|
|
||||||||

* Department of Pharmacology, Microbiology, and Food Hygiene, The Norwegian School of Veterinary Science, N-0033 Oslo
Department of Animal Science, Agricultural University of Norway, N-1432 Ås
Corresponding author:
Siri Kulberg; e-mail:
siri.kulberg{at}veths.no.
| ABSTRACT |
|---|
|
|
|---|
All cows had increased total white blood cells, neutrophils, and neutrophil/lymphocyte ratio as parturition approached, with peak levels at parturition. Cows selected for low clinical mastitis had lower levels of total white blood cells and neutrophils compared with cows selected for high protein yield throughout all three periods. The difference was significant prepartum. Cows from the low clinical mastitis group also had lower neutrophil/lymphocyte ratios and lower serum cortisol levels than those in the high protein yield group. A significant positive correlation was found between cortisol and total white blood cells and neutrophils, respectively.
We conclude that selection for lower mastitis incidence in the low clinical mastitis group is associated with a significant decrease in total white blood cells and neutrophils in blood prepartum. Results from the present study, and the genetic difference in mastitis incidence observed in the groups, indicates that increased leukocyte mobilization at certain stages of lactation may be associated with increased susceptibility to mastitis.
Abbreviation key: HPY = High protein yield, LCM = low clinical mastitis, N/L ratio = neutrophil/lymphocyte ratio, NRF = Norwegian Cattle
Key Words: dairy cow selected cattle group parturition leukocyte
| INTRODUCTION |
|---|
|
|
|---|
Despite the immunosuppression, increasing numbers of total white blood cells are found in the peripheral blood of periparturient animals, mainly due to a rise in neutrophil concentration. Peak values are observed at parturition, but levels decline shortly after and reach basal conditions within 2 wk postpartum (Detilleux et al., 1995; Kimura et al., 1999; Klinkon and Zadnik, 1999).
Neutrophils are crucial in protecting the body against invading pathogens. As a first group of defense they migrate rapidly into tissue where they phagocytose and kill invaders at sites of inflammation. Physiologically high periparturient blood concentrations of glucocorticoids have been suggested as an explanation for the increased neutrophil count around parturition (Burton et al., 1995; Lee and Kehrli, 1998; Kehrli et al., 1999).
Cortisol is a well-accepted indicator for stress in animals and man. A release of adrenocorticotropic hormone from the pituitary gland occurs in late pregnancy. Adrenocorticotropic hormones induce synthesis and secretion of glucocorticoids from the adrenal cortex. It is well-known that glucocorticoids have an immunosuppressive effect (Kehrli et al., 1999). Although the exact mechanism is not completely understood, it is believed that glucocorticoids increase the bone marrow output of neutrophils and induces an enzymatic cleavage of L-selectin on the neutrophil surface. Reduced expression of L-selectin shifts the neutrophil population from the marginating pool to the peripheral circulation, and results in a larger circulating pool of neutrophils less capable of migrating towards sites of inflammation. Consequently animals become more susceptible to infections (Roth and Kaeberle, 1982; Burton et al., 1995; Lee and Kehrli, 1998).
The unfavorable genetic correlation between mastitis and milk yield (Heringstad et al., 2000), implies that single trait selection for increased milk production is expected to increase susceptibility to mastitis. In 1989, GENO Breeding and A. I. Association and Department of Animal Science, Agricultural University of Norway started a dairy cattle selection experiment that includes two selection groups of Norwegian Cattle (NRF); one high protein yield (HPY) group and one low clinical mastitis (LCM) group. More than 10 years of selection have resulted in two groups of cows with genetically different levels of susceptibility to clinical mastitis. The cows from this experiment represent a unique resource for studies of the underlying mechanisms causing a realized reduction in mastitis frequency among animals selected for improved health.
The aim of the present study was to observe the effect of different lactation stages on serum cortisol levels and hematological properties, especially neutrophil count, in the two groups of NRF selected for high protein yield or low clinical mastitis, respectively. We hypothesize that there might be differences in total white blood cells and neutrophil counts between the two groups.
| MATERIALS AND METHODS |
|---|
|
|
|---|
Cows in the HPY group were mated to the two to four highest ranking proven sires with regard to protein yield each year, and cows in the LCM group were mated to the two to four best proven sires for mastitis resistance each year.
All cows present in the herd when the selection experiment started that were mated with one of the defined sires and produced a daughter that made a first lactation record of mastitis and/or protein yield were defined as cow-generation 0, their daughters were defined as cow generation 1, and so on. After three cow generations, the genetic difference between HPY and LCM cows was equivalent to 15.4 kg protein (305 d lactation yield, kg protein) and 8.6% clinical mastitis (B. Heringstad, 2002, unpublished). These estimated genetic differences were based on EBV for cows from linear animal model analysis of data for the total NRF population from 1978 to 1999. The mean EBV for the 34 sires used in the HPY group and the 28 sires used in the LCM group in the period from 1989 to 1998 was 0.0179 and –0.0727 for clinical mastitis and 24.36 and 2.83 for protein yield, respectively (B. Heringstad, 2002, unpublished).
Animals and Blood Sampling
Due to practical reasons, the four herds closest to the Norwegian School of Veterinary Science were chosen for this study (Buskerud, Jønsberg, Kalnes, and Mære). These herds do not differ from the other four herds except in location. Animals were chosen for sampling according to their expected calving dates, and the number of animals sampled in each herd according to the size of the herd. Animals, both primiparous and multiparous, with overlapping calving dates were chosen. An equal number of animals were sampled from each selection group.
Blood samples were taken from the jugular vein 2 wk prior to calving, as soon as possible after calving, and 1 mo after calving. Milk samples were collected from all four quarters as soon as possible after calving and 1 mo after calving. Because of the distances and the number of animals sampled, local veterinarians took most samples.
On each occasion, a 10 ml sample of blood was withdrawn into Venoject glass tubes containing EDTA as anticoagulant (Terumo Europe N.V., Leuven, Belgium) for hematological analysis, while another 10 ml sample was withdrawn into Venoject glass tubes with no anticoagulant (Terumo) to provide serum samples. The hematological analyses were performed shortly after sampling. The serum samples were stored frozen at –20°C until analyzed.
Because several persons were involved in the sampling procedure, and predicting calving dates accurately is quite difficult, some of the samples were taken too late or too early in relation to the actual calving date or were not taken at all. In addition, some of the animals were culled during the sampling period. Finally, a total of 121 animals were sampled from the four herds (Table 1
). 16 animals were sampled once, 26 animals were sampled twice, and 81 animals were sampled three times. There were no major differences in average age of the cows sampled, or in mean sampling day pre- or postpartum (Table 2
).
|
|
A total of 85 cows were sampled outside the sampling periods and excluded (Table 1
). There was no discrepancy between the groups. A total of 51 samples were excluded due to indications of udder infection (Table 1
). The discrepancy seen between the groups were due to different numbers of animals sampled from each group. Additionally, some of the cows with indications of udder infection had a SCC within normal ranges, and it is questionable if these samples represent true udder infections. However, these samples were excluded the same way as samples with both pathogens in the milk and an increased SCC. In the LCM group, ten of these samples were found, compared with only four in the HPY group.
At the end, 91 cows were included in the study (Table 1
), and more than 75% of these were from cow-generation 2 and 3; 4, 7, and 12 cows were from cow-generation 0, 1, and 4, respectively. The mean EBV for clinical mastitis for sires of the 53 LCM cows and sires of the 38 HPY cows were –0.004 and –0.079, respectively, a difference equivalent to 7.5% clinical mastitis. The corresponding mean EBV for protein yield (305 d lactation yield, kg protein) were 3.35 and 26.19 for sires of the LCM and HPY cows.
Leukocyte Counting and Cortisol Analysis
Haematological analyses were performed at the Central Laboratory, Norwegian School of Veterinary Science, Oslo. Total white blood cells (x 106 /ml) and differential counts (x 106 /ml) of monocytes, lymphocytes, neutrophils, eosinophils, and basophils, were calculated using a Technicon ADVIA 120 (Technicon Industrial Systems, Tarrytown, NY).
Serum cortisol levels were analyzed by radioimmunoassay (Simensen et al., 1978) at the Department of Reproductive Physiology and Pathology, Norwegian School of Veterinary Science, Oslo.
Statistical Analysis
All results are expressed as mean values with 95% confidence intervals constructed using the Student procedure (Altmann, 1991). Standard deviation was used as the index for dispersion. Differences between groups were considered statistically significant if the P-value was less than or equal to a level of 5%.
P-values were adjusted for multi-significance using the Bonferroni correction. An ANOVA expectation component model was used to compare groups, when variables could be assumed normally distributed (Altmann, 1991).
Assumption of normality was rejected for neutrophils and cortisol. In these cases, the Wilcoxon two-sample test was used (Lehmann, 1975).
The Pearson linear correlation analysis was used to study the connection pattern between the included variables (Altmann, 1991). All 169 single samples from the 91 animals included were used in the correlation analysis. The residual correlation for the four parameters, total white blood cells, neutrophils, the neutrophil/lymphocyte (N/L) ratio, and cortisol was calculated by correction for animal group and stage of lactation.
| RESULTS |
|---|
|
|
|---|
|
|
|
0.02), neutrophil concentration (P
0.002), and the N/L ratio (P
0.01) was significant prepartum, but not at parturition nor at peak lactation. As for the other leukocytes, monocyte and basophil levels increased during parturition, whereas lymphocyte and eosinophil levels decreased. There were no significant differences between the groups.
Significant, positive correlations were found between total white blood cells and neutrophils (r = 0.81, P < 0.001), total white blood cells and N/L ratio (r = 0.39, P < 0.001), and neutrophils and N/L ratio (r = 0.78, P < 0.001). The results shown are correlations corrected for differences due to genetic group and stage of lactation.
Cortisol
Serum cortisol levels (Figure 4
) increased as parturition approached, with peak levels at parturition. Slightly decreasing levels were found postpartum. Throughout all three periods, serum cortisol levels were higher in the HPY group than in the LCM group, although not significantly.
|
| DISCUSSION |
|---|
|
|
|---|
The N/L ratio in both selection groups increased as parturition approached. In the HPY group, the N/L ratio was higher than in the LCM group on all sampling occasions. The difference was significant prepartum. An increased N/L ratio has been described previously in animals under conditions of stress (McGlone et al., 1993; Stull and Rodiek, 2000).
Serum cortisol levels were higher in the HPY group than the LCM group at all stages of lactation. Levels increased up to parturition, where they reached their peak. After parturition, the cortisol concentration slowly decreased.
Experimental injections with synthetic glucocorticoids to create artificial stress similar to the stress seen in the periparturient period induced marked neutrophilia and down-regulation of L-selectin on neutrophil membranes. The neutrophils became less able to migrate and the susceptibility to infections increased (Burton et al., 1995; Nakagawa et al., 1999). The present study detected a significant positive correlation between neutrophils and the plasma cortisol level in the two groups.
The periparturient period induces physiological stress in cattle. It has been shown that transport-induced stress in different species leads to elevated cortisol blood concentrations, increases in total white blood cell and neutrophil levels, and an elevated N/L ratio (Dalin et al., 1993; McGlone et al., 1993; Stull and Rodiek, 2000). Animals highly susceptible to stress would almost certainly produce higher levels of cortisol in the periparturient period than less susceptible individuals, and would as a result have higher total white blood cells and neutrophil concentrations in blood than the less susceptible animals.
Different basal cortisol levels in blood have been observed among breeds with different susceptibility to stress (Zavy et al., 1992; Garcia-Belenguer et al., 1996), suggesting a genetic effect of this parameter. This results in permanently higher physiological total white blood cell and neutrophil numbers. On the other hand, cattle selected for fighting and dominance ability, although having the same basal cortisol levels as other breeds, were less fearful to surprising-effect tests, and experienced a less pronounced rise in cortisol levels, than other breeds, when subjected to stress (Plusquellec and Bouissou, 2001).
Whether the rise in cortisol level is permanently present, or only under temporary conditions of stress, it nevertheless may result in neutrophils being less capable of migration and phagocytosis, and consequently in animals with higher susceptibility to infectious diseases.
In the present study, increases in total white blood cells, neutrophils and the N/L ratio were significantly higher in the HPY group than in the LCM group prepartum.
Hematological changes prepartum are stress related, previously connected with the periparturient period. It is not possible to conclude solely based on this investigation whether the changes observed prepartum are permanent or if they are induced by the periparturient period. However, it is reasonable to suggest that the changes observed are either due to periparturient stress, and that the HPY group may be more affected by this than the LCM group, or that the differences may be permanent and due to higher basal cortisol levels in the HPY group. A difference in leukocyte recruitment between the groups is also a possibility and may represent a stronger inflammatory response upon local stimulation in the HPY group.
| CONCLUSION |
|---|
|
|
|---|
Further studies are needed to draw a firm conclusion. Analysis of neutrophil counts and baseline cortisol levels in the groups at other stages of lactation, will also reveal possible differences here, or support the hypothesis that the differences are related to the periods investigated in this study.
| ACKNOWLEDGEMENTS |
|---|
|
|
|---|
Received for publication October 9, 2001. Accepted for publication June 7, 2002.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
K. K. M. O'Driscoll, M. M. Schutz, A. C. Lossie, and S. D. Eicher The effect of floor surface on dairy cow immune function and locomotion score J Dairy Sci, September 1, 2009; 92(9): 4249 - 4261. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Saebo, T. Almoy, B. Heringstad, G. Klemetsdal, and A. H. Aastveit Genetic Evaluation of Mastitis Resistance Using a First-Passage Time Model for Wiener Processes for Analysis of Time to First Treatment J Dairy Sci, February 1, 2005; 88(2): 834 - 841. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |