Effect of endotoxic mastitis on epithelial cell numbers in the milk of dairy cows

Sarah A. Wagner Department of Veterinary Diagnostic and Production Animal Medicine, College of Veterinary Medicine, Iowa State University, Ames, IA 50011.

Search for other papers by Sarah A. Wagner in
Current site
Google Scholar
PubMed
Close
 DVM, PhD
,
Douglas E. Jones Department of Veterinary Pathology, College of Veterinary Medicine, Iowa State University, Ames, IA 50011.

Search for other papers by Douglas E. Jones in
Current site
Google Scholar
PubMed
Close
 VMD, PhD
, and
Michael D. Apley Department of Veterinary Diagnostic and Production Animal Medicine, College of Veterinary Medicine, Iowa State University, Ames, IA 50011.

Search for other papers by Michael D. Apley in
Current site
Google Scholar
PubMed
Close
 DVM, PhD

Click on author name to view affiliation information

Abstract

Objective—To measure epithelial cell percentages and somatic cell counts (SCCs) in milk and determine whether isoflupredone acetate reduces mammary gland epithelial cell sloughing in cows with acute endotoxin-induced mastitis.

Animals—13 lactating Holstein cows.

Procedures—Determination of SCC and flow cytometric analysis of cytokeratin-positive (epithelial) cells in milk were performed before and 12 hours after induction of mastitis via intramammary administration of bacterial endotoxin in 8 cows and at the same time points in 5 cows without mastitis. Endotoxin-treated cows received isoflupredone acetate (20 mg) or saline (0.9% NaCl) solution (n = 4/group) IV after signs of mastitis developed.

Results—At the 12-hour time point, mean ± SD percentage of epithelial cells in milk increased from 2.74 ± 1.93% to 42.11 ± 36.21% and decreased from 5.73 ± 4.52% to 5.31 ± 1.93% in milk from cows with and without mastitis, respectively. Median (range) SCC in milk increased from 195,000 cells/mL (17,000 to 442,000 cells/mL) to 5,437,500 cells/mL (69,000 to 11,036,000 cells/mL) and from 19,000 cells/mL (9,000 to 125,000 cells/mL) to 51,000 cells/mL (10,000 to 835,000 cells/mL) in cows with and without mastitis, respectively. Changes in these variables were significantly greater in mastitis-affected cows. Administration of isoflupredone acetate did not affect epithelial cell percentage or SCC in milk.

Conclusions and Clinical Relevance—During the early phase of endotoxin-induced mastitis in dairy cows, large numbers of epithelial cells were sloughed into the milk. Epithelial cell damage likely precedes an influx of immune cells into affected mammary glands and may contribute to breakdown of the blood-milk barrier.

Abstract

Objective—To measure epithelial cell percentages and somatic cell counts (SCCs) in milk and determine whether isoflupredone acetate reduces mammary gland epithelial cell sloughing in cows with acute endotoxin-induced mastitis.

Animals—13 lactating Holstein cows.

Procedures—Determination of SCC and flow cytometric analysis of cytokeratin-positive (epithelial) cells in milk were performed before and 12 hours after induction of mastitis via intramammary administration of bacterial endotoxin in 8 cows and at the same time points in 5 cows without mastitis. Endotoxin-treated cows received isoflupredone acetate (20 mg) or saline (0.9% NaCl) solution (n = 4/group) IV after signs of mastitis developed.

Results—At the 12-hour time point, mean ± SD percentage of epithelial cells in milk increased from 2.74 ± 1.93% to 42.11 ± 36.21% and decreased from 5.73 ± 4.52% to 5.31 ± 1.93% in milk from cows with and without mastitis, respectively. Median (range) SCC in milk increased from 195,000 cells/mL (17,000 to 442,000 cells/mL) to 5,437,500 cells/mL (69,000 to 11,036,000 cells/mL) and from 19,000 cells/mL (9,000 to 125,000 cells/mL) to 51,000 cells/mL (10,000 to 835,000 cells/mL) in cows with and without mastitis, respectively. Changes in these variables were significantly greater in mastitis-affected cows. Administration of isoflupredone acetate did not affect epithelial cell percentage or SCC in milk.

Conclusions and Clinical Relevance—During the early phase of endotoxin-induced mastitis in dairy cows, large numbers of epithelial cells were sloughed into the milk. Epithelial cell damage likely precedes an influx of immune cells into affected mammary glands and may contribute to breakdown of the blood-milk barrier.

Epithelial cells are a component of milk from cows without mastitis; estimates of the prevalence of epithelial cells as a percentage of total somatic cells in milk vary from 0% to 45%.1,2,a When a cow develops endotoxic mastitis attributable to infection with Escherichia coli, epithelial cells in the affected mammary gland are damaged and may slough from the basement membrane.3 Administration of the glucocorticoid drug dexamethasone prior to the onset of clinical signs may preserve milk production in cows with mastitis.4 This effect might be explained in part by a reduction in cellular damage in the affected mammary glands. The purpose of the study reported here was to measure epithelial cell numbers in milk during acute endotoxic mastitis in cows and to determine whether the glucocorticoid drug isoflupredone acetate reduces epithelial cell sloughing. The hypothesis was that the administration of a glucocorticoid drug would decrease inflammatory damage to epithelial cells in affected mammary glands, resulting in a decreased number of epithelial cells being sloughed into the milk during the acute phase of mastitis.

Materials and Methods

Animals—All study procedures were approved by the Committee on Animal Care at Iowa State University. Thirteen nonpregnant Holstein cows in the first 30 to 60 days of their second or greater lactation were used in the study. Each cow was selected on the basis of a history of no mastitis in the current lactation, no drug treatment during the preceding 30 days, a negative result of bacterial culture of milk samples collected from both front mammary glands 72 hours prior to the induction of mastitis, and a negative result of a California mastitis test performed in the front mammary gland assigned for experimental treatment prior to the start of the study procedures.

Experimental treatment and sample collection—During the morning milking session, a milk sample (approx 30 mL) was collected (designated as 0 hours) from 1 front mammary gland of each of the 13 cows. In 5 cows, mastitis was not experimentally induced, nor was any other treatment administered. In 8 cows, the udder hair was removed from the designated front mammary gland by use of electric clippers, and 4 marks were made on the surface of the mammary gland by use of a rectangular template and a permanent marker. The distance between the marks was measured by use of a flexible ruler to obtain the baseline surface area of the aforementioned mammary gland. Mastitis was then induced in the designated mammary gland of each of those 8 cows via intramammary administration of 100 μg of purified endotoxinb in 20 mL of sterile isotonic saline (0.9% NaCl) solution.

Cows with mastitis were ranked according to their ear tag numbers and assigned in rank order to 1 of 2 treatment groups (4 cows/group) by use of a random numbers list. Development of clinical mastitis was indicated by a rectal temperature ≥ 40°C and an increase in surface area of the affected mammary gland, which was measured by the aforementioned protocol. These signs of clinical mastitis were detected between 4 and 7 hours following endotoxin administration in all 8 cows. When signs of clinical mastitis were detected, 4 cows were subsequently treated with isoflupredone acetatec (20 mg, IV) via a jugular vein, and the other 4 cows received no treatment. At the next milking, a second milk sample (approx 30 mL) was collected from each affected mammary gland in cows with mastitis and the designated mammary gland in cows without mastitis. The second collection was 11 to 12 hours after intramammary endotoxin administration (designated as 12 hours).

For each of the 8 cows with endotoxin-induced mastitis, the morning milk sample was obtained from a bucket into which the milk from the affected mammary gland was collected in the parlor at the regular milking time, whereas evening samples were obtained via hand stripping just prior to milking. To separate effects on SCC and epithelial cell percentage attributable to the different sample collection methods from effects attributable to mastitis or drug treatment, a milk sample was collected from the designated front mammary gland in each of the 5 cows without mastitis by use of the same collection protocol.

Sample analysis—Milk samples for SCC determination were preserved by use of bronopol and mailed to a diagnostic laboratoryd for analysis by use of a cell counter.e Analysis of milk samples for determination of percentage of cytokeratin-positive epithelial cells was performed at the College of Veterinary Medicine at Iowa State University. On the day of collection, those samples were washed with PBS solution containing 0.1% bovine serum albumin and 0.1% sodium azide to remove milk fat and whey. The samples were then preserved by use of 1% paraformaldehyde and stored at 4°C until staining and flow cytometric analysis, which were performed within 48 hours. Before flow cytometric analysis, intracellular staining for cytokeratin was performed by incubating an aliquot from each sample with a 0.1% saponin solution at 4°C for 10 minutes to make the cell membranes permeable. This was followed by incubation with FITC-conjugated monoclonal mouse anti-human cytokeratin IgG1f (1:30 dilution) at 4°C for 1 hour. Anti-human cytokeratin IgG is considered to be a marker of epithelial cells in bovine milk.1 Additional aliquots from each milk sample were processed in the same manner, except that FITC-conjugated mouse IgG1 directed against Aspergillus niger glucose oxidaseg (an enzyme that is not present in mammalian tissues) was used; these aliquots were used as isotype control samples. Flow cytometric analysish variables that were recorded included forward scatter, side scatter, and fluorescent staining properties. To determine the number of cytokeratin-positive epithelial cells in each sample, FITC-positive cells were selected and counted.

Statistical analysis—Changes in SCC and percentage of epithelial cells between milk samples collected at 0 and 12 hours from affected glands in isoflupredone acetate–treated and untreated cows with mastitis and from designated glands in cows without mastitis were assessed. By use of an ANOVA for matched pairs,i the magnitudes of change in these variables in cows with and without mastitis were compared; similarly, the magnitudes of change in isoflupredone acetate–treated and untreated cows with mastitis were compared. Significance was set at a value of P < 0.05.

Results

For the 5 days immediately preceding research procedures, mean daily milk production among the 8 cows in which mastitis was induced was 44.5 kg/d (range, 32.6 to 53.1 kg/d). The mean daily milk production during this period among the 5 cows in which mastitis was not induced was 44.3 kg/d (range, 32.9 to 52.4 kg/d).

Cytokeratin-positive epithelial cells were detected in milk samples stained with FITC-linked anti-human cytokeratin antibodies but not in the isotype control samples. At 0 hours (prior to the induction of mastitis) in the 8 cows that were administered endotoxin via intramammary infusion, the mean ± SD percentage of epithelial cells in milk samples was 2.74 ± 1.93%, and median SCC was 195,000 cells/mL (range, 17,000 to 442,000 cells/mL). At 0 hours in the 5 cows in which mastitis was not induced experimentally, the mean percentage of epithelial cells in milk samples was 5.73 ± 4.52%, and median SCC was 19,000 cells/mL (range, 9,000 to 125,000 cells/mL).

In the milk samples collected at 12 hours, cytokeratin-positive epithelial cell percentages were increased in all cows with mastitis, compared with findings at 0 hours. Except for 1 cow in the isoflupredone acetate–treated group, SCC was also comparatively increased at 12 hours in all cows with endotoxin-induced mastitis. In cows with endotoxin-induced mastitis, mean percentage of epithelial cells increased to 42.11 ± 36.21%, and median SCC increased to 5,237,500 cells/mL (range, 69,000 to 11,036,000 cells/mL). In control cows without mastitis, mean percentage of epithelial cells decreased to 5.31 ± 1.93%, and median SCC increased to 51,000 cells/mL (range, 10,000 to 835,000 cells/mL). At the 12-hour time point, cows with mastitis had a significantly (P = 0.039) greater change in the percentage of cytokeratin-positive epithelial cells and a significantly (P = 0.014) greater change in SCC in milk, compared with the corresponding changes in cows without mastitis. Sample collection technique did have an effect on percentage of epithelial cells and SCC. However, despite the differences attributable to sample collection technique, the changes caused by endotoxin-induced mastitis remained significant. Increases in percentage of epithelial cells and SCC in cows with mastitis were not ameliorated by isoflupredone acetate treatment following the development of clinical signs (P = 0.80 for percentage of epithelial cells; P = 0.61 for SCC). The variability in SCCs and percentages of epithelial cells in milk at 0 and 12 hours among cows with mastitis (that were or were not treated with isoflupredone acetate) and cows without mastitis was evident graphically (Figure 1).

Figure 1—
Figure 1—

Somatic cell counts (A) and percentage of epithelial cells (B) in milk samples collected from 1 mammary gland of cows in their second or greater lactation before (0 hours) and 12 hours after infusion of that mammary gland with 100 μgof bacterial endotoxin to induce mastitis and IV administration of isoflupredone acetate (20 mg, [triangles; n = 4]) or saline (0.9% NaCl) solution (squares; 4) and in samples collected at the same time points from 1 mammary gland of untreated cows without mastitis (circles; 5). For the cows with endotoxin-induced mastitis, treatments were administered after the development of clinical signs of mastitis, approximately 4 to 7 hours after endotoxin administration.

Citation: American Journal of Veterinary Research 70, 6; 10.2460/ajvr.70.6.796

Discussion

In the study reported here, the percentage of cytokeratin-positive cells in milk samples collected from cows before experimental induction of mastitis was within the range of values previously reported1,a for epithelial cells in milk from cows without mastitis. In milk from cows without mastitis, the mean percentage of epithelial cells as evaluated by use of light microscopy was 16.5% after Wright's staining and 17.5% after cytokeratin monoclonal antibody staining.a When evaluated by use of electron microscopy, the percentage range of epithelial cells in milk from cows without mastitis was 0% to 7%.1

The magnitude of the increase in the percentage of epithelial cells in milk samples collected from cows with mastitis in the present study differs from the value reported previously by Miller et al.5 In that study, there was a large influx of neutrophils into the milk following the intramammary administration of endotoxin, with a corresponding reduction in the percentages of other cells types, including epithelial cells; neutrophils comprised 78.0% to 89.8% of cells in the milk of cows following intramammary administration of endotoxin.5 Percentages of other cell types in milk were not assessed in the study reported here, but high percentages of neutrophils would be incompatible with the moderately high percentage of epithelial cells detected after endotoxin administration. The difference between findings of the present study and those of the study by Miller et al5 may be explained by the difference in the interval from endotoxin administration to sample collection; samples in the present study were obtained 12 hours after endotoxin administration, whereas samples in the other study5 were obtained 24 hours after endotoxin administration. The differences in procedures and results between that study5 and the study reported here suggest that a flush of epithelial cells into the milk precedes an increase in neutrophil numbers in milk from cows with endotoxin-induced mastitis.

This sequence of events—marked sloughing of epithelial cells followed by an increase in neutrophil numbers in the mammary gland—was evident over a shorter time period by use of a different method in which mammary glands were infused with 100 million CFUs of pathogenic E coli and examined at various time points after inoculation by use of light microscopy and electron miscroscopy.6 That inoculum of 100 million CFUs is large compared with inocula used in other studies; the authors stated that this number was chosen because “the high dose was expected to mask differences in animal susceptibility” (yet the authors still detected a variable response among experimental animals).6 When much smaller numbers of E coli were used to induce mastitis (50 or 200 CFUs), results of microscopic examination revealed only mild changes in the mammary gland epithelial layer 10 hours after inoculation, but at 14 hours, there were extensive areas where mammary gland epithelium had been completely denuded and the basement membrane was exposed.3 Desquamated areas were more prevalent in the distal portions of the mammary gland, with damage concentrated in the teat sinus; there was less epithelium lost from the lactiferous sinuses and little apparent damage to the glandular tissue.

It has been suggested that endotoxin may not cause epithelial cell damage during coliform mastitis because histologic evidence of necrosis was detected in mammary glands into which sterile E coli culture filtrate was administered, but was not detected in mammary glands into which purified endotoxin was administered.7 In contrast, results of the study reported here suggested that endotoxin does damage mammary gland epithelium, resulting in sloughing of epithelial cells off the basement membrane and into the milk. Lesions in the mammary gland epithelium attributable to E coli infection are rapidly covered by migrating epithelial cells6; perhaps this is why mammary gland epithelial lesions were undetected via histologic examination following intramammary administration of endotoxin in the previous study.7 In that study,7 epithelial cells may have been damaged early and sloughed into the milk; the milk was not examined for epithelial cells. However, by the time the cows were euthanatized and the mammary tissues were collected, the lesions from which the epithelial cells were sloughed may have been covered by migrating epithelial cells.

In the present study, the timing of milk sample collection enabled the observation of high numbers of epithelial cells in the milk. Epithelial cell damage occurs early in the course of endotoxin-induced mastitis. Because the milk samples were collected from the cows at 12 hours after administration of endotoxin, a large influx of epithelial cells into the milk was detected. To our knowledge, this early-phase influx of epithelial cells into the milk has not been previously reported. The influx of neutrophils, which has been described as peaking from 10 to 16 or as many as 27 hours after intramammary administration of endotoxin,8,9 is apt to overwhelm the earlier increase in epithelial cell count. To our knowledge, endotoxin as the sole agent of induction of sloughing of large numbers of mammary epithelial cells into the milk early in the course of mastitic disease has not been previously reported.

Treatment with isoflupredone acetate following the development of clinical signs of mastitis in cows did not limit the magnitude of the increase in percentage of epithelial cells or the SCC in the milk. This lack of treatment effect may be associated with the fact that the drug was not administered until clinical signs were detected at 4 to 7 hours following the intramammary administration of endotoxin. Future research may be directed at determining when neutrophils become the predominant cell type and epithelial cell numbers begin to decrease in milk during the course of endotoxin-induced mastitis and may focus on investigating the correlation between epithelial cell damage and severity of disease. The contribution of sloughing epithelium to increased permeability of the blood-milk barrier and the magnitude of release of inflammatory mediators from the mammary gland may also be investigated because these aspects of epithelial cell damage may affect disease progression and subsequent milk production.

ABBREVIATIONS

FITC

Fluorescein isothiocyanate

SCC

Somatic cell count

a.

Paape MJ, Miller RH. Identification of epithelial cells in bovine milk (abstr). J Dairy Sci 1988;71(suppl 1):232.

b.

Purified lipopolysaccharide from E coli strain 0111:B4, Sigma-Aldrich, St Louis, Mo.

c.

Predef 2x, Pfizer Animal Health, Kalamazoo, Mich.

d.

Iowa Dairy Herd Improvement Association Laboratory, Dubuque, Iowa.

e.

Fossomatic 400 cell counter, FOSS North America, Eden Prairie, Minn.

f.

Product code No. F 0859, DAKO USA, Carpinteria, Calif.

g.

Product code No. X 0943, DAKO USA, Carpinteria, Calif.

h.

FACScan Flow Cytometer, Becton-Dickinson, Franklin Lakes, NJ.

i.

JMP Statistical Software, version 7, SAS institute Inc, Cary, NC.

References

  • 1.

    Lee C-S, Wooding FBP, Kemp P. Identification, properties, and differential counts of cell populations using electron microscopy of dry cow secretions, colostrum and milk from normal cows. J Dairy Res 1980;47:3950.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 2.

    Leitner G, Shoshani E, Krifucks O, et al. Milk leukocyte population patterns in bovine udder infection of different aetiology. J Vet Med B Infect Dis Vet Public Health 2000;47:581589.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 3.

    Frost AJ, Hill AW. Pathogenesis of experimental bovine mastitis following a small inoculum of Escherichia coli. Res Vet Sci 1982;33:105112.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 4.

    Lohuis JA, Van Leeuwen W, Verheijden JH. Effect of dexamethasone on experimental Escherichia coli mastitis in the cow. J Dairy Sci 1988;71:27822798.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 5.

    Miller RH, Paape MJ, Filep R, et al. Flow cytometric analysis of neutrophils in cows' milk. Am J Vet Res 1993;54:19751979.

  • 6.

    Frost AJ, Hill AW, Brooker BE. The early pathogenesis of bovine mastitis due to Escherichia coli. Proc ≥ Soc Lond B Biol Sci 1980;209:431439.

  • 7.

    Frost AJ, Brooker BE, Hill AW. The effect of Escherichia coli endotoxin and culture filtrate on the lactating bovine mammary gland. Aust Vet J 1984;61:7782.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 8.

    Saad AM, Ostensson K. Flow cytofluorometric studies on the alteration of leukocyte populations in blood and milk during endotoxin-induced mastitis in cows. Am J Vet Res 1990;51:16031607.

    • Search Google Scholar
    • Export Citation
  • 9.

    Burton JL, Erskine RJ. Immunity and mastitis: some new ideas for an old disease. Vet Clin North Am Food Anim Pract 2003;19:145.

All Time Past Year Past 30 Days
Abstract Views 54 0 0
Full Text Views 414 302 30
PDF Downloads 124 69 8
Advertisement