Intramammary infections (IMI) are prevalent in non-lactating dairy cattle and their occurrence during periods of significant mammary growth and development (i.e. pregnant heifers and dry cows) is believed to interfere with growth, development, and subsequent milk production. However, direct study of IMI impacts on non-lactating but developing mammary glands is lacking. The objectives of this study were to (1) define how IMI affected total and differential mammary secretion somatic cell counts in mammary glands stimulated to rapidly grow using estradiol and progesterone, and (2) characterize changes in mammary morphology in response to IMI. Mammary growth was stimulated in 19 non-pregnant, non-lactating cows and 2 quarters of each cow were subsequently infused with either saline (n = 19) or Staphylococcus aureus (n = 19). Mammary secretions were taken daily until mammary tissues were collected at either 5 or 10 days post-challenge. Staph. aureus quarter secretions yielded greater concentrations of somatic cells than saline quarters and contained a greater proportion of neutrophils. Staph. aureus mammary tissues exhibited higher degrees of immune cell infiltration in luminal and intralobular stroma compartments than saline quarters. Infected tissues also contained reduced areas of epithelium and tended to have greater amounts of intralob- ular stroma. Results indicate that IMI in non-lactating glands that were stimulated to grow, produced immune cell infiltration into mammary tissues and secretions, which was associated with changes in mammary tissue structure. The observed reduction of mammary epithelium indicates that IMI impair mammary development in rapidly growing mammary glands, which may reduce future reduced milk yields. Introduction Regardless of these well-documented effects in the lac - Bovine mastitis is almost exclusively the result of a bac- tating bovine mammary gland, it is recognized that both terial intramammary infection (IMI) and continues to be non-lactating heifers [4, 5] and dry cows  can also a major challenge for the US and global dairy industries. develop IMI. The effects of mastitis are most apparent, and appreci - Quarter IMI prevalence in nulliparous heifers is esti- ated, in lactating cattle given the considerable volume of mated to be approximately 43%, based on a weighted literature that has described the increase in milk somatic average of summarized survey studies  and approxi- cell count (SCC) , reduced milk yields , and histo- mately 8–25% of quarters in dry cows are expected to pathological changes  that occur in response to IMI. acquire new infections during the dry period between lactations . The occurrence of such IMI in non-lactat - ing mammary glands is concerning given the consider- *Correspondence: email@example.com 1 able mammary growth that occurs during these distinct Dairy Science Department, Virginia Polytechnic Institute and State University, Blacksburg, VA 24060, USA physiological states. For instance, the greatest amount of Full list of author information is available at the end of the article © The Author(s) 2018. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creat iveco mmons .org/licen ses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creat iveco mmons .org/ publi cdoma in/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Enger et al. Vet Res (2018) 49:47 Page 2 of 14 mammary growth and development that transpires dur- Staphylococcus aureus (Staph. aureus) to characterize ing an animal’s life occurs during first gestation in prepa - how presence of IMI influences mammary morphol - ration for lactation . Significant growth also occurs in ogy. The specific objectives of this study were to quantify dry cow mammary glands between lactations, contrib- total and differential somatic cells in mammary secre - uting to the observed increased milk yields in succes- tions from saline and Staph. aureus infused mammary sive lactations [10, 11]. Previously, the effects of IMI in glands and define the infiltration of immune cells into non-lactating, non-pregnant, mammary growth quies- mammary tissues. Additionally, a histological evaluation cent heifer glands were examined, and marked changes was applied to characterize mammary tissue structure to in glandular structure such as reduced areas of epithe- determine if IMI influenced mammary development. lium and increased mammary stroma were described . These results demonstrated that IMI negatively affects non-lactating mammary gland structure, which Materials and methods is expected to impact mammary function (future growth Animal selection and study design and development and milk yields). Despite this recogni- This work was approved by the Virginia Polytechnic tion that IMI impacts non-lactating mammary glands, no Institute and State University Institutional Animal Care studies have investigated how IMI influences histologi - and Use Committee (Protocol #15-196). A total of 19 ani- cal or morphometric changes in glands that are rapidly mals were selected from the Virginia Tech milking dairy growing and developing. herd for this study, and included non-pregnant, clinically The mammary growth and development occurring healthy Holstein or Jersey cows that were being culled for during first gestation and subsequent dry periods is reproductive or production reasons, not for reasons con- largely attributed to the key pregnancy associated hor- cerning udder health. Selected cows were identified from mones, estradiol and progesterone. Substantial literature a larger cohort of milking cows for study inclusion by describes the pivotal roles that these hormones have in collecting 3 aseptic quarter foremilk samples, with 1 day driving mammary epithelial cell proliferation  and between each sampling  during the last week of lac- glandular morphogenesis [14–16] to support subsequent tation for bacterial examination and SCC quantification. lactation and have been reviewed previously . Given Bacterial examination followed the methods outlined by the key role of these hormones, their utility, in a model the National Mastitis Council  in which a 10-µL ali- setting, to stimulate rapid mammary growth and devel- quot of fresh milk was streaked onto blood agar plates opment so that molecular mechanisms involved in mam- (Columbia Blood Agar, Hardy Diagnostics, Santa Maria, mary growth and development may be elucidated shows CA, USA). A local Dairy Herd Information Association promise. Laboratory quantified milk somatic cells using a Fos - It is logical to suspect that an IMI occurring when somatic FC (FOSS North America, Eden Prairie, MN, mammary parenchyma is rapidly growing and develop- USA). Quarters were classified as infected if 2 of the 3 ing, (i.e., in the heifer during late gestation or in the mul- samples taken were culture positive for the same patho- tiparous cow during the second half of the dry period) gen . would be problematic. In short, a key question remains: To be included in the study, cows must have had at Is rapid mammogenesis compatible with the immune least 2 uninfected quarters. Furthermore, cows produc- responses initiated to combat a newly developed IMI, and ing foremilk SCC ≤ 200 000/mL for all quarter samplings can the processes of mammogenesis and IMI eradication were preferentially selected over cows yielding higher coexist without consequence? If there are conflicts, how SCC; none of the utilized animals were treated for Staph. are these manifested? While there are differences in the aureus IMI during the current lactation. Cows were dynamics between the mammary growth experienced identified and dried off in groups due to limited animal in late gestation heifers (parenchyma expansion into the availability. The first group, Group A, contained 6 cows; fat pad) and dry cows during the dry period (cessation of Groups B and C contained 9 and 4 cows, respectively milk secretion, regression, and redevelopment, etc.), both (Table 1). All quarters of all cows were aseptically infused situations entail rapid mammogenesis, driven by estra- via the partial insertion technique  with a commercial diol and progesterone, to establish the parenchymal tis- dry cow therapy product ( ToMORROW , Boehringer sue necessary for lactation. In this study, non-pregnant, Ingelheim Vetmedica Inc., St. Joseph, MO, USA) at dry- dry cows, that had undergone a more extensive involu- off and then moved to pasture. At 35 days dry, cows were tion than that typically experienced by a dry cow due administered a single dose of dinoprost tromethamine to the lack of concurrent pregnancy , were injected (Zoetis, Parsippany, NJ, USA) to synchronize animals to with estradiol and progesterone to stimulate rapid mam- a similar day of estrus and relocated to a sawdust bedded mogenesis. Subsequently, animals were challenged with barn. All quarters were aseptically sampled again at 39, Enger et al. Vet Res (2018) 49:47 Page 3 of 14 Table 1 Experimental animal group demographics and corresponding Staph. aureus Novel inoculums infused into challenge quarters Group Breed Mean last week milk Lactations Mean days Mean last week milk Staph. aureus SCC (cells/mL) completed in milk yield (kg/day) inoculum (CFU) Group A Holstein (n = 6) 141 000 1 (n = 5) 568 25.7 4300 3 (n = 1) Group B Holstein (n = 6); 126 000 1 (n = 7) 391 28.4 6400 Jersey (n = 3) 2 (n = 1) 6 (n = 1) Group C Holstein (n = 4) 251 000 1 (n = 2) 611 22.0 7500 2 (n = 1) 4 (n = 1) 41, and 43 days dry for bacterial examination to confirm Mammary secretion sampling and examination that at least 2 quarters were free of IMI. Mammary secretion samples were obtained from cows Cows began a mammary growth induction protocol by removing gross debris from teats and the base of the at 45 days after dry off. The first day of this induction udder via a single-use paper towel. Teats were dipped protocol marks the beginning of the core experimental in a commercial iodine teat disinfectant (TEAT-KOTE approach used in this trial and will hereafter be referred 10/III, GEA United States, Colombia, MD, USA), which remained on teat skin for at least 30 s  before removal to as day 1. Cows received consecutive daily injections with a single-use paper towel. Teat ends were scrubbed of estradiol and progesterone on days 1 through 7 as with 10 × 10 cm described below to stimulate mammary growth and cotton squares soaked in 70% ethanol, development. Cows were aseptically sampled again on and mammary secretions were aseptically expressed into days 8 and 9 to confirm that mammary glands remained sterile 5-mL round bottom polystyrene tubes. Secretions culture negative. Two culture negative quarters from were immediately placed on ice and transported to the each cow were randomly assigned to receive either an laboratory for culture and somatic cell quantification and differentiation. intramammary infusion of sterile phosphate buffered Secretion samples were first processed for bacterio saline (PBS) (n = 19) or Staph. aureus (n = 19). Secretion - samples were taken immediately before intramammary logical examination using the methods discussed earlier infusion on day 10, and then again on days 11, 12, 13, 14, and then used to determine somatic cell concentration 16, 18, and 20 for bacterial examination and quantifica - (cell/mL) and then differentiate somatic cells. Somatic tion and differentiation of somatic cells. During this sam - cells were quantified using the methods outlined by the pling period, cows were randomly selected for euthanasia National Mastitis Council Subcommittee on Screen- at either 5 days post-challenge (day 15; n = 10) or 10 days ing Tests . Fresh secretion samples were first diluted post-challenge (day 20; n = 9) for collection of mammary either 1:4 (Group A) or 1:10 (Group B and C) in PBS tissues. containing 2.2% bovine serum albumin (BSA). Dupli- cate smears were prepared by spreading 10 µL of the diluted secretion within a 1-cm circle on milk somatic Estradiol and progesterone injections cell counting slides (Bellco Glass Inc., Vineland, NJ, USA) Cows were administered daily subcutaneous injections and then dried at 45 °C on a slide warmer. Smears were of estradiol (0.1 mg/kg BW; Sigma-Aldrich Co., St. Louis, subsequently stained for 2 min by flooding the slide with MO, USA) and progesterone (0.25 mg/kg BW; Sigma- Newman’s Modified Stain Solution (Sigma-Aldrich Co.). Slides were drained of excess stain, dried, and rinsed in Aldrich Co.) on alternating sides of the neck on days 1 3 changes of tap water. Stained smears were visualized through 7 . Estradiol and progesterone were dis- under oil immersion with a 5-mm square reticle (Micro solved in absolute ethanol, mixed with benzyl benzoate, - and sterilized using a 0.45-µm filter. The filtrate was then scope World, Carlsbad, CA, USA), which produced a mixed with autoclaved corn oil, which served as the main countable strip width of 0.050 mm for the microscope injectable carrier. The final injectable solution, containing and reticle combination. Stained cells were enumerated dissolved estradiol and progesterone, was 10% ethanol, by counting the number of cells across the diameter of 20% benzyl benzoate, and 70% corn oil by volume. the circle (11.28 mm) within the defined strip width. Each Enger et al. Vet Res (2018) 49:47 Page 4 of 14 smear was counted by 2 independent counters; thus, a mammary glands was confirmed by diluting and plating total of 4 counts were completed per secretion sample. the final dilution on trypticase soy agar (Becton, Dick - Enumerated smears were used to calculate the SCC of inson and Company) and is reported for each group of the undiluted secretion sample, and these SCC were aver- cows in Table 1. aged to produce a single SCC estimate for each secretion In preparation for infusion, gross debris was removed sample. Final secretion SCC were log transformed. from teats and the base of the udder via a disposable The procedures used to differentiate secretion somatic paper towel, and teats were disinfected using a com- cells were adapted from those described previously . mercial aerosol teat disinfectant (Fight Bac, Deep Valley Briefly, 10 µL of fresh secretion was loaded into the top Farm Inc., Brooklyn, CT, USA). Teats were dried using and bottom chambers of a double cytocentrifuge fun- a single use paper towel after allowing a teat disinfect- nel, and 70-µL of PBS containing 2.2% BSA was added ant contact time of at least 30 s, and teats ends were then to each chamber. The cytocentrifuge funnel, fitted with scrubbed with 10 × 10 cm cotton squares soaked in 70% a slide, was centrifuged at 110 × g in a Shandon Cyto- ethanol. Aseptic secretion samples were collected and Spin 2 (Thermo Fisher Scientific, Waltham, MA, USA) teats were cleaned again using the commercial aerosol for 10 min, which produced duplicate smears of the same teat disinfectant and cotton squares used before infusion. sample on each slide. Slides were dried at room tempera- All quarters were infused via the partial insertion method ture then stained with a Wright–Giemsa stain (Electron using sterile teat cannulas (Jorgenson Labs, Loveland, Microscopy Sciences, Hatfield, PA, USA) for 2.5 min. CO, USA) affixed to loaded syringes. Quarters assigned Slides were then drained of excess stain, placed in a stain to the saline treatment were always infused before Staph. primed phosphate buffer (6.8 pH; electron microscopy aureus challenge quarters. Gloves were changed between sciences) for 4 min, rinsed with deionized water, and cows and if gloves became soiled. dried again at room temperature. Stained slides were cov- erslipped and somatic cells were visualized and differen - Tissue collection and processing tiated by a single operator and classified as being either: Cows were euthanized by captive bolt and exsanguina- (1) neutrophils; (2) macrophages, which would have tion for tissue collection. Udders were labeled for orienta- included any epithelial cells present; or (3) lymphocytes. tion, removed, cleaned of gross debris using paper towels, A total of 100 cells were differentiated for each duplicate and placed on aluminum dissecting trays with the teats smear resulting in 200 cells being differentiated and used facing upward for dissection. Mammary parenchyma tis- to calculate percentages for each cell type. sues were collected from saline and Staph. aureus infused quarters. Mammary tissues were collected from paren- Intramammary challenge chyma proximal to the teat, dorsal to the gland cistern The Staph. aureus Novel strain  was used as the chal- for histological evaluation and fixed in 10% formalin for lenge organism because of its demonstrated ability to 72 h. Formalin fixed tissues were transferred and stored induce apoptosis of bovine mammary epithelial cells in 70% ethanol before being dehydrated in a graded etha- in vitro . Briefly, a single colony of Staph. aureus nol series and embedded in paraffin using an automated Novel was removed from a Columbia Blood Agar plate tissue processor (Leica TP 1020; Leica Biosystems Inc, that had been incubated for 24 h and placed in a flask Buffalo Grove, IL, USA). containing trypticase soy broth (Becton, Dickinson and Company, Franklin Lakes, NJ, USA). The inoculated Tissue histologic analysis flask was incubated at 37 °C for 6 h in a shaking incuba - Paraffin embedded mammary tissues were sectioned tor rotating at 250 rpm . At the end of the incuba- 5 µm thick using a rotary microtome (Model HM 340 E, tion, the bacterial culture was centrifuged at 1600 × g Microm International GmbH, Waldoff, Germany) and for 10 min at 4 °C. The resulting pellet was resuspended floated in a water bath at 42 °C. Relaxed sections were in sterile PBS and washed twice more using these same mounted on Superfrost Plus microscope slides (Thermo procedures. The final bacterial pellet was resuspended Fisher Scientific), drained, and dried at 37 °C for 24 h on in sterile PBS to a concentration of 5 × 10 colony form- a slide warmer. Slides were deparaffinized in 3 changes of ing units (CFU)/mL based on absorbance measured at a xylene substitute (Clear-Rite 3, Thermo Fisher Scien - 600 nm. The adjusted Staph. aureus Novel suspension tific) for 5 min each and rehydrated to deionized water was serially diluted with sterile PBS to a target concentra- using a graded ethanol series. Sections were subsequently tion of 5000 CFU/mL. For intramammary infusion, 1 mL stained with hematoxylin and eosin and coverslipped of the final dilution was aseptically loaded into tuber - using the procedures described previously . culin syringes and transported to the farm on ice. The A single hematoxylin and eosin stained section for actual number of Staph. aureus Novel CFU infused into each experimental quarter was visualized in its entirety, Enger et al. Vet Res (2018) 49:47 Page 5 of 14 and 8 representative lobules were identified and imaged to characterize the degree of immune cell infiltration at 100× to capture a representative profile of the lob - in intralobular stroma and luminal area compartments ules in each section. Areas of interlobular stroma were independently. Intralobular stroma infiltration scores avoided, and only lobules were imaged; this was done to ranged from 1 to 4; a score of 1 was a lobule with no evi- focus this analysis on the functional epithelium within dence of immune cell infiltration (Figure 1, score 1) and the mammary gland. Imaged lobules were classified by a a score of 4 was a lobule that was more than 2/3 invaded scorer blinded to treatments using a graded scoring scale by immune cells (Figure 1, score 4). Luminal infiltration Figure 1 Intralobular stromal immune cell invasion scoring. The presented quarter lobules were used to characterize the degree of intralobular stromal immune cell invasion observed in saline and Staph. aureus quarter lobules; scores 1–4. A Depicts a score of 1 with no infiltration; B depicts a score of 2 where small isolated pockets of infiltration (arrows) are present; C exemplifies a score of 3 where cellular infiltration affects approximately 1/3 of the lobule; and D depicts a score of 4, marked infiltration affecting more than 2/3 of the lobule. Scale bars = 200 µm. Enger et al. Vet Res (2018) 49:47 Page 6 of 14 scores ranged from 1 to 3; a score of 1 was a lobule that Immune cell intralobular stromal invasion and luminal had scant infiltration of immune cells into luminal spaces invasion scores were averaged across the 8 representative (Figure 2, score 1) and a score of 3 reflected a lobule lobules imaged to obtain mean scores for each experi- exhibiting marked luminal infiltration where almost all mental quarter sampled. Intralobular stroma and lumi- lumens contained immune cells (Figure 2, score 3). nal invasion scores served as the dependent variables in A second examination of these same imaged lobules two separate models using the MIXED procedure. Both was conducted to measure the area occupied by different models included the fixed, independent effects of quarter tissue structures, e.g., epithelium, intralobular stroma, treatment (n = 2) and day euthanized (n = 2); the interac- and lumen. Lobule and intralobular tissue structures tive term of quarter treatment and day euthanized was were traced and measured using Image-Pro Plus 7.0 not included based on non-significance (P ≥ 0.30). Cow (Media Cybernetics, Rockville, MD, USA; Figure 3). This nested within day of euthanasia was specified as a ran - was achieved by first tracing and measuring the area of an dom effect in both models, but group of cows (n = 3) was imaged lobule and subsequently tracing and measuring removed as a random effect because it was non-signifi - the epithelial tissue structures contained within the lob- cant in all models when tested as a fixed effect (P ≥ 0.15). ule as they interface with the intralobular stroma. There - Resultant least squares means were contrasted using fore, these epithelial tracings would include the epithelial Fisher’s least significant differences test. structure itself and any luminal areas contained within Measured tissue areas were used to calculate the per- the structure. Finally, lumens alone, within the origi- centage of lobule area occupied by: (1) intralobular nal traced lobule and epithelial structures, were traced, stroma; (2) epithelial structures; and (3) luminal space measured, and recorded. These tissue areas were used for each lobule. These measures were subsequently to calculate the epithelial area alone and the intralobular averaged across the 8 representative lobules imaged to stromal area for each lobule. obtain mean percentages for each experimental quarter sampled. Intralobular stroma, epithelial structure, and Statistical analysis luminal space percentages served as dependent vari- Total secretion SCC were analyzed using the MIXED ables in 3 separate models, which used the MIXED pro- procedure in SAS 9.4 (SAS Institute Inc., Cary, NC, USA) cedure. These models were identical to those previously using log transformed SCC as the dependent variable. described and used to analyze tissue immune cell inva- The final model included the fixed, independent effects sion scores. Least squares means estimated by the models of quarter treatment (n = 2), trial day when secretion were contrasted using Fisher’s least significant differences was collected (n = 8), and their interaction. Cow nested test. within day of euthanasia and cow nested within day of euthanasia interacting with quarter treatment were Results included as random effects. Group of cows (n = 3) was Success of challenge not included as a random effect because it was non-sig - Intramammary Staph. aureus Novel challenge estab- nificant when tested as a fixed effect (P > 0.05). Total log lished IMI in 18 of the 19 infused quarters. All Staph. SCC were analyzed using a repeated measures approach aureus infections persisted until tissues were collected at where trial day served as the repeated time point and cow either 5 or 10 days post-challenge, and all saline infused nested within day of euthanasia interacting with quarter quarters remained culture negative throughout. Chal- treatment was defined as the measure repeated. Least lenged quarters did not display clinical signs of inflam - squares means estimated by the model were compared mation, such as quarters being red, swollen, or hot to using a slice procedure to determine if differences existed the touch, but small flakes were occasionally observed in between treatments within the days secretions were challenged quarter secretions. Secretion and tissue sam- collected. ples collected from the cow that did not develop an IMI Differential cell counts were also analyzed using the in the challenged quarter were not utilized in any of the MIXED procedure of SAS using either, neutrophil, mac- preceding described analyses. As a result, secretions and rophage, or lymphocyte cell type percentages as the tissues were examined for 9 cows that were euthanized dependent variable. Each cell type was analyzed in a 5 days post-challenge and 9 cows euthanized 10 days separate model. The models used to analyze the respec - post-challenge. tive cell types were identical to the model used to analyze the total secretion SCC. A slice procedure was again used Secretion somatic cells to compare the estimated least squares means for each The mean secretion Staph. aureus quarter SCC treatment within day of secretion collection. (7.45 ± 0.06 log cells/mL) was greater than the mean saline quarter SCC (6.77 ± 0.06 log cells/mL; P < 0.001). 10 Enger et al. Vet Res (2018) 49:47 Page 7 of 14 Figure 2 Luminal immune cell invasion scoring. The presented quarter lobules were used to characterize the degree of luminal immune cell invasion observed in saline and Staph. aureus quarter lobules; scores 1–3 are presented. A Depicts a score of 1 which signifies no luminal infiltration; B depicts a score of 2 with infiltration in fewer than half the lobule lumens (arrows); and C denotes a score of 3 which is marked infiltration in most lumens. Scale bars = 200 µm. Enger et al. Vet Res (2018) 49:47 Page 8 of 14 percentages, which were lower in challenge quarters for every day sampled post-challenge (P < 0.001; Figure 4C). In addition, some neutrophils collected from challenge quarters in the present study were observed to contain intracellular Staph. aureus (Figure 5B). Lymphocyte per- centages were lower in challenged quarters than saline quarters for the first 3 days following challenge (P < 0.05), but not for the remainder of the days sampled (P > 0.05; Figure 4D). Eosinophils were also observed in secretions (Fig- ure 5A) but made up less than 1% of the differential cell count, preventing comparisons from being made between saline and challenged quarters. Binucleated giant cells and lumen resident cells undergoing mitosis were also sporadically observed (Figures 5C and D). Tissue measures Immune cell infiltration scores were not affected by day Figure 3 Measurement of lobule tissue area percentages in imaged lobules. Example image of the tracings applied to imaged of euthanasia (P ≥ 0.25) but were greater for challenged lobules to measure lobule, intralobular stroma, epithelial, and luminal quarter lobules than saline quarter lobules for both the areas. Scale bar = 200 µm. luminal (1.68 vs 1.13 ± 0.09; P < 0.001) and intralobular stroma compartments (1.85 vs 1.50 ± 0.09; P = 0.005) (Figure 6). Saline quarter lobules were essentially devoid Additionally, secretion SCC were significantly influ - of neutrophils in both the luminal and intralobular stro- enced by treatment interacting with trial day (P < 0.001; mal compartments (Figure 7A), but neutrophils were fre- Figure 4A). Overall, secretion SCC appeared unchanged quently observed in both compartments of Staph. aureus in saline quarters throughout the trial’s duration (Fig- challenged quarter lobules (Figure 7B). Lymphocytes ure 4A) and these SCC were significantly lower for all could be observed in both saline and challenged quar- days sampled post-challenge relative to challenged quar- ters but were more abundant in the latter. It is notewor- ters (P < 0.05). thy to state that lymphocytes appeared to preferentially Neutrophil, macrophage, and lymphocyte percentages accrue in intralobular stromal compartments rather than measured in saline and challenged quarter mammary luminal spaces (Figure 7C). Plasma cells were abundant secretions are stratified by day of trial and are illus - in both saline and challenged quarter tissues, but did not trated in panels B–D in Figure 4; representative images grossly appear to be more abundant in one vs the other of each cell type are shown in Figures 5A and B. Over- (Figure 7D). all, the mean percentage of neutrophils in challenged Lobules in Staph. aureus challenged quarters exhib- quarters (47.2 ± 2.3%) was greater than the mean neutro- ited a greater percentage of luminal space (7.7% vs phil percentage in saline quarters (7.1 ± 2.3%; P < 0.001); 5.4% ± 0.6%; P = 0.004), a reduced percentage of epithe- conversely, the mean percentages of macrophages and lial area (33.3% vs 38.1% ± 1.1%; P < 0.0001), and tended lymphocytes in challenge quarters (17.7 ± 3.0% and to have a greater percentage of intralobular stromal 34.4 ± 2.1%, respectfully) were lower than those meas- area (59.0% vs 56.5% ± 1.3%; P = 0.1) than saline infused ured in saline quarters (51.0 ± 3.0% and 40.1 ± 2.1%, glands (Figure 8). respectfully; P ≤ 0.03). Aside from these main effects, a significant interac - Discussion tion existed between quarter treatment and trial day Secretion somatic cells in its effect on all the measured cell type percentages The first objective of this study was to characterize the (P < 0.001). In general, saline quarter cell type percentages somatic cell and differential cell count response result - remained stable throughout the trial, but cell type per- ing from infusion of saline and Staph. aureus Novel centages in Staph. aureus quarters changed in response into non-lactating mammary glands. The absence of to challenge. For instance, neutrophil percentages were an increase in secretion SCC in saline infused quarters greater for every day sampled post-challenge in chal- was expected. This indicates that no significant immune lenged quarters compared to saline quarters (P < 0.001; response resulted from saline infusion and complements Figure 4B) and this appeared to impact macrophage the observation that saline quarters remained culture Enger et al. Vet Res (2018) 49:47 Page 9 of 14 Figure 4 Total SCC and differential cell type percentages in collected mammary secretions. Total secretion SCC A and corresponding differential cell type percentages B–D collected from saline (n = 18) and Staph. aureus (n = 18) infused quarters. Error bars represent the standard error of the respective means. *P ≤ 0.05. negative throughout the trial. Conversely, the signifi - in the bovine mammary gland that respond to IMI cant increase in secretion SCC observed in response to  and similar changes have been described in other Staph. aureus challenge was expected and establishes challenge trials in heifers  and lactating cows . that an immune response resulted in these quarters. No Neutrophils have also been described as being the pre- reports could be identified describing the SCC response dominate cell type in infected dry cow gland secretions of non-lactating mammary glands to Staph. aureus chal- . Lymphocytes were the second most predomi- lenge but it has been reported that secretions collected nate cell type observed in challenged quarters but, this from uninfected quarters of pregnant dry cows, contain observation is not consistent with two previous reports a lower SCC than those collected from infected quar- [31, 37] that reported that macrophages were the sec- ters . Furthermore, the reported SCC of uninfected ond most predominant cell type infected dry cow and infected quarters in the previous report  are secretions. Reasons for this disparity are unclear but comparable to the SCC observed here, indicating that are posited to be attributed to the fact that the secre- the SCC response was similar to that of the pregnant tions collected in those two previous studies were from dry cow. Aside from the SCC response in non-lactating quarters naturally and chronically infected with an glands, previous reports in lactating cows [32–34] have assortment of different mastitis pathogens. The secre - described increases in SCC resulting from Staph. aureus tions collected herein were collected from quarters challenge. responding to a Staph. aureus challenge and are thus The observed increase in neutrophil percentages in more associated with a rapid immune response rather Staph. aureus infused quarters were expected given than an immune response linked to chronic infec- neutrophils are the main innate immune effector cell tions. Also, the immune response generated herein was Enger et al. Vet Res (2018) 49:47 Page 10 of 14 Figure 5 Somatic cells observed in collected mammary secretions. An eosinophil (E), lymphocyte (L), and macrophage (M) are depicted in A. B Depicts two neutrophils collected from a challenged quarter with the bottom neutrophil containing intracellular Staph. aureus (arrow). A binucleated macrophage is shown in C (arrow). These cells are suspected to originate from lumen resident macrophages that undergo incomplete cell division like the mitotic cell (arrow) shown in D. Scale bars = 10 µm. specific to Staph. aureus and it has been previously within secretions containing binucleated cells. Addition- demonstrated that different mastitis pathogens elicit ally, binucleated epithelial cells have also been described different cytokine profiles and immune responses   and would contribute to the presence of these cells which would, in part, explain the discrepancy between in milk or mammary secretions should they be sloughed these results. from the basement membrane into the lumen. Binucleated giant cells, like those observed here, have been reported and discussed previously [26, 39, 40], Tissue measures but how these cells originate in the gland is not entirely The second key objective of this study was to character - clear. It is possible that a subset of these cells originated ize mammary tissue structure in quarters infused with from what appear to be lumen resident macrophages saline and Staph. aureus to define how Staph. aureus undergoing incomplete cell division (Figure 5D) given challenge affected mammary gland structure and devel - that these mitotic cells were most commonly observed opment. The observed infiltration of immune cells into Enger et al. Vet Res (2018) 49:47 Page 11 of 14 stroma than saline lobules as similarly reported here. Differing from the results of this previous study was the observation that challenged quarters exhibited larger luminal areas than saline infused quarters. The larger luminal areas observed herein are suspected to be con- sequence of the initial immune cell influx into the gland’s lumen, bringing fluid across the epithelium, given that IMI reduces epithelial integrity and results in increased concentrations of BSA  and ions  in milk from affected quarters. The reason for the lack of agreement between the previous study and the results of the pre- sent may also be attributed to differences infection dura - tion. The cows used here were euthanized 5 and 10 days post-challenge, whereas the previous study euthanized Figure 6 Mean immune cell infiltration scores for lumen and heifers 2–3 weeks post-challenge . This longer infec - intralobular stroma areas. Mammary tissues were collected from tion duration would have allowed the sustained immune 18 saline and 18 Staph. aureus infused quarters and 8 representative response to affect glandular structure over a longer lobules were scored for each experimental quarter. Error bars period of time. As a result, continued deposition and represent the standard error of the respective mean immune cell accumulation of connective tissues, leading to fibrosis, infiltration scores. Asterisks denote differences between saline and Staph. aureus quarter treatments within intralobular stoma and would result and begun to displace luminal space as fluid luminal areas. **P ≤ 0.01, ***P ≤ 0.001. from infected quarters was reabsorbed, given the initial immune response had begun to subside. The dry cows used in this study were treated with estra - Staph. aureus infected tissues was expected and comple- diol and progesterone to stimulate mammary growth ments the influx of immune cells into challenged quarter and development so that IMI impact in growing and mammary secretions that would result from leukocyte developing mammary glands could be investigated. In diapedesis from blood vessels to mammary gland lumens this context, the reductions in epithelial areas and ten- in response to the presence of Staph. aureus. The abun - dency for challenged glands to contain greater areas of dance of plasma cells observed in mammary tissues from stromal tissue indicate that challenged glands failed to both treatments is believed to be consequence of the develop comparable amounts of epithelium and experi- estradiol and progesterone injections given their signifi - enced varying degrees of connective tissue deposition in cance in colostrogenesis [42, 43]. Not surprisingly, a simi- the gland as a result of IMI. It is currently unknown what lar hormonal induction model to that used here has been chief mechanisms are responsible for these changes in used to investigate bovine colostrogenesis mechanisms glandular structure, but the deposition and accumulation . Examination of colostrum formation and immuno- of connective tissues in affected tissues, displacing mam - globulin transport was not an objective in the present mary epithelium, and the immune response produced study but the abundance of plasma cells in the collected to address the presence of bacteria may interfere with tissues may warrant consideration for future studies mammary epithelial cell proliferation and alter gland investigating colostrogenesis mechanisms, particularly development, perhaps in the long term. Such changes those concerned with immunoglobulin production and in glandular development are expected to contribute, in transport. part, to the reduced milk yields reported for heifers that No reports examining the histopathological response freshen with IMI [47, 48] as well as the reduced milk of a mastitis challenge in non-pregnant, dry cows could yields described for cow quarters that freshen with IMI be identified with which to compare the tissue area per - compared to paired, uninfected, lateral quarters within centages reported here. However, a previous report the same cow . described the histopathological response of mammary Interestingly, neither day of euthanasia for tissue col- tissue in non-pregnant heifers after Staph. aureus chal- lection nor the interaction between day of euthanasia lenge  and reported a similar reduction in the epithe- and treatment significantly influenced any of the respec - lial areas relative to uninfected quarters. This previous tive lobule area percentages measured (P ≥ 0.18). This study also reported that Staph. aureus infected tissues was unexpected, but is not entirely surprising given this contained greater areas of stroma tissue than uninfected study’s experimental design. This study was designed to quarters and complements the tendency of Staph. aureus first allow for treatment comparisons to be made within quarter lobules to contain greater areas of intralobular animal to control for inter-animal variation. When an Enger et al. Vet Res (2018) 49:47 Page 12 of 14 Figure 7 Cellularity of tissues from saline and Staph. aureus quarter lobules. A Depicts tissues from a saline infused quarter exhibiting diffuse intralobular stroma and non-secretory epithelium. Neutrophilic infiltration (arrows) of luminal and intralobular stoma compartments is depicted in B for tissues from a Staph. aureus infused quarter lobule. C Exemplifies the preferential infiltration of lymphocytes into intralobular stroma areas (dashed outline) that could be observed in saline quarters but were more frequent in Staph. aureus quarters. Plasma cells (arrows) could be observed in both saline and Staph. aureus quarter lobules D. Scale bars in A and C = 50 µm, B and D = 10 µm. examination of day of euthanasia was applied to this challenge; collection of tissues closer to the initial chal- design, resulting in the nesting of animals within day lenge may have allowed for changes over time in gland euthanized, control for between animal variation was structure to be better appreciated. lost, which significantly influenced the studies ability to In conclusion, Staph. aureus challenge of rapidly detect quarter treatment differences between animals growing, non-lactating mammary glands increased euthanized at 5 or 10 days post-challenge. Furthermore, immune cell invasion in mammary secretions and perhaps examining tissues 5 days post-challenge was too both intralobular stroma and luminal compartments late to capture the temporal changes occurring in gland of the mammary gland. This invasion was associated morphology resulting from intramammary Staph. aureus with changes in mammary structure as Staph. aureus Enger et al. Vet Res (2018) 49:47 Page 13 of 14 Ethics approval and consent to participate Not applicable. Author details Dairy Science Department, Virginia Polytechnic Institute and State University, Blacksburg, VA 24060, USA. Animal and Dairy Science Department, University of Georgia, Athens, GA 30602, USA. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in pub- lished maps and institutional affiliations. Received: 20 April 2018 Accepted: 9 May 2018 References 1. Dohoo IR, Leslie KE (1991) Evaluation of changes in somatic cell counts as indicators of new intramammary infections. Pre Vet Med 10:225–237 2. Jones GM, Pearson RE, Clabaugh GA, Heald CW (1984) Relation- Figure 8 Lobule area percentages occupied by luminal space, ships between somatic cell counts and milk production. J Dairy Sci epithelium, and intralobular stroma in experimental quarters. 67:1823–1831 Mammary tissues were collected from 18 saline and 18 Staph. aureus 3. Akers RM, Nickerson SC (2011) Mastitis and its impact on structure and function in the ruminant mammary gland. J Mammary Gland Biol Neo- infused quarters and 8 representative lobules from each quarter plasia 16:275–289 were used to quantify lobule areas occupied by intralobular stroma, 4. Trinidad P, Nickerson SC, Alley TK (1990) Prevalence of intramammary epithelium, and luminal space. Error bars represent the standard infection and teat canal colonization in unbred and primigravid dairy error of the respective mean percentages. Dagger symbol and heifers. J Dairy Sci 73:107–114 asterisks denote differences between saline and Staph. aureus quarter 5. Fox LK, Chester ST, Hallberg JW, Nickerson SC, Pankey JW, Weaver LD treatments within tissue structures; P = 0.1, **P ≤ 0.01, ***P ≤ 0.001. (1995) Survey of intramammary infections in dairy heifers at breeding age and first parturition. J Dairy Sci 78:1619–1628 6. Eberhart RJ (1986) Management of dry cows to reduce mastitis. J Dairy challenged quarters exhibited reduced areas of epithe- Sci 69:1721–1732 lium and tended to have greater areas of intralobular 7. Fox LK (2009) Prevalence, incidence and risk factors of heifer mastitis. Vet Microbiol 134:82–88 stroma relative to saline infused quarters. When these 8. Arruda AG, Godden S, Rapnicki P, Gorden P, Timms L, Aly SS, Lehen- histological changes are taken together, it suggests that bauer TW, Champagne J (2013) Randomized noninferiority clinical trial IMI in rapidly growing non-lactating mammary glands evaluating 3 commercial dry cow mastitis preparations: I. Quarter-level outcomes. J Dairy Sci 96:4419–4435 limit mammary growth and development, which is 9. Tucker HA (1987) Quantitative estimates of mammary growth during expected to negatively impact future milk yield, milk various physiological states: a review. J Dairy Sci 70:1958–1966 quality, and productivity of the animal in the herd. 10. Capuco AV, Akers RM, Smith JJ (1997) Mammary growth in Holstein cows during the dry period: quantification of nucleic acids and histology. J Dairy Sci 80:477–487 11. Capuco AV, Akers RM (1999) Mammary involution in dairy animals. J Abbreviations Mammary Gland Biol Neoplasia 4:137–144 IMI: intramammary infection; SCC: somatic cell count; Staph. aureus: Staphy- 12. Trinidad P, Nickerson SC, Adkinson RW (1990) Histopathology of staphylo- lococcus aureus; PBS: phosphate buffered saline; BSA: bovine serum albumin; coccal mastitis in unbred dairy heifers. J Dairy Sci 73:639–647 CFU: colony forming units. 13. Woodward TL, Beal WE, Akers RM (1993) Cell interactions in initiation of mammary epithelial proliferation by oestradiol and progesterone in Competing interests prepubertal heifers. J Endocrinol 136:149–157 The authors declare that they have no competing interests. 14. Sud SC, Tucker HA, Meites J (1968) Estrogen-progesterone requirements for udder development in ovariectomized heifers. J Dairy Sci 51:210–214 Authors’ contributions 15. Howe JE, Heald CW, Bibb TL (1975) Histology of induced bovine lac- BDE, SCN, and RMA designed the study and BDE, CEC, TTY, KME, and CLMP togenesis. J Dairy Sci 58:853–860 executed the animal trial. BDE and CEC quantified SCC and BDE differentiated 16. Croom WJ, Collier RJ, Bauman DE, Hays RL (1976) Cellular studies of mam- somatic cells. BDE preformed all histological measures. BDE and TTY pre- mary tissue from cows hormonally induced into lactation: histology and formed the described statistical analyses. The manuscript was drafted by BDE, ultrastructure. J Dairy Sci 59:1232–1246 SCN, and RMA. All authors read and approved the final manuscript. 17. Akers RM (2017) A 100-year review: mammary development and lacta- tion. J Dairy Sci 100:10332–10352 Acknowledgements 18. Capuco AV, Li M, Long E, Ren S, Hruska KS, Schorr K, Furth PA (2002) Con- Dr Lawrence K. Fox, professor, Washington State University, Pullman, WA, USA current pregnancy retards mammary involution: effects on apoptosis and is thanked for his kind gift of the Staph. aureus Novel strain. The Virginia Tech proliferation of the mammary epithelium after forced weaning of mice. Farm Staff are thanked for their assistance with conducting this study. This Biol Reprod 66:1471–1476 work was supported by a USDA-NIFA-AFRI competitive predoctoral fellowship 19. Dohoo I, Andersen S, Dingwell R, Hand K, Kelton D, Leslie K, Schukken (2017-67011-26049), awarded to B. D. Enger, a Virginia Agricultural Council Y, Godden S (2011) Diagnosing intramammary infections: compari- Grant ( VAC Project No. 685) awarded to R. M. Akers, and Professorship funds son of multiple versus single quarter milk samples for the identifica- (Horace E. and Elizabeth F. Alphin Professorship, Grant VT 438934) awarded to tion of intramammary infections in lactating dairy cows. J Dairy Sci R. M. Akers. 94:5515–5522 Enger et al. Vet Res (2018) 49:47 Page 14 of 14 20. Hogan JS, González RN, Harmon RJ, Nickerson SC, Oliver SP, Pankey JW, 35. Paape M, Mehrzad J, Zhao X, Detilleux J, Burvenich C (2002) Defense of Smith KL (1999) Laboratory handbook on bovine mastitis. National Masti- the bovine mammary gland by polymorphonuclear neutrophil leuko- tis Council, Madison cytes. J Mammary Gland Biol Neoplasia 7:109–121 21. Andersen S, Dohoo IR, Olde Riekerink R, Stryhn H, Conference MRW 36. Jackson KA, Nickerson SC, Kautz FM, Hurley DJ (2012) Technical note: (2010) Diagnosing intramammary infections: evaluating expert opinions development of a challenge model for Streptococcus uberis mastitis in on the definition of intramammary infection using conjoint analysis. J dairy heifers. J Dairy Sci 95:7210–7213 Dairy Sci 93:2966–2975 37. Sordillo LM, Nickerson SC, Akers RM, Oliver SP (1987) Secretion composi- 22. Boddie RL, Nickerson SC (1986) Dry cow therapy: effects of method of tion during bovine mammary involution and the relationship with drug administration on occurrence of intramammary infection. J Dairy mastitis. Int J Biochem 19:1165–1172 Sci 69:253–257 38. Bannerman DD, Paape MJ, Lee JW, Zhao X, Hope JC, Rainard P (2004) 23. Ball S, Polson K, Emeny J, Eyestone W, Akers RM (2000) Induced lactation Escherichia coli and Staphylococcus aureus elicit differential innate in prepubertal Holstein heifers. J Dairy Sci 83:2459–2463 immune responses following intramammary infection. Clin Diagn Lab 24. Enger BD, Fox LK, Gay JM, Johnson KA (2015) Reduction of teat skin mas- Immunol 11:463–472 titis pathogen loads: differences between strains, dips, and contact times. 39. Nickerson SC, Sordillo LM (1985) Role of macrophages and multinucleate J Dairy Sci 98:1354–1361 giant cells in the resorption of corpora amylacea in the involuting bovine 25. Brazis AR, Jasper DE, Levowitz D, Newbould FH, Postle DS, Schultze WD, mammary gland. Cell Tissue Res 240:397–401 Smith JW, Ullmann WW (1968) Direct microscopic somatic cell count in 40. Nickerson SC, Sordillo LM (1987) Origin, fate, and properties of multinu- milk. J Milk Food Technol 31:350–354 cleated giant cells and their association with milk-synthesizing tissues of 26. Williams JE, Price WJ, Shafii B, Yahvah KM, Bode L, McGuire MA, McGuire the bovine mammary gland. Immunobiology 174:200–209 MK (2017) Relationships among microbial communities, maternal 41. Rios AC, Fu NY, Jamieson PR, Pal B, Whitehead L, Nicholas KR, Lindeman cells, oligosaccharides, and macronutrients in human milk. J Hum Lact GJ, Visvader JE (2016) Essential role for a novel population of binucleated 33:540–551 mammary epithelial cells in lactation. Nat Commun 7:11400 27. Smith TH, Fox LK, Middleton JR (1998) Outbreak of mastitis caused by one 42. Barrington GM, McFadden TB, Huyler MT, Besser TE (2001) Regulation of strain of Staphylococcus aureus in a closed dairy herd. J Am Vet Med Assoc colostrogenesis in cattle. Livest Prod Sci 70:95–104 212:553–556 43. Weisz-Carrington P, Roux ME, McWilliams M, Phillips-Quagliata JM, Lamm 28. Bayles KW, Wesson CA, Liou LE, Fox LK, Bohach GA, Trumble WR (1998) ME (1978) Hormonal induction of the secretory immune system in the Intracellular Staphylococcus aureus escapes the endosome and induces mammary gland. Proc Natl Acad Sci U S A 75:2928–2932 apoptosis in epithelial cells. Infect Immun 66:336–342 44. Stark A, Wellnitz O, Dechow C, Bruckmaier R, Baumrucker C (2015) Colos- 29. Kelsey JA, Bayles KW, Shafii B, McGuire MA (2006) Fatty acids and mono - trogenesis during an induced lactation in dairy cattle. J Anim Physiol acylglycerols inhibit growth of Staphylococcus aureus. Lipids 41:951–961 Anim Nutr 99:356–366 30. Tucker HL, Parsons CL, Ellis S, Rhoads ML, Akers RM (2016) Tamoxifen 45. Chockalingam A, Paape MJ, Bannerman DD (2005) Increased milk levels impairs prepubertal mammary development and alters expression of transforming growth factor-α, β1, and β2 during Escherichia coli- of estrogen receptor alpha (ESR1) and progesterone receptors (PGR). induced mastitis. J Dairy Sci 88:1986–1993 Domest Anim Endocrinol 54:95–105 46. Linzell JL, Peaker M (1972) Day-to-day variations in milk composition in 31. Jensen DL, Eberhart RJ (1981) Total and differential cell counts in the goat and cow as a guide to the detection of subclinical mastitis. Br secretions of the nonlactating bovine mammary gland. Am J Vet Res Vet J 128:284–285 42:743–747 47. Owens WE, Nickerson SC, Washburn PJ, Ray CH (1991) Efficacy of a 32. Schukken YH, Leslie KE, Barnum DA, Mallard BA, Lumsden JH, Dick cephapirin dry cow product for treatment of experimentally induced PC, Vessie GH, Kehrli ME (1999) Experimental Staphylococcus aureus Staphylococcus aureus mastitis in heifers. J Dairy Sci 74:3376–3382 intramammary challenge in late lactation dairy cows: quarter and cow 48. Oliver SP, Lewis MJ, Gillespie BE, Dowlen HH, Jaenicke EC, Roberts RK effects determining the probability of infection. J Dairy Sci 82:2393–2401 (2003) Prepartum antibiotic treatment of heifers: milk production, milk 33. Middleton JR, Luby CD, Viera L, Tyler JW, Casteel S (2004) Short commu- quality and economic benefit. J Dairy Sci 86:1187–1193 nication: influence of Staphylococcus aureus intramammary infection on 49. Smith A, Dodd FH, Neave FK (1968) The effect of intramammary infection serum copper, zinc, and iron concentrations. J Dairy Sci 87:976–979 during the dry period on the milk production of the affected quarter at 34. Nickerson SC (1980) Histological and cytological response of the bovine the start of the succeeding lactation. J Dairy Res 35:287–290 mammary gland to experimental S. aureus infection. Virginia Polytechnic Institute and State University, Blacksburg Ready to submit your research ? Choose BMC and benefit from: fast, convenient online submission thorough peer review by experienced researchers in your ﬁeld rapid publication on acceptance support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions
Veterinary Research – Springer Journals
Published: Jun 5, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
All the latest content is available, no embargo periods.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud