a detailed study of apoptosis

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

1. Introduction

Apoptosis is an evolutionarily conserved, highly regulated process of programmed, active, cell death, morphologically and biochemically different from necrosis, and is important in normal development and physiological homeostasis of multicellular organisms [1, 2 and 3]. Cells dying by apoptosis maintain membrane integrity until late in the process but display several morphological and biochemical alterations, including chromatin condensation, nuclear segmentation, internucleosomal DNA fragmentation, cytoplasmic vacuolisation, cell shrinkage and membrane blebbing with shedding of apoptotic bodies [4 and 5]. Since dying apoptotic cells and apoptotic bodies are rapidly phagocytosed by neighboring cells, mainly macrophages, before leakage of the cellular contents, this cell death process does not usually result in an inflammatory response [5 and 6]. In contrast, necrosis is an accidental form of cell death, resulting from physically or chemically induced membrane damage and results in swollen cells that leak their cytoplasmic contents. This leakage usually induces an inflammatory response [5].

Many bacterial pathogens are able to manipulate for their own benefit the host apoptotic programme [7]. Some obligate intracellular pathogens, such as Brucella suis, Rickettsia rickettsii and Chlamydiae, can exert an anti-apoptotic effect upon host cells, so that these remain as a site for their growth and multiplication [8, 9 and 10]. Pathogen-induced activation of the host cell-death pathway may, in turn, serve to eliminate key immune cells, evading host defenses that otherwise would act to limit infection [11]. Induction of host cell apoptosis by bacteria has been suggested as an important factor in the pathogenesis of some infections [7 and 11], since elimination of immune cells can facilitate invasion and spreading of the infectious agent.

Photobacterium damselae subsp. piscicida (previously Pasteurella piscicida) is the causative agent of fish pasteurellosis, a serious bacterial disease affecting several economically important marine fish species such as yellowtail, gilthead seabream, striped jack and sea bass [12 and 13]. Different studies have identified several factors/mechanisms important in the virulence of Ph. damselae subsp. piscicida (see [14]). Despite these studies, information on the pathogenesis of pasteurellosis in sea bass is still scarce and the interaction of the bacteria with the host phagocytes is not sufficiently understood.

In this study, we investigated the interaction of Ph. damselae subsp. piscicida with sea bass phagocytes in vitro and in fish experimentally infected by the intraperitoneal (i.p.) route. Here, we report on the occurrence of apoptosis of sea bass macrophages and neutrophils as a result of the host-pathogen interaction in vivo and discuss the possible role of the Ph. damselae-induced apoptosis in the pathogenesis of fish pasteurellosis.

2. Materials and methods

2.1. Experimental fish

Sexually immature sea bass were purchased from a commercial fish farm. Fish weighing 16.5±2.9 g were used for the determination of the virulence of Ph. damselae subsp. piscicida isolates. Fish weighing 113.6±23.9 g were used for all the other experiments because they allowed the collection of a larger number of peritoneal cells. The animals were maintained in recirculating aerated seawater, at 18-19 °C. The fish used in experiments with live bacteria were previously acclimatized to 23-24 °C. Water quality was maintained with mechanical and biological filtration and fish were fed ad libitum on commercial pellets. Only healthy fish, as indicated by their activity and exterior appearance, were used in the experiments.

2.2. Bacteria

The origin and virulence of the Ph. damselae subsp. piscicida strains used in this study are listed in Table 1. Strains DI 21, B51 and EPOY 8803-II were provided by Professor Alicia E. Toranzo (Departamento de Microbiología y Parasitología, Facultad de Biologia, University of Santiago de Compostela, Spain), strains MT1415, PP3, MT1375, MT1588 and MT1594 were provided by Dr Andrew C. Barnes (Marine Laboratory, Aberdeen, UK) and strain PTAVSA95 was supplied by Dr Nuno M.S. dos Santos (Institute for Molecular and Cell Biology, University of Porto, Portugal). The strain ATCC 29690 was obtained from the American Type Culture Collection, USA. The virulence of each isolate was determined by injecting groups of 12-16 sea bass (mean weight 16.5±2.9 g) with different doses of bacteria. Virulence was defined according to previous studies [15]. All the virulent strains used in this work killed 100% of the fish when 107CFU were injected i.p. In contrast, less than 50% mortality was recorded when 108CFU of the non-virulent strains were injected.

Table 1. Origin and virulence of the Ph. damselae subsp. piscicida isolates used in the present study

Bacteria were routinely cultured at 22 °C in tryptic soy broth (TSB) or tryptic soy agar (TSA) (both from Difco Laboratories, Detroit, MI, USA) supplemented with NaCl to a final concentration of 1% (w/v) (TSB-1 and TSA-1, respectively) and were stored frozen at −70 °C in TSB-1 supplemented with 15% (v/v) glycerol.

To prepare the inocula for injection into the peritoneal cavities of fish, the stocked bacteria were cultured for 48 h at 22 °C on TSA-1 and then inoculated into TSB-1 and cultured overnight at the same temperature, with shaking (100 rpm). Exponentially growing bacteria were collected by centrifugation at 3000-g for 30 min, washed once and resuspended in phosphate buffered saline (PBS) at the indicated concentrations. After anesthetization with 0.03% (v/v) ethylene glycol monophenyl ether (Merck, Darmstadt, Germany), groups of sea bass were injected i.p. with 100 μl of the bacterial suspensions. Plating serial dilutions of the suspensions onto TSA-1 plates and counting the number of CFU following incubation at 22 °C confirmed bacterial concentrations of the inocula. In some experiments, UV-killed bacterial cells were used. For UV-treatment, exponentially growing bacteria collected as described previously and resuspended in PBS were treated with UV-light for 1 h. The non-viability of the bacteria was confirmed by inoculation of TSA-1 plates.

2.3. Bacterial culture supernatants

For preparing culture supernatants, bacterial strains were cultured overnight in TSB-1 at 22 °C with shaking (100 rpm) and sub-cultured at a dilution of 1:100 in the same conditions, until the middle exponential phase (OD of 0.650 at 600 nm). The bacteria were removed by centrifugation (3000-g, 30 min, 4 °C) and the culture supernatants were collected, filtered through a 0.22 μm-pore-size filter (Schleicher and Shuell, Dassel, Germany) and concentrated approximately 100-fold using a Vivaflow 200 concentrator (Sartorius AG, Goettingen, Germany). Concentrated supernatants were dialysed against 20 mM Tris-HCl (pH 8.0) and were then aliquoted and stored at −70 °C until use. Protein concentrations were determined using a Micro BCA Protein Assay Reagent Kit (Pierce, Rockford, IL, USA). The sterility of the culture filtrates was confirmed by the absence of colonies after plating on TSA-1 plates. Concentrated supernatants were diluted in PBS to the indicated concentrations prior to the i.p. injection of 100 μl per fish. In some experiments, bacterial culture supernatants diluted in PBS were boiled for 10 min before testing.

2.4. Collection of peritoneal leukocytes

The peritoneal cells were collected from undisturbed peritoneal cavities (control) or at the indicated times after injection of bacteria or culture supernatants, by a procedure described in detail elsewhere [16]. Briefly, after being killed by an overexposure to ethylene glycol monophenyl ether (Merck, Darmstadt, Germany) at a concentration of 0.06% (v/v), the animals were exsanguinated by cutting the ventral aorta, and the abdominal side of the fish was then cleaned with ethanol. PBS with osmotic strength adjusted to 355 mOsm and supplemented with 20 U heparin ml−1and 5% (w/v) glucose was injected into the peritoneal cavity (5 ml per fish). PBS containing the peritoneal cells was then collected and placed on ice until processed, as follows.

2.5. Light microscopy

Total peritoneal cell counts were performed with a hemocytometer and cytospin preparations were made using a Shandon Cytospin 2 apparatus. The cytospins were fixed with formol-ethanol (10% of 37% formaldehyde in absolute ethanol) for 45 s and stained with Wright's stain (Haemacolor, Merck, Darmstadt, Germany). Detection of peroxidase activity to label neutrophils [17] was done by the Antonow's technique [18]. The macrophages and neutrophils in the peritoneal exudates were differentially counted and the percentage of both cell types established after counting a minimum of 300 cells per slide. Additionally, where indicated, the percentages of apoptotic macrophages and neutrophils were determined.

2.6. TUNEL staining

DNA fragmentation in individual cells was detected by terminal deoxynucleotidyltransferase-mediated dUTP nick end labeling (TUNEL) of DNA strand breaks (In Situ Cell Death Detection Kit, Fluorescein; Roche Diagnostics, Mannheim, Germany). Since preliminary observations showed that the TUNEL staining could be done in cytospins processed for peroxidase detection by the Antonow method, cytospin preparations fixed with formol-ethanol and stained for peroxidase, as described above, were subjected to the TUNEL reaction according to the manufacturer's instructions. The cytospin preparations were mounted with the anti-fading agent Vectashield (Vector Laboratories, Peterborough, UK) containing 4 μg ml−1propidium iodide and were observed under a Zeiss Axioskop epifluorescence microscope equipped with an HBO-100 mercury lamp, filter set 40 (BP360/51, BP485/17, BP560/18) from Zeiss, excitation filter BP450-490, beam splitter FT510 and emission filter LP520. Apoptotic cells were identified by yellow-green fluorescence within the nucleus, due to the fluorescein-dUTP incorporated at the 3′OH ends of fragmented DNA. Images were acquired with a Spot 2 camera (Diagnostics Instruments), recording separately for each microscopic field the images of peroxidase staining (bright field) and TUNEL (fluorescence). Comparison between prints of bright field and fluorescence images of the same cells allowed the identification of four leukocyte categories: cells positive for both peroxidase and TUNEL, cells negative for both stainings and cells positive for only peroxidase or TUNEL. TUNEL-positive cells with abundant peroxidase-positive cytoplasmic granules were counted as apoptotic neutrophils. Apoptotic macrophages were identified as large mononuclear, peroxidase-negative, TUNEL-positive, leukocytes.

For detection of apoptosis in tissue sections, sea bass head-kidneys were fixed in 10% (v/v) buffered neutral formalin, embedded in paraffin and sectioned. Sections were then processed for TUNEL staining, following the manufacturer's instructions for staining of sections.

2.7. Electron microscopy

The ultrastructure of peritoneal leukocytes was examined by transmission electron microscopy. Peritoneal exudates were fixed in 2.5% (v/v) glutaraldehyde in cacodylate buffer (0.1 M, pH 7.2) and post-fixed in 1% (w/v) osmium tetroxide in the same buffer. The samples were then embedded in Epon resin (TAAB Laboratories Equipment Ltd, Berkshire, UK) after dehydration in a graded series of ethanol. Ultrathin sections were cut with an RMC MT-7 microtome and contrasted with uranyl acetate and lead citrate as described [19]. Observations and micrographs were done under a Zeiss EM10 electron microscope.

2.8. Analysis of DNA fragmentation by gel electrophoresis

Endonuclease-mediated internucleosomal DNA fragmentation is a characteristic feature of apoptosis [20] that can be detected by the occurrence of a ladder pattern in agarose gel electrophoresis of DNA extracted from apoptotic cells.

Peritoneal leukocytes (2-106) collected as described above were centrifuged at 350-g for 5 min at 4 °C. The supernatants were discarded and the cell pellets were stored frozen at −20 °C until processed for extraction of DNA following standard procedures [21]. The DNA was incubated for 2 h at 37 °C with 20 μg ml−1RNAse, before being electrophoresed in a 1.5% (w/v) agarose gel containing 0.5 μg ml−1ethidium bromide to allow visualization of DNA. A 100 base-pair DNA ladder marker (Amersham Pharmacia Biotech Inc., Piscataway, NJ, USA) was always included as a reference. Gels were photographed under UV illumination using a Polaroid camera and the photographs were digitalized and adjusted for brightness and contrast using Adobe Photoshop.

2.9. In vitro macrophage assays

Monolayers of sea bass macrophages were obtained essentially as described [22]. Briefly, the head-kidney was removed and pushed through a 100 μm nylon mesh with L-15 medium (Gibco, Paisley, UK) containing 2% fetal bovine serum (FBS, Gibco, Paisley, UK), 1% penicillin/streptomycin (P/S, Gibco, Paisley, UK) and 10 U/ml heparin. The cell suspension was layered onto a 31-45% Percoll density gradient and, following centrifugation at 400-g for 30 min at 4 °C, the band of cells layering above the 31-45% interface was collected and washed with L-15 containing 0.1% FBS. The cell suspension in L-15 medium containing 0.1% FBS and 1% P/S was adjusted to 107cells ml−1, and 1 ml was added to each well of a 24-well culture plate (Nunc, Denmark) containing a Thermanox coverslip (Nunc, Naperville, IL, USA). The plates were incubated at 20 °C and after 24 h, the non-adherent cells were removed by washing twice with L-15 medium and the remaining monolayer was fed with 1 ml L-15 medium supplemented with 5% FBS and 1% P/S and maintained at 20 °C for 2-3 days before use. Adherent cells were >94% of the monocyte/macrophage lineage, according to morphological characteristics observed after Wright staining (Haemacolor, Merck, Darmstadt, Germany), and peroxidase detection [17 and 18]. The apoptotic activity of virulent bacterial culture supernatants was assessed by incubating macrophage monolayers with 0.1, 1 or 10 μl ml−1of MT1415 concentrated supernatants prepared as described above and diluted in L-15 medium containing 0.1% FBS and 1% P/S. Mock-treated wells and wells treated with the concentrated non-virulent ATCC 29690 bacterial supernatant were used as controls. After 3, 6 and 24 h, coverslips were removed (in triplicate), fixed with 10% formalin in absolute ethanol, stained with Wright Stain (Haemacolor, Merck, Darmstadt, Germany) and observed under the light microscope for the presence of cells with apoptotic morphology.

2.10. Ex vivo peritoneal exudate assays

For ex vivo assays, the sea bass peritoneal exudate cells were collected from control sea bass essentially as above, with minor changes. The collection of the peritoneal cells was performed in sterile conditions using L-15 medium supplemented with 10% FBS, 1% P/S and 30 U/ml heparin instead of PBS. The cell density was determined as above. The collected cell suspensions were placed into a 10 ml tube and incubated at 22 °C, with gentle shaking. Exudate cell suspensions were infected with exponentially growing DI21 virulent bacterial cells at a bacteria:leukocyte ratio of 10:1 or 100:1 or treated with 10 μl ml−1virulent (DI21, B51) and non-virulent (EPOY 8803-II) Ph. damselae subsp. piscicida concentrated supernatants. After 6 h of incubation for the experiments with bacterial cells and after 6 or 22 h with bacterial supernatants, aliquots of the cell suspensions were collected and analyzed for the presence of apoptotic cells by light microscopic techniques as described earlier.

2.11. Reproducibility of results

For each experimental situation, at least six fish were used. Where indicated, statistical analysis of the data was done using the Student's t-test. P<0.05 was considered significant.

3. Results

Preliminary observations showed abundant cells with a morphology suggestive of apoptosis in samples of peritoneal exudates and head-kidney from sea bass dying after i.p. injection of lethal doses of Ph. damselae subsp. piscicida virulent strains. That such a morphology was indeed apoptotic was supported by the observation of abundant TUNEL-positive cells both in peritoneal exudates and in head-kidney. These observations prompted us to study in more detail the apoptotic process using the peritoneal model of infection, which allows a precise cytological analysis [16].

3.1. Virulent Ph. damselae subsp. piscicida cells induce apoptosis of sea bass peritoneal macrophages and neutrophils in vivo

Following the i.p. injection of 106-108CFU of the virulent strain MT1415, macrophages and neutrophils with apoptotic morphological alterations were seen by conventional light microscopy, starting at 6 h post-inoculation (Fig. 1a). The apoptotic nature of these alterations was confirmed by TUNEL staining (Fig. 2a-c), electron microscopy (Fig. 3 (a), Fig. 3 (b) and Fig. 3 (c)) and DNA electrophoresis (Fig. 4a, lanes 1-3). The analysis of fluorescence and bright field images of the same cells in peritoneal samples processed for peroxidase detection and TUNEL staining (as documented below; see Fig. 5a,b), showed that the apoptotic process preferentially affects macrophages and neutrophils, with only a small percentage of apoptotic small mononuclear leukocytes (thrombocytes and lymphocytes) and no apoptotic eosinophilic granular cells (EGCs) being present. The percentage of apoptotic cells was found to correlate positively with the number of bacteria used for infection (data not shown). At 6 h after the i.p. injection, apoptotic macrophages were much more abundant than apoptotic neutrophils, a difference that was attenuated at subsequent times. After i.p. injection of 107CFU of the virulent strain MT1415, most of the fish died within 48 h and some were already moribund at 18 h. Extensive apoptosis of peritoneal macrophages and neutrophils occurred in all the studied moribund fish (Fig. 2c). Electron microscopy of peritoneal samples collected from fish in advanced stages of infection, showed free and intracellular bacteria, and extensive lysis of the two types of phagocytic cells and of apoptotic bodies ( Fig. 3 (a), Fig. 3 (b) and Fig. 3 (c)). Some of the lysing phagocytes (identified by the presence of intracellular bacteria) showed ultrastructural alterations typical of apoptosis ( Fig. 3 (a), Fig. 3 (b) and Fig. 3 (c)).

Full-size image(58K)

Fig. 1. Light microscopic images of sea bass peritoneal exudates stained for the detection of peroxidase and counterstained by the Wright technique. (a) Sample collected 6 h after the i.p. injection of 107CFU of the virulent strain MT1415 of Ph. damselae subsp. piscicida. Note peroxidase-positive and peroxidase-negative cells with alterations typical of apoptosis. (b) Sample collected 24 h after the i.p. injection of 107CFU of the non-virulent strain EPOY 8803-II. Note the presence of numerous neutrophils (identified by the peroxidase activity) indicating the occurrence of an inflammatory reaction. No cell with apoptotic morphology is present.

Full-size image(35K)

Fig. 2. Fluorescence microscopy of peritoneal exudates processed for the detection of DNA fragmentation by TUNEL staining. (a-c) Samples collected from sea bass injected i.p. with 107CFU of the virulent strain MT1415 of Ph. damselae subsp. piscicida, showing abundant TUNEL-positive cells. (a) After 6 h; (b) after 24 h; (c) from a moribund fish, 18 h after injection. (d) Sample collected 48 h after the i.p. injection of 107CFU of the non-virulent strain EPOY 8803-II. Only a very small number of cells are TUNEL-positive.

Full-size image(49K)

Fig. 3 (a).

Full-size image(47K)

Fig. 3 (b).

Full-size image(53K)

Fig. 3 (c). Fig. 3. Transmission electron microscopy of peritoneal exudates of sea bass collected after the i.p. injection of 107CFU of the virulent strain MT1415 of Ph. damselae subsp. pisicida. Sections contrasted with uranyl acetate and lead citrate. Bar=2 μm. (a) Sample collected after 6 h. Note the leukocytes with apoptotic alterations including cell shrinkage, chromatin condensation, nuclear fragmentation and cell blebbing. The macrophage on the upper right corner has intracellular apoptotic bodies. (b) Sample collected after 12 h. Note the leukocytes in advanced apoptosis with condensed chromatin. Arrows indicate bacteria inside apoptotic phagocytes. (c) Sample collected after 12 h. Note the lytic alterations in most leukocytes. Arrows indicate bacteria inside an apoptotic phagocyte.

Full-size image(32K)

Fig. 4. Agarose gel electrophoresis of DNA extracted from peritoneal leukocytes of sea bass. (a) DNA extracted from peritoneal leukocytes collected after the i.p. injection of 107CFU of Ph. damselae subsp. piscicida MT1415 (lanes 1-3), EPOY 8803-II (lanes 4-6) and ATCC 29690 (lanes 7-9) strains. Lanes 1, 4 and 7-after 6 h; lanes 2, 5 and 8-after 24 h; lanes 6 and 9-after 48 h; lane 3-from a moribund fish 18 h after injection; M-100 base-pair ladder marker. (b) DNA extracted from peritoneal leukocytes collected after the i.p. injection of concentrated culture supernatants from the Ph. damselae subsp. piscicida strains MT1415 (lanes 1-6) and ATCC 29690 (lanes 7-9). Lanes 1 to 3-after injection of 100 μl of PBS with 0.1 μl concentrated culture supernatant (containing approximately 0.7 μg of protein) from the strain MT1415; lanes 4 to 6-after injection of 100 μl PBS with 1 μl of MT1415 concentrated culture supernatant (containing approximately 7 μg of protein); lanes 7 to 9-after injection of 1 μl, in 100 μl PBS, of ATCC 29690 concentrated culture supernatants (containing approximately 7 μg of protein); lanes 1, 4 and 7-after 6 h; lanes 2, 5 and 8-after 12 h; lanes 3, 6 and 9-after 24 h; lane 10-DNA from peritoneal leukocytes of a control fish; M-100 base-pair ladder marker.

Full-size image(74K)

Fig. 5. Peritoneal exudate collected from a sea bass 6 h after the i.p. injection of 100 μl of PBS with 0.1 μl of concentrated culture supernatant (containing approximately 0.7 μg of total protein) of the virulent strain MT1415. Cytospin processed for the simultaneous detection of DNA fragmentation (TUNEL-staining) and peroxidase activity (Antonow technique). (a) Fluorescence microscopy. (b) Bright field of the same microscopic field. Note, in (a), the high proportion of TUNEL-positive cells. TUNEL and peroxidase-positive cells are marked with white asterisks. TUNEL-positive, peroxidase-negative cells are marked with black asterisks. The analysis of (b) shows that most of the TUNEL-positive cells are large mononuclear, peroxidase-negative cells, that is, macrophages. Some cells positive for both TUNEL and peroxidase, that is, apoptotic neutrophils, can also be observed.

Besides strain MT1415, all the other virulent Ph. damselae subsp. piscicida strains listed in Table 1 induced apoptosis of peritoneal macrophages and neutrophils when 106-107CFU were injected i.p.

No increase in the very low number of apoptotic cells usually present in control peritoneal exudates was detected by TUNEL staining either at 6, 24, 48 h or 4 days after the i.p. injection of 107CFU of the non-virulent Ph. damselae subsp. piscicida strains ATCC 29690 and EPOY 8803-II (Fig. 2d). Electron microscopy and DNA electrophoresis ( Fig. 4a, lanes 4-9) confirmed this observation. In the peritoneal cavities of these fish, a transient inflammatory response with significant neutrophil influx was seen ( Fig. 1b).

3.2. In vivo apoptosis of sea bass peritoneal macrophages and neutrophils is induced by Ph. damselae subsp. piscicida culture supernatants but not by UV-killed bacteria

In order to characterize the components of Ph. damselae subsp. piscicida involved in the induction of apoptosis of sea bass phagocytes, we tested UV-killed bacterial cells and bacterial culture supernatants for the ability to induce macrophage and neutrophil apoptosis in vivo. No increase over the control levels of apoptosis could be observed in peritoneal exudates collected 6 h after the i.p. injection of 108UV-killed virulent bacterial cells (data not shown). In contrast, extensive apoptosis of macrophages and neutrophils was detected, in a dose-dependent way, in exudates collected 6, 12 and 24 h after i.p. injection of culture supernatants of virulent bacterial cultures (strains MT1415, B51 and DI21). The apoptotic process induced by the injection of 100 μl PBS with 0.1 or 1 μl concentrated supernatants, corresponding, respectively, to about 0.7 or 7 μg protein and to 10 or 100 μl of the initial culture supernatants, was characterized by TUNEL analysis (Fig. 5a), electron microscopy and DNA gel electrophoresis ( Fig. 4b; lanes 1-6). As in the case of the i.p. injection of live virulent Ph. damselae subsp. piscicida cells, the analysis of bright field and fluorescence images of the same cells in peritoneal exudates processed for TUNEL staining and peroxidase detection (Fig. 5a,b, respectively) showed that only a small percentage of thrombocytes and lymphocytes and no EGCs were TUNEL-positive, indicating that the apoptotic process induced by the injection of virulent culture supernatants affects preferentially macrophages and neutrophils. Supernatants from non-virulent bacterial cultures, containing amounts of total protein similar to those of virulent supernatants, were not able to induce apoptosis of sea bass peritoneal phagocytes, as shown by DNA electrophoresis ( Fig. 4b, lanes 7-9) and TUNEL staining (not shown).

The apoptogenic activity of culture supernatants from the virulent strains tested (MT1415, B51 and DI21) was abolished by boiling for 10 min, as assessed by TUNEL staining and DNA electrophoresis of cells harvested 6 h after injection (not shown).

A more detailed study on the evolution of the apoptotic process induced by 1 μl of concentrated culture supernatant of strain MT1415 was carried out. This study showed that following the injection of the supernatant there is an initial continuous increase of apoptotic macrophages accompanied by a continuous decrease of non-apoptotic macrophages so that at 6 h post-injection, apoptotic macrophages represented 61.9±31.0% of total peritoneal macrophages; this proportion increased to 86.1±36.1% at 24 h, when 1.2±0.5-107apoptotic macrophages per peritoneal cavity were present. As a consequence of this process, the number of non-apoptotic macrophages during the initial 24 h was always below the total number at time zero and below the number in fish injected with the non-virulent supernatants.

On the other hand, the number of apoptotic neutrophils also increased continuously during the initial 24 h, but at a slower pace as compared to the macrophages. At 6 h post-injection only 18.7±12.4% of the peritoneal neutrophils were apoptotic, and apoptotic macrophages were 10.8±6.5 times more abundant than apoptotic neutrophils (P=0.004), but at 24 h 0.9±0.4-107neutrophils were apoptotic, which represents 66.8±34.2% of total neutrophils.

The further analysis of the evolution of the process showed that in the peritoneal exudates of fish that were moribund as a consequence of the injection of 1 μl of the concentrated virulent supernatant, most of the macrophages and neutrophils were apoptotic and lysing. In contrast, the surviving fish showed a progressive control of the apoptotic process, almost no apoptotic phagocytes being present after 4 days post-injection. In these fish, 15 days after the injection the numbers of peritoneal macrophages and neutrophils were decreasing and approaching the values of the controls.

3.3. Ph. damselae subsp. piscicida do not induce apoptosis of sea bass peritoneal phagocytes in vitro

The number of apoptotic macrophages in in vitro head-kidney macrophage cultures or of apoptotic neutrophils and macrophages in peritoneal exudates incubated ex vivo, treated with virulent supernatants or virulent bacteria, under the described conditions, was low and similar to that in untreated samples or in samples treated with avirulent supernatants (data not shown). In the ex vivo experiments with Ph. damselae subsp. piscicida, phagocytosis of the bacteria by macrophages and neutrophils was seen.

4. Discussion

Bacterial pathogens have developed different strategies to survive inside the host and to overcome its natural defenses, and thus cause disease. One of these strategies is the utilization of mechanisms that rapidly and efficiently kill host cells involved in antimicrobial defense. In the case of bacteria that proliferate extracellularly, destruction of host phagocytes by the invading microorganism will deprive the infected host of the crucial defense mechanism represented by phagocytosis. In many organisms, including fish, the phagocytic system consists of two types of specialized leukocytes. One is the fixed, long-living, resident macrophage or macrophage-like cell, present in all compartments of the body, which, therefore, plays a very important role in the initial stages of infection, as it is the first phagocyte to encounter invading microorganisms. The other professional phagocyte is the neutrophil, a leukocyte armed with potent antimicrobial molecules that is rapidly recruited to the infectious foci from reserve pools in the hemopoietic organs and blood and that, consequently, also is important in the phagocytic defense mechanism.

Zychlinsky et al. [23] were the first to show that macrophages can undergo apoptosis in vitro as a result of infection. Since then, several authors have reported on the triggering of apoptosis of host immune cells by several bacterial species and proposed this as a pathogenic mechanism (see reviews in [7 and 11]). Almost all of the reported instances of microbial-induced apoptotic death of host phagocytes are in vitro studies with single cell populations. Most concern macrophages [24, 25, 26, 27, 28, 29, 30, 31, 32, 33 and 34], a few refer to neutrophils [31, 35 and 36] but none describes the simultaneous destruction of macrophages and neutrophils.

The present paper describes the occurrence of extensive apoptosis in the peritoneal cavity and in the head-kidney of sea bass experimentally infected with Ph. damselae subsp. piscicida. Further analysis of the apoptotic process, using the peritoneal model, showed for the first time the occurrence of extensive apoptosis of neutrophils and macrophages during an infection by a microbial pathogen. We found that virulent, but not non-virulent, isolates of Ph. damselae subsp. piscicida induce, in experimentally infected sea bass, apoptosis of macrophages and neutrophils. The occurrence of such a process was detected by light and electron microscopic observation of peritoneal neutrophils and macrophages with morphological features of apoptotic cells, by specific in situ detection of DNA fragmentation in peritoneal cells by TUNEL staining, and by the occurrence of a characteristic ladder pattern after agarose gel electrophoresis of DNA extracted from peritoneal leukocytes.

Our present results do not identify the bacterial factor(s) responsible for the induction of sea bass macrophage and neutrophil apoptosis by Ph. damselae subsp. piscicida. Since UV-killed virulent bacterial cells do not trigger the apoptotic process, secreted bacterial products emerge as the likely candidates for the apoptogenic factor. In support of this interpretation is the observation that extensive apoptosis of macrophages and neutrophils was observed in peritoneal exudates after injection of virulent, but not of non-virulent, bacterial culture supernatants. Since the apoptogenic activity of culture supernatants was abolished by heat-treatment, our data are consistent with the secretion of one (or several) protein(s) by virulent isolates of Ph. damselae subsp. piscicida, with apoptogenic activity towards macrophages and neutrophils in experimentally infected sea bass.

The percentage of macrophages undergoing apoptosis early after the onset of the process was found to be greater than that of neutrophils. This result most probably reflects the predominance of macrophages (5.7±3.4-107per fish) over neutrophils (0.69±0.37-107per fish) in the resting peritoneal cavity when the supernatant was injected. Moreover, the kinetics of the apoptotic process affecting neutrophils was significantly affected by the inflammatory response to the i.p. injection of the supernatants. It is widely documented that the early response to a peritoneal aggression is characterized by a quick and extensive influx of neutrophils, while the macrophage response is more delayed [17 and 18]. It follows, therefore, that the high numbers of incoming neutrophils during the initial 24 h of the process significantly increase the counts of non-apoptotic neutrophils. In contrast, a similar situation does not occur with macrophages.

The results regarding the apoptotic process that follows the i.p. injection of lethal doses of Ph. damselae subsp. piscicida cells or culture supernatants show that both peritoneal macrophages and neutrophils are extensively affected, the majority of macrophages and neutrophils being apoptotic 24 h after injection and reaching values around 107per peritoneal cavity for both cell types. Moreover, it is relevant that in those situations the number of non-apoptotic macrophages was always below the number in the resting peritoneal cavity. On the other hand, electron microscopy showed that at the late stages of this process, extensive lysis of macrophages, neutrophils and apoptotic bodies occurs. One relevant consequence of the process is the severe depletion of functional phagocytes with the failure of the antibacterial phagocytic defenses. Virulent Ph. damselae subsp. piscicida cells, although able to invade sea bass epithelial cells, multiply extracellularly in the infected fish [37]. We have recently shown that peritoneal neutrophils and macrophages of sea bass extensively phagocytose i.p. injected Ph. damselae subsp. piscicida [17]. On the other hand, it has been demonstrated that this fish pathogen is killed by sea bass macrophages in vitro, superoxide anions being implicated in the killing process [38]. The susceptibility of these bacteria to the macrophage killing activity is supported by the results of Barnes et al. [39] showing that this pathogen is not well equipped for the protection against attack by the reactive oxygen species produced by macrophages and neutrophils [39]. It is, therefore, likely that extensive destruction of sea bass phagocytes by the apoptogenic activity of virulent Ph. damselae subsp. piscicida would be an important factor in the pathogenesis of sea bass pasteurellosis by preventing phagocytosis and killing of the pathogen and by releasing bacteria already inside the phagocytes.

The simultaneous destruction of sea bass neutrophils and macrophages by Ph. damselae subsp. piscicida has another consequence advantageous to the pathogen. Indeed, macrophages, besides having an efficient antibacterial phagocytic activity, also are the scavenger cells with the critical role of eliminating the potentially cytotoxic, moribund, apoptotic neutrophils and neutrophilic apoptotic bodies in infectious/inflammatory foci [40]. Therefore, the inefficient clearance by macrophages of dying apoptotic neutrophils and neutrophil apoptotic bodies results in the progress towards secondary necrosis [5 and 41] and eventual lysis of the apoptotic, moribund neutrophils and apoptotic bodies exposing host tissues to the highly cytotoxic neutrophilic contents [40, 42 and 43].

The present observation that no apoptogenic activity towards sea bass phagocytes of virulent Ph. damselae subsp. piscicida cells or culture supernatants was seen in vitro suggests an indirect effect of the apoptogenic factor and may well explain how the bacterium is so pathogenic in vivo but is efficiently killed by macrophages in vitro [38].

Several studies have been carried out aiming at the identification of the factors/mechanisms involved in the virulence of Ph. damselae subsp. piscicida. However, understanding of the pathogenesis of pasteurellosis in sea bass is still limited and the facets of the interaction of the bacteria with the host phagocytes are far from clear. The presently reported induction of simultaneous macrophage and neutrophil apoptosis by Ph. damselae subsp. piscicida in sea bass, with eventual lysis of high numbers of the two phagocytes appears as a novel and very powerful pathogenic strategy of this fish pathogen. Examination of fish suffering from natural Ph. damselae subsp. piscicida infection will determine if induction of apoptosis is important for the pathogenesis of natural pasteurellosis.