Effect of Dietary Vitamin A and p-Carotene on Polymorphonuclear Leukocyte and Lymphocyte Function in Dairy Cows During the Early Dry Period1 L. W. TJOELKER, B. P. CHEW, T. S. TANAKA, and L. R. DANIEL Department of Animal Sciences Washington State University Pullman 99164-6320 ABSTRACT
(Key words: lymphocyte)
Vitamin A and ~-carotene improved mammary health in dairy cows around dry off. To define possible mechanisms, cows were fed 1) 53,000 IV vitamin A, 2) 213,000 IV vitamin A, or 3) 53,000 IV vitamin A plus 400 mg ~-carotene/cow per d (n = to/treatment) from 6 wk before to 2 wk after dry off. Blood polymorphonuclear neutrophil function (phagocytosis, kill, and chemotaxis) and lymphocyte proliferation were measured at wk -{), 0 (dry off), and 2. Concentrations of vitamin A in serum did not differ across vitamin treatments. ~-Carotene in serum was elevated in cows fed ~-carotene. Treatment did not influence phagocytosis or kill. Kill ability increased after dry off in all treatment groups, but phagocytosis tended to decrease after dry off in cows fed vitamin A only. Lymphocyte blastogenesis stimulated by concanavalin A on wk 2 for cows fed 53,000 IV vitamin A but did not vary in the other two groups. Lipopolysaccharide-stimulated blastogenesis peaked at wk 0 and then decreased to pretreatment values by wk 2 in cows fed 213,000 IV vitamin A. These data indicate lymphocyte function is influenced by vitamin A supplementation and that ~-carotene supplementation seems to exert a stabilizing effect on neutrophil and lymphocyte function during the period around dry off.
leukocytes,
INTRODUCTION
Received May 1. 1989. Accepted September 25.1989. IResearch supponed by the USDA Animal Health Granl 86-CSRS-Z-286Z. Washington Dairy Products CommiSSIOn and the College of Agriculture and Home Economics Research Center, Washington State University. Pullman. Published as Scientific Paper Number 7r09. 1990 J Dairy Sci 73:1017-1022
~-carotene,
Dairy cows fed high amounts of vitamin A and ~-carotene during the prepartum period had lower postpartum milk sce than cows fed vitamin A alone (6). Additionally, lower plasma vitamin A has been associated with higher milk sec (13). Subsequent studies reported lower incidences of mammary infections and higher spontaneous cure rates in cows supplemented with vitamin A and ~-carotene than in cows fed vitamin A alone during the dry period, even though concentrations of plasma vitamin A were similar among treatment groups (7). Therefore, vitamin A and ~-carotene may promote mammary health. The mechanisms b~ which vitamin A and ~-carotene exert their protective effects against mastitis have not been defined. Two major components of host defense, the phagocytic cells and lymphocytes, are important in mammary immunity. Polymorphonuclear neutrophils (PMN) are considered the most important line of defense against invading mammary pathogens (25). However, induction of PMN infiltration into bovine (22) and guinea pig (9) mammary glands depen~s largely upon T lymphocytes. Furthermore, opttmum phagocytosis of bacteria is dependent upon bacteria-specific immunoglobulin, the production of which involves the interactions of macrophages and B and T lymphocytes (3). Therefore, it seems likely that the protective effects of vitamin A and ~-carotene may be mediated by enhanced phagocyte and lymphocyte function. Numerous studies lend credence to our hypothesis. Ongsakul et aI. (24) reported decreased phagocytic capacity of blood PMN from vitamin A-deficient rats. Tjoelker et al. (2?) demonstrated the stimulatory influence of
1017
1018
TJOELKER ET AL.
retinol and retinoic acid on in vitro phagocytosis and intracellular kill of live Staphylococcus aureus by bovine milk PMN. These observations may be partly explained by reports that certain retinoids (retinol. retinoic acid, alltrans-retinal) stimulate superoxide production in vitro by human and guinea pig PMN (1, 12, 17). Superoxide is an important intermediate in the oxygen-dependent antimicrobial function of the PMN (15). Numerous reports illustrate similar beneficial effects of vitamin A and j3-carotene on lymphocyte function (4). Vitamin A deficiency is reflected in compromised lymphocyte function; whereas vitamin A sufficiency, both in vivo and in vitro, promotes proper lymphocyte function. In addition, Daniel et al. (8) reported that ~-carotene, as well as vitamin A, enhanced mitogen-stimulated lymphocyte proliferation of bovine blood lymphocytes in vitro; the ~-carotene and vitamin A status of these animals was unknown. Clearly, vitamin A and j3-carotene are important for proper PMN and lymphocyte function. However, it is not known whether it is through these avenues that the vitamins promote mammary health. Therefore, the objective of this study was to examine the effects of vitamin A and ~-carotene supplementation to dairy cows around dry off on in vitro PMN and lymphocyte function. MATERIALS AND METHODS
Thirty Holstein cows were used. The study was conducted at Washington State University Knott Dairy Center from May 1986 through February 1987 and portions of the study have been reported (28. 29). Cows were selected on the basis of lactation number and mammary health. Prior to inclusion in the study, cows in their third or earlier lactation were screened for mammary infection. Quarter milk samples were collected aseptically three separate times within a 3-wk period (18). Only cows with one or no quarters infected were used. Six weeks before dry off, cows were assigned to one of three dietary treatments: 1) 53,000 IV vitamin A palmitate/cow per d (AVA), 2) 213,000 IV vitamin A palmitate/ cow/d (HVA), or 3) 53.000 IV vitamin A palmitate + 400 mg j3-carotene/cow per d (VA+BC). We chose to use 400 mg ~-carotene, Journal of Dairy Science Vol. 73,
No.4, 1990
because 300 mg used in our previous study (7) were inadequate for maintaining plasma ~-caro tene at a constant high concentration; instead. ~-carotene in plasma gradually declined. Both vitamin A palmitate and ~-carotene were supplied by Hoffmann-La Roche Inc. (Nutley, NJ). Cows fed VA+BC received the equivalent of 213,000 IU vitamin A based on a conversion ratio of 1 mg ~-carotene = 400 IV vitamin A (19). This allowed a meaningful assessment of any ~-carotene effects to be made by comparing treatments HVA and VA+BC. All cows were fed a 75:25 mixture of an alfalfa grass silage and low quality grass hay (containing no detectable ~-carotene). Attempts to increase the proportion of the low quality hay to reduce ~ carotene intake failed due to refusal. Concentrates fed during lactation were formulated to contain no added preformed vitamin A but were otherwise balanced for all other essential nutrients (21). Vitamin supplements were topdressed daily at 0700 h from 6 wk before through 2 wk after dry off. Cows were blocked in groups of three, and the cows within each block randomly assigned to the treatments. Blood was collected on wk --6, 0 (day of dry off), and 2. Blood collected for vitamin A and ~-carotene analyses were processed under subdued light and serum was stored at -20T in darkness until analysis. For isolating lymphocytes and PMN, blood was collected aseptically from the coccygeal vein into tubes containing 10% sodium citrate (wtl vol) (.5 mlliO ml blood). The isolation of blood lymphocytes (29) and PMN (28) have been described. The methods used in the lymphocyte blastogenesis assay are as described by Tjoelker et aI. (29). Concanavalin A (Con A; 10 Ilg/ml) and Escherichia coli lipopolysaccharide (LPS; 100 Ilglml) were the mitogens used. Assays for phagocytic (PI) and intracellular kill (KI) abilities of blood PMN have also been described (29). Chemotaxis by blood PMN was measured by a migration under agarose assay as previously described (20). This assay compared random migration with directional migration toward a chemoattractant as a means of assessing the chemotactic responsiveness of the PMN. The chemoattractant used was zymosan-activated serum (ZAS). Zymosan (Sigma Chemical Co., St. Louis. MO) was washed in phosphate-
1019
MAMMARY HEALTH AND DIET
buffered saline (25 mg zymosan/ml) by vigorous vortexing followed by centrifugation (1000 x g, 10 min). The supernatant was discarded and the zymosan pellet mixed with serum (25 mg/ml) pooled from 3 healthy mature cows. The mixture was vortexed, incubated for 30 min at 37"C, and again centrifuged. The supernatant was stored in 20(HJ.I aliquots at -20T until used. Medium 199 containing 1% agarose (Sigma) and 10% fetal bovine serum, pH 7.4, were pipetted into 60 x 15-mm petri dishes (Falcon, Oxnard, CA) and allowed to solidify. Using a template, 8 sets of triplicate wells, 2.4 mm in diameter and 2.4 mm apart, were cut into the agarose in an outwardly radiating configuration. With each triplicate, the outer well received 10 III of ZAS. the central well received 10 III of PMN suspension (2.5 x 107/ml) and the inner well received 10 III of control medium (Roswell Park Memorial Institute, RPMI-I640). Cells from each cow were assayed in quadruplicate. Following a 2-h incubation at 37°C in a humidified 95% air:5% C02 atmosphere, the agarose was removed and the adherent PMN stained with Wright stain. Migration was determined by measuring the distance between the edge of the center well and the leading front of cells migrating towards the ZAS and the control wells. The chemotactic index (CI) was calculated by: CI
= (Distance +
to ZAS Distance to control) - 1.
Statistical Analysis
Data were analyzed by the General Linear Models procedure of SAS (2). Least squares analysis using a randomized complete block, split-plot design was used to test treatment and sampling differences in Kl, PI and CI, in lymphocyte proliferation stimulated by Con A and LPS, and in serum vitamin A and J3-carotene concentrations. The statistical model used was: Yijkl = 11 + Ti + Bj (Tj x Bj) + Pk + (Tj x Pic.) + eijkl where:
o
ADEQU,"TE VITAMI
/>.
lSS':l HIGH VITAM I A 1m VITAMIN A + ff-CAR:JTENE 12.0 -
-
X W 0
110
U
10.5 l-
~
10.0
0
9.5
z
u
115
'-' «
9.0
CL
8.5
I
8.0
:
r
~ ~
T
I
i
tI
if i II -6
j
.
0
~
2
WEEK RELATIVE TO DRYOFF (DRYOFF = 0) Figure I. Phagocytosis of Slaphylococcus aureus by blood polymorphonuclear leukocytes from cows fed/cow per d adequate vitamin A (53.000 IV). high vitamin a (213.000 IU), or vitamin A (53.000 IU) plus Ii--earotene (400 mg). Data represent least squares means ± SEM.
Yijld
11 Ti
Bj Ti x Bj Pk Ti x Pk eijld
= KI,
PI and CI, proliferation stimulated by Con A- and LPS, and concentrations of serum vitamin A and ~-carotene; = overall mean; = effect of the treatment i, i = I to 3 where 1 = group AVA, 2 = group HVA, and 3 = group VA+BC; = effect of the block j; = interaction of the treatment and the block j (error A); = effect of the sampling period k; = interaction of the treatment i and the sampling period k; and = random residuals.
RESULTS Polymorphonuclear Neutrophil Function
Phagocytosis. Vitamin treatment had no effect on PI at each sampling period (Figure 1). The PI was similar for all treatments between wk -6 and 0 but tended to decrease by wk 2 in cows in groups AVA and HVA but not in group VA+BC. Intracellular Kill. Kill index was not different between treatment groups at all sampling Jownal of Dairy Science Vol. 73.
No.4, 1990
1020
TJOELKER ET AL. c=J ADEQUATE VITAMIN A
o
16
ADEQUATE VITAMIN A HIGH VITAMIN A i555 VITAMIN A + P-CAROTENE
~
~
.60
15
~55
x
2::
14
~
.50
13 ~
U
.45
Z
()
z
::J ---l
HIGH VITAMIN A + P-CAROTENE
i555 VITAMIN A
f= U
~
12
o
~
~
11
W I U
-6
0
2
WEEK RELATIVE TO DRYOFF (DRYOFF = 0) Figure 2. Kill of Staphylococcus aureus by blood polymorphonuclear leukocytes from cows fed/cow per d adequate vitamin A (53,000 IV), high vitamin A (213,000 IU). or vitamin A (53,000 IV) plus ~-earolene (400 mg). Data represent least squares means ± SEM.
periods (Figure 2). There was no significant difference in KI between wk -6 and O. However, KI increased (P
(Key words: lymphocyte)
Vitamin A and ~-carotene improved mammary health in dairy cows around dry off. To define possible mechanisms, cows were fed 1) 53,000 IV vitamin A, 2) 213,000 IV vitamin A, or 3) 53,000 IV vitamin A plus 400 mg ~-carotene/cow per d (n = to/treatment) from 6 wk before to 2 wk after dry off. Blood polymorphonuclear neutrophil function (phagocytosis, kill, and chemotaxis) and lymphocyte proliferation were measured at wk -{), 0 (dry off), and 2. Concentrations of vitamin A in serum did not differ across vitamin treatments. ~-Carotene in serum was elevated in cows fed ~-carotene. Treatment did not influence phagocytosis or kill. Kill ability increased after dry off in all treatment groups, but phagocytosis tended to decrease after dry off in cows fed vitamin A only. Lymphocyte blastogenesis stimulated by concanavalin A on wk 2 for cows fed 53,000 IV vitamin A but did not vary in the other two groups. Lipopolysaccharide-stimulated blastogenesis peaked at wk 0 and then decreased to pretreatment values by wk 2 in cows fed 213,000 IV vitamin A. These data indicate lymphocyte function is influenced by vitamin A supplementation and that ~-carotene supplementation seems to exert a stabilizing effect on neutrophil and lymphocyte function during the period around dry off.
leukocytes,
INTRODUCTION
Received May 1. 1989. Accepted September 25.1989. IResearch supponed by the USDA Animal Health Granl 86-CSRS-Z-286Z. Washington Dairy Products CommiSSIOn and the College of Agriculture and Home Economics Research Center, Washington State University. Pullman. Published as Scientific Paper Number 7r09. 1990 J Dairy Sci 73:1017-1022
~-carotene,
Dairy cows fed high amounts of vitamin A and ~-carotene during the prepartum period had lower postpartum milk sce than cows fed vitamin A alone (6). Additionally, lower plasma vitamin A has been associated with higher milk sec (13). Subsequent studies reported lower incidences of mammary infections and higher spontaneous cure rates in cows supplemented with vitamin A and ~-carotene than in cows fed vitamin A alone during the dry period, even though concentrations of plasma vitamin A were similar among treatment groups (7). Therefore, vitamin A and ~-carotene may promote mammary health. The mechanisms b~ which vitamin A and ~-carotene exert their protective effects against mastitis have not been defined. Two major components of host defense, the phagocytic cells and lymphocytes, are important in mammary immunity. Polymorphonuclear neutrophils (PMN) are considered the most important line of defense against invading mammary pathogens (25). However, induction of PMN infiltration into bovine (22) and guinea pig (9) mammary glands depen~s largely upon T lymphocytes. Furthermore, opttmum phagocytosis of bacteria is dependent upon bacteria-specific immunoglobulin, the production of which involves the interactions of macrophages and B and T lymphocytes (3). Therefore, it seems likely that the protective effects of vitamin A and ~-carotene may be mediated by enhanced phagocyte and lymphocyte function. Numerous studies lend credence to our hypothesis. Ongsakul et aI. (24) reported decreased phagocytic capacity of blood PMN from vitamin A-deficient rats. Tjoelker et al. (2?) demonstrated the stimulatory influence of
1017
1018
TJOELKER ET AL.
retinol and retinoic acid on in vitro phagocytosis and intracellular kill of live Staphylococcus aureus by bovine milk PMN. These observations may be partly explained by reports that certain retinoids (retinol. retinoic acid, alltrans-retinal) stimulate superoxide production in vitro by human and guinea pig PMN (1, 12, 17). Superoxide is an important intermediate in the oxygen-dependent antimicrobial function of the PMN (15). Numerous reports illustrate similar beneficial effects of vitamin A and j3-carotene on lymphocyte function (4). Vitamin A deficiency is reflected in compromised lymphocyte function; whereas vitamin A sufficiency, both in vivo and in vitro, promotes proper lymphocyte function. In addition, Daniel et al. (8) reported that ~-carotene, as well as vitamin A, enhanced mitogen-stimulated lymphocyte proliferation of bovine blood lymphocytes in vitro; the ~-carotene and vitamin A status of these animals was unknown. Clearly, vitamin A and j3-carotene are important for proper PMN and lymphocyte function. However, it is not known whether it is through these avenues that the vitamins promote mammary health. Therefore, the objective of this study was to examine the effects of vitamin A and ~-carotene supplementation to dairy cows around dry off on in vitro PMN and lymphocyte function. MATERIALS AND METHODS
Thirty Holstein cows were used. The study was conducted at Washington State University Knott Dairy Center from May 1986 through February 1987 and portions of the study have been reported (28. 29). Cows were selected on the basis of lactation number and mammary health. Prior to inclusion in the study, cows in their third or earlier lactation were screened for mammary infection. Quarter milk samples were collected aseptically three separate times within a 3-wk period (18). Only cows with one or no quarters infected were used. Six weeks before dry off, cows were assigned to one of three dietary treatments: 1) 53,000 IV vitamin A palmitate/cow per d (AVA), 2) 213,000 IV vitamin A palmitate/ cow/d (HVA), or 3) 53.000 IV vitamin A palmitate + 400 mg j3-carotene/cow per d (VA+BC). We chose to use 400 mg ~-carotene, Journal of Dairy Science Vol. 73,
No.4, 1990
because 300 mg used in our previous study (7) were inadequate for maintaining plasma ~-caro tene at a constant high concentration; instead. ~-carotene in plasma gradually declined. Both vitamin A palmitate and ~-carotene were supplied by Hoffmann-La Roche Inc. (Nutley, NJ). Cows fed VA+BC received the equivalent of 213,000 IU vitamin A based on a conversion ratio of 1 mg ~-carotene = 400 IV vitamin A (19). This allowed a meaningful assessment of any ~-carotene effects to be made by comparing treatments HVA and VA+BC. All cows were fed a 75:25 mixture of an alfalfa grass silage and low quality grass hay (containing no detectable ~-carotene). Attempts to increase the proportion of the low quality hay to reduce ~ carotene intake failed due to refusal. Concentrates fed during lactation were formulated to contain no added preformed vitamin A but were otherwise balanced for all other essential nutrients (21). Vitamin supplements were topdressed daily at 0700 h from 6 wk before through 2 wk after dry off. Cows were blocked in groups of three, and the cows within each block randomly assigned to the treatments. Blood was collected on wk --6, 0 (day of dry off), and 2. Blood collected for vitamin A and ~-carotene analyses were processed under subdued light and serum was stored at -20T in darkness until analysis. For isolating lymphocytes and PMN, blood was collected aseptically from the coccygeal vein into tubes containing 10% sodium citrate (wtl vol) (.5 mlliO ml blood). The isolation of blood lymphocytes (29) and PMN (28) have been described. The methods used in the lymphocyte blastogenesis assay are as described by Tjoelker et aI. (29). Concanavalin A (Con A; 10 Ilg/ml) and Escherichia coli lipopolysaccharide (LPS; 100 Ilglml) were the mitogens used. Assays for phagocytic (PI) and intracellular kill (KI) abilities of blood PMN have also been described (29). Chemotaxis by blood PMN was measured by a migration under agarose assay as previously described (20). This assay compared random migration with directional migration toward a chemoattractant as a means of assessing the chemotactic responsiveness of the PMN. The chemoattractant used was zymosan-activated serum (ZAS). Zymosan (Sigma Chemical Co., St. Louis. MO) was washed in phosphate-
1019
MAMMARY HEALTH AND DIET
buffered saline (25 mg zymosan/ml) by vigorous vortexing followed by centrifugation (1000 x g, 10 min). The supernatant was discarded and the zymosan pellet mixed with serum (25 mg/ml) pooled from 3 healthy mature cows. The mixture was vortexed, incubated for 30 min at 37"C, and again centrifuged. The supernatant was stored in 20(HJ.I aliquots at -20T until used. Medium 199 containing 1% agarose (Sigma) and 10% fetal bovine serum, pH 7.4, were pipetted into 60 x 15-mm petri dishes (Falcon, Oxnard, CA) and allowed to solidify. Using a template, 8 sets of triplicate wells, 2.4 mm in diameter and 2.4 mm apart, were cut into the agarose in an outwardly radiating configuration. With each triplicate, the outer well received 10 III of ZAS. the central well received 10 III of PMN suspension (2.5 x 107/ml) and the inner well received 10 III of control medium (Roswell Park Memorial Institute, RPMI-I640). Cells from each cow were assayed in quadruplicate. Following a 2-h incubation at 37°C in a humidified 95% air:5% C02 atmosphere, the agarose was removed and the adherent PMN stained with Wright stain. Migration was determined by measuring the distance between the edge of the center well and the leading front of cells migrating towards the ZAS and the control wells. The chemotactic index (CI) was calculated by: CI
= (Distance +
to ZAS Distance to control) - 1.
Statistical Analysis
Data were analyzed by the General Linear Models procedure of SAS (2). Least squares analysis using a randomized complete block, split-plot design was used to test treatment and sampling differences in Kl, PI and CI, in lymphocyte proliferation stimulated by Con A and LPS, and in serum vitamin A and J3-carotene concentrations. The statistical model used was: Yijkl = 11 + Ti + Bj (Tj x Bj) + Pk + (Tj x Pic.) + eijkl where:
o
ADEQU,"TE VITAMI
/>.
lSS':l HIGH VITAM I A 1m VITAMIN A + ff-CAR:JTENE 12.0 -
-
X W 0
110
U
10.5 l-
~
10.0
0
9.5
z
u
115
'-' «
9.0
CL
8.5
I
8.0
:
r
~ ~
T
I
i
tI
if i II -6
j
.
0
~
2
WEEK RELATIVE TO DRYOFF (DRYOFF = 0) Figure I. Phagocytosis of Slaphylococcus aureus by blood polymorphonuclear leukocytes from cows fed/cow per d adequate vitamin A (53.000 IV). high vitamin a (213.000 IU), or vitamin A (53.000 IU) plus Ii--earotene (400 mg). Data represent least squares means ± SEM.
Yijld
11 Ti
Bj Ti x Bj Pk Ti x Pk eijld
= KI,
PI and CI, proliferation stimulated by Con A- and LPS, and concentrations of serum vitamin A and ~-carotene; = overall mean; = effect of the treatment i, i = I to 3 where 1 = group AVA, 2 = group HVA, and 3 = group VA+BC; = effect of the block j; = interaction of the treatment and the block j (error A); = effect of the sampling period k; = interaction of the treatment i and the sampling period k; and = random residuals.
RESULTS Polymorphonuclear Neutrophil Function
Phagocytosis. Vitamin treatment had no effect on PI at each sampling period (Figure 1). The PI was similar for all treatments between wk -6 and 0 but tended to decrease by wk 2 in cows in groups AVA and HVA but not in group VA+BC. Intracellular Kill. Kill index was not different between treatment groups at all sampling Jownal of Dairy Science Vol. 73.
No.4, 1990
1020
TJOELKER ET AL. c=J ADEQUATE VITAMIN A
o
16
ADEQUATE VITAMIN A HIGH VITAMIN A i555 VITAMIN A + P-CAROTENE
~
~
.60
15
~55
x
2::
14
~
.50
13 ~
U
.45
Z
()
z
::J ---l
HIGH VITAMIN A + P-CAROTENE
i555 VITAMIN A
f= U
~
12
o
~
~
11
W I U
-6
0
2
WEEK RELATIVE TO DRYOFF (DRYOFF = 0) Figure 2. Kill of Staphylococcus aureus by blood polymorphonuclear leukocytes from cows fed/cow per d adequate vitamin A (53,000 IV), high vitamin A (213,000 IU). or vitamin A (53,000 IV) plus ~-earolene (400 mg). Data represent least squares means ± SEM.
periods (Figure 2). There was no significant difference in KI between wk -6 and O. However, KI increased (P
