rising THI values at week 14 of the experiment (Herbein et al., 1985). The reduction in cortisol levels supports the existence of a direct relation between climatic conditions and glucose changes (Alvarez and Johnson, 1973). An inverse relation between glucose and inorganic phosphorus concentrations have also been found by Calamari et al. (1991) in the plasma of calves subjected to restricted feeding. These authors postulated that the increment of inorganic phosphorus levels may be the result of an increased glucose catabolism with a pending energy de®cit situation. Accordingly, in the present study, lower glucose levels were associated with higher inorganic phosphorus concentrations in the blood serum of outdoor reared animals at weeks 5 and 14 of the experiment. The differences in blood serum urea concentrations at week 14 might be ascribed to either a nitrogen intake in excess of requirements to indoor ewes or a reduced use of amino acids for gluconeogeneis in outdoor animals (Calamari et al., 1991). However, an increased urinary nitrogen excretion, that has been described as a metabolic reaction of dairy animals to thermal stress (Habeeb et al., 1992), was not excluded to interpret the changes of urea levels in the blood serum of outdoor ewes at this time. Differences in ALT/GPT and AST/GOT blood levels among groups are not easy to explain. Ronchi et al. (1999) observed a transaminase increase in the plasma of lactating cows subjected to restricted feeding, hypothesizing that it was an adaptive response to energy de®cit leading to an enhanced metabolic activity of the liver. Reduced energy and nitrogen availability to the mammary gland resulted in a transient drop of milk yield and deterioration of milk quality in the outdoor group. The SCC did not signi®cantly change in the milk from outdoor ewes throughout the experiment. Conversely, the indoor group had a marked SCC increase in milk during the second half of the trial, probably due to a progressive worsening of air and litter condition, as a consequence of fecal accumulation and trampling by indoor reared animals. 5. Conclusion In this study, housing system did not affect the endocrine and immune responses of lactating ewes,
although indoor and outdoor reared animals had different behavioral responses, as far as locomotor activity and inactive behavior. Outdoor enclosure of ewes resulted in a lower milk SCC than indoor con®nement as well as in reduced yield and nutritional quality of milk. These ®ndings suggest that outdoor enclosure could be bene®cial to the behavioral needs of lactating ewes, by satisfying their motivation for walking around (Petherick and Rushen, 1997), though no signi®cant advantage in terms of stress related indicators were found. Acknowledgements Research supported by the Italian Ministero della Ricerca Scienti®ca e Tecnologica (MURST ex 40%) ``Effect of housing system on sheep behavior, welfare and production performance'' Scienti®c director: Prof. Donato Casamassima. Thanks are due to Dr. Julie Morrow-Tesch for her precious help with English. References Alliston, J.C., Lucas, I.A.M., 1979. Response of Welsh mountain ewes to ¯ushing and to housing for all or part of the winter. Anim. Prod. 28, 257±269. Alvarez, M.B., Johnson, H.D., 1973. Environmental heat exposure on cattle plasma catecholamine and glucocorticoids. J. Dairy Sci. 56, 189±194. AOAC, 1990. Of®cial Methods of Analysis, 15th Edition. Association of Of®cial Analytical Chemists, Washington, DC, pp. 69±88. Armstrong, D.V., 1994. Heat stress interaction with shade and cooling. J. Dairy Sci. 77, 2044±2050. Barnett, J.L., Hemsworth, P.H., Cronin, G.M., Newman, E.A., McCallum, T.H., Chilton, D., 1992. Effects of pen size, partial stalls and method of feeding on welfare-related behavioral and physiological responses of group-housed pig. Appl. Anim. Behav. Sci. 34, 207±220. Berge, E., 1997. Housing of sheep in cold climate. Livest. Prod. Sci. 49, 139±149. Brosh, A., Aharoni, Y., Degen, A.A., Wright, A., Young, B.A., 1998. Effects of solar radiation, dietary energy and time of feeding on thermoregulatory responses and energy balance in cattle in a hot environment. J. Anim. Sci. 76, 2671±2677. Calamari, L., Bertoni, G., Cappa, V., 1991. Endocrine and metabolic effect of prolonged fasting in Friesian calves. In: Proceedings of the 9th Congresso Nazionale ASPA. Rome, pp. 299±310. Carson, K., 1985. Kinesis. In: Fraser, A.F. (Ed.), Ethology of Farm Animals, Elsevier, Amsterdam, NL, pp. 209±214.
D. Casamassima et al. / Small Ruminant Research 41 (2001) 151±161 Celi, R., Cosentino, E., Melodia, L., Di Trana, A., 1993. Changes in production performance and body weight of lactating ewe as affected by rearing system and concentrate supplementation during pregnancy. Zootecnica Nutrizione Animale 19, 141± 149. Cockram, M.S., Ranson, M., Imlah, P.J., Burrells, C., Harkiss, G.D., 1994. The behavioral, endocrine and immune responses of sheep to isolation. Anim. Prod. 58, 389±399. Curtis, S.E., 1983. Environmental Management in Animal Agriculture. Iowa State Press, Ames. Done-Currie, J.R., Hecker, J.F., Wodzicka-Tomaszewska, M., 1984. Behavior of sheep transferred from pasture to an animal house. Appl. Anim. Behav. Sci. 12, 121±130. Finco, D.R., 1989. Kidney function. In: Kaneko, J.J. (Ed.), Clinical Biochemistry of Domestic Animals, 4th Edition. Academic Press, Saint Diego, CA, pp. 497±538. Fraser, A.F., Broom, D.M., 1990. Farm Animal Behavior and Welfare. Bailliere Tindall, London. Friend, T.H., Dellmeier, G.R., Gbur, E.E., 1985. Comparison of four methods of calf con®nement. I. Physiology. J. Anim. Sci. 60, 1095±1101. Gluesing, E.A., Balph, D.F., 1980. An aspect of feeding behavior and its importance to grazing systems. J. Range Manage. 33, 426±427. Habeeb, A.A.M., Marai, I.F.M., Kamal, T.H., 1992. Heat stress. In: Phillips, C., Piggins, D. (Eds.), Farm Animals and the Environment. CAB International, Wallingford, UK, pp. 27±47. Hanlon, A.J., Rhind, S.M., Reid, H.W., Burrells, C., Lawrence, A.B., Milne, J.A., McMillen, S.R., 1994. Relationship between immune response, liveweight gain, behavior and adrenal function in red deer (Cervus elaphus) calves derived from wild and farmed stock, maintained at two housing densities. Appl. Anim. Behav. Sci. 41, 243±255. Hart, I.C., 1983. Endocrine control of nutrient partition in lactating ruminants. Proc. Nutr. Soc. 42, 181±194. Hartung, J., 1994. The effect of airborne particulate on livestock health and production. In: Ap Dewi, I., Axford, R.F.E., Marai, I.F.M., Omed, H. (Eds.), Pollution in Livestock Production Systems. CAB International, Wallingford, UK, pp. 55±79. Herbein, J.H., Aiello, R.J., Eckler, L.I., Pearson, R.E., Akers, R.M., 1985. Glucagon, insulin, growth hormone, and glucose concentrations in blood plasma of lactating dairy cows. J. Dairy Sci. 68, 320±325. Hughes, B.O., Duncan, I.J.H., 1988. The notion of ethological need, models of motivation and animal welfare. Anim. Behav. 36, 1696±1707. IDF, 1990. Determination of Milk Fat, Protein and Lactose Content Ð Guide for the Operation of Mid-Infra-Red Instruments. IDF
161
Standard No. 141B. International Dairy Federation, Brussels, Belgium. IDF, 1995. Enumeration of Somatic Cells. IDF Standard No. 148A. International Dairy Federation, Brussels, Belgium. Kelly, C.F., Bond, T.E., 1971. Bioclimatic Factors and Their Measurement: A Guide to Environmental Research on Animals. National Academy of Sciences, Washington, DC. Lynch, J.J., Hinch, G.N., Adams, D.B., 1992. The Behavior of Sheep. Biological Principles and Implications for Production. CAB International, Wallingford, UK. McNatty, K.P., Young, A., 1973. Diurnal changes of plasma cortisol levels in sheep adapting to a new environment. J. Endocrinol. 56, 329±330. Palestrini, C., Ferrante, V., Mattiello, S., Canali, E., Carenzi, C., 1998. Relationship between behavior and heart rate as an indicator of stress in domestic sheep under different housing systems. Small Rum. Res. 27, 177±181. Pearson, R.A., Mellor, D.J., 1976. Some behavioral and physiological changes in pregnant goats and sheep during adaptation to laboratory conditions. Res. Vet. Sci. 20, 215±217. Petherick, J.C., Rushen, J., 1997. Behavioral restriction. In: Appleby, M.C., Hughes, B.O. (Eds.), Animal Welfare. CAB International, Wallingford, UK, pp. 89±105. Ronchi, B., Bernabucci, U., Lacetera, N., Verini Supplizi, A., Nardone, A., 1999. Distinct and common effects of heat stress and restricted feeding on metabolic status of Holstein heifers. Zootecnica Nutrizione Animale 25, 11±20. SAS, 1990. SAS/STAT User's Guide, Version 6, 4th Edition. Statistical Analysis System Inst., Cary, NC. Sevi, A., Massa, S., Annicchiarico, G., Dell'Aquila, S., Muscio, A., 1999. Effect of stocking density on ewes milk yield, udder health and microenvironment. J. Dairy Res. 66, 489±499. Shapiro, S.S., Wilk, M., 1965. An analysis of variance test for normality. Biometrika 52, 591±601. Shref¯er, C., Hohenboken, W., 1980. Circadian behavior, including thermoregulatory activities, in feedlot lambs. Appl. Anim. Behav. Sci. 6, 241±246. Wemelsfelder, F., 1993. The concept of animal boredom and its relationship to stereotyped behavior. In: Lawrence, A.B., Rushen, J. (Eds.), Stereotypic Animal Behavior: Fundamentals and Applications to Welfare. CAB International, Wallingford, UK, pp. 65±95. Wemelsfelder, F., Birke, L., 1997. Environmental challenge. In: Appleby, M.C., Hughes, B.O. (Eds.), Animal Welfare. CAB International, Wallingford, UK, pp. 35±47. Zannoni, M., Annibaldi, S., 1981. Standardization of the renneting ability of milk by Formagraph. Scienza e Tecnica LattieroCasearia 32, 9±94.
Effects of two different housing systems on behavior, physiology and milk yield of Comisana ewes D. Casamassimaa, A. Sevib,*, M. Palazzoa, R. Ramacciatoa, G.E. Colellaa, A. Bellittia a
Dipartimento di Scienze Animali, Vegetali e dell'Ambiente, via De Sanctis, 86100 Campobasso, Italy Istituto di Produzioni e Preparazioni Alimentari, FacoltaÁ di Agraria di Foggia, via Napoli 25, 71100 Foggia, Italy
b
Accepted 9 January 2001
Abstract Two groups of 20 early-lactating Comisana ewes were used in this study, and were allocated to either an indoor or outdoor daytime environment. The indoor environment was a 3 m 12 m straw bedded pen inside a pre-fabricated building. The outdoor environment was a 200 m2 paddock during daytime (09.00±19.00 h) with ewes being moved to the shed, as described for the indoor treatment, at night. Behavior of ewes was recorded at 14 days interval from 09.00 to 19.00 h. A phytohemagglutinin (PHA) skin test was performed at weeks 6, 11, 15 and 18 of the experiment to induce a non-speci®c delayed-type hypersensitivity in ewes. Jugular blood samples were taken at the beginning and at weeks 5, 10, 14 and 18 to determine endocrine and metabolic responses of animals to housing system and to changes in climatic conditions and stage of lactation. Ewe milk yield was recorded daily and individual milk samples were analyzed for milk composition, coagulating properties and somatic cell count (SCC) at 14 days interval. No differences were found between groups for endocrine and immune responses. Outdoor ewes had increased locomotor activities (P < 0:01) and decreased idling (P < 0:05) compared to indoor animals. The outdoor group had higher levels of blood creatinine (P < 0:01) and inorganic phosphorus (P < 0:05) as well as lower urea (P < 0:01) and glucose (P < 0:001) concentrations compared to the indoor group. Milk yield and composition were not changed by the housing system, though signi®cant time treatment interactions were found for milk constituents, with indoor-housed ewes having higher (P < 0:05) milk protein, fat and lactose concentrations during the middle of the trial. Outdoor ewes had lower SCC (P < 0:003) and pH (P < 0:001) in their milk than indoor ewes, whereas renneting parameters were not different across treatments. Results suggest that ewe welfare and productivity were not substantially affected by the housing system. The provision of feeding rations that meet the greater energy demand for maintenance is required to sustain productivity in outdoor reared sheep. # 2001 Elsevier Science B.V. All rights reserved. Keywords: Sheep; Housing system; Behavior; Welfare; Metabolism; Milk yield
1. Introduction Increased size of specialized ovine milk producing ¯ocks, especially in the northern countries of the *
Corresponding author. Tel.: 39-881-589222; fax: 39-881-740211. E-mail address: [email protected] (A. Sevi).
Mediterranean basin, has recently warranted the attention of scientists wanting to assess the interrelationships between microclimate, management practices and sheep welfare. In intensive production systems, animals do not face unpredictable and challenging environments and are generally protected from climate extremes and inadequate nutrition; undoubtedly this can result in improved welfare and
0921-4488/01/$ ± see front matter # 2001 Elsevier Science B.V. All rights reserved. PII: S 0 9 2 1 - 4 4 8 8 ( 0 1 ) 0 0 2 0 1 - 2
152
D. Casamassima et al. / Small Ruminant Research 41 (2001) 151±161
health of livestock (Berge, 1997; Brosh et al., 1998). However, published evidence proves that a number of factors, primarily space allocation (Sevi et al., 1999) and indoor climate and hygiene (Curtis, 1983; Hartung, 1994), are critical to ensure the health and welfare of housed animals. Furthermore, living in highly predictable and structured environments may also cause problems to farm animals (Wemelsfelder and Birke, 1997), that cannot perform many of the behaviors they would normally use to attain their functional goals (Petherick and Rushen, 1997). There is a widespread belief that behavioral restriction or deprivation will cause animals to suffer (Hughes and Duncan, 1988). It is known that sheep habituate to con®nement less easily than other farm animals. Indeed, sheep are gregarious and relatively defenseless animals and husbandry operations, such as simple handling, may result in the motivational state of fear, anxiety and frustration (Lynch et al., 1992). It has been found that housing sheep in enclosures induces an increase in heart rates in rams (Palestrini et al., 1998) as well as long-term physiological disturbance and increased cortisol secretion in ewes (McNatty and Young, 1973; Pearson and Mellor, 1976). Done-Currie et al. (1984) also found that movement of sheep from pasture to an indoor environment results in a higher incidence of stereotypic activities and that the performance of abnormal behaviors increases with length of con®nement. Thus, adopting less restrictive housing conditions might reduce the impact of con®nement on sheep well-being. When assessing the effects of the rearing system on lactating ewes, Celi et al. (1993) found higher body weight gains and similar yields of milk in indoorhoused animals, compared to grazing. Alliston and Lucas (1979) did not observe any signi®cant difference in ewe reproductive performance and lamb growth rate when comparing outdoor rearing to housing for all or part of the winter. Number of lambs born, wool production and feed intake are reported by Berge (1997) to be similar for sheep under different housing conditions. There is a paucity of data describing the effect of housing system on ewe milk yield and welfare. The present study was undertaken to assess the behavioral, endocrine and metabolic responses and production performance of lactating ewes when
housed indoors or maintained in an outdoor enclosure during the day and housed indoors during the night. 2. Materials and methods 2.1. Experimental design and animal management The experimental site was approximately 10 km northwest of Campobasso, Molise, Italy (latitude: 418340 N and longitude: 148390 E), with an elevation of about 650 m above sea level. The climate of this area is Mediterranean, with about 650 mm of annual rainfall, mainly distributed in late autumn and winter, and 15.7 and 8.48C mean maximum and minimum temperatures over the last 40 years. In this area, ¯ocks are maintained on pasture during the daytime and moved to a shed at night, except for the middle part of the winter season during which sheep are not on pasture, due to weather severity and reduced herbage availability. The experiment, which lasted 18 weeks, was conducted in the spring and summer (April± August) of 1999 and involved 40 Comisana ewes that had lambed in winter. At weaning (50 3 days after lambing), ewes were divided into two groups of 20 each, balanced for age, parity, time of lambing, number of lambs suckled, body weight (44:6 0:7 kg, means S:E.) and milk yield and composition. The indoor group was housed in a 3 m 12 m straw bedded pen inside a pre-fabricated building throughout the study period, whereas outdoor group was maintained in a 200 m2 paddock, designated with a wire-mesh fence, during daytime (09.00±19.00 h) and moved to a pen of the same area and adjacent to the pen that indoor ewes were housed in during nighttime. The experimental building was 4 m high and provided with transom windows placed at a height of 2.5 m from the ¯oor. The paddock was on ¯at ground and herbage was mowed to ground level at the commencement of the experiment. Pens and paddock were provided with two mangers and a hay crib (feeder space 0:4 m per animal). The feeding area in the paddock being covered by a 3 m 4 m wooden canopy. Experimental pens were not cleaned during the study period, but a layer of straw (about 0.3 kg/m2) was added daily to each pen. Ewes were fed on a diet composed of 67.5% alfalfa hay and 32.5% pelleted concentrate, which was
D. Casamassima et al. / Small Ruminant Research 41 (2001) 151±161
administered in three meals daily (09.00, 14.30 and 21.00 h) in equal amounts to the experimental groups. The chemical composition of dry matter was determined by AOAC (1990) methods and was 16.3% crude protein, 3.1% fat, by ether extract, 20.7% crude ®bre and 8.6% ash. Throughout the trial, the amount of food given to the ewes was modi®ed based on the mean productive levels, keeping the forage to concentrate ratio 0.625±0.375 constant. Average daily DM intake was 2.2 kg per ewe during the study period. Ambient temperature and relative humidity inside the pre-fabricated building and in the paddock were monitored throughout the trial using thermo-hygrographs (LSI, 20090 Settala Premenugo, Italy) placed at a height of 2 m from the ¯oor. Temperature±humidity index (THI) was calculated using the formula of Kelly and Bond (1971). 2.2. Behavioral observation Behavioral observations were recorded by two trained observers equipped with video cameras from 09.00 to 19.00 h at fortnightly intervals. Three different focal animals were chosen at random each time in both groups. All three of the focal animals were videotaped and the tapes were then reviewed continuously for each animal. Times spent in each of the following seven behavioral categories were recorded: eating, drinking, ruminating, lying, walking, idling and other activities (social and aggressive interactions, scratching oneself, bleating and fence or manger or bucket biting). 2.3. In vivo cell-mediated immunity The phytohemagglutinin (PHA) skin test was performed to induce a non-speci®c delayed-type hypersensitivity. At weeks 6, 11, 15 and 18 of the experiment, 1 mg of PHA (Sigma-Aldrich, 20151 Milan, Italy) dissolved in 1 ml of sterile saline solution was injected intradermally into the middle of two 2 cm wide circles stamped on shaved skin in the upper side of each shoulder. The skinfold thickness was determined before PHA injection and 24 h after with a caliper. For each animal an average increase in skinfold thickness (24 h thickness Ð pre-injection thickness) was compared using an average of the two measurements gathered from each shoulder.
153
2.4. Blood sampling and analyses Jugular blood samples were taken from ewes at the beginning of the experiment and at weeks 5, 10, 14 and 18 of the trial. Blood samplings were collected in 10 ml vacuum tubes at 07.00 h. Samples were centrifuged for 20 min at 3500 rpm and blood serum was stored at 208C except for a sub-sample which was immediately analyzed enzymatically for serum concentrations of glucose. The following metabolites and enzymes were determined within 3 days from blood sampling using standard procedures and a semi-automatic UVICON 941 spectrophotometer (Kontron, Milan, Italy): total protein, albumin, total bilirubin, NEFA, calcium and inorganic phosphorus, using colorimetrical methods, and total cholesterol, triglycerides, urea, creatinine, aspartate-amintransferase (AST/GOT), alanine-aminotransferase (ALT/GPT), lactate-dehydrogenase (LDH) and alkaline phosphatase (ALP), using enzymatical methods. The concentrations of cortisol, free-triiodothyronine (F-T3) and free-thyroxine (F-T4) were determined using an enzyme-linked ¯uorescent assay (ELFA) and commercial reagents (Biomerieux, F-69280 Marcy l'Etoile, France). 2.5. Sampling and analyses of milk Ewes were hand milked twice daily (08.00 and 20.00 h). Morning and evening milk was collected in metal buckets and weighed using an electronic scale. Milk samples, consisting of proportional volumes of morning and evening milk, were individually collected in plastic containers at 14 days interval, refrigerated (48C) and transported to our laboratory. Upon arrival at the laboratory (30± 60 min after collection), the following measurements were made on milk: pH, total protein, fat and lactose content, using an IR spectrophotometer (Milko Scan 133B, Foss Electric, DK-3400 Hillerùd, Denmark) according to the IDF (1990) standard; somatic cell count (SCC), using a Foss Electric Fossomatic 90 cell counter (IDF, 1995); and renneting parameters (clotting time, rate of clot formation and clot ®rmness after 30 min) using a Foss Electric Formagraph and the method of Zannoni and Annibaldi (1981). At the beginning of the experiment and on the day before each milk sampling, all the ewes were
154
D. Casamassima et al. / Small Ruminant Research 41 (2001) 151±161
examined carefully to assess the absence of signs of clinical mastitis, such as fever, pain or gland swelling. A small quantity of milk was checked visually for signs of mastitis. Body weights of the ewes were recorded at the beginning and at weeks 9 and 18 of the study period after the machine milking in the morning but before feeding. 2.6. Statistical analysis Data from two ewes, one from the indoor and the other from the outdoor group, were excluded from the data set due to suspected mastitis. Behavioral data were analyzed using the Wilcoxon rank-sum test in the NPAR1WAY procedure of the SAS (1990) statistical software, having housing system and time as categorical factors. Milk and blood variables were tested for normality using the Shapiro and Wilk (1965) test and milk SCC were log-transformed to normalize their frequency distributions before performing statistical analysis. A linear model that included the effects of housing system, time, housing system time and residual error was used (SAS, 1990). Pre-treatment values were used as covariates for blood parameters. Body weights and body weight changes of the ewes
were analyzed using a linear model that included housing system and residual error. 3. Results 3.1. Meteorological data Averages of indoor and outdoor temperatures during daytime (09.00±19.00 h) were substantially similar throughout the experiment (Table 1). Differences only exceeded 38C during weeks 5, 13 and 17. Through the trial, humidity ranged from 41 to 63% in the shed and from 33 to 72% in the paddock. During weeks 15, 17 and 18, averages of indoor and outdoor THI exceeded 72, which is considered the upper critical THI at least for lactating cows (Armstrong, 1994). Outdoor THI values were also over 72 during weeks 9, 10 and 13. 3.2. Behavior Feeding activity (Fig. 1) was not affected by treatment (26.9 versus 21.7% in indoor and outdoor group, respectively, P 0:079), whereas drinking was signi®cantly lower in housed than in outdoor reared
Table 1 Averages of indoor and outdoor temperatures (8C), humidities (%) and temperature±humidity indexes (THI) during daytime (09.00±19.00 h) Trial week
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Temperature
Relative humidity
THI
Indoor
Outdoor
Indoor
Outdoor
Indoor
Outdoor
14.0 11.7 14.6 20.7 17.8 20.8 18.4 21.9 24.7 24.1 20.6 21.1 25.0 24.0 26.9 22.6 30.4 31.5
12.8 10.9 14.2 23.0 22.0 21.7 17.0 23.4 25.4 25.6 21.7 22.6 28.1 22.9 25.1 21.6 26.9 28.5
55.0 56.9 63.1 59.6 61.7 51.0 56.4 51.4 41.0 44.0 56.3 41.1 47.6 51.0 67.7 61.7 51.6 47.5
61.0 59.0 67.0 53.4 72.0 64.1 72.2 52.7 53.7 68.4 44.3 46.1 62.1 42.1 55.9 54.7 33.6 32.9
57.4 54.2 58.2 66.7 62.8 66.3 63.4 67.8 70.5 70.1 66.4 66.1 71.5 70.6 76.4 69.6 79.1 79.8
55.7 53.1 57.6 69.4 69.5 68.5 61.9 69.9 72.7 74.6 67.1 68.3 77.4 68.4 72.2 67.7 72.2 73.9
D. Casamassima et al. / Small Ruminant Research 41 (2001) 151±161
155
Fig. 1. Behavioral traits of ewes when allocated to an indoor or outdoor environment during daytime. Values are least square means S:E.
animals (0.9 versus 2.1%). Time spent in ambulatory activities was about three-fold greater in the outdoor than in indoor ewes (13.4 versus 4.5%, P < 0:001). Conversely, the incidence of idling was higher in indoor compared to outdoor ewes (21.5 versus 13.3%, P < 0:05). Irrespective of housing system, eating and ruminating decreased (P < 0:001) over time, whereas lying, idling and drinking activities increased (P < 0:001). Aggressive interactions and
anomalous behaviors were sporadically observed in both indoor and outdoor ewes. 3.3. Immune response The skinfold thickness after PHA injection (Fig. 2) was similar across treatments throughout the experiment (5.31 and 5.11 mm, in indoor and outdoor groups, respectively, P 0:407).
Fig. 2. Immune response to PHA injection in ewes when allocated to an indoor or outdoor environment during daytime. Values are least square means S:E.
156
D. Casamassima et al. / Small Ruminant Research 41 (2001) 151±161
3.4. Endocrine response Neither cortisol, nor F-T3, or F-T4 blood levels (Fig. 3) were changed by housing system. However, cortisol concentrations tended to decrease in
outdoor compared to indoor ewes at week 14 of the study (8.36 versus 9.67 ng/ml, P 0:081). An effect of time of sampling was observed for F-T3 (P < 0:05) and F-T4 (P < 0:01), with ewes having the lowest blood levels of both thyroid hormones
Fig. 3. Plasma levels of cortisol, free-triiodothyronine and free-thyroxine in ewes when allocated to an indoor or outdoor environment during daytime. Values are least square means S:E.
D. Casamassima et al. / Small Ruminant Research 41 (2001) 151±161
157
Table 2 Changes in blood metabolites and enzymes in ewes when allocated to an indoor or outdoor environmenta Item
Trial week
Glucose (mmol/l)
Urea (mmol/l)
Group
S.E.
Indoor (n 19)
Outdoor (n 19)
5 10 14 18
3.37 a 2.80 3.68 a 4.07
3.06 b 2.71 2.18 b 3.69
5 10 14 18
6.36 7.69 9.54 a 5.38
5.68 6.26 6.86 b 6.14
Effects, P Treatment
Time
Treatment time
0.2 0.1 0.2 0.2
0.0001
0.0767
0.0001
0.6 0.6 0.6 0.4
0.0075
0.4986
0.0336
Creatinine (mmol/l)
5 10 14 18
51.3 b 30.5 11.5 13.0
81.3 a 35.2 13.6 14.6
4.1 2.5 2.1 1.4
0.0011
0.0001
0.0051
AST/GOT (U/l)
5 10 14 18
58.2 a 40.5 46.0 b 38.4
32.3 b 38.5 58.6 a 33.6
2.6 3.1 2.9 3.1
0.0180
0.2532
0.0001
5 10 14 18
13.8 8.6 b 8.4 7.9
11.6 16.9 a 10.0 8.5
1.2 2.6 0.7 0.7
0.0605
0.8205
0.0023
0.1 0.1 0.2 0.1
0.0464
0.9961
0.3685
ALT/GPT (U/l)
Inorganic phosphorus (mmol/l)
a
5 10 14 18
1.31 b 1.78 1.32 b 1.19
1.86 a 1.93 2.22 a 1.19
Values are least square means S.E. Means followed by different letters differ signi®cantly at P < 0:05.
Fig. 4. Milk yield (g per day) of ewes when allocated to an indoor or outdoor environment during daytime. Values are least square means S:E.
158
D. Casamassima et al. / Small Ruminant Research 41 (2001) 151±161
at week 18 of the experiment regardless of housing system. 3.5. Blood metabolites and enzymes Blood glucose levels were higher in indoor than in outdoor ewes at weeks 5 (P < 0:05) and 14 (P < 0:001) of the experiment (Table 2), whereas blood concentrations of glucose only tended to increase (P 0:06) in housed compared to outdoor reared animals at week 18. Indoor ewes also exhibited higher urea (P < 0:001) and lower AST/GOT levels
(P < 0:01) than outdoor animals at week 14. In contrast, the latter group had higher levels of blood creatinine (P < 0:001) at week 5, of ALT/GPT (P < 0:05) at week 10 and of inorganic phosphorus (P < 0:01) at weeks 5 and 14 of the experiment. No differences were found for any of the other parameters measured. 3.6. Milk yield Average milk yield and constituents were not affected by housing system. However, signi®cant
Fig. 5. Total protein and fat content (%) in the milk of ewes when allocated to an indoor or outdoor environment during daytime. Values are least square means S:E.
D. Casamassima et al. / Small Ruminant Research 41 (2001) 151±161
159
Table 3 Somatic cell count (SCC), pH and renneting parameters of milk in ewes when allocated to an indoor or outdoor environmenta Item
Group
SCC (cells/ml 103) pH Clotting time (min) Rate of clot formation (min) Clot firmness (mm) a
S.E.
Indoor (n 19)
Outdoor (n 19)
437 6.45 9.82 2.16 59.8
224 6.37 9.56 2.14 62.8
49 0.01 0.39 0.07 1.61
Effects, P Treatment
Time
Treatment time
0.0031 0.0001 0.6385 0.7985 0.3111
0.1947 0.0001 0.0001 0.0003 0.0001
0.0464 0.0001 0.0761 0.6511 0.2004
Values are least square means S.E.
time treatment interactions were found for all parameters investigated. Indeed, indoor ewes yielded greater volumes of milk (P < 0:05) at weeks 2 and 6 of the experiment than outdoor ewes (Fig. 4). In addition, housed animals had higher milk contents (P < 0:05) of protein at weeks 8 and 14, of fat (Fig. 5) at weeks 6 and 14 and of lactose at week 8 of the experiment compared to outdoor reared ewes. Milk from outdoor ewes had a lower pH (P < 0:001) and somatic cell counts (P 0:003) than indoor animals, with the difference being greater during the second half of the trial (Table 3). In contrast, no signi®cant differences were found in the coagulating properties of milk between groups. 3.7. Body weight change Body weight and body weight changes were similar across treatments. On average, weight gains were 44:6 17:2 and 45:0 21:1 g per day in the indoor and outdoor groups, respectively. 4. Discussion Outdoor ewes spent more time walking than indoor animals, due to a greater available space and possibly to a stronger kinetic drive to investigate the environment (Carson, 1985). Walking is part of exploratory behavior (Gluesing and Balph, 1980) and it is well known that sheep spend a long time exploring any new ®eld they are put in (Fraser and Broom, 1990). Indoor ewes spent more time idling compared to outdoor animals, probably as a consequence of an increased threshold of stimuli perception due to a narrower range of stimuli experienced (Wemelsfelder, 1993).
Inactive behaviors in group-housed animals have also been interpreted as a strategy to reduce physical interactions with conspeci®cs (Fraser and Broom, 1990; Barnett et al., 1992; Hanlon et al., 1994). The reduction in milk yield and the rise in ambient temperature with the advancement of the trial were probably responsible for decreased feeding and ruminating activities of ewes and for increased thermoregulatory behaviors, such as lying, idling and drinking (Shref¯er and Hohenboken, 1980). Friend et al. (1985) did not observe signi®cant changes of neutrophil and lymphocyte counts in the blood of Holstein calves when housed in pens, hutches and yards. Accordingly, no differences in the immune response were found in this study between indoor and outdoor reared animals. These results are consistent with the lack of differences in blood cortisol concentrations, because the levels of circulating corticosteroids are known to have a direct effect on ewe immune function (Cockram et al., 1994). The changes in climatic conditions through the study period may explain the behavior of both cortisol and thyroid hormones through the experiment. Indeed, animals respond to rising THI values with a reduction in the secretion of thermogenic hormones to lower their basal metabolism (Habeeb et al., 1992). Reduction in ewe metabolic rate with the decline of lactation also contributed to lowering of blood serum levels of thyroid hormones at week 18 of the experiment (Hart, 1983). Increased energy demand and muscular tissue metabolism, due to enhanced locomotion, resulted in higher blood levels of creatinine and lower blood glucose concentrations in outdoor compared to indoor ewes at week 5 of the experiment (Finco, 1989). The reduction of glucose levels in outdoor reared animals
160
D. Casamassima et al. / Small Ruminant Research 41 (2001) 151±161
was probably magni®ed by the exposure to rising THI values at week 14 of the experiment (Herbein et al., 1985). The reduction in cortisol levels supports the existence of a direct relation between climatic conditions and glucose changes (Alvarez and Johnson, 1973). An inverse relation between glucose and inorganic phosphorus concentrations have also been found by Calamari et al. (1991) in the plasma of calves subjected to restricted feeding. These authors postulated that the increment of inorganic phosphorus levels may be the result of an increased glucose catabolism with a pending energy de®cit situation. Accordingly, in the present study, lower glucose levels were associated with higher inorganic phosphorus concentrations in the blood serum of outdoor reared animals at weeks 5 and 14 of the experiment. The differences in blood serum urea concentrations at week 14 might be ascribed to either a nitrogen intake in excess of requirements to indoor ewes or a reduced use of amino acids for gluconeogeneis in outdoor animals (Calamari et al., 1991). However, an increased urinary nitrogen excretion, that has been described as a metabolic reaction of dairy animals to thermal stress (Habeeb et al., 1992), was not excluded to interpret the changes of urea levels in the blood serum of outdoor ewes at this time. Differences in ALT/GPT and AST/GOT blood levels among groups are not easy to explain. Ronchi et al. (1999) observed a transaminase increase in the plasma of lactating cows subjected to restricted feeding, hypothesizing that it was an adaptive response to energy de®cit leading to an enhanced metabolic activity of the liver. Reduced energy and nitrogen availability to the mammary gland resulted in a transient drop of milk yield and deterioration of milk quality in the outdoor group. The SCC did not signi®cantly change in the milk from outdoor ewes throughout the experiment. Conversely, the indoor group had a marked SCC increase in milk during the second half of the trial, probably due to a progressive worsening of air and litter condition, as a consequence of fecal accumulation and trampling by indoor reared animals. 5. Conclusion In this study, housing system did not affect the endocrine and immune responses of lactating ewes,
although indoor and outdoor reared animals had different behavioral responses, as far as locomotor activity and inactive behavior. Outdoor enclosure of ewes resulted in a lower milk SCC than indoor con®nement as well as in reduced yield and nutritional quality of milk. These ®ndings suggest that outdoor enclosure could be bene®cial to the behavioral needs of lactating ewes, by satisfying their motivation for walking around (Petherick and Rushen, 1997), though no signi®cant advantage in terms of stress related indicators were found. Acknowledgements Research supported by the Italian Ministero della Ricerca Scienti®ca e Tecnologica (MURST ex 40%) ``Effect of housing system on sheep behavior, welfare and production performance'' Scienti®c director: Prof. Donato Casamassima. Thanks are due to Dr. Julie Morrow-Tesch for her precious help with English. References Alliston, J.C., Lucas, I.A.M., 1979. Response of Welsh mountain ewes to ¯ushing and to housing for all or part of the winter. Anim. Prod. 28, 257±269. Alvarez, M.B., Johnson, H.D., 1973. Environmental heat exposure on cattle plasma catecholamine and glucocorticoids. J. Dairy Sci. 56, 189±194. AOAC, 1990. Of®cial Methods of Analysis, 15th Edition. Association of Of®cial Analytical Chemists, Washington, DC, pp. 69±88. Armstrong, D.V., 1994. Heat stress interaction with shade and cooling. J. Dairy Sci. 77, 2044±2050. Barnett, J.L., Hemsworth, P.H., Cronin, G.M., Newman, E.A., McCallum, T.H., Chilton, D., 1992. Effects of pen size, partial stalls and method of feeding on welfare-related behavioral and physiological responses of group-housed pig. Appl. Anim. Behav. Sci. 34, 207±220. Berge, E., 1997. Housing of sheep in cold climate. Livest. Prod. Sci. 49, 139±149. Brosh, A., Aharoni, Y., Degen, A.A., Wright, A., Young, B.A., 1998. Effects of solar radiation, dietary energy and time of feeding on thermoregulatory responses and energy balance in cattle in a hot environment. J. Anim. Sci. 76, 2671±2677. Calamari, L., Bertoni, G., Cappa, V., 1991. Endocrine and metabolic effect of prolonged fasting in Friesian calves. In: Proceedings of the 9th Congresso Nazionale ASPA. Rome, pp. 299±310. Carson, K., 1985. Kinesis. In: Fraser, A.F. (Ed.), Ethology of Farm Animals, Elsevier, Amsterdam, NL, pp. 209±214.
D. Casamassima et al. / Small Ruminant Research 41 (2001) 151±161 Celi, R., Cosentino, E., Melodia, L., Di Trana, A., 1993. Changes in production performance and body weight of lactating ewe as affected by rearing system and concentrate supplementation during pregnancy. Zootecnica Nutrizione Animale 19, 141± 149. Cockram, M.S., Ranson, M., Imlah, P.J., Burrells, C., Harkiss, G.D., 1994. The behavioral, endocrine and immune responses of sheep to isolation. Anim. Prod. 58, 389±399. Curtis, S.E., 1983. Environmental Management in Animal Agriculture. Iowa State Press, Ames. Done-Currie, J.R., Hecker, J.F., Wodzicka-Tomaszewska, M., 1984. Behavior of sheep transferred from pasture to an animal house. Appl. Anim. Behav. Sci. 12, 121±130. Finco, D.R., 1989. Kidney function. In: Kaneko, J.J. (Ed.), Clinical Biochemistry of Domestic Animals, 4th Edition. Academic Press, Saint Diego, CA, pp. 497±538. Fraser, A.F., Broom, D.M., 1990. Farm Animal Behavior and Welfare. Bailliere Tindall, London. Friend, T.H., Dellmeier, G.R., Gbur, E.E., 1985. Comparison of four methods of calf con®nement. I. Physiology. J. Anim. Sci. 60, 1095±1101. Gluesing, E.A., Balph, D.F., 1980. An aspect of feeding behavior and its importance to grazing systems. J. Range Manage. 33, 426±427. Habeeb, A.A.M., Marai, I.F.M., Kamal, T.H., 1992. Heat stress. In: Phillips, C., Piggins, D. (Eds.), Farm Animals and the Environment. CAB International, Wallingford, UK, pp. 27±47. Hanlon, A.J., Rhind, S.M., Reid, H.W., Burrells, C., Lawrence, A.B., Milne, J.A., McMillen, S.R., 1994. Relationship between immune response, liveweight gain, behavior and adrenal function in red deer (Cervus elaphus) calves derived from wild and farmed stock, maintained at two housing densities. Appl. Anim. Behav. Sci. 41, 243±255. Hart, I.C., 1983. Endocrine control of nutrient partition in lactating ruminants. Proc. Nutr. Soc. 42, 181±194. Hartung, J., 1994. The effect of airborne particulate on livestock health and production. In: Ap Dewi, I., Axford, R.F.E., Marai, I.F.M., Omed, H. (Eds.), Pollution in Livestock Production Systems. CAB International, Wallingford, UK, pp. 55±79. Herbein, J.H., Aiello, R.J., Eckler, L.I., Pearson, R.E., Akers, R.M., 1985. Glucagon, insulin, growth hormone, and glucose concentrations in blood plasma of lactating dairy cows. J. Dairy Sci. 68, 320±325. Hughes, B.O., Duncan, I.J.H., 1988. The notion of ethological need, models of motivation and animal welfare. Anim. Behav. 36, 1696±1707. IDF, 1990. Determination of Milk Fat, Protein and Lactose Content Ð Guide for the Operation of Mid-Infra-Red Instruments. IDF
161
Standard No. 141B. International Dairy Federation, Brussels, Belgium. IDF, 1995. Enumeration of Somatic Cells. IDF Standard No. 148A. International Dairy Federation, Brussels, Belgium. Kelly, C.F., Bond, T.E., 1971. Bioclimatic Factors and Their Measurement: A Guide to Environmental Research on Animals. National Academy of Sciences, Washington, DC. Lynch, J.J., Hinch, G.N., Adams, D.B., 1992. The Behavior of Sheep. Biological Principles and Implications for Production. CAB International, Wallingford, UK. McNatty, K.P., Young, A., 1973. Diurnal changes of plasma cortisol levels in sheep adapting to a new environment. J. Endocrinol. 56, 329±330. Palestrini, C., Ferrante, V., Mattiello, S., Canali, E., Carenzi, C., 1998. Relationship between behavior and heart rate as an indicator of stress in domestic sheep under different housing systems. Small Rum. Res. 27, 177±181. Pearson, R.A., Mellor, D.J., 1976. Some behavioral and physiological changes in pregnant goats and sheep during adaptation to laboratory conditions. Res. Vet. Sci. 20, 215±217. Petherick, J.C., Rushen, J., 1997. Behavioral restriction. In: Appleby, M.C., Hughes, B.O. (Eds.), Animal Welfare. CAB International, Wallingford, UK, pp. 89±105. Ronchi, B., Bernabucci, U., Lacetera, N., Verini Supplizi, A., Nardone, A., 1999. Distinct and common effects of heat stress and restricted feeding on metabolic status of Holstein heifers. Zootecnica Nutrizione Animale 25, 11±20. SAS, 1990. SAS/STAT User's Guide, Version 6, 4th Edition. Statistical Analysis System Inst., Cary, NC. Sevi, A., Massa, S., Annicchiarico, G., Dell'Aquila, S., Muscio, A., 1999. Effect of stocking density on ewes milk yield, udder health and microenvironment. J. Dairy Res. 66, 489±499. Shapiro, S.S., Wilk, M., 1965. An analysis of variance test for normality. Biometrika 52, 591±601. Shref¯er, C., Hohenboken, W., 1980. Circadian behavior, including thermoregulatory activities, in feedlot lambs. Appl. Anim. Behav. Sci. 6, 241±246. Wemelsfelder, F., 1993. The concept of animal boredom and its relationship to stereotyped behavior. In: Lawrence, A.B., Rushen, J. (Eds.), Stereotypic Animal Behavior: Fundamentals and Applications to Welfare. CAB International, Wallingford, UK, pp. 65±95. Wemelsfelder, F., Birke, L., 1997. Environmental challenge. In: Appleby, M.C., Hughes, B.O. (Eds.), Animal Welfare. CAB International, Wallingford, UK, pp. 35±47. Zannoni, M., Annibaldi, S., 1981. Standardization of the renneting ability of milk by Formagraph. Scienza e Tecnica LattieroCasearia 32, 9±94.