| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Laboratoire de Sciences Animales, U.C. INRA 12 340, Ecole Nationale Supérieure d’Agronomie et des Industries Alimentaires (ENSAIA), 54 505 Vandoeuvre-lès-Nancy, France
2 Ghent University, Faculty of Veterinary Medicine, Department of Physiology, Biochemistry and Biometrics, Salisburylaan 133, 9820 Merelbeke, Belgium
Corresponding author: F. Moussaoui; e-mail: Fatima.Moussaoui{at}ensaia.inpl-nancy.fr.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Key Words: Escherichia coli mastitis • milk • proteolysis • polymorphonuclear neutrophil
Abbreviation key: PA = plasmin activity, PIH = post-infusion hour, PMN = polymorphonuclear neutrophils, PP = proteose peptones, t-PA = tissue plasminogen activator, u-PA = urokinase-type plasminogen activator
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Previous work has been dedicated to assessing the changes in milk composition, especially proteins, and protease activities, including plasmin and PMN enzymes (Moussaoui et al., 2002, 2003) during LPS experimental mastitis. To study the level of protein breakdown, proteose peptone (PP) fraction was studied both qualitatively and quantitatively, because this is considered to be an indicator of proteolysis (Le Roux et al., 1995; Moussaoui et al., 2002, 2003). The identification of some fragments from PP that were released from CN breakdown underlined the role of PMN in endogenous proteolysis (Moussaoui et al., 2003).
Lipopolysaccharide from Escherichia coli allows clinical mastitis to be imitated, avoiding, however, the involvement of bacteria and making it possible to study endogenous proteolysis. The presence of bacteria, including its endotoxin, would enable a relevant comparison with a model very much closer to reality, as far as both the direct and the nondirect roles of the bacteria are concerned. In addition, E. coli itself contains numerous proteases including metallo-, serine- and aspartic-proteases, capable of degrading CN and, in particular, some abnormal proteins such as oxidized proteins (Rozkov, 2001).
Mastitis is mostly induced by bacteria. The recovered pathogen and the corresponding relative frequency vary from country to country. In addition, E. coli is one of the most important pathogens causing mastitis in dairy cow (Bradley, 2002). The outcome of experimental E. coli mastitis ranges from mild to severe in individual dairy cows and can be quantified by bacterial counts in milk after infection (Lohuis et al., 1990). The clinical effects of the infection can, however, be very severe, leading to death during the periparturient period (Sordillo and Babiuk, 1991). Proteolysis in milk during E. coli mastitis has received very little attention. Very little information is available either on the involvement of the bacteria on CN breakdown or on the intensity of proteolytic machinery during the infection.
Therefore, the aim of this work was to explore milk proteolysis during E. coli experimental mastitis primarily to investigate sequentiality, to precisely define the roles of both endogenous proteases (including plasmin and leukocyte enzymes) and bacterial proteases. The PP protein fraction was used as an indicator of proteolysis. Both PP content and qualitative composition were investigated; PP composition change was compared with CN breakdown by leukocytes. In addition, plasmin activity (PA), SCC, and cfu were observed for several hours after the induction of E. coli mastitis.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONACKNOWLEDGEMENTSREFERENCES |
|---|
For inclusion in the intramammary inoculation trial, no treatment of clinical diseases (mastitis) was allowed within 10 d before the intramammary inoculation. Therefore, only clinically healthy animals, free of major mastitis pathogens on 2 consecutive bacteriological negative examinations, with an SCC <200,000/mL on quarter level, were finally included in the intramammary E. coli challenge.
Mechanical milking was performed twice daily using a quarter milking device (Packo & Fullwood, Zedelgem, Belgium).
Intramammary Inoculation Procedure
A stock of E. coli strain P4:O32 (Bramley, 1976) was maintained in lyophilization medium at –20°C. For experimental use, the bacteria were subcultured in brain-heart infusion broth (CM225; Oxoid, Nepean, ON, Canada) at 37°C for 3 consecutive days. The bacterial suspension was washed 3 times with pyrogen-free PBS, and resuspended in PBS.
Before the challenge, the teat ends were disinfected with 70% ethanol containing 0.5% chlorhexidin. Just before inoculation, the suspension was diluted in pyrogen-free PBS to a final concentration of 1 x 104 cfu/mL (n = 4) or 1 x 106 cfu/mL (n = 4). On d 0, 30 min after morning milking (1.5 h after feeding), the cows were inoculated in the left front and rear quarters with a suspension containing 1 x 104 cfu or 1 x 106 cfu E. coli P4:O32 in a total volume of 10 mL of pyrogen-free saline solution (0.9%) per quarter. After IMI, the bacterial suspension was thoroughly distributed into the udder cistern through a 30-s massage. All bacterial suspensions were infused into the teat cistern using a sterile, pyrogen-free teat cannula (7 cm; Me.Ve.Mat, Deinze, Belgium).
Sampling Procedure
Milk samples were collected before and at postinfusion hours (PIH): –24, 0, 3, 6, 9, 12, 15, 18, 21, 24, 33, 48, 57, 72, 144, and 216.
Foremilk (5 mL) was aseptically collected for bacteriologic examination for major pathogens before inoculation and for quantification of the E. coli population level (cfu) from PIH 3 onward. Furthermore, milk samples (100 mL) were collected for the determination of SCC and the preparation of milk whey through centrifugation. All samples were kept on melting ice (1°C) during transportation.
Clinical Examination
Standard clinical parameters were examined. Rectal temperature, heart rate, respiration rate, rumen motility, fecal appearance, appetite, general behavior, and aspects of the mammary gland (abnormal milk, swelling, teat relaxation, and milk leakage) (Massart-Leën et al., 1988) were recorded by a veterinarian each time blood and milk samples were collected.
SCC and Milk Composition
Somatic cell count was determined using a fluoro-opto electronic method (Fossomatic 400 cell counter; Foss Electrics, Hillerød, Denmark). The PP content was determined as described previously (Le Roux et al., 1995).
Quarter Milk Production
Daily quarter milk production, the yield of the evening and subsequent morning milking, was measured using a quarter milking device (Packo and Fullwood, Zedelgem, Belgium). Quarter milk production, expressed as L/d, was measured at d –4, –1, 0, 1, 2, and 3.
Colony-Forming Units in the Inoculated Quarters
The E. coli population level (cfu/mL) after experimental inoculation was determined by appropriate 10-fold dilutions of the sample in PBS. Ten microliters of these dilutions were plated out on Columbia Sheep Blood agar (Biokar Diagnostics, Beauvois, France). All dilutions were performed in duplicate. Colonies were counted after 24-h incubation at 37°C. The colony count was converted to cfu/mL based on the dilution factor, and the final results were expressed as log10(cfu)/mL.
PA
Plasmin activity was measured by a method based on the release of a yellow compound (measured at 405 nm) when a synthetic plasmin-sensitive substrate (D-Val-Leu-Lys p-nitroanilide dihydrochloride; Sigma) was hydrolyzed after the addition of a dissolving reagent (Linden et al., 1987).
Peptide Characterization
An SDS-PAGE was performed according to the method of Laemmli and Favre (1973) with a 5% (wt/vol) polyacrylamide stacking gel in 0.125 M Tris-HCl (pH 6.8) and with 15% (wt/vol) polyacrylamide separating gel in 0.38 M Tris-HCl (pH 8.8) in the presence of 0.1% (wt/vol) SDS and 5% (vol/vol) 2-mercaptoethanol. Quantities of 60 µg protein for PP and 100 µg for milk per well were loaded into the gel for each sample.
Proteins were fixed in gel using 12% (wt/vol) TCA and stained with 0.1% (wt/vol) Coomassie blue R250 in 50% (vol/vol) ethanol and 10% (vol/vol) acetic acid for at least 2 h. Destaining was performed with a solution of 30% (vol/vol) ethanol and 7.5% (vol/vol) acetic acid. Quantification of the electrophoretic bands was performed by densitometry at 633 nm (Ultrascan XL densitometer; Pharmacia Fine Chemicals, Uppsala, Sweden), and the amount of BSA in milk was deduced from the amount of protein loaded per well in the gel. Results were expressed as a ratio between the BSA amount at time t and the amount before challenge PIH 0. The ratio at PIH 0 was set at 100%.
For microsequencing analysis, proteins were electro-transferred after electrophoresis onto a polyvinylidene difluoride membrane (Immobilon-P; Millipore, Milford, MA) for 5 h at 4°C in 10 mM 3-[cyclohexylamino]-1-propanesulfonic acid buffer (pH 11), containing 10% (vol/vol) methanol (Towbin et al., 1979). Proteins were stained with 0.2% (wt/vol) Ponceau S Red, and the membrane was washed with ultrapure water. After excision, bands of interest were amino-terminal microsequenced by Edman degradation on an automated 476 A protein microsequencer (Applied Biosystems, Foster City, CA).
Statistical Analysis
A mixed model was fitted to evaluate the effect of an IMI on PP, PA, and BSA with time, dose, and their interaction as categorical fixed effects and cow as a random effect. Because the dose and its interaction with time were not significant (see results section), a mixed model with only time as categorical fixed effect and cow as random effect was fitted.
Within this model, PP and PA at each time point were compared with the baseline value at time 0 using Dunnett’s multiple comparisons adjustments technique with an overall type I error equal to 5%.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Milk Protein Compositional Change
Protein from serum, including lactoferrin, BSA, and Ig G, were slightly visible in milk before challenge (Figure 1A
). The 3 proteins were easily visible from PIH 6 after challenge with a maximum at PIH 12 to 15 for BSA and Ig G on the one hand and PIH 24 to 72 for lactoferrin on the other hand. Milk proteins, including CN, ß-lactoglobulin, and 
-lactalbumin, were present during the complete kinetic. A decrease in CN band intensity was, however, noted and particularly at PIH 12 and 15.
|
). It peaked at PIH 15 with 5.3 times the initial level and returned to normal level after PIH 72.
PP content.
For PP, there was no significant difference between the 2 doses (P = 0.6238), but PP did vary over time (P < 0.0001). There was no interaction between the dose and the time (P = 0.7848) so that it can be assumed that PP evolved in a similar way for the 2 doses. The PP content before challenge averaged 0.97 g/L (±0.20), and it differed significantly from this baseline value from PIH 12 (adjusted P = 0.027) onward until PIH 48 (adjusted P = 0.034). It reached its maximum of 6.52 g/L (±0.22) at PIH 33. The average increase was 7.3 x (±1.7). Return to normal values was observed at PIH 216 (PIH 0 vs. PIH 216; P < 0.05; Figure 2).
|
).
|
).
|
).
|
). During the challenge, 2 types of profiles appeared. The first type showed that at PIH 215 additional electrophoretic bands appeared, although not visible before the challenge. These were named E1, P1, P2, E2, and E3. The second type showed that at PIH 24 a single band appeared named P2. The respective molecular masses of the cited bands were 27.0, 17.5, 21.0, 15.5, and 9.0 kDa. Table 1 gives the corresponding sequences with CN fragments for all bands and an additional fragment of BSA for E3.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONACKNOWLEDGEMENTSREFERENCES |
|---|
The increased permeability of the mammary epithelial barrier was associated with inflammation and was supported here by the influx of serum proteins, including lactoferrin, BSA, and IgG, into milk, which is also in accordance with previous work (Hamann and Krömker, 1997).
There are several explanations that may explain the long duration and the later occurrence of the increased permeability after E. coli challenge in comparison with LPS challenge. In milk, E. coli releases very low concentrations of LPS in comparison with a direct intramammary injection of LPS, thereby explaining the later and lower response observed with the whole bacteria. Another explanation could be the engulfment of bacteria by epithelial cells (Döpfer et al., 2000, 2001) during the highest response of the immune system and the release of E. coli intact when the immune defense starts to decrease.
Sequentiality in Milk Proteolysis
Somatic cells.
The SCC increased from PIH 3 and then reached a plateau from PIH 6 to 216. For LPS challenge, SCC decreased already from PIH 16 onward. In addition, maximum SCC was lower after E. coli challenge (3.98 x 106 cells/mL) than after LPS challenge (27 x 106 cells/mL; Moussaoui et al., 2002). This supported previous clinical response and PP content results, as far as the immune response is concerned. In addition, the massive influx of leukocytes was supported by the increasing permeability of the epithelial barrier according to the BSA increase in milk over time.
E. coli.
Colony-forming unit increase was effective from PIH 3, and the plateau lasted until PIH 216. The bacteriostactic effect combined with the lag time could last between 3 and 6 h after E. coli inoculation, which corresponded to weak bacterial growth caused by some milk proteins with an inhibitory effect including lactoferrin, lysozyme, and some CN fragments resulting from proteolysis (Lahov and Regelson, 1996; Smithers et al., 1996).
PA.
Plasmin activity was significantly higher than baseline between PIH 15 and 24. The increase in PA was lower after E. coli challenge (maximum of 41.4 µmol of p-nitroanilide/h per L) than after LPS challenge (75.8 µmol of p-nitroanilide/h per L; Moussaoui et al., 2002). Plasmin is known to be the main protease involved in CN breakdown in milk (Kaartinen et al., 1988; Le Roux et al., 1995). The increase in its activity is linked to that of the permeability of the milk epithelial barrier during inflammation and, therefore, to the increase in BSA amount over time. This is supported by previous data related to the increase in permeability (Nickerson and Pankey, 1984) and therefore to an important influx of plasminogen, its zymogen (Kaartinen et al., 1988). In addition, numerous activators of plasmin and its zymogen come from blood stream, PMN, and bacteria. During mastitis, the serine protease activator concentration, including plasmin and plasminogen in blood and also in milk, increases sharply according to some researchers (Yamagata et al., 2001; Le Roux et al., 2003;). Polymorphonuclear neutrophils have a pool of plasminogen activators, such as urokinase type (u-PA) and, to a lesser extent, tissue (t-PA), according to Moir et al. (2001). Escherichia coli is reported to have both plasmin and plasminogen receptors on its surface, and these enhance enzyme activation and express plasminogen activators. An indirect plasmin activation channel is reported as being due to the secretion of t-PA by epithelial cells, thanks to cell stimulation by IL-1, in turn resulting from the release of monocytes, consequent to the contact with E. coli (Lähteenmäki et al., 2001).
PP.
The PP fraction is widely considered as a relevant indicator of proteolysis (Le Roux et al., 1995; Urech et al., 1999; Moussaoui et al., 2002). The increase in PP content as well as the maximum content (from PIH 12 to 33) was effective later than that following LPS challenge (Moussaoui et al., 2002). The PP increase was much more marked for E. coli (0.97 g/L at PIH 0 vs. 6.52 g/L at PIH 33) than for LPS challenge (1.02 g/L at PIH 0 vs. 2.49 g/L at PIH 8; Moussaoui et al., 2002). In addition, the increase lasted longer with E. coli than after LPS challenge. Casein breakdown occurred with a lag time, but was more intense in the case of E. coli mastitis. One explanation of the lag time could be the longer delay required for the immune response in the presence of E. coli. The slight increase in caseinolysis would be linked to the first immune barrier that was non-specific and corresponded to mainly macrophages.
Sequentiality.
The increases in SCC, cfu, and PA over time suggested their involvement in a global mechanism responsible for the increase of proteolysis in milk after E. coli challenge. Somatic cell count and E. coli may be involved from PIH 3 to 216, and PA involvement might be highlighted during the maximum proteolysis, from PIH 15 to 24. The maximal increase in PP content might correspond to very high protease concentrations in the milk. This very high PP content (up to 6.52 g/L) could result from an interaction between E. coli, the system plasmin-plasminogen, and neutrophils. Indeed, the regulatory mechanisms remain key in explaining the non-additional increase of proteolytic activities. Indeed, proteases of neutrophils as well as plasmin are known to permit the chemotaxis of cells in the site of inflammation. Furthermore, they are involved in the limitation in time of the immune response, e.g., by the cleavage of some cytokines such as IL-2, IL-6, IL-8, etc. Among these proteases, collagenase IV (also called MMP-9) activates cathepsin G as well as plasmin and inhibits serpines, the inhibitors of serine proteases. During the inflammation of the mammary gland, numerous protease activators and inhibitors come from the blood stream such as 2-macroglobulin, 1-anti-plasmin, serpines, u-PA, and t-PA (Le Roux et al., 2003). The involvement of E. coli consists of an activator effect on progelatinase A (also called pro-MMP-2) by E. coli-derived serine proteases (Takeda et al., 2000). The third (late) phase, from PIH 33 to 216, corresponded to the return to the initial level, with an increased level of SCC and cfu. In conclusion, sequentiality in the mechanism responsible for the increase of proteolysis in milk after E. coli challenge was suggested according to SCC, PA, and cfu data variation over time. Both leukocyte and E. coli taking parts were noted from PIH 3 to 216, but PA was significantly increased during the maximal proteolysis, from PIH 15 to 33.
CN Breakdown and the Resulting Peptides
During challenge, a compositional change in the PP fraction was noted with the apparition of 5 electrophoretic bands, including P1 and P2, previously reported during LPS challenge (Moussaoui et al., 2003) as well as E1, E2, and E3 specific to the E. coli challenge. The endogenous origins of CN breakdown are illustrated by previous studies. These demonstrate that concomitant increases in SCC and neutrophil enzymes play a key role in caseinolysis and are described by a recent work dealing with some CN peptides generated during LPS mastitis (Moussaoui et al., 2003). Indeed, P1 and P2 are known to be generated after CN breakdown by PMN proteases (Moussaoui et al., unpublished data) and P1 by elastase (Moussaoui et al., 2003), which elicits the involvement of PMN during the inflammatory response to E. coli challenge. Polymorphonuclear neutrophils degrade CN in extracellular compartments during de-granulation (Owen and Campbell, 1999) as well as in intracellular compartments such as CN fragments, which are internalized by neutrophils (Paape and Guidry, 1977). The activity of neutral proteases would be effective by degranulation; however, for acidic proteases, 2 mechanisms can be hypothesized: 1) activity at local pH levels in the close surroundings of the cell and 2) lysosomal action after the internalization of CN fragments followed by a release into the milk. As for the exogenous origins of CN breakdown, E. coli is known to contain several proteases likely to be involved. Nevertheless, the protease of E. coli best known to degrade CN is called ClpAP (caseinolytic protease with ATPase), according to Katayama et al. (1988). This protease, also called Ti, consists of a chaperone component, ClpA, and a proteolytic component, ClpP, which is a serine protease (Hwang, 1988). Numerous other proteases of E. coli can be cited, including serine and metallo-proteases, a majority of which are known to degrade CN (Rozkov, 2001). In addition to the large number of proteases and also that of enzyme activity during mastitis caused by bacteria, it is noteworthy that many of them degrade specifically abnormal proteins such as oxidized ones (Rozkov, 2001). Indeed, oxidized proteins forming aggregates, are selectively recognized and degrade by the intracellular proteasomal system (Merker and Grune, 2000) of activated macrophages (Gieche et al., 2001) as well as proteases from E. coli (Davies and Lin, 1988). This is highlighted by the production of reactive oxygen and nitrogen species during the inflammatory process, which can enhance CN breakdown. Indeed, reactive oxygen species and nitric oxide production during inflammation, and more precisely during LPS and E. coli mastitis, are well known (Blum et al., 2000; Sethi et al., 2001). Interestingly, this oxidative stress is strongly increased by serine proteases during inflammation (Aoshiba et al., 2001), which is highlighted by the increase in some serine protease activity during inflammation, e.g., elastase and cathepsins G during LPS mastitis (Le Roux et al., 2003). In conclusion, E. coli challenge induced the generation of 5 peptide fractions in PP: 2 had previously been recovered following LPS challenge and 3 were specific to E. coli challenge. Besides endogenous origins, proteolysis during E. coli challenge exhibits 2 specific origins: the large panel of bacterial proteases and the protein oxidation that increases their susceptibility to proteolysis.
|
CONCLUSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONACKNOWLEDGEMENTSREFERENCES |
|---|
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Received for publication October 31, 2003. Accepted for publication April 2, 2004.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONACKNOWLEDGEMENTSREFERENCES |
|---|
and nitrite/nitrate responses during acute mastitis induced by Escherichia coli infection and endotoxin in dairy cows. Domest. Anim. Endocrinol. 19:223–235.[Medline]
s1- and ß-caseins. Food Chem. 76:59–67.
This article has been cited by other articles:
|
|
 |
|
  X. Zhao and P. Lacasse Mammary tissue damage during bovine mastitis: Causes and control J Anim Sci, March 1, 2008; 86(13_suppl): 57 - 65. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. K. Park, H. C. Koo, S. H. Kim, S. Y. Hwang, W. K. Jung, J. M. Kim, S. Shin, R. T. Kim, and Y. H. Park The Analysis of Milk Components and Pathogenic Bacteria Isolated from Bovine Raw Milk in Korea J Dairy Sci, December 1, 2007; 90(12): 5405 - 5414. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
