- Email: [email protected]
Streptococcus uberis internalizes and persists in bovine mammary epithelial cells
Microbial Pathogenesis 40 (2006) 279–285 www.elsevier.com/locate/micpath
Streptococcus uberis internalizes and persists in bovine mammary epithelial cells Batcha Tamilselvam a, Rau´l A. Almeida a,*, John R. Dunlap b, Stephen P. Oliver a a
Department of Animal Science, Institute of Agriculture, Food Safety Center of Excellence, The University of Tennessee, 60 McCord Hall, Knoxville, TN 37996, USA b Department of Botany, The University of Tennessee, Knoxville, TN 37996, USA Received 18 July 2005; received in revised form 3 January 2006; accepted 28 February 2006
Abstract Streptococcus uberis is one of the most important emerging bovine mastitis pathogens and chronic persistent intramammary infections (IMI) are often described. To define the ability of S. uberis to persist intracellularly, studies on time-dependent internalization and survival of S. uberis strains in bovine mammary epithelial cells were conducted. Two S. uberis strains (UT366 and UT888) and a Staphylococcus aureus strain used as positive control, all isolated from cows with clinical mastitis were cocultured with bovine mammary epithelial cells (MAC-T) and persistent survival in host epithelial cells for extended periods (120 h) studied. Of S. uberis strains tested, UT366 showed highest internalization values at 60 min of incubation whereas at 8 h of incubation the corresponding values for UT888 were the highest. Of both strains of S. uberis tested, UT366 seems to internalize bovine mammary cells more efficiently initially, however, during the first 8 h, UT888 seems to survive intracellularly better than UT366. Results showed that both S. uberis strains could survive intracellularly up to 120 h without apparent loss of host cells viability. S. aureus internalized more efficiently than all strains tested and host cell death was observed after 72 h of incubation. These results indicate that S. uberis can survive within mammary epithelial cells for extended time without apparent loss of host cells viability. Intracellular persistence of S. uberis may be associated with the spread of the infection to deeper tissues and development of persistent IMI. q 2006 Elsevier Ltd. All rights reserved. Keywords: Intracellular persistence; Streptococcus uberis; Bovine mastitis; Pathogenesis
1. Introduction Mastitis remains economically the most important disease of dairy cattle throughout the world, accounting 38% of the direct costs of the common production diseases [1]. Streptococcus uberis is an environmental pathogen responsible for a high proportion of cases of clinical and subclinical mastitis in lactating cows. S. uberis is the predominant organism isolated from mammary glands during the non-lactating period accounting for approximately 90% total cases of environmental streptococcal mastitis in heifers, within the first 5 days of lactation [2,3]. Bovine intramammary infections (IMI) caused by S. uberis could progress with clinical or subclinical symptoms and very often, acute IMI lead to chronic infections that can persist in the infected mammary gland for more than one lactation [2]. * Corresponding author. Tel.: C1 865 974 0991; fax: C1 865 974 3394. E-mail address: [email protected] (R.A. Almeida).
0882-4010/$ - see front matter q 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.micpath.2006.02.006
Internalization of bovine mastitis causing pathogens into host cells including S. uberis has been previously described [4–7]. An intracellular microenvironment provides these pathogens with protection from the immune system and non-specific antibacterial factors present in bovine milk. Earlier work from our laboratory showed that for internalization S. uberis exploits host cell cytoskeleton and signal transduction, as well as expresses de novo proteins [8–11]. Although information is available on adherence to and internalization of S. uberis into bovine mammary epithelial cells, little is known about postinternalization events. Therefore, the objective of this study was to further characterize post-internalization period focusing on time-dependent internalization and fate of intracellular S. uberis using classical internalization and fluorescent assays.
2. Results 2.1. Internalization of S. uberis into MAC-T cells Results from the first hour of coculture showed that S. uberis UT366 internalize significantly more efficiently than S. uberis
B. Tamilselvam et al. / Microbial Pathogenesis 40 (2006) 279–285
2.2. Intracellular persistence of S. uberis in MAC-T cells In order to assess the survival ability of internalized S. uberis strains over a 6 day period, MAC-T cells were cocultured with S. uberis strains (UT888 and UT366) for 1 h, washed and treated with antibiotic solution for 2 h and then incubated with 0.25! diluted antibiotic solution for 120 h. CFU/ml internalized S. uberis
A 5.E+04 UT888 UT366
4.E+04 3.E+04 2.E+04 1.E+04 0.E+00
1
4
6
8
Time (hours) Log10 cfu/ml internalized bacteria
B 100000000 10000000 UT888 UT366 UT955
1000000 100000 10000 1000
1
4
6
8
Time (hours)
Fig. 1. Internalization of Streptococcus uberis into bovine mammary epithelial cells. S. uberis UT888 ( , :) S. uberis UT366 (Q, &) and Staphylococcus aureus UT 955 (C) were cocultured with a bovine mammary epithelial cell line (MAC-T cells) and colony forming units per milliliter (cfu/ml) of internalized bacteria calculated at 1, 4, 6 and 8 h. Data are presented as colony forming units per milliliter (A) or log10 cfu/ml (B) of internalized bacteria are the mean of three experiments run in triplicate. Error bars in panel A represent the standard error of the mean.
A cfu/ml internalized S. uberis
UT888 (PR0.005). Although results from the first 6 h showed that S. uberis UT366 internalized better, results obtained at 8 h of coculture suggest that S. uberis UT888 survived intracellularly better than UT366 (Fig. 1(A)). Contrary to the steady loss of intracellular viability observed in S. uberis UT366, UT888 intracellular colony forming units per milliliter were constant during the assay. The progressive decline of S. uberis UT366 viability was marked enough that at 8 h of incubation the percentage of intracellular S. uberis UT366 was lower than that of S. uberis UT888 reaching values 66% lower than that observed at 1 h of coculture (Fig. 1). When internalization values of the controls were compared with S. uberis, Staphylococcus aureus UT955 showed overall highest internalization values (Fig. 2). Maximum internalization for this pathogen was at 6 h of incubation and were approximately 2.5 log10 higher than those observed at 1 h of coculture, and 4.5 log higher than the corresponding for S. uberis UT888 and UT366 at the same sampling point (6 h). Cultures of supernatant showed no bacterial growth (data not shown) suggesting that extracellular bacteria were killed by the antibiotic treatment. No internalization of Escherichia coli DH5a into MAC-T cell was detected at any sampling time.
5.E+05 4.E+05 3.E+05 2.E+05 1.E+05 0.E+00
2
3 Time (days)
4
5
3
4
5
B Log10 cfu/ml internalized bacteria
280
1.E+08 1.E+07 1.E+06 1.E+05 1.E+04 1.E+03 1
2
Time (days)
Fig. 2. Intracellular survival of S. uberis into bovine mammary epithelial cells. S. uberis UT888 ( , :) S. uberis UT366 (Q, &) and S. aureus UT 955 (C) were cocultured with a bovine mammary epithelial cell line (MAC-T cells) and colony forming units per milliliter (cfu/ml) of internalized bacteria calculated at 1, 2, 3, 4 and 5 days. Data are presented as colony forming units per milliliter (A) or log10 cfu/ml (B) of internalized bacteria and are the mean of three experiments run in triplicate. Error bars in panel A represent the standard error of the mean.
Cultures of supernatant showed no bacterial growth (data not shown) suggesting that extracellular bacterial were killed by the antibiotic treatment. Colony forming units per milliliter of internalized S. uberis recovered after 24 h of incubation was slightly higher (UT888, PZ0.26; UT366, PZ0.04) than the recovered at 8 h of coculture (Fig. 2). After the initial 24 h of incubation, internalization of host cell was significantly higher for UT888 than for UT366 (PZ0.02) and such strain effect was observed during the remaining of the assay. Results presented in Fig. 2A show that colony forming units per milliliter of internalized S. uberis increased 1.35 and 1.43 log10 for UT888 and 1.63 and 1.72 log10 for UT366 at days 2 and 3, respectively, compared with values observed on day 1. From day 4 and to the end of the experiment, S. uberis UT888 and UT366 colony forming units per milliliter values decreased with no significant differences from values observed in day 1 and 5. S. aureus UT 955 showed an increasing trend of internalization (Fig. 2B). After reaching maximum values at day 2, internalized S. aureus decreased markedly reaching approximately the same values as observed for S. uberis UT888 and UT366 at day 5. When numbers of internalized bacteria were expressed as percentage of the multiplicity of infection (MOI) (i.e. for MOI of 10, six internalized bacteria represent 60%) S. aureus UT955 showed the highest values as compared with S. uberis strains. Nevertheless, S. uberis UT888 values at 96 h were higher than S. uberis UT366 and S. aureus UT955 (Fig. 3). Fig. 4A–C shows fluorescent-labeled S. uberis UT888, and UT366 at 96 h of coculture, and S. aureus UT 955 at 72 h, respectively. Internalization of S. aureus UT955 into MAC-T
B. Tamilselvam et al. / Microbial Pathogenesis 40 (2006) 279–285
cells caused severe damage that lead to sloughing and death of host cells (data not shown). In contrast, MAC-T cells did not show apparent damage when internalized by S. uberis UT888 or UT366. The fluorescent staining of MAC-T cells, likely to be originated from bleeding of fluorescent nucleic acid bacterial stains (i.e. fluorescent SYTO 9 and red-fluorescent propidium iodide) showed minimal to no nuclear damage in S. uberis internalized host cells. In contrast, extensive nuclear damage as evidenced by red–orange fluorescent staining was observed in bovine mammary epithelial cells internalized by S. aureus UT955 (Fig. 4D). Additional results from fluorescent-labeled S. uberis MAC-T cells cocultures showed that both S. uberis strains maintained intracellular viability at each sampling point. When numbers of intracellular viable bacteria were calculated, a similar trend as described for the first hour of internalization (Fig. 1A) was obtained. At 2 h of incubation numbers of internalized S. uberis UT366 were higher than S. uberis UT888, but the opposite was observed at the remaining sampling times. Among all bacteria tested, S. aureus showed the highest internalization and viability, except at 96 h where numbers of internalized S. uberis UT888 were the highest. No viable internalized E. coli D5a was detected at any time. Taken together, results from these assays show the ability of both S. uberis strains tested to survive Average internalized bacteria per cell
A
281
intracellularly for extended periods inducing minimal damage to the host cells. 3. Discussion Survival of pathogenic microorganisms has depended on the evolution of strategies for evasion of host defenses. Associated with this evolution is the expression of a variety of virulence determinants that favor persistence of bacteria in the face of a massive inflammatory cell infiltration. It has been unequivocally demonstrated that many pathogens considered extr acellular are capable of triggering a self-uptake signal leading to their internalization into non-phagocytic host cell with limited loss of pathogen’s viability. Previous studies conducted in our laboratory demonstrated the ability of S. uberis to attach and internalize into bovine mammary epithelial cells exploiting host cells signal transduction pathways and cytoskeleton [4,12]. Data from the present investigation provide new insights into post-internalization events of S. uberis into bovine mammary epithelial cells. Result of this study not only confirmed the previously described ability of S. uberis strains to internalize bovine mammary epithelial cells [4], but also that both S. uberis strains tested were capable of maintaining a low level of intracellular presence. Results also
70 60 50 40 30 20 10 0
2
24
48 Time (hours)
72
96
Percentage multiplicity of infection
B 200
150
100
50
0
2
24
48 Time (hours)
72
96
Fig. 3. Internalization of S. uberis and S. aureus into bovine mammary epithelial cells. Fluorescent-labeled (Live/DeadwBacLight Bacterial Viability Kit) (Molecular Probes, Inc., Eugene, OR) S. uberis UT888 ( ), S. uberis UT366 (Q) or S. aureus UT 955 (C) were cocultured with bovine mammary epithelial cell and viable intracellular bacteria (green) counted. Data are expressed as the average of bacteria per cell (A, 250 cells in five random fields) or as a percentage of theoretical multiplicity of infection (S. aureus MOIZ38, S. uberis UT366 MOIZ38, S. uberis UT888 MOIZ23). Horizontal checkered bar in B indicates theoretical MOI and considered as 100%. Bars represent the standard error of the mean of three independent experiments run in triplicate.
282
B. Tamilselvam et al. / Microbial Pathogenesis 40 (2006) 279–285
Fig. 4. Micrograph of bovine mammary epithelial cells internalized by fluorescent-labeled bacteria. Fluorescent-labeled (Live/DeadwBacLight Bacterial Viability Kit) (Molecular Probes, Inc., Eugene, OR) S. uberis UT888 (A), S. uberis UT366 (B) or S. aureus UT 955 (C) and (D) were cocultured with bovine mammary epithelial cell and analyzed by confocal laser microscopy. Panels A and B are S. uberis UT888 and 366 at 96 h of coculture. Panel C and D are S. aureus UT 955 at 72 h of coculture. Notice abundance of dead cells (red–orange nucleus) in monolayers cocultured with S. aureus UT 955. Arrows indicate intracellular bacteria and the lack of them in S. uberis cocultures.
showed that different from S. aureus, S. uberis did not cause apparent extensive damage and cell destruction. When internalization and persistence patterns of S. uberis stains were compared, differences were noticed. Differences in the internalization profile between S. uberis strains were initially observed after 1 h of coculture and showed that internalization values of S. uberis UT366 were 2.6 times higher than UT888, suggesting that UT366 induced a better uptake signal than UT888. Over time, a gradual loss of intracellular viability was observed, and at 8 h of coculture UT366 internalization values were lower than that for UT888. These data suggest that a progressive UT366 losts of viability likely due to host cell intracellular killing mechanisms occurred. These putative intracellular killing mechanisms seem not to affect UT888 or even UT366 after a low level of internalization was reached. When internalization values were analyzed, strain differences in favor of UT888 were observed and maximum internalization values were detected at 48 and 72 h of coculture. We were unable to definitively confirm if increased internalization values observed at 48 and 72 h were due to the single or combined effect of intracellular growth and/or decreased
intracellular killing. However, when internalized data were analyzed as percentage of theoretical MOI, it was evident that a net intracellular growth was detected for S. aureus. It could be speculated that the increased internalization values observed with S. uberis were due to intracellular replication of S. uberis as previously described [14], or due to extracellular bacteria released from decaying or dead cells that multiplied extracellularly and re-internalize into host cells. However, the presence of antibiotics as well as the lack of bacterial growth in coculture supernatants reject this later hypothesis and support the theory that S. uberis intracellular growth could explain the increased internalization values observed. Nevertheless, data obtained indicate that both S. uberis strains survived well inside host cells and that it is likely that mechanisms leading to evasion of host-cell killing mechanisms were employed by both S. uberis strains tested. Concerning strain differences, it is important to indicate that S. uberis UT366 typically produce hyper-acute IMI with severe local and generalized symptoms [13], whereas S. uberis UT888 produce less severe acute IMI which, typically lead to chronic and persistent infections [14]. Therefore, it could be that the
B. Tamilselvam et al. / Microbial Pathogenesis 40 (2006) 279–285
ability of S. uberis UT888 to induce chronic and persistent infections could be linked to the capability to persist intracellularly for extended periods. Results from fluorescent assays confirmed the results obtained from classical internalization assays and showed the ability of S. uberis strains tested to survive intracellularly. It also confirmed, although not through a specifically designed assay, the ability of S. uberis to survive without affecting host cells viability. This differed from S. aureus, which caused severe damage and cell death and ultimately self-limiting their chances of intracellular persistence as noticed from day 2 to the end of the experiment. An important conclusion of this investigation is that S. uberis seems to have evolved mechanisms to survive intracellularly while maintaining a low level of viable microorganisms that likely serve as a reservoir for persistent infections. Future research should define these mechanisms and confirm if these in vitro observations could be linked to characteristics of natural S. uberis IMI such as persistent infections In conclusion, the results of this study showed the survival of S. uberis strains in MAC-T cells for extended time without causing apparent cell damage or death. Current findings will help to understand the pathogenesis of S. uberis IMI in dairy cows. 4. Materials and methods 4.1. Bacterial strains Two strains of S. uberis (UT888 and UT366) originally isolated from dairy cows with mastitis were used. S. uberis UT888 was originally isolated from a cow with chronic mastitis; this strain has been extensively used in our laboratory to induce mild clinical mastitis during experimental challenge exposure studies. S. uberis UT366 is a highly virulent strain isolated from a cow with clinical acute mastitis previously used for experimental induction of bovine intramammary infections [13]. These strains were characterized extensively by PCRbased DNA fingerprinting and biochemical analysis [15,16]. S. aureus UT955 was isolated from a dairy cow with clinical mastitis and used as positive control. The non-pathogenic strain of E. coli DH5a (Gibco, BRL, Bethesda, MD, USA) was used as a negative control. 4.2. Preparation of bacterial cultures S. uberis UT888 and UT366, S. aureus UT955 and E. coli DH5a stored at K70 8C in 10% skim milk were thawed in a 37 8C water bath, and 10 ml of each bacterial suspension was plated onto trypticase soy agar plate supplemented with 5% defibrinated sheep blood (BAP, Becton Dickinson and Company, Franklin Lakes, NJ), and incubated overnight at 37 8C. After incubation, bacterial lawns were harvested and S. uberis and S. aureus strains were resuspended in 20 ml BactoTodd Hewitt broth (THB, Becton Dickinson Co., Sparks, MD) whereas E. coli DH5a was subcultured in 20 ml Luria
283
broth and all the cultures incubated with shaking (150 rpm) for 2 h at 37 8C. Bacterial suspensions were then washed three times by centrifugation (2500g, 15 min at 4 8C) in phosphate buffer saline (PBS, pH 7.4), resuspended to original volume in PBS (pH 7.4) and diluted 1:100 in Dulbecco’s Modified Eagle’s (DMEM, Gibco, Grand Island, NY) at an approximately bacterial concentration of w107 colony forming units per milliliter (cfu/ml). 4.3. Mammary epithelial cells and culture conditions A bovine mammary epithelial cell line (MAC-T) described by Huynh et al. [18] was used. MAC-T cells were cultured in T75 cell culture flasks using cell growth media (CGM) described previously [12]. MAC-T cells monolayers were harvested by trypsinzation, washed by centrifugation (5 min at 1000g) with PBS (pH 7.4) transferred into Petri dishes containing coverslips and incubated at 378C in 5% CO2:95% air (vol/vol). For fluorescence microscopy studies, MAC-T cells were gown to confluence on four-well polystyrene tissue culture LabTek chambers (Nunc, Inc., Naperville, IL). 4.4. Coverslips preparation A modification of the protocol described by Jones and Portnoy [17] was used. Briefly, round glass coverslips (1! 15 mm) stored in 70% ethanol were dipped in 95% (vol/vol) ethanol. Excessive alcohol was blot off using tissue paper and quickly flamed. Eight sterile coverslips were then placed in a sterile Petri dish (60!20) mm. Coverslip overlapping was avoided to ensure uniform monolayer formation. 4.5. Bacterial internalization assay After removing supernatant, bacterial inoculum (w107 cfu/ml in GCM) was added to Petri dishes and incubated for at 37 8C, ensuring that coverslips were not disturbed. At regular time intervals (1, 2, 4, 6 and 8 h) three coverslips per Petri dish were removed and transferred into 50 ml conical tubes containing 5 ml of sterile PBS solution for gentle washing. One milliliter of CGM containing penicillin (5 mg/ml, Sigma Chemical Co.) and gentamicin (100 mg/ml, Sigma Chemical Co.) were added to monolayers cocultured with S. uberis and E. coli or 1 ml of CGM containing lysostaphin (1 mg/ml, Sigma Chemical Co.) was added to S. aureus cocultured monolayers. Coverslips were incubated for 2 h at 37 8C and after removing CGM with antibiotics, monolayers were washed three times with PBS and lysed with a mixture containing trypsin (0.25%, Sigma Chemical Co.) and Triton X-100 (0.01%, Amersham, Arlington Heights, IL, USA). Lysates were 10-fold serially diluted, plated onto BAP or Luria plates and incubated overnight at 37 8C and colony forming units per milliliter determined. Colony forming units per milliliter in, lysates, monolayers supernatants and bacterial inoculum were determined by standard colony counting techniques. Effectiveness of antibiotic treatment in killing extracellular
284
B. Tamilselvam et al. / Microbial Pathogenesis 40 (2006) 279–285
bacteria was assessed by culturing co-culture supernatants on BAP. Experiments were conducted in triplicate and repeated three times. Multiplicity of infection for all the internalization assays described were 38, 38, and 23 for S. aureus UT955, S. uberis UT366, S. uberis UT888, respectively. 4.6. Long-term survival of intracellular bacteria To determine the internalization and intracellular survival of S. uberis in bovine mammary epithelial cells, two methods were employed. The first method was conducted following the internalization protocol described by Almeida et al. [12] with modifications. Briefly, mammary epithelial cells were cocultured with bacterial inoculum for 60 min, supernatants were removed by aspiration, and monolayers were washed three times with PBS (pH 7.4) as described above. Monolayers were then incubated for 120 h with CGM containing penicillin (5 mg/ml, Sigma Chemical Co.) and Gentamicin (100 mg/ml, Sigma Chemical Co.) Since Gentamicin may slowly penetrate the eukaryotic membrane [18], its concentration was decreased fourfold after initial 2 h of incubation. For S. aureus cocultures, lysostaphin (1 mg/ml in GCM, Sigma Chemical Co.) was used. Monolayers were washed three times with PBS (pH 7.4) lysed with trypsin (0.01% in PBS, Sigma Chemical Co.) and Triton X-100 (0.5% in sterile H2O, Sigma Chemical Co.) and colony forming units per milliliter of internalized bacteria in cell lysates were determined as described above. The second method used followed the protocol described by Barker et al. [19]. To determine the intracellular survival of S. uberis in bovine mammary cells, the Live/DeadwBacLight Bacterial Viability Kit (Molecular Probes, Inc., Eugene, OR) was used. Live/DeadwBacLight Bacterial Viability Kit has mixtures of nucleic acid stains green fluorescent SYTO 9 and red-fluorescent propidium iodide which differ in spectral characteristics and in their ability to penetrate healthy bacterial cells. SYTO 9 stain generally labels all bacteria with intact membranes and those with damaged membranes while propidium iodide penetrates only bacteria with damaged membranes, causing a reduction in the SYTO 9 fluorescence when both dyes are present. For bacterial staining, PBS washed bacterial suspensions (w108 cfu/ml) were microfuged at 14,000g for 5 min and resuspended in sterile saline solution. Bacterial cells were then labeled with fluorescent dye and incubated for 30 min at 4 8C. After incubation, pre-labeled bacteria were diluted in DMEM to w107 cfu/ml and cocultured with MAC-T cell cultures for 2 h at 37 8C in 5% CO2: 95% air (vol/vol). After incubation, monolayers were washed three times with PBS (pH 7.4), treated with antibiotics as described before and incubated for 96 h at 37 8C in 5% CO2: 95% air (vol/vol). At each sampling time, chambers were removed from the slides and monolayers washed three times with PBS (pH 7.4). Slides were then mounted with cover slides, sealed with nail polish, and observed
under confocal epifluorescence microscope (Leica TCS SP2, Leica Microsystems Heidelberg GmbH, Mannheim, Germany) and analyzed using Leica Lite software (Leica Microsystems Heidelberg GmbH). Further analysis of these images was conducted using Image J software (www.rsb.info.nih.gov/ij/). 4.7. Statistics Internalization and intracellular survival assays were performed three times with each condition in triplicate. Means from each experiment were analyzed by ANOVA and those showing statistically significant differences (P%0.01) were further analyzed by Student’s t-test using ProStat (Poly Software International, Salt Lake City, UT) statistical software. Acknowledgements This project was supported by the National Research Initiative of the USDA Cooperative State Research, Education and Extension Service, grant number 2004-35204-14739. References [1] Kossaibati MA, Esslemont RJ. The costs of production diseases in dairy herds in England. Vet J 1997;154:649–53. [2] Oliver SP. Frequency of isolation of environmental mastitis-causing pathogen and incidence of new intramammary infections during the nonlactating period. Am J Vet Res 1998;49:1789–93. [3] Todhunter DA, Smith KL, Hogan JS. Environmental streptococcal intramammary infections of the bovine mammary gland. J Dairy Sci 1995;78:2366–74. [4] Matthews KR, Almeida RA, Oliver SP. Bovine mammary epithelial cell invasion by Streptococcus uberis. Infect Immun 1994;62:5641–6. [5] Almeida RA, Oliver SP. Invasion of bovine mammary epithelial cells by Streptococcus dysgalactiae. J Dairy Sci 1995;78:1310–7. [6] Almeida RA, Matthews KR, Cifrian E, Guidry AJ, Oliver SP. Staphylococcus aureus invasion of bovine mammary epithelial cells. J Dairy Sci 1996;70:1021–6. [7] Dopfer D, Almeida RA, Lam TJGM, Nederbragt H, Oliver SP, Gaastra W. Adhesion and invasion of Escherichia coli from recurrent clinical cases of bovine mastitis. J Vet Microbiol 2000;74:331–43. [8] Matthews KR, Luther DA, Oliver SP. Protein expression by Strept ococcus uberis in the presence of bovine mammary epithelial cells. J Dairy Sci 1994;77:3490. [9] Gilbert FB, Luther DA, Oliver SP. Induction of surface-associated proteins of Streptococcus uberis by cultivation with extracellular matrix components and bovine mammary epithelial cells. FEMS Microbiol Lett 1997;156:161–4. [10] Almeida RA, Calvinho LF, Oliver SP. Influence of protein kinase inhibitors on Streptococcus uberis internalization into bovine mammary epithelial cells. Microb Pathog 2000;28:9–16. [11] Almeida RA, Oliver SP. Role of collagen in adherence of Streptococcus uberis to bovine mammary epithelial cells. J Vet Med Ser B 2001;48: 759–63. [12] Almeida RA, Luther DA, Kumar SJ, Calvinho LF, Bronze MS, Oliver SP. Adherence of Streptococcus uberis to bovine mammary epithelial cells and to extracellular matrix proteins. J Vet Med Ser B 1996;43:385–92.
B. Tamilselvam et al. / Microbial Pathogenesis 40 (2006) 279–285 [13] Doane RM, Oliver SP, Walker RD, Shull EP. Experimental infection of lactating bovine mammary glands with Streptococcus uberis in quarters colonized by Corynebacterium bovis. Am J Vet Res 1987;48: 749–54. [14] Oliver SP, Gillespie BE, Jayarao BM. Detection of new and persistent Streptococcus uberis and Streptococcus dysgalactiae intramammary infections by polymerase chain reaction-based DNA fingerprinting. FEMS Microbiol Lett 1998;160:69–73. [15] Jayarao BM, Bassam BJ, Caetano-Anolles G, Gresshoff PM, Oliver SP. Subtyping of Streptococcus uberis by DNA amplification fingerprinting. J Clin Microbiol 1992;30:1347–50.
285
[16] Jayarao BM, Oliver SP. Polymerase chain reaction-based DNA fingerprinting for identification of Streptococcus and Enterococcus species isolated from bovine milk. J Food Prot 1994;57:240–8. [17] Jones S, Portnoy DA. Intracellular growth of bacteria. Methods Enzymol 1994;256:463–7. [18] Huynh HT, Robitaille G, Turner JD. Establishment of bovine mammary epithelial cells (MAC-T): an in vitro model for bovine lactation. Exp Cell Res 1991;197:191–9. [19] Barker LP, George KM, Falkow S, Small PLC. Differential trafficking of live and dead Mycobacterium marinum organisms in macrophages. Infect Immun 1997;65:1497–504.
Streptococcus uberis internalizes and persists in bovine mammary epithelial cells Batcha Tamilselvam a, Rau´l A. Almeida a,*, John R. Dunlap b, Stephen P. Oliver a a
Department of Animal Science, Institute of Agriculture, Food Safety Center of Excellence, The University of Tennessee, 60 McCord Hall, Knoxville, TN 37996, USA b Department of Botany, The University of Tennessee, Knoxville, TN 37996, USA Received 18 July 2005; received in revised form 3 January 2006; accepted 28 February 2006
Abstract Streptococcus uberis is one of the most important emerging bovine mastitis pathogens and chronic persistent intramammary infections (IMI) are often described. To define the ability of S. uberis to persist intracellularly, studies on time-dependent internalization and survival of S. uberis strains in bovine mammary epithelial cells were conducted. Two S. uberis strains (UT366 and UT888) and a Staphylococcus aureus strain used as positive control, all isolated from cows with clinical mastitis were cocultured with bovine mammary epithelial cells (MAC-T) and persistent survival in host epithelial cells for extended periods (120 h) studied. Of S. uberis strains tested, UT366 showed highest internalization values at 60 min of incubation whereas at 8 h of incubation the corresponding values for UT888 were the highest. Of both strains of S. uberis tested, UT366 seems to internalize bovine mammary cells more efficiently initially, however, during the first 8 h, UT888 seems to survive intracellularly better than UT366. Results showed that both S. uberis strains could survive intracellularly up to 120 h without apparent loss of host cells viability. S. aureus internalized more efficiently than all strains tested and host cell death was observed after 72 h of incubation. These results indicate that S. uberis can survive within mammary epithelial cells for extended time without apparent loss of host cells viability. Intracellular persistence of S. uberis may be associated with the spread of the infection to deeper tissues and development of persistent IMI. q 2006 Elsevier Ltd. All rights reserved. Keywords: Intracellular persistence; Streptococcus uberis; Bovine mastitis; Pathogenesis
1. Introduction Mastitis remains economically the most important disease of dairy cattle throughout the world, accounting 38% of the direct costs of the common production diseases [1]. Streptococcus uberis is an environmental pathogen responsible for a high proportion of cases of clinical and subclinical mastitis in lactating cows. S. uberis is the predominant organism isolated from mammary glands during the non-lactating period accounting for approximately 90% total cases of environmental streptococcal mastitis in heifers, within the first 5 days of lactation [2,3]. Bovine intramammary infections (IMI) caused by S. uberis could progress with clinical or subclinical symptoms and very often, acute IMI lead to chronic infections that can persist in the infected mammary gland for more than one lactation [2]. * Corresponding author. Tel.: C1 865 974 0991; fax: C1 865 974 3394. E-mail address: [email protected] (R.A. Almeida).
0882-4010/$ - see front matter q 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.micpath.2006.02.006
Internalization of bovine mastitis causing pathogens into host cells including S. uberis has been previously described [4–7]. An intracellular microenvironment provides these pathogens with protection from the immune system and non-specific antibacterial factors present in bovine milk. Earlier work from our laboratory showed that for internalization S. uberis exploits host cell cytoskeleton and signal transduction, as well as expresses de novo proteins [8–11]. Although information is available on adherence to and internalization of S. uberis into bovine mammary epithelial cells, little is known about postinternalization events. Therefore, the objective of this study was to further characterize post-internalization period focusing on time-dependent internalization and fate of intracellular S. uberis using classical internalization and fluorescent assays.
2. Results 2.1. Internalization of S. uberis into MAC-T cells Results from the first hour of coculture showed that S. uberis UT366 internalize significantly more efficiently than S. uberis
B. Tamilselvam et al. / Microbial Pathogenesis 40 (2006) 279–285
2.2. Intracellular persistence of S. uberis in MAC-T cells In order to assess the survival ability of internalized S. uberis strains over a 6 day period, MAC-T cells were cocultured with S. uberis strains (UT888 and UT366) for 1 h, washed and treated with antibiotic solution for 2 h and then incubated with 0.25! diluted antibiotic solution for 120 h. CFU/ml internalized S. uberis
A 5.E+04 UT888 UT366
4.E+04 3.E+04 2.E+04 1.E+04 0.E+00
1
4
6
8
Time (hours) Log10 cfu/ml internalized bacteria
B 100000000 10000000 UT888 UT366 UT955
1000000 100000 10000 1000
1
4
6
8
Time (hours)
Fig. 1. Internalization of Streptococcus uberis into bovine mammary epithelial cells. S. uberis UT888 ( , :) S. uberis UT366 (Q, &) and Staphylococcus aureus UT 955 (C) were cocultured with a bovine mammary epithelial cell line (MAC-T cells) and colony forming units per milliliter (cfu/ml) of internalized bacteria calculated at 1, 4, 6 and 8 h. Data are presented as colony forming units per milliliter (A) or log10 cfu/ml (B) of internalized bacteria are the mean of three experiments run in triplicate. Error bars in panel A represent the standard error of the mean.
A cfu/ml internalized S. uberis
UT888 (PR0.005). Although results from the first 6 h showed that S. uberis UT366 internalized better, results obtained at 8 h of coculture suggest that S. uberis UT888 survived intracellularly better than UT366 (Fig. 1(A)). Contrary to the steady loss of intracellular viability observed in S. uberis UT366, UT888 intracellular colony forming units per milliliter were constant during the assay. The progressive decline of S. uberis UT366 viability was marked enough that at 8 h of incubation the percentage of intracellular S. uberis UT366 was lower than that of S. uberis UT888 reaching values 66% lower than that observed at 1 h of coculture (Fig. 1). When internalization values of the controls were compared with S. uberis, Staphylococcus aureus UT955 showed overall highest internalization values (Fig. 2). Maximum internalization for this pathogen was at 6 h of incubation and were approximately 2.5 log10 higher than those observed at 1 h of coculture, and 4.5 log higher than the corresponding for S. uberis UT888 and UT366 at the same sampling point (6 h). Cultures of supernatant showed no bacterial growth (data not shown) suggesting that extracellular bacteria were killed by the antibiotic treatment. No internalization of Escherichia coli DH5a into MAC-T cell was detected at any sampling time.
5.E+05 4.E+05 3.E+05 2.E+05 1.E+05 0.E+00
2
3 Time (days)
4
5
3
4
5
B Log10 cfu/ml internalized bacteria
280
1.E+08 1.E+07 1.E+06 1.E+05 1.E+04 1.E+03 1
2
Time (days)
Fig. 2. Intracellular survival of S. uberis into bovine mammary epithelial cells. S. uberis UT888 ( , :) S. uberis UT366 (Q, &) and S. aureus UT 955 (C) were cocultured with a bovine mammary epithelial cell line (MAC-T cells) and colony forming units per milliliter (cfu/ml) of internalized bacteria calculated at 1, 2, 3, 4 and 5 days. Data are presented as colony forming units per milliliter (A) or log10 cfu/ml (B) of internalized bacteria and are the mean of three experiments run in triplicate. Error bars in panel A represent the standard error of the mean.
Cultures of supernatant showed no bacterial growth (data not shown) suggesting that extracellular bacterial were killed by the antibiotic treatment. Colony forming units per milliliter of internalized S. uberis recovered after 24 h of incubation was slightly higher (UT888, PZ0.26; UT366, PZ0.04) than the recovered at 8 h of coculture (Fig. 2). After the initial 24 h of incubation, internalization of host cell was significantly higher for UT888 than for UT366 (PZ0.02) and such strain effect was observed during the remaining of the assay. Results presented in Fig. 2A show that colony forming units per milliliter of internalized S. uberis increased 1.35 and 1.43 log10 for UT888 and 1.63 and 1.72 log10 for UT366 at days 2 and 3, respectively, compared with values observed on day 1. From day 4 and to the end of the experiment, S. uberis UT888 and UT366 colony forming units per milliliter values decreased with no significant differences from values observed in day 1 and 5. S. aureus UT 955 showed an increasing trend of internalization (Fig. 2B). After reaching maximum values at day 2, internalized S. aureus decreased markedly reaching approximately the same values as observed for S. uberis UT888 and UT366 at day 5. When numbers of internalized bacteria were expressed as percentage of the multiplicity of infection (MOI) (i.e. for MOI of 10, six internalized bacteria represent 60%) S. aureus UT955 showed the highest values as compared with S. uberis strains. Nevertheless, S. uberis UT888 values at 96 h were higher than S. uberis UT366 and S. aureus UT955 (Fig. 3). Fig. 4A–C shows fluorescent-labeled S. uberis UT888, and UT366 at 96 h of coculture, and S. aureus UT 955 at 72 h, respectively. Internalization of S. aureus UT955 into MAC-T
B. Tamilselvam et al. / Microbial Pathogenesis 40 (2006) 279–285
cells caused severe damage that lead to sloughing and death of host cells (data not shown). In contrast, MAC-T cells did not show apparent damage when internalized by S. uberis UT888 or UT366. The fluorescent staining of MAC-T cells, likely to be originated from bleeding of fluorescent nucleic acid bacterial stains (i.e. fluorescent SYTO 9 and red-fluorescent propidium iodide) showed minimal to no nuclear damage in S. uberis internalized host cells. In contrast, extensive nuclear damage as evidenced by red–orange fluorescent staining was observed in bovine mammary epithelial cells internalized by S. aureus UT955 (Fig. 4D). Additional results from fluorescent-labeled S. uberis MAC-T cells cocultures showed that both S. uberis strains maintained intracellular viability at each sampling point. When numbers of intracellular viable bacteria were calculated, a similar trend as described for the first hour of internalization (Fig. 1A) was obtained. At 2 h of incubation numbers of internalized S. uberis UT366 were higher than S. uberis UT888, but the opposite was observed at the remaining sampling times. Among all bacteria tested, S. aureus showed the highest internalization and viability, except at 96 h where numbers of internalized S. uberis UT888 were the highest. No viable internalized E. coli D5a was detected at any time. Taken together, results from these assays show the ability of both S. uberis strains tested to survive Average internalized bacteria per cell
A
281
intracellularly for extended periods inducing minimal damage to the host cells. 3. Discussion Survival of pathogenic microorganisms has depended on the evolution of strategies for evasion of host defenses. Associated with this evolution is the expression of a variety of virulence determinants that favor persistence of bacteria in the face of a massive inflammatory cell infiltration. It has been unequivocally demonstrated that many pathogens considered extr acellular are capable of triggering a self-uptake signal leading to their internalization into non-phagocytic host cell with limited loss of pathogen’s viability. Previous studies conducted in our laboratory demonstrated the ability of S. uberis to attach and internalize into bovine mammary epithelial cells exploiting host cells signal transduction pathways and cytoskeleton [4,12]. Data from the present investigation provide new insights into post-internalization events of S. uberis into bovine mammary epithelial cells. Result of this study not only confirmed the previously described ability of S. uberis strains to internalize bovine mammary epithelial cells [4], but also that both S. uberis strains tested were capable of maintaining a low level of intracellular presence. Results also
70 60 50 40 30 20 10 0
2
24
48 Time (hours)
72
96
Percentage multiplicity of infection
B 200
150
100
50
0
2
24
48 Time (hours)
72
96
Fig. 3. Internalization of S. uberis and S. aureus into bovine mammary epithelial cells. Fluorescent-labeled (Live/DeadwBacLight Bacterial Viability Kit) (Molecular Probes, Inc., Eugene, OR) S. uberis UT888 ( ), S. uberis UT366 (Q) or S. aureus UT 955 (C) were cocultured with bovine mammary epithelial cell and viable intracellular bacteria (green) counted. Data are expressed as the average of bacteria per cell (A, 250 cells in five random fields) or as a percentage of theoretical multiplicity of infection (S. aureus MOIZ38, S. uberis UT366 MOIZ38, S. uberis UT888 MOIZ23). Horizontal checkered bar in B indicates theoretical MOI and considered as 100%. Bars represent the standard error of the mean of three independent experiments run in triplicate.
282
B. Tamilselvam et al. / Microbial Pathogenesis 40 (2006) 279–285
Fig. 4. Micrograph of bovine mammary epithelial cells internalized by fluorescent-labeled bacteria. Fluorescent-labeled (Live/DeadwBacLight Bacterial Viability Kit) (Molecular Probes, Inc., Eugene, OR) S. uberis UT888 (A), S. uberis UT366 (B) or S. aureus UT 955 (C) and (D) were cocultured with bovine mammary epithelial cell and analyzed by confocal laser microscopy. Panels A and B are S. uberis UT888 and 366 at 96 h of coculture. Panel C and D are S. aureus UT 955 at 72 h of coculture. Notice abundance of dead cells (red–orange nucleus) in monolayers cocultured with S. aureus UT 955. Arrows indicate intracellular bacteria and the lack of them in S. uberis cocultures.
showed that different from S. aureus, S. uberis did not cause apparent extensive damage and cell destruction. When internalization and persistence patterns of S. uberis stains were compared, differences were noticed. Differences in the internalization profile between S. uberis strains were initially observed after 1 h of coculture and showed that internalization values of S. uberis UT366 were 2.6 times higher than UT888, suggesting that UT366 induced a better uptake signal than UT888. Over time, a gradual loss of intracellular viability was observed, and at 8 h of coculture UT366 internalization values were lower than that for UT888. These data suggest that a progressive UT366 losts of viability likely due to host cell intracellular killing mechanisms occurred. These putative intracellular killing mechanisms seem not to affect UT888 or even UT366 after a low level of internalization was reached. When internalization values were analyzed, strain differences in favor of UT888 were observed and maximum internalization values were detected at 48 and 72 h of coculture. We were unable to definitively confirm if increased internalization values observed at 48 and 72 h were due to the single or combined effect of intracellular growth and/or decreased
intracellular killing. However, when internalized data were analyzed as percentage of theoretical MOI, it was evident that a net intracellular growth was detected for S. aureus. It could be speculated that the increased internalization values observed with S. uberis were due to intracellular replication of S. uberis as previously described [14], or due to extracellular bacteria released from decaying or dead cells that multiplied extracellularly and re-internalize into host cells. However, the presence of antibiotics as well as the lack of bacterial growth in coculture supernatants reject this later hypothesis and support the theory that S. uberis intracellular growth could explain the increased internalization values observed. Nevertheless, data obtained indicate that both S. uberis strains survived well inside host cells and that it is likely that mechanisms leading to evasion of host-cell killing mechanisms were employed by both S. uberis strains tested. Concerning strain differences, it is important to indicate that S. uberis UT366 typically produce hyper-acute IMI with severe local and generalized symptoms [13], whereas S. uberis UT888 produce less severe acute IMI which, typically lead to chronic and persistent infections [14]. Therefore, it could be that the
B. Tamilselvam et al. / Microbial Pathogenesis 40 (2006) 279–285
ability of S. uberis UT888 to induce chronic and persistent infections could be linked to the capability to persist intracellularly for extended periods. Results from fluorescent assays confirmed the results obtained from classical internalization assays and showed the ability of S. uberis strains tested to survive intracellularly. It also confirmed, although not through a specifically designed assay, the ability of S. uberis to survive without affecting host cells viability. This differed from S. aureus, which caused severe damage and cell death and ultimately self-limiting their chances of intracellular persistence as noticed from day 2 to the end of the experiment. An important conclusion of this investigation is that S. uberis seems to have evolved mechanisms to survive intracellularly while maintaining a low level of viable microorganisms that likely serve as a reservoir for persistent infections. Future research should define these mechanisms and confirm if these in vitro observations could be linked to characteristics of natural S. uberis IMI such as persistent infections In conclusion, the results of this study showed the survival of S. uberis strains in MAC-T cells for extended time without causing apparent cell damage or death. Current findings will help to understand the pathogenesis of S. uberis IMI in dairy cows. 4. Materials and methods 4.1. Bacterial strains Two strains of S. uberis (UT888 and UT366) originally isolated from dairy cows with mastitis were used. S. uberis UT888 was originally isolated from a cow with chronic mastitis; this strain has been extensively used in our laboratory to induce mild clinical mastitis during experimental challenge exposure studies. S. uberis UT366 is a highly virulent strain isolated from a cow with clinical acute mastitis previously used for experimental induction of bovine intramammary infections [13]. These strains were characterized extensively by PCRbased DNA fingerprinting and biochemical analysis [15,16]. S. aureus UT955 was isolated from a dairy cow with clinical mastitis and used as positive control. The non-pathogenic strain of E. coli DH5a (Gibco, BRL, Bethesda, MD, USA) was used as a negative control. 4.2. Preparation of bacterial cultures S. uberis UT888 and UT366, S. aureus UT955 and E. coli DH5a stored at K70 8C in 10% skim milk were thawed in a 37 8C water bath, and 10 ml of each bacterial suspension was plated onto trypticase soy agar plate supplemented with 5% defibrinated sheep blood (BAP, Becton Dickinson and Company, Franklin Lakes, NJ), and incubated overnight at 37 8C. After incubation, bacterial lawns were harvested and S. uberis and S. aureus strains were resuspended in 20 ml BactoTodd Hewitt broth (THB, Becton Dickinson Co., Sparks, MD) whereas E. coli DH5a was subcultured in 20 ml Luria
283
broth and all the cultures incubated with shaking (150 rpm) for 2 h at 37 8C. Bacterial suspensions were then washed three times by centrifugation (2500g, 15 min at 4 8C) in phosphate buffer saline (PBS, pH 7.4), resuspended to original volume in PBS (pH 7.4) and diluted 1:100 in Dulbecco’s Modified Eagle’s (DMEM, Gibco, Grand Island, NY) at an approximately bacterial concentration of w107 colony forming units per milliliter (cfu/ml). 4.3. Mammary epithelial cells and culture conditions A bovine mammary epithelial cell line (MAC-T) described by Huynh et al. [18] was used. MAC-T cells were cultured in T75 cell culture flasks using cell growth media (CGM) described previously [12]. MAC-T cells monolayers were harvested by trypsinzation, washed by centrifugation (5 min at 1000g) with PBS (pH 7.4) transferred into Petri dishes containing coverslips and incubated at 378C in 5% CO2:95% air (vol/vol). For fluorescence microscopy studies, MAC-T cells were gown to confluence on four-well polystyrene tissue culture LabTek chambers (Nunc, Inc., Naperville, IL). 4.4. Coverslips preparation A modification of the protocol described by Jones and Portnoy [17] was used. Briefly, round glass coverslips (1! 15 mm) stored in 70% ethanol were dipped in 95% (vol/vol) ethanol. Excessive alcohol was blot off using tissue paper and quickly flamed. Eight sterile coverslips were then placed in a sterile Petri dish (60!20) mm. Coverslip overlapping was avoided to ensure uniform monolayer formation. 4.5. Bacterial internalization assay After removing supernatant, bacterial inoculum (w107 cfu/ml in GCM) was added to Petri dishes and incubated for at 37 8C, ensuring that coverslips were not disturbed. At regular time intervals (1, 2, 4, 6 and 8 h) three coverslips per Petri dish were removed and transferred into 50 ml conical tubes containing 5 ml of sterile PBS solution for gentle washing. One milliliter of CGM containing penicillin (5 mg/ml, Sigma Chemical Co.) and gentamicin (100 mg/ml, Sigma Chemical Co.) were added to monolayers cocultured with S. uberis and E. coli or 1 ml of CGM containing lysostaphin (1 mg/ml, Sigma Chemical Co.) was added to S. aureus cocultured monolayers. Coverslips were incubated for 2 h at 37 8C and after removing CGM with antibiotics, monolayers were washed three times with PBS and lysed with a mixture containing trypsin (0.25%, Sigma Chemical Co.) and Triton X-100 (0.01%, Amersham, Arlington Heights, IL, USA). Lysates were 10-fold serially diluted, plated onto BAP or Luria plates and incubated overnight at 37 8C and colony forming units per milliliter determined. Colony forming units per milliliter in, lysates, monolayers supernatants and bacterial inoculum were determined by standard colony counting techniques. Effectiveness of antibiotic treatment in killing extracellular
284
B. Tamilselvam et al. / Microbial Pathogenesis 40 (2006) 279–285
bacteria was assessed by culturing co-culture supernatants on BAP. Experiments were conducted in triplicate and repeated three times. Multiplicity of infection for all the internalization assays described were 38, 38, and 23 for S. aureus UT955, S. uberis UT366, S. uberis UT888, respectively. 4.6. Long-term survival of intracellular bacteria To determine the internalization and intracellular survival of S. uberis in bovine mammary epithelial cells, two methods were employed. The first method was conducted following the internalization protocol described by Almeida et al. [12] with modifications. Briefly, mammary epithelial cells were cocultured with bacterial inoculum for 60 min, supernatants were removed by aspiration, and monolayers were washed three times with PBS (pH 7.4) as described above. Monolayers were then incubated for 120 h with CGM containing penicillin (5 mg/ml, Sigma Chemical Co.) and Gentamicin (100 mg/ml, Sigma Chemical Co.) Since Gentamicin may slowly penetrate the eukaryotic membrane [18], its concentration was decreased fourfold after initial 2 h of incubation. For S. aureus cocultures, lysostaphin (1 mg/ml in GCM, Sigma Chemical Co.) was used. Monolayers were washed three times with PBS (pH 7.4) lysed with trypsin (0.01% in PBS, Sigma Chemical Co.) and Triton X-100 (0.5% in sterile H2O, Sigma Chemical Co.) and colony forming units per milliliter of internalized bacteria in cell lysates were determined as described above. The second method used followed the protocol described by Barker et al. [19]. To determine the intracellular survival of S. uberis in bovine mammary cells, the Live/DeadwBacLight Bacterial Viability Kit (Molecular Probes, Inc., Eugene, OR) was used. Live/DeadwBacLight Bacterial Viability Kit has mixtures of nucleic acid stains green fluorescent SYTO 9 and red-fluorescent propidium iodide which differ in spectral characteristics and in their ability to penetrate healthy bacterial cells. SYTO 9 stain generally labels all bacteria with intact membranes and those with damaged membranes while propidium iodide penetrates only bacteria with damaged membranes, causing a reduction in the SYTO 9 fluorescence when both dyes are present. For bacterial staining, PBS washed bacterial suspensions (w108 cfu/ml) were microfuged at 14,000g for 5 min and resuspended in sterile saline solution. Bacterial cells were then labeled with fluorescent dye and incubated for 30 min at 4 8C. After incubation, pre-labeled bacteria were diluted in DMEM to w107 cfu/ml and cocultured with MAC-T cell cultures for 2 h at 37 8C in 5% CO2: 95% air (vol/vol). After incubation, monolayers were washed three times with PBS (pH 7.4), treated with antibiotics as described before and incubated for 96 h at 37 8C in 5% CO2: 95% air (vol/vol). At each sampling time, chambers were removed from the slides and monolayers washed three times with PBS (pH 7.4). Slides were then mounted with cover slides, sealed with nail polish, and observed
under confocal epifluorescence microscope (Leica TCS SP2, Leica Microsystems Heidelberg GmbH, Mannheim, Germany) and analyzed using Leica Lite software (Leica Microsystems Heidelberg GmbH). Further analysis of these images was conducted using Image J software (www.rsb.info.nih.gov/ij/). 4.7. Statistics Internalization and intracellular survival assays were performed three times with each condition in triplicate. Means from each experiment were analyzed by ANOVA and those showing statistically significant differences (P%0.01) were further analyzed by Student’s t-test using ProStat (Poly Software International, Salt Lake City, UT) statistical software. Acknowledgements This project was supported by the National Research Initiative of the USDA Cooperative State Research, Education and Extension Service, grant number 2004-35204-14739. References [1] Kossaibati MA, Esslemont RJ. The costs of production diseases in dairy herds in England. Vet J 1997;154:649–53. [2] Oliver SP. Frequency of isolation of environmental mastitis-causing pathogen and incidence of new intramammary infections during the nonlactating period. Am J Vet Res 1998;49:1789–93. [3] Todhunter DA, Smith KL, Hogan JS. Environmental streptococcal intramammary infections of the bovine mammary gland. J Dairy Sci 1995;78:2366–74. [4] Matthews KR, Almeida RA, Oliver SP. Bovine mammary epithelial cell invasion by Streptococcus uberis. Infect Immun 1994;62:5641–6. [5] Almeida RA, Oliver SP. Invasion of bovine mammary epithelial cells by Streptococcus dysgalactiae. J Dairy Sci 1995;78:1310–7. [6] Almeida RA, Matthews KR, Cifrian E, Guidry AJ, Oliver SP. Staphylococcus aureus invasion of bovine mammary epithelial cells. J Dairy Sci 1996;70:1021–6. [7] Dopfer D, Almeida RA, Lam TJGM, Nederbragt H, Oliver SP, Gaastra W. Adhesion and invasion of Escherichia coli from recurrent clinical cases of bovine mastitis. J Vet Microbiol 2000;74:331–43. [8] Matthews KR, Luther DA, Oliver SP. Protein expression by Strept ococcus uberis in the presence of bovine mammary epithelial cells. J Dairy Sci 1994;77:3490. [9] Gilbert FB, Luther DA, Oliver SP. Induction of surface-associated proteins of Streptococcus uberis by cultivation with extracellular matrix components and bovine mammary epithelial cells. FEMS Microbiol Lett 1997;156:161–4. [10] Almeida RA, Calvinho LF, Oliver SP. Influence of protein kinase inhibitors on Streptococcus uberis internalization into bovine mammary epithelial cells. Microb Pathog 2000;28:9–16. [11] Almeida RA, Oliver SP. Role of collagen in adherence of Streptococcus uberis to bovine mammary epithelial cells. J Vet Med Ser B 2001;48: 759–63. [12] Almeida RA, Luther DA, Kumar SJ, Calvinho LF, Bronze MS, Oliver SP. Adherence of Streptococcus uberis to bovine mammary epithelial cells and to extracellular matrix proteins. J Vet Med Ser B 1996;43:385–92.
B. Tamilselvam et al. / Microbial Pathogenesis 40 (2006) 279–285 [13] Doane RM, Oliver SP, Walker RD, Shull EP. Experimental infection of lactating bovine mammary glands with Streptococcus uberis in quarters colonized by Corynebacterium bovis. Am J Vet Res 1987;48: 749–54. [14] Oliver SP, Gillespie BE, Jayarao BM. Detection of new and persistent Streptococcus uberis and Streptococcus dysgalactiae intramammary infections by polymerase chain reaction-based DNA fingerprinting. FEMS Microbiol Lett 1998;160:69–73. [15] Jayarao BM, Bassam BJ, Caetano-Anolles G, Gresshoff PM, Oliver SP. Subtyping of Streptococcus uberis by DNA amplification fingerprinting. J Clin Microbiol 1992;30:1347–50.
285
[16] Jayarao BM, Oliver SP. Polymerase chain reaction-based DNA fingerprinting for identification of Streptococcus and Enterococcus species isolated from bovine milk. J Food Prot 1994;57:240–8. [17] Jones S, Portnoy DA. Intracellular growth of bacteria. Methods Enzymol 1994;256:463–7. [18] Huynh HT, Robitaille G, Turner JD. Establishment of bovine mammary epithelial cells (MAC-T): an in vitro model for bovine lactation. Exp Cell Res 1991;197:191–9. [19] Barker LP, George KM, Falkow S, Small PLC. Differential trafficking of live and dead Mycobacterium marinum organisms in macrophages. Infect Immun 1997;65:1497–504.