Bovine Blood Neutrophil Acyloxyacyl Hydrolase (AOAH) Activity During

Bovine blood neutrophil acyloxyacyl hydrolase (AOAH) activity during endotoxin and coliform mastitis Jalil Mehrzad, Hilde Dosogne, Bart de Spiegeleer, Luc Duchateau, Christian Burvenich

To cite this version:

Jalil Mehrzad, Hilde Dosogne, Bart de Spiegeleer, Luc Duchateau, Christian Burvenich. Bovine blood neutrophil acyloxyacyl hydrolase (AOAH) activity during endotoxin and coliform mastitis. Veterinary Research, BioMed Central, 2007, 38 (5), pp.655-668. 10.1051/vetres:2007024. hal-00902885

HAL Id: hal-00902885 https://hal.archives-ouvertes.fr/hal-00902885 Submitted on 1 Jan 2007

HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. Vet. Res. 38 (2007) 655–668 Available online at: c INRA, EDP Sciences, 2007 www.vetres.org DOI: 10.1051/vetres:2007024 Original article

Bovine blood neutrophil acyloxyacyl hydrolase (AOAH) activity during endotoxin and coliform mastitis

Jalil Ma,c, Hilde Db,c,BartD Sd, Luc Dc, Christian Bc*

a Ferdowsi University of Mashhad, Faculty of Veterinary Medicine, Department of Pathobiology, Section Immunology, PO Box 91775-1793, Mashhad, Iran b DeLaval International AB, Product Portfolio Milk Quality and Animal Health, Industriepark 8-10, 9031 Drongen, Belgium c Ghent University, Faculty of Veterinary Medicine, Department of Physiology and Biometrics, member of Phimadran, Salisburylaan 133, 9820 Merelbeke, Belgium d Ghent University, Faculty of Pharmaceutical Sciences, Drug Quality and Registration (DruQuaR) group, member of Phimadran, Harelbekestraat 72, 9000 Gent, Belgium

(Received 7 December 2006; accepted 19 March 2007)

Abstract – The dynamics of blood neutrophil acyloxyacyl hydrolase (AOAH) activity, the appearance of endotoxin (lipopolysaccharide, LPS) in blood and the role of blood neutrophil AOAH in the severity of Escherichia coli and endotoxin mastitis were investigated in early postpartum dairy cows experimentally challenged with either endotoxin (n = 6) or E. coli (n = 6). The AOAH activity of blood neutrophils started to decrease significantly at post challenge hours (PCH) 6–24 and 12–24 in the endotoxin and E. coli-challenged groups, respectively; it returned to pre-challenged values at PCH 48 in both endotoxin- and E. coli-challenged groups. The cows were classified as moderate and severe responders according to milk production loss in the non-challenged quarters at PCH 48. There were no severe responders in the endotoxin-challenged group. In the E. coli-challenged group, only 1 severe responder was identified. The pre-challenge neutrophil AOAH activity of the severe responder was ∼30% lower than that of moderate responders. No LPS was detected in the plasma of endotoxin-challenged cows; neither was it found in the plasma of moderate responders in the E. coli- challenged group at any PCH. However, at PCH 6, a remarkable amount of LPS was detected in the plasma of the severe responder from the E. coli-challenged group. Furthermore, neutrophil AOAH activity was increased by ∼70% in the severe responder at PCH 6, but it increased by only ∼15% in moderate responders. This was followed by a decreased neutrophil AOAH activity at PCH 12–24 and 24–72 in moderate and severe responders, respectively; the decreased AOAH activity at those PCH was more pronounced in the severe responder. The pronounced decreased neutrophil AOAH activity during mastitis often coincided with extreme leukopenia, neutropenia and a maximal number of immature neutrophils in the blood. Our results demonstrate that a decrease in neutrophil AOAH activity results in the appearance of LPS in the blood, and low blood neutrophil deacylation activity could be considered as a risk factor for severe clinical coliform mastitis.

acyloxyacyl hydrolase (AOAH) / blood / endotoxin / mastitis / neutrophils

* Corresponding author: [email protected]

Article available at http://www.vetres.org or http://dx.doi.org/10.1051/vetres:2007024 656 J. Mehrzad et al.

1. INTRODUCTION and bactericidal-permeability increasing protein [15, 16, 25]. The 2-subunit lipase, An invading pathogen must be held in acyloxyacyl hydrolase (AOAH), which is check by the innate immune system until a also present in bovine neutrophil granules specific immune response can be mounted. [18, 19], hydrolyses and removes the In the case of Gram negative bacteria, secondary acyl chains of Gram-negative the principal stimulator of the innate im- bacterial lipid A of endotoxin [12]; this mune system is lipopolysaccharide (LPS) results in a substantial decreased toxicity or endotoxin, a compound of the bacte- of LPS while retaining much of the rial outer membrane. In the udder, this LPS immunostimulatory potency of native evokes several functional responses in the LPS [23]. Blood neutrophil AOAH activity neutrophils that contribute to the innate has been found to be decreased in postpar- immunity of the udder. The recruitment tum cows; this decrease was particularly and activation of these short-lived, bone pronounced in some cows, whereas in marrow-derived cells to the site of inflam- other cows no changes were observed [8]. mation or infection are pivotal to limit the Thus, in addition to other antibacterial severity of mastitis [2, 3, 20, 22, 28, 30]. or anti-LPS defense mechanisms such as Around calving, the risk for severe clinical the production of reactive oxygen species mastitis is extremely high; the main under- (ROS) [14, 20, 22], decreased AOAH lying reason for this high risk can be im- activity could represent a risk factor for paired blood neutrophil function [3,22,25]. severe clinical coliform mastitis. Although LPS has been recognized as Indeed, since LPS is considered as an important mediator of local and sys- a toxic compound to trigger a massive temic symptoms during coliform and en- TNF-α release [1], and endotoxin shock, dotoxin mastitis [17, 20], the role of blood decreased LPS detoxification could play neutrophils in defense against LPS dur- a critical role in the outcome of coliform ing Escherichia coli mastitis is not com- mastitis. To date, no one has demonstrated pletely understood. While the presence the changes in neutrophil AOAH activity of LPS in plasma is only short-lasting, and plasma LPS concentrations during col- its resorption from the mammary gland iform/endotoxin mastitis. Furthermore, the into the blood stream is possible during role of neutrophil AOAH activity as a risk E. coli mastitis due to increased permeabil- factor for the severity of E. coli mastitis in ity of the blood-milk barrier; this triggers early postpartum dairy cows has not been a marked increase of tumor necrosis fac- investigated so far. Therefore, the objec- tor alpha (TNF-α) in the blood. These tives of the present study were the follow- events are restricted to severely diseased ing: (1) investigate the temporal changes cows [1, 8, 20–22]. Severity of E. coli mas- in neutrophil AOAH activity following ex- titis seems to be related to the enhanced posure of the mammary gland to LPS and release of secondary induced inflammatory E. coli; (2) detect LPS in the blood; and (3) mediators such as TNF-α [1, 14], which relate this with the occurrence of endotox- could result from impaired LPS detoxifica- emia in cattle. tion in the blood. There are many-known and un- known mechanisms, enzymatic and 2. MATERIALS AND METHODS non-enzymatic, that neutralize LPS in the blood [15, 16, 18, 19]. Neutrophil granules This experiment was approved by the contain an array of anti-LPS defense Ethics Committee of Ghent University, proteins such as lactoferrin, lysozyme Faculty of Veterinary Medicine (Belgium). Coliform mastitis and neutrophil AOAH activity 657

2.1. Experimental animals and mixed with chlorhexidine as previously described [20]. Endotoxin mastitis In total 12 healthy Holstein-Friesian was induced – after the morning milk- cows in their first or second lactation and ing – into the left front and hind quarters between 2 and 6 weeks after parturition by a single intramammary (i.mam) injec- were used as experimental animals. The tion of 20 mL LPS solution per quarter animals, on a zero-grazing system from ar- (25 µgLPS/mL), final concentration) using rival till the end of the experiment, were a sterile teat cannula. After injection, each put into an individual stall and were fed quarter was massaged for 30 s to distribute with a special ration for pregnancy and lac- the LPS solution in the mammary gland. tation. The cows had calved normally and In the second study, one hour after the showed no signs of periparturient diseases. morning milking, 104 cfu of E. coli P4:O32 They had free access to water and hay. (H37, ß-glucuronidase +, haemolysin −) After gestation, clinically healthy cows were intramammarily injected into the showing no signs of typical periparturi- 2 left udder quarters of each cow. Prior to ent diseases were selected on the basis inoculation, the bacteria had been cultured of 2 consecutive bacteriological negative in Brain Heart Infusion broth (DIFCO Lab- ◦ milk samples and a milk somatic cell count oratories, Detroit, MI) for 18 h at 37 C, (SCC) of < 2.105/mL milk per individ- washed, resuspended and diluted with ster- ual quarter. One week before the start ile 0.01 M phosphate-buffered saline solu- of the experiment the animals were fed tion (PBS) as previously described [21,22]. a daily ration of approximately 8 kg of After cleaning of the udder and steriliza- concentrate and had free access to water tion of the teat ends, a total volume of and hay. They were milked twice daily at 10 mL containing 104 E. coli/mL was in- 8 a.m. and 5 p.m. with a 4-quarter milking jected into the left quarters using a sterile machine (Packo & Fullwood, Zeddelgem, teat cannula; to better observe the signs of Belgium). clinical mastitis, two quarters were used for the challenge. After injection of the E. coli suspension, each quarter was mas- 2.2. Induction of mastitis saged for 30 s to distribute the bacteria in the quarters. In the first study, mastitis was experimentally induced by LPS infusion (n = 6) and in the second study, mastitis was in- 2.3. Blood sampling duced by infusion of living E. coli bacteria (n = 6). In the first study, 10 mg LPS from Blood samples for leukocyte count- E. coli O111:B4 (Sigma Chemical Co., St. ing, differentiation, enumeration and iso- Louis, MO, USA) was diluted in 100 mL lation of neutrophils and determination pyrogen-free (9 g/L) saline solution and of neutrophil AOAH activity were col- aliquoted in bottles of 5 mL LPS solution lected at post challenge hours (PCH) 0, (500 µgin5mL).Allairwasremoved 6, 12, 24, 48, 72, 144 and 216 after from the bottles by a sterile N2–flow. LPS endotoxin and E. coli challenge. A to- solutions were stored at −20 ◦C until use. tal of 40 mL blood was collected at Frozen LPS solutions were thawed im- each sampling by venipuncture in Vac- mediately before a challenge experiment uette tubes (Greiner, 9 mL) containing and 15 mL of pyrogen-free saline solution lithium heparin and stored on ice until pro- was added. Before LPS injection the teat cessed. For LPS quantification in plasma, ends were disinfected with ethanol (70%) additional samplings were performed at 658 J. Mehrzad et al. those PCH. To minimize exogenous LPS procedure were determined to be pyrogen- contamination, samples were collected free using a qualitative limulus amebo- in a pyrogen-free manner by puncture cyte lysate assay (BioWhittaker Inc., Walk- of the external jugular vein with dis- ersville, MD, USA). The final pellet of posable needles mounted on disposable neutrophils was resuspended in 2.5 mL tubes (Vacutainer, Terumo, Belgium); PBS (0.01 M). The total number of leuko- endotoxin-free vacuum blood collection cytes in whole blood and isolated blood tubes containing 120 i.u. sodium hep- cells was determined using an electronic arin (Chromogenix AB, Mölndal, Sweden) particle counter [20, 22]. The total num- were used. Plasma samples were frozen at ber of different circulating leukocytes was −20 ◦C and stored until use for LPS quan- determined by combination of the total tification. number of blood leukocytes and the differentiation on smear preparations of blood samples [20, 22]. Differential cell counts 2.4. Classification of mastitis challenged and staining procedures were performed cows into severe and moderate on whole blood similar to the isolates on responders eosin-Giemsa-stained (Merck Diagnostica, Darmstadt, Germany) smears, using light Milk production (MP) of individual microscopy (Nikon labophot-2) at magnifi- quarters was measured using a 4-quarter cation ×1000. Cell identification was based milking machine in order to determine MP on morphological characteristics as previ- in challenged and non-challenged quarters. ously described [20, 22]. The MP data from evening and morn- To quantify percentages of each cell ing milk were pooled to obtain the daily type in the samples, a total of 200 cells per MP. The MP of non-challenged quarters at slide were classified as neutrophils (ma- 2 days after challenge, compared with pre- ture and immature), monocytes, lympho- challenge MP, was used for classification cytes and eosinophils. The total number of of mastitis challenged cows into severe (S, cells isolated was determined by using the MP < 50% of pre-challenge MP) and mod- cell counter, and the percentage of neu- erate (M, MP > 50% of pre-challenge MP) trophils was determined by direct exam- responders according to [7, 13, 22]. ination. The isolation technique routinely provided large numbers of neutrophils that were determined to be viable by the pro- 2.5. Blood leukocyte enumeration, pidium iodide exclusion method [21] and neutrophil isolation, preparation greater than 95% neutrophils in content. To and differentiation prepare neutrophil lysates containing the AOAH enzyme, the isolated neutrophils The number of leukocytes in whole were lysed with a buffer (100 mM KCI; blood was determined using an electronic 3.9 mM NaCl; 3.5 mM MgCl2;10mM particle counter (Coulter counter Z2, Coul- HEPES (N-2-hydroxyethylpiperazine- N’- ter Electronics, Hialeah, FL, USA) af- 2-ethanesulfonic acid) (pH 7.4) that con- ter addition of 200 µL of Zap-Oglobin tained 1% (vol/vol) Nonidet P-40, 15 mM II lytic reagent (Beckman Coulter, Paris, EDTA, and 75 g of phenylmethylsulfonyl France) to 105-fold diluted blood sam- fluoride per mL). The lysis buffer was ples and gentle mixing. Neutrophils were added (1 mL.10−7 viable neutrophils), and isolated from whole blood as previously the cells were incubated at room tem- described [4, 20–22]. Double distilled wa- perature for 10 min with frequent mix- ter and buffers used for the isolation ing; then the mixtures were centrifuged at Coliform mastitis and neutrophil AOAH activity 659

2000 × g for 10 min. Supernatant aliquots amœbocyte lysate (LAL) test (Biowhit- were stored at −70 ◦C, for quantification of taker Inc, Walkersville, Maryland, neutrophil AOAH. USA) according to Dosogne et al. [9]. Briefly, one endotoxin unit (EU) per mL platelet-rich plasma (PRP) corresponds to 2.6. Detection of SCC and CFU 100 pg LPS/mL PRP. The LPS detection in challenged quarters limit of the assay for bovine plasma was 0.036 EU/mL PRP. The LPS values were To examine the dynamics of SCC and only considered positive when it was CFU in challenged quarters and to re- higher than 2 times the detection limit; that late these changes to our main findings, is, > 0.072 EU/mL PRP or > 7.2 pg/mL milk samples were taken at PCH 0, 6, LPS. The results are expressed in pg/mL. 12, 24, 48, 72, 144 and 216 of the endotoxin and E. coli challenge. The SCC was determined by a Fluoro-optoelectronic 2.9. Statistics cell counting procedure (Fossomatic 360; Foss Electronic, Eden Prairie, MN, USA). Statistical analysis of the data was per- The maximal detection capacity of the formed using a mixed model with the cow 7/ as the random effect, time as a categori- method for SCC values was 10 mL. The ff CFU was performed throughout the study, cal fixed e ect and milk reduction and its interaction with time as continuous fixed using bacterial plate culture, as previously ff ff described [21, 22]. e ects. Significance of the di erences was determined at P < 0.05∗, P < 0.01∗∗ and P < 0.001∗∗∗. 2.7. Neutrophil acyloxyacyl hydrolase (AOAH) activity 3. RESULTS

AOAH activity in neutrophil lysates was 3.1. Clinical findings and identification determined according to McDermott et al. of severe (S) and moderate (M) [18, 19], modified by Dosogne et al. [8]. responders 3Hand14C radioactivity were measured simultaneously with a liquid scintillation Intramammary E. coli and LPS pro- counter (1219 Rackbeta, LKB Wallac). voked clinical signs (both local and sys- AOAH activity was measured as the ra- temic) of acute mastitis such as fever, dioactivity of 3H-labeled fatty acids re- tachycardia, decreased rumen motility and leased from LPS whereas 14C was used for increase of CFU and SCC with decreased correction of LPS backbone contamination MP in all quarters of the cows. The MP of the fatty acid extracts. AOAH activity in challenged quarters substantially de- was expressed as picomole (pM) fatty acid creased; even at PCH 12–54 it hardly released per 107 neutrophils per hour. All yielded 100 mL for the experiment (data AOAH determinations were performed in not shown). This decrease resulted from duplicate. the damage to the gland caused by both systemic and local effects of mastitis. Most clinical signs peaked at PCH 6–24 and 2.8. Determination of LPS in plasma almost completely restored at PCH > 72 (Tab. I). The amount of LPS in plasma was There were no S responders when us- determined using a quantitative limulus ing MP loss of non-challenged quarters 660 J. Mehrzad et al.

Table I. Comparison of some circulating leukocytes (WBC), mature neutrophils, immature neutrophils (sum of myelocytes and metamyelocytes) and AOAH (acyloxyacyl hydrolase) activity (pM f.a./106 neutrophils/h) of blood neutrophils and appearance of LPS in serum as well as the SCC and cfu dynamics in inflamed quarters of cows experimentally challenged with LPS (n = 6) and E. coli (moderate responders; n = 5, and severe responder; n = 1). ND = not detected; PCH = post challenge hours; time 0 = hour of mastitis induction. The values are means ± SEM. The significance of the difference between pre and post-infusion of LPS/E. coli is indicated with asterisks (*P < 0.05, ** P < 0.01 and *** P < 0.001).

PCH Parameter Endotoxin mastitis E. coli mastitis group group (n = 6) Moderate responders Severe responder (n = 5) (n = 1) WBC/µL 7314 ± 213 9633 ± 658 8540 Mature neutrophils /µL 1903 ± 80 2886 ± 290 1708 Immature neutrophils /µL 216 ± 34 467 ± 97 256 0 LPS in Serum (pg/mL) ND ND ND AOAH activity 73.0 ± 2.8 70.8 ± 2.5 56.7 SCC/mL (×1000) 81 ± 17 86.4 ± 6.1 71 CFU/mL (×1000) 0 0 0 WBC/µL 2368 ± 299*** 5974 ± 160*** 6650 Mature neutrophils /µL 454 ± 70*** 1611 ± 30*** 133 Immature neutrophils /µL 554 ± 79** 636 ± 67* 532 6 LPS in Serum (pg/mL) ND ND 26 AOAH activity 47.9 ± 2.0** 81.0 ± 7.7* 92.9 SCC/mL (×1000) 4516 ± 657 3900 ± 472 110 CFU/mL (×1000) 0 14.1 ± 11.2 390 WBC/µL 4442 ± 1180** 2783 ± 373** 8160 Mature neutrophils /µL 568 ± 141*** 590 ± 85*** 82 Immature neutrophils /µL 868 ± 232** 400 ± 48 979 12 LPS in Serum (pg/mL) ND ND 93 AOAH activity 59.8 ± 2.1** 48.6 ± 4.1*** 58.6 SCC/mL (×1000) > 10000 7680 ± 674 4900 CFU/mL (×1000) 0 25.3 ± 9.4 1400 WBC/µL 12372 ± 2125** 5182 ± 389** 3340 Mature neutrophils /µL 1965 ± 420 988 ± 110** 67 Immature neutrophils /µL 2030 ± 411** 1138 ± 137 1536 24 LPS in Serum (pg/mL) ND ND 9.7 AOAH activity 65.4 ± 1.7* 55.4 ± 8.5*** 24.6 SCC/mL (×1000) > 10000 4800 ± 424 > 10000 CFU/mL (×1000) 0 8.9 ± 4.68 420 WBC/µL 12222 ± 1244** 8944 ± 236 2580 Mature neutrophils /µL 1967 ± 284 1629 ± 100* 26 Immature neutrophils /µL 1283 ± 103** 1495 ± 168** 1367 48 LPS in Serum (pg/mL) ND ND ND AOAH activity 70.0 ± 1.5 72.4 ± 1.2 18.2 SCC/mL (×1000) 3390 ± 613 2080 ± 305 > 10000 CFU/mL (×1000) 0 1.9 ± 0.62 180 Coliform mastitis and neutrophil AOAH activity 661

Table I. Continued.

PCH Parameter Endotoxin mastitis E. coli mastitis group group (n = 6) Moderate responders Severe responder (n = 5) (n = 1) WBC/µL 8155 ± 1656 11315 ± 464** 3560 Mature neutrophils /µL 1360 ± 305* 2014 ± 248 36 Immature neutrophils /µL 512 ± 119* 1349 ± 167** 2100 72 LPS in Serum (pg/mL) ND ND ND AOAH activity 72.0 ± 1.6 96.6 ± 5.9** 14.8 SCC/mL (×1000) 1127 ± 167 867 ± 297 7950 CFU/mL (×1000) 0 0.45 ± 0.2 35 WBC/µL 10230 ± 1317* 11352 ± 475** 7110 Mature neutrophils /µL 1735 ± 214 1902 ± 128 995 Immature neutrophils /µL 430 ± 44* 1226 ± 107** 1635 144 LPS in Serum (pg/mL) ND ND ND AOAH activity 74.3 ± 1.4 70.7 ± 4.1 68.7 SCC/mL (×1000) 217 ± 9 274 ± 39 5500 CFU/mL (×1000) 0 0 5.5 WBC/µL 9055 ± 246* 10462 ± 565* 8640 Mature neutrophils/µL 1808 ± 69 1956 ± 159 1123 Immature neutrophils/µL 257 ± 30 818 ± 84* 1642 216 LPS in Serum (pg/mL) ND ND ND AOAH activity 68.2 ± 1.4 77.2 ± 3.2 69.2 SCC/mL (×1000) 107 ± 12 112 ± 15 4100 CFU/mL (×1000) 0 0 5.6 at PCH 48 in the endotoxin-challenged leukopenia, leukocytosis, neutropenia and group. In the E. coli-challenged group, neutrophilia was observed (Tab. I). Sig- 1 cow was identified as an S responder, nificant leukopenia was observed between whereas the other 5 cows in the E. coli PCH 6 and 12 and 6–24 in endotoxin model were M responders. The MP of and E. coli mastitis, respectively, regain- non-challenged quarters was decreased to ing, even exceeding, pre-challenge values 5% of the pre-challenge MP in the S cow after those PCH (Tab. I). Compared to and extremely low until 5 days after chal- pre-challenged values, the number of cir- lenge (Fig. 1). In M cows, the MP of culating mature neutrophils was signifi- non-challenged quarters 2 days after chal- cantly (P < 0.001) minimal at PCH 6 lenge was almost the same volume as the and 12 in endotoxin and E. coli mas- pre-challenge MP and was completely re- titis groups, and in E. coli it still re- covered at 3 days after challenge. mained significantly low till PCH 48. This decrease often coincided with an increased number of immature neutrophils 3.2. Blood leukocyte enumeration, (metamyelocyte and myelocytes). Blood differentiation and neutrophil neutrophil numbers changed significantly immaturity over time (P < 0.0001). Blood neutrophil numbers decreased significantly with in- After challenge, both in endotoxin creasing milk reduction (P = 0.050). In and E. coli challenged cows, a transient the S responder with the highest reduction 662 J. Mehrzad et al.

Figure 1. Daily milk production (liters/udder halves) from the non-challenged contralateral quarters from LPS-challenged cows (n = 6 ± SEM; triangles), E. coli-challenged cows (moderate; n = 5 ± SEM; squares) and E. coli-challenged cows (severe; n = 1; circles) during mastitis. Day 0 = day of mastitis induction. the decrease was even reached to 133 In all groups before the experiment, and neutrophils/µL at PCH 6. The reduction in the endotoxin group throughout the ex- in blood neutrophil numbers of this cow periment, the CFU was always zero. The was severe and remained extremely low CFU in milk of the S cow increased much between PCH 6 and 72 (less than 5% of faster (at PCH 6) and did not reach zero pre-challenged values; Tab. I). throughout the infection, whereas in M Mature blood neutrophil num- cows it was not substantially increased and bers changed significantly over time it reached zero at PCH 144 and onwards (P < 0.0001). Mature blood neutrophil (Tab. I). numbers decreased significantly with increasing milk reduction (P = 0.002). The pattern of decrease in immature neu- 3.4. Blood neutrophil acyloxyacyl trophils (metamyelocyte and myelocytes) hydrolase activity was remarkable in the S responder (Tab. I). Although both groups were moderate responders, the pattern of changes in neu- 3.3. SCC and CFU in challenged trophil AOAH activity in the endotoxin- quarters challenged group was different from those of the E. coli-challenged (the M) group The overall pre-challenged SCC was the (Tab. I, Fig. 2). After challenge, in the same for endotoxin, M and S groups. In en- E. coli (M) group, AOAH activity in- dotoxin challenged quarters, at PCH 12–24 creased (at PCH 6; P < 0.05), then the SCC value exceeded the maximal de- remarkably decreased (at PCH 12–24; tection capacity of the method (Tab. I). The P < 0.001), reaching pre-challenged val- SCC in endotoxin and M groups, however, ues at PCH 48 and again increased (at PCH increased faster (maximal at PCH 12). In 72; P < 0.001), and finally reached the the S cow the maximal SCC appeared pre-challenged values; whereas in the en- much later (at PCH 24). Throughout the dotoxin group, the neutrophil AOAH activ- experiment the SCC values did not reach ity first decreased (at PCH 6–24; P < 0.01) the normal in the S cow (see Tab. I). then reached the pre-challenged values at Coliform mastitis and neutrophil AOAH activity 663

Figure 2. Changes in AOAH activity (pM f.a./106 neutrophils/h) of blood neutrophils and its re- lation with milk production reduction of non-challenged quarters at 2 days after challenge during experimentally induced endotoxin mastitis (n = 6; circles) and E. coli mastitis (n = 6; triangles); PCH = post challenge hours. The triangle, which is located at the most right part of the ﬁgures, represents the values for the severe responder.

PCH 48 and onwards (Tab. I, Fig. 2). reduction (P = 0.0004). There was, how- Therefore, a rebound effect was observed ever, also a significant interaction between at PCH 72 only in M cows after re- time and milk reduction (P < 0.0001). establishment of blood neutrophil AOAH This was due to the fact that the neutrophil activity at PCH 48. AOAH activity of the severe responder in- Neutrophil AOAH activity changed sig- creased dramatically at PCH 6, reaching nificantly during infection (P < 0.0001). pre-challenged values and again hugely de- On average neutrophil AOAH activity decreased (Tab. I), which was not the case for creased significantly with increasing milk the endotoxin and M groups. 664 J. Mehrzad et al.

Milk reduction increased with lower been shown to be an appropriate parameter value of pre-challenged neutrophil AOAH to estimate systemic illness during E. coli activity; indeed, before challenge, the neu- mastitis [2, 7, 13, 21, 22]. trophil AOAH activity of the S responder Many host-derived cytokines such as was ∼30% lower than that of M respon- interleukin-1 (IL-1), IL-6 [25, 26] and ders (Tab. I). At PCH 6, neutrophil AOAH TNF-α [14] are produced during endo- activity was increased by ∼70% in the toxin/coliform mastitis; they cause sys- S responder, which was remarkably dif- temic effects such as fever and increase ferent from those of other groups. Over- the bone marrow output of leukocytes to all, throughout the study, the changes on compensate for the decrease in the circu- AOAH activity was the most pronounced lating pool [20–22]. Consequently, there is in the severe responder (Tab. I, Fig. 2). a transient increase in the number of circulating immature neutrophils, increased permeability of capillaries and the blood/milk 3.5. LPS in plasma barrier, and a concomitant exudation of some blood proteins in the mammary In endotoxin-challenged cows as well as gland [14, 21]. Increased permeability of in the E. coli challenged cows (M group), the blood/milk barrier was indicated by the no LPS was detected in the plasma at any appearance of clots in milk at PCH 6–24. time point following intramammary chal- Between PCH 6 and 24, the influx of a lenge. A remarkable amount of LPS (26, large number of neutrophils in the chal- 93 and 9.7 pg/mL at PCH 6, 12 and 24, lenged quarters resulted in a substantial in- respectively) was detected in the plasma crease of the SCC (see Tab. I). The recruit- (Tab. I) but only in the S cow of the E. coli ment of neutrophils into the challenged challenged group. quarters, as reflected by the SCC, together with their increased functionality [20–22] are both important aspects of the defense 4. DISCUSSION mechanism against E. coli/endotoxin during coliform mastitis. In this study the two The major goal of our experiment was left quarters were injected in order to better to examine the temporal changes on neu- assess the signs of clinical mastitis, and the trophil deacylation activity and serum LPS cows’ responses to the endotoxin/E. coli concentration during both endotoxin and challenge would have been less intensive E. coli mastitis. The fluctuations in blood if we had used only one quarter. neutrophil AOAH activity during both en- The increased milk SCC in challenged dotoxin and E. coli mastitis were con- quarters coincided with decreased CFU. sistent with other findings [18, 19]. This Although no significant differences on study shows, however, a novel-detailed in- pre-challenged SCC were observed be- formation on temporal changes, especially tween all groups, the SCC increased much during the critically earliest stages of mas- faster in endotoxin and M groups (see titis, during which the nature of the cow’s Tab. I). This is a classic immunological innate immune response governs whether phenomenon [3, 21, 22, 25] that was more and how to eliminate an intramammary appropriate in endotoxin and M groups, infection from the gland [3, 22, 28]. In ad- substantially lessening CFU (Tab. I). The dition, we distinguished between severely reason why a remarkable amount of LPS and moderately diseased cows based on the was detected at PCH 6–12 would be that MP loss in non-challenged quarters. The at these PCH there were very low levels MP loss in non-challenged quarters has of mature neutrophils, and that an array Coliform mastitis and neutrophil AOAH activity 665 of LPS-neutralizing proteins (e.g., cationic ing the toxic effects of endotoxin during peptides) produced by neutrophils may E. coli/endotoxin mastitis. The underlying be at low levels due to low numbers of mechanism of these differences remains to mature neutrophils (see Tab. I, Fig. 2). be investigated. Furthermore, at those PCH the extremely Accordingly [8, 9, 18, 19], before chal- rapid/unstoppable increase of CFU (E. coli lenge and during early hours of mastitis, growth) resulted in a huge imbalance be- the major source of LPS neutralizing fac- tween AOAH and its substrate (LPS) in tors are predominantly neutrophil AOAH. the S cow, and the body’s AOAH capacity This might be a decisive factor in deacyla- was insufficient to deacylate LPS. There- tion and thereby detoxification of LPS thus fore, the initial E. coli growth may result prevents prolonged mastitis. Interestingly, in an extremely high LPS production and before challenge, blood neutrophil AOAH AOAH-LPS deacylation imbalance. This activity was lower in S than in M cows. might result from a far slower increase Our data, therefore, suggest decreased of SCC in the gland at those PCH (see LPS detoxification mechanisms. Indeed, in Tab. I). This exciting topic needs to be fur- the cow with impaired blood neutrophil ther studied. AOAH activity before challenge, MP was Local signs of mastitis such as swelling strongly decreased for several days fol- and pain of the challenged quarters, ap- lowing intramammary E. coli challenge pearance of flecks and milk leakage and LPS were detected in plasma, whereas in challenged quarters were observed at in the other cows MP was restored very PCH 6 to 24 (data not shown) for all rapidly and no LPS were found in plasma. groups. These local signs were more pro- In addition to an impairment of blood nounced and prolonged in the S cow, ex- neutrophil AOAH activity, the number of actly indicating a greater breakdown of the blood neutrophil was also drastically de- blood-milk barrier in the severe respon- creased in the S cow (see Tab. I), sug- der, potentially enhancing efflux of LPS to gesting a further decrease of total LPS the blood. However, at PCH 48–78, dur- detoxification capacity in the blood. The ing which the AOAH activity and mature results from our study agreed with other neutrophils were the lowest in this cow, studies [8,13,22,26,30] suggesting that the we expected to have detectable LPS, but occurrence of LPS in plasma of severely the LPS was not detected in the blood. diseased cows could be the result of a de- The most acceptable reason for these dis- creased detoxification capacity of blood parities would be the “non-enzymatic sub- neutrophils. In our study, it was possible stances” in mastitis milk and serum that that in M cows the E. coli endotoxins were might contribute to endotoxin neutraliza- effectively controlled by milk neutrophil tion [18, 19, 22]. For example, inhibition AOAH, though milk neutrophil AOAH ac- of neutrophil AOAH activity during natu- tivity was not assessed. ral cases of mastitis caused by both Gram The AOAH activity per neutrophil in negative and Gram positive organisms endotoxin mastitis was also changed, how- [18, 19] supports the involvement of non- ever, the time of the alteration was different enzymatic local-and-systemic mediators from that of the E. coli challenged group like LPS binding proteins (LBP), lactofer- (see Tab. I, Fig. 2). The rapid decreased rin, TNF-α [1, 32] in LPS neutralization. neutrophil AOAH activity after LPS chal- Although the LPS molecule is not de- lenge in comparison with E. coli challenge graded by, for instance, LBP, the LPS-LBP might be due to a faster inflammatory re- bond could result in decreased bioavail- sponse caused by intramammary admin- ability of the LPS molecule, thereby reduc- istration of LPS. In contrast, intravenous 666 J. Mehrzad et al. administration of LPS to rabbits resulted in consistent with previous results [18, 19] a rapid (within 90 min) increase of plasma and can be explained by the stimulatory AOAH activity [10]. So, a different route of effect of secondary induced inflammatory LPS administration leads to different tem- mediators. In a previous experiment, the poral changes in neutrophil AOAH activ- priming effect for enhancement of neu- ity. Surprisingly, intramammary adminis- trophil ROS production in M responders at tration of E. coli resulted in faster increased PCH 72 was tentatively explained by the AOAH activity of neutrophils (PCH 6). action of endogenous inflammatory media- Therefore, it is possible that, in contrast to tors such as interferon-gamma but not LPS endotoxin, live bacteria elicit an increase in [13, 22], as in our study. Indeed, enhance- AOAH activity in both M and S responders ment of neutrophil AOAH activity was not immediately after infection. This might be observed in the S cow at PCH 72, during due to the involvement of a phagocytosis which no LPS was detected in its plasma. phenomenon even by mammary epithelia Therefore, the impairment of neutrophil during maximal immune response which AOAH activity at those PCH can partly be then releases E. coli in the tissue, intact [6]. ascribed to extreme neutropenia with in- In the LPS model this phenomenon is al- creased neutrophil immaturity. most non-existent. In this study only intracellular AOAH In our study, blood neutrophil AOAH activity of neutrophils was investigated. activity was decreased at PCH 12–24 and Therefore, our results might explain the 6, 24–72 in M and S groups, respectively, notion that LPS first had to be endocytosed with a much more pronounced effect in by neutrophils in order for the intracel- the S cow. This has not been previously lular AOAH-dependent detoxification (and reported. These novel findings could be other LPS detoxification mechanisms) to explained by neutrophil immaturity, since take place. It is generally accepted that neutrophil granule enzymes are acquired LPB first forms a complex with LPS, along their maturation pathway [11, 29]. which then binds to the CD14 molecule The release of immature neutrophils during and TLR, especially TLR4, on phago- E. coli mastitis was consistent with previ- cytes [27]. The β2-integrin CD11b/CD18 ous reports [13, 21, 22, 30] and can be as- has also been described as an LPS re- cribed to indirect hematopoietic effects of ceptor on bovine neutrophils [24]. In- LPS on the bone marrow. The long-lasting tramammary LPS infusion induced an impairment of blood neutrophil AOAH ac- upregulation of CD11b/CD18 and CD14 tivity in the S cow could be explained by its on neutrophils [5, 24, 30] for further LPS prolonged shift to the left (intense produc- recognition. Extracellular AOAH activity tion of immature neutrophils; see Tab. I). was not measured in the present study. Be- Together, our findings suggest a direct cause it has been demonstrated that extra- or indirect stimulatory effect of LPS on cellular and intracelluar AOAH activities blood neutrophil AOAH activity during of neutrophils show the same trend during E. coli mastitis. In the S cow, intense stim- induced E. coli mastitis [18, 19], it may be ulatory effects were observed very early expected that extracellular AOAH activity (PCH 6), whereas in M cows, despite a also showed the same trends as intracellu- mild increase at PCH 6, an intense increase lar neutrophil AOAH activity in our study. was at a later stage of the inflammation The current study has at least (PCH 72); this could be due to rapid LPS demonstrated that a decreased intracellular resorption and increased neutrophil imma- neutrophil AOAH activity together with turity in the S cow. The rebound effect of extreme leukopenia, neutropenia and a neutrophil AOAH activity in M cows was maximal number of immature neutrophils Coliform mastitis and neutrophil AOAH activity 667

was associated with the occurrence of LPS [5] Diez-Fraille A., Mehrzad J., Meyer E., in plasma, indicating a significant role Duchateau L., Burvenich C., Comparison of L-selectin and Mac-1 expression on blood of not only the quantity of neutrophils, and milk neutrophils during experimental but also, as importantly, the quality of Escherichia coli-induced mastitis in cows, neutrophils (AOAH production capacity) Am. J. Vet. Res. (2004) 65:1164–1171. during E. coli mastitis. Besides AOAH, [6] Döpfer D., Almeida R.A., Lam T.J.G.M., bovine neutrophil granules also contain Nederbragt H., Oliver S.P., Gaastra W., other LPS binding cationic proteins such as Adhesion and invasion of Escherichia coli lactoferrin, and a wide variety of cationic from single and recurrent clinical cases of bovine mastitis in vitro, Vet. Microbiol. peptides and oxidative enzymes that (2000) 74:331–343. contribute to enhance neutrophil quality [7] Dosogne H., Burvenich C., van Werven and LPS neutralization [15, 16, 21, 22, 31]. T., Roets E., Noordhuizen-Stassen E.N., In short, this study demonstrates that a Goddeeris B., Increased surface expression decrease in neutrophil AOAH activity re- of CD11b receptors on polymorphonuclear ffi sults in the appearance of LPS in the blood, leukocytes is not su cient to sustain phagocytosis during Escherichia coli mastitis in and low blood neutrophil deacylation ac- early postpartum dairy cows, Vet. Immunol. tivity could be considered as a risk factor Immunopathol. (1997) 60:47–59. for severe clinical coliform mastitis. [8] Dosogne H., Capuco A.V., Paape M.J., Roets E., Burvenich C., Fenwick B., Reduction of acyloxyacyl hydrolase activity in circulating ACKNOWLEDGEMENTS neutrophils from cows after parturition, J. Dairy Sci. (1998) 81:672–677. This work was supported in part by the [9] Dosogne H., Meyer E., Sturk G., Van loon ff Flemish Institute for the Encouragement of Re- J., Massart-Leen A.M., Burvenich C., E ect of enrofloxacin treatment on plasma endo- search in the Industry (IWT-grant No. 030784) toxin during bovine Escherichia coli masti- and the Ferdowsi University of Mashhad. tis, Inflamm. Res. (2002) 51:201–205. [10] Erwin A.L., Munford R.S., Plasma lipopolysaccharide-deacylating activity (acylo- REFERENCES xyacyl hydrolase) increases after lipopolysaccharide administration to rabbits, Lab. [1] Blum J.W., Dosogne H., Hoeben D., Invest. (1991) 65:138–144. Massart-Leën A.M., Hammon H.M., [11] Gullberg U., Andersson E., Garwicz D., Bruckmaier R.M., Burvenich C., Tumor Lindmark A., Olsson I., Biosynthesis, pro- necrosis factor-alpha and nitrite/nitrate cessing and sorting of neutrophil proteins: responses during acute mastitis induced by insight into neutrophil granule development, Escherichia coli infection and endotoxin Eur. J. Haematol. (1997) 58:137–153. in dairy cows, Domest. Anim. Endocrinol. (2000) 19:223–235. [12] Hall C.L., Munford R.S., Enzymatic deacy- [2] Burvenich C., Paape M.J., Guidry A.J., lation of the lipid A moiety of Salmonella Miller R.H., Heyneman R., Kremer W.D.J., typhimurium lipopolysaccharides by human Brand A., Role of the neutrophil leukocyte neutrophils, Proc. Natl. Acad. Sci. USA in the local and systemic reactions during ex- (1983) 80:6671–6675. perimentally induced E. coli mastitis in cows [13] Heyneman R., Burvenich C., Vercauteren immediately after calving, Vet. Q. (1994) R., Interaction between the respiratory burst 16:45–49. activity of neutrophil leukocytes and experi- [3] Burvenich C., Van Merris V., Mehrzad J., mentally induced Escherichia coli mastitis in Diez-Fraile A., Duchateau L., Severity of cows, J. Dairy Sci. (1990) 73:985–994. E. coli mastitis is mainly determined by cow [14] Hoeben D., Burvenich C., Trevisi E., Bertoni factors, Vet. Res. (2003) 34:521–562. G., Hamann J., Blum J.W., Role of endotoxin [4] Carlson G.P., Kaneko J.J., Isolation of leuko- and TNF-α in the pathogenesis of experi- cytes from bovine peripheral blood, Proc. mentally induced mastitis in periparturient Soc. Exp. Biol. Med. (1973) 142:853–856. cows, J. Dairy Res. (2000) 67:503–514. 668 J. Mehrzad et al.

[15] Levy O., Ooi C.E., Elsbach P., Doerfler [25] Paape M., Mehrzad J., Zhao X., Detilleux M.E., Lehrer R.I., Weiss J., Antibacterial J., Burvenich C., Defense of the bovine proteins of granulocytes differ in inter- mammary gland by polymorphonuclear neu- action with endotoxin: Comparison of trophil leukocytes, J. Mammary Gland Biol. bactericidal/permeability-increasing protein, Neoplasia (2002) 7:109–121. p15s, and defensins, J. Immunol. (1995) [26] Shuster D.E., Lee E.K., Kehrli M.E., 154:5403–5410. Bacterial growth, inflammatory cytokine [16] Levy O., Antibiotic proteins of polymor- production, and neutrophil recruitment dur- phonuclear leukocytes, Eur. J. Haematol. ing coliform mastitis in cows within ten days (1996) 56:263–277. after calving, compared with cows at mid- [17] Lohuis J.A., Verheijden J.H., Burvenich C., lactation, Am. J. Vet. Res. (1996) 57:1569– Van Miert A.S., Pathophysiological effects 1575. of endotoxins in ruminants, 2. Metabolic as- [27] Tobias P.S., Ulevitch R.J., Lipopoly- pects, Vet. Q. (1988) 10:117–125. saccharide binding protein and CD14 in [18] McDermott C.M., Cullor J.S., Fenwick LPS dependent macrophage activation, B.W., Intracellular and extracellular enzy- Immunobiology (1993) 187:227–232. matic deacylation of bacterial endotoxin [28] Vandeputte-Van Messom G., Burvenich C., during localised inflammation induced by Roets E., Massart-Leën A.M., Heyneman R., Escherichia coli, Infect. Immun. (1991) Kremer W.D.J., Brand A., Classification of 59:478–485. newly calved cows into moderate and se- [19] McDermott C.M., Morrill J.L., Fenwick vere responders to experimentally induced B.W., Deacylation of endotoxin during nat- Escherichia coli mastitis, J. Dairy Res. ural cases of bovine mastitis, J. Dairy Sci. (1993) 60:19–29. (1991) 74:1227–1234. [29] Van Merris V., Meyer E., Burvenich C., [20] Mehrzad J., Dosogne H., Meyer E., Functional maturation during bovine gran- Burvenich C., Local and systemic effects ulopoiesis, J. Dairy Sci. (2002) 85:2859– of endotoxin mastitis on the chemilumines- 2868. cence of milk and blood neutrophils in dairy [30] Van Werven T., Noordhuizen-Stassen E.N., cows, Vet. Res. (2001) 32:131–144. Daemen A.J.J.M., Schukken Y.H., Brand A., [21] Mehrzad J., Duchateau L., Burvenich C., Burvenich C., Preinfection in vitro chemo- taxis, phagocytosis, oxidative burst, and ex- Viability of milk neutrophils and severity of / bovine coliform mastitis, J. Dairy Sci. (2004) pression of CD11 CD18 receptors and their 87:4150–4162. predictive capacity on the outcome of mastitis induced in dairy cows with Escherichia [22] Mehrzad J., Duchateau L., Burvenich C., coli, J. Dairy Sci. (1997) 80:67–74. High milk neutrophil chemiluminescence limits severity of bovine coliform mastitis, [31] Wang D., Pabst K.M., Aida Y., Pabst M.J., Vet. Res. (2005) 36:101–116. Lipopolysaccharide-inactivating activity of neutrophils is due to lactoferrin, J. Leukoc. [23] Munford R.S., Hall C.L., Detoxification of Biol. (1995) 57:865–874. bacterial lipopolysaccharides (endotoxins) by a human neutrophil enzyme, Science [32] Zimecki M., Spiegel K., Wlaszczyk A., (1986) 234:203–205. Kubler A., Kruzel M.L., Lactoferrin increases the output of neutrophil precursors [24] Paape M.J., Lilius E.M., Wiitanen P.A., and attenuates the spontaneous production Kontio M.P., Miller R.H., Intramammary de- of TNF-alpha and IL-6 by peripheral blood fense against infections by Escherichia coli, cells, Arch. Immunol. Ther. Exp. Warsz. Am. J. Vet. Res. (1996) 57:477–482. (1999) 47:113–118.