Veterinary Microbiology, 28 ( 1991 ) 189-198 Elsevier Science Publishers B.V., Amsterdam
189
Controlling acute Escherichia coli mastitis during the periparturient period with recombinant bovine interferon gamma Lorraine M. Sordillo and Lorne A. Babiuk ~2JterinaryInfectious Disease Organization, University of Saskatchewan, Saskatoon, Saskatchewan STN OWO Canada (Accepted 10 January 1991 )
ABSTRACT Sordillo, L.M. and Babiuk, L.A. 1991. Controlling acute Escherichia coli mastitis during the periparturient period with recombinant bovine interferon gamma. Vet. MicrobioL 28:189-198. The efficacy of recombinant bovine interferon (rBolFN)-gamma against experimentally induced Escherichia coli mastitis during the periparturient period was investigated. Dairy cows intramammarily treated with rBolFN-gamma 24 h before the E. coli challenge had fewer infected quarters, lower clinical scores, and infections of shorter duration when compared to placebo-treated animals. All rBolFN-gamma treated cows survived the experimental E. coli challenge. However, placebo treated cows had a 42% mortality rate attributed to coliform mastitis within 3 days of the challenge. Results from this study suggest that intramammary infusion of rBolFN-gamma can prevent the rapid, unrestricted growth of E. coli within the mammary gland and inhibit the subsequent development of an unlimited inflammatory response under experimental conditions. It is likely that controlling severe local inflammatory reactions may also decrease the pathological alterations to mammary parenchymal tissue that often accompanies acute coliform mastitis during the periparturient period. The potential for prophylactic treatment of perinatal dairy cows with rBolFN-gamma to regulate the rate, severity, and duration of naturally occurring coliform mastitis during periods of heightened susceptibility is discussed.
INTRODUCTION
Bovine coliform mastitis is regarded as a serious problem because it is severe, sometimes fatal, and difficult to control. Mastitis during the periparturient and early lactating periods is often caused by Gram negative bacteria, Escherichia coli being the most prevalent (Eberhart, 1977; Smith et al., 1985 ). Clinical E. coli mastitis is rare in middle and late lactation, and is most severe during the first few weeks of lactation, at which time it is a major cause of acute toxic mastitis (Eberhart, 1977; Hill, 1981; Smith et al., 1985). Establishment and severity ofE. coli mastitis with respect to stage of lactation have been explained by rate of growth of the organism within the gland, the elabo0378-1135/91/$03.50
© 1 9 9 1 - Elsevier Science Publishers B.V.
190
L.M. SORDILLO AND L.A. BABIUK
ration and absorption of toxins, and the susceptibility of the host. A rapid and intense inflammatory response in quarters from mid-lactation cows was observed following experimental challenge with E. coli and in most cases, the bacteria were eliminated promptly without causing damage to secretory parenchymal tissue. In contrast, quarters infected with E. coli during the immediate postpartum period generally have minimal neutrophil influx to the site of infection, resulting in the failure to eliminate the organism from the gland (Hill, 1981 ). Studies have shown that interferon (IFN)-gamma may be important in both specific i m m u n e reactions and nonspecific primary defenses. Interferongamma enhances natural killer cell activity, antibody-dependent cellular cytotoxicity, and cytotoxic T-lymphocyte activity (Lawman et al., 1990). Exposure to IFN-gamma also enhances the phagocytic and bacterial activities of macrophages and neutrophils (Lawman et al., 1990). We recently have shown that in vitro treatment of bovine m a m m a r y gland neutrophils with IFNgamma reversed the suppressive effects of m a m m a r y secretions and increased their antibacterial capabilities (Sordillo and Babiuk, 1990). Results from these in vitro data suggest that IFN-gamma may modulate neutrophil functions in vivo and facilitate the rapid clearance of bacteria from infected m a m m a r y glands. It has been hypothesized that IFN-gamma also may regulate host responses to lipopolysaccharide (LPS)-induced inflammation (Billiau, 1987; Heremans et al., 1987 ). When administered parenterally shortly before LPS challenge, IFN-gamma was shown to inhibit local inflammatory reactions. In contrast, local application of IFN-gamma intensified responses within the inflammatory focus (Heremans et al., 1987). Although the exact mode of action is still unclear, IFN-gamma appears to have an important role in the regulation of acute phase responses and uncontrolled inflammatory reactions which may lead to pathological alterations to host tissues. The objective of this study was to evaluate the influence of recombinant bovine (rBo)IFN-gamma on the establishment and severity of experimentally induced E. coli mastitis in dairy cattle during periods of increased susceptibility to intramammary infections. MATERIALS AND METHODS
Experimental animals Fourteen Holstein-Friesian dairy cows in their last trimester of pregnancy were used in this study. Beginning one month prior to experimental manipulation, duplicate quarter foremilk samples were collected aseptically at weekly intervals to ensure that all quarters were free of intramammary infection. Only cows with healthy appearances and normal rectal temperatures, heart rates, and respiration were used in this trial.
CONTROLLING MASTITIS WITH INTERFERON GAMMA
191
Experimental design Cows were divided randomly to either a rBoIFN-gamma treated (n -- 7 ) or placebo (n = 7) group within the first week of lactation. Treated cows were intramammarily infused into all four quarters with 105 U rBoIFN-gamma per quarter in 10 ml of sterile saline. The placebo treated group was infused intramammarily with 10 ml of sterile saline. All animals were challenged with approximately 50 CFU ofE. coli into each quarter at 24 h following experimental treatment. Mammary secretion samples were collected and rectal temperatures were recorded prior to challenge and again at 4 h intervals for 12 h. Additional samples were collected at 12 h intervals from 24 h to 60 h after challenge. Sampling was also done at 7 and 14 days after challenge as well. Mammary secretion samples were used to establish infection status, and to determine total milk somatic cell counts (SCC) of each quarter. Prior to each milking, a strip cup was used to observe evidence of mastitis in the first few streams of foremilk. The clinical status of all quarters was recorded immediately prior to experimental challenge and for 28 consecutive observations made at each milking. The clinical scores were recorded on a scale from 1 to 5 where: 1 is normal milk with no quarter swelling; 2 is questionable milk with no quarter swelling; 3 is obvious abnormal milk with no quarter swelling; 4 is abnormal milk with a swollen or tender quarter; and 5 is acute mastitis with systemic involvement (Smith et al., 1985 ).
Recombinant bovine interferon-gamma The rBoIFN-gamma was synthesized in E. coli by recombinant DNA technology and purified to homogeneity as determined by high-pressure liquid chromatography and sodium dodecyl sulfate-polyacrylamide gel electrophoresis (CIBA-GEIGY Limited, Basel, Switzerland). The stock solutions contained 5 mg protein/ml with a titer of 5 × 1 0 6 U / m g protein when titrated on MDBK cells challenged with vesicular stomatitis virus and calibrated to laboratory reference standards. Preparations contained less than 0.1 #g/ml of endotoxin as determined by the limulus amebocyte lysate test (Whittaker Corp., Walkersville, MD).
Challenge inoculum The challenge strain of E. coli was isolated from an acute case of bovine mastitis from the University of Saskatchewan Research Herd. Stock cultures of the serum resistant encapsulated strain was stored at - 70 ° C in skim milk until needed. The bacterial challenge was prepared by streaking the stock culture onto Esculin Blood Agar plates containing 5% whole blood. After 48 h incubation at 37 °C, a single colony was used to inoculate 100 ml of ultra high temperature ( U H T ) pasteurized milk and incubated for 12 h at 37°C. The 12 h culture was mixed well and a 100 #1 aliquot was removed to inoculate a second 100 ml volume of U H T milk. After a 6 h incubation at 37°C, the
192
L.M.SORDILLOANDL.A.BABIUK
culture was diluted serially in 10-fold increments using pyrogen-free saline. The C F U / m l value of each dilution was determined by plate pouring 100 #1 of the bacterial suspension onto McConkey agar plates. The dilution containing approximately 50 C F U of E. coli per ml of the pyrogen-free saline was selected for each trial. Inocula were plated post challenge to confirm viability counts.
Microbiological procedures Aseptic foremilk samples were collected at each sampling period to determine infection status of each quarter. Presumptive identification of bacterial isolates was determined by methods described by Brown et al. ( 1981 ). Gram negative isolates were identified to the species level using the API 20E System (Analytab Products, Plainview, NY). New intramammary infections following intracisternal challenge were identified by the isolation ofE. coli with the same API 20E profile as the challenge strain for three consecutive samplings.
Statistical analysis Data were analyzed by least squares analysis of variance using the general linear model procedure in the Statistical Analytical System (SAS, 1982) to determine effects of treatment and sample period on clinical scores. Preplanned comparisons of least square means were made by a pairwise t test. Means were contrasted between treatment groups within a sample period and between sample period within a treatment group. No other comparisons were made. Reductions in rate of new E. coli intramammary infections and persistance of E. coli mastitis in treated and control quarters were calculated using the normal approximation for comparing two binomially distributed percentages ( Steel and Torrie, 1982 ). The following formula was used: t = (PI - P 2 ) / [ ( P I Q I IN1 ) d- (P2Q2/N2) ]~ were t approximates a standard Student's t test, P = percent infected quarters, Q--percent uninfected quarters, and N = n u m b e r of quarters tested. Treatment groups are represented by the subscripts. RESULTS
Prior to the start of the experiment, periparturient cows were clinically normal. All quarters were free of intramammary infections and SCC values were < 500 000/ml. No changes in rectal temperature, clinical scores, or SCC were observed during the 12 h period following intramammary infusion of either rBoIFN-gamma or placebo (data not shown). Changes in mean SCC following experimental challenge with E. coli are shown in Fig. 1. Mean SCC for both treatment groups increased significantly
CONTROLLING MASTITISWITH INTERFERON GAMMA
193
10 8 -~-
m iz
o
C~
PLACEBO
--~
IFN
10 ~
w
0
10 ~
.05 ).
(P~< 0.01 ) beginning at 24 h and continuing until 60 h following the experimental challenge. No significant differences in SCC were observed between the rBolFN-gamma and placebo treatment groups up to 48 h following challenge. However, mean SCC values of placebo treated cows were elevated significantly over values for the rBolFN-gamma treated cows from 60 h to 336 h following experimental challenge. Mean SCC of rBolFN-gamma treated cows at the end of the trial were not significantly different from pre-challenge values. The effects of pre-treating animals with rBolFN-gamma on the rate of new E. coli intramammary infection is summarized in Table 1. Twenty one new E. coli infections or 91.3% of quarters were confirmed in the placebo treated
194
L.M. SORDILLO AND L.A. BABIUK
cows following experimental challenge. In the rBoIFN-gamma treated cows, 17 new E. coli infections or 70.8% o f quarters were confirmed. Compared with the placebo treatment group, cows intramammarily treated with rBoIFNgamma prior to experimental challenge had a 22.5% reduction in the number of new E. coli infections. Although the rate of new E. coli intramammary infections was lower in the rBoIFN-gamma treated cows, the reduction was not significantly different when compared to control animals. The effects o f pre-treating cows with rBoIFN-gamma on the persistance of E. coli mastitis during the experimental efficacy period is summarized in Table 2. All 21 new E. coli intramammary infections in the placebo treated cows persisted for 2 weeks following experimental challenge. In the rBoIFN-gamma treated cows, only 6 intramammary infections or 35% of the new infections persisted for 2 weeks following experimental challenge. Persistance of E. coli mastitis was reduced significantly (P~< 0.01 ) in rBoIFN-gamma treated cows when compared to placebo treated controls (64.7%). Effects of rBoIFN-gamma treatment on the severity of experimental E. coli TABLE2 Effects of rBolFN-gamma on persistance ofE. coli mastitis during the experimental challenge efficacy period. Treatment
No. new infections 2
No. infections after 14 days
Infections
groups 1 IFN Placebo
17 21
6 21
35.3 100
(%)
Reduction 3 (% ) 64.7
~Dairy cows were treated intramammarily with wither 105 U rBoIFN-gamma (IFN) or physiological saline (Placebo) at 24 h prior to experimental challenge. 2Number of quarters becoming infected after intracisternal challenge with E. coli. 3Significantly different from placebo treated cows (P~< 0.01 ). TABLE 3 Effects of r B o l F N - g a m m a on the severity of experimental E. coli mastitis. Treatment groups ~
No. eligible
Milkings
quarters 2
clinical 2
Average score 3
(%) I FN
24
16.7**
1.80**
Placebo
23
70.0
2.28
~Dairy cows were treated intramammarily with either 105 U rBoIFN-gamma (IFN) or physiological saline (Placebo) at 24 h prior to experimental challenge. 2Clinical score of >~ 2 recorded at each milking. 3Calculated from 28 consecutive milkings with clinical scores ranging from 1 to 5. **Significantly different from placebo-treated cows (P~< 0.01 ).
CONTROLLING
MASTITIS WITH INTERFERON
i
A
19 5
GAMMA
PLACEBO
IFN
-e
u.l O
O < O J
O
112,'
•
"~4~.T~. \ ~\~ ~I~ ~
*** 0
~.~
~
0
~ 4
~
~ 8
MILKINGS
t
12
_
~
_
~
I
16
FOLLOWING
I
Z~
20
i
2
I
24
I 28
CHALLENGE
Fig. 2. Effects of rBolFN-gamma on mean clinical scores throughout the 14 day experimental period. Each arrow indicates the death of a placebo-treated cow.
~
7
•
6
" O
5
~
4
>.> ~
3
e
1
O
O
2
42% Mortality
~//'~
IFN-GAMMA
~
PLACEBO
0
r
O~ ~ 2
--
2~4~'6
~8
6 " 0 ~ ~ "7
Hours
!L
1~ ' Days
j
Time After Challenge Fig. 3. Effects of rBolFN-gamma on the mortality rate associated with experimental E. coli mastitis.
mastitis is summarized in Table 3 and Figs. 2 and 3. Clinical scores for both treatment groups began to increase as early as 12 h following experimental challenge. Scores were elevated significantly (P~< 0.001 ) by 24 h (2.82 +_0.24) and reached peaked levels (3.67 ___0.24) by 36 h following E. coli challenge (Fig. 2 ). Cows treated with rBolFN-gamma exhibited clinical signs of mastitis at fewer ( P > 0.001 ) m i l k i n g s (16.7%) when compared to the placebo treatment group which exhibited clinical signs of mastitis at 70% of the milk-
196
L.M. SORDILLO AND L.A. BABIUK
ings (Table 3). In the rBoIFN-gamma treatment group, clinical signs were less pronounced (P~< 0.01 ) over the entire sampling period when compared to the placebo-treated cows (Table 3 ). Obvious differences in the number of cows surviving the experimental challenge were noted between the treatment groups (Figure 3). All cows infused intramammarily with rBoIFN-gamma survived the experimental E. coli challenge. However, cows treated with the saline placebo had a 42% mortality rate attributed to E. coli mastitis within 3 days of the experimental challenge. DISCUSSION
The acute symptoms most often associated with coliform mastitis are due to the rapid and unrestricted growth of the organism, the release ofendotoxin, and the subsequent development of an unlimited inflammatory reaction. Mammary gland neutrophils play a pivotal role in preventing or eliminating intramammary infections (Paape et al., 1981 ). Several studies have shown that the effectiveness of this defense system against E. coli mastitis depends on the promptness and magnitude of neutrophil migratory responses and their subsequent bactericidal capabilities at the foci of infection (Erskine et al., 1989; Hill, 1981; Lohuis et al., 1990). It is well documented that the antibacterial activities of m a m m a r y gland neutrophils are reduced in the presence of m a m m a r y gland secretions and comprised during physiological transitions of the m a m m a r y gland (Nagahata et al., 1988; Paape et al., 1981; Sordillo and Babiuk, 1990). The diminished antibacterial capacity of m a m m a r y gland neutrophils is attributable to several factors such as indiscriminate ingestion of milk constituents, degranulation, lack of energy source, and low concentrations of opsonins (Paape et al., 1981 ). Overcoming the immunosuppressive influence of m a m m a r y gland secretions on neutrophil function during the periparturient period may have a pronounced influence on the establishment and severity of E. coli mastitis. One approach to increase non-specific resistance to mastitis is to enhance neutrophil influx into the gland. Previous studies have shown that enhanced neutrophil migration to the m a m m a r y gland can control bacterial numbers during the early stages of infection and reduce the severity of coliform mastitis (Erskine et al., 1989). Improving the bactericidal efficiency of mammary gland neutrophils may also improve the cow's ability to limit bacterial growth. Recent studies have shown that in vitro treatment of m a m m a r y gland neutrophils with rBoIFN-gamma reversed the suppressive effects of mammary gland secretions and significantly increased the phagocytic and bactericidal activity when compared with untreated control cells (Sordillo and Babiuk, 1990 ). Although speculative at this point, the influence of local rBoIFNgamma administration on the establishment and severity of experimental E. coli mastitis in this study may reflect the effect of this cytokine on neutrophil
C O N T R O L L I N G MASTITIS WITH I N T E R F E R O N G A M M A
197
function in vivo. Prophylactic administration of rBoIFN-gamma did not significantly increase mean SCC over the placebo treated cows. However, cows treated with rBoIFN-gamma did have lower new infection rates and infections of shorter duration following experimental E. coli challenge when compared with controls. These results suggest that the bacterial capacity of neutrophils in rBoIFN-gamma treated quarters were more efficient at the time of experimental challenge. Enhancing an otherwise compromised defense mechanism during the initial stages of infection may control growth of the organism and minimize the presence of endotoxins in the gland, which would explain the lower infection rate and clinical scores in rBoIFN-gamma treated quarters. The clinical systemic signs associated with acute coliform mastitis result from the host's reaction to endotoxins released from killed bacteria. Pathophysiologic responses to endotoxins occur indirectly through the release and subsequent absorption of endogenous mediators from infected m a m m a r y glands and into the circulation (Lohuis et al., 1988 ). Cytokines (interleukin1, tumor necrosis factor, and IFN) are among these mediator molecules induced by LPS and which have an important role in the development of the acute inflammatory response (Billiau, 1987 ). Inflammatory changes occur in response to infectious pathogens. When cells or tissues are injured, the host attempts to protect and heal itself by initiating the inflammatory process. However, an exaggerated acute or protracted chronic inflammatory responses can cause extensive necrosis of m a m m a r y parenchymal tissues, loss of milk production, and mortality in dairy cows (Hill, 1981; Lohuis et al., 1990; Lohuis et al., 1988). Evidence indicates that IFN-gamma is a potent regulator of the acute phase response (Billiau, 1987 ). When administered prior to LPS challenge, IFN-gamma has been shown to down-regulate local inflammatory reactions (Billiau, 1987; Heremans et al., 1987). In this study, cows treated with rBoIFN-gamma prior to E. coli challenge had milder clinical reactions and lower mortality when compared with control cows. It is possible that rBoIFN-gamma inhibited the generation of endogenous inflammatory mediators in infected quarters which are responsible for systemic signs of coliform mastitis. The mechanisms by which rBoIFN-gamma modifies the outcome ofE. coli were subject to conjecture. However, intramammary infusion of rBoIFNgamma can prevent the rapid, unrestricted growth ofE. coli within the mammary gland and inhibit the subsequent development of an unlimited inflammatory response under experimental conditions. Controlling severe inflammation may decrease the pathological alterations to m a m m a r y parenchymal tissue and the tremendous milk production loss that often accompanies acute coliform mastitis during the periparturient period, thereby reducing the impact of this economically important form of mastitis. Prophylactic treatment of dairy cows with rBoIFN-gamma may potentially regulate the rate, severity,
198
L.M. SORDILLO AND L.A. BABIUK
and duration of naturally occurring coliform mastitis during periods of heightened susceptibility to infection. ACKNOWLEDGMENTS
The authors acknowledge with thanks the expert technical assistance of Lynn Gilbert. The assistance of the Animal Support Staff is appreciated also. This work was supported by CIBA-GEIGY Canada Limited, Mississauga, Ontario and the Natural Sciences and Engineering Research Council of Canada. Published with the approval of the Director of VIDO as journal series No. 105.
REFERENCES Billiau, A., 1987. Interferons and Inflammation. J. Interferon Res., 7:559. Brown, R.W., Barnum, D.A., Jasper, J.S., McDonald, J.S. and Schultze, W.D., 1981. Microbiological procedures for use in the diagnosis of bovine mastitis. 2nd Edn. Natl. Mastitis Counc., Inc., Arlington, VA, pp. 6-32. Eberhart, R.J., 1977. Coliform mastitis. J. Am. Vet. Med. Assoc., 170:1160. Erskine, R.J., Eberhart, R.J., Grasso, P.J. and Scholz, R.W., 1989. Induction ofEscherichia coli mastitis in cows fed selium-deficient or selium-supplemented diets. Am. J. Vet. Res., 50:20932100. Heremans, H., Dijkmans, R., Sobis, H., Vandekerckhove, F. and Billiau, A., 1987. Regulation by interferons of the local inflammatory response to bacterial lipopolysaccharide. J. Immunol., 138:4175. Hill, A.W., 1981. Factors influencing the outcome of Escherichia coli mastitis in the dairy cow. Res. Vet. Sci., 31:107. Lawman, M.J.P., Campos, M., Bielefeldt Ohmann, H., Griebel, P. and Babiuk, L.A., 1989. Recombinant cytokines and their potential therapeutic value in veterinary medicine. Comprehensive Biotechnology. Pergamon Press, London, pp. 63-106. Lohuis, J.A.C.M., Verheijden, J.H.M., Burvenich, C. and Van Miert, A.S.J.P.A.M., 1988. Pathophysiological effects ofendotoxins in ruminants. 1. Metabolic aspects. Vet. Q. 10:117. Lohuis, J.A.C.M., Schukken, Y.H., Verheijden, J.H.M., Brand, A. and Van Mien, A.S.J.P.A.M., 1990. Effect of severity of systemic signs during the acute phase of experimentally induced Escherichia coli mastitis on milk production losses. J. Dairy Sci. 73:333. Nagahata, H., Makino, S., Takeda, S., Takahashi, H. and Noda, H., 1988. Assessment of neutrophil function in the dairy cow during the perinatal period. J. Vet. Med. 35:747. Paape, M.J., Wergin, W.P., Guidry, A.J. and Schultze, W.D., 1981. Phagocytic defense of the ruminant mammary gland. Adv. Exp. Med. Biol., 137:555. SAS User's Guide: Statistics. 1982. SAS Inst. Inc., Cary, N.C. Smith, K.L., Todhunter, D.A. and Schoenberger, P.S., 1985. Environmental mastitis: cause, prevalence, prevention. J. Dairy Sci.,68:1531. Sordillo, L.M. and Babiuk, L.A., 1990. Modulation of bovine mammary neutrophil function during the periparturient period following in vitro exposure to recombinant bovine interferon gamma. Vet. Immunol. Immunopathol., 27: 393. Steel, R.G.D. and Torrie, J.H., 1982. Principles and procedures of statistics. McGraw-Hill Book Co., New York, N.Y.
189
Controlling acute Escherichia coli mastitis during the periparturient period with recombinant bovine interferon gamma Lorraine M. Sordillo and Lorne A. Babiuk ~2JterinaryInfectious Disease Organization, University of Saskatchewan, Saskatoon, Saskatchewan STN OWO Canada (Accepted 10 January 1991 )
ABSTRACT Sordillo, L.M. and Babiuk, L.A. 1991. Controlling acute Escherichia coli mastitis during the periparturient period with recombinant bovine interferon gamma. Vet. MicrobioL 28:189-198. The efficacy of recombinant bovine interferon (rBolFN)-gamma against experimentally induced Escherichia coli mastitis during the periparturient period was investigated. Dairy cows intramammarily treated with rBolFN-gamma 24 h before the E. coli challenge had fewer infected quarters, lower clinical scores, and infections of shorter duration when compared to placebo-treated animals. All rBolFN-gamma treated cows survived the experimental E. coli challenge. However, placebo treated cows had a 42% mortality rate attributed to coliform mastitis within 3 days of the challenge. Results from this study suggest that intramammary infusion of rBolFN-gamma can prevent the rapid, unrestricted growth of E. coli within the mammary gland and inhibit the subsequent development of an unlimited inflammatory response under experimental conditions. It is likely that controlling severe local inflammatory reactions may also decrease the pathological alterations to mammary parenchymal tissue that often accompanies acute coliform mastitis during the periparturient period. The potential for prophylactic treatment of perinatal dairy cows with rBolFN-gamma to regulate the rate, severity, and duration of naturally occurring coliform mastitis during periods of heightened susceptibility is discussed.
INTRODUCTION
Bovine coliform mastitis is regarded as a serious problem because it is severe, sometimes fatal, and difficult to control. Mastitis during the periparturient and early lactating periods is often caused by Gram negative bacteria, Escherichia coli being the most prevalent (Eberhart, 1977; Smith et al., 1985 ). Clinical E. coli mastitis is rare in middle and late lactation, and is most severe during the first few weeks of lactation, at which time it is a major cause of acute toxic mastitis (Eberhart, 1977; Hill, 1981; Smith et al., 1985). Establishment and severity ofE. coli mastitis with respect to stage of lactation have been explained by rate of growth of the organism within the gland, the elabo0378-1135/91/$03.50
© 1 9 9 1 - Elsevier Science Publishers B.V.
190
L.M. SORDILLO AND L.A. BABIUK
ration and absorption of toxins, and the susceptibility of the host. A rapid and intense inflammatory response in quarters from mid-lactation cows was observed following experimental challenge with E. coli and in most cases, the bacteria were eliminated promptly without causing damage to secretory parenchymal tissue. In contrast, quarters infected with E. coli during the immediate postpartum period generally have minimal neutrophil influx to the site of infection, resulting in the failure to eliminate the organism from the gland (Hill, 1981 ). Studies have shown that interferon (IFN)-gamma may be important in both specific i m m u n e reactions and nonspecific primary defenses. Interferongamma enhances natural killer cell activity, antibody-dependent cellular cytotoxicity, and cytotoxic T-lymphocyte activity (Lawman et al., 1990). Exposure to IFN-gamma also enhances the phagocytic and bacterial activities of macrophages and neutrophils (Lawman et al., 1990). We recently have shown that in vitro treatment of bovine m a m m a r y gland neutrophils with IFNgamma reversed the suppressive effects of m a m m a r y secretions and increased their antibacterial capabilities (Sordillo and Babiuk, 1990). Results from these in vitro data suggest that IFN-gamma may modulate neutrophil functions in vivo and facilitate the rapid clearance of bacteria from infected m a m m a r y glands. It has been hypothesized that IFN-gamma also may regulate host responses to lipopolysaccharide (LPS)-induced inflammation (Billiau, 1987; Heremans et al., 1987 ). When administered parenterally shortly before LPS challenge, IFN-gamma was shown to inhibit local inflammatory reactions. In contrast, local application of IFN-gamma intensified responses within the inflammatory focus (Heremans et al., 1987). Although the exact mode of action is still unclear, IFN-gamma appears to have an important role in the regulation of acute phase responses and uncontrolled inflammatory reactions which may lead to pathological alterations to host tissues. The objective of this study was to evaluate the influence of recombinant bovine (rBo)IFN-gamma on the establishment and severity of experimentally induced E. coli mastitis in dairy cattle during periods of increased susceptibility to intramammary infections. MATERIALS AND METHODS
Experimental animals Fourteen Holstein-Friesian dairy cows in their last trimester of pregnancy were used in this study. Beginning one month prior to experimental manipulation, duplicate quarter foremilk samples were collected aseptically at weekly intervals to ensure that all quarters were free of intramammary infection. Only cows with healthy appearances and normal rectal temperatures, heart rates, and respiration were used in this trial.
CONTROLLING MASTITIS WITH INTERFERON GAMMA
191
Experimental design Cows were divided randomly to either a rBoIFN-gamma treated (n -- 7 ) or placebo (n = 7) group within the first week of lactation. Treated cows were intramammarily infused into all four quarters with 105 U rBoIFN-gamma per quarter in 10 ml of sterile saline. The placebo treated group was infused intramammarily with 10 ml of sterile saline. All animals were challenged with approximately 50 CFU ofE. coli into each quarter at 24 h following experimental treatment. Mammary secretion samples were collected and rectal temperatures were recorded prior to challenge and again at 4 h intervals for 12 h. Additional samples were collected at 12 h intervals from 24 h to 60 h after challenge. Sampling was also done at 7 and 14 days after challenge as well. Mammary secretion samples were used to establish infection status, and to determine total milk somatic cell counts (SCC) of each quarter. Prior to each milking, a strip cup was used to observe evidence of mastitis in the first few streams of foremilk. The clinical status of all quarters was recorded immediately prior to experimental challenge and for 28 consecutive observations made at each milking. The clinical scores were recorded on a scale from 1 to 5 where: 1 is normal milk with no quarter swelling; 2 is questionable milk with no quarter swelling; 3 is obvious abnormal milk with no quarter swelling; 4 is abnormal milk with a swollen or tender quarter; and 5 is acute mastitis with systemic involvement (Smith et al., 1985 ).
Recombinant bovine interferon-gamma The rBoIFN-gamma was synthesized in E. coli by recombinant DNA technology and purified to homogeneity as determined by high-pressure liquid chromatography and sodium dodecyl sulfate-polyacrylamide gel electrophoresis (CIBA-GEIGY Limited, Basel, Switzerland). The stock solutions contained 5 mg protein/ml with a titer of 5 × 1 0 6 U / m g protein when titrated on MDBK cells challenged with vesicular stomatitis virus and calibrated to laboratory reference standards. Preparations contained less than 0.1 #g/ml of endotoxin as determined by the limulus amebocyte lysate test (Whittaker Corp., Walkersville, MD).
Challenge inoculum The challenge strain of E. coli was isolated from an acute case of bovine mastitis from the University of Saskatchewan Research Herd. Stock cultures of the serum resistant encapsulated strain was stored at - 70 ° C in skim milk until needed. The bacterial challenge was prepared by streaking the stock culture onto Esculin Blood Agar plates containing 5% whole blood. After 48 h incubation at 37 °C, a single colony was used to inoculate 100 ml of ultra high temperature ( U H T ) pasteurized milk and incubated for 12 h at 37°C. The 12 h culture was mixed well and a 100 #1 aliquot was removed to inoculate a second 100 ml volume of U H T milk. After a 6 h incubation at 37°C, the
192
L.M.SORDILLOANDL.A.BABIUK
culture was diluted serially in 10-fold increments using pyrogen-free saline. The C F U / m l value of each dilution was determined by plate pouring 100 #1 of the bacterial suspension onto McConkey agar plates. The dilution containing approximately 50 C F U of E. coli per ml of the pyrogen-free saline was selected for each trial. Inocula were plated post challenge to confirm viability counts.
Microbiological procedures Aseptic foremilk samples were collected at each sampling period to determine infection status of each quarter. Presumptive identification of bacterial isolates was determined by methods described by Brown et al. ( 1981 ). Gram negative isolates were identified to the species level using the API 20E System (Analytab Products, Plainview, NY). New intramammary infections following intracisternal challenge were identified by the isolation ofE. coli with the same API 20E profile as the challenge strain for three consecutive samplings.
Statistical analysis Data were analyzed by least squares analysis of variance using the general linear model procedure in the Statistical Analytical System (SAS, 1982) to determine effects of treatment and sample period on clinical scores. Preplanned comparisons of least square means were made by a pairwise t test. Means were contrasted between treatment groups within a sample period and between sample period within a treatment group. No other comparisons were made. Reductions in rate of new E. coli intramammary infections and persistance of E. coli mastitis in treated and control quarters were calculated using the normal approximation for comparing two binomially distributed percentages ( Steel and Torrie, 1982 ). The following formula was used: t = (PI - P 2 ) / [ ( P I Q I IN1 ) d- (P2Q2/N2) ]~ were t approximates a standard Student's t test, P = percent infected quarters, Q--percent uninfected quarters, and N = n u m b e r of quarters tested. Treatment groups are represented by the subscripts. RESULTS
Prior to the start of the experiment, periparturient cows were clinically normal. All quarters were free of intramammary infections and SCC values were < 500 000/ml. No changes in rectal temperature, clinical scores, or SCC were observed during the 12 h period following intramammary infusion of either rBoIFN-gamma or placebo (data not shown). Changes in mean SCC following experimental challenge with E. coli are shown in Fig. 1. Mean SCC for both treatment groups increased significantly
CONTROLLING MASTITISWITH INTERFERON GAMMA
193
10 8 -~-
m iz
o
C~
PLACEBO
--~
IFN
10 ~
w
0
10 ~
.05 ).
(P~< 0.01 ) beginning at 24 h and continuing until 60 h following the experimental challenge. No significant differences in SCC were observed between the rBolFN-gamma and placebo treatment groups up to 48 h following challenge. However, mean SCC values of placebo treated cows were elevated significantly over values for the rBolFN-gamma treated cows from 60 h to 336 h following experimental challenge. Mean SCC of rBolFN-gamma treated cows at the end of the trial were not significantly different from pre-challenge values. The effects of pre-treating animals with rBolFN-gamma on the rate of new E. coli intramammary infection is summarized in Table 1. Twenty one new E. coli infections or 91.3% of quarters were confirmed in the placebo treated
194
L.M. SORDILLO AND L.A. BABIUK
cows following experimental challenge. In the rBoIFN-gamma treated cows, 17 new E. coli infections or 70.8% o f quarters were confirmed. Compared with the placebo treatment group, cows intramammarily treated with rBoIFNgamma prior to experimental challenge had a 22.5% reduction in the number of new E. coli infections. Although the rate of new E. coli intramammary infections was lower in the rBoIFN-gamma treated cows, the reduction was not significantly different when compared to control animals. The effects o f pre-treating cows with rBoIFN-gamma on the persistance of E. coli mastitis during the experimental efficacy period is summarized in Table 2. All 21 new E. coli intramammary infections in the placebo treated cows persisted for 2 weeks following experimental challenge. In the rBoIFN-gamma treated cows, only 6 intramammary infections or 35% of the new infections persisted for 2 weeks following experimental challenge. Persistance of E. coli mastitis was reduced significantly (P~< 0.01 ) in rBoIFN-gamma treated cows when compared to placebo treated controls (64.7%). Effects of rBoIFN-gamma treatment on the severity of experimental E. coli TABLE2 Effects of rBolFN-gamma on persistance ofE. coli mastitis during the experimental challenge efficacy period. Treatment
No. new infections 2
No. infections after 14 days
Infections
groups 1 IFN Placebo
17 21
6 21
35.3 100
(%)
Reduction 3 (% ) 64.7
~Dairy cows were treated intramammarily with wither 105 U rBoIFN-gamma (IFN) or physiological saline (Placebo) at 24 h prior to experimental challenge. 2Number of quarters becoming infected after intracisternal challenge with E. coli. 3Significantly different from placebo treated cows (P~< 0.01 ). TABLE 3 Effects of r B o l F N - g a m m a on the severity of experimental E. coli mastitis. Treatment groups ~
No. eligible
Milkings
quarters 2
clinical 2
Average score 3
(%) I FN
24
16.7**
1.80**
Placebo
23
70.0
2.28
~Dairy cows were treated intramammarily with either 105 U rBoIFN-gamma (IFN) or physiological saline (Placebo) at 24 h prior to experimental challenge. 2Clinical score of >~ 2 recorded at each milking. 3Calculated from 28 consecutive milkings with clinical scores ranging from 1 to 5. **Significantly different from placebo-treated cows (P~< 0.01 ).
CONTROLLING
MASTITIS WITH INTERFERON
i
A
19 5
GAMMA
PLACEBO
IFN
-e
u.l O
O < O J
O
112,'
•
"~4~.T~. \ ~\~ ~I~ ~
*** 0
~.~
~
0
~ 4
~
~ 8
MILKINGS
t
12
_
~
_
~
I
16
FOLLOWING
I
Z~
20
i
2
I
24
I 28
CHALLENGE
Fig. 2. Effects of rBolFN-gamma on mean clinical scores throughout the 14 day experimental period. Each arrow indicates the death of a placebo-treated cow.
~
7
•
6
" O
5
~
4
>.> ~
3
e
1
O
O
2
42% Mortality
~//'~
IFN-GAMMA
~
PLACEBO
0
r
O~ ~ 2
--
2~4~'6
~8
6 " 0 ~ ~ "7
Hours
!L
1~ ' Days
j
Time After Challenge Fig. 3. Effects of rBolFN-gamma on the mortality rate associated with experimental E. coli mastitis.
mastitis is summarized in Table 3 and Figs. 2 and 3. Clinical scores for both treatment groups began to increase as early as 12 h following experimental challenge. Scores were elevated significantly (P~< 0.001 ) by 24 h (2.82 +_0.24) and reached peaked levels (3.67 ___0.24) by 36 h following E. coli challenge (Fig. 2 ). Cows treated with rBolFN-gamma exhibited clinical signs of mastitis at fewer ( P > 0.001 ) m i l k i n g s (16.7%) when compared to the placebo treatment group which exhibited clinical signs of mastitis at 70% of the milk-
196
L.M. SORDILLO AND L.A. BABIUK
ings (Table 3). In the rBoIFN-gamma treatment group, clinical signs were less pronounced (P~< 0.01 ) over the entire sampling period when compared to the placebo-treated cows (Table 3 ). Obvious differences in the number of cows surviving the experimental challenge were noted between the treatment groups (Figure 3). All cows infused intramammarily with rBoIFN-gamma survived the experimental E. coli challenge. However, cows treated with the saline placebo had a 42% mortality rate attributed to E. coli mastitis within 3 days of the experimental challenge. DISCUSSION
The acute symptoms most often associated with coliform mastitis are due to the rapid and unrestricted growth of the organism, the release ofendotoxin, and the subsequent development of an unlimited inflammatory reaction. Mammary gland neutrophils play a pivotal role in preventing or eliminating intramammary infections (Paape et al., 1981 ). Several studies have shown that the effectiveness of this defense system against E. coli mastitis depends on the promptness and magnitude of neutrophil migratory responses and their subsequent bactericidal capabilities at the foci of infection (Erskine et al., 1989; Hill, 1981; Lohuis et al., 1990). It is well documented that the antibacterial activities of m a m m a r y gland neutrophils are reduced in the presence of m a m m a r y gland secretions and comprised during physiological transitions of the m a m m a r y gland (Nagahata et al., 1988; Paape et al., 1981; Sordillo and Babiuk, 1990). The diminished antibacterial capacity of m a m m a r y gland neutrophils is attributable to several factors such as indiscriminate ingestion of milk constituents, degranulation, lack of energy source, and low concentrations of opsonins (Paape et al., 1981 ). Overcoming the immunosuppressive influence of m a m m a r y gland secretions on neutrophil function during the periparturient period may have a pronounced influence on the establishment and severity of E. coli mastitis. One approach to increase non-specific resistance to mastitis is to enhance neutrophil influx into the gland. Previous studies have shown that enhanced neutrophil migration to the m a m m a r y gland can control bacterial numbers during the early stages of infection and reduce the severity of coliform mastitis (Erskine et al., 1989). Improving the bactericidal efficiency of mammary gland neutrophils may also improve the cow's ability to limit bacterial growth. Recent studies have shown that in vitro treatment of m a m m a r y gland neutrophils with rBoIFN-gamma reversed the suppressive effects of mammary gland secretions and significantly increased the phagocytic and bactericidal activity when compared with untreated control cells (Sordillo and Babiuk, 1990 ). Although speculative at this point, the influence of local rBoIFNgamma administration on the establishment and severity of experimental E. coli mastitis in this study may reflect the effect of this cytokine on neutrophil
C O N T R O L L I N G MASTITIS WITH I N T E R F E R O N G A M M A
197
function in vivo. Prophylactic administration of rBoIFN-gamma did not significantly increase mean SCC over the placebo treated cows. However, cows treated with rBoIFN-gamma did have lower new infection rates and infections of shorter duration following experimental E. coli challenge when compared with controls. These results suggest that the bacterial capacity of neutrophils in rBoIFN-gamma treated quarters were more efficient at the time of experimental challenge. Enhancing an otherwise compromised defense mechanism during the initial stages of infection may control growth of the organism and minimize the presence of endotoxins in the gland, which would explain the lower infection rate and clinical scores in rBoIFN-gamma treated quarters. The clinical systemic signs associated with acute coliform mastitis result from the host's reaction to endotoxins released from killed bacteria. Pathophysiologic responses to endotoxins occur indirectly through the release and subsequent absorption of endogenous mediators from infected m a m m a r y glands and into the circulation (Lohuis et al., 1988 ). Cytokines (interleukin1, tumor necrosis factor, and IFN) are among these mediator molecules induced by LPS and which have an important role in the development of the acute inflammatory response (Billiau, 1987 ). Inflammatory changes occur in response to infectious pathogens. When cells or tissues are injured, the host attempts to protect and heal itself by initiating the inflammatory process. However, an exaggerated acute or protracted chronic inflammatory responses can cause extensive necrosis of m a m m a r y parenchymal tissues, loss of milk production, and mortality in dairy cows (Hill, 1981; Lohuis et al., 1990; Lohuis et al., 1988). Evidence indicates that IFN-gamma is a potent regulator of the acute phase response (Billiau, 1987 ). When administered prior to LPS challenge, IFN-gamma has been shown to down-regulate local inflammatory reactions (Billiau, 1987; Heremans et al., 1987). In this study, cows treated with rBoIFN-gamma prior to E. coli challenge had milder clinical reactions and lower mortality when compared with control cows. It is possible that rBoIFN-gamma inhibited the generation of endogenous inflammatory mediators in infected quarters which are responsible for systemic signs of coliform mastitis. The mechanisms by which rBoIFN-gamma modifies the outcome ofE. coli were subject to conjecture. However, intramammary infusion of rBoIFNgamma can prevent the rapid, unrestricted growth ofE. coli within the mammary gland and inhibit the subsequent development of an unlimited inflammatory response under experimental conditions. Controlling severe inflammation may decrease the pathological alterations to m a m m a r y parenchymal tissue and the tremendous milk production loss that often accompanies acute coliform mastitis during the periparturient period, thereby reducing the impact of this economically important form of mastitis. Prophylactic treatment of dairy cows with rBoIFN-gamma may potentially regulate the rate, severity,
198
L.M. SORDILLO AND L.A. BABIUK
and duration of naturally occurring coliform mastitis during periods of heightened susceptibility to infection. ACKNOWLEDGMENTS
The authors acknowledge with thanks the expert technical assistance of Lynn Gilbert. The assistance of the Animal Support Staff is appreciated also. This work was supported by CIBA-GEIGY Canada Limited, Mississauga, Ontario and the Natural Sciences and Engineering Research Council of Canada. Published with the approval of the Director of VIDO as journal series No. 105.
REFERENCES Billiau, A., 1987. Interferons and Inflammation. J. Interferon Res., 7:559. Brown, R.W., Barnum, D.A., Jasper, J.S., McDonald, J.S. and Schultze, W.D., 1981. Microbiological procedures for use in the diagnosis of bovine mastitis. 2nd Edn. Natl. Mastitis Counc., Inc., Arlington, VA, pp. 6-32. Eberhart, R.J., 1977. Coliform mastitis. J. Am. Vet. Med. Assoc., 170:1160. Erskine, R.J., Eberhart, R.J., Grasso, P.J. and Scholz, R.W., 1989. Induction ofEscherichia coli mastitis in cows fed selium-deficient or selium-supplemented diets. Am. J. Vet. Res., 50:20932100. Heremans, H., Dijkmans, R., Sobis, H., Vandekerckhove, F. and Billiau, A., 1987. Regulation by interferons of the local inflammatory response to bacterial lipopolysaccharide. J. Immunol., 138:4175. Hill, A.W., 1981. Factors influencing the outcome of Escherichia coli mastitis in the dairy cow. Res. Vet. Sci., 31:107. Lawman, M.J.P., Campos, M., Bielefeldt Ohmann, H., Griebel, P. and Babiuk, L.A., 1989. Recombinant cytokines and their potential therapeutic value in veterinary medicine. Comprehensive Biotechnology. Pergamon Press, London, pp. 63-106. Lohuis, J.A.C.M., Verheijden, J.H.M., Burvenich, C. and Van Miert, A.S.J.P.A.M., 1988. Pathophysiological effects ofendotoxins in ruminants. 1. Metabolic aspects. Vet. Q. 10:117. Lohuis, J.A.C.M., Schukken, Y.H., Verheijden, J.H.M., Brand, A. and Van Mien, A.S.J.P.A.M., 1990. Effect of severity of systemic signs during the acute phase of experimentally induced Escherichia coli mastitis on milk production losses. J. Dairy Sci. 73:333. Nagahata, H., Makino, S., Takeda, S., Takahashi, H. and Noda, H., 1988. Assessment of neutrophil function in the dairy cow during the perinatal period. J. Vet. Med. 35:747. Paape, M.J., Wergin, W.P., Guidry, A.J. and Schultze, W.D., 1981. Phagocytic defense of the ruminant mammary gland. Adv. Exp. Med. Biol., 137:555. SAS User's Guide: Statistics. 1982. SAS Inst. Inc., Cary, N.C. Smith, K.L., Todhunter, D.A. and Schoenberger, P.S., 1985. Environmental mastitis: cause, prevalence, prevention. J. Dairy Sci.,68:1531. Sordillo, L.M. and Babiuk, L.A., 1990. Modulation of bovine mammary neutrophil function during the periparturient period following in vitro exposure to recombinant bovine interferon gamma. Vet. Immunol. Immunopathol., 27: 393. Steel, R.G.D. and Torrie, J.H., 1982. Principles and procedures of statistics. McGraw-Hill Book Co., New York, N.Y.
