Despite decades of research and the use of multiple methods for controlling bacterial infections, including antiseptic teat dips, teat sealants, improvements in milking equipment and animal housing, vaccines against specific pathogens, and new antimicrobial preparations for use during and at the end of lactation, mastitis remains a major source of economic losses for dairy producers globally. Animal genetics and husbandry practices can substantially affect mammary gland immunity and susceptibility to intramammary infection.1
Beyond the anatomic barriers in the teats, protection of mammary glands from microbial invasion requires a complex and coordinated interaction between both the innate and acquired immune responses. Cattle have periods of increased susceptibility to clinical mastitis during the periparturient period beginning 1 to 2 weeks before parturition and continuing for up to 3 weeks after calving.2 Neutrophils play a key role in defense during intramammary infections and are the primary cell type in mammary glands at the onset of an infection.3 Impairment of immune responses during the periparturient period has been well characterized.4 Investigators in 1 study5 found that neutrophil functions are impaired in cows during the periparturient period. Results of that study5 indicate bovine neutrophils have decreases in oxidative burst and myeloperoxidase-H2O2-halide activity, both of which play important roles in neutrophil-mediated antimicrobial activity.
Granulocyte colony-stimulating factor is an endogenous hematopoietic growth factor that stimulates the production and differentiation of neutrophils by progenitor cells located in the bone marrow.6,7 Recombinant human G-CSF is used to increase circulating neutrophil numbers in patients undergoing myelosuppressive chemotherapy for cancer. Recombinant bG-CSF has activities similar to those of human G-CSF, and daily administration of bG-CSF to periparturient cows induces pronounced neutrophilia and increased phagocytic and cytotoxic activity of neutrophils.8,9
The pharmacodynamic response to native-sequence bG-CSF as determined by changes in absolute neutrophil counts indicates that daily administration of the protein would be required for therapeutic benefit. In modern dairy operations, husbandry practices used to manage periparturient cattle during the transition period are not consistent with administering daily injections for several weeks to restore protective neutrophil function throughout the periparturient period. Modifying native proteins by covalent binding of water-soluble polymers such as PEG extends the duration of activity by increasing the hemodynamic volume of the protein, reducing first-pass renal clearance, and reducing proteolytic degradation.10 A single dose of PEG–human G-CSF can provide efficacy equivalent to that after 11 daily doses of native human G-CSF.11
A variant of PEG–bG-CSF was found12 to be capable of increasing absolute neutrophil counts for approximately 7 to 10 days following a single SC injection. Administration of PEG–bG-CSF to periparturient dairy cattle approximately 7 days prior to the anticipated date of parturition and within 24 hours after calving enhances the myeloperoxidase-H2O2-halide activity of neutrophils, which is depressed during the periparturient period.12 On the basis of these results, the authors hypothesized that periparturient cattle receiving PEG–bG-CSF would have improved resistance to clinical mastitis. The objective of the study reported here was to evaluate the effects of various doses of PEG–bG-CSF on the incidence of naturally occurring clinical mastitis in periparturient dairy cattle.
Materials and Methods
ANIMALS
Pregnant Holstein cows and heifers (n = 211) from a commercial dairy herd located in Hanford, California, were enrolled in the study, which was conducted at that dairy. Approximately 25% of the cattle were heifers and 75% were multiparous (second and third lactation) cows. Each animal was enrolled as it approached the anticipated date of parturition (day of parturition = day 0). On day −14, pregnant cattle were moved to a designated pen, and a physical examination that included evaluation of all major body systems, respiratory rate, heart rate, and rectal temperature was conducted. The physical examination was repeated on day −7. Cattle that were clinically normal on days −14 and −7 were eligible for enrollment in the study.
Owners provided informed consent for participation of the cattle in the study. The study protocol was reviewed by the Elanco Animal Health Institutional Review Board to ensure compliance with applicable guidelines for conduct of research on animals. Edible products from cattle treated with PEG–bG-CSF were discarded for 7 days after the last dose of PEG–bG-CSF in compliance with a slaughter authorization issued by the US FDA.
PREPARATION OF PEG–BG-CSF
Bovine G-CSF was cloned and expressed in Escherichia coli by use of recombinant technology. Expressed protein was isolated from inclusion bodies and refolded. It was then chromatographically purified and covalently bound to a 20-kDa activated PEG at position T133, in accordance with the methods of Cho et al,13 to yield mono-PEG–bG-CSF. The protein was formulated at a concentration of 7.83 mg/mL in 10mM sodium acetate buffer (pH, 4.0), sorbitol (50 mg/mL), and polysorbate 20 (0.033 mg/mL). Sterile saline (0.9% NaCl) solution was used as a negative control treatment. The PEG–bG-CSF was stored frozen at −70°C until use. Saline solution was stored under ambient conditions until use.
ANIMAL HUSBANDRY
Cattle were housed as a group in a drylot pen with partial shade and dirt floors prior to parturition. After they had calved, cattle were moved to another drylot pen with partial shade and a dirt floor, which was kept wet to facilitate exposure of the mammary glands to mastitis-inducing pathogens. On days when there was no precipitation, water was added to the dirt with an oscillating sprinkler placed on the pen floor. The surface was maintained damp but without freestanding puddles of water.
Cows were milked twice daily in a herringbone parlor with automated milking equipment. Premilking and postmilking teat dipping were performed with an iodine disinfectant. Cattle for which parturition was imminent were fed a total mixed ration composed of alfalfa, corn, straw, distiller's grains, minerals, corn gluten, and corn silage prior to parturition. After calving, cattle were fed a total mixed ration composed of alfalfa, barley, corn, canola, wet distiller's grain, and corn silage. Both rations met National Research Council nutrient requirements for dairy cattle. Cattle had ad libitum access to water through automatic waterers located in the pens.
Cows were vaccinated against E coli with a J5 bacterin at the end of lactation and approximately 30 days prior to the subsequent parturition. They also were treated with an antimicrobiala (penicillin G procaine and dihydrostreptomycin) at the end of lactation.
STUDY DESIGN
The study was conducted as a completely randomized design. No blocking factors were used. Cattle were assigned to each of 4 treatment groups. Cattle in the 4 treatment groups received SC injections of PEG–bG-CSF at a dose of 5 μg/kg (n = 51 [38 cows and 13 heifers]), 10 μg/kg (54 [41 cows and 13 heifers]), or 20 μg/kg (53 [40 cows and 13 heifers]) or saline solution (control treatment) at a volume equal to that used for the 20 μg/kg dose of PEG –bG-CSF (53 [39 cows and 14 heifers]); treatments were administered on day −7 and within 24 hours after parturition (day 0). Separate randomization tables containing a list of treatment groups were used for heifers and multiparous cows to ensure equal proportions of heifers and multiparous cows were included in each group. Cattle were randomly assigned to groups on the basis of the sequence in the randomization tables on day −7. Animals that calved < 24 hours or > 10 days after receiving the first dose were removed from the study and replaced with another animal of the same parity assigned to the same treatment.
TREATMENT ADMINISTRATION
Cattle were weighed on day −7 to enable calculation of individual doses, which were administered on days −7 and 0. The PEG–bG-CSF was thawed immediately before administration, and the solution was mixed by gentle inversion of the vials. A syringe and 18-gauge, 1-inch needle were used to administer the appropriate dose of PEG–bG-CSF or saline solution by SC injection in the prescapular region of the neck of each animal.
ASSESSMENT OF ABSOLUTE NEUTROPHIL COUNT
In a previous study,12 treatment of cattle with PEG–bG-CSF induced significant increases in absolute neutrophil counts. Therefore, blood samples were collected by jugular venipuncture immediately before treatment administration on day −7 and again on day −6 (approx 24 hours after treatment administration). Additional blood samples were collected from all cattle on day 7.
Blood samples were analyzed with an automated cell counter,b and light microscopy was used for WBC differentiation. Automated cell counts and results for light microscopy were used to calculate the absolute neutrophil count of each sample. Both segmented neutrophils and immature neutrophils (eg, band cells, myelocytes, and metamyelocytes) were included in calculations of absolute neutrophil counts.
ASSESSMENT OF MASTITIS
The clinical status of each mammary gland was recorded at the morning and evening milking on days 1 to 28 by the milker, who was unaware of the treatment assignments for each animal. Clinical score was assigned by use of a 5-point scale as follows: 1 = visually or palpably normal mammary gland and visually normal milk, 2 = visually or palpably normal mammary gland but visually questionable milk, 3 = visually or palpably normal mammary gland but visually abnormal milk, 4 = visually or palpably abnormal mammary gland and visually abnormal milk, and 5 = visually or palpably abnormal mammary gland, visually abnormal milk, and evidence of systemic inflammation. When an animal had a clinical score ≥ 2, a CMT (scores of negative, trace, 1, 2, and 3) was performed on milk from the affected mammary gland. When the CMT result was greater than trace, duplicate milk samples from the affected mammary gland were obtained and the rectal temperature of the animal was determined. Cattle with a rectal temperature ≥ 40°C were considered to have systemic inflammation (clinical score, 5).
A clinical score ≥ 2 and CMT score greater than trace at any of the evaluations were considered indicative of clinical mastitis. Each new case of clinical mastitis was recorded as a separate event. A new case of clinical mastitis was defined as 2 mastitic episodes separated by an evaluation that yielded a clinical score of 1 (ie, visually or palpably normal mammary gland and visually normal milk).
BACTERIOLOGIC EVALUATION
Milk samples were aseptically collected from the affected mammary glands of cattle with signs of clinical mastitis and used for identification of microbial pathogens. Samples were collected at the time clinical mastitis criteria were met; samples were frozen at −70°C and stored at the study site. After all cattle enrolled in the study had reached 28 days of lactation, frozen milk samples were sent to a diagnostic laboratoryc for analysis. Each sample was cultured on blood agar and Mycoplasma direct plates. Biochemical tests were used to identify the genus and species of each isolate.
MILK PRODUCTION AND COMPOSITION
To assess the effect of treatment on milk yield, milk production was recorded at each milking for cattle with no morbidity events or for cattle up to the time of their clinical morbidity event (ie, healthy animals only) on days 1 to 28. Mean daily milk yield was determined for each group for the first 28 days of lactation.
To assess the effect of treatment on SCC and milk composition, composite milk samples were collected from all healthy mammary glands at the morning milking on days 3, 5, 7, and 10. Samples were sent to the Tulare County Dairy Herd Improvement Association Laboratory for analysis of SCC and concentrations of fat, lactose, protein, and milk solids.
DURATION OF GESTATION AND PERCENTAGE OF LIVE BIRTHS
To determine whether administration of PEG–bG-CSF had an effect on the duration of gestation, the number of days between the first dose (estimated day −7) and the day of parturition (day 0) was determined for each animal, and mean values were calculated. The total number of calves born alive for each treatment group was tabulated and compared to determine whether PEG–bG-CSF had detrimental effects on the viability of calves in utero.
STATISTICAL ANALYSIS
Statistical analysis was performed with the aid of commercial software.d For all analyses, values of P ≤ 0.05 were considered significant.
The primary efficacy criterion was the failure rate as measured by the incidence of clinical mastitis events. Clinical mastitis incidence rates were evaluated 2 ways. First, an animal was denoted as clinically affected if 1 or more mammary glands met the criteria for clinical mastitis (clinical score ≥ 2 and CMT score greater than trace) at any evaluation time during the first 28 days of lactation. A generalized mixed model for binomial data was used to compare the failure rates for the control treatment with the failure rates for each PEG–bG-CSF treatment. Parity (heifers vs multiparous cows) and treatment were fixed effects. If the parity-by-treatment interaction was not significant, treatments were compared across both levels of parity. If the parity-by-treatment interaction was significant, treatments were compared within each of the 2 levels of parity. A generalized mixed model for binomial data was used to model the relationship among treatment failure rates. If the parity-by-treatment interaction was not significant, dose response was compared across both levels of parity. If the parity-by-treatment interaction was significant, the dose response was evaluated for each of the 2 levels of parity. Second, the total number of clinical mastitis events per group was evaluated. Clinical mastitis events that occurred at any time during the study were determined (eg, if an animal had > 1 clinical mastitis event or clinical mastitis in > 1 mammary gland). The resulting data were expressed as count data and were analyzed by the method of analysis described for the clinical mastitis incidence data.
Duration of abnormal clinical status scores was determined for each morbidity event for each animal. Depending on the distribution of the data, an ANOVA or an appropriate nonparametric method was used for analysis and to compare treatment means (or medians) with the control mean (or median). A 2-way analysis was performed to evaluate the parity-by-treatment interaction. If the parity-by-treatment interaction was not significant, treatments were compared across both levels of parity. If the parity-by-treatment interaction was significant, the treatments were compared within each of the 2 levels of parity.
Secondary efficacy variables (milk yield, milk composition, and absolute neutrophil count) that were continuous in nature were analyzed by use of an ANOVA or an appropriate nonparametric method for each time point. The treatment-by-parity interaction was compared within each time point. Secondary efficacy variables that were qualitative (eg, clinical scores) were analyzed by use of the Cochran–Mantel-Haenszel row mean test at each time point, whereby results for the control group were compared with results for each treatment group, both across and within both levels of parity.
Mean SCC and concentrations of fat, protein, lactose, and milk solids were determined on days 3, 5, 7, and 10 at the morning milking for each group. Repeated-measures ANOVA was used, whereby parity, treatment, and day were fixed effects and animal nested within treatment by parity was considered a random effect. The possible covariance structures evaluated were compound symmetry, compound symmetry heterogeneous variances, autoregressive, autoregressive heterogeneous variances, and unstructured. The minimum Akaike information criterion was used to select the covariance structure. The least squares mean was used to compare the effects of PEG–bG-CSF with those of the control treatment on the basis of unadjusted P values. When compound symmetry, compound symmetry heterogeneous variances, and unstructured were used as covariance structures, the random effect of cattle nested within treatment by parity was eliminated from the model specification. Repeated-measures ANOVA identical to that described for milk composition was used for analysis of total daily milk yield.
Repeated-measures ANOVA was used for analysis of absolute neutrophil counts, with particular interest in the means for treatment-by-day interactions. These means provided information on the effect of exposure to PEG–bG-CSF.
Results
ABSOLUTE NEUTROPHIL COUNTS
Saline solution–treated cattle had relatively constant absolute neutrophil counts across the time points (Figure 1). In contrast, cattle treated with PEG–bG-CSF had marked increases in absolute neutrophil counts within 24 hours after administration. Absolute neutrophil counts remained elevated, relative to those in control cattle, 7 days after the second dose of PEG-bG-CSF at 10 and 20 μg/kg but not at 5 μg/kg. Results of statistical analysis of group comparisons were summarized (Table 1).
Mean ± SEM absolute neutrophil count for 211 periparturient Holstein cows and heifers injected SC with PEG–bG-CSF at a dose of 5 μg/kg (stippled bars; n = 51 [38 cows and 13 heifers]), 10 μg/kg (diagonal-striped bars; 54 [41 cows and 13 heifers]), or 20 μg/kg (white bars; 53 [40 cows and 13 heifers]) or saline (0.9% NaCl) solution (control treatment) at a volume equal to that used for the 20 μg/kg dose of PEG–bG-CSF (black bars; 53 [39 cows and 14 heifers]). Cattle received 2 injections (the first on day −7 and the second within 24 hours after parturition). Day of parturition was designated as day 0.
Citation: American Journal of Veterinary Research 76, 3; 10.2460/ajvr.76.3.231
Mean ± SEM absolute neutrophil count for 211 periparturient Holstein cows and heifers injected SC with PEG–bG-CSF at a dose of 5 μg/kg (stippled bars; n = 51 [38 cows and 13 heifers]), 10 μg/kg (diagonal-striped bars; 54 [41 cows and 13 heifers]), or 20 μg/kg (white bars; 53 [40 cows and 13 heifers]) or saline (0.9% NaCl) solution (control treatment) at a volume equal to that used for the 20 μg/kg dose of PEG–bG-CSF (black bars; 53 [39 cows and 14 heifers]). Cattle received 2 injections (the first on day −7 and the second within 24 hours after parturition). Day of parturition was designated as day 0.
Citation: American Journal of Veterinary Research 76, 3; 10.2460/ajvr.76.3.231
Mean ± SEM absolute neutrophil count for 211 periparturient Holstein cows and heifers injected SC with PEG–bG-CSF at a dose of 5 μg/kg (stippled bars; n = 51 [38 cows and 13 heifers]), 10 μg/kg (diagonal-striped bars; 54 [41 cows and 13 heifers]), or 20 μg/kg (white bars; 53 [40 cows and 13 heifers]) or saline (0.9% NaCl) solution (control treatment) at a volume equal to that used for the 20 μg/kg dose of PEG–bG-CSF (black bars; 53 [39 cows and 14 heifers]). Cattle received 2 injections (the first on day −7 and the second within 24 hours after parturition). Day of parturition was designated as day 0.
Citation: American Journal of Veterinary Research 76, 3; 10.2460/ajvr.76.3.231
Results of statistical analyses (P value) of absolute neutrophil count in 211 periparturient Holstein cows and heifers injected SC with PEG–bG-CSF at a dose of 5 μg/kg (n = 51 [38 cows and 13 heifers]), 10 μg/kg (54 [41 cows and 13 heifers]), or 20 μg/kg (53 [40 cows and 13 heifers]) or saline (0.9% NaCl) solution (control treatment) at a volume equal to that used for the 20 μg/kg dose of PEG–bG-CSF (53 [39 cows and 14 heifers]).
| Day of study | |||
|---|---|---|---|
| Comparison | –7 | –6 | 7 |
| Saline solution vs 5 μg/kg | 0.685 | < 0.001 | 0.149 |
| Saline solution vs 10 μg/kg | 0.737 | < 0.001 | < 0.001 |
| Saline solution vs 20 μg/kg | 0.887 | < 0.001 | < 0.001 |
| 5 μg/kg vs 10 μg/kg | 0.944 | 0.665 | 0.017 |
| 5 μg/kg vs 20 μg/kg | 0.794 | 0.042 | < 0.001 |
| 10 μg/kg vs 20 μg/kg | 0.848 | 0.108 | < 0.001 |
Values were considered significant at P ≤ 0.05.
Cattle received 2 injections (the first on day −7 and the second within 24 hours after parturition). Day of parturition was designated as day 0.
CLINICAL MASTITIS INCIDENCE
The proportion of cattle in each group with signs of clinical mastitis in at least 1 mammary gland was summarized (Table 2). Approximately one-third of the cattle receiving saline solution developed clinical mastitis during the first 28 days of lactation. Cattle treated with PEG-bG-CSF at 10 or 20 μg/kg had significant reductions in the incidence of clinical mastitis, relative to the incidence for the control group. Cattle treated with PEG-bG-CSF at 5 μg/kg had a nonsignificant reduction in the incidence of clinical mastitis. There were no significant differences in clinical mastitis rates among the PEG–bG-CSF treatment groups. Similarly, the treatment-by-parity interaction was not significant, which suggested that multiparous cows and heifers had similar responses to treatment.
Incidence of clinical mastitis in cattle treated with saline solution or various doses of PEG–bG-CSF.
| Group | No. (%) of cattle with clinical mastitis |
|---|---|
| Saline solution (n = 53) | 18 (34) |
| 5 μg/kg (n = 51) | 10 (19.6) |
| 10 μg/kg (n = 54) | 9 (16.7)* |
| 20 μg/kg (n = 53) | 5 (9.4)† |
Value differs significantly (P = 0.044) from the value for saline solution.
Value differs significantly (P = 0.004) from the value for saline solution.
The number of mammary glands with clinical mastitis in the various groups was summarized (Table 3). Cattle in all 3 PEG-bG-CSF treatment groups had significant decreases in the number of affected mammary glands and a nonsignificant reduction in the proportion of cattle with multiple infected mammary glands, relative to results for the control group. There were no significant differences in clinical mastitis rates among the PEG-bG-CSF treatment groups. Similarly, the treatment-by-parity interaction was not significant, which suggested that multiparous cows and heifers had similar responses to treatment.
Incidence of clinical mastitis in mammary glands of cattle treated with saline solution or various doses of PEG–bG-CSF.
| Group | No. of mammary glands with clinical mastitis | No. of cattle with multiple clinical mastitis events* |
|---|---|---|
| Saline solution | 22 | 4 |
| 5 μg/kg | 10† | 0 |
| 10 μg/kg | 10‡ | 1 |
| 20 μg/kg | 7§ | 2 |
Cattle had > 1 mammary gland with clinical mastitis or cattle had > 1 episode of clinical mastitis.
Value differs significantly (P = 0.035) from the value for saline solution.
Value differs significantly (P = 0.020) from the value for saline solution.
Value differs significantly (P = 0.003) from the value for saline solution.
Relative severity of each case of clinical mastitis at the time of diagnosis was assessed on the basis of clinical score (Table 4). The distribution of clinical scores was similar among groups, and most affected mammary glands had a clinical score of 3 at the time of clinical mastitis diagnosis. The effect of treatments on the duration of abnormal milk or mammary glands for each clinical mastitis case was summarized. Mean duration of clinical mastitis episodes did not differ significantly among groups.
Severity of clinical mastitis at the time of diagnosis and duration of clinical mastitis for cattle treated with saline solution or various doses of PEG–bG-CSF.
| Clinical score at diagnosis | |||||
|---|---|---|---|---|---|
| Group | 2 | 3 | 4 | 5 | Mean ± SEM duration (d) |
| Saline solution | 1 | 21 | 0 | 0 | 10.8 ± 1.4 |
| 5 μg/kg | 1 | 8 | 0 | 1 | 5.3 ± 2.1 |
| 10 μg/kg | 0 | 8 | 0 | 2 | 8.4 ± 2.1 |
| 20 μg/kg | 0 | 7 | 0 | 0 | 6.9 ± 2.5 |
Clinical score was assigned by use of a 5-point scale as follows: 1 = visually or palpably normal mammary gland and visually normal milk, 2 = visually or palpably normal mammary gland but visually questionable milk, 3 = visually or palpably normal mammary gland but visually abnormal milk, 4 = visually or palpably abnormal mammary gland and visually abnormal milk, and 5 = visually or palpably abnormal mammary gland, visually abnormal milk, and evidence of systemic inflammation.
MICROBIOLOGICAL ANALYSIS
Fifty-one milk samples collected from individual mammary glands with clinical signs of mastitis were evaluated to identify bacteria associated with each infection (Table 5). Results of the microbiological analyses indicated that the clinical mastitis cases in this study population were associated with both gram-positive and gram-negative bacteria. The most frequent isolates were common mastitis pathogens, including Streptococcus uberis, Streptococcus dysgalactiae, coagulase-negative Staphylococcus spp, and E coli. There were also a variety of isolates, including Aerococcus spp, Enterococcus spp, and Lactococcus spp, that are not usually associated with intramammary infections.
Frequency distribution of bacterial isolates cultured from milk samples collected from individual mammary glands with clinical signs of mastitis for cattle treated with saline solution or various doses of PEG–bG-CSF.
| Bacterial isolate | Saline solution | 5 μg/kg | 10 μg/kg | 20 μg/kg |
|---|---|---|---|---|
| Streptococcus uberis | 5 | 2 | 5 | 5 |
| Streptococcus dysgalactiae | 1 | 0 | 1 | 1 |
| Streptococcus acidominimus | 0 | 0 | 1 | 0 |
| Aerococcus urinae | 1 | 0 | 0 | 0 |
| Aerococcus viridans | 0 | 1 | 1 | 0 |
| Granulicatella adiacens | 0 | 0 | 0 | 1 |
| Enterococcus durans | 1 | 1 | 0 | 0 |
| Lactococcus lactis | 0 | 0 | 1 | 0 |
| Escherichia coli | 2 | 0 | 1 | 0 |
| Staphylococcus xylosus | 2 | 1 | 0 | 0 |
| Staphylococcus hyicus | 2 | 0 | 1 | 0 |
| Staphylococcus chromogenes | 0 | 0 | 1 | 2 |
| Staphylococcus lentis | 0 | 1 | 0 | 0 |
| No growth | 8 | 4 | 2 | 1 |
MILK PRODUCTION AND COMPOSITION
Mean daily milk production for healthy cattle in each group was summarized (Figure 2). All groups had steadily increasing milk production over the first 28 days of lactation. There were no significant differences in mean daily milk production among groups in the study.
Effect of administration of PEG–bG-CSF at a dose of 5 μg/kg (circles; n range = 41 to 51), 10 μg/kg (triangles; n range = 45 to 54), or 20 μg/kg (squares; n range = 48 to 53) or saline solution (diamonds; n range = 35 to 53) on mean ± SEM daily milk production by healthy cattle during the first 28 days of lactation. Cattle received 2 injections (day −7 and within 24 hours after parturition). Day of parturition was designated as day 0.
Citation: American Journal of Veterinary Research 76, 3; 10.2460/ajvr.76.3.231
Effect of administration of PEG–bG-CSF at a dose of 5 μg/kg (circles; n range = 41 to 51), 10 μg/kg (triangles; n range = 45 to 54), or 20 μg/kg (squares; n range = 48 to 53) or saline solution (diamonds; n range = 35 to 53) on mean ± SEM daily milk production by healthy cattle during the first 28 days of lactation. Cattle received 2 injections (day −7 and within 24 hours after parturition). Day of parturition was designated as day 0.
Citation: American Journal of Veterinary Research 76, 3; 10.2460/ajvr.76.3.231
Effect of administration of PEG–bG-CSF at a dose of 5 μg/kg (circles; n range = 41 to 51), 10 μg/kg (triangles; n range = 45 to 54), or 20 μg/kg (squares; n range = 48 to 53) or saline solution (diamonds; n range = 35 to 53) on mean ± SEM daily milk production by healthy cattle during the first 28 days of lactation. Cattle received 2 injections (day −7 and within 24 hours after parturition). Day of parturition was designated as day 0.
Citation: American Journal of Veterinary Research 76, 3; 10.2460/ajvr.76.3.231
The effects of PEG–bG-CSF administration on SCC and milk composition (concentrations of fat, lactose, protein, and milk solids) in healthy animals were summarized (Table 6). Administration of PEG–bG-CSF at the 3 doses evaluated had no significant effect on SCC or milk composition.
Effects of PEG-bG-CSF on SCC and milk composition for healthy cattle treated with saline solution or various doses of PEG-bovine G-CSF.
| Variable | Day | Saline solution | 5 μg/kg | 10 μg/kg | 20 μg/kg |
|---|---|---|---|---|---|
| SCC (X 103 cells/mL) | 3 | 989 ± 313 | 668 ± 144 | 858 ± 183 | 1,212 ± 288 |
| 5 | 987 ± 337 | 524 ± 165 | 394 ± 115 | 364 ± 108 | |
| 7 | 415 ± 178 | 317 ± 105 | 160 ± 44 | 481 ± 155 | |
| 10 | 415 ± 183 | 164 ± 38 | 115 ± 30 | 172 ± 56 | |
| Fat (%) | 3 | 1.97 ± 0.18 | 2.15 ± 0.18 | 2.15 ± 0.28 | 2.73 ± 0.31 |
| 5 | 1.95 ± 0.20 | 2.16 ± 0.29 | 2.16 ± 0.20 | 2.23 ± 0.32 | |
| 7 | 1.32 ± 0.11 | 1.64 ± 0.32 | 1.58 ± 0.12 | 1.67 ± 0.22 | |
| 10 | 1.06 ± 0.09 | 1.56 ± 0.24 | 1.36 ± 0.18 | 1.18 ± 0.09 | |
| Lactose (%) | 3 | 4.17 ± 0.13 | 4.41 ± 0.06 | 4.39 ± 0.07 | 4.30 ± 0.08 |
| 5 | 4.43 ± 0.09 | 4.59 ± 0.06 | 4.56 ± 0.07 | 4.54 ± 0.07 | |
| 7 | 4.65 ± 0.06 | 4.73 ± 0.06 | 4.76 ± 0.04 | 4.72 ± 0.05 | |
| 10 | 4.79 ± 0.06 | 4.82 ± 0.05 | 4.90 ± 0.04 | 4.87 ± 0.04 | |
| Protein (%) | 3 | 4.51 ± 0.08 | 4.55 ± 0.06 | 4.61 ± 0.16 | 4.59 ± 0.08 |
| 5 | 4.13 ± 0.05 | 4.02 ± 0.06 | 4.13 ± 0.07 | 4.12 ± 0.05 | |
| 7 | 3.78 ± 0.05 | 3.76 ± 0.05 | 3.81 ± 0.05 | 3.81 ± 0.05 | |
| 10 | 3.50 ± 0.04 | 3.58 ± 0.05 | 3.52 ±0.04 | 3.50 ± 0.04 | |
| Milk solids (%) | 3 | 9.50 ± 0.14 | 9.81 ± 0.08 | 9.82 ± 0.15 | 9.71 ± 0.11 |
| 5 | 9.41 ± 0.11 | 9.48 ± 0.10 | 9.55 ± 0.10 | 9.51 ± 0.09 | |
| 7 | 9.30 ± 0.09 | 9.36 ± 0.10 | 9.44 ± 0.07 | 9.41 ± 0.09 | |
| 10 | 9.16 ± 0.09 | 9.28 ± 0.09 | 9.31 ± 0.07 | 9.27 ± 0.07 |
Values reported are mean ± SEM. Day of parturition was designated as day 0.
DURATION OF GESTATION AND PERCENTAGE OF LIVE BIRTHS
Effects of PEG–bG-CSF administration on the duration of gestation and percentage of live births were summarized (Table 7). Administration of PEG–bG-CSF had no significant effect on the duration of gestation after the initial dose (P = 0.47) or percentage of live births (P = 0.14).
Effect of PEG-bG-CSF on mean ± SEM duration of gestation and percentage of live births for cattle treated with saline solution or various doses of PEG-bG-CSF.
| Group | Duration of gestation from first dose to parturition (d) | Live births* |
|---|---|---|
| Saline solution | 6.5 ± 0.4 | 54/56 (96.4) |
| 5 μg/kg | 6.8 ± 0.4 | 52/53 (98.1) |
| 10 μg/kg | 7.0 ± 0.4 | 52/54 (96.3) |
| 20 μg/kg | 6.0 ± 0.4 | 50/56 (89.3) |
Cattle received 2 injections, the first of which was administered on day −7 and the second within 24 hours after parturition; day of parturition was designated as day 0.
Value reported is number of live births/number of treated cattle (percentage).
Discussion
Cattle treated with PEG–bG-CSF at 10 and 20 μg/kg 7 days prior to the anticipated date of parturition and again within 24 hours after calving had sustained increases in absolute neutrophil counts. These results are consistent with the biological function of G-CSF as an inducer of proliferation and differentiation of bone marrow myeloid precursor cells.14 Previous studies15,16 conducted to evaluate the effects of human G-CSF or bG-CSF indicate significant increases in absolute neutrophil counts could be achieved with daily administration of these cytokines at doses ranging from 1 to 5 μg/kg. Polyethylene glycol conjugation of proteins increases the duration of activity primarily by reducing first-pass renal clearance.17 In the present study, absolute neutrophil counts in the peripheral circulation were increased for at least 14 days after 2 injections of PEG–bG-CSF at 10 and 20 μg/kg administered approximately 1 week apart. Suppression of neutrophil function in periparturient cattle persists for several weeks.5 Polyethylene glycol–conjugated bG-CSF provides an extended duration of activity, which should improve administration convenience.
The incidence of clinical mastitis during the periparturient period was assessed by observing the condition of the mammary glands and milk at each milking during the first 28 days of lactation. This procedure is consistent with National Mastitis Council guidelines for diagnosis of clinical mastitis and consistent with practices used at commercial dairies. Approximately one-third of the cattle treated with sterile saline solution had at least 1 case of clinical mastitis during the first 28 days of lactation. Approximately one-third of the milk samples from these clinical mastitis cases did not contain recoverable microbial isolates. These results are consistent with those of a previous study18 that indicate a significant proportion of milk samples collected from clinical cases of mastitis yield negative culture results. Most milk samples from clinical mastitis events containing bacterial isolates included gram-positive and gram-negative bacteria consistent with pathogens typically present for the commercial farm where the present study was conducted.e
Administration of PEG–bG-CSF at 10 and 20 μg/kg resulted in a lower incidence of clinical mastitis on a peranimal basis (9/54 and 5/53, respectively), relative to that of animals treated with sterile saline solution (18/53). Similarly, PEG–bG-CSF significantly reduced the number of mammary glands with clinical mastitis (10 and 7 for 10 and 20 μg/kg, respectively), relative to that for animals treated with sterile saline solution (22). These results are consistent with those of a previous study19 that indicate administration of recombinant human G-CSF prevents the occurrence of Staphylococcus aureus–induced mastitis by use of an intramammary challenge-exposure technique.
Administration of PEG–bG-CSF did not significantly alter the severity of clinical score at the time each clinical mastitis event was diagnosed. In addition, the duration of each clinical mastitis event was not significantly different among treatments. It is reasonable to hypothesize that neutrophils primed for improved microbicidal activity by exposure to PEG–bG-CSF in the circulation may be better able to eliminate infections before an inflammatory response is apparent. However, once the immune response is overwhelmed and an acute inflammatory response is apparent, the severity of the inflammatory response and amount of time required for resolution remain unchanged. In the study reported here, cattle were exposed to environmental conditions that are known to result in mastitis. To confirm whether neutrophils from PEG–bG-CSF are more effective at killing invading bacteria, milk samples for microbiological testing would need to be collected immediately before and after signs of clinical mastitis are apparent.
Generating and maintaining immune competence require energy. It is possible that administration of PEG–bG-CSF to alter neutrophil numbers and function during the periparturient period could divert energy from milk production in healthy cows. Administration of PEG–bG-CSF had no significant effect on daily milk production in cows with healthy mammary glands.
Results of milk SCC assays indicated that cattle with healthy mammary glands were similar among groups. These results indicated that the increased numbers of circulating neutrophils in cattle treated with PEG–bG-CSF were not associated with elevated numbers of somatic cells in healthy mammary glands. In vitro exposure of bovine neutrophils to bG-CSF results in increased expression of adhesion molecules CD18 and CD11b, which play a role in extravasation of the cells.8 In the present study, milk SCC was not determined at the time clinical mastitis was diagnosed. Therefore, it is not possible to determine whether cattle treated with PEG–bG-CSF had a more robust influx of neutrophils into an affected mammary gland early in the disease process.
Administration of PEG–bG-CSF did not significantly alter concentrations of fat, protein, lactose, or milk solids in healthy mammary glands. This suggested that administration of PEG–bG-CSF to cattle will not negatively impact processing of milk-derived products such as cheese and yogurt.
The impairment of neutrophil function in pregnant cattle during the periparturient period could be an adaptation to practices associated with the husbandry of cows for milk production. Therefore, it is possible that altering neutrophil function during the periparturient period could have negative consequences for the ability of dams to maintain pregnancy late in gestation or could impair the survival of calves in utero. Administration of the first dose of PEG–bG-CSF before parturition had no significant effect on the duration of gestation or percentage of live births. These data suggested that manipulation of neutrophil numbers and function during late gestation had no negative effects on the dam's ability to maintain pregnancy or on the viability of calves in utero.
Analysis of the results of the present study indicated that administration of PEG–bG-CSF reduced the incidence of naturally occurring clinical mastitis in periparturient dairy cows and heifers. The mechanism of action by which PEG–bG-CSF was able to resolve impaired neutrophil function during a time when cattle are more susceptible to disease represents a novel approach to controlling mastitis and is worthy of further investigation.
Acknowledgments
Supported by Elanco Animal Health, Greenfield, IN.
This manuscript represents a portion of the documentation submitted to the US FDA Center for Veterinary Medicine by Elanco Animal Health to fulfill requirements for registration of a new animal drug.
The authors thank Dr. George Milliken and April Milliken-Mackinnon for assistance with the statistical analysis.
ABBREVIATIONS
| bG-CSF | Bovine granulocyte colony-stimulating factor |
| CMT | California mastitis test |
| G-CSF | Granulocyte-stimulating factor |
| PEG | Polyethylene glycol |
| SCC | Somatic cell count |
Footnotes
Quartermaster DC, Zoetis, Florham Park, NJ.
ADVIA, Siemens, Munich, Germany.
Dairy Food Safety Lab, Veterinary Medicine Teaching and Research Center, University of California-Tulare, Tulare, Calif.
SAS, version 9.2, SAS Institute Inc, Cary, NC.
TeVelde B, Lone Oak Farms, Hanford, Calif: Personal communication, 2014.
References
1. Sordillo LM. Factors affecting mammary gland immunity and mastitis susceptibility. Livest Prod Sci 2005; 98: 89–99.
2. Oliver SP, Sordillo LM. Udder health in the periparturient period. J Dairy Sci 1988; 71: 2584–2606.
3. Paape MJ, Bannerman DD, Zhao X, et al. The bovine neutrophil: structure and function in blood and milk. Vet Res 2003; 34: 597–627.
4. Sordillo LM, Shafer-Weaver K, DeRosa D. Immunobiology of the mammary gland. J Dairy Sci 1997; 80: 1851–1865.
5. Kehrli ME, Nonnecke BJ, Roth JA. Alterations in bovine neutrophil function during the periparturient period. Am J Vet Res 1989; 50: 207–214.
6. Nagata S. Gene structure and function of granulocyte colony stimulating factor. Bioessays 1989; 10: 113–117.
7. Nomura H, Imazeki I, Oheda M, et al. Purification and characterization of human granulocyte colony stimulating factor (G-CSF). EMBO J 1986; 5: 871–876.
8. Heidari M, Harp JA, Kehrli ME. Expression, purification and in vitro biological activities of recombinant bovine granulocyte-colony stimulating factor. Vet Immunol Immunopathol 2001; 81: 45–57.
9. Kehrli ME, Goff JP, Stevens MG, et al. Effects of granulocyte colony-stimulating factor administration to periparturient cows on neutrophils and bacterial shedding. J Dairy Sci 1991; 74: 2448–2458.
10. Molineux G. PEGylation: engineering improved biopharmaceuticals for oncology. Pharmacotherapy 2003; 23: 3S–8S.
11. Molineux G. The design and development of pegfilgrastim (PEG-rmetHuG-CSF, Neulasta). Curr Pharm Des 2004; 10: 1235–1244.
12. Kimura K, Goff JP, Canning P, et al. Effect of recombinant bovine granulocyte colony-stimulating factor covalently bound to polyethylene glycol injection on neutrophil number and function in periparturient dairy cows. J Dairy Sci 2014; 97: 4842–4851.
13. Cho H, Daniel T, Buechler YJ, et al. Optimized clinical performance of growth hormone with an expanded genetic code. Proc Natl Acad Sci U S A 2011; 108: 9060–9065.
14. Basu S, Dunn A & Ward A. G-CSF. Function and modes of action. Int J Mol Med 2002; 10: 3–10.
15. Cullor JS, Fairley N, Smith WL, et al. Hemogram changes in lactating dairy cows given recombinant granulocyte colony stimulating factor (r-MethuG-CSF). Vet Pathol 1990; 27: 311–316.
16. Cullor JS, Smith W, Zinki JG, et al. Hematologic and bone marrow changes after short- and long-term administration of two recombinant bovine granulocyte colony-stimulating factors. Vet Pathol 1992; 29: 521–527.
17. Harris JM, Martin NA, Modi M. PEGylation: a novel process for modifying pharmacokinetics. Clin Pharmacokinet 2001; 40: 539–551.
18. Hogan JS, Smith JL, Hoblet KH, et al. Field survey of clinical mastitis in low somatic cell count herds. J Dairy Sci 1989; 72: 1547–1556.
19. Nickerson SC, Owens WE, Watts JL. Effects of recombinant granulocyte colony-stimulating factor on Staphylococcus aureus mastitis in lactating dairy cows. J Dairy Sci 1989; 72: 3286–3294.
