|
|
||||||||
Ghent University, Faculty of Veterinary Medicine Department of Physiology, Biochemistry and Biometrics Salisburylaan 133 9820 Merelbeke, Belgium
Corresponding author:
C. Burvenich; e-mail:
christian.burvenich{at}rug.ac.be.
| ABSTRACT |
|---|
|
|
|---|
Cells were obtained from the bone marrow of clinically healthy cows and separated into different stages of maturation using density gradient centrifugation. Three cellular fractions were obtained that were enriched for either early immature, late immature or mature granulocytic cells. Functions and receptor expressions assessed in the three maturation stages were: Fc-IgG2 receptor and CD11b expression, phagocytosis of Escherichia coli, respiratory burst activity, and cellular myeloperoxidase activity. Immature cells expressed already Fc-IgG2 receptor and CD11b on their cytoplasma membrane. Phagocytic ability was acquired in the myelocytic stage, but only the more mature forms were readily capable of phagocytosis. Promyelocytes, myelocytes and metamyelocytes showed no respiratory burst activity. Only band and segmented cells produced reactive oxygen species. Myeloperoxidase was present at all stages of maturity. Thus, each of the maturation stages was characterized by a selective expression of one or more functions and receptors. Therefore, sequential biochemical maturation is postulated during bovine granulopoiesis.
Key Words: bovine bone marrow functional maturation granulopoiesis
Abbreviation key: FBS = fetal bovine serum, FITC = fluorescein isothiocyanate, FS/SS = forward scatter side scatter, F = fluorescence, MFI = mean fluorescence intensity, MPO = myeloperoxidase, PI = propidium iodide, PMA = phorbol 12-myristate 13-acetate, PMN = polymorphonuclear leukocytes, ROS = reactive oxygen species
| INTRODUCTION |
|---|
|
|
|---|
The number of circulating polymorphonuclear leukocytes (PMN) is the result of influx and efflux of leukocytes. The efflux depends on the migration of the cells towards the marginated pool and towards tissues. Neutrophils do not recirculate. There is a unidirectional movement from the bone marrow into the blood, and from circulation towards the tissues. A well-controlled equilibrium exists between the circulating and the marginated pool in the blood (Figure 1
). Upon demand, a physiological mechanism appears for the rapid release of mature granulocytes into the circulation and accelerates the maturation of cells in the bone marrow (Moreira da Silva et al., 1994). When maturation is impaired, e.g. due to the increased proliferation rate, a higher number of immature cells will appear in circulation. This phenomenon occurs in periods of stress (e.g. parturition, mastitis).
|
Until now, bone marrow research in the bovine species was scarce. Recently, Hoeben et al. (1999) provided a first indication of the importance of the granulopoiesis by studying the effect of hormones and metabolites, of which the concentration changes dramatically during the periparturient period, on bovine bone marrow cells in vitro. Acetoacetic acid, ß-hydroxybutyric acid, hydrocortisone-21-acetate and bovine pregnancy-associated glycoprotein were found to have inhibitory effects on the proliferation of progenitor cells.
The objective of the present study was to investigate the maturation sequence of the functional properties of bovine granulocytes during proliferation and maturation in the bone marrow. Therefore, functions and receptors playing a crucial role in udder defense and found to be deficient during periparturient E. coli mastitis were studied. The Fc-IgG2 receptor and CD11b expression, the occurrence of phagocytosis, the respiratory burst activity, and the cellular myeloperoxidase activity were studied in three fractions of granulocytes with different maturation characteristics.
| MATERIALS AND METHODS |
|---|
|
|
|---|
Density Gradient Separation
Bone marrow cells were fractionated into different stages of maturation using a three step Percoll gradient originally described by Cowland and Borregaard (1999) for the human, and adapted for the bovine by Van Merris et al. (2001a). Briefly, the marrow suspension was layered on top of a discontinuous three-layer Percoll gradient with increasing densities, 1.030, 1.060, and 1.080 g/ml respectively. Following centrifugation, three cellular bands were selectively retrieved: fractions I, II, and III. To characterize each fraction, cell suspensions were submitted to a forward scatter/side scatter (FS/SS) dotplot analysis in the flow cytometer. Smears were analyzed visually under the light microscope. The fractions corresponded to three main maturation stages: early immature (fraction I), late immature (fraction II), and mature (fraction III). In each fraction, the viability of the cells was determined through the propidium-iodide (PI; Sigma Chemical Co, MO) exclusion test (Hoeben et al., 1997).
Fc-IgG2 Receptor Expression
Fc-IgG2 receptor expression was assessed by flow cytometry through direct labeling with fluorescein isothiocyanate (FITC) conjugated sheep anti-bovine IgG2 antibody (Serotec, Oxford, UK), as described by Worku et al. (1994). Each maturation fraction containing 1.106 cells in 100 µl staining buffer [PBS supplemented with 1% FBS and 0.09% (w/v) sodium azide] was incubated with a saturating amount of the antibody for 30 min at 4°C in the dark. Cells were washed twice with 300 µl staining buffer, pelleted by centrifugation (200 x g, 10 min, 4°C) and resuspended in 400 µl PBS. Subsequently, the fluorescence was measured by flow cytometry.
Fluorescence microscopy was also executed on the same samples for visualization of the Fc-IgG2 receptor expression.
CD11b Expression
CD11b expression was assessed by flow cytometry through an indirect labeling technique. Each maturation fraction was adjusted to 1.106 cells in 100 µl staining buffer. Cells were incubated with a saturating amount of mouse anti-bovine CD11b (Probio Ltd., Kent, England) for 30 min at room temperature. Cells were washed twice with 300 µl staining buffer and pelleted by centrifugation (200 x g, 10 min, 4°C). Pellets were incubated in the dark with a F(ab)2 goat anti-mouse IgG conjugated to FITC (Sigma) for 30 min at 4°C. Cells were washed twice with staining buffer, and resuspended in 400 µl PBS. This staining technique was previously used on circulating bovine PMN and resulted in the labeling of 95 to 99% of the cells (Dosogne et al., 1997).
For the determination of the total CD11b content, the primary bovine antibody was added both before and after the fixation and permeabilisation procedure (Lundahl et al., 1991), carried out with a Cytofix/Cytoperm fixation kit (PharMingen, San Diego). In this way, both cytoplasma membrane expression and intracellular CD11b content were measured.
Fluorescence microscopy was also executed on the same samples for visualization of the CD11b expression.
Phagocytosis of Escherichia coli
To test phagocytic capacity, the method described by Hoeben et al. (1997) was used. In this method, phagocytosis is tested without discrimination between attachment and ingestion of E. coli bacteria. Following isolation, cell suspensions were analyzed by flow cytometry FS/SS dotplot to assess the percentage of granulocytic cells within each fraction (Van Merris et al., 2001a). The three fractions were adjusted to 1.106 granulocytic cells in 100 µl HBSS (Gibco BRL). In chronological order, PI-labeled E. coli bacteria were added to the samples at a ratio of 1:25 and incubated in a shaking bath at 37°C with 5% autologous plasma. Phagocytosis was stopped after 20 min by addition of 0.1 mM N-ethylmaleimide (Sigma Chemical Co) in the same chronological order. Samples were further diluted with 400 µl PBS, stored for a few minutes on ice and immediately analyzed by flow cytometry. Using this method, both nonimmunological (i.e. mediated by complement, hydrophobic properties, or lectins) and immunological (i.e. antibody-dependent) phagocytosis was detected.
Fluorescence microscopy was also executed on the same samples to visualize the attachment and engulfment of E. coli bacteria.
Respiratory Burst Activity
The respiratory burst activity was measured as described by Smits et al. (1997), using dihydrorhodamine as substrate. Following isolation, cell suspensions were analyzed by flow cytometry in a FS/SS dotplot to assess the percentage of granulocytic cells within each fraction (Van Merris et al., 2001a). Subsequently, the three fractions were adjusted to 1.106 granulocytic cells in 100 µl HBSS and incubated with 5% autologous plasma and 10 µM dihydrorhodamine 123 (Molecular Probes, Leiden, The Netherlands) in a shaking bath at 37°C for 5 min. Samples were further diluted with 400 µl PBS and the spontaneous reactive oxygen species (ROS) production was measured by flow cytometry. The stimulated respiratory burst activity was determined by adding 10 µl 0.1 µM phorbol 12-myristate 13-acetate (PMA; Sigma Chemicals Co) during 15 min, following the first incubation.
Cellular Myeloperoxidase Activity
Myeloperoxidase (MPO; EC 1.11.1.7) positive granules were microscopically analyzed in each maturation fraction by staining the slides with benzidine dihydrochloride (Sigma Chemical Co) (Metcalf et al., 1986).
Flow Cytometric Analysis
Fluorescence was measured using a FACScan flow cytometer (Becton Dickinson Immunocytometry Systems, San Jose, California) equipped with a 488 nm argon-ion laser. Percent fluorescence (F) and mean fluorescence intensity (MFI) of selectively gated granulocytic cells in each maturation fraction (Van Merris et al., 2001a) were used to quantify the receptor expression and the functional activity of bone marrow cells. Percent fluorescence is defined as the percentage of granulocytic cells with a fluorescence higher than autofluorescence, attributed to respectively receptor expression (Fc-IgG2 receptor, and CD11b), engulfment of bacteria (phagocytosis) or ROS production (respiratory burst). MFI was used as an index for the average number of receptors expressed, bacteria ingested or ROS produced per cell. For CD11b expression, the MFI was corrected by subtraction of background fluorescence obtained using an isotype-matched antibody control.
Differential Counting and Microscopic Analysis
The distribution of granulocytic precursors in the three maturation fractions was assessed by differential counting of 100 cells in each fraction. Morphologic identification of bovine bone marrow cells was carried out according to Wilde (1964). Cell suspensions were adjusted to a concentration of 1 x 105 ml–1 in PBS with 10% FBS. For light microscopy, 200 µl aliquots were centrifuged onto FBS-coated slides in a cytospin (Shandon, Runcorn, UK). The slides were air dried at room temperature, fixed in methanol and stained with Hemacolor (Merck Diagnostica, Darmstadt, Germany). Fluorescence microscopy was used as a qualitative control for all properties measured with flow cytometry. Cell suspensions were fixed in 1% paraformaldehyde, incubated in the dark with a nuclear dye (DAPI; Molecular Probes) for 20 min and washed twice with PBS prior to cytocentrifugation.
Statistical Analysis
The analyses were computed using statistical software (Statistix 4.1 NH Analytical Software). The normality of the data was tested using the Wilk-Shapiro/rankit plot. The significance of differences in receptor expression or functional activity of the bone marrow cells in the three maturation stages (fraction I, II, III) was determined by a LSD (t) pairwise comparisons of means. Data were expressed as means ± SEM.
| RESULTS |
|---|
|
|
|---|
|
|
Results were obtained on eight cows and are summarized in Table 2
.
|
|
All results were obtained on nine cows and are summarized in Table 3
and Figure 3
.
|
Results were obtained on 11 cows and are summarized in Table 4
.
|
Following PMA stimulation, no differences in percent fluorescence could be detected in early and late immature fractions compared to the basal ROS production. However, the MFI increased three-fold in the late immature fraction following stimulation (fraction II, P < 0.001). On average 57.1 ± 2.70% fluorescence was detected in band and segmented cells (fraction III). The MFI showed a ten-fold increase in comparison to the spontaneous burst levels (P < 0.001).
All results were obtained on 11 cows and are summarized in Table 5
and Figure 5
.
|
|
| DISCUSSION |
|---|
|
|
|---|
The knowledge about the acquisition of functional properties during bovine granulopoiesis is based on studies in human bone marrow (Glasser and Fiederlein, 1987). There is only little information about the functional differentiation of bovine cells. A more in depth investigation of bovine bone marrow cells is therefore essential for the determination of the immune status in the bovine, because species variation in neutrophil function exists (Styrt, 1989).
The inflammatory reaction is initiated by extravasation of PMN (diapedesis). This is followed by a directed movement towards the site of infection (chemotaxis), where phagocytosis and killing of the pathogen will take place. Extravasation of activated neutrophils occurs after adhesion of the cells to the endothelial surface and is mediated by specific membrane adhesion molecules, e.g. the ß2-integrin CD11/CD18. The entire process of phagocytosis entails recognition and adhesion, followed by internalization of the pathogen that will be degraded through oxygen-dependent and -independent mechanisms. In order for phagocytosis to proceed, neutrophils must recognize the bacteria. Immunological recognition is mainly accomplished by IgG2 antibodies which recognize the bacteria through F(ab)-regions and bind to the neutrophils via Fc-receptors on their cytoplasma membrane. Activation of complement also promotes phagocytosis. Therefore, CD11b functions as a receptor for complement factor C3bi. Once complement and immunoglobulins bind to receptors on the neutrophil surface, the cell becomes activated and generates ROS. The respiratory burst system is very effective for eliminating Gram-negative bacteria such as E. coli (Burvenich et al., 1994). Moreira da Silva et al. (1998) observed a minimal respiratory burst of neutrophils three days after calving. This time period coincides with the maximum number of immature neutrophils found in circulation. The question arises whether the fall in the oxidative burst is the direct consequence of the appearance of immature cells in circulation.
Because neutrophils mainly constitute the first line defense in the udder, receptors and functions that play a role in the inflammatory reaction were studied during granulopoiesis: Fc-IgG2 receptor and CD11b expression, complement mediated phagocytosis of E. coli, the respiratory burst activity, and the cellular myeloperoxidase activity.
In the present study the Fc-IgG2 receptor was detected in bovine bone marrow as early as the myeloblasts and promyelocytes. The percent Fc-IgG2 positive cells increased with the stage of maturation. The MFI was highest in the mature fraction, the immature cells having fewer binding sites for IgG2. No comparative data for the bovine species are described in literature. Only one paper describes that the Fc-IgG2 receptor was present on bovine neonatal blood PMN and that its expression was reduced in comparison to circulating PMN isolated from adult cattle (Zwahlen et al., 1992). On human granulocytic precursors, Fc receptors were also found to be expressed early in differentiation, more specifically at the promyelocytic stage (Glasser and Fiederlein, 1987).
We demonstrated that CD11b surface antigen was expressed in all maturation stages of bovine bone marrow cells. The CD11b expression on the cytoplasma membrane increased significantly as precursor cells underwent granulocytic differentiation. However, in a comparable study on the characterization of bovine progenitors, CD11b was detected only on maturing cells, mostly myelocytes, metamyelocytes and segmented granulocytes (Fritsch et al., 1991). We propose that this discrepancy in results could be accounted for either by the difference in isolation procedure and/or the monoclonal antibody used. In humans, CD11b has been shown to be present in increasing amounts on cell surface starting from the myelocytic stage (Arnaout, 1990). Although Rosmarin et al. (1989) observed CD11b expression at low levels already in myeloblasts.
Additionally, we determined the total CD11b content in bovine bone marrow cells. In humans as well as in bovine, circulating PMN can increase their cytoplasma membrane expression of CD11b upon stimulation through mobilization of preformed CD11b, stored in specific granules. A significant intracellular CD11b pool was detected in bovine myelocytes, metamyelocytes, band and segmented cells.
The present results confirm that immature bovine cells do not exhibit a phagocytic ability. Progenitor cells started to attach and ingest E. coli bacteria at the myelocytic stage; only in band and segmented cells was detected a significant phagocytic capacity. Moreover, only 50% of these apparently mature granulocytes exhibited phagocytosis. This remarkable observation indicates that cellular characteristics required for optimal phagocytosis are acquired in conjunction with their release from the bone marrow (Altman and Stossel, 1974). Although CD11b is already expressed early in maturation, its presence does not guarantee complement mediated phagocytosis by immature bovine bone marrow cells. According to Lichtman and Weed (1972), immature cells may be less capable of forming pseudopods following phagocytic activation. Our findings are in general agreement with the observations of Silva et al. (1989), who studied immune phagocytosis in bone marrow cells of cows. However, Guidry and Paape (1976), inducing a left-shift in bovine following intramammary E. coli endotoxin infusion, observed that the immature peripheral granulocytes are equally phagocytic as their mature counterparts. Decreased phagocytosis as compared to circulating PMN has also been demonstrated in human immature as well as morphologically mature bone marrow cells (Altman and Stossel, 1974; Glasser and Fiederlein, 1987).
Assessment of the respiratory burst activity demonstrated that immature cells isolated from bovine bone marrow were not capable to produce ROS, neither spontaneously, nor after stimulation. Following PMA-stimulation, myelocytes and metamyelocytes showed a significant respiratory burst activity, whereas they did not exhibit any spontaneous ROS production. The band and segmented cells on the contrary had a spontaneous respiratory burst activity and PMA induced a significant increase in ROS production.
Two enzymes play a keyrole in the ROS production: myeloperoxidase and the membrane-bound NADPH-oxidase. Myeloperoxidase is found exclusively in azurophilic granules and is an abundant early myeloid protein (Rosmarin, 1989). Also in the present study on bovine bone marrow cells, MPO positive granules were found at all maturation stages, starting with the promyelocyte. Following our results, we suggest that only from the metamyelocytic stage, bovine granulocytic cells dispose of a functionally oxidative burst. Comparative data for the bovine species are scarce. In a study conducted by Silva et al. (1989), the nitroblue tetrazolium reduction, an alternative parameter for ROS production, was only observed in segmented granulocytes from the bovine bone marrow. Also in human bone marrow, only band and segmented cells showed a substantial oxygen-dependent microbial killing using nitroblue tetrazolium reduction (Glasser and Fiederlein, 1987). The minor differences in the data could be attributed to the different techniques used for assessing the ROS production.
| CONCLUSION |
|---|
|
|
|---|
Fc-IgG2 receptor expression
CD11b expression
complement mediated phagocytosis
respiratory burst activity (Figure 6
|
|
| ACKNOWLEDGEMENTS |
|---|
|
|
|---|
Received for publication February 5, 2002. Accepted for publication May 9, 2002.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
J. Detilleux, F. Vangroenweghe, and C. Burvenich Mathematical Model of the Acute Inflammatory Response to Escherichia coli in Intramammary Challenge. J Dairy Sci, September 1, 2006; 89(9): 3455 - 3465. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Mehrzad, L. Duchateau, and C. Burvenich Viability of Milk Neutrophils and Severity of Bovine Coliform Mastitis J Dairy Sci, December 1, 2004; 87(12): 4150 - 4162. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. Monfardini, V. Van Merris, M. Paape, L. Duchateau, and C. Burvenich L-Selectin and Chemotaxis Throughout Bone Marrow Granulocyte Maturation in the Bovine J Dairy Sci, October 1, 2004; 87(10): 3350 - 3357. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Diez-Fraile, E. Meyer, L. Duchateau, and C. Burvenich L-Selectin and {beta}2-Integrin Expression on Circulating Bovine Polymorphonuclear Leukocytes During Endotoxin Mastitis J Dairy Sci, July 1, 2003; 86(7): 2334 - 2342. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |